Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The Effect of Including Benchmark Prevalence Data of Common Imaging Findings in Spine Image Reports on Health Care Utilization Among Adults Undergoing Spine Imaging

The Effect of Including Benchmark Prevalence Data of Common Imaging Findings in Spine Image... RESULTS We enrolled 250 401 participants (of whom 238 886 [95.4%] met eligibility for this Supplemental content analysis, with 137 373 [57.5%] women and 105 497 [44.2%] aged >60 years) from 3278 primary care Author affiliations and article information are clinicians. A total of 117 455 patients (49.2%) were randomized to the control group, and 121 431 listed at the end of this article. patients (50.8%) were randomized to the intervention group. There was no significant difference in cumulative spine-related RVUs comparing intervention and control conditions through 365 days. The adjusted median (interquartile range) RVU for the control group was 3.56 (2.71-5.12) compared with 3.53 (2.68-5.08) for the intervention group (difference, −0.7%; 95% CI, −2.9% to 1.5%; P = .54). Rates of subsequent RVUs did not differ between groups by specific clinical findings in the report but did differ by type of index imaging (eg, computed tomography: difference, −29.3%; 95% CI, −42.1% to −13.5%; magnetic resonance imaging: difference, −3.4%; 95% CI, −8.3% to 1.8%). We observed a (continued) Open Access. This is an open access article distributed under the terms of the CC-BY License. JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 1/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults Abstract (continued) small but significant decrease in the likelihood of opioid prescribing from a study clinician within 1 year of the intervention (odds ratio, 0.95; 95% CI, 0.91 to 1.00; P = .04). CONCLUSIONS AND RELEVANCE In this study, inserting benchmark prevalence information in lumbar spine imaging reports did not decrease subsequent spine-related RVUs but did reduce subsequent opioid prescriptions. The intervention text is simple, inexpensive, and easily implemented. TRIAL REGISTRATION ClinicalTrials.gov Identifier: NCT02015455 JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 Introduction 1,2 Spine imaging often reveals incidental findings among individuals without back pain, which can 3,4 5 lead to unnecessary and possibly harmful tests and treatments. Roland and van Tulder proposed adding statements to plain film reports describing the prevalence of degenerative findings in people 6,7 without back pain. In small observational studies, we and others have found that primary care patients undergoing lumbar spine imaging were less likely to receive certain subsequent diagnostic and therapeutic interventions if imaging reports contained information describing the prevalence of common imaging findings among individuals without back pain. These results suggest that benchmark information may reassure both patients and physicians, resulting in fewer downstream interventions. Since beginning our trial, others have published research suggesting that 7-9 contextualizing imaging information can affect both health care professionals and patients. We now report the results of a large, prospective randomized clinical trial of this intervention, the Lumbar Imaging with Reporting of Epidemiology (LIRE) trial. Our primary hypothesis was that patients of primary care professionals (PCPs) who received lumbar spine imaging reports with age-appropriate and imaging modality–appropriate benchmark prevalence data would have less spine-related health care utilization, as measured by our primary outcome, spine-related relative 10,11 value units (RVUs) (eAppendix 1 in Supplement 1). RVUs are based on Current Procedural Terminology (CPT) and provide a common metric for comparing health care utilization resulting from physician services. We also report the impact of the intervention on the prespecified secondary outcome of subsequent opioid prescriptions and prespecified subgroup analyses examining initial (index) imaging type and index report findings. Methods Study Design We conducted a multicenter, stepped-wedge, cluster randomized clinical trial assigning primary care clinics at 4 large health systems to when they would begin receiving lumbar spine imaging reports containing age-appropriate and modality-appropriate epidemiological benchmarks for common imaging findings. We previously published our study protocol, and it is available in Supplement 2. We designed LIRE to be highly pragmatic (eAppendix 2 in Supplement 1) to measure effects in routine care settings. We chose clinic-level cluster randomization because of the strong concern regarding contamination from intervention PCPs to control PCPs. We chose a stepped-wedge randomization because of the appeal of all clusters receiving the intervention by the end of the trial, facilitating implementation and the ability to perform both within-cluster (ie, before and after) and between-cluster comparisons. JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 2/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults Each health care system’s institutional review board or ethics committee reviewed the project, and all institutional review boards classified our study as minimal risk, granting waivers of both informed consent and Health Insurance Portability and Accountability Act authorization. This study followed the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline. Participants We enrolled clinics and their patients at 4 integrated health care systems: Kaiser Permanente Northern California; Henry Ford Health System in Michigan; Kaiser Permanente Washington; and Mayo Clinic Health System in Minnesota and Wisconsin. These systems have comprehensive electronic medical record (EMR) systems to capture health care utilization data. Clinic, PCP, and Patient Eligibility Criteria At each system, we identified adult primary care clinics and their physicians in family medicine, general internal medicine, and associated mid-level clinicians. We defined a LIRE clinician as a PCP whose main practice was at 1 clinic providing primary care and who ordered at least 1 qualifying imaging examination during the study period. We enrolled patients aged 18 years and older whose PCP from an eligible clinic ordered an imaging test of the lumbar spine between October 1, 2013, and September 30, 2016. We included all patients receiving eligible imaging studies at participating clinics who had not had lumbar spine imaging within the prior 12 months. We excluded only those patients who had opted out of research studies. Patient Identification We identified eligible patients and PCPs using the electronic ordering systems. When a PCP ordered an eligible examination, the system automatically determined whether the patient, PCP, and clinic were eligible. Randomization We used a stepped-wedge randomization scheme, randomly assigning clinics in each system to begin receiving the intervention at 1 of 5 dates at 6-month intervals from April 2014 through April 2016. We classified clinics in tertiles by their number of PCPs. The data coordinating center randomly selected clinics using urn-based randomization (without replacement) stratified by system and clinic size stratified by tertile (small, medium, and large). Clinic sizes were represented equally in each randomization wave. Because of the stepped-wedge temporal randomization scheme, we labeled clinics control if inclusion of the intervention text had not started and intervention after starting inclusion of the intervention text. Masking of the participating clinics was not feasible because of the nature of the intervention. Except for the biostatistician who received and cleaned the data, all investigators at the data coordinating center remained masked to clinic and participant assignment until the final stages of data analysis. Procedures The intervention text consisted of age-specific and modality-specific epidemiological benchmarks indicating the prevalence of common findings from imaging in people without back pain (eAppendix 5,6,15 3in Supplement 1). Using an automated approach through either the radiology information system or the EMR, we inserted the intervention text into lumbar spine imaging reports at intervention clinics. PCPs in control clinics received usual imaging reports. Data and Collection Methods We collected all data passively from the EMR and electronic administrative data systems. We performed 2 types of data queries from each system. To verify that the systems deployed the intervention appropriately, we queried systems 2 to 4 weeks after the start of each randomization wave for all patients who received an eligible lumbar imaging study. Text matching verified that the JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 3/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults reports contained the correct intervention text. One year after the first randomization wave and then every 6 months thereafter, we performed an additional query that included safety and outcome variables. Systems submitted both types of queries as limited data sets (deidentified except for dates of service) to the data coordinating center, providing unique study identifiers for each patient. We collected diagnosis and utilization data for patients 12 months before index imaging to characterize the cohort at the patient level. Each health system provided prescription data using national drug code or a similar classification from their pharmacy databases. This National Institutes of Health–sponsored trial required the collection of race and ethnicity data, which we obtained through the EMR. Outcomes Our primary outcome was the cumulative spine-related RVUs 365 days after index imaging. Spine- related RVUs are a composite measure of back pain interventions that combine the overall intensity of resource utilization for back pain care in a single metric. Our summary spine-related RVU incorporated procedures (CPT codes), diagnoses (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] and International Statistical Classification of Diseases, Tenth Revision, Clinical Modification [ICD-10-CM] codes), PCP visits, and inpatient hospitalizations and was based on a validated algorithm. eAppendix 1 in Supplement 1 provides examples of spine- related procedures and associated RVUs. To obtain a spine-related summary RVU from CPT and ICD-9-CM and ICD-10-CM codes, we used an existing validated algorithm when possible and used a 18,19 modified version for codes not accounted for by the algorithm. We aggregated relevant CPT codes through 1 year after the index imaging test to obtain total spine-related RVUs. The data coordinating center performed all calculations for assessing spine-related RVUs. We listed the following prespecified secondary outcomes in our published protocol: (1) an indicator of opioid prescribing after the index imaging; (2) cumulative spine-related total RVUs 2 years after index imaging; (3) subsequent advanced imaging (ie, number of magnetic resonance imaging [MRI] or computed tomography [CT] studies) within 90 days and 12 months after index imaging study; (4) spine injections and spine surgeries; and (5) other back-related medical costs during 2 years. In our original study protocol, the opioid outcome was the number of patients with a subsequent opioid prescription written by a study PCP. However, following discussions among the study team, we concluded that the total morphine equivalent dose (MED) prescribed per patient would be a better metric and thus included the number of MEDs prescribed per patient as the opioid outcome in our protocol paper. However, we were unable to obtain the necessary data for this detailed calculation. Instead, we analyzed whether patients had received an opioid prescription from a LIRE PCP within 1 year of index imaging; this was the outcome in our pilot project. We also report whether an opioid prescription was received within 90 days of index imaging, an outcome not prespecified on ClinicalTrials.gov. Extracting Imaging Results We used machine learning natural language processing to extract imaging findings from radiology text reports. We identified common imaging findings that are likely less clinically important (eg, disc bulge, disc space narrowing) vs likely more important (eg, moderate to severe spinal canal 1,21 stenosis, nerve root compression (eAppendix 4 in Supplement 1). Statistical Analysis To evaluate the impact of the intervention, we used multilevel linear mixed-effects models or generalized linear mixed models that cluster on clinic and then PCP within clinic, coupled with the use of robust standard errors for all primary and secondary outcome measures (Supplement 2). In secondary analyses, we used generalized estimating equations, adopting simple exchangeable correlation models at the clinic level to determine whether conclusions were sensitive to model specification (eAppendix 5 in Supplement 1). All analyses used the intention-to-treat principle. JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 4/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults We used a log transformation of RVU [log(RVU + 1)] in primary outcome models to address right skew of the utilization data. A constant (ie, 1) was added to RVU prior to transformation so that participants with 0 RVUs could be included in the analyses. We conducted sensitivity analyses that varied the constant added to RVU before the transformation (eAppendix 5 in Supplement 1). We also constructed a model examining subsequent RVUs in a subgroup of patients from clinics in which patients were less likely to have sought outside care who had utilization within their system through at least 12 months to address the issue of care received out of system. We used a similar analytic approach for opioid prescriptions as we used for spine-related RVUs but adapted generalized linear models to use logistic regression for this binary outcome. Realizing the potential importance of confounding due to secular trends in opioid prescribing, we conducted additional post hoc opioid analyses exploring sensitivity to alternative modeling of time. We also performed analyses on an outcome that incorporated opioid prescriptions from both LIRE and non- LIRE PCPs. We had 2 prespecified subgroup analyses. We used findings extracted from the reports to determine whether the findings in the imaging report influenced the effects of the intervention. We also examined whether the intervention effect was modified by modality of index imaging. We tested these hypotheses as interaction terms using the Wald test. A post hoc subgroup analysis distinguished between those patients who were and were not prior opioid users because the intervention might have been more likely to prevent opioid prescriptions from being written for opioid-naive patients. We defined prior use as at least 1 opioid prescription written within 120 days before the index imaging date, and we included an interaction term of prior opioid prescription status with intervention status in those models. We used model results and patient covariates to calculate predicted RVUs and predicted probability of opioid prescription for each participant under both the control and intervention conditions and aggregated the results to report median adjusted RVUs and adjusted opioid prescriptions by intervention status and subgroups. We used SAS software version 9.4 (SAS Institute) for all analyses. Statistical significance was set at P < .05, and all tests were 2-tailed. Power for Primary Outcome We calculated statistical power for the primary outcome, spine-related RVUs. The study had 89% power to detect reductions of 5.0% or greater. Data Safety Monitoring Two external safety officers monitored emergency department visits within 90 days and deaths within 6 months of index imaging. The safety officers used absolute relative risk ratio monitoring thresholds of 1.15 and 1.10 for comparing 90-day emergency department visit and death rates by intervention group, with adjustment for patient-specific characteristics (ie, age, sex, Charlson 23 24 comorbidity index ), health care system, image modality, time, season, and clinic size. Results We randomly allocated intervention start dates to 98 clinics with 3278 PCPs and 250 401 patients. A total of 11 515 patients were excluded for the following reasons: prior lumbar spine image within 12 months (11 149 [96.8%]), imaging report finalization date more than 4 days after image completion date (354 [3.1%]), image completion date prior to report finalization date (3 [<0.1%]), and unable to link to utilization data (9 [0.1%]). This resulted in a final sample of 238 886 patients (95.4%; 137 373 [57.5%] women; 105 497 [44.2%] aged >60 years) with 3257 PCPs (99.4%). Three health systems were of comparable size and enrolled 41 882 patients (17.5%) from 936 PCPs (28.7%) while the fourth health system enrolled 197 004 patients (82.5%) from 2321 PCPs (71.3%) (Figure 1). We did JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 5/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults not observe any substantial differences in the baseline characteristics between the control and intervention groups (Table). Our primary outcome, 12-month spine-related RVU, was not significantly different for the intervention group compared with the control group (adjusted median [interquartile range], 3.53 [2.68-5.08] vs 3.56 [2.71-5.12]; difference, −0.7%; 95% CI, −2.9% to 1.5%; P = .54) (Figure 2). Injections and surgery accounted for a higher proportion of subsequent spine-related RVUs for patients who had magnetic resonance imaging or computed tomography for their index examination compared with radiographs, while physical therapy and imaging were proportionally higher for patients who had radiographs as the index imaging test (eAppendix 7 in Supplement 1). Our prespecified secondary outcome, opioid prescriptions by a LIRE PCP within 1 year of index imaging, demonstrated a small but statistically significant reduction in the odds of receiving at least 1 prescription for an opioid for patients in the intervention group compared with patients in the control group (adjusted opioid proportion, 36.2% vs 37.0%; odds ratio, 0.95; 95% CI, 0.91-1.00; P = .04) (Figure 3). Sensitivity analyses with alternative modeling of time yielded similar results (Figure 3). Comparison of opioid prescribing between control and intervention groups within 90 days following index imaging showed a similar small reduction in the odds of receiving an opioid prescription for the intervention group compared with the control group (adjusted opioid proportion, 28.9% vs 29.8%; odds ratio, 0.95; 95% CI, 0.90-0.99; P = .02) (eAppendix 6 in Supplement 1). Safety monitoring demonstrated no evidence of increased deaths or emergency department visits in the intervention vs control group within 6 months of the index test (adjusted emergency department visit rate, 11.1% vs 11.3%; OR, 0.98; 95% CI, 0.94-1.01) (Figure 4). Figure 1. CONSORT Stepped-Wedge Allocation of Trial Subjects Clinics under control condition Clinics under intervention condition b b Step 0 Step 1 Step 2 Step 3 Step 4 Step 5 Clinic October 2013- April 2014- October 2014- April 2015- October 2015- April 2016- Total Group March 2014 September 2014 March 2015 September 2015 March 2016 September 2016 41 558 Analyzed 10 630 Analyzed 52 188 Analyzed 34 219 Intervention 78 Intervention 2394 Excluded (19 clinics) 7339 No intervention 970 Excluded 1424 Excluded 31 611 Analyzed 15 605 Analyzed 47 216 Analyzed 29 167 Intervention 4 Intervention 2158 Excluded (20 clinics) 2444 No intervention 1134 Excluded 1024 Excluded 30 157 Analyzed 29 628 Analyzed 59 785 Analyzed 25 313 Intervention 394 Intervention 2766 Excluded (20 clinics) 4944 No intervention 1788 Excluded 978 Excluded 21 970 Analyzed 10 277 Analyzed 32 247 Analyzed 194 Intervention 9433 Intervention 1887 Excluded (18 clinics) 1428 Excluded 844 No Intervention 459 Excluded 39 622 Analyzed 10 277 Analyzed 47 450 Analyzed 7411 Intervention 114 Intervention 2310 Excluded (21 clinics) 2037 Excluded 417 No Intervention 273 Excluded Totals 7828Analyzed 117 455 Analyzed 121 431 Analyzed 238 886 Analyzed 7411Intervention All 784 Intervention 105 543 Intervention 11 515 Excluded (98 clinics) 417 No intervention 7357 Excluded 15 888 No intervention 273 Excluded 4158 Excluded For clinics under the control condition, intervention indicates the intervention text was By pretrial design, for 1 clinic, step 0 extended through May 2014, and step 1 began mistakenly included in the image report. For clinics under the intervention condition, June 1, 2014. intervention indicates that the intervention text was successfully included in the image report and no intervention indicates that the intervention text was not included. Two small clinics randomized to groups 2 and 5 were dropped before the first data submission because of clinic closure and are not included in the clinic counts. JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 6/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults Table. Baseline Characteristics No. (%) Characteristic Control (n = 117 455) Intervention (n = 121 431) Site A 6950 (5.9) 7388 (6.1) B 96 275 (82.0) 100 729 (83.0) C 7846 (6.7) 7736 (6.4) D 6384 (5.4) 5588 (4.6) Age, y 18-39 21 237 (18.1) 22 105 (18.2) 40-60 45 032 (38.3) 44 995 (37.1) ≥61 51 186 (43.6) 54 331 (44.7) Sex Women 67 915 (57.8) 69 458 (57.2) Men 49 534 (42.2) 51 965 (42.8) Race American Indian or Alaska Native 806 (0.7) 880 (0.7) Asian 13 311 (11.3) 13 197 (10.9) Black or African American 11 919 (10.1) 11 649 (9.6) Native Hawaiian or other Pacific Islander 905 (0.8) 709 (0.6) White 76 431 (65.1) 79 142 (65.2) Multiracial or other 459 (0.4) 546 (0.4) Unknown or not reported 13 624 (11.6) 15 308 (12.6) Ethnicity Hispanic or Latino 17 754 (15.1) 18 475 (15.2) Not Hispanic or Latino 19 867 (16.9) 19 276 (15.9) Not available 79 834 (68.0) 83 680 (68.9) Modality RG 93 465 (79.6) 98 970 (81.5) CT 494 (0.4) 449 (0.4) MR 23 496 (20) 22 012 (18.1) Charlson Comorbidity Index 0 75 106 (63.9) 77 973 (64.2) 1 20 675 (17.6) 21 193 (17.5) 2 11 451 (9.7) 11 760 (9.7) ≥3 10 223 (8.7) 10 505 (8.7) Finding status None 27 770 (23.6) 27 776 (22.9) LIRE finding without clinically important finding 72 127 (61.4) 77 065 (63.5) Clinically important finding 17 558 (14.9) 16 590 (13.7) ≥1 Opioid prescriptions prior to index 32 225 (27.4) 29 306 (24.1) Primary insurance at index Medicare 44 362 (37.8) 46 479 (38.3) Medicaid or state-subsidized 5546 (4.7) 6510 (5.4) Commercial 65 375 (55.7) 66 368 (54.7) VA 117 (0.1) 131 (0.1) Self-pay 731 (0.6) 570 (0.5) Unknown or not reported 1324 (1.1) 1373 (1.1) Socioeconomic index, mean (SD) 57 (6) 57 (7) Health care professional type MD 105 359 (89.7) 108 165 (89.1) DO 8131 (6.9) 9157 (7.5) Extender, eg, NP, PA 3965 (3.4) 4109 (3.4) (continued) JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 7/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults Table. Baseline Characteristics (continued) No. (%) Characteristic Control (n = 117 455) Intervention (n = 121 431) Health care professional specialty Family medicine 56 795 (48.4) 60 277 (49.6) Internal medicine 59 684 (50.8) 60 158 (49.5) Other 976 (0.8) 996 (0.8) Female health care professional 62 840 (53.5) 62 680 (51.6) Health care professional age, mean (SD), y 49 (9) 49 (9) Abbreviations: CT, computed tomography; DO, doctor of osteopathy; MD, medical doctor; LIRE, Lumbar Imaging with Reporting of Epidemiology; MR, magnetic resonance; NP, nurse practitioner; PA, physician’s assistant; RG, radiograph; VA, Veterans Administration. Does not include 14 patients (<0.1%) with other or unknown gender. Due to the manner in which race and ethnicity are collected at 1 health system (ie, sometimes the concepts are conflated and sometimes Hispanic ethnicity is captured by a single checkbox), it is not possible to reliably distinguish between “not Hispanic” and “did not answer.” Does not include 6810 patients (2.7%) with unknown socioeconomic index. Sites mapped participant addresses to Federal Information Processing System codes at the block-group level using geocoding software. These codes were mapped to socioeconomic indices derived from data available from the 2010 Census Summary File 1 and the American Community Survey, 2007 to 2011, 5-year estimate data. Does not include 424 patients (0.1%) for whom provider age was unknown. Figure 2. Model Results for Spine-Related Relative Value Units (RVUs) at 1 Year Adjusted Adjusted median RVU median RVU Favors Favors Population control intervention Difference (95% CI), % intervention control P value Whole cohort 3.56 3.53 –0.7 (–2.9 to 1.5) .54 All models adjust for health system, clinic size, age Index imaging modality range (ie, 18-39, 40-60, and61 years), sex, imaging RG 3.19 3.19 0.1 (–2.0 to 2.3) modality, Charlson Comorbidity Index category (ie, 0, CT 10.40 7.07 –29.3 (–42.1 to –13.5) .01 1, 2, and3), and health system specific time trends. MR 7.67 7.37 –3.4 (–8.3 to 1.8) Models include hierarchical random effects for clinic Image finding type (intercept and treatment) and primary care Likely CI 9.26 8.83 –4.2 (–9.0 to 0.9) professional (intercept only). P values for subgroup LIRE finding without likely CI 3.60 3.58 –0.4 (–2.6 to 1.9) models (ie, index imaging type and image finding type) Neither finding type 2.35 2.36 0.3 (–2.7 to 3.3) are for Wald tests for effect modification. CI indicates –30 –20 –10 0 10 clinically important, CT, computed tomography; RG, Difference (95% CI), % radiograph; and MR, magnetic resonance. The prespecified subgroup analysis of whether the intervention differentially affected spine- related RVUs by imaging modality revealed that the small number of patients (943 [0.4%]) of patients who had computed tomography as the index imaging had markedly lower subsequent median RVUs if exposed to the intervention (mean difference, −29.3%; 95% CI,−42.1% to −13.5%). The nearly 20% of patients (45 508 [19.1%]) who had magnetic resonance imaging had lower subsequent RVUs in the intervention group (difference, −3.4%; 95% CI, −8.3% to 1.8%), although this was not statistically significant (Figure 2). The second prespecified subgroup analysis that examined whether image finding type differentially affected spine-related RVUs revealed no differences in subsequent median RVUs in the intervention compared with the control group (Figure 2). In a post hoc subgroup analysis, the adjusted proportion of control patients without a prior opioid prescription who received an opioid prescription from a LIRE PCP within 1 year following index imaging was 25% compared with 72% for control patients with a prior opioid prescription. However, there was no intervention effect modification by prior opioid prescription status (test for effect modification, P = .58) (Figure 3). When we included prescriptions from non-LIRE PCPs who were not exposed to the intervention in the 1-year opioid outcome, the intervention effect was attenuated JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 8/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults (adjusted opioid proportion, 47.1% intervention vs 47.5% control; OR, 0.98; 95% CI, 0.94-1.02; P = .27) (Figure 3). Discussion The LIRE intervention did not reduce subsequent spine-related RVUs for the population as a whole. However, patients in the intervention group were less likely (OR, 0.95; 95% CI, 0.90-0.99; P =.02) to receive a subsequent opioid prescription compared with patients not receiving the intervention. The intervention also reduced subsequent spine-related RVUs for the small proportion of patients with CT as the index imaging. Pragmatic trials must be simple to implement and the populations relatively unselected. Thus, a 25-27 negative primary result is not unusual. This suggests the likely importance of heterogeneous intervention effects, prespecified subgroup analyses, and prespecified secondary outcomes. An explanation for the differential effect by imaging modality is that patients undergoing CT for their index imaging were more likely to receive back pain interventions than patients receiving other modalities, and thus, the intervention was more effective at reducing subsequent interventions in patients who were most likely to receive those interventions in the first place (eAppendix 7 in Supplement 1). Our finding of no greater subsequent emergency department visits and deaths in the intervention group provides reassurance that the intervention did not cause deleterious undertreatment. Given the climate of overdiagnosis and overtreatment of back pain in the United Figure 3. Model Results for Opioid Prescriptions Within 12 months Adjusted Adjusted opioid rate opioid rate Favors Favors Population control, % intervention, % OR (95% CI) intervention control P value Whole cohort 37.0 36.2 0.95 (0.91-1.00) .04 Prior opioid prescription No 24.8 24.1 0.96 (0.91-1.01) NA Yes 72.2 70.8 0.93 (0.85-1.02) NA Alternative time definition Spline 1 knot 37.0 36.2 0.95 (0.91-1.00) .04 Spline 2 knots 37.0 36.2 0.95 (0.91-1.00) .03 Alternative source definition Prescription from any source 47.5 47.1 0.98 (0.94-1.02) .27 Prescription from non-LIRE source 23.4 23.4 1.00 (0.96-1.04) .96 –30 –20 –10 0 10 OR (95% CI) All models adjust for health system, clinic size, age range (ie, 18-39, 40-60, and61 the 120 days prior to index imaging. A Lumbar Imaging with Reporting of Epidemiology years), sex, imaging modality, Charlson Comorbidity Index category (ie, 0, 1, 2, and3), (LIRE) source is any health care professional who ordered an index lumbar spine image prior opioid use, and health system specific time trends. Models include hierarchical for 1 or more participants in the LIRE trial. It need not be the same individual who ordered random effects for clinic (intercept and treatment) and primary care professional the patient’s index image. A non-LIRE source is any other health care professional. Any (intercept only). Prior opioid prescription is defined as having 1 or more prescriptions in source includes both LIRE and non-LIRE clinicians. NA indicates not applicable. Figure 4. Safety Outcomes Adjusted rate Adjusted rate Favors Favors Outcome control, % intervention, % OR (95% CI) intervention control ED visit within 90 days 11.3 11.1 0.98 (0.94-1.01) 6-month mortality 11.3 0.81 1.03 (0.88-1.20) 0.90 0.95 1.00 1.05 1.10 OR (95% CI) All models adjust for health system, clinic size, age range (ie, 18-39, 40-60, and61 visit model includes hierarchical random effects for clinic (intercept and treatment) and years), sex, imaging modality, Charlson Comorbidity Index category (ie, 0, 1, 2, and3), primary care professional (intercept only). The mortality model uses general estimating seasonality, and health system specific time trends. The emergency department (ED) equations with clustering on clinic. JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 9/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults States, undertreatment may be less likely to occur in the United States than elsewhere. Our intervention provided an opportunity to increase the knowledge of patients and health care professionals. Because we did not detect any harm of the intervention and we did detect a possible benefit, including the intervention should safely allow patients and health care professionals to make better informed decisions. Finally, our primary null result may have been different if we had studied different health systems. For example, if we had enrolled clinics with higher baseline utilization of tests for back pain patients, we may have found a positive result. Limitations This study has limitations. Opioid prescribing decreased in the United States during our study. Although we made multiple efforts to account for this potential confounding in our modeling, residual confounding may exist. Because we did not collect patient-reported outcomes, we cannot comment on outcomes such as functional status, pain, or psychosocial functioning. The decision not to collect patient-reported data was deliberate, based on the recognition that it could jeopardize the feasibility of this large pragmatic trial of more than 250 000 patients. We also did not capture patient care not included in the EMRs. However, we found similar results to those of our primary analysis when we examined subsequent RVUs from patients less likely to seek outside care (eAppendix 5 in Supplement 1). Previous studies have shown high degrees of accuracy when EMR data were validated by manual medical record reviews. All of our participating health systems were integrated delivery systems and nonprofit. There is evidence that nonprofit hospitals may be less responsive to the type of intervention that we tested than for-profit hospitals. However, this conservative bias emphasizes the robustness of the positive impact that we observed with respect to opioid prescribing. Our findings may also not be generalizable to systems having greater restrictions on advanced imaging. We do not know the indication for imaging, including whether the patient had a red flag, so we cannot comment on the appropriateness. Conclusions In this study, adding benchmark prevalence information for spine imaging findings did not reduce subsequent spine-related RVUs, but it slightly reduced the likelihood of subsequent opioid prescribing, an important prespecified secondary outcome. Reporting benchmark information is a fundamental change to the imaging reporting paradigm that may be relevant for other conditions and could easily be applied to other diagnostic tests (eg, other imaging tests, genetic testing). Finally, unmeasured benefits of the intervention may result from patients and health care professionals having a better understanding of the clinical meaning of imaging findings. ARTICLE INFORMATION Accepted for Publication: June 12, 2020. Published: September 4, 2020. doi:10.1001/jamanetworkopen.2020.15713 Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Jarvik JG et al. JAMA Network Open. Corresponding Author: Jeffrey G. Jarvik, MD, MPH, Department of Radiology, University of Washington, 325 Ninth Ave, Box 359728, Seattle, WA 98104-2499 (jarvikj@uw.edu). Author Affiliations: Department of Radiology, University of Washington, Seattle (Jarvik, James, Gold, Bresnahan); Department of Neurological Surgery, University of Washington, Seattle (Jarvik); Department of Health Services, University of Washington, Seattle (Jarvik, Kessler); Comparative Effectiveness, Cost, and Outcomes Research Center, University of Washington, Seattle (Jarvik, James, Gold, Rundell, Friedly, Turner, Bresnahan); Department JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 10/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults of Biostatistics, University of Washington, Seattle (Meier, Tan, Comstock, Heagerty); Center for Biomedical Statistics, University of Washington, Seattle (Meier, Tan, Comstock, Heagerty); Flatiron Health, New York, New York (Tan); Rehabilitation Care Services, VA Puget Sound Health Care System, Seattle, Washington (Suri); Department of Rehabilitation Medicine, University of Washington, Seattle (Suri, Rundell, Friedly, Turner); Department of Radiology, Mayo Clinic, Rochester, Minnesota (Kallmes, Luetmer); Kaiser Permanente Washington, Seattle (Cherkin, Sherman); Departments of Family Medicine and Internal Medicine, Oregon Health and Science University, Portland (Deyo); Department of Radiology, Henry Ford Hospital, Detroit, Michigan (Halabi, Griffith); Department of Radiology, Stanford University School of Medicine, Palo Alto, California (Halabi); Division of Research, Kaiser Permanente Northern California, Oakland, California (Avins); Surgical Outcomes Research Center, University of Washington, Seattle (Lavallee); Department of Psychiatry and Behavioral Sciences, University of Washington, Seattle (Stephens, Turner). Author Contributions: Dr Jarvik had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: Jarvik, Kallmes, Deyo, Comstock, Turner, Bresnahan, Heagerty. Acquisition, analysis, or interpretation of data: Jarvik, Meier, James, Gold, Tan, Kessler, Suri, Cherkin, Sherman, Halabi, Comstock, Luetmer, Avins, Rundell, Griffith, Friedly, Lavallee, Stephens, Turner, Bresnahan, Heagerty. Drafting of the manuscript: Jarvik, Meier, Gold, Suri, Lavallee, Stephens, Turner, Heagerty. Critical revision of the manuscript for important intellectual content: Jarvik, Meier, James, Gold, Tan, Kessler, Suri, Kallmes, Cherkin, Deyo, Sherman, Halabi, Comstock, Luetmer, Avins, Rundell, Griffith, Friedly, Turner, Bresnahan, Heagerty. Statistical analysis: Jarvik, Meier, Gold, Tan, Kessler, Comstock, Stephens, Heagerty. Obtained funding: Jarvik, Comstock, Avins, Heagerty. Administrative, technical, or material support: Jarvik, James, Tan, Suri, Cherkin, Deyo, Halabi, Luetmer, Lavallee, Stephens, Heagerty. Supervision: Jarvik, James, Kessler, Kallmes, Luetmer, Lavallee, Heagerty. Conflict of Interest Disclosures: Dr Jarvik reported receiving royalties from Springer Publishing and Wolters/ Kluwer/UpToDate and receiving travel reimbursement from GE–Association of University Radiologists outside the submitted work. Dr Tan reported being employed by Flatiron Health and owning stock in Roche outside the submitted work. Dr Deyo reported having an endowed professorship funded by Kaiser Permanente and receiving personal fees from UpToDate outside the submitted work. Dr Rundell reported receiving grants from the Institute of Translational Health Sciences, the Center for Large Data Research and Data Sharing in Rehabilitation, the Center on Health Services Training and Research, and the Scan Design Foundation; having a contract with the Agency for Healthcare Research and Quality; and receiving personal fees from the Department of Defense and ATI Physical Therapy outside the submitted work. Dr Friedly reported receiving grants from the Department of Defense and salary support from the American Academy of Physical Medicine and Rehabilitation for serving as editor in chief outside the submitted work. No other disclosures were reported. Funding/Support: This work was supported within the National Institutes of Health (NIH) Health Care Systems Re- search Collaboratory by the NIH Common Fund through cooperative agreement U24AT009676 from the Office of Strategic Coordination within the Office of the NIH Director and cooperative agreements UH2AT007766 and UH3AR066795 from the National Institute of Arthritis and Musculoskeletal and Skin Diseases. It was also supported by the University of Washington Clinical Learning, Evidence, And Research Center for Musculoskeletal Disorders funded by NIH/National Institute of Arthritis and Musculoskeletal and Skin Diseases grant P30AR072572. Role of the Funder/Sponsor: The NIH Healthcare Systems Collaboratory provided oversight of the design and conduct of the study, the collection, management, analysis, and interpretation of the data. They did not participate in the preparation, review, or approval of the manuscript, nor the decision to submit the manuscript for publication. Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Data Sharing Statement: See Supplement 3. REFERENCES 1. Jarvik JJ, Hollingworth W, Heagerty P, Haynor DR, Deyo RA. The Longitudinal Assessment of Imaging and Disability of the Back (LAIDBack) Study: baseline data. Spine (Phila Pa 1976). 2001;26(10):1158-1166. doi:10.1097/ 00007632-200105150-00014 2. Brinjikji W, Luetmer PH, Comstock B, et al. Systematic literature review of imaging features of spinal degeneration in asymptomatic populations. AJNR Am J Neuroradiol. 2015;36(4):811-816. JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 11/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults 3. Jarvik JG, Gold LS, Comstock BA, et al. Association of early imaging for back pain with clinical outcomes in older adults. JAMA. 2015;313(11):1143-1153. doi:10.1001/jama.2015.1871 4. Graves JM, Fulton-Kehoe D, Jarvik JG, Franklin GM. Health care utilization and costs associated with adherence to clinical practice guidelines for early magnetic resonance imaging among workers with acute occupational low back pain. Health Serv Res. 2014;49(2):645-665. doi:10.1111/1475-6773.12098 5. Roland M, van Tulder M. Should radiologists change the way they report plain radiography of the spine? Lancet. 1998;352(9123):229-230. doi:10.1016/S0140-6736(97)11499-4 6. McCullough BJ, Johnson GR, Martin BI, Jarvik JG. Lumbar MR imaging and reporting epidemiology: do epidemiologic data in reports affect clinical management? Radiology. 2012;262(3):941-946. doi:10.1148/radiol. 7. Fried JG, Andrew AS, Ring NY, Pastel DA. Changes in primary care health care utilization after inclusion of epidemiologic data in lumbar spine MR imaging reports for uncomplicated low back pain. Radiology. 2018;287(2): 563-569. doi:10.1148/radiol.2017170722 8. Karran EL, Yau YH, Hillier SL, Moseley GL. The reassuring potential of spinal imaging results: development and testing of a brief, psycho-education intervention for patients attending secondary care. Eur Spine J. 2018;27(1): 101-108. doi:10.1007/s00586-017-5389-8 9. Medalian Y, Moseley GL, Karran EL. An online investigation into the impact of adding epidemiological information to imaging reports for low back pain. Scand J Pain. 2019;19(3):629-633. doi:10.1515/sjpain-2019-0023 10. Hsiao WC, Braun P, Dunn D, Becker ER. Resource-based relative values: an overview. JAMA. 1988;260(16): 2347-2353. doi:10.1001/jama.1988.03410160021004 11. Hsiao WC, Braun P, Yntema D, Becker ER. Estimating physicians’ work for a resource-based relative-value scale. N Engl J Med. 1988;319(13):835-841. doi:10.1056/NEJM198809293191305 12. American Medical Association. CPT (Current Procedural Terminology). Accessed August 4, 2020. https://www. ama-assn.org/amaone/cpt-current-procedural-terminology 13. Jarvik JG, Comstock BA, James KT, et al. Lumbar Imaging With Reporting Of Epidemiology (LIRE)—protocol for a pragmatic cluster randomized trial. Contemp Clin Trials. 2015;45(Pt B):157-163. doi:10.1016/j.cct.2015.10.003 14. Johnson KE, Neta G, Dember LM, et al. Use of PRECIS ratings in the National Institutes of Health (NIH) Health Care Systems Research Collaboratory. Trials. 2016;17:32. doi:10.1186/s13063-016-1158-y 15. Brinjikji W, Diehn FE, Jarvik JG, et al. MRI findings of disc degeneration are more prevalent in adults with low back pain than in asymptomatic controls: a systematic review and meta-analysis. AJNR Am J Neuroradiol. 2015;36 (12):2394-2399. doi:10.3174/ajnr.A4498 16. World Health Organization. International Classification of Diseases, Ninth Revision (ICD-9). World Health Organization; 1977. 17. World Health Organization. International Statistical Classification of Diseases, Tenth Revision (ICD-10). World Health Organization; 1992. 18. Martin B, Mirza SK, Lurie JD, Tosteson ANA, Deyo RA. Validation of an administrative coding algorithm to identify back-related degenerative diagnoses. Paper presented at: International Society for the Study of the Lumbar Spine (ISSLS); May 14, 2013; Scottsdale, AZ. 19. Martin BI, Mirza SK, Franklin GM, Lurie JD, MacKenzie TA, Deyo RA. Hospital and surgeon variation in complications and repeat surgery following incident lumbar fusion for common degenerative diagnoses. Health Serv Res. 2013;48(1):1-25. doi:10.1111/j.1475-6773.2012.01434.x 20. Tan WK, Hassanpour S, Heagerty PJ, et al. Comparison of natural language processing rules-based and machine-learning systems to identify lumbar spine imaging findings related to low back pain. Acad Radiol. 2018;25 (11):1422-1432. doi:10.1016/j.acra.2018.03.008 21. Jarvik JG, Hollingworth W, Heagerty PJ, Haynor DR, Boyko EJ, Deyo RA. Three-year incidence of low back pain in an initially asymptomatic cohort: clinical and imaging risk factors. Spine (Phila Pa 1976). 2005;30(13):1541-1548. doi:10.1097/01.brs.0000167536.60002.87 22. Turner JA, Shortreed SM, Saunders KW, LeResche L, Von Korff M. Association of levels of opioid use with pain and activity interference among patients initiating chronic opioid therapy: a longitudinal study. Pain. 2016;157(4): 849-857. doi:10.1097/j.pain.0000000000000452 23. Charlson M, Szatrowski TP, Peterson J, Gold J. Validation of a combined comorbidity index. J Clin Epidemiol. 1994;47(11):1245-1251. doi:10.1016/0895-4356(94)90129-5 JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 12/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults 24. Quan H, Li B, Couris CM, et al. Updating and validating the Charlson comorbidity index and score for risk adjustment in hospital discharge abstracts using data from 6 countries. Am J Epidemiol. 2011;173(6):676-682. doi: 10.1093/aje/kwq433 25. Dember LM, Lacson E Jr, Brunelli SM, et al. The TiME trial: a fully embedded, cluster-randomized, pragmatic trial of hemodialysis session duration. J Am Soc Nephrol. 2019;30(5):890-903. doi:10.1681/ASN.2018090945 26. Huang SS, Septimus E, Kleinman K, et al; ABATE Infection Trial Team. Chlorhexidine versus routine bathing to prevent multidrug-resistant organisms and all-cause bloodstream infections in general medical and surgical units (ABATE Infection trial): a cluster-randomised trial. Lancet. 2019;393(10177):1205-1215. doi:10.1016/S0140-6736 (18)32593-5 27. Coronado GD, Petrik AF, Vollmer WM, et al. Effectiveness of a mailed colorectal cancer screening outreach program in community health clinics: the STOP CRC cluster randomized clinical trial. JAMA Intern Med. 2018;178 (9):1174-1181. doi:10.1001/jamainternmed.2018.3629 28. Pezalla EJ, Rosen D, Erensen JG, Haddox JD, Mayne TJ. Secular trends in opioid prescribing in the USA. J Pain Res. 2017;10:383-387. doi:10.2147/JPR.S129553 29. Patel NK, Moses RA, Martin BI, Lurie JD, Mirza SK. Validation of using claims data to measure safety of lumbar fusion surgery. Spine (Phila Pa 1976). 2017;42(9):682-691. doi:10.1097/BRS.0000000000001879 30. Horwitz JR. Making profits and providing care: comparing nonprofit, for-profit, and government hospitals. Health Aff (Millwood). 2005;24(3):790-801. doi:10.1377/hlthaff.24.3.790 SUPPLEMENT 1. eAppendix 1. Outcome Sources and Definitions eAppendix 2. PRECIS-2 Diagram eAppendix 3. Intervention Text eAppendix 4. Imaging Findings eAppendix 5. Sensitivity Analyses for Spine-Related Relative Value Units (RVUs) Outcome eAppendix 6. Opioid Prescription Within 90 Days eAppendix 7. Patient Characteristics and Outcomes by Index Modality eReferences. SUPPLEMENT 2. Trial Protocol and Statistical Analysis Plan SUPPLEMENT 3. Data Sharing Statement JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 13/13 Lumbar Imaging Reporting with Epidemiology (LIRE) Study Protocol Version 3.0 ! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Table!of!Contents! 1.0 Project!Summary!…………………………………………………………………………………………………………………4! ! 1.1!!!!!!!Project!Organization! Schema'1:'Overall'Organization'Chart' Schema'2:'Site'Organization'Chart' Personnel'Directory' 1.2!! Interaction!with!Collaboratory!Coordinating!Center! 2.0 Background!and!Rationale………………………………………………………………………………………..…………13! Figure'1:'MRI 'of'spine'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''' Figure'2:'Epidemiologic'Statement'Included'In'Lumbar'Spine'MR'Imaging'Reports '''''''''''' Table'1:'Outcomes'Of'Patients'Whose'Imaging'Did'And'Did'Not'Include'A'Statement' Containing'Epidemiological'Benchmarks' 3.0! Specific!Aims…………………………………………………………………………………………………………………..…16! 4.0! Study!Details……………………………………………………………………………………………………………………..17! 4.1!Eligibility!Criteria! 4.2!Consent!procedure! 4.3!Inclusion!and!Exclusion!Criteria! 5.0! Research!Design!and!Methods……………………………………………………………………………………..……18! 5.1!Clinic/Practitioner/Patient!Identification! ! 5.2!Randomization! Figure'3:'Proposed'Randomization'Schedule' 5.3!Clinic/Patient!Enrollment! Table'2:'Lumbar'Examinations'to'be'Included'in'Pragmatic'Trial' 5.4!Data!Collection! 5.5!Aims!for!UH2!Phase! Table'3:'AgeMspecific'Rates'of'Lumbar'Spine'Imaging'Findings'' Figure'4:'Comparison'of'Parallel,'Crossover'and'Stepped'Wedge'Designs'' 5.6!Working!Groups! Table'4:'Milestones'for'UH2'Planning'Year'Needed'to'Transition'to'UH3'Impleme ntation' 5.7!Aims!for!UH3!Phase! Table'5:'Sample'of'RVUs'and'CMS Mbased'payment'amounts'for'lumbar'imaging' Table'6:''Milestones'for'UH3'(Implementation'Phase) M'Years'2M5' Table'7:'Timeline'for'the'UH3'Phase'(Years'2M5)' 2013409426/kj! 2! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! 6.0! Statistical!Considerations…………………………………………………………………………………………………31! 7.0! Human!Subjects!………………………………………………………………………………………………………………31! 7.1!Human!Subjects!Involvement!and!Characteristics!! 7.2!Research!Data!!! 7.3!Potential!risks! 7.4!Adequacy!of!Protection!Against!Risks! 7.5!Potential!Benefits!of!the!Proposed!Research!to!the!Subjects!! 7.6!Importance!o!the!Knowledge!to!be!Gained! 8.0! Data!and!Safety!Monitoring……………………………………………………………………………………….……33! 9.0! References………………………………………………………………………………………………………………………34! 10.0! Appendices! ! ' Appendix'A:'Article'Extraction'form'(Working'Group'1)' Appendix'B:'Literature'Search'and'Articles'used'in'Intervention'text'(Working'Group'1)' Appendix'C:'Pilot'Implementation'Site'Checklist'(Working'Group'2)' Appendix'D:'Literature'Search'for'RVUMbased'Assessment'(Working'Group'3)' ! ! 2013409426/kj! 3! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! 1.!Project!Summary! Low!back!pain,!an!Institute!of!Medicine!priority!condition!for!comparative!effectiveness!research,!is!of! major! public! health! importance.! It! is! one! of! the! most! common! reasons! for! physician! visits! and! an! important!cause!of!functional!limitation!and!disability.!Imaging!is!frequently!performed!as!part!of!the! diagnostic!evaluation!and!is!an!important!contributor!to!the!cost!of!back!pain!care,!which!totaled!more! than!$86!billion!in!2005.!It!is!well!known!that,!even!without!back!pain,!magnetic!resonance!(MR)!imaging! of! the! lumbar! spine! frequently! reveals! findings! such! as! disc! desiccation! or! bulging.! Patients! and! their! providers! may! attribute! greater! importance! to! these! findings,! which! are! often! age\related,! than! they! should,!because!they!do!not!have!an!appropriate!frame!of!reference!in!which!to!interpret!the!findings.! These!“incidental”!findings!may!initiate!a!cascade!of!events!leading!possibly!even!to!surgery,!without! improving!patient!outcomes.!! The! overall! goal! of! the! Lumbar! Image! Reporting! with! Epidemiology! (LIRE)! trial,! is! to! perform! a! large,! pragmatic,!randomized!controlled!trial!to!determine!the!effectiveness!of!a!simple,!inexpensive!and!easy! to!deploy!intervention!–!of!inserting!epidemiological!benchmarks!into!lumbar!spine!imaging!reports!–!at! reducing! subsequent! tests! and! treatments.! The! long\term! public! health! significance! is! that! our! intervention!has!the!potential!to!substantially!reduce!unnecessary!and!expensive!care!not!only!for!back! pain,!but!also!for!a!wide!range!of!other!conditions,!since!it!could!easily!be!applied!to!other!diagnostic! tests! (e.g.! other! imaging! tests,! laboratory! tests,! genetic! testing).! If! our! study! is! positive,! adding! epidemiologic! benchmarks! to! diagnostic! test! reporting! could! become! the! dominant! paradigm! for! communicating!all!diagnostic!information.! We! propose! an! efficient,! novel,! cluster! randomized! design! referred! to! as! a! “stepped! wedge”! design,! permitting!longitudinal!comparisons!while!controlling!for!temporal!trends.!We!plan!to!passively!collect! primary! outcome! measures! of! healthcare! utilization! both! pre\! and! post\intervention,! using! robust! electronic!medical!records!at!the!participating!sites.!We!hypothesize!that!for!patients!of!primary!care! providers,!inserting!epidemiological!benchmarks!in!lumbar!spine!imaging!reports!will!reduce!subsequent! diagnostic!and!therapeutic!interventions,!including!MR!and!CT,!opioid!prescriptions,!spinal!injections!and! surgery.! The! rationale! is! that! the! epidemiologic! data! may! provide! a! context! for! both! physicians! and! patients!to!better!interpret!imaging!findings.!! The!University!of!Washington!will!serve!as!the!over!data!coordination!center!(DCC)!for!the!project!that! will!take!place!at!four!performance!sites:!!Group!Health!Cooperative,!Kaiser!Permanente!of!Northern! California,!Henry!Ford!Health!Systems,!and!Mayo!Clinic!Health!Systems.!!The!role!of!the!DCC!is!to! coordinate!study!efforts,!thus!overseeing!the!technical!implementation!of!the!intervention!across!the! sites.!!The!DCC!will!oversee!the!transfer!and!storage!of!study!data,!provide!biostatistical!and!analysis! expertise,!as!well!as!lead!manuscript!writing!efforts.!!Each!performance!site!is!tasked!with!the! implementation!of!the!randomized!intervention!at!the!primary!care!clinics!within!their!system,!as!well!as! the!technical!abstraction!(and!transfer)!of!electronic!medical!record!(EMR)!and!administrative!data!from! their!system!to!the!DCC.!!See!Schemas!1!and!2!below!for!organizational!overview.! 2013409426/kj! 4! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! 1.1!Project!Organization:!! Schema!1:!Overall!Organization!Chart! Schema!2:!Site!Organization!Chart! 2013409426/kj! 5! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Personnel!Directory! University!of!Washington ! Data!Coordination!Center! Jeffrey!G.!Jarvik!MD,!MPH! Principal'Investigator' Professor,!Radiology!and!Neurological!Surgery! Adjunct!Professor,!Health!Services! Director,!Comparative!Effectiveness,!Cost!and!Outcomes!Research!Center!(CECORC)! University!of!Washington! Box!359455! 4333!Brooklyn!Ave!NE! Seattle,!WA!98195! Tel:!(206)!616\2941! ! Fax:!(206)!616\3135! Email:!jarvikj@uw.edu! Patrick!Heagerty!PhD! CoMInvestigator' Professor,!Biostatistics! Director,!Center!for!Biomedical!Statistics! University!of!Washington! Box!357232! Seattle,!WA!!98104! Tel:!(206)!616\2720! ! ! ! ! Fax:!(206)!543\3286! Email:!heagerty@uw.edu! Judith!Turner!PhD! CoMInvestigator' Professor,!Psychiatry!and!Behavioral!Sciences,!Rehabilitation!Medicine! Adjunct!Professor,!Anesthesiology!and!Pain!Medicine! University!of!Washington! Box!356560! Seattle,!WA!!98195! Tel:!(206)!543\3997! ! Fax:!(206)!685\1139! Email:!jturner@uw.edu! Brian!Bresnahan!PhD! Health'Economist' Research!Assistant!Professor,!Radiology! University!of!Washington! Box!359736! 2013409426/kj! 6! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! 325!Ninth!Ave.! Seattle,!WA!!98104! Tel:!(206)!744\1805! ! ! ! ! Fax:!(206)!744\9962! Email:!bres@uw.edu! Bryan!Comstock,!MS! Biostatistician' Operations!Director,!Center!for!Biomedical!Statistics! University!of!Washington! Box!359461! Seattle,!WA!!98195! Tel:!(206)!543\1882! ! ! Fax:!(206)!543\5881! Email:!bac4@uw.edu! Kathryn!James!PAWC,!MPH! Overall'Project'Director' Administrative!&!Operations!Director,!Comparative!Effectiveness,!Cost!&!Outcomes!Research! Center!(CECORC)! Box!359455! 4333!Brooklyn!Ave!NE! Seattle,!WA!98195! Tel:!(206)!221\7444! ! Fax:!(206)!616\3135! Email:!jarvikj@uw.edu! UW!Grant!Contact :!My\Hanh!Tong!\!mhtt@uw.edu! Oregon!Health!Sciences!University! Subcontracted!Site!for!Consultant! Richard!A.!Deyo!MD,!MPH! CoMInvestigator' Kaiser!Permanente!Professor!of!Evidence\Based!Family!Medicine! Director,!KL2!Multidisciplinary!Clinical!Research!Career!Development!Program! Director,!OCTRI!Community!and!Practice\based!Research!Program! Departments!of!Family!Medicine!and!Internal!Medicine! Oregon!Health!and!Science!University! 3181!SW!Sam!Jackson!Park!Rd.! Portland,!Oregon!97239\3098! Tel:!(503)!494\1694! ! ! ! ! Fax:!(503)!494\2746! Email:!deyor@ohsu.edu! OHSU!Grant!Contact:!Connie!YuW!yuco@ohsu.edu! 2013409426/kj! 7! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! University!of!CaliforniaW!Davis! Subcontracted!Site!for!Consultant! ! Nicolas!Anderson,!PhD! ! CoMInvestigator' ' Nick!Anderson,!Ph.D.! Robert!D.!Cardiff!Professor!of!Informatics! Director!of!Informatics!Research! Department!of!Pathology!and!Laboratory!Medicine! University!of!California,!Davis! Tel:!(916)!703!6976! ! UC!Davis!Grant!Contract:!Kate!Marie\!kate.marie@ucdmc.ucdavis.edu! Group!Health!Cooperative! Performance!Site! Dan!Cherkin,!PhD! Site'Principal'Investigator' Group!Health!Research!Institute! 1730!Minor!Ave,!Ste!1600! Seattle,!WA!!98101! Email:!!cherkin.d@ghc.org! Heidi!Berthoud!MPH! Project'Manager' Group!Health!Research!Institute! 1730!Minor!Ave,!Ste!1600! Seattle,!WA!!98101' Email:!berthoud.h@ghc.org! GHC!Grant!Contact:!David!Hawkes\!hawkes.d@ghc.org! Henry!Ford!Health!System! Performance!Site! Safwan!Halabi!MD! Site'Principal'Investigator' Associate!Professor,!Radiology! Director,!Imaging!Informatics! 2799!West!Grand!Boulevard!! Detroit,!MI!48202' Email:!safwanh@rad.hfhs.edu! 2013409426/kj! 8! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! David!Nerenz!PhD! Site'CoMInvestigator! Director,!Outcomes!Research! ! Neuroscience!Institute! Department!of!Neurosurgery! Henry!Ford!Hospital! 2799!West!Grand!Boulevard,!K\11,!W\1136! Detroit,!MI!!48202\2689! Tel:!(313)!916\5454! ! ! ! ! Fax:!(313)!874\7137! Email:!dnerenz1@hfhs.org! Brooke!Wessman! Project'Manager' Tel:!(313)!916\0829' Email:!brookew@rad.hfhs.org! HF!Grant!Contact:!Kim!Sadlocha!\ksadloc1@hfhs.org! Kaiser!Permanente!of!Northern!California!! Performance!Site! Andrew!Avins!MD,!MPH!!! Site'Principal'Investigator! Clinical!Professor,!Medicine! Adjunct!Professor,!Epidemiology!and!Biostatistics! Kaiser!Permanente!Division!of!Research! 2000!Broadway! Oakland,!CA!!94612\!2304! Office:!(510)!891\3557! ! ! ! Fax:!(510)!891\3606!! ! ! Cell:!(415)!302\5986! Email:!andrew.avins@ucsf.edu! Luisa!M.!Hamilton! Project'Manager ,!KPNC!Division!of!Research! 2000!Broadway! Oakland,!CA!!94612!\2304! Tel:!(510)!891\3712! ! ! ! Fax:!(510)!891\3802! Email:!luisa.M.Hamilton@kp.org! KPNC!Grant!Contact:!Anna!DelaneyWHeathW!delaney@kp.org! 2013409426/kj! 9! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Mayo!Clinic!Health!System! Performance!Site! David!F.!Kallmes!MD! Site'Principal'Investigator' Professor,!Radiology! Adjunct!Professor,!Neurological!Surgery! 200!First!St.!SW! Rochester,!MN!55905! Tel:!507\266\3350! Email:!kallmes.david@mayo.edu! Administrative!Contact:!! Kimberly!Collins:!!Collins.kimberly@mayo.edu! Patrick!Leutmer!MD! Site'CoMInvestigator' Assistant!Professor,!Radiology! 200!First!St.!SW! Rochester,!MN!55905! Tel:!507\284\2097! Email:!leutmer.patrick@mayo.edu! Jyotishman!Pathak!PhD! Site'IT'Consultant' Associate!Professor,'Biomedical!Statistics!and!Informatics' 200!First!St.!SW! Rochester,!MN!55905! Tel:!507\284\5541! Fax:!507\284\0460' Email:!!pathak.jyotishman@mayo.edu! Administrative!Contact:! Stacy!Tapp:!!tapp.stacy2@mayo.edu! Beth!Connelly! Project'Manager,'' Associate!Clinical!Research!Coordinator,!Department!of!Radiology! Tel:!507\538\3928! Pager:!507\293\4510! Email:!connelly.beth@mayo.edu! 2013409426/kj! 10! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Kristina!Schmidtknecht!! Site'IRB'Contact' Protocol!Development!Coordinator,!!Department!of!Radiology!!! Tel:!507!266\2082!! ! ! ! ! ! Fax:!507\284\8249!!! Pager:!507!538\2477!!! Email:!!!schmidtknecht.kristina@mayo.edu! Mayo!Grant!Contact:!Tracey!AndersonW!anderson.tracey@mayo.edu! 2013409426/kj! 11! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Sponsor!Information:! National!Institutes!of!Health!(NIH)!! Funding!Award:!1UH2AT007766\01! !! Health!Care!Systems!Research!Collaboratory:! UH2/UH3!mechanism!with!UH2!as!a!planning!year!and!UH3!a!separate!award!for!Yr!2\5!awarded!after! competitive!review!based!on!progress!against!UH2!milestones.!! NIH!press!release!regarding!award:!http://nccam.nih.gov/news/2012/092512!! Collaboratory!Website:!!www.nihcollaboratory.org! !! Timeline:! Budget!Period:!!09/30/2012!\!12/31/2013!! Project!Period:!!01/01/2014!\!12/31/2017!! !!! Participating!Institutions:! Data!Coordinating!Center!(DCC)!and!Prime!Awardee:!!! University!of!Washington\!Seattle,!WA!! Principal!Investigator:!Jeffrey!G.!Jarvik,!MD,!MPH!! Performance!Sites:! Group!Health!Cooperative!(GHC)!and!Group!Health!Research!Institute:!!Site!PI:!Dan!Cherkin,!PhD!! Henry!Ford!Health!System!(HFHS)!Site!PI:!Safwan!Halabi,!MD!! Kaiser!Permanente!of!Northern!California!(KPNC)!Site!PI:!Andy!Avins,!MD,!MPH!! Mayo!Clinic!Health!System!(MCHS)!Site!PI:!!David!Kallmes,!MD!! Subcontracted!Sites:! Oregon!Health!Sciences!University!(OHSU):!Rick!Deyo,!MD,!MPH! University!of!California\!Davis!(UCD):!Nicholas!Anderson,!PhD!(subcontract!in!UH2!phase!only,!contractor! in!UH3)! 1.2!LIRE!Interaction!with!Collaboratory!Coordinating!Center!at!Duke!Clinical!Research!Institute:! The!Collaboratory!Coordinating!Center!has!several!“Cores”!aimed!at!organizing!topic\specific!working! groups!across!the!seven!demonstration!projects.!!LIRE!is!contributing!to!the!Collaboratory!by!assigning! key!team!members!to!these!Cores!in!the!following!ways:! ¥ Electronic!health!records\!Anderson,!Jarvik,!Comstock,!and!James! ¥ Provider!Health!Systems!Interactions\!Jarvik!and!James! ¥ Regulatory/Ethics\!James! ¥ Biostatistics/Study!design\!Heagerty!and!Comstock! ¥ Stakeholder!engagement\!Jarvik!and!James! ¥ Pheontype!and!Data!Standards\!Anderson! 2013409426/kj! 12! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! 2.!Background!and!Rationale! Summary'of rationale:!A!common!problem!with!many!diagnostic!tests!is!the!discovery!of!incidental! findings!unrelated!to!patient!symptoms!or!complaints.!Such!findings!can!lead!to!wasteful!subsequent! testing!and!intervention,!sometimes!with!avoidable!complications.!Our!overall!goal!is!to!test!a!strategy! for!mitigating!these!“cascade!effects”!of!incidental!findings.!We!focus!on!the!example!of!lumbar!spine! imaging,!where!incidental!findings!are!extremely! common.!We!propose!a!pragmatic!randomized! trial!of!the!strategy!of!inserting!epidemiological! evidence!into!routine!spine!imaging!reports. If!the!study!is!positive,!the!method!is likely!to!be! generalizable!to!many!other!conditions!and!to! other!kinds!of!testing!(eg,!laboratory!tests).!!So! while!back!pain,!especially!the!back!pain!that! primary!care!providers!see!and!treat,!is!incredibly! important,!our!project!can!also!be!viewed!as!a! "proof!of!concept"!study!that!could!open!the! doors!to!many!similar!interventions.!!Moreover,! the!potential!cost/effectiveness!of!this! intervention,!if!successful,!is!enormous.!!The!cost! of!the!intervention!itself!is!minimal,!yet! substantial!clinical!and!financial!benefits!could! result.!Few!medical!interventions!can!make!that! claim.! Back!pain!is!one!of!the!most!important!causes!of! functional!limitation!and!disability!worldwide!and! is!an!Institute!of!Medicine!priority!condition.(1,2)! It!is!one!of!the!most!common!reasons!for! physician!visits.(3)!The!American!College!of!Physicians!(ACP)!instituted!a!program!in!2011!called!High\ Value,!Cost\Conscious!Health!Care!(HVCCHC).(4)!The!purpose!of!the!program!is!“…!to!help!physicians! and!patients!understand!the!benefits,!harms,!and!costs!of!an!intervention!and!whether!it!provides!good! value,!and!to!slow!the!unsustainable!rate!of!health!care!costs!while!preserving!high\value,!high\quality! care.”!The!importance!of!back!pain!is!highlighted!by!the!first!recommendations!of!the!program!being! focused!on!the!appropriate!use!of!spine!imaging.!(5)!In!April!2012,!the!ACP!in!combination!with!the! ABIM!Foundation!released!their!5!top!“Things!that!Patients!and!Physicians!Should!Question.”!!Number! two!on!the!list!was!“Don’t!obtain!imaging!studies!in!patients!with!non\specific!low!back!pain”!(6).!! Luo!and!colleagues!estimate!that!the!1998!direct!costs!of!low!back!pain!in!the!U.S.!were!over!$26!billion.! More!recently,!Martin!et!al,!estimated!that!the!2005!direct!costs!were!over!$86!billion.!(7)!Diagnostic! imaging!is!a!critical!step!in!the!work\up!of!back!pain.!It!can!quickly!lead!to!a!precise!and!actionable! diagnosis,!such!as!severe!central!spinal!stenosis!with!cauda!equina!compression!that!may!require!rapid! surgical!consultation.!But!imaging!examinations!of!the!lumbar!spine!frequently!reveal!numerous!findings,! including!disk!desiccation,!height!loss,!or!bulging,!with!questionable!relevance!to!patient!symptoms.! 2013409426/kj! 13! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Figure!1,!from!our!study,!Longitudinal!Assessment!of!Imaging!and!Disability!of!the!Back!(LAIDBack)!in! 2001!demonstrates!a!lumbar!spine!MR!without!degenerative!changes!(Figure!1A)!and!a!subject!with! extensive!degenerative!changes!(Figure!1B).!Neither!of!these!subjects!had!low!back!pain.!(8)!These! findings!are!common!in!asymptomatic!adults,!with!prevalences!in!this!group!as!high!as!90%.!(8\10)! Moreover,!multiple!studies!have!failed!to!demonstrate!clinical!benefit!with!the!use!of!early!MR!imaging! for!low!back!pain!(LBP)!compared!with!radiographs!alone!or!no!imaging!at!all;!furthermore,!the!imaging! results!may!negatively!affect!patients’!sense!of!well\being!(7,13,14).!But!diagnostic!imaging!of!the! lumbar!spine!can!also!lead!to!a!cascade!of!subsequent!tests!and!treatments!that!may!have!little! beneficial!impact!on!a!patient’s!outcome!and!may!even!be!deleterious.!(11,!12)!Because!incidental! findings!are!nearly!ubiquitous!with!spine!imaging,!it!is!important!to!have!a!good!understanding!of!the! prevalence!of!various!findings!in!asymptomatic!patients.!! While!spine!imaging!may!be!one!of!the!most!common!examples!of!incidental!findings!on!diagnostic! testing!resulting!in!a!cascade!of!subsequent!tests!and!treatments,!this!situation!is!by!no!means!limited!to! spinal!diagnosis.!Lung!cancer!screening!with!CT!was!recently!shown!to!be!beneficial!in!a!high\risk! population,!but!one!of!the!concerns!with!such!screening!are!the!frequent!benign!nodules!that!are! discovered.!(13)!Adrenal!nodules!seen!on!body!CTs!(14),!thyroid!nodules!seen!on!neck!and!chest!CTs!(14),! sinus!mucosal!thickening!seen!on!head!MR!(15)!and!CTs!(16)!could!all!lead!to!subsequent!diagnostic!and! therapeutic!interventions!if!their!prevalence!in!patients!without!disease!was!not!well\understood.! While!spine!specialists!are!well!aware!of!these!prevalence!data,!non\specialists!such!as!family! practitioners!and!general!internists!may!not!know!that!a!finding!such!as!an!annular!fissure!is!seen!in! about!one\third!of!asymptomatic!patients!and!if!present,!is!likely!not!related!to!a!patient’s!pain.! Several!years!ago!our!group!at!the!University!of!Washington!implemented!into!our!clinical!practice,!the! recommendation!of!Roland!and!van!Tulder!(17)!to!include!epidemiologic!information!in!the!radiology! report!to!help!physicians!interpret!findings!frequently!seen!on!lumbar!spine!imaging!(Figure!2).!By! providing!a!context!for!these!common!findings,!we!hoped!to!mitigate concern!and!dampen!any! subsequent!cascade!of!inappropriate!testing!and!treatment.! ! FIGURE 2: EPIDEMIOLOGIC STATEMENT INCLUDED IN LUMBAR SPINE MR IMAGING REPORTS Multiple!randomized!controlled!trials! have!shown!that!the!early!use!of!imaging! for!LBP!is!not!associated!with!improved! outcomes!and!may!be!harmful!to!the! patient!(11,!18\23).!The!American!College! of!Physicians!recently!re\issued!guidelines! for!imaging!patients!with!LBP!emphasizing! not!only!the!inefficiencies!of!early!imaging! but!also!the!potential!harms!(24).! Furthermore,!as!rates!of!MR!imaging!of! the!lumbar!spine!have!increased,!so!too! have!treatments;!including!narcotics! prescriptions,!lumbosacral!injections,!and! spinal!surgery,!often!without!benefit!(25\32).!Not!only!do!these!treatments!result!in!increased! expenditures!(7,!32,!33),!but,!more!importantly,!they!pose!serious!risks!to!the!health!of!the!patient.! Narcotics!are!associated!with!multiple!side!effects,!including!respiratory!depression,!cognitive! 2013409426/kj! 14! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! impairment,!constipation,!and!even!death,!as!well!as!the!development!of!tolerance!and!dependency!(34,! 35).!Complications!from!spinal!surgeries,!especially!more!invasive!fusions,!include!wound!complications,! major!medical!complications,!and!death!(32).!! In!2012!our!group!published!a!pilot!study!demonstrating!insertion!of!such!epidemiological!evidence!was! associated!with!reduced!narcotic!prescriptions!and!a!non\significant!reductions!in!subsequent!MR,!CT! and!physical!therapy!as!well.!Taken!together,!these!findings!suggest!primary!care!providers!were!more! reserved!in!their!management!of!patients!whose!MR!report!included!the!epidemiological!evidence! statement.!(Table!1:!from!McCoullough!et!al,!2012)!(36)!Additionally,!if!patients!learned!about!the! statement,!as!some!undoubtedly!did!since!patients!at!the!study!site!have!direct!access!to!their!medical! records,!knowledge!that!their!spine!findings!are!common!in!patients!without!back!pain!might!alleviate! anxiety,!which!is!known!to!have!an!important!influence!on!pain.!(37)! Table 1: Outcomes of Patients Whose Imaging Did and Did Not Include a Statement Containing Epidemiological Benchmarks (from McCoullough et al, 2012) (36) The!relatively!new!field!of!clinical!genomics!is!on!the!verge!of!a!virtual!explosion!of!genetic!tests!that!will! be!inexpensive!and!readily!available.!(38,!39)!!However,!genetic!testing!faces!the!same!challenges!of! communicating!risk!information!that!more!traditional!diagnostic!testing!has!faced!for!decades.!(39)! Lessons!learned!from!diagnostic!imaging!may!be!applied!to!genomic!testing!and!vice\versa.!! Because!our!intervention!is!simple,!inexpensive!and!can!be!automated,!it!is!easy!to!implement!on!a!large! scale,!making!it!nearly!ideal!to!study!in!the!context!of!a!large,!pragmatic!trial!in!multiple!health!systems.! We!decided!to!confine!our!participating!sites!to!large!health!systems!that!have!sophisticated!electronic! medical!records!allowing!us!to!passively!collect!our!outcomes!through!electronic!queries.!! Our!method!of!random!assignment!is!also!relatively!novel.!We!propose!to!use!a!stepped!wedge!cluster! design,!where!the!order!in!which!clinics!receive!the!intervention!is!determined!at!random!and!by!the! end!of!the!random!allocation,!all!clinics!will!have!received!the!intervention.!(40,!41)! Finally,!given!the!rapid!spread!and!adoption!of!IT!clinical!tools,!like!the!EMR!and!templates!for!radiology! readings,!finding!ways!to!capitalize!on!the!technology!itself!to!positively!influence!the!process!of!care! will!make!the!mammoth!nationwide!clinical!IT!investment!much!more!compelling.!This!project!is!truly! 2013409426/kj! 15! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! emblematic!of!the!kinds!of!innovative!thinking!that!needs!to!be!applied!to!the!clinical!IT!world!to!derive! the!maximum!benefit!of!the!tools!meant!to!deliver!better!and!more!efficient!care.! 3.!Specific!Aims! This!study!is!a!pragmatic!cluster!randomized!controlled!trial,!randomly!assigning!primary!care!clinics!at! four!sites,!to!receive!either!standard!lumbar!spine!imaging!reports!or!reports!containing!epidemiological! benchmarks!for!common!imaging!findings.!Our!primary!outcome!will!be!a!metric!of!back\related! intervention!intensity,!measured!passively!using!the!electronic!medical!record!(EMR).!The!primary! analysis!will!focus!on!clinic\level!changes!by!using!aggregate!patient\level!data.! Aim!1:!To!determine!whether!inserting!a!description!of!age\specific!prevalence!of!imaging!findings! among!asymptomatic!subjects!into!lumbar!spine!imaging!reports!decreases!back\related!interventions! !imaging,!injections,!surgeries,!etc.)!over!the!subsequent!year.! Aim!1a:!To!determine!if!inserting!epidemiological!evidence!reduces!Relative!Value!Units!(RVUs)! attributable!to!spine!interventions!(imaging,!injections,!specialist!referrals,!surgeries,!etc.).! Hypothesis!1a:!After!primary!care!clinics!are!randomly!assigned!to!receive!the!modified!report,!they!will! have!a!lower!average!overall!RVU!(technical!and!professional)!per!imaged\patient!attributable!to!spine! interventions!than!when!clinics!are!not!receiving!the!modified!reports.!Spine!interventions!reflect!visits,! tests,!and!procedures!and!are!patient!centered,!having!both!direct!and!indirect!impacts!on!patients.! Aim!1b:!To!determine!if!inserting!epidemiological!data!decreases!opioid!prescriptions.! Hypothesis!1b:!Time!periods!during!which!clinics!are!randomly!assigned!to!receive!the!modified!imaging! reports!will!have!a!lower!rate!of!subsequent!opioid!prescriptions!than!time!periods!during!which!clinics! do!not!receive!modified!reports.! Aim!1c:!To!determine!if!inserting!epidemiological!evidence!decreases!subsequent!cross\sectional! imaging!!magnetic!resonance!(MR)!and!computed!tomography!(CT).! Hypothesis!1c:!Time!periods!during!which!randomly!assigned!clinics!receive!modified!imaging!reports! will!have!a!lower!rate!of!subsequent!cross\sectional!imaging!than!time!periods!for!which!clinics!do!not.! Aim!1d:!To!explore!whether!adding!epidemiological!evidence!decreases!overall!costs!of!care!for!low! back!pain!based!on!CMS!reimbursement.! Hypothesis!1d:!Clinics!that!are!randomly!assigned!to!receive!the!modified!imaging!reports!will!have! lower!back!pain\related!estimated!payer!costs!than!clinics!whose!patients!do!not!receive!modified! reports.!Costs!are!another!outcome!that!are!highly!relevant!to!both!patients!and!health!systems.! Aim!2:!To!determine!whether!inserting!age\specific!prevalence!of!imaging!findings!in!asymptomatic! subjects!has!a!differential!effect!on!subsequent!back\related!interventions!if!inserted!into!lumbar!spine! MR!and!CT!imaging!reports!compared!with!plain!films.! 2013409426/kj! 16! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Hypothesis!2:!Inserting!epidemiological!information!into!plain!film!reports!will!result!in!a!greater! decrease!in!subsequent!back\related!interventions!than!similar!information!put!into!MR!and!CT!reports.! Given!that!plain!films!are!generally!obtained!earlier!in!the!course!of!back!pain!and!more!frequently!than! MR!and!CT,!the!potential!impact!of!inserting!epidemiological!information!into!plain!film!reports!is!large.! Aim!3:!To!determine!if!specific!imaging!findings!influence!subsequent!interventions.! Hypothesis!3:!Inserting!the!statement!will!result!in!a!greater!decrease!in!subsequent!interventions!for! patients!without!clinically!important!findings!compared!with!patients!who!have!clinically!important! imaging!findings.!Our!work!and!others!have!shown!that!certain!imaging!findings!are!likely!to!be!clinically! more!important!than!others!(e.g.!nerve!root!compression,!moderate!to!severe!central!stenosis,!disc! extrusions).!We!expect!that!patients!without!these!more!important!findings!will!have!a!greater! reduction!in!subsequent!interventions.! 4.!Study!Details! 4.1!Eligibility!Criteria! Because!this!is!a!pragmatic!trial,!we!have!minimized!eligibility!restrictions,!making!the!inclusion!criteria! as!broad!as!possible.!Clinics!will!be!the!primary!unit!of!randomization!and!analysis,!while!the! intervention!will!be!applied!at!the!individual!patient!level.!Thus!two!sets!of!eligibility!criteria!are! necessary:!clinic!and!patient.! The!criteria!for!clinic'eligibility!are!that!the!health!care!providers!are!a!distinct,!readily!identifiable!group! that!has!at!least!a!subgroup!of!primary!care!providers!who!do!not!practice!at!another!clinic!that!will!also! be!part!of!the!trial.!This!requirement!of!being!based!primarily!at!one!site!is!to!minimize!cross\ contamination!(having!the!use!of!epidemiological!benchmarks!at!one!site!influence!another!site!not! receiving!the!benchmarks).! The!criteria!for!patient'eligibility!are!that!they!have!had!an!imaging!study!of!the!lumbar!spine!requested! by!a!primary!care!provider.!We!will!include!all!conventional!lumbar!spine!imaging!(plain!films,!CT,!MR)! ordered!by!primary!caregivers.!! 4.2!Consent!procedure! Because!the!intervention!will!be!administered!at!the!clinic!level,!consent!of!either!individual!patients!or! primary!caregivers!is!neither!feasible!nor!warranted.!Moreover,!the!intervention!is!relatively!benign!(the! insertion!of!additional!epidemiological!information!into!the!radiology!report)!and!poses!minimal!risk!to! caregivers!and!patients.!The!performance!sites!are!enthusiastic!about!incorporating!the!epidemiological! benchmarks!into!their!reports!and!may!well!eventually!adopt!them!regardless!of!the!project,!our!study! simply!allows!for!systematic!study!of!the!effects!of!a!well\controlled!implementation!of!the!insertion!of! the!benchmark!information.!The!randomization!scheme!defines!when!each!clinic!begins!including!the! epidemiological!information!into!the!reports,!with!all!sites!eventually!receiving!the!intervention!of! interest.! 4.3!Inclusion!and!Exclusion!Criteria! We!will!define!a!clinic!as!a!primary!care!clinic!if!a!majority!of!the!practitioners!at!that!clinic!are!providing! 2013409426/kj! 17! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! primary!care.!We!will!include!general!internal!medicine!and!family!practice!physicians!as!primary! caregivers!as!well!as!mid\level!providers!working!with!physicians!such!as!nurse!practitioners!and! physician!assistants.! We!will!include!all!adult!patients!of!eligible!caregivers!who!have!had!a!lumbar!spine!imaging!study!plain! film,!CT!or!MR)!ordered!by!their!primary!care!practitioner.! 5.!Research!Design!and!Methods! 5.1!Clinic/Practitioner/Patient!Identification! The!site!PI!will!identify!eligible!clinics!within!their!health!system,!working!closely!with!their! administrative!and!information!technology!(IT)!staff!to!assure!complete!inclusion!of!primary!care!clinics.! The!site!PIs!will!then!categorize!practitioners!within!each!clinic!by!specialty,!designating!general! internists,!family!practitioners!and!obstetrician/gynecologists!as!primary!care!practitioners.!Mid\level! providers!(e.g.!nurse!practitioners!and!physician!assistants)!working!as!primary!caregivers!will!also!be! classified!as!primary!care!practitioners.!The!health!information!system!will!be!used!to!automatically! identify!when!a!practitioner!from!a!particular!clinic!orders!a!lumbar!spine!imaging!study.! 5.2!Randomization! At!each!site!we!will!identify!the!settings!where!primary!care!is!delivered!and!designate!an!appropriate! unit!that!will!constitute!a!functional!“clinic”!for!randomization!and!analysis.!We!will!randomly!assign!all! predetermined!clinics!at!each!site!to!receive!the!intervention!at!one!of!five!fixed!time\points,!rolling! interventions!out!every!six!months!beginning!at!the!start!of!the!second!quarter!of!Year!2.!Using!cutoffs! determined!in!the!UH2!project!phase,!we!will!sort!clinics!by!number!of!primary!care!providers!into! tertiles!(e.g.!small,!medium,!large!clinics).!From!each!tertile!we!will!randomly!select!clinics!using!urn\ based!randomization!(without!replacement)!stratified!by!site!and!clinic!size!such!that!clinics!of!small,! medium,!and!large!size!are!equally!represented!in!each!randomization!wave.!!For!more!details!regarding! the!Analysis!plan,!please!refer!to!the!UH3!transition!request!proposal!and!accompanying!Appendix!10,! Analysis!plan!that!reflects!modifications!made!to!the!original!plan!we!outlined!for!UH2.! Figure!3:!Proposed!Randomization!Schedule! 2013409426/kj! 18! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! 5.3!Clinic/Patient!Enrollment! Using!the!site!administrative!data!systems,!we!will!identify!all!primary!care!providers!(PCP)!at!a!given! clinic.!When!an!identified!PCP!from!a!randomized!clinic!submits!a!request!for!a!lumbar!spine!imaging! study,!the!report!will!automatically!be!flagged.!The!PCP’s!name!will!be!cross\referenced!with!the! randomization!assignment!and!those!PCPs!who!work!in!clinics!assigned!to!receive!the!intervention!will! have!the!epidemiological!benchmark!information!automatically!inserted!into!their!imaging!reports.! Those!PCPs!who!work!in!clinics!not!yet!scheduled!to!receive!the!benchmark!information!will!get!the! usual!imaging!report!issued!by!their!radiologists.!Since!the!intervention!will!be!applied!at!the!PCP!and! clinic!level,!all!patients!receiving!lumbar!spine!imaging!studies!at!those!clinics!will!be!part!of!the!trial.! The!lumbar!spine!imaging!studies!that!we!plan!to!include!in!the!trial!are!plain!films,!magnetic!resonance!! (MR)!imaging!examinations!and!computerized!tomography!(CT).!Table!2!lists!the!proposed!CPT!codes! that!we!will!flag!for!inclusion.!We!are!currently!not!planning!on!including!nuclear!medicine!studies!(e.g.! bone!scans,!both!planar!and!SPECT)!both!because!they!are!infrequently!ordered!by!primary!care! clinicians!as!well!as!because!there!is!inadequate!benchmarking!information.! Table!2:!Lumbar!Examinations!to!be!Included!in!Pragmatic!Trial! CPT!Code! Examination!Description!!!!!!!!!!!!!!!!!!!!!!!!!!!! 72080!!!!!!!!!!!!THORACOLUMBAR!SPINE,!2!views! 72100! LUMBAR!SPINE!2!VIEWS! 72110! LUMBAR!SPINE!3\4!VIEWS! 72114! LUMBAR!SPINE!5!VIEWS! 72131! CT!L!SPINE!W/O!CONTRAST! 72132! CT!L!SPINE!W/!CONTRAST! 72133! CT!L!SPINE!W/O!&!W/!CONTRAST! 72148! MRI!LUMBAR!SPINE!W/O!CONTRAST! 72149! MRI!LUMBAR!SPINE!W/!CONTRAST! 72158! MRI!L!SPINE!W/!&!W/O!CONTRAST! 5.4!Data!collection! We!will!collect!all!baseline!and!follow\up!data!from!the!electronic!information!systems!which,! depending!on!the!site,!will!include!both!the!electronic!medical!record!(EMR)!as!well!as!administrative! data!systems.! Baseline'Data'Collection:!We!will!include!all!patients!receiving!lumbar!spine!imaging!studies!(plain!films,! MR!and!CT)!in!the!last!quarter!of!Year!1!and!the!first!quarter!of!Year!2!as!a!part!of!a!baseline!accrual! period!to!establish!baseline!parameters!for!the!primary!care!physicians!in!participating!clinics.!Since!the! randomization!will!occur!at!the!clinic!level,!the!baseline!data!will!reflect!clinic!level!ordering!patterns!of! diagnostic!and!therapeutic!interventions.! FollowMup'Data'Collection:!We!will!capture!EMR!data!on!patients!for!a!minimum!of!one!and!up!to!two! years!after!the!index!imaging!test.!All!patients!will!have!a!minimum!of!one\year!follow\up.!Eighty! percent!of!patients!will!have!two\year!follow\up!due!to!the!staggered!implementation!of!the! 2013409426/kj! 19! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! intervention.!The!six\month!length!of!each!patient!accrual!period!is!sufficiently!long!to!account!for! temporal!or!system\level!trends!in!the!measured!outcomes!over!the!course!of!the!study.!! 5.5!Aims!for!UH2!Phase! Our!goal!is!to!use!the!planning!UH2!phase!of!the!grant!to!accomplish!the!following:!First,!we!will!refine! the!epidemiological!benchmarks!that!we!will!insert!into!the!radiology!report.!Second,!we!will!develop! and!test!our!site\specific!deployment!method!for!the!cluster!randomization.!Third,!we!will!develop!a! metric!that!reflects!the!intensity!of!interventions!for!back!pain\related!care!and!develop!CMS\based! standardized!cost!estimates!associated!with!resource!use!intensity!that!can!be!applied!uniformly!among! health!systems.!We!will!validate!this!metric!using!data!from!the!health!care!systems!electronic!medical! record.!Fourth,!we!will!develop!and!validate!our!methods!for!extracting!outcome!data!from!the! electronic!medical!record.!Fifth,!once!we!have!defined!the!above,!we!will!obtain!Institutional!Review! Board!approval!for!the!implementation!phase!of!the!study.!We!will!also!use!this!time!to!assemble! subcontracts!for!administrative!review!at!each!site.! Aim!1:!Refine!the!information!to!be!included!in!the!radiology!report!so!that!it!is!specific!for!imaging! modality!and!patient!age.!! In!our!original!implementation,!we!only!inserted!the!epidemiological!benchmarks!into!reports!of!lumbar! spine!MRs!whereas!in!the!current!project!we!propose!to!insert!the!information!into!reports!of!MR,!CT! and!plain!films.!Moreover,!we!used!epidemiological!data!from!a!single!study!published!by!our!group! (Table!3).(8)!Other!groups!have!published!similar!data!for!MR!as!well!as!other!modalities.!(9,!10,!42\59)! In!addition!to!updating!the!epidemiological!benchmarks!and!expanding!them!to!other!modalities,!we! will!also!gather!data!regarding!age\specific!rates!for!various!imaging!findings.!While!eventually!we!would! envision!a!decision!support!tool!that!could!recognize!specific!patient!attributes,!such!as!age!or!the! presence!of!a!particular!finding,!and!insert!customized!benchmark!data!for!that!individual,!such!a!system! is!beyond!the!scope!of!this!project.!Instead!we!plan!to!insert!benchmarks!that!are!simply!stratified!by! age!ranges.!We!will!perform!a!systematic!review!of!the!literature!so!that!we!are!inserting!the!most! recent!and!complete!epidemiological!evidence!into!the!radiology!report.! Table!3:!Age\specific!Rates!of!Lumbar!Spine!Imaging!Findings!(from!Jarvik!et!al,!2001)! 2013409426/kj! 20! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Aim!2:!Develop!siteWspecific!deployment!methods!for!the!stepped!wedge,!cluster!randomization! scheme.!! The!unit!of! Figure!4:!Comparison!of!Parallel,!Crossover!and!Stepped!Wedge! randomization!for! Designs!(from!Hussey!et!al.,!2007)! this!project!will!be!at! the!level!of!the!clinic.! The!stepped!wedge! design!is!a!one\way! cluster,!crossover! design!that! temporally!spaces!the! intervention!and! assures!that!each! participating!clinic! will!eventually! receive!the! intervention.!Figure!4,!from!Hussey!et!al,!(41)!compares!the!stepped!design!with!parallel!and!crossover! designs:! Our!experience!with!the!BOLD!project!and!other!multicenter!studies!informs!us!that!the!procedures!and! hurdles!at!each!site!will!be!different!so!that!our!approach!for!implementing!the!cluster!randomization! must!be!customized!at!least!to!some!extent.!In!the!UH2!phase!of!the!project,!we!will!work!closely!with! the!informatics!groups!at!each!health!system!to!develop!the!schedule!of!clinics!to!be!randomized!to! receive!the!insertion!of!epidemiological!data!into!the!relevant!reports.!A!key!component!of!deploying! the!intervention!is!that!it!be!automated!and!not!require!a!radiologist!to!actively!insert!benchmark! statements!into!the!report.!We!also!plan!to!create!a!process!for!notifying!sites!of!intervention! deployment,!being!careful!to!minimize!opportunities!for!internal!or!external!sources!of!contamination.! Aim!3:!Develop!and!validate!a!composite!measure!of!spine!intervention!intensity!that!combines!into!a! single!metric!the!overall!intensity!of!resource!utilization!for!back!pain!care.!! Relative!value!units!(RVUs)!are!a!measure!of!work!effort!associated!with!a!particular!medical!service.! Although!there!are!potential!drawbacks!to!RVUs,!including!overvaluing!certain!services!relative!to!others,! it!is!a!widely!used!metric!and!one!potential!method!for!measuring!the!intensity!of!services!provided!(or! resources!utilized)!for!back!pain!treatment.!During!the!planning!year,!our!group!would!need!to! specifically!identify!services!that!would!comprise!the!back!care!intensity!metric.!! Once!defined,!we!could!then!attempt!to!validate!the!metric!using!the!World!Health!Organization’s! stepped!care!approach!to!pain!treatment,!which!amounts!to!an!escalating!ladder!of!treatment! intensiveness.!! Aim!4:!Develop!and!validate!electronic!data!methods!and!tools!to!capture!the!outcomes!of!interest! (subsequent!diagnostic!testing,!opioid!prescriptions,!spinal!injections,!spine!surgeries).!! 2013409426/kj! 21! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! While!all!of!the!health!care!systems!that!we!propose!to!involve!in!our!project!have!sophisticated!and! comprehensive!electronic!medical!records!systems,!we!know!from!experience!that!accessing!the! relevant!data!and!transmitting!it!to!the!project’s!data!coordinating!center!will!require!careful!planning! and!individualized!approaches!for!each!health!system.!We!have!already!successfully!implemented! methods!for!collecting!some!of!this!information!from!2!of!our!proposed!sites!(Kaiser!Northern!California! and!Henry!Ford!Health!System,!Detroit,!MI)!and!would!refine!and!deploy!these!methods!at!two!new! sites,!Group!Health!Cooperative!in!Seattle,!WA!and!the!Mayo!Clinic!Health!System!in!Minnesota!and! Wisconsin.! 5.6!Working!Groups ! The!planning!UH2!phase!of!the!grant!will!be!used!to!accomplish!the!following:!First,!we!will!refine!the! epidemiological!benchmarks!that!we!will!insert!into!the!radiology!report.!Second,!we!will!develop!and! test!our!site\specific!deployment!method!for!the!cluster!randomization.!Third,!we!will!develop!a!metric! that!reflects!the!intensity!of!interventions!for!back!pain\related!care!and!develop!Medicare\based! standardized!cost!estimates!associated!with!resource!use!intensity!that!can!be!applied!uniformly!among! health!systems.!We!will!validate!this!metric!using!data!from!BOLD!and!Medicare.!Fourth,!we!will!develop! and!validate!our!methods!for!extracting!outcome!data!from!the!electronic!medical!record.! The!scope!of!work!for!the!UH2!planning!year!has!been!outlined!and!assigned!to!one!of!four!working! groups!that!align!with!the!aims!described!above.! !! Working!Group!1 !will!focus!on!refining!the!information!to!insert!into!the!radiological!report.!They!will! lead!a!critical!review!of!the!literature!focusing!on!the!age\specific!prevalence!of!common!findings!seen! on!plain!films,!CT!and!MR!in!people!without!low!back!pain.!We!will!use!meta\analytic!methods!to! combine!the!prevalence!estimates!from!multiple!sources,!weighting!by!study!quality!and!relevance!to! the!LIRE!population.!We!will!summarize!the!epidemiological!information!so!that!it!can!be!inserted!into! plain!film,!MR!and!CT!reports!of!lumbar!spine!imaging.!Dr.!Jarvik!will!lead!this!working!group!and!work! closely!with!Dr.!Kallmes!at!the!Mayo!Clinic!Health!System.! Working!Group!1!progress :!!Members!of!this!working!group!developed!and!pilot!tested!the!intervention! text!and!took!several!steps!in!this!process.! 1. Comprehensive!literature!review!for!relevant!articles!regarding!radiologic!findings! !2.!!Two!independent!reviews!and!data!extraction!of!findings!from!relevant!articles.!See!! Appendices!1!and!2!for!the!Abstraction!form!used!and!list!of!final!articles!contributing!data!to!the! prevalence!rates!included!in!the!intervention!text.!! !3.!!!Data!cleaning!and!compilation! !4.!!!Data!analysis!and!modeling!to!determine!a)!which!findings!had!sufficient!data!to!report!on! and!b)!age!cut!points!! !7.!!!Consult!with!the!Program!for!Readability!In!Science!&!Medicine!(PRISM)!scientific!writing! group!at!Group!Health.!!! !8.!!!Key!informant!interviews!with!two!patient!advisors!for!feedback!on!format!and!readability!! !9.!!!On\line!survey!with!over!20!patient!advisors!for!feedback!and!comment! Participating!patient!representatives!were!provided!with!a!sample!radiology!report!and!four! different!versions!of!the!intervention!macro.!!They!were!asked!the!following!questions:! 2013409426/kj! 22! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! ¥ What!is!your!age?! ¥ Have!you!ever!sought!medical!care!for!low!back!pain?! ¥ Which%format%option%did%you%prefer?! ¥ Why!did#you#prefer#that#format#option#over#the#others?#! ¥ Based&on&this&information,&would&you&say&disc&degeneration&is&common&or&rare?! ¥ What%is%clear!about&the&information&presented?&! ¥ What%is%confusing)about&the&information&presented?&! ¥ How$would$you$use$the$information(presented(in(the(option(that(you(preferred?(! ¥ If#you#received#a#radiology#report#with#the#information#in#the#option#that#you#preferred,# would&you&feel&less&or&more&concerned&about&having&the&imaging&finding(s)&that&are& common%in%people%without%back!pain?&!For$example,$would$you$feel$less$or$more$ concerned'about'having'a'degenerated'disk?! !10.!!On\line!survey!with!seven!primary!care!physicians!for!feedback!and!comment.!!Participating! providers!were!provided!with!a!sample!radiology!report!and!three!different!versions!of!the! intervention!macro.!!They!were!asked!the!following!questions:! ¥ What!proportion!of!your!patient!visits!are!related!to!low!back!pain?!!! ¥ Which%format%option%do%you%prefer?! ¥ Why!did#you#prefer#that#format#option#over#the#others?#! ¥ Based&on!this%information,%would%you%say%disc%degeneration%is%common%or%rare?! ¥ What%could%be%improved%about%your!preferred&option?&! ¥ If#you#received#a#radiology#report#with#the#information#in#the#option#that#you#preferred,# would&you&be&less&or&more&concerned&about&a!given&patient's&imaging&finding?&!For$ example,(would(you(feel(less(or(more(concerned(about(a(patient(having(a(degenerated( disk?! ¥ How$would$the$information$presented$in$the$option$that$you$preferred$inform$your$ clinical&decision&making?! The!final!text!is!a!product!of!the!efforts!described!above!and!will!serve!as!the!final!intervention.!! Different!wording!is!offered!based!on!1)!Age!(<40,!40\60,!and!>60!years!of!age)!and!2)!Modality! (plain!film,!CT,!MR).! Working!Group!2 !will!focus!on!methods!to!practically!deploy!the!stepped!wedge!cluster!randomization! scheme.!This!will!require!working!closely!with!the!informatics!group!at!each!of!the!four!sites!to! determine!the!optimal!method!for!inserting!the!intervention!text!into!their!reports.!This!might!occur!at! the!level!of!the!radiology!reporting!software!(RRS),!the!radiology!information!system!(RIS)!or!the! hospital!information!system!(HIS).!Sites!must!demonstrate!the!ability!to!selectively!insert!the! intervention!text!only!into!reports!where!the!clinic!is!randomly!allocated!to!receive!the!intervention.!Dr.! Heagerty!and!Mr.!Comstock!will!lead!this!working!group!out!the!Center!for!Biomedical!Statistics!(CBS)! that!is!part!of!the!Institute!for!Translational!Health!Sciences!(ITHS),!the!University!of!Washington’s! Clinical!and!Translational!Science!Award!(CTSA).!Dr.!Heagerty!is!Professor!and!Associate!Chair!of! Biostatistics!and!the!Director!of!the!CBS.! Working!Group!2! progress:!!A!site\readiness!tool!has!been!developed!to!guide!efforts!for!pilot!testing! the!insertion!of!the!intervention!macro!across!the!clinics!at!a!given!site!(see!Appendix!3!for!an!example! of!this!pilot!testing!document!and!site!checklist).!!At!the!conclusion!of!this!pilot,!the!number!and!size!of! 2013409426/kj! 23! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! participating!clinics!at!each!site!will!be!verified!as!well!a!final!determination!regarding!technical! feasibility!of!clinic!randomization!and!intervention!text!insertion.! Working!Group!3 !will!develop!and!validate!the!composite!measure!of!spine!intervention!intensity,!likely! to!be!based!on!units!and!types!of!services!used.!We!will!validate!this!metric!using!data!from!a!variety!of! sources!including!the!Back!pain!Outcomes!using!Longitudinal!Data!(BOLD)!registry.!This!will!be!an!on\ going!effort!through!much!of!the!planning!year!as!we!review!the!literature!and!obtain!expert!opinion!to! optimize!the!factors!comprising!the!composite!measure,!and!validate!the!measure.!Drs.!Deyo!and! Bresnahan!will!co\lead!this!effort.!Working!Group!3!will!also!develop!cost!estimates!to!apply!to!overall! spine!intervention!intensity!and!to!individual!services.! Working!Group!3!progress: !!A!comprehensive!review!of!the!literature!was!conducted!searching!for! articles!pertaining!to!RVU\related!assessment!(see!Appendix!4!for!a!complete!list!of!articles).!Mapping!of! CPT!codes!to!relative!value!units!has!been!completed!using!the!BOLD!Registry!data.!!A!manuscript!of!this! work!is!currently!being!drafted.!Mapping!of!codes!determined!to!be!“spine\related”!has!been!completed! utilizing!previous!work!performed!by!a!colleague!currently!at!Dartmouth,!Brook!Martin,!PhD,!MPH.!! Working!Group!4 !will!develop!and!validate!the!methods!to!extract!the!necessary!data!to!passively! measure!outcomes!from!each!site’s!EMR.!The!group!will!perform!test!data!pulls!from!each!site!of!the! key!variables!identified!by!Working!Group!3.!Anonymized!data!will!be!transmitted!to!the!Data! Coordinating!Center!at!the!UW,!housed!at!the!CBS.!UW!would!only!receive!limited!data!sets!without! protected!health!information.!!Dr.!Nick!Anderson!will!lead!this!group,!working!closely!with!informatics! experts!at!each!of!the!sites.!Dr.!Anderson!was!Associate!Director!of!the!Bioinformatics!Core!at!the!UW! ITHS!at!the!start!of!the!project!and!has!since!taken!a!position!with!University!of!California\!Davis!but!will! remain!involved!in!the!project.! Working!Group!4!progress:!!An!overall!plan!has!been!developed!to!leveraging!PopMedNet!!to!virtually! connect!the!various!implementation!sites!to!UW!who!is!to!serve!as!the!DCC!for!this!project!in! anticipation!of!the!data!exchange!which!will!take!place!in!the!UH3!phase.!!For!EHR!extraction,!we! anticipate!utilizing!the!Virtual!Data!Warehouse!at!the!three!HMORN!sites!as!much!as!possible,! developing!customized!programming!pieces!that!are!necessary!beyond!that!to!further!capture!all!the! necessary!data!elements.!!Mayo!clinic!does!not!use!the!VDW!so!all!data!extraction!programming!will! have!to!be!customized!at!that!site.! The! working! group! activities! as! well! as! general! project! coordination! will! utilize! the! web\based! tool,! Basecamp!(www.BaseCamp.com)!to!streamline!efforts.!!All!faculty!and!staff!at!the!DCC,!as!well!as!Site! PI’s!and!study!personnel!at!each!site,!will!be!given!access!to!Basecamp.!In!addition,!each!working!group! has! a! dedicated! “project”! within! Basecamp! that! will! be! used! to! facilitate! discussions! among! group! members,!organize!files,!keep!notes,!and!centralize!study\related!documents.! In!order!to!coordinate!efforts!at!each!of!the!individual!sites!and!assure!the!UH2!milestones!are!met!in!a! timely! fashion,! site! milestones! have! been! outlined! and! site! PIs! and! their! respective! research! and! technical!teams!are!working!towards!these!goals.! 1.“Radiology!buy\in”!! Assignee:!Site!PI/!Jarvik! 2013409426/kj! 24! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Objective:!Facilitate!consensus!among!Radiologists!that!they!are!willing!to!have!the!intervention! used! Deliverable!#1:!!Letter!of!support!from!Radiology!chair!(template!provided)! Due!Date:!July!31,!2013! 2.“IRB!approval”! Assignee:!!Site!PI/!James! Objective:!Coordinate!IRB!review!such!that!waiver!of!consent!and!HIPAA!for!patients!and!waiver! of!consent!for!physicians!is!in!in!place! Deliverable!#2:!!Final!IRB!approval!documentation! Due!Date:!!Final!approval!of!application!August!31,!2013!!! 3.!“Randomization!Pilot”! Assignee:!!Site!PI/!Comstock! Objective:!Demonstrate!ability!to!insert!template!into!radiology!report!on!schedule,!randomized! by!clinic! Deadline:!July!31,!2013! 4.!“EMR!data!extraction”! Assignee:!Site!PI/!Anderson! IT!resource!identified!who!Installs!and!authenticates!the!PopMedNet!Client! Deadline:!July!31,!2013! Identify!a!programming!resource!and!validate!a!"starter!set"!VDW!query!that!has!been!mapped! against!the!LIRE!data!set! Deadline:!!August!31,!2013! 2013409426/kj! 25! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Table 4: Milestones for UH2 Planning Year Needed to Transition to UH3 Implementation Suitability for Importance to UH3 UH2 Phase Milestone Timeline Assessing UH2 Success Success Aim 1:Complete critical literature Critically important: Highly suitable: review to refine the information to be Development and Necessary to develop included in the radiology report so refinement of the text describing age- that it is specific for imaging modality months intervention text is specific epidemiologic and patient age. necessary for the timely benchmarks for plain Jarvik, Bresnahan, Deyo, Halabi, implementation of the films, CT and MR. Kallmes, Turner, Luetmer, Avins intervention. Aim 2: Develop site-specific Critically important: Sites deployment methods for the stepped Highly suitable: Will must have proven ability wedge, cluster randomization scheme need detailed to selectively implement Comstock, Anderson, Avins, 3 implementation intervention text in the Cherkin, Halabi, Heagerty, James, months protocols prior to radiologic reports Jarvik, Pathak, Murphy, Ciarelli, implementation of generated for providers at Needed: KP technical resource intervention. a given primary care Needed: GHC technical resource clinic. Important: The Aim 3: Develop and validate a achievement is important Suitable: Definition of composite measure of spine but not critical. While intervention intensity that combines composite measure other metrics could be to be used as primary into a single metric the overall used successfully, intensity of resource utilization for outcome and including single months back pain care and develop cost validation using parameters, a composite existing BOLD data. estimates associated with units of measure enables a more resource used and intensity of use. comprehensive estimate Bresnahan, Deyo, James, Jarvik of overall care received for back pain. Aim 4:Develop and validate electronic data methods and tools to Highly suitable: Data capture the outcomes of interest Critically Important: dictionary and (subsequent diagnostic testing, opioid Feasibility must be 6 protocol to query the prescriptions, spinal injections, demonstrated in order to months electronic medical specialist visits, spine surgeries, etc.) passively collect record (EMR) Anderson, Comstock, James, outcomes using the EMR. necessary Jarvik, Turner Highly suitable: Drafting of DSMP, Additional Goals: Data Safety Critically Important: A submission to Monitoring Plan (DSMP) formulation 6 safety officer will need to IRBs,funding agency, and designation of safety officer months review and approve the and designation of Heagerty, Comstock, James, Jarvik study DSMP. Safety Officer needed prior to study initiation Highly suitable: Draft Critically Important: The Additional Goals: Draft study the study protocol so study protocol will need protocol it incorporates the to be reviewed by the months James, Comstock, Jarvik decisions made for IRBs before the study Aims 1-4 receives final approval. 2013409426/kj! 26! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Additional Goals: IRB approval Highly suitable: Critically Important: IRB James, Avins, Cherkin, Halabi, Conditional IRB 9-12 approval is required Jarvik, Kallmes, Project Managers approval at multiple months before study procedures from each site (Hamilton, Connelly, sites always can be initiated. Wessman, Hawkes) challenging Additional Goals: Establish subcontracts with sites Draft and submit Critically Important: 9-12 James, Avins, Cherkin, Halabi, subcontracts for the Subcontracting with sites months Jarvik, Kallmes, Post-award UH3 phase at all sites is a required process. personnel at each site 5.7!Aims!for!UH3!Phase! Aim!1:!To!determine!whether!inserting!a!description!of!ageWspecific!prevalence!of!imaging!findings! among!asymptomatic!subjects!into!lumbar!spine!imaging!reports!decreases!backWrelated!interventions! during!the!subsequent!year.! Aim!1a:!!To!determine!if!inserting!epidemiological!evidence!reduces!RVUs!attributable!to!spine! interventions.! Hypothesis'1a:'After'primary'care'clinics'are'randomly'assigned'to'receive'the'modified'report,'they'will' have'a'lower'average'overall'RVU'(technical'and'professional)'per'imagedMpatient'attributable'to'spine' interventions'than'clinics'that'are'not'receiving'the'modified'reports.' We!will!calculate!an!overall!spine\related!RVU!for!each!patient!in!the!study!by!summing!all!RVUs! attributable!to!spine\interventions!within!one!and!two!years!after!the!date!of!return!of!the!index!image! report!(the!plain!film,!MR!or!CT!of!the!lumbar!spine!imaging!study!whose!report!either!does!or!does!not! contain!the!epidemiological!benchmark!data).!For!each!patient!accrual!period!in!Figure!3,!we!will! calculate!a!total!spine\related!RVU!per!primary!care!provider!who!orders!at!least!one!lumbar!imaging! exam!in!Table!2.!We!will!aggregate!spine\related!RVUs!across!the!study\eligible!patient!panel.!The! calculated!primary!care!provider!RVU!will!serve!as!the!primary!outcome!measure!of!this!study.!As!noted,! we!will!also!apply!standardized!CMS\based!costs!to!RVU!calculations.! We!will!use!generalized!linear!mixed!models!with!jackknifed!standard!to!model!the!change!in!post\ intervention!RVU!from!pre\intervention!RVU.!We!will!use!random!effects!for!the!clinic!and!for!the!effect! of!intervention!defined!by!the!indicator!of!exposure!to!the!LIRE!intervention.!We!will!use!a!three\month! pre\intervention!washout!period!where!patients!with!index!visits!in!this!window!will!be!excluded!from! the!analysis.!We!will!adjust!the!model!for!period!of!time!(period!0!through!period!5)!as!a!fixed!effect!to! adjust!for!general!calendar!trends!in!RVUs.!We!will!also!adjust!the!model!for!the!type!of!image!ordered! at!the!initial!index!visit!as!a!categorical!variable!(plain!film,!CT,!MR).! Aim!1b:!To!determine!if!inserting!epidemiological!evidence!decreases!subsequent!opioid!prescriptions.!!! Hypothesis'1b:'Clinics'that'are'randomly'assigned'to'receive'the'modified'imaging'reports'will'have'a' lower'rate'of'subsequent'opioid'prescriptions'than'clinics'whose'patients'do'not'receive'modified'reports.' 2013409426/kj! 27! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Our!pilot!work!suggested!that!including!epidemiological!evidence!was!associated!with!a!nearly!3.5\fold! reduction!in!opioid!prescriptions.!We!feel!that!it!is!important!to!test!this!hypothesis!given!the!growing! recognition!that!the!overuse!of!opioids!is!an!important!issue!in!the!management!of!back!pain!patients.! We!will!calculate!a!binomial!outcome!of!the!number!of!patients!(out!of!the!eligible!patient!panel)!with!a! prescription!for!opioids!that!occurred!after!the!lumbar!spine!imaging!report!was!finalized.!We!will! examine!this!for!each!primary!care!provider!across!each!of!the!six!patient!accrual!periods.!We!will!utilize! a!generalized!linear!mixed!model!to!assess!the!impact!of!the!LIRE!intervention!on!subsequent!written! opioid!prescription!rates.!Since!some!patients!will!already!have!an!active!prescription!for!opioids!at!the! time!of!the!index!visit,!we!will!also!conduct!a!subgroup!analysis!by!conducting!analyses!separately!for! those!patients!with!and!without!an!active!prescription!at!baseline.!As!described!earlier,!we!will!convert! all!prescriptions!into!MEDs.!This!will!allow!us!to!examine!temporal!trends!in!not!only!prescriptions!but! also!dose.! ! ! Aim!1c:!To!determine!if!inserting!epidemiological!evidence!decreases!subsequent!crossWsectional! imaging!(magnetic!resonance!(MR)!and!computed!tomography).! Hypothesis'1c:'Clinics'that'are'randomly'assigned'to'receive'the'modified'imaging'reports'will'have'a' lower'rate'of'subsequent'crossMsectional'imaging'than'clinics'whose'patients'do'not'receive'modified' reports.' The!number!of!patients!who!receive!cross\sectional!re\imaging!within!1!and!2!years!(out!of!the!eligible! patient!panel)!will!be!calculated!as!a!binomial!outcome!measure!for!each!primary!care!provider!across! each!of!the!six!patient!accrual!periods.!We!will!again!utilize!a!generalized!linear!mixed!model!to!assess! the!impact!of!the!LIRE!intervention!on!subsequent!rates!of!cross\sectional!reimaging,!including!random! effects!for!the!baseline!rate!of!cross\sectional!re\imaging!and!for!an!indicator!of!exposure!to!the!LIRE! intervention.!We!will!adjust!the!model!for!period!of!time!(period!0!through!period!5)!as!a!fixed!effect!to! adjust!for!general!longitudinal!trends!in!re\imaging!rates.!We!will!also!adjust!the!model!for!the!type!of! image!ordered!at!the!initial!index!visit!as!a!categorical!variable!(plain!film,!CT,!MR).! Aim!1d:!To!explore!whether!adding!epidemiological!evidence!decreases!overall!costs!of!care!for!low! back!pain!based!on!CMS!reimbursement.! Hypothesis'1d:'Clinics'that'are'randomly'assigned'to'receive'the'modified'imaging'reports'will'have'lower' back'painMrelated'estimated'payer'costs'than'clinics'whose'patients'do'not'receive'modified'reports.' We!will!apply!CMS\based!reimbursement!amounts!to!RVU!calculations!in!order!to!standardize!unit!cost! estimation!among!our!sites.!We!will!use!CPT!and!diagnostic!codes!to!determine!whether!interventions! are!associated!with!back!pain,!and!multiply!the!total!back\treatment!related!RVUs!at!sites!by!the!unit! price!payment!amounts!for!respective!RVU!calculations.! Table!5:!Sample!of!RVUs!and!CMSWbased!payment!amounts!for!lumbar!imaging!(US$!2012)!! CPT!! Diagnostic!Imaging!Exam! Hospital!(facility)! !! !! Professional!! Hosp!!!!! Hosp! Pro! Pro! Pro! !! !!!! RVU! payment! !! wRVU! tRVU! payment! 72100! 2!view!x\ray!exam!lower! 0.6399! $48.22! !! 0.22! 0.34! !!!$12.08! 2013409426/kj! 28! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! spine! 4!view!x\ray!exam!lower! 72120! spine! 0.6399! $48.22! !! 0.22! 0.35! $12.47! ! ! ! ! ! ! ! ! 72131! CT!lumbar!spine!w/o!dye! 2.746! $206.54! !! 1! 1.44! $51.57! 72132! CT!lumbar!spine!w/dye! 4.2918! $323.42! !! 1.22! 1.75! $62.58! CT!lumbar!spine!w/o!&! 72133! w/dye! 4.7716! $359.59! !! 1.27! 1.82! $65.10! ! ! ! ! ! ! ! ! 72148! MRI!lumbar!spine!w/o!dye! 4.8333! $364.24! !! 1.48! 2.14! $76.44! 72149! MRI!lumbar!spine!w/dye! 6.2248! $469.10! !! 1.78! 2.57! $91.88! MRI!lumbar!spine!w/o!&! 72158! w/dye! 7.6273! $574.79! !! 2.36! 3.41! $121.79! Aim!2:!To!determine!whether!inserting!ageWspecific!prevalence!of!imaging!findings!in!asymptomatic! subjects!has!a!differential!effect!on!subsequent!backWrelated!interventions!if!inserted!into!lumbar! spine!magnetic!resonance!(MR)!and!computed!tomography!(CT)!imaging!reports!compared!with!plain! films.!! Hypothesis'2:'Inserting'epidemiological'information'into'plain'film'reports'will'result'in'a'greater' decrease'in'subsequent'backMrelated'interventions'than'similar'information'put'into'MR'and'CT'reports.' Our!pilot!work!examined!only!MR.!However,!given!that!plain!films!are!generally!obtained!earlier!in!the! course!of!back!pain!and!more!frequently!than!cross\sectional!imaging,!the!potential!impact!of!inserting! epidemiological!information!into!plain!film!reports!is!large.!In!each!separate!analysis!of!Aim!1!outcomes,! we!will!add!an!indicator!of!imaging!modality!(plain!film!versus!MR!versus!CT)!and!an!imaging!modality!by! treatment!interaction!term!in!the!model.!Primary!inference!will!be!on!the!interaction!term,!where!we! expect!that!patients!with!more!advanced!imaging!will!have!a!significantly!greater!reduction!in! subsequent!interventions!than!those!who!receive!a!plain!film!image.! Aim!3:!To!determine!whether!the!presence!of!certain!imaging!findings!influence!subsequent! interventions!!! Hypothesis'3:'Inserting'epidemiological'information'will'result'in'a'greater'decrease'in'subsequent' interventions'for'patients'without'findings'that'are'clearly'clinically'important'compared'with'patients' who'have'clinically'important'imaging'findings.' Our!work!and!others!have!shown!that!certain!imaging!findings!are!likely!clinically!more! important!than!others!(e.g.!nerve!root!compression,!moderate!to!severe!central!stenosis,!disc! extrusions).!To!address!this!hypothesis,!we!will!use!the!returned!result!from!radiology!to!categorize! imaging!findings!into!clinically!important!versus!not!clinically!important.!!We!have!identified!central! canal!stenosis,!nerve!root!compression!and!disc!extrusion!(a!type!of!herniation)!as!the!clinically!most! important!imaging!findings.!!This!is!in!contrast!to!findings!that!are!less!clinically!important!(disc!bulge,! disc!narrowing,!Modic!change,!annular!fissure,!etc).! In!each!separate!analysis!of!Aim!1!outcomes,!we!will!add!an!indicator!of!clinical!importance!and!a! clinical!importance!by!treatment!interaction!term!in!the!model.!Similar!to!Aim!2,!primary!inference!will! 2013409426/kj! 29! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! again!be!on!the!interaction!term,!where!we!expect!that!patients!without!these!more!important!findings! will!have!a!significantly!greater!reduction!in!subsequent!interventions!compared!to!patients!with! important!findings.!!! Table!6:!Milestones!for!UH3!(Implementation!Phase)W!Years!2W5! Year of UH3 Milestone Comment Project Year 2 To be completed before Wave 1 of Final testing of intervention deployment randomization scheduled for April, 2014 Planned staggered implementation using Intervention implemented at 40% of clinical sites stepped wedge design will require close monitoring of progress. This will continue the work started as Algorithm finalized for electronic medical record UH2 Milestone #4. Each site will require extraction and tested at all sites a customized algorithm- hence the need for site-specific development and testing We will prepare a manuscript describing Protocol paper submitted for publication our study protocol and procedures. Year 3 Planned staggered implementation using Randomized intervention implemented at 80% of stepped wedge design will require close clinical sites monitoring of progress. Medical record extraction complete for 12mo Data extraction ongoing for duration of outcomes on randomization waves 1-2 project for 12 and 24mo time-points. Comparison of abstraction methods for radiology reports (natural language processing vs. Amazon Turk) Year 4 All clinical sites randomized to Intervention implementation completed intervention by this time. Medical record extraction complete for 12mo outcomes on randomization waves 3-4 and 24mo outcomes on randomization waves 1-2 Abstraction of radiology reports through 12mo for waves 1-4 using preferred method from yr 3 Year 5 Medical record extraction including imaging reports complete for 12mo outcomes on randomization wave 5 and 24 mo outcomes on randomization waves 3-4 2013409426/kj! 30! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Data analysis, manuscript writing & dissemination Manuscripts for 12mo outcomes and of results at national meetings 24mo outcomes submitted for publication Table!7:!Timeline!for!the!UH3!Phase!(Years!2W5)! YEAR222(2014) YEAR232(2015) YEAR242(2016) YEAR252(2017) J F MA M J J A S O N D J F MA M J J A S O N D J F MA M J J A S O N D J F MA M J J A S O N D Randomization2Wave* Wave%#1 Wave%#2 Wave#3 Wave%#4 Wave%#5 Validation2datasets** Validation%sets%transferred%from%sites%to%DCC ✖ ✖ ✖ ✖ ✖ ✖ Comprehensive2datasets*** Comprehensive%datasets%transferred%from%sites%to%DCC%% ✖ ✖ ✖ ✖ ✖ ✖ Data2Quality2Assessment Quality%Assessment%on%set%#1 ✖ Quality%Assessment%between%set%#1%and%#2 ✖ Data2Safety2&2Monitoring2Reports ! ! ! ! ! Reports%every%six%monthes *Insertion%of%Intervention%text%,%randomized%at%clinic%level **Index%file%with%CPT%codes,%LIRE%IDs,%radiology%image%reports%verifying%insertion%of%intervention%text ***Electronic%Medical%Record%and%administrative/billing%data%for%12mo%and%24mo%outcomes 6.!Statistical!Considerations! For!details!regarding!sample!size!and!power!calculations!as!well!as!all!other!statistical!considerations,! please!refer!to!the!LIRE!Statistical!Analysis!Plan.!! 7.!Human!Subjects! Because!the!intervention!will!be!administered!at!the!clinic!level,!consent!of!either!individual!patients!or! primary!caregivers!is!neither!feasible!nor!warranted.!!Moreover,!the!intervention!is!relatively!benign! (the!insertion!of!additional!epidemiological!information!into!the!radiology!report)!and!poses!minimal! risk!to!caregivers!and!patients.!Because!leadership!at!the!Healthcare!Systems!making!up!the! performance!sites!are!enthusiastic!about!incorporating!the!epidemiological!benchmarks!into!their! reports!and!may!well!eventually!adopt!them!regardless!of!the!project,!our!study!simply!allows!us!to! systematically!study!the!effects!of!a!well\controlled!implementation!of!the!insertion!of!the!benchmark! information.!The!randomization!scheme!defines!when!each!clinic!begins!including!the!epidemiological! information!into!the!reports,!with!all!sites!eventually!receiving!the!intervention!of!interest.!! !! 7.1!Human!Subjects!Involvement!and!Characteristics!! Eligibility!criteria:!A!patient!will!be!eligible!for!inclusion!in!the!study!if!they!are!at!least!18!years!old!and! referred!by!their!primary!care!provider!for!plain!films,!CT!or!MR!of!the!lumbar!spine!to!evaluate!low! back!or!leg!pain.!!We!will!access!patient!medical!records!6!months!prior!to!the!index!image!and!for!two! years!after!the!index!image!in!order!to!track!patient!outcomes!before!and!after!the!intervention.! Subjects!will!receive!usual!care,!and!neither!their!diagnostic!evaluation!nor!their!therapy!will!be! 2013409426/kj! 31! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! constrained!by!study!considerations.!We!anticipate!enrolling!~100,000,!patients!who!underwent!lumbar! spine!imaging!examinations!across!four!different!health!systems.!!! 7.2!Research!Data!!! Research!data!will!consist!of!individual!subjects’!medical!record!data!and!information!on!clinics!and! providers.!We!will!collect!all!data!passively!with!automated!data!extractions.!!Data!extracted!from!the! medical!record!will!include!demographic!data,!variables!related!to!imaging,!pharmacy,!procedures,! hospitalizations,!and!other!factors!related!to!healthcare!utilization.!We!will!not!collect!patient!reported! outcomes!unless!they!are!part!of!the!medical!record.!!! We!will!collect!demographic!data!on!primary!care!providers!and!will!code!the!data!in!such!a!way!that!an! individual!practitioner!is!not!identifiable.!!We!will!use!patient!data!to!derive!pre!and!post!randomization! rates!of!spine!related!interventions!(diagnostic!imaging,!opioid!prescriptions,!spine!related!procedures,! physical!therapy!etc.)!among!a!provider’s!patient!panel.!!!!!!! We!will!code!with!a!unique!study!identification!number,!without!reference!to!patient!or!provider! identity.!!The!code!key!will!be!kept!secured!at!the!recruitment!site,!separate!from!the!data.!!Only!the! site!researchers!will!have!access!to!the!code!key!(not!the!researchers!at!the!DCC).!!! 7.3!Potential!risks! The!research!activities!in!this!trial!are!very!low!risk.!Perhaps!the!most!important!risk!is!a!breach!of! confidentiality!of!clinical!information.!!! !! Individual!subjects!will!not!be!contacted!or!consented!for!this!project.!!No!patient!reported!outcomes! are!being!collected!and!so!no!patient!interviews!will!be!performed.!!The!intervention!is!being! administered!at!the!clinic!level;!therefore,!consent!of!either!individual!patients!or!providers!is!neither! feasible!nor!warranted.!Moreover,!the!intervention!is!relatively!benign!(the!insertion!of!epidemiological! benchmark!data!into!the!radiology!report)!and!poses!virtually!no!risk!to!either!providers!or!patients.!!We! will!not!constrain!the!choice!of!tests!or!treatments!offered!to!subjects.!!!! !! The!main!risk!associated!with!this!project!will!be!loss!of!confidentiality!as!medical!record!access!will!be! necessary!in!order!to!assess!the!impact!of!the!intervention.!We!will!make!extensive!efforts!to!assure! that!records!are!kept!in!locked!files!and!are!not!identifiable!to!anyone!but!the!investigators.!All!PHI!will! be!stored!securely!locally.!Non\PHI!data!will!be!uploaded!via!a!web\based!system!to!the!Data! Coordinating!Center!at!the!Center!for!Biomedical!Informatics!and!Biomedical!Statistics!at!the!University! of!Washington.!Anonymized!data!will!be!stored!on!a!server!located!at!Biomedical!Informatics,!where!no! names!or!hospital!numbers!are!included!and!only!study!numbers!will!be!attached!to!the!data!files.!Data! will!be!kept!on!a!server!that!requires!a!password!for!entry!and!in!a!locked!office.!! !! As!the!identities!and!clinical!information!gathered!on!patients!will!be!guarded,!so!too,!will!the!identities! and!data!collected!on!clinic!providers.!!All!identifying!information!will!be!stored!securely!at!the!local! recruitment!sites.!!Only!coded,!limited!data!set!will!be!transferred!to!the!DCC!such!that!an!individual! provider!from!a!given!clinic!within!a!health!system;!cannot!be!identified.!!!! 7.4!Adequacy!of!Protection!Against!Risks! We!anticipate!that!each!site!will!work!within!their!own!health!system!to!identify!primary!care!clinics,! 2013409426/kj! 32! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! primary!care!providers,!and!operationalize!the!technical!aspects!regarding!the!intervention.!!The! intervention!itself!is!the!addition!of!epidemiologic!data!relevant!to!the!imaging!modality!and!age!range! (in!deciles)!of!a!given!patient!for!whom!a!radiologic!image!was!ordered.!!This!data!will!be!automatically! added!to!existing!template!radiology!reports!in!the!intervention!group.!!The!randomization!schedule!will! be!allocated!at!the!clinic!site!rather!than!at!the!individual!patient!or!provider!level.!!Only!group!results! will!be!reported.!! Per!Health!and!Human!Services!Policy!for!Protection!of!Human!Research!Subjects,!Section!46.102.i:!! “Minimal!risk!means!that!the!probability!and!magnitude!of!harm!or!discomfort!anticipated!in!the! research!are!not!greater!in!and!of!themselves!than!those!ordinarily!encountered!in!daily!life!or!during! the!performance!of!routine!physical!or!psychological!examinations!or!tests”.!! !! We!will!seek!a!waiver!of!consent!from!the!IRB’s!at!each!of!the!participating!health!care!systems!since!the! risk!to!individuals!is!minimal,!the!intervention!is!relatively!benign!and!consent!of!patients!and!providers! is!not!practical.!! 7.5!Potential!Benefits!of!the!Proposed!Research!to!the!Subjects!! We!believe!that!the!risks!to!subjects!are!minimal!and!that!the!relevant!knowledge!gains!may!be!great.!! Individual!subjects!in!this!study!are!not!likely!to!benefit!immediately!from!this!new!knowledge,!although! it!could!influence!their!subsequent!treatment,!and!may!influence!the!treatment!of!others!with!a!similar! condition.!!Knowledge!of!benefit!or!lack!thereof)!will!inform!providers!and!patients!in!the!future!about! the!usefulness!providing!epidemiologic!context!to!radiologic!results!in!the!management!of!low!back!pain.!! !! 7.6!Importance!o!the!Knowledge!to!be!Gained! This!study!will!assess!the!impact!epidemiologic!data!tailored!to!radiologic!modality!and!age!range!of!a! given!patient)!has!on!treatment!outcomes!among!those!with!low!back!pain!in!the!primary!care!setting.!! Low!back!pain!is!prevalent,!imaging!is!routinely!used!in!its!assessment!and!evaluation,!and!radiologic! results!can!heavily!inform!providers’!clinical!decision!making.!!Since!the!risks!to!research!subjects!are! minor!and!there!is!the!potential!for!improved!patient!management,!this!research!should!be!pursued.!! All!implementation!sites!now!have!IRB!approval!in!place.!!Mayo!Clinic!and!Henry!Ford!each!went! through!their!own!IRB!approval!process!for!minimal!risk!applications!and!received!approval.!!Group! Health!Cooperative!has!agreed!to!the!IRB!of!record!for!the!study!and!UW!and!Kaiser’s!IRB’s!both!ceded! authority!to!them!for!monitoring!the!study!moving!forward.! 8.0!Data!and!Safety!Monitoring!! We!have!drafted!a!data!safety!and!monitoring!plan!(DSMP)!and!have!designated!two!Safety!Officers,! Steven!Atlas,!MD!and!Constantine!Gatsonis,!PhD who!has!agreed!to!review!study!data!at!regular! intervals!for!safety!concerns.!! 2013409426/kj! 33! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! References! 1.! Lidgren!L.!The!bone!and!joint!decade!2000\2010.!Bull!World!Health!Organ.!2003;81(9):629.! PMCID:!2572548.! 2.! Initial!National!Priorities!for!Comparative!Effectiveness!Research.!Washington!DC:!Institute!of! Medicine;!2009!June!30,!2009!Contract!No.:!Document!Number|.! 3.! Schappert!SM,!Rechtsteiner!EA.!Ambulatory!medical!care!utilization!estimates!for!2006.!Natl! Health!Stat!Report.!2008(8):1\29.! 4.! Owens!DK,!Qaseem!A,!Chou!R,!Shekelle!P.!High\value,!cost\conscious!health!care:!concepts!for! clinicians!to!evaluate!the!benefits,!harms,!and!costs!of!medical!interventions.!Ann!Intern! Med.154(3):174\80.! 5.! Chou!R,!Qaseem!A,!Owens!DK,!Shekelle!P.!Diagnostic!imaging!for!low!back!pain:!advice!for!high\ value!health!care!from!the!American!College!of!Physicians.!Ann!Intern!Med.154(3):181\9.! 6.! Foundation!ABoIM,!Physicians!ACo.!Choosing!Wisely!Campaign.!2012![updated!2012;!cited];! Available!from:!choosingwisely.org/.! 7.! Martin!BI,!Deyo!RA,!Mirza!SK,!Turner!JA,!Comstock!BA,!Hollingworth!W,!et!al.!Expenditures!and! health!status!among!adults!with!back!and!neck!problems.!JAMA.!2008;299(6):656\64.! 8.! Jarvik!JJ,!Hollingworth!W,!Heagerty!P,!Haynor!DR,!Deyo!RA.!The!Longitudinal!Assessment!of! Imaging!and!Disability!of!the!Back!(LAIDBack)!Study:!baseline!data.!Spine.!2001;26(10):1158\66.! 9.! Boden!SD,!Davis!DO,!Dina!TS,!Patronas!NJ,!Wiesel!SW.!Abnormal!magnetic\resonance!scans!of! the!lumbar!spine!in!asymptomatic!subjects.!A!prospective!investigation.!J!Bone!Joint!Surg![Am].! 1990;72(3):403\8.! 10.! Jensen!MC,!Brant\Zawadzki!MN,!Obuchowski!N,!Modic!MT,!Malkasian!D,!Ross!JS.!Magnetic! resonance!imaging!of!the!lumbar!spine!in!people!without!back!pain![see!comments].!N!Engl!J! Med.!1994;331(2):69\73.! 11.! Jarvik!JG,!Hollingworth!W,!Martin!B,!Emerson!SS,!Gray!DT,!Overman!S,!et!al.!Rapid!magnetic! resonance!imaging!vs!radiographs!for!patients!with!low!back!pain:!a!randomized!controlled!trial.! JAMA.!2003;289(21):2810\8.! 12.! Lurie!JD,!Birkmeyer!NJ,!Weinstein!JN.!Rates!of!advanced!spinal!imaging!and!spine!surgery.!Spine! (Phila!Pa!1976).!2003;28(6):616\20.! 13.! Aberle!DR,!Adams!AM,!Berg!CD,!Black!WC,!Clapp!JD,!Fagerstrom!RM,!et!al.!Reduced!lung\cancer! mortality!with!low\dose!computed!tomographic!screening.!N!Engl!J!Med.!2011;365(5):395\409.! 14.! Johnson!PT,!Horton!KM,!Megibow!AJ,!Jeffrey!RB,!Fishman!EK.!Common!incidental!findings!on! MDCT:!survey!of!radiologist!recommendations!for!patient!management.!J!Am!Coll!Radiol.! 2011;8(11):762\7.! 15.! Iwabuchi!Y,!Hanamure!Y,!Ueno!K,!Fukuda!K,!Furuta!S.!Clinical!significance!of!asymptomatic!sinus! abnormalities!on!magnetic!resonance!imaging.!Arch!Otolaryngol!Head!Neck!Surg.! 1997;123(6):602\4.! 16.! Havas!TE,!Motbey!JA,!Gullane!PJ.!Prevalence!of!incidental!abnormalities!on!computed! tomographic!scans!of!the!paranasal!sinuses.!Arch!Otolaryngol!Head!Neck!Surg.!1988;114(8):856\ 9.! 17.! Roland!M,!van!Tulder!M.!Should!radiologists!change!the!way!they!report!plain!radiography!of!the! spine?!Lancet.!1998;352(9123):229\30.! 2013409426/kj! 34! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! 18.! Modic!MT,!Obuchowski!NA,!Ross!JS,!Brant\Zawadzki!MN,!Grooff!PN,!Mazanec!DJ,!et!al.!Acute!low! back!pain!and!radiculopathy:!MR!imaging!findings!and!their!prognostic!role!and!effect!on! outcome.!Radiology.!2005;237(2):597\604.! 19.! Kendrick!D,!Fielding!K,!Bentley!E,!Kerslake!R,!Miller!P,!Pringle!M.!Radiography!of!the!lumbar!spine! in!primary!care!patients!with!low!back!pain:!randomised!controlled!trial.!BMJ.! 2001;322(7283):400\5.!PMCID:!26570.! 20.! Kerry!S,!Hilton!S,!Dundas!D,!Rink!E,!Oakeshott!P.!Radiography!for!low!back!pain:!a!randomised! controlled!trial!and!observational!study!in!primary!care.!Br!J!Gen!Pract.!2002;52(479):469\74.! PMCID:!1314322.! 21.! Deyo!RA,!Diehl!AK,!Rosenthal!M.!Reducing!roentgenography!use.!Can!patient!expectations!be! altered?!Arch!Intern!Med.!1987;147(1):141\5.! 22.! Gilbert!FJ,!Grant!AM,!Gillan!MG,!Vale!LD,!Campbell!MK,!Scott!NW,!et!al.!Low!back!pain:!influence! of!early!MR!imaging!or!CT!on!treatment!and!outcome\\multicenter!randomized!trial.!Radiology.! 2004;231(2):343\51.! 23.! Ash!LM,!Modic!MT,!Obuchowski!NA,!Ross!JS,!Brant\Zawadzki!MN,!Grooff!PN.!Effects!of!diagnostic! information,!per!se,!on!patient!outcomes!in!acute!radiculopathy!and!low!back!pain.!AJNR!Am!J! Neuroradiol.!2008;29(6):1098\103.! 24.! Chou!R,!Qaseem!A,!Owens!DK,!Shekelle!P.!Diagnostic!imaging!for!low!back!pain:!advice!for!high\ value!health!care!from!the!American!College!of!Physicians.!Ann!Intern!Med.!2011;154(3):181\9.! 25.! Deyo!RA,!Mirza!SK,!Turner!JA,!Martin!BI.!Overtreating!chronic!back!pain:!time!to!back!off?!J!Am! Board!Fam!Med.!2009;22(1):62\8.!PMCID:!2729142.! 26.! Friedly!J,!Chan!L,!Deyo!R.!Increases!in!lumbosacral!injections!in!the!Medicare!population:!1994!to! 2001.!Spine!(Phila!Pa!1976).!2007;32(16):1754\60.! 27.! Luo!X,!Pietrobon!R,!Hey!L.!Patterns!and!trends!in!opioid!use!among!individuals!with!back!pain!in! the!United!States.!Spine!(Phila!Pa!1976).!2004;29(8):884\90;!discussion!91.! 28.! Deshpande!A,!Furlan!A,!Mailis\Gagnon!A,!Atlas!S,!Turk!D.!Opioids!for!chronic!low\back!pain.! Cochrane!Database!Syst!Rev.!2007(3):CD004959.! 29.! Airaksinen!O,!Brox!JI,!Cedraschi!C,!Hildebrandt!J,!Klaber\Moffett!J,!Kovacs!F,!et!al.!Chapter!4.! European!guidelines!for!the!management!of!chronic!nonspecific!low!back!pain.!Eur!Spine!J.! 2006;15!Suppl!2:S192\300.! 30.! Deyo!RA,!Gray!DT,!Kreuter!W,!Mirza!S,!Martin!BI.!United!States!trends!in!lumbar!fusion!surgery! for!degenerative!conditions.!Spine!(Phila!Pa!1976).!2005;30(12):1441\5;!discussion!6\7.! 31.! Mirza!SK,!Deyo!RA.!Systematic!review!of!randomized!trials!comparing!lumbar!fusion!surgery!to! nonoperative!care!for!treatment!of!chronic!back!pain.!Spine!(Phila!Pa!1976).!2007;32(7):816\23.! 32.! Deyo!RA,!Mirza!SK,!Martin!BI,!Kreuter!W,!Goodman!DC,!Jarvik!JG.!Trends,!major!medical! complications,!and!charges!associated!with!surgery!for!lumbar!spinal!stenosis!in!older!adults.! JAMA.!2010;303(13):1259\65.!PMCID:!2885954.! 33.! Martin!BI,!Turner!JA,!Mirza!SK,!Lee!MJ,!Comstock!BA,!Deyo!RA.!Trends!in!health!care! expenditures,!utilization,!and!health!status!among!US!adults!with!spine!problems,!1997\2006.! Spine!(Phila!Pa!1976).!2009;34(19):2077\84.! 34.! Bartleson!JD.!Evidence!for!and!against!the!use!of!opioid!analgesics!for!chronic!nonmalignant!low! back!pain:!a!review.!Pain!Med.!2002;3(3):260\71.! 35.! Paulozzi!LJ,!Ryan!GW.!Opioid!analgesics!and!rates!of!fatal!drug!poisoning!in!the!United!States.!Am! J!Prev!Med.!2006;31(6):506\11.! 2013409426/kj! 35! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! 36.! McCullough!BJ,!Johnson!GR,!Martin!BI,!Jarvik!JG.!Lumbar!MR!imaging!and!reporting! epidemiology:!do!epidemiologic!data!in!reports!affect!clinical!management?!Radiology.! 2012;262(3):941\6.!PMCID:!3285226.! 37.! Scholich!SL,!Hallner!D,!Wittenberg!RH,!Hasenbring!MI,!Rusu!AC.!The!relationship!between!pain,! disability,!quality!of!life!and!cognitive\behavioural!factors!in!chronic!back!pain.!Disabil!Rehabil.! 38.! Bick!D,!Dimmock!D.!Whole!exome!and!whole!genome!sequencing.!Curr!Opin!Pediatr.23(6):594\ 600.! 39.! Majewski!J,!Schwartzentruber!J,!Lalonde!E,!Montpetit!A,!Jabado!N.!What!can!exome!sequencing! do!for!you?!J!Med!Genet.48(9):580\9.! 40.! Brown!CA,!Lilford!RJ.!The!stepped!wedge!trial!design:!a!systematic!review.!BMC!Med!Res! Methodol.!2006;6:54.!PMCID:!1636652.! 41.! Hussey!MA,!Hughes!JP.!Design!and!analysis!of!stepped!wedge!cluster!randomized!trials.!Contemp! Clin!Trials.!2007;28(2):182\91.! 42.! Coventry!MB,!Ghormley!RK,!Kernohan!JW.!Intervertebral!disc;!its!microscopic!anatomy!and! pathology;!anatomy,!development,!and!physiology.!J!Bone!Joint!Surg.!1945;27:105.! 43.! McRae!DL.!Asymptomatic!intervertebral!disc!protrusions.!Acta!Radiologica.!1956;46:9\27.! 44.! Hitselberger!WE,!Witten!RM.!Abnormal!myelograms!in!asymptomatic!patients.!J!Neurosurg.! 1968;28(3):204\6.! 45.! Wiesel!S,!Tsourmas!N,!Feffer!H,!Citrin!C,!Patronas!N.!A!study!of!computer\assisted!tomography.!I.! The!incidence!of!positive!CAT!scans!in!an!asymptomatic!group!of!patients.!Spine.!1984;9:549\51.! 46.! Powell!MC,!Wilson!M,!Szypryt!P,!Symonds!EM,!Worthington!BS.!Prevalence!of!lumbar!disc! degeneration!observed!by!magnetic!resonance!in!symptomless!women.!Lancet.! 1986;2(8520):1366\7.! 47.! Weinreb!JC,!Wolbarsht!LB,!Cohen!JM,!Brown!CE,!Maravilla!KR.!Prevalence!of!lumbosacral! intervertebral!disk!abnormalities!on!MR!images!in!pregnant!and!asymptomatic!nonpregnant! women.!Radiology.!1989;170(1!Pt!1):125\8.! 48.! Szypryt!EP,!Twining!P,!Mulholland!RC,!Worthington!BS.!The!prevalence!of!disc!degeneration! associated!with!neural!arch!defects!of!the!lumbar!spine!assessed!by!magnetic!resonance!imaging.! Spine.!1989;14(9):977\81.! 49.! Tertti!MO,!Salminen!JJ,!Paajanen!HE,!Terho!PH,!Kormano!MJ.!Low\back!pain!and!disk! degeneration!in!children:!a!case\control!MR!imaging!study.!Radiology.!1991;180(2):503\7.! 50.! Bartolozzi!C,!Caramella!D,!Zampa!V,!Dal!Pozzo!G,!Tinacci!E,!Balducci!F.![The!incidence!of!disk! changes!in!volleyball!players.!The!magnetic!resonance!findings].!Radiol!Med!(Torino).! 1991;82(6):757\60.! 51.! Buirski!G,!Silberstein!M.!The!symptomatic!lumbar!disc!in!patients!with!low\back!pain.!Magnetic! resonance!imaging!appearances!in!both!a!symptomatic!and!control!population.!Spine.! 1993;18(13):1808\11.! 52.! Parkkola!R,!Rytokoski!U,!Kormano!M.!Magnetic!resonance!imaging!of!the!discs!and!trunk!muscles! in!patients!with!chronic!low!back!pain!and!healthy!control!subjects.!Spine.!1993;18(7):830\6.! 53.! Boos!N,!Rieder!R,!Schade!V,!Spratt!KF,!Semmer!N,!Aebi!M.!1995!Volvo!Award!in!clinical!sciences.! The!diagnostic!accuracy!of!magnetic!resonance!imaging,!work!perception,!and!psychosocial! factors!in!identifying!symptomatic!disc!herniations.!Spine.!1995;20(24):2613\25.! 54.! Salminen!JJ,!Erkintalo!M,!Laine!M,!Pentti!J.!Low!back!pain!in!the!young.!A!prospective!three\year! follow\up!study!of!subjects!with!and!without!low!back!pain.!Spine.!1995;20(19):2101\7;! discussion!8.! 2013409426/kj! 36! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! 55.! Burns!JW,!Loecker!TH,!Fischer!JR,!Jr.,!Bauer!DH.!Prevalence!and!significance!of!spinal!disc! abnormalities!in!an!asymptomatic!acceleration!subject!panel.!Aviat!Space!Environ!Med.! 1996;67(9):849\53.! 56.! Paajanen!H,!Erkintalo!M,!Parkkola!R,!Salminen!J,!Kormano!M.!Age\dependent!correlation!of!low\ back!pain!and!lumbar!disc!regeneration.!Arch!Orthop!Trauma!Surg.!1997;116(1\2):106\7.! 57.! Savage!RA,!Whitehouse!GH,!Roberts!N.!The!relationship!between!the!magnetic!resonance! imaging!appearance!of!the!lumbar!spine!and!low!back!pain,!age!and!occupation!in!males.!Eur! Spine!J.!1997;6(2):106\14.! 58.! Stadnik!TW,!Lee!RR,!Coen!HL,!Neirynck!EC,!Buisseret!TS,!Osteaux!MJ.!Annular!tears!and!disk! herniation:!prevalence!and!contrast!enhancement!on!MR!images!in!the!absence!of!low!back!pain! or!sciatica.!Radiology.!1998;206(1):49\55.! 59.! Weishaupt!D,!Zanetti!M,!Hodler!J,!Boos!N.!MR!imaging!of!the!lumbar!spine:!prevalence!of! intervertebral!disk!extrusion!and!sequestration,!nerve!root!compression,!end!plate!abnormalities,! and!osteoarthritis!of!the!facet!joints!in!asymptomatic!volunteers.!Radiology.!1998;209(3):661\6.! 60.! Rao!VM,!Levin!DC,!Parker!L,!Frangos!AJ,!Sunshine!JH.!Trends!in!utilization!rates!of!the!various! imaging!modalities!in!emergency!departments:!nationwide!Medicare!data!from!2000!to!2008.!J! Am!Coll!Radiol.!2011;8(10):706\9.! 61.! Picano!E.!Sustainability!of!medical!imaging.!BMJ.!2004;328:578\80.! 62.! Fineberg!H.!Computerized!cranial!tomography:!effect!on!diagnostic!and!therapeutic!plans.!JAMA.! 1977;38:224\7.! 63.! Fryback!D,!Thornbury!J.!The!efficacy!of!diagnostic!imaging.!Med!Decis!Making.!1991;11:88\94.! 64.! Sox!!HJ,!Margulies!I,!Sox!C.!Psychologically!mediated!effects!of!diagnostic!tests.!Ann!Intern!Med.! 1981;95:680\5.! 65.! Franklin!GM,!Stover!BD,!Turner!JA,!Fulton\Kehoe!D,!Wickizer!TM.!Early!opioid!prescription!and! subsequent!disability!among!workers!with!back!injuries:!the!Disability!Risk!Identification!Study! Cohort.!Spine!(Phila!Pa!1976).!2008;33(2):199\204.! 66.! Franklin!GM,!Mai!J,!Turner!J,!Sullivan!M,!Wickizer!T,!Fulton\Kehoe!D.!Bending!the!prescription! opioid!dosing!and!mortality!curves:!impact!of!the!Washington!State!opioid!dosing!guideline.!Am!J! Ind!Med.55(4):325\31.! 67.! Kallmes!DF,!Comstock!BA,!Heagerty!PJ,!Turner!JA,!Wilson!DJ,!Diamond!TH,!et!al.!A!randomized! trial!of!vertebroplasty!for!osteoporotic!spinal!fractures.!N!Engl!J!Med.!2009;361(6):569\79.! 68.! Jarvik!JG,!Comstock!BA,!Kliot!M,!Turner!JA,!Chan!L,!Heagerty!PJ,!et!al.!Surgery!versus!non\surgical! therapy!for!carpal!tunnel!syndrome:!a!randomised!parallel\group!trial.!Lancet.! 2009;374(9695):1074\81.! 69.! Wei!LJ,!Lachin!JM.!Properties!of!the!urn!randomization!in!clinical!trials.!Control!Clin!Trials.! 1988;9(4):345\64.! 70.! Health!WK.!Drug!facts!and!compaarisons;!2004!Contract!No.:!Document!Number|.! 71.! Efron!B,!Stein!C.!The!jackknife!estimate!of!variance.!Annals!of!Statistics.!1981;9(3):586\96.! 72.! Briggs!A,!Claxton!K,!Sculpher!M.!Decision!modeling!for!health!economic!evaluation.!Oxford:! Oxford!University!Press;!2008.! 2013409426/kj! 37! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Appendices! Appendix!A:!Article!Extraction!form!(Working!Group!1) ! Appendix!B:!Literature!Search!and!Articles!used!in!Intervention!text!(Working!Group!1) ! Appendix!C:!Pilot!Implementation!Site!Checklist!(Working!Group!2) ! Appendix!D:!Literature!Search!for!RVUWbased!Assessment!(Working!Group!3) ! 2013409426/kj! 38! Version!3! ! LIRE%Article%Abstraction%Form% Abstractor!Name:! ! ! ! Lead!author:! ! ! Journal:! ! Year:! ! 1.!Review!Article!__ __yes!(do!not!abstract!prevalence!data)!!!!!!!!__ __no! 2.!Clearly!asymptomatic!LBP!population:! no!(do!not!abstract)!!!! yes!!!!! ambiguous! 3.!Article!not!relevant!for!other!reason:!! yes!(do!not!abstract)!! no! Prevalence!of!Imaging!Findings!in!People!Without!Low!Back!Pain!(include!point! estimate!and!if!available!CI)! Finding! Overall!! Age! Age! Age! Age! Age! Age! Mean! range!! range! range! range! range! range! Age! ! ! ! ! ! ! range! #!subjects! ! ! ! ! ! ! ! without!LBP! Disc! ! ! ! ! ! ! ! degeneration! Disc!signal! ! ! ! ! ! ! ! loss! Disc!height! ! ! ! ! ! ! ! loss! ! ! ! ! ! ! ! Disc!bulge! Disc! ! ! ! ! ! ! ! protrusion! Annular! ! ! ! ! ! ! ! fissure! ! ! ! ! ! ! ! Modic!1! change! Modic!2! ! ! ! ! ! ! ! change! Facet! ! ! ! ! ! ! ! degeneration! (any)! Facet! ! ! ! ! ! ! ! degeneration! (modSsev)! Other!1! ! ! ! ! ! ! ! ! ! ! ! ! ! ! Other!2! Comments!(e.g.!population!characteristics,!potential!problems!using!data):! ! ! Last!modified!12/5/12!2:25pm! LIRE literature search strategy for Intervention Text 1. The LIRE WG1 team worked with University of Washington librarians to develop our search strategy and list of search terms. 2. Using PubMed, we used the following terms for our initial WG1 literature search: ("Morbidity"[Mesh] OR ("epidemiology"[Subheading] OR "epidemiology"[All Fields] OR "prevalence"[All Fields] OR "prevalence"[MeSH Terms]) OR "Epidemiology"[Mesh] OR "epidemiology"[Subheading] OR "Epidemiologic Factors"[Mesh] OR "Incidental Findings"[Mesh] OR incidental[All Fields] OR "Asymptomatic Diseases"[Mesh] OR asymptomatic[All Fields] OR "Unnecessary Procedures"[MeSH Terms]) AND ("Lumbar Vertebrae"[Mesh] OR "Low Back Pain"[Mesh] OR "Intervertebral Disc Displacement"[Mesh] OR "Sciatica"[Mesh] OR "Spinal Stenosis"[Mesh] OR "Synovial Cyst"[Mesh] OR "Scoliosis"[Mesh] OR "Spondylolysis"[Mesh] OR "Spinal Osteochondrosis"[Mesh] OR "disc degeneration"[All Fields] OR "disc height loss"[All Fields] OR "disc bulge"[All Fields] OR "disc protrusion"[All Fields] OR annular[All Fields] OR anular[All Fields] OR "high intensity zone"[All Fields] OR anulus[All Fields] OR annulus[All Fields] OR listhesis[All Fields] OR (disc[All Fields] AND ("desiccation"[MeSH Terms] OR "desiccation"[All Fields] OR "dessication"[All Fields])) OR "disc dehydration"[All Fields] OR (modic[All Fields] AND endplate[All Fields] AND ("Change"[Journal] OR "change"[All Fields])) OR "nerve root displacement"[All Fields] OR "nerve root compression"[All Fields] OR "disc sequestration"[All Fields] OR "intravertebral herniations"[All Fields] OR (intravertebral[All Fields] AND ("hernia"[MeSH Terms] OR "hernia"[All Fields])) OR (intradiscal[All Fields] AND ("hernia"[MeSH Terms] OR "hernia"[All Fields])) OR (intradiscal[All Fields] AND ("hernia"[MeSH Terms] OR "hernia"[All Fields] OR "herniation"[All Fields])) OR "intradural herniation"[All Fields] OR (intravertebral[All Fields] AND ("hernia"[MeSH Terms] OR "hernia"[All Fields])) OR (intravertebral[All Fields] AND ("hernia"[MeSH Terms] OR "hernia"[All Fields] OR "herniation"[All Fields])) OR "prolapsed disc"[All Fields] OR "radial fissure"[All Fields]) AND ("Magnetic Resonance Imaging"[Mesh] OR "Tomography, X-Ray Computed"[Mesh] OR "Radiography"[Mesh]) 3. Using PubMed advance search tool, the above search criteria generates N=2957 identified articles. LIRE WG1 used the following reviewers and inclusion criteria to evaluate the abstracts. Start End reviewer 1 reviewer 2 1 330 avins bresnahan 331 660 bresnahan chen 661 990 chen deyo 991 1320 deyo halabi 1321 1650 halabi jarvik 1651 1980 jarvik kallmes 1981 2310 kallmes luetmer 2311 2640 luetmer turner 2641 2957 turner avins Inclusion criteria for articles: 1. article included subjects without low back pain (LBP) 2. listed prevalence of imaging finding in patients without LBP 3. subjects were >=18 (exclude series that were strictly peds) 4. subjects were human and alive (no cadaver or animal studies) 5. imaging study prevalence data was for either MR, CT or plain film 6. prevalence for at least one of the following was included : 4. spinal stenosis, 5. disc bulge, 6. disc protrusion, 7. disc extrusion, 8. disc herniation, 9. disc degeneration, 10. disc dessication (or dehydration), 11. disc height loss, 12. nerve root involvement (contact, displacement or compression), 13. anular fissure (or anular tear or HIZ), 14. spondylolysis, 15. spondylolisthesis, 16. modic change, 17. SchmorlÕs node, 18. synovial cyst, 19. osteochondrosis References(for(LIRE(Intervention(Text( 1. Boden(SD,(Davis(DO,(Dina(TS,(Patronas(NJ,(Wiesel(SWAbno .( rmal(magnetic> resonance(scans(of(the(lumbar(spine(in(asymptomatic(subjects.(A(prospective( investigation.(J(Bone(Joint(Surg(Am.(1990(Mar;72(3):403>8.( 2. Boden(SD,(Riew(KD,(Yamaguchi(K,(Branch(TP,(Schellinger(D,(Wiesel(SW.( Orientation(of(the(lumbar(facet(joints:(association(with(degenerative(disc( disease.(J(Bone(Joint(Surg(Am.(1996(Mar;78(3):40>311.( 3. Boos(N,(Rieder(R,(Schade(V,(Spratt(KF,(Semmer(N,(Aebi(M. (1995(Volvo(Award(in( clinical(sciences.(The(diagnostic(accuracy(of(magnetic(resonance(imaging,(work( perception,(and(psychosocial(factors(in(identifying(symptomatic(disc(herniations.( Spine((Phila(Pa(1976).(1995(Dec(15;20(24):2613>25.( 4. Capel(A.(et(al.(Magnetic(resonance(study(of(lumbar(disks(in(female(dancers.(Am(J( Sports(Med.(2009(Jun;37(6):1208>13.(Epub(2009(May(5(. 5. Carragee(E,(Alamin(T,(Cheng(I,(Franklin(T,(van(den(Haak(E,(Hurwitz(E.(Are(fir> st time(episodes(of(serious(LBP(associated(with(new(MRI(findings?(Spine(J.(2006( Nov>Dec;6(6):624>35.(Epub(2006(Oct(11(. 6. Danielson(B(et(al.(Axially(loaded(magnetic(resonance(image(of(the(lumbar(spine( in(asymptomatic(individuals.(Spine((Phila(Pa(1976).(2001(Dec(1;26(23):260>16.( 7. Dora(C(et(al.(The(significance(of(spinal(canal(dimensions(in(discriminating( symptomatic(from(asymptomatic(disc(herniations.(Eur(Spine(J.(2002( Dec;11(6):575>81.(Epub(2002(Aug(28(. 8. Edmondston(et(al.(MRI(evaluation(of(lumbar(spine(flexion(and(extension(in( asymptomatic(individuals.(Manual(Therapy((2000)(5(3),(158>164( 9. Feng(T,(Zhao(P,(Liang(G.(Clinical(significance(on(protruded(nucleus(pulposus:(a( comparative(study(of(44(patients(with(lumbar(intervertebral(disc(protrusion(and( 73(asymptomatic(control(in(tridimentional(computed(tomography.Z(hongguo( Zhong(Xi(Yi(Jie(He(Za(Zhi([Chinese(Journal(of(Integrated(Traditiona l(and(Western( Medicine].(2000(May;20(5):347>9.( 10. Greenberg(JO,(Schnell(RG.(Magnetic(resonance(imaging(of(the(lumbar(spine(in( asymptomatic(adults.(Cooperative(study>>American(Society(of(Neuroimaging.(J( Neuroimaging.(1991(Feb;1(1):2>7.( ( 11. Hamanishi(C,(Kawabata(T,(Yosii(T,(Tanaka(S.(Schmorl's(nodes(on(magnetic( resonance(imaging.(Their(incidence(and(clinical(relevance.(Spine((Phila(Pa(1976).( 1994(Feb(15;19(4):450>3.( 12. Healy(JF,(Healy(BB,(Wong(WH,(Olson(EM.(Cervical(and(lumbar(MRI(in( asymptomatic(older(male(lifelong(athletes:(frequency(of(degenerative(findings.(J( Comput(Assist(Tomogr.(1996(Jan>Feb;20(1):107>12.( 13. ( (Jarvik(JJ,(Hollingworth(W,(Heagerty(P,(Haynor(DR,(Deyo(RA.(The(Longitudinal( Assessment(of(Imaging(and(Disability(of(the(Back((LAIDBack) (Study:(baseline( data.(Spine((Phila(Pa(1976).(2001(May(15;26(10):115>866.( 14. Jensen(MC,(Brant>Zawadzki(MN,(Obuchowski(N,(Modic(MT,(Malkasian(D,(Ross(JS.( Magnetic(resonance(imaging(of(the(lumbar(spine(in(people(without(back(pain.(N( Engl(J(Med.(1994(Jul(14;331(2):69>73.( 15. ( Kalichman,(L,(Li(L,(Hunter(DJ,(Been(E.(Association(between(computed( tomography>evaluated(lumbar(lordosis(and(features(of(spinal(degeneration,( evaluated(in(supine(position.(Spine(J.(2010(Apr;11(4);(308>15.( 16. Kanayama(M,(Togawa(D,(Takahashi(C,(Terai(T,(Hashimoto(T.(Cross>sectional( magnetic(resonance(imaging(study(of(lumbar(disc(degeneration(in ( 200(healthy(individuals.(J(Neurosurg(Spine.(2009(Oct;11(4):50>17.( 17. Karakida(O,(Ueda(H,(Ueda(M,(Miyasaka (T.(Diurnal(T2(value(changes(in(the(lumbar( intervertebral(discs.(Clin(Radiol.(2003(May;58(5):389>92.( 18. Kjaer(P,(Leboeuf>Yde(C,(Korsholm(L,(Sorensen(JS,(Bendix(TMag .( netic(resonance( imaging(and(low(back(pain(in(adults:(a(diagnostic(imaging(study(of(40>year>old( men(and(women.(Spine((Phila(Pa(1976).(2005(May(15;30(10):1173>80.( 19. Ranson(et(al.(Magnetic(resonance(imaging(of(the(lumbar(spine(in(asymptomatic( professional(fast(bowlers(in(cricket.(J(Bone(Joint(Surg([Br](2005;8>7B:1111>16.( 20. Savage(RA,(Whitehouse(GH,(Roberts(N.(The(relationship(between(the(magnetic( resonance(imaging(appearance(of(the(lumbars(pine(and(low(back(pain,(age(and( occupation(in(males.(Eur(Spine(J.(1997;6(2):106>14.( 21. Silcox(DH(3rd,(Horton(WC,(Silverstein(AM.(MRI(of(lumbar(intervertebral(discs.( Diurnal(variations(in(signal(intensities.(Spine((Phila(Pa(1976).(1995(Apr( 1;20(7):807>11;(discussion(811>2.( 22. Stadnik(TW,(Lee(RR,(Coen(HL,(Neirynck(EC,(Buisseret(TS,(Osteaux(MJAnnular .( ( tears(and(disk(herniation:(prevalence(and(contrast(enhancement(on(MR(images( in(the(absence(of(low(back(pain(or(sciatica. (Radiology.(1998(Jan;206(1):49>55.( ( 23. Szypryt(EP,(Twining(P,(Mulholland(RC,(Worthington(BS.(The(prevalence(of(disc( degeneration(associated(with(neural(arch(defects(of(th(elumbar(spine(assessed( by(magnetic(resonance(imaging.(Spine((Phila(Pa(1976).(1989(Sep;14(9):977>81.( 24. Takatalo(J(et(al.(Association(of(modic(changes,(Schmorl’s (nodes,(spondylolytic( defects,(high>intensity(zone(lesions,(disc(herniations,(and(radial(tears(with(low( back(symptom(severity(among(young(Finnish(adults.((Spine.(2012(June( 15;37(14)1231>9.( 25. Weinreb(JC,(Wolbarsht(LB,(Cohen(JM,(Brown(CE,(Maravilla(KR. Pr(evalence(of( lumbosacral(intervertebral(disk(abnormalities(on(MR(images(in(pregnant(and( asymptomatic(nonpregnant(women.(Radiology.(1989(Jan;170(1(Pt(1):125>8.( 26. Weishaupt(D,(Zanetti(M,(Hodler(J,(Boos(N.MR( ( imaging(of(the(lumbar(spine:( prevalence(of(intervertebral(disk(extrusion(and(sequestration,(nerve(root( compression,(end(plate(abnormalities,(and(osteoarthritis(of(the(facet(joints(in( asymptomatic(volunteers.(Radiology.(1998(Dec;209(3):661( 27. Zobel(BB(et(al.(A(Magnetic(resonance(imaging(quantification(of(early(lumbar( intervertebral(disc(degeneration(in(healthy(your(adults.((Spine.(2012(June( 15;37(14)1224>30.( ( LIRE%Projec% t % % % % Intervention%Pilot%Testing%P%lan% % April% ;% May%201% 3 1.#LIRE#study#design#overview# LIRE%is% % clu a ster%randomized%trial%to%study%the%impact%of%i texnser t% detsicng ri%a ptio% n% of%ag;especific%prevalence%of%imaging%findings%among%asymptomatic,% %su into% bjects lumbar%spine%imaging%rep.%We%a orts im%to%st % suubdy sequent% back;related% interventions%(narcotic%prescriptions,%subsequent%imaging,%injections,%surgeries,%etc)% over%thefo%llowing% 1%and%2%y.e% W ares%are%using%a%stepped%wedge%design%and%will% randomize%each%clinic%to%begin%implementing%the%intervention%textp%a re;t%one%of%5% specified%dates% :% April%1,%2014 % October%1,%20% 14 April%1,%2015 % October%1,%20% 15 April%1,%2016 % At%clinics%randomized%to%receive%the%intervention,%the%text%will%be%inserted%into%the% radiology%report%whenever%one%of%the%follow coindge% s% CP isT%%generate7d2:1 %00,% 72110,%72114,%72131,1 %3722,%72133,%72148,%72149,%72158,%72080%for% %1p8a% tients and%older.%Once%the%intervention%is%implemented%at%a%given%clinic,%it%will%remain%“on”% indefinitely%unless%the%study%stopped%earlys% onfos% r%orf% easafety%or%the%health% %system wants%to%turn%it%“off”%after%the%study%period%(April%2018)%although%this%is%not% anticipated.% 2.#Pilot#test#implementation# The%goal%of%this%pilot%implementation%is%to%verify%clinic%eligibility%for%the%LIRE%study% and%to%demonstrate%successful%insert% tiohn%o e%LIfRE%text%into%radiology%reports.%We% are%capturing%data%in%Table%2%of%this%document%that%will%be%included%in%the%UH2% progress%report%to%NIH%in%our%request%foor% %tfruanndsiintigo% n% t to%the%UH3%phase.% (So,%this%is%important% …!)% Your%approach%towardse%p %thilot%implementation%needs%to%include%the%f %ollowing: ! Assess%clinic%eligibility:%for%each%clinic%identified%in%Table%2,%verify%the% questions%in%Table%1%and%mark%in%Table%2%whether%the%site%meets%or%does%not% meet%the%inclusion%criteria%or%if%you%a e.%P re%u lease%p nsur rovide%comments%if% necessary.% ! Proof%of%implementation%of%intervention%text%in%Section%3%(Options%1%or%2):% using%dummy%records,%successfully%demonstrate%that%for%an%eligible%CPT%code% (and%patient%a>g% e1% 8)%the%text%has%been%inserted%intoo% logthye%%rreapdoirt.% An%example%printout%of%the%text%would%suffice%to%show%this.% ! Complete%pilot%implementation%befMoare% y#31,#2013.%This%is%a%hard%deadline% as%the%transition%report%must%be%written%and%submitted%to%the%NIH%soon%after% this.% ! Since%we%do%not%hav% aep% pIrRoB val,%this%test%needs%to%be%conducted%%in%a%test environment%or%by%using%a%dummy%case.% # LIRE%Projec% t % % % % Intervention%Pilot%Testing%P%lan% % April% ;% May%201% 3 Table#1.#LIRE#clinic#inclusion#criteria# Intervention implementation inclusion criteria to verify in pilot test Required Inclusion Criteria 1. Can the intervention text be delivered based upon a specific CPT code (Xray: 72100, 72110, 72114, 72080; CT: 72131, 72132, 72133; MR: 72148, 72149, 72158)? 2. Can modality-specific (Xray, CT, MR) intervention text be inserted? 3. Can the intervention text be delivered based upon patient age (patients 18 and older)? 4. Can the intervention text be delivered to clinics on a scheduled basis at the 5 pre- specified dates listed above? 5. Through an electronic medical record or radiology information system data pull, can you verify that the text was inserted into a patientÕs record with an eligible imaging CPT code? Must meet one of the following two criteria: 6.1 Can age range-specific (Section 3, Option 1) text be displayed in the radiology report depending on patient age? 6.2 Can tabular information by age (Section 3, Option 2) be displayed in the radiology report? 3.#Intervention#Pilot#Testing#Text# One%of%the%following%options%would%be%inserted%specific%to%imaging%modality% indicated%by%CPT%c.% ode Option'1:'Age-specific'intervention'text' “Comment:%The%following%findings%are%so%common%in%people%without%low%back%pain% that%while%we%report%their%presence,%they%must%be%interpreted%with%caution%and%in% the%context%of%the%clinical%situation%(R–% efJearrevnikc% e% et%al,%Spine%% 2001) Findings:%(prevalence%in%patientsX%Xa-YgY e% % without%low%back%pain),%Disk% degeneration%(decreased%T2%signal,%height%lgoes)s% ,%(9 b1 u% l ),%Disk% –% T si2g% nal%loss% (83%),%Disk%height%loss%(56%),%Disk%bulge% sk(6 %4p% ro),t% ruDsiion%(32%)n,% nuAlar% fissure%(38%)% ” Option'2:'Age-tabulated'intervention'text' The following MRI findings are so common in people without low back pain that while we report their presence, they must be interpreted with caution and in the context of the clinical situation. Disk Disk T2 Disk Disk Disk Annular Degeneration Signal Height Bulge Protrusion Fissure Loss Loss 21-30 A % A % A % A % A % A % 1 2 3 4 5 6 31-40 B % B % B % B % B % B % 1 2 3 4 5 6 41-50 C % C % C % C % C % C % 1 2 3 4 5 6 51-60 D % D % D % D % D % D % 1 2 3 4 5 6 61-70 E % E % E % E % E % E % 1 2 3 4 5 6 71-80 F % F % F % F % F % F % 1 2 3 4 5 6 81-90 G % G % G % G % G % G % 1 2 3 4 5 6 >91 H % H % H % H % H % H % 1 2 3 4 5 6 LIRE%Projec% t % % % % Intervention%Pilot%Testing%P%lan% % April% ;% May%201% 3 4.#Group#Health#clinics#identified#for#the#LIRE#project.#Please#verify#that#the# clinic#meets#the#inclusions#criteria#listed#in#Table#1.%% Table#2.#Site#eligibility#evaluation.# Meets Meets Inclusion Inclusion Criteria Criteria Reasons # Group Health Clinic Name #PCPs 1 - 5 6.1 or 6.2 for Failure 1 Bellevue Medical Center 12 2 Burien Medical Center 15 3 Capitol Hill Campus 32 4 Downtown Seattle Medical Center 7 5 Everett Medical Center 20 6 Factoria Medical Center 12 7 Federal Way Medical Center 14 8 Kent Medical Center 7 Spokane-Lidgerwood Medical 9 Center 13 10 Northgate Medical Center 31 11 Northshore Medical Center 8 12 Olympia Medical Center 41 13 Port Orchard Medical Center 18 14 Poulsbo Medical Center 8 15 Puyallup Medical Center 11 16 Rainier Medical Center 8 17 Redmond Medical Center 10 18 Renton Medical Center 13 Spokane-Riverfront Medical 19 Center 17 20 Silverdale Medical Center 16 21 Spokane-South Hill Medical Center 5 22 Tacoma Medical Center 10 23 Tacoma South Medical Center 15 24 Spokane-Veradale Medical Center 8 25 Lynnwood Medical Center 15 Additional#comments#from#pilot#test#implementation# <comments here> % LIRE Working Group 3 Appendix of Relevant Articles Using RVU-based Assessment 1. Adelgais KM, Grossman DC, Langer SG, Mann FA. Use of helical computed tomography for imaging the pediatric cervical spine. Academic emergency medicine : official journal of the Society for Academic Emergency Medicine. Mar 2004;11(3):228-236. 2. Allen B, Jr., Donovan WD, McGinty G, et al. Professional component payment reductions for diagnostic imaging examinations when more than one service is rendered by the same provider in the same session: an analysis of relevant payment policy. Journal of the American College of Radiology : JACR. Sep 2011;8(9):610-616. 3. Appleton P, Chacko A, Rodriguez EK. Financial implications of nonoperative fracture care at an academic trauma center. Journal of orthopaedic trauma. Nov 2012;26(11): 617-619. 4. Archila AM, Jarecke AJ, Damiani TN, Boorady J. Tri-service optometry productivity findings and benchmark recommendation. Military medicine. Apr 2007;172(4):418-420. 5. Arenson RL, Lu Y, Elliott SC, Jovais C, Avrin DE. Measuring the academic radiologist's clinical productivity: applying RVU adjustment factors. Academic radiology. Jun 2001;8(6):533-540. 6. Barnes SL, Robinson BR, Richards JT, et al. The devil is in the details: maximizing revenue for daily trauma care. Surgery. Oct 2008;144(4):670-675; discussion 675-676. 7. Becker ER, Hall K. Physician services in an academic neurology department: using the resource-based relative-value scale to examine physician activities. Journal of health care finance. Summer 2001;27(4):79-91. 8. Becker ER, Mauldin PD, Culler SD, Kosinski AS, Weintraub WS, King SB. Applying the resource-based relative value scale to the Emory angioplasty versus surgery trial. The American journal of cardiology. Mar 15 2000;85(6):685-691. 9. Begaz T, Decker MC, Treat R, Tews M. No relationship between measures of clinical efficiency and teaching effectiveness for emergency medicine faculty. Emergency medicine journal : EMJ. Jan 2011;28(1):37-39. 10. Bender JR, Johnson AJ, Schenk TW. The economic value of automotive occupational health services: business metrics for performance management. Journal of occupational and environmental medicine / American College of Occupational and Environmental Medicine. Feb 2008;50(2):138-145. 11. Bergersen L, Gauvreau K, McElhinney D, et al. Capture of Complexity of Specialty Care in Pediatric Cardiology by Work RVU Measures. Pediatrics. Feb 2013;131(2):258-267. 12. Bernard AM, Hayward RA, Rosevear J, Chun H, McMahon LF. Comparing the hospitalizations of transfer and non-transfer patients in an academic medical center. Academic medicine : journal of the Association of American Medical Colleges. Mar 1996;71(3):262-266. 13. Bittner JGt, Kang R, Stern PJ. Management of flexor tendon sheath ganglions: a cost analysis. The Journal of hand surgery. Jul 2002;27(4):586-590. 14. Bree RL, Kazerooni EA, Katz SJ. Effect of mandatory radiology consultation on inpatient imaging use. A randomized controlled trial. JAMA : the journal of the American Medical Association. Nov 20 1996;276(19):1595-1598. 15. Brennan DF, Silvestri S, Sun JY, Papa L. Progression of emergency medicine resident productivity. Academic emergency medicine : official journal of the Society for Academic Emergency Medicine. Sep 2007;14(9):790-794. 16. Bronstein JM, Adams EK. Rural-urban differences in health risks, resource use and expenditures within three state medicaid programs: implications for medicaid managed care. The Journal of rural health : official journal of the American Rural Health Association and the National Rural Health Care Association. Winter 2002;18(1):38-48. 17. Buntin MJ, Escarce JJ, Goldman D, Kan H, Laugesen MJ, Shekelle P. Increased Medicare expenditures for physicians' services: what are the causes? Inquiry : a journal of medical care organization, provision and financing. Spring 2004;41(1):83-94. 18. Chaudhry R, Kottke TE, Naessens JM, et al. Busy physicians and preventive services for adults. Mayo Clinic proceedings. Mayo Clinic. Feb 2000;75(2):156-162. 19. Cheriff AD, Kapur AG, Qiu M, Cole CL. Physician productivity and the ambulatory EHR in a large academic multi-specialty physician group. International journal of medical informatics. Jul 2010;79(7):492-500. 20. Chin MH, Zhang JX, Merrell K. Specialty differences in the care of older patients with diabetes. Medical care. Feb 2000;38(2):131-140. 21. Chung KC, Walters MR, Greenfield ML, Chernew ME. Endoscopic versus open carpal tunnel release: a cost-effectiveness analysis. Plastic and reconstructive surgery. Sep 1998;102(4):1089-1099. 22. Conoley PM, Vernon SW, Burtram SG. Growth profiles for radiology in the United States, 1973-1989: commentary based upon analysis of relative value units. European journal of radiology. Oct 1992;15(3):264-269. 23. Conoley PM, Vernon SW, Burtram SG. Productivity of radiologists in the United States by imaging technique: a 16-year analysis based upon relative value units. European journal of radiology. Oct 1992;15(3):258-263. 24. Conrad D, Fishman P, Grembowski D, et al. Access intervention in an integrated, prepaid group practice: effects on primary care physician productivity. Health services research. Oct 2008;43(5 Pt 2):1888-1905. 25. Cook SP. Candidate's Thesis: Laryngotracheal separation in neurologically impaired children: long-term results. The Laryngoscope. Feb 2009;119(2):390-395. 26. Cullen EJ, Jr., Lawless ST, Hertzog JH, et al. A model of determining a fair market value for teaching residents: who profits? Pediatrics. Jul 2003;112(1 Pt 1):40-48. 27. Davenport DL, Ferraris VA, Hosokawa P, Henderson WG, Khuri SF, Mentzer RM, Jr. Multivariable predictors of postoperative cardiac adverse events after general and vascular surgery: results from the patient safety in surgery study. Journal of the American College of Surgeons. Jun 2007;204(6):1199-1210. 28. Davis KA, Cabbad NC, Schuster KM, et al. Trauma team oversight improves efficiency of care and augments clinical and economic outcomes. The Journal of trauma. Dec 2008;65(6):1236-1242; discussion 1242-1234. 29. Davis PL. Assessing the potential versus the actual earnings of academic radiologists: effects of unequal duty service assignments. Academic radiology. Aug 2001;8(8): 782-791. 30. Dobrez D, Sasso AL, Holl J, Shalowitz M, Leon S, Budetti P. Estimating the cost of developmental and behavioral screening of preschool children in general pediatric practice. Pediatrics. Oct 2001;108(4):913-922. 31. Dorr DA, Wilcox A, McConnell KJ, Burns L, Brunker CP. Productivity enhancement for primary care providers using multicondition care management. The American journal of managed care. Jan 2007;13(1):22-28. 32. Dudas RA, Monroe D, McColligan Borger M. Community pediatric hospitalists providing care in the emergency department: an analysis of physician productivity and financial performance. Pediatric emergency care. Nov 2011;27(11):1099-1103. 33. Duszak R, Jr., Harris AB. Percutaneous abscess drainage: use of related radiology services and associated economic impact on a radiology practice. Journal of vascular and interventional radiology : JVIR. May 2003;14(5):597-601. 34. Fahy BN, Bold RJ, Schneider PD, Khatri V, Goodnight JE, Jr. Cost-benefit analysis of biopsy methods for suspicious mammographic lesions; discussion 994-5. Archives of surgery (Chicago, Ill. : 1960). Sep 2001;136(9):990-994. 35. Fairchild DG, McLoughlin KS, Gharib S, et al. Productivity, quality, and patient satisfaction: comparison of part-time and full-time primary care physicians. Journal of general internal medicine. Oct 2001;16(10):663-667. 36. Fleming NS, Becker ER, Culler S, Cheng D, McCorkle R, Ballard DJ. Financial performance of primary care physician practices prior to electronic health record implementation. Proceedings (Baylor University. Medical Center). Apr 2009;22(2): 112-118. 37. Grace C, Alston WK, Ramundo M, Polish L, Kirkpatrick B, Huston C. The complexity, relative value, and financial worth of curbside consultations in an academic infectious diseases unit. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. Sep 15 2010;51(6):651-655. 38. Green PH, Neugut AI, Naiyer AJ, Edwards ZC, Gabinelle S, Chinburapa V. Economic benefits of increased diagnosis of celiac disease in a national managed care population in the United States. Journal of insurance medicine (New York, N.Y.). 2008;40(3-4): 218-228. 39. Griffin TC, Hutter JJ, Johnson KK, Moscow JA. A survey of clinical productivity and current procedural terminology (CPT) coding patterns of pediatric hematologist/ oncologists. Pediatric blood & cancer. Aug 2004;43(2):140-147. 40. Henke PK, Kubus J, Englesbe MJ, Harbaugh C, Campbell DA. A statewide consortium of surgical care: a longitudinal investigation of vascular operative procedures at 16 hospitals. Surgery. Oct 2010;148(4):883-889; discussion 889-892. 41. Henley MB, Mann FA, Holt S, Marotta J. Trends in case-mix-adjusted use of radiology resources at an urban level 1 trauma center. AJR. American journal of roentgenology. Apr 2001;176(4):851-854. 42. Hillman BJ, Olson GT, Griffith PE, et al. Physicians' utilization and charges for outpatient diagnostic imaging in a Medicare population. JAMA : the journal of the American Medical Association. Oct 21 1992;268(15):2050-2054. 43. Johnson RG, Arozullah AM, Neumayer L, Henderson WG, Hosokawa P, Khuri SF. Multivariable predictors of postoperative respiratory failure after general and vascular surgery: results from the patient safety in surgery study. Journal of the American College of Surgeons. Jun 2007;204(6):1188-1198. 44. Jordan SW, Mioton LM, Smetona J, et al. Resident Involvement Affects Plastic Surgery Outcomes: An Analysis of 10,356 Patients From the NSQIP Database. Plastic and reconstructive surgery. Dec 17 2012. 45. Katz SJ, McMahon LF, Manning WG. Comparing the use of diagnostic tests in Canadian and US hospitals. Medical care. Feb 1996;34(2):117-125. 46. Khandelwal CM, Meyers MO, Yeh JJ, et al. Relative value unit impact of complex skin closures to academic surgical melanoma practices. American journal of surgery. Sep 2012;204(3):327-331. 47. Khorasani R, Goel PK, Ma'luf NM, Fox LA, Seltzer SE, Bates DW. Trends in the use of radiology with inpatients: what has changed in a decade? AJR. American journal of roentgenology. Apr 1998;170(4):859-861. 48. Kravet SJ, Jones H, Howell EE, Wright SM. Pilot study comparing patients' valuation of health-care services with Medicare's relative value units. Health expectations : an international journal of public participation in health care and health policy. Dec 2008;11(4):391-399. 49. Laditka SB, Mastanduno MP, Laditka JN. Health care use of individuals with diabetes in an employer-based insurance population. Archives of internal medicine. May 28 2001;161(10):1301-1308. 50. Latimer EA, Verrilli D, Welch WP. Utilization of physician services at the end of life: differences between the United States and Canada. Inquiry : a journal of medical care organization, provision and financing. Spring 1999;36(1):90-100. 51. Levin DC, Rao VM, Parker L, Frangos AJ, Sunshine JH. Recent trends in utilization of cardiovascular imaging: how important are they for radiology? Journal of the American College of Radiology : JACR. Sep 2005;2(9):736-739. 52. Lind BK, Gerkovich MM, Cherkin DC, Deyo RA, Sherman KJ, Lafferty WE. Effect of Risk Adjustment Method on Comparisons of Health Care Utilization Between Complementary and Alternative Medicine Users and Nonusers. Journal of alternative and complementary medicine (New York, N.Y.). Oct 4 2012. 53. Lu Y, Zhao S, Chu PW, Arenson RL. An update survey of academic radiologists' clinical productivity. Journal of the American College of Radiology : JACR. Jul 2008;5(7): 817-826. 54. Lynch F. The use of an interactive computerized daily schedule in a busy interventional radiology practice increases efficiency. Journal of the American College of Radiology : JACR. Dec 2004;1(12):965-971. 55. Maitino AJ, Levin DC, Parker L, Rao VM, Sunshine JH. Practice patterns of radiologists and nonradiologists in utilization of noninvasive diagnostic imaging among the Medicare population 1993-1999. Radiology. Sep 2003;228(3):795-801. 56. Maxwell S, Zuckerman S, Berenson RA. Use of physicians' services under Medicare's resource-based payments. The New England journal of medicine. May 3 2007;356(18): 1853-1861. 57. Mortenson MM, Ho HS, Bold RJ. An analysis of cost and clinical outcome in palliation for advanced pancreatic cancer. American journal of surgery. Sep 2005;190(3):406-411. 58. Noll RE, Jr., Tonnessen BH, Kim J, Money SR, Sternbergh WC, 3rd. Long-term postplacement cost comparison of AneuRx and Zenith endografts. Annals of vascular surgery. Nov 2008;22(6):710-715. 59. Noll RE, Jr., Tonnessen BH, Mannava K, Money SR, Sternbergh WC, 3rd. Long-term postplacement cost after endovascular aneurysm repair. Journal of vascular surgery. Jul 2007;46(1):9-15; discussion 15. 60. Phibbs CS, Bhandari A, Yu W, Barnett PG. Estimating the costs of VA ambulatory care. Medical care research and review : MCRR. Sep 2003;60(3 Suppl):54S-73S. 61. Privette AR, Shackford SR, Osler T, Ratliff J, Sartorelli K, Hebert JC. Implementation of resident work hour restrictions is associated with a reduction in mortality and provider- related complications on the surgical service: a concurrent analysis of 14,610 patients. Annals of surgery. Aug 2009;250(2):316-321. 62. Raja AS, Mortele KJ, Hanson R, Sodickson AD, Zane R, Khorasani R. Abdominal imaging utilization in the emergency department: trends over two decades. International journal of emergency medicine. 2011;4:19. 63. Rubinsky AD, Sun H, Blough DK, et al. AUDIT-C alcohol screening results and postoperative inpatient health care use. Journal of the American College of Surgeons. Mar 2012;214(3):296-305 e291. 64. Shine D, Jessen L, Bajaj J, Pencak D, Panush R. Actual and potential effects of medical resident coverage on reimbursement for inpatient visits by attending physicians. Journal of general internal medicine. Jun 2002;17(6):428-434. 65. Showstack J, Lin F, Learman LA, et al. Randomized trial of medical treatment versus hysterectomy for abnormal uterine bleeding: resource use in the Medicine or Surgery (Ms) trial. American journal of obstetrics and gynecology. Feb 2006;194(2):332-338. 66. Siegel JE, Clancy CM. Relative value in healthcare: cost-effectiveness of interventions. Journal of nursing care quality. Apr-Jun 2006;21(2):99-103. 67. Solomon A, Martino S. Relative value units: practical productivity measurement. Radiology management. Winter 1991;13(1):33-37. 68. Soremekun OA, Noble VE, Liteplo AS, Brown DF, Zane RD. Financial impact of emergency department ultrasound. Academic emergency medicine : official journal of the Society for Academic Emergency Medicine. Jul 2009;16(7):674-680. 69. Talley JD, Mauldin PD, Becker ER, Stikovac M, Leesar MA. Cost and therapeutic modification of intracoronary ultrasound-assisted coronary angioplasty. The American journal of cardiology. Jun 15 1996;77(15):1278-1282. 70. Weintraub WS, Culler SD, Kosinski A, et al. Economics, health-related quality of life, and cost-effectiveness methods for the TACTICS (Treat Angina With Aggrastat [tirofiban]] and Determine Cost of Therapy with Invasive or Conservative Strategy)-TIMI 18 trial. The American journal of cardiology. Feb 1 1999;83(3):317-322. 71. Welch WP, Verrilli D, Katz SJ, Latimer E. A detailed comparison of physician services for the elderly in the United States and Canada. JAMA : the journal of the American Medical Association. May 8 1996;275(18):1410-1416. 72. Wiersema MJ, Mergener K. Current procedural terminology, Resource-based Relative Value Scale, and the Center for Medicare and Medicaid Services: overview. Gastrointestinal endoscopy clinics of North America. Oct 2006;16(4):775-787. 73. Williams TR. A geologic survey of the medicare RBRVS system. Journal of the American College of Radiology : JACR. Mar 2004;1(3):192-198. 74. Zuber TJ, Purvis JR. Coding and reimbursement of primary care debridement and excision procedures. The Journal of family practice. Dec 1992;35(6):663-672. 75. Zwolak RM. Resource-Based Relative Value Scale (RBRVS), coding, and Medicare reimbursement. Seminars in vascular surgery. Jun 1997;10(2):119-127. Lumbar Imaging Reporting with Epidemiology (LIRE) STATISTICAL ANALYSIS PLAN Research Aims Aim 1: To determine whether inserting a description of age-specific prevalence of imaging findings among asymptomatic subjects into lumbar spine imaging reports decreases back-related interventions (imaging, injections, surgeries, etc.) over the subsequent year. Aim 1a: To determine if inserting epidemiological evidence reduces Relative Value Units (RVUs) attributable to spine interventions (imaging, injections, specialist referrals, surgeries, etc.). Aim 1b: To determine if inserting epidemiological data decreases opioid prescriptions. Aim 1c: To determine if inserting epidemiological evidence decreases subsequent cross-sectional imaging (magnetic resonance (MR) and computed tomography (CT)). Aim 1d: To explore whether adding epidemiological evidence decreases overall costs of care for low back pain based on CMS reimbursement. Aim 2: To determine whether inserting age-specific prevalence of imaging findings in asymptomatic subjects has a differential effect on subsequent back-related interventions if inserted into lumbar spine MR and CT imaging reports compared with plain films. Aim 3: To determine if specific imaging findings influence subsequent interventions. Intervention: This trial will study two groups of patients within providers, each will have had a lumbar imaging CPT code. Patients and providers at intervention clinics will receive additional prevalence summary data of incidental findings as a part of their radiology report. Control patients and providers will receive the standard imaging report without the LIRE text. Design: Using a stepped wedge cluster randomized design , we will randomly assign all predetermined clinics at each site to receive the intervention at one of five fixed time-points. Interventions will roll out every six months at the start of the second quarter of UH3 Year 2 according to the schedule shown in Figure 1. During the UH2 project phase, Figure 1: Proposed Randomization Schedule we obtained a current and accurate enumeration of PCPs within clinics. Within each recruitment site, we sorted clinics by number of primary care providers into tertiles (e.g. small, medium, large clinics). From each tertile we will randomly select clinics using urn-based randomization (without replacement) stratified by site and clinic size such that clinics of small, medium, and large size are equally represented in each randomization wave. Table 1 Created on 11/11/19 10:31:00 AM Page 1 of 8 Lumbar Imaging Reporting with Epidemiology (LIRE) STATISTICAL ANALYSIS PLAN displays the site-specific strata definitions and size. In total, we will randomize 110 clinics with 1,824 PCPs as units of observation within those clinics. Note that we have chosen to use site-specific definitions for the size of the clinic with the goal of having balance of clinic size within each site. In addition, by balancing randomization on size we will be sure to have comparable time on Table 1. Within-site stratified randomization schedule of clinics by number of PCPs. control and intervention for Units of each clinic size strata. Randomization PCP strata size boundaries (# of PCPs) (# clinics) In the original project application, we assumed 128 Recruitment Site Small Medium Large clinics and 1,898 PCPs would participate in the LIRE Group Health Cooperative 25 (370) 5 to 10 (9) 11 to 15 (8) 16 to 41 (8) project. After input from the Collaboratory Biostatistics Henry Ford Health System 26 (230) 3 to 6 (9) 7 to 9 (9) 10 to 24 (8) Core, we excluded all clinics Kaiser Permanent N. CA 20 (865) 17 to 29 (7) 33 to 39 (5) 43 to 106 (8) with a single PCP (n=18) Mayo Clinic Health System 39 (359) 2 to 4 (15) 5 to 9 (12) 11 to 34 (12) from the primary study and statistical analysis and will Total 110 (1,824) only include clinics with 2 or more PCPs. Primary Outcome: We have devoted substantial effort towards developing and refining the primary outcome measure: a summary back-specific relative value unit (RVU). The back-specific RVU is a composite measure of spine intervention intensity that combines the overall intensity of resource utilization for back pain care into a single metric. To develop the composite RVU measure, we used data from our large cohort of patients with back pain who comprise the Back pain Outcomes using Longitudinal Data (BOLD) Project, Agency for Healthcare Research and Quality (AHRQ)-funded study. During our work with the BOLD Project we developed algorithms to abstract electronic medical record (EMR) data across three health systems (two of which overlap with LIRE): Kaiser Northern California, Henry Ford Health System and Harvard Vanguard/Harvard Pilgrim. For the 5,239 BOLD cohort participants, we obtained extensive EMR data on pharmacy records, healthcare utilization (CPT codes), diagnoses and provider visits (ICD-9 codes), and inpatient hospitalization data. Using the Medicare Physician Fee Schedule (http://www.cms.gov/) we generated and tested a mapping algorithm to assign more than 10,000 unique CPT codes to RVUs. A sample of RVUs from the 2012 CMS file is shown in Table 2. Using the BOLD cohort EMR Table 2. Example spine-related CPT codes and associated RVUs. data, we developed and CPT Code Description RVU tested an algorithm for aggregating individual 72100 X-ray exam of lower spine - 2 views 1.07 RVUs across procedures 97001 PT Evaluation 2.18 over a time interval for a 99214 Detailed office visit 2.26 given patient, as well as 99284 Emergency department visit - high moderate intensity 3.37 across primary care 64483 Epidural injection for lumbar spinal stenosis 3.37 providers or clinics. 72131 CT lumbar spine w/o dye 6.27 To obtain a spine-related 72148 MRI Lumbar Spine w/o Contrast 11.31 summary RVU from CPT 63047 Removal of spinal lamina 32.89 and ICD-9 codes, we used 22804 Fusion of the spine 71.60 an existing algorithm Created on 11/11/19 10:31:00 AM Page 2 of 8 Lumbar Imaging Reporting with Epidemiology (LIRE) STATISTICAL ANALYSIS PLAN 2-4 developed by a colleague at Dartmouth College. !Aggregating across CPT codes identified by this algorithm yields the back-specific RVU. We are currently preparing a manuscript describing this development work as well as a manuscript that directly influences and informs our LIRE UH3 proposal. Using BOLD cohort data, we identified a subset of patients who have had an early lumbar image (MRI/CT or plain film) following an office visit for back pain. Our BOLD cohort manuscript (in preparation) compares the one-year cumulative RVU of early-imaged patients to carefully matched BOLD cohort controls who did not have an early lumbar image. Preliminary results indicate a substantial downstream increase in healthcare utilization for patients who received an early image compared to propensity score matched controls. Patients who underwent a lumbar MRI or CT had a mean one-year RVU of 150 +/- 410, versus 120 +/- 450 for those who had an early plain film, versus 43 +/- 120 for carefully matched controls. Mapping the relative increases of utilization of nearly 80 and 110 RVUs for the plain film and advanced imaging modalities to the example codes shown in Table 2, we see that imaged patients undergo substantially more procedures. Our expectation for the UH3 project is that the insertion of normative prevalence data into lumbar imaging reports will reduce subsequent inappropriate healthcare utilization. Secondary Outcomes: In addition to back-specific RVU, important secondary outcomes will be obtained and derived using electronic medical record data pulls and include: an indicator of opioid prescriptions written within 30 and 90 days after the index image (Aim 1b); subsequent cross-sectional re-imaging within 90 days and 12 months (Aim 1c); and medical costs (Aim 1d). In the BOLD project, we developed mapping algorithms based upon the United States Food and Drug Administration National Drug Codes (NDC) that generate an indicator of whether or not an individual pharmacy record is an opioid analgesic. Similarly, we have enumerated and categorized a listing of CPT codes that indicate cross-sectional lumbar imaging (CT, or MRI). General Analytic Strategy: To evaluate the effectiveness of inserting epidemiologic evidence into an imaging report we will use longitudinal regression methods such as linear mixed effects models (LMMs) or generalized linear mixed models (GLMMs) for all primary and secondary outcome measures. Mixed models provide an efficient method for analysis of longitudinal or multilevel data and will be the basis of our primary analysis approach. However, correct model specification is required to ensure valid results when using LMMs or GLMMs and we will therefore use robust standard errors for our primary analysis. Therefore, we are effectively adopting a “working” correlation structure through the specification of flexible multilevel models (LMM or GLMM) but will rely on non-parametrically valid robust standard errors for inference where we cluster on the clinic. Secondary analysis will directly use generalized estimating equations (GEE) adopting simple exchangeable correlation models at the clinic level to determine whether conclusions appear sensitive to model specification. In each analysis we will also consider a ‘washout period’ in the three months prior to the intervention being activated at a clinic, as determined by the randomization schedule. The rationale for a washout period is to reduce or eliminate within-provider cross-contamination of patient outcomes and utilization in the transition period between control and intervention. Including a washout period reduces the risk of having a patient initially treated in the control time period return to their primary care provider for subsequent care after the primary care provider has been exposed to the intervention through other patients. This reduces the potential bias due to within-provider cross-contamination of outcomes on the estimated inte rvention effect. Primary Analysis: The primary longitudinal model for back pain specific RVUs will use a time-varying intervention status indicator Status (0 = control, 1 = intervention, for clinic k at time t). Use of the time- kt dependent intervention status indicator permits both within-clinic contrasts that inform intervention effects (post- versus pre-intervention) as well as contrasts across clinics with different intervention statuses within each time period. The specific regression model will adopt a functional form given below, with fixed effects for time (linear), age (18-39, 40-59, 60+, using two dummy variables), imaging modality type (plain film, CT, MRI using two dummy variables), and clinic size (small, medium, large, using two dummy variables), and site (Group Health Cooperative, Henry Ford, Kaiser Permanente, Mayo Clinic, using three dummy variables) in addition to random Created on 11/11/19 10:31:00 AM Page 3 of 8 Lumbar Imaging Reporting with Epidemiology (LIRE) STATISTICAL ANALYSIS PLAN effects for provider, clinic, and intervention status: T T T T Y = b + b × Time + b × Age + b × Modality + b × Size + b × Site + l × Status + mean model ijk 0 1 t 2 ijk 3 ijk 4 k 5 k 0 kt b + b × Status + clinic random effects k,0 k,1 kt a + e provider random effects and errors jk,0 ijk We will collect the outcome measure Y on patient i (i= 1,2,…,n ) under primary care provider j (j =1,2,…,n ) ijk j k enrolled in time period t (t = 0,1,2,… ,5) in order to evaluate the overall effect of the intervention at the level of the clinic k (k = 1,2,…,110). Note that we will collect a single outcome measure for each subject recording the total utilization (RVU) over the 12 months after the index imaging event. Given that the random effects structure may contain additional elements (see below) we will use a robust standard error to test the null hypothesis that l = 0. For example, using SAS PROC MIXED we can use the “empirical” option in Figure 2: Hypothetical example of study data showing order to obtain robust standard errors. random clinic intercepts and intervention effects. Each line Alternatively, use of the jackknife (at the clinic shows the expected profile for a specific clinic. Here clinic 1 level) provides a robust standard error initiates intervention at quarter=2 while clinic 2 initiates at estimate (if using R and lmer) that is quarter=4. simple to compute. Key Model Parameters: The primary parameter of interest is l , which represents the average effect of the ● intervention adjusting for temporal trends ● − (Time ), clinic characteristics (Site , Size ), t k k and individual covariates (Age , ijk ● clinic 2 effect Modality ). In order to interpret the ● ijk random effects structure we focus on clinic 1 effect ●● clinic level means removing covariate effects where we have: adjusted mean at clinic k for times prior to intervention = b 0 ● + b ; and the adjusted mean at clinic k k,0 for times after start of the intervention = b 0 1 2 3 4 5 + b + l + b . For clinic-specific means k,0 0 k,1 study.quarter we average over both providers (ajk,0) and patients (e ). Using this representation ijk we interpret b as the pre-intervention adjusted overall mean outcome averaging across all clinics, and b is the 0 k,0 difference between that adjusted overall mean and the pre-intervention (baseline) mean for clinic k. The variance, var(b ), is a measure of the variation in the baseline mean outcome across clinics. The change in the k,0 adjusted mean outcome for clinic k is given by: (post-intervention adjusted mean) – (pre-intervention adjusted mean) = (b + b + l + b ) – (b + b ) = l + b . Here l represents the average intervention effect across all 0 k,0 0 k,1 0 k,0 0 k,1 0 clinics and b represents the difference between that average intervention effect and the intervention effect for k,1 clinic k. The variance, var(b ), is a measure of the variation in the change associated with intervention across k,1 clinics, or a measure of the heterogeneity of the intervention effect. Our primary regression model acknowledges the fundamental multilevel structure of individual-level data collected in health care systems with patients nested within providers, and providers nested within clinics. Although, the basic intervention contrast is the pre-post change associated with the initiation of intervention for each clinic, we do not propose using clinic-level summary measures for inference since the weighting of both patients and providers is not simple when heterogeneity of cluster sizes exists (e.g. PCPs per clinic, and patients per PCP). A proper multilevel model allows for optimal weighting based on the estimated variance components Created on 11/11/19 10:31:00 AM Page 4 of 8 Primary Outcome 0.6 0.7 0.8 0.9 1.0 1.1 1.2 Lumbar Imaging Reporting with Epidemiology (LIRE) STATISTICAL ANALYSIS PLAN (e.g. Gauss-Markov) and yields both an efficient summary of the overall intervention effect, as well an estimate of the variability in the magnitude of effect across clinics. However, we will not rely on the covariance model being correct for statistical inference and will use a robust (empirical) standard error. With greater than 100 total clusters (clinics) we expect valid inference and proper test size and do not anticipate needing to perform any correction such as the jackknife (recommended when the number of clusters is small). In our analysis we effectively assume that individual patients are nested within a single provider. However, in practice a patient may change providers during the follow-up year over which the primary outcome is captured. However, our basic mixed model covariance structure will simply use the assigned primary provider at the index imaging time. Therefore, we do not rely on model-based standard errors since the covariance structure may not match the true within-clinic covariance structure. We will use robust standard errors clustering at the clinic level, and therefore our analysis is valid even if there are changes in patient provider leading to an incorrectly specified covariance structure. Robust standard errors remain valid when a covariance model is not correctly specified. Furthermore, key secondary analysis of the primary outcome will directly use GEE and only cluster at the clinic level and provider level linkages are not used (nor needed) for simple GEE analysis. Secondary Analyses of Primary Outcome: We will conduct additional secondary analyses that evaluate the sensitivity of the multilevel model to the assumed basic random effects structure. We have included in the primary model multilevel random intercepts and a random effect for the clinic-level intervention. However, we will expand the random effects structure to also permit random slopes on time for both clinics and providers. Given the relatively short duration of follow-up with only six (6) total measurement times we do not expect strong heterogeneity across providers or clinics in cluster-specific temporal trends. Figure 2 shows an example of hypothetical data series for two clinics (assuming aggregation of providers to a clinic summary) and illustrates both the staggering of the crossover time and the potential to observe clinic-specific intervention effects. This figure also illustrates the fact that separating random effects of time (linear) from random effects of intervention would be difficult since time and intervention status are correlated give the unidirectional crossover from control to intervention. In addition, we will use GEE as a covariance model robust inference method and therefore can produce valid point estimates and confidence intervals without relying on correct covariance specification. Details of model choice and comparison of alternative models for longitudinal cluster level crossover trials is presented in French and Heagerty (2008) and comparison of alternative approaches is recommended. Models for time and intervention effect: Our primary analysis adopts a linear adjustment for calendar time in order to remove any large-scale temporal trends that may bias estimates of intervention effects. However, our basic regression model assumes a common (adjusted) mean for all times after the initiation of intervention. In practice there may be a delay in the impact of intervention so alternative models will be considered that incorporate a delayed and/or gradual effect of intervention. For example, the basic coding of the time-dependent variable Status takes the value 0 pre-intervention and the value 1 post-intervention. Delay in the impact of kt intervention can be accommodated using alternatives such as: 0 pre-intervention; 0.5 for quarter 1 after intervention; and 1 for all other post-intervention quarters. Such a modified model would allow full impact of the intervention to require two quarters of exposure. We will conduct secondary analyses to explore alternative models for the accumulation or delay of the intervention effect. Secondary Outcome Analysis: We will also analyze the impact of intervention on the rate of opioid prescription using Generalized Linear Mixed Models (GLMMs). For Aim 1b let Y =1 if opioids were prescribed within a given ijk timeframe (e.g. 30 days or 90 days) to patient i (i= 1,2,…,n ) seen by primary care provider j (j =1,2,…,n ) within j k clinic k (k = 1,2,…,110). Analysis for this outcome will use a logistic mixed model given as: T T T T logit(p ) = b + b × Time + b × Age + b × Modality + b × Size + b × Site + l × Status + mean model ijk 0 1 t 2 ijk 3 ijk 4 k 5 k 0 kt b + b × Status + clinic random effects k,0 k,1 kt a provider random effects jk,0 Created on 11/11/19 10:31:00 AM Page 5 of 8 Lumbar Imaging Reporting with Epidemiology (LIRE) STATISTICAL ANALYSIS PLAN where p denotes the probability that Y =1. Our secondary outcome analysis parallels the primary and will be ijk ijk based on a natural multilevel mixed model, with additional robust secondary analysis provided by GEE. For Aim 1c we will use Y =1 if CT or MR imaging occurs within a specified timeframe (e.g. 90 days or 12 months) after ijk the index imaging event. Medical costs (Aim 1d): Spine-related costs of care will be estimated using two approaches. First, we will use the spine-related RVU calculated in Aim 1a and estimate clinical-level, spine-intervention expenditures using the annual Medicare-determined payment amount per RVU (e.g., CY2013 = $34.023 per RVU). (reference: http://www.cms.gov/Outreach-and-Education/Medicare-Learning-Network- LN/MLNProducts/downloads/medcrephysfeeschedfctsht.pdf ) Second, as a proxy for costs of spine care, we will use a standard set of reimbursement amounts, i.e., CMS- based payments, and estimate clinic-level spine-related aggregate expenditures by applying CPT-based payment amounts to specific spine-intervention events (e.g., imaging, office visits, procedures, other). We will present monthly and annual means, medians, and ranges of clinic-level cost estimates, prior to and subsequent to implementing the epidemiological intervention. We will assess the level of right-skewness in the expenditure estimates and use t-tests to compare arithmetic means for clinic-level expenditures. In the case of considerable skewness, we will test for differences in logarithmically transformed mean clinic-level expenditures (before and after implementing intervention). We will also describe categories of prescriptions ordered, when available in the electronic medical records for a health system, and estimate costs for prescribed spine-related medications Analysis for Aim 2: The hypothesis of Aim 2 is that there will be a differential effect of the intervention according to the imaging modality used. In order to test this hypothesis we will analyze patient-level data according to the appropriate LMM or GLMM given above, but including the interactions between Modality indicators (modeled ijk using two indicator variables coding CT and MR, with plain film as the reference) and Status . A test of the kt interaction terms (2 degree of freedom Wald test) will be used to test the null hypothesis that the effect of the intervention does not vary according the imaging modality. Analysis for Aim 3: The hypothesis of Aim 3 is that there will be a differential effect of the intervention according to the results that are found in the imaging report. We will use an additional variable, ImageFinding , that takes ijk the value 1 if a significant image finding is present, and 0 otherwise (see detail regarding variable specification in protocol). We will test the null hypothesis that the interaction between ImageFinding and Status is zero using a ijk kt Wald test. Power Calculations: Our UH2 efforts with respect to sample size and statistical power focused on two key items. First, an important aim of our UH2 Working Group 2 was to obtain an accurate clinic and provider count for each health system. We will now randomize n=110 clinics (1,824 PCPs), which is slightly lower than the n=128 clinics (1,898 PCPs) assumed in our initial project application. However, the majority of the clinics that were dropped were those with only one PCP and therefore would not have contributed much information to the analysis. Second, in our Working Group 3 we sought to develop and characterize a composite RVU summary to be used as the primary outcome measure in this study. In our UH2 project application we discussed statistical power in the context of an important secondary outcome measure, a reduction in subsequent opioid prescription rates. We now present statistical power for the primary outcome measure using data from the BOLD Registry to inform key design parameter estimates. To our knowledge, off-the-shelf calculators do not exist that would adequately characterize statistical power for a stepped wedge cluster randomized trial with a varying number of sampling units between clusters. We therefore utilized simulation methods to generate and analyze data that closely mimics the design characteristics we anticipate for this study. With a simulation approach, we were able to include estimates of both patient and clinic- level variability and implement the proposed primary analysis methods: random intercept linear mixed effects models for RVU outcomes; and generalized linear mixed models for opioid prescription rates. All simulations Created on 11/11/19 10:31:00 AM Page 6 of 8 Lumbar Imaging Reporting with Epidemiology (LIRE) STATISTICAL ANALYSIS PLAN were conducted using R (version 3.0.1) with the lmer and glmmPQL functions implementing mixed model estimation. Power for Primary Outcome: In the BOLD Registry we identified 639 patients in the Kaiser Permanente and Henry Ford health systems that had a qualifying lumbar image within 6 weeks of a PCP visit, the majority (74%) of which occurred within 7 days. As one would expect with a measure of health care utilization intensity, patient- level RVUs are positively skewed. In our simulations and in the future analysis of study data, we therefore utilize an approximately normalizing transformation of log(RVU + 1) but will Figure 2: Statistical power for make interpretations regarding effect size back on the original RVU scale. spine-related RVU With log-transformed BOLD Registry RVU data, we fit a linear mixed effects model adjusting for image type (advanced vs. plain film) and study recruitment site and estimated the variance components for clinic (0.026) and the residual error term (1.230). The observed intra-class correlation coefficient (ICC) across clinics was 0.013 (95% CI: 0.000 to 0.046). In this subset of BOLD data, the number of PCPs with multiple patients was too few to inform the PCP-level variance component and it is therefore conservatively included in the error term variance for power simulations. The numbers of clinics and primary care providers were considered fixed 0% 2% 4% 6% 8% 10% for each simulation and we assumed that each provider would provide Percentage Reduction in Median PCP spine RVU data for all study time periods. For a range of potential RVU effect sizes, we generated 1,000 simulated data sets and performed mixed model estimation with each data set. In Figure 2, we show statistical power for the primary outcome measure of PCP spine-related RVU under the proposed study design. The study has greater than 90% power to detect reductions in the median spine-related RVU of 5.0% or larger. For a patient receiving a lumbar CT, a 5% reduction in spine RVU translates into one fewer additional lumbar CT scan on average compared to a patient unexposed to the LIRE intervention. Power for Secondary Outcome: Using the updated clinic and provider listing, we repeated the UH2 power analyses for a reduction in subsequent opioid prescriptions. We again used an average baseline opioid prescription rate of 22%, suggested from the pilot manuscript to anchor the effect size of percent reduction in the baseline rate. A clinic-specific baseline opioid prescription rate was drawn from a Beta(a=6, b=20) distribution; we used this rate to draw a baseline opioid prescription rate random effect for each primary care provider using a clinic-specific log-normal distribution. We generated 1,000 simulated data sets using the effect size of a 7.5% reduction in the opioid reported in our UH2 application and evaluated each using a GLMM assuming random intercepts. With 110 clinics randomized, the study remains well powered (88.9% power) to detect a reduction in the rate of opioid prescriptions of 7.5% or larger (e.g. 22% down to 20.4%). Statistical Analysis Plan References 1. Hussey MA, Hughes JP. Design and analysis of stepped wedge cluster randomized trials. Contemp Clin Trials 2007;28:182-91. 2. Martin B, Mirza SK, Lurie JD, et al. Validation of an administrative coding algorithm to identify back-related degenerative diagnoses. International Society for the Study of the Lumbar Spine (ISSLS). Scottsdale, AZ, 3. Martin BI, Gerkovich MM, Deyo RA, et al. The association of complementary and alternative medicine use and health care expenditures for back and neck problems. Med Care 2012;50:1029-36. 4. Martin BI, Mirza SK, Franklin GM, et al. Hospital and surgeon variation in complications and repeat surgery following incident lumbar fusion for common degenerative diagnoses. Health Serv Res 2013;48:1-25. 5. http://www.fda.gov/drugs/informationondrugs/ucm142438.htm, accessed September 13, 2013. Created on 11/11/19 10:31:00 AM Page 7 of 8 Statistical Power 0% 20% 40% 60% 80% 100% Lumbar Imaging Reporting with Epidemiology (LIRE) STATISTICAL ANALYSIS PLAN 6. Diggle PJ, Heagerty PJ, Liang KY, Zeger SL. Analysis of Longitudinal Data. Second Edition ed: Oxford University Press; 2002. 7. Efron B, Stein C. The jackknife estimate of variance. Annals of Statistics. 1981;9(3):586-96. 8. French B, Heagerty PJ. Analysis of longitudinal data to evaluate a policy change. Statistics in Medicine 2008; 27(24):5005-25. 9. McCullough BJ, Johnson GR, Martin BI, Jarvik JG. Lumbar MR imaging and reporting epidemiology: do epidemiologic data in reports affect clinical management? Radiology. 2012;262(3):941-6. PMCID: 3285226. Created on 11/11/19 10:31:00 AM Page 8 of 8 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png JAMA Network Open American Medical Association

The Effect of Including Benchmark Prevalence Data of Common Imaging Findings in Spine Image Reports on Health Care Utilization Among Adults Undergoing Spine Imaging

Download PDF
89 pages

Loading next page...
 
/lp/american-medical-association/the-effect-of-including-benchmark-prevalence-data-of-common-imaging-2949kLME1X

References (183)

Publisher
American Medical Association
Copyright
Copyright 2020 Jarvik JG et al. JAMA Network Open.
eISSN
2574-3805
DOI
10.1001/jamanetworkopen.2020.15713
Publisher site
See Article on Publisher Site

Abstract

RESULTS We enrolled 250 401 participants (of whom 238 886 [95.4%] met eligibility for this Supplemental content analysis, with 137 373 [57.5%] women and 105 497 [44.2%] aged >60 years) from 3278 primary care Author affiliations and article information are clinicians. A total of 117 455 patients (49.2%) were randomized to the control group, and 121 431 listed at the end of this article. patients (50.8%) were randomized to the intervention group. There was no significant difference in cumulative spine-related RVUs comparing intervention and control conditions through 365 days. The adjusted median (interquartile range) RVU for the control group was 3.56 (2.71-5.12) compared with 3.53 (2.68-5.08) for the intervention group (difference, −0.7%; 95% CI, −2.9% to 1.5%; P = .54). Rates of subsequent RVUs did not differ between groups by specific clinical findings in the report but did differ by type of index imaging (eg, computed tomography: difference, −29.3%; 95% CI, −42.1% to −13.5%; magnetic resonance imaging: difference, −3.4%; 95% CI, −8.3% to 1.8%). We observed a (continued) Open Access. This is an open access article distributed under the terms of the CC-BY License. JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 1/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults Abstract (continued) small but significant decrease in the likelihood of opioid prescribing from a study clinician within 1 year of the intervention (odds ratio, 0.95; 95% CI, 0.91 to 1.00; P = .04). CONCLUSIONS AND RELEVANCE In this study, inserting benchmark prevalence information in lumbar spine imaging reports did not decrease subsequent spine-related RVUs but did reduce subsequent opioid prescriptions. The intervention text is simple, inexpensive, and easily implemented. TRIAL REGISTRATION ClinicalTrials.gov Identifier: NCT02015455 JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 Introduction 1,2 Spine imaging often reveals incidental findings among individuals without back pain, which can 3,4 5 lead to unnecessary and possibly harmful tests and treatments. Roland and van Tulder proposed adding statements to plain film reports describing the prevalence of degenerative findings in people 6,7 without back pain. In small observational studies, we and others have found that primary care patients undergoing lumbar spine imaging were less likely to receive certain subsequent diagnostic and therapeutic interventions if imaging reports contained information describing the prevalence of common imaging findings among individuals without back pain. These results suggest that benchmark information may reassure both patients and physicians, resulting in fewer downstream interventions. Since beginning our trial, others have published research suggesting that 7-9 contextualizing imaging information can affect both health care professionals and patients. We now report the results of a large, prospective randomized clinical trial of this intervention, the Lumbar Imaging with Reporting of Epidemiology (LIRE) trial. Our primary hypothesis was that patients of primary care professionals (PCPs) who received lumbar spine imaging reports with age-appropriate and imaging modality–appropriate benchmark prevalence data would have less spine-related health care utilization, as measured by our primary outcome, spine-related relative 10,11 value units (RVUs) (eAppendix 1 in Supplement 1). RVUs are based on Current Procedural Terminology (CPT) and provide a common metric for comparing health care utilization resulting from physician services. We also report the impact of the intervention on the prespecified secondary outcome of subsequent opioid prescriptions and prespecified subgroup analyses examining initial (index) imaging type and index report findings. Methods Study Design We conducted a multicenter, stepped-wedge, cluster randomized clinical trial assigning primary care clinics at 4 large health systems to when they would begin receiving lumbar spine imaging reports containing age-appropriate and modality-appropriate epidemiological benchmarks for common imaging findings. We previously published our study protocol, and it is available in Supplement 2. We designed LIRE to be highly pragmatic (eAppendix 2 in Supplement 1) to measure effects in routine care settings. We chose clinic-level cluster randomization because of the strong concern regarding contamination from intervention PCPs to control PCPs. We chose a stepped-wedge randomization because of the appeal of all clusters receiving the intervention by the end of the trial, facilitating implementation and the ability to perform both within-cluster (ie, before and after) and between-cluster comparisons. JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 2/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults Each health care system’s institutional review board or ethics committee reviewed the project, and all institutional review boards classified our study as minimal risk, granting waivers of both informed consent and Health Insurance Portability and Accountability Act authorization. This study followed the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline. Participants We enrolled clinics and their patients at 4 integrated health care systems: Kaiser Permanente Northern California; Henry Ford Health System in Michigan; Kaiser Permanente Washington; and Mayo Clinic Health System in Minnesota and Wisconsin. These systems have comprehensive electronic medical record (EMR) systems to capture health care utilization data. Clinic, PCP, and Patient Eligibility Criteria At each system, we identified adult primary care clinics and their physicians in family medicine, general internal medicine, and associated mid-level clinicians. We defined a LIRE clinician as a PCP whose main practice was at 1 clinic providing primary care and who ordered at least 1 qualifying imaging examination during the study period. We enrolled patients aged 18 years and older whose PCP from an eligible clinic ordered an imaging test of the lumbar spine between October 1, 2013, and September 30, 2016. We included all patients receiving eligible imaging studies at participating clinics who had not had lumbar spine imaging within the prior 12 months. We excluded only those patients who had opted out of research studies. Patient Identification We identified eligible patients and PCPs using the electronic ordering systems. When a PCP ordered an eligible examination, the system automatically determined whether the patient, PCP, and clinic were eligible. Randomization We used a stepped-wedge randomization scheme, randomly assigning clinics in each system to begin receiving the intervention at 1 of 5 dates at 6-month intervals from April 2014 through April 2016. We classified clinics in tertiles by their number of PCPs. The data coordinating center randomly selected clinics using urn-based randomization (without replacement) stratified by system and clinic size stratified by tertile (small, medium, and large). Clinic sizes were represented equally in each randomization wave. Because of the stepped-wedge temporal randomization scheme, we labeled clinics control if inclusion of the intervention text had not started and intervention after starting inclusion of the intervention text. Masking of the participating clinics was not feasible because of the nature of the intervention. Except for the biostatistician who received and cleaned the data, all investigators at the data coordinating center remained masked to clinic and participant assignment until the final stages of data analysis. Procedures The intervention text consisted of age-specific and modality-specific epidemiological benchmarks indicating the prevalence of common findings from imaging in people without back pain (eAppendix 5,6,15 3in Supplement 1). Using an automated approach through either the radiology information system or the EMR, we inserted the intervention text into lumbar spine imaging reports at intervention clinics. PCPs in control clinics received usual imaging reports. Data and Collection Methods We collected all data passively from the EMR and electronic administrative data systems. We performed 2 types of data queries from each system. To verify that the systems deployed the intervention appropriately, we queried systems 2 to 4 weeks after the start of each randomization wave for all patients who received an eligible lumbar imaging study. Text matching verified that the JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 3/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults reports contained the correct intervention text. One year after the first randomization wave and then every 6 months thereafter, we performed an additional query that included safety and outcome variables. Systems submitted both types of queries as limited data sets (deidentified except for dates of service) to the data coordinating center, providing unique study identifiers for each patient. We collected diagnosis and utilization data for patients 12 months before index imaging to characterize the cohort at the patient level. Each health system provided prescription data using national drug code or a similar classification from their pharmacy databases. This National Institutes of Health–sponsored trial required the collection of race and ethnicity data, which we obtained through the EMR. Outcomes Our primary outcome was the cumulative spine-related RVUs 365 days after index imaging. Spine- related RVUs are a composite measure of back pain interventions that combine the overall intensity of resource utilization for back pain care in a single metric. Our summary spine-related RVU incorporated procedures (CPT codes), diagnoses (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] and International Statistical Classification of Diseases, Tenth Revision, Clinical Modification [ICD-10-CM] codes), PCP visits, and inpatient hospitalizations and was based on a validated algorithm. eAppendix 1 in Supplement 1 provides examples of spine- related procedures and associated RVUs. To obtain a spine-related summary RVU from CPT and ICD-9-CM and ICD-10-CM codes, we used an existing validated algorithm when possible and used a 18,19 modified version for codes not accounted for by the algorithm. We aggregated relevant CPT codes through 1 year after the index imaging test to obtain total spine-related RVUs. The data coordinating center performed all calculations for assessing spine-related RVUs. We listed the following prespecified secondary outcomes in our published protocol: (1) an indicator of opioid prescribing after the index imaging; (2) cumulative spine-related total RVUs 2 years after index imaging; (3) subsequent advanced imaging (ie, number of magnetic resonance imaging [MRI] or computed tomography [CT] studies) within 90 days and 12 months after index imaging study; (4) spine injections and spine surgeries; and (5) other back-related medical costs during 2 years. In our original study protocol, the opioid outcome was the number of patients with a subsequent opioid prescription written by a study PCP. However, following discussions among the study team, we concluded that the total morphine equivalent dose (MED) prescribed per patient would be a better metric and thus included the number of MEDs prescribed per patient as the opioid outcome in our protocol paper. However, we were unable to obtain the necessary data for this detailed calculation. Instead, we analyzed whether patients had received an opioid prescription from a LIRE PCP within 1 year of index imaging; this was the outcome in our pilot project. We also report whether an opioid prescription was received within 90 days of index imaging, an outcome not prespecified on ClinicalTrials.gov. Extracting Imaging Results We used machine learning natural language processing to extract imaging findings from radiology text reports. We identified common imaging findings that are likely less clinically important (eg, disc bulge, disc space narrowing) vs likely more important (eg, moderate to severe spinal canal 1,21 stenosis, nerve root compression (eAppendix 4 in Supplement 1). Statistical Analysis To evaluate the impact of the intervention, we used multilevel linear mixed-effects models or generalized linear mixed models that cluster on clinic and then PCP within clinic, coupled with the use of robust standard errors for all primary and secondary outcome measures (Supplement 2). In secondary analyses, we used generalized estimating equations, adopting simple exchangeable correlation models at the clinic level to determine whether conclusions were sensitive to model specification (eAppendix 5 in Supplement 1). All analyses used the intention-to-treat principle. JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 4/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults We used a log transformation of RVU [log(RVU + 1)] in primary outcome models to address right skew of the utilization data. A constant (ie, 1) was added to RVU prior to transformation so that participants with 0 RVUs could be included in the analyses. We conducted sensitivity analyses that varied the constant added to RVU before the transformation (eAppendix 5 in Supplement 1). We also constructed a model examining subsequent RVUs in a subgroup of patients from clinics in which patients were less likely to have sought outside care who had utilization within their system through at least 12 months to address the issue of care received out of system. We used a similar analytic approach for opioid prescriptions as we used for spine-related RVUs but adapted generalized linear models to use logistic regression for this binary outcome. Realizing the potential importance of confounding due to secular trends in opioid prescribing, we conducted additional post hoc opioid analyses exploring sensitivity to alternative modeling of time. We also performed analyses on an outcome that incorporated opioid prescriptions from both LIRE and non- LIRE PCPs. We had 2 prespecified subgroup analyses. We used findings extracted from the reports to determine whether the findings in the imaging report influenced the effects of the intervention. We also examined whether the intervention effect was modified by modality of index imaging. We tested these hypotheses as interaction terms using the Wald test. A post hoc subgroup analysis distinguished between those patients who were and were not prior opioid users because the intervention might have been more likely to prevent opioid prescriptions from being written for opioid-naive patients. We defined prior use as at least 1 opioid prescription written within 120 days before the index imaging date, and we included an interaction term of prior opioid prescription status with intervention status in those models. We used model results and patient covariates to calculate predicted RVUs and predicted probability of opioid prescription for each participant under both the control and intervention conditions and aggregated the results to report median adjusted RVUs and adjusted opioid prescriptions by intervention status and subgroups. We used SAS software version 9.4 (SAS Institute) for all analyses. Statistical significance was set at P < .05, and all tests were 2-tailed. Power for Primary Outcome We calculated statistical power for the primary outcome, spine-related RVUs. The study had 89% power to detect reductions of 5.0% or greater. Data Safety Monitoring Two external safety officers monitored emergency department visits within 90 days and deaths within 6 months of index imaging. The safety officers used absolute relative risk ratio monitoring thresholds of 1.15 and 1.10 for comparing 90-day emergency department visit and death rates by intervention group, with adjustment for patient-specific characteristics (ie, age, sex, Charlson 23 24 comorbidity index ), health care system, image modality, time, season, and clinic size. Results We randomly allocated intervention start dates to 98 clinics with 3278 PCPs and 250 401 patients. A total of 11 515 patients were excluded for the following reasons: prior lumbar spine image within 12 months (11 149 [96.8%]), imaging report finalization date more than 4 days after image completion date (354 [3.1%]), image completion date prior to report finalization date (3 [<0.1%]), and unable to link to utilization data (9 [0.1%]). This resulted in a final sample of 238 886 patients (95.4%; 137 373 [57.5%] women; 105 497 [44.2%] aged >60 years) with 3257 PCPs (99.4%). Three health systems were of comparable size and enrolled 41 882 patients (17.5%) from 936 PCPs (28.7%) while the fourth health system enrolled 197 004 patients (82.5%) from 2321 PCPs (71.3%) (Figure 1). We did JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 5/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults not observe any substantial differences in the baseline characteristics between the control and intervention groups (Table). Our primary outcome, 12-month spine-related RVU, was not significantly different for the intervention group compared with the control group (adjusted median [interquartile range], 3.53 [2.68-5.08] vs 3.56 [2.71-5.12]; difference, −0.7%; 95% CI, −2.9% to 1.5%; P = .54) (Figure 2). Injections and surgery accounted for a higher proportion of subsequent spine-related RVUs for patients who had magnetic resonance imaging or computed tomography for their index examination compared with radiographs, while physical therapy and imaging were proportionally higher for patients who had radiographs as the index imaging test (eAppendix 7 in Supplement 1). Our prespecified secondary outcome, opioid prescriptions by a LIRE PCP within 1 year of index imaging, demonstrated a small but statistically significant reduction in the odds of receiving at least 1 prescription for an opioid for patients in the intervention group compared with patients in the control group (adjusted opioid proportion, 36.2% vs 37.0%; odds ratio, 0.95; 95% CI, 0.91-1.00; P = .04) (Figure 3). Sensitivity analyses with alternative modeling of time yielded similar results (Figure 3). Comparison of opioid prescribing between control and intervention groups within 90 days following index imaging showed a similar small reduction in the odds of receiving an opioid prescription for the intervention group compared with the control group (adjusted opioid proportion, 28.9% vs 29.8%; odds ratio, 0.95; 95% CI, 0.90-0.99; P = .02) (eAppendix 6 in Supplement 1). Safety monitoring demonstrated no evidence of increased deaths or emergency department visits in the intervention vs control group within 6 months of the index test (adjusted emergency department visit rate, 11.1% vs 11.3%; OR, 0.98; 95% CI, 0.94-1.01) (Figure 4). Figure 1. CONSORT Stepped-Wedge Allocation of Trial Subjects Clinics under control condition Clinics under intervention condition b b Step 0 Step 1 Step 2 Step 3 Step 4 Step 5 Clinic October 2013- April 2014- October 2014- April 2015- October 2015- April 2016- Total Group March 2014 September 2014 March 2015 September 2015 March 2016 September 2016 41 558 Analyzed 10 630 Analyzed 52 188 Analyzed 34 219 Intervention 78 Intervention 2394 Excluded (19 clinics) 7339 No intervention 970 Excluded 1424 Excluded 31 611 Analyzed 15 605 Analyzed 47 216 Analyzed 29 167 Intervention 4 Intervention 2158 Excluded (20 clinics) 2444 No intervention 1134 Excluded 1024 Excluded 30 157 Analyzed 29 628 Analyzed 59 785 Analyzed 25 313 Intervention 394 Intervention 2766 Excluded (20 clinics) 4944 No intervention 1788 Excluded 978 Excluded 21 970 Analyzed 10 277 Analyzed 32 247 Analyzed 194 Intervention 9433 Intervention 1887 Excluded (18 clinics) 1428 Excluded 844 No Intervention 459 Excluded 39 622 Analyzed 10 277 Analyzed 47 450 Analyzed 7411 Intervention 114 Intervention 2310 Excluded (21 clinics) 2037 Excluded 417 No Intervention 273 Excluded Totals 7828Analyzed 117 455 Analyzed 121 431 Analyzed 238 886 Analyzed 7411Intervention All 784 Intervention 105 543 Intervention 11 515 Excluded (98 clinics) 417 No intervention 7357 Excluded 15 888 No intervention 273 Excluded 4158 Excluded For clinics under the control condition, intervention indicates the intervention text was By pretrial design, for 1 clinic, step 0 extended through May 2014, and step 1 began mistakenly included in the image report. For clinics under the intervention condition, June 1, 2014. intervention indicates that the intervention text was successfully included in the image report and no intervention indicates that the intervention text was not included. Two small clinics randomized to groups 2 and 5 were dropped before the first data submission because of clinic closure and are not included in the clinic counts. JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 6/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults Table. Baseline Characteristics No. (%) Characteristic Control (n = 117 455) Intervention (n = 121 431) Site A 6950 (5.9) 7388 (6.1) B 96 275 (82.0) 100 729 (83.0) C 7846 (6.7) 7736 (6.4) D 6384 (5.4) 5588 (4.6) Age, y 18-39 21 237 (18.1) 22 105 (18.2) 40-60 45 032 (38.3) 44 995 (37.1) ≥61 51 186 (43.6) 54 331 (44.7) Sex Women 67 915 (57.8) 69 458 (57.2) Men 49 534 (42.2) 51 965 (42.8) Race American Indian or Alaska Native 806 (0.7) 880 (0.7) Asian 13 311 (11.3) 13 197 (10.9) Black or African American 11 919 (10.1) 11 649 (9.6) Native Hawaiian or other Pacific Islander 905 (0.8) 709 (0.6) White 76 431 (65.1) 79 142 (65.2) Multiracial or other 459 (0.4) 546 (0.4) Unknown or not reported 13 624 (11.6) 15 308 (12.6) Ethnicity Hispanic or Latino 17 754 (15.1) 18 475 (15.2) Not Hispanic or Latino 19 867 (16.9) 19 276 (15.9) Not available 79 834 (68.0) 83 680 (68.9) Modality RG 93 465 (79.6) 98 970 (81.5) CT 494 (0.4) 449 (0.4) MR 23 496 (20) 22 012 (18.1) Charlson Comorbidity Index 0 75 106 (63.9) 77 973 (64.2) 1 20 675 (17.6) 21 193 (17.5) 2 11 451 (9.7) 11 760 (9.7) ≥3 10 223 (8.7) 10 505 (8.7) Finding status None 27 770 (23.6) 27 776 (22.9) LIRE finding without clinically important finding 72 127 (61.4) 77 065 (63.5) Clinically important finding 17 558 (14.9) 16 590 (13.7) ≥1 Opioid prescriptions prior to index 32 225 (27.4) 29 306 (24.1) Primary insurance at index Medicare 44 362 (37.8) 46 479 (38.3) Medicaid or state-subsidized 5546 (4.7) 6510 (5.4) Commercial 65 375 (55.7) 66 368 (54.7) VA 117 (0.1) 131 (0.1) Self-pay 731 (0.6) 570 (0.5) Unknown or not reported 1324 (1.1) 1373 (1.1) Socioeconomic index, mean (SD) 57 (6) 57 (7) Health care professional type MD 105 359 (89.7) 108 165 (89.1) DO 8131 (6.9) 9157 (7.5) Extender, eg, NP, PA 3965 (3.4) 4109 (3.4) (continued) JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 7/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults Table. Baseline Characteristics (continued) No. (%) Characteristic Control (n = 117 455) Intervention (n = 121 431) Health care professional specialty Family medicine 56 795 (48.4) 60 277 (49.6) Internal medicine 59 684 (50.8) 60 158 (49.5) Other 976 (0.8) 996 (0.8) Female health care professional 62 840 (53.5) 62 680 (51.6) Health care professional age, mean (SD), y 49 (9) 49 (9) Abbreviations: CT, computed tomography; DO, doctor of osteopathy; MD, medical doctor; LIRE, Lumbar Imaging with Reporting of Epidemiology; MR, magnetic resonance; NP, nurse practitioner; PA, physician’s assistant; RG, radiograph; VA, Veterans Administration. Does not include 14 patients (<0.1%) with other or unknown gender. Due to the manner in which race and ethnicity are collected at 1 health system (ie, sometimes the concepts are conflated and sometimes Hispanic ethnicity is captured by a single checkbox), it is not possible to reliably distinguish between “not Hispanic” and “did not answer.” Does not include 6810 patients (2.7%) with unknown socioeconomic index. Sites mapped participant addresses to Federal Information Processing System codes at the block-group level using geocoding software. These codes were mapped to socioeconomic indices derived from data available from the 2010 Census Summary File 1 and the American Community Survey, 2007 to 2011, 5-year estimate data. Does not include 424 patients (0.1%) for whom provider age was unknown. Figure 2. Model Results for Spine-Related Relative Value Units (RVUs) at 1 Year Adjusted Adjusted median RVU median RVU Favors Favors Population control intervention Difference (95% CI), % intervention control P value Whole cohort 3.56 3.53 –0.7 (–2.9 to 1.5) .54 All models adjust for health system, clinic size, age Index imaging modality range (ie, 18-39, 40-60, and61 years), sex, imaging RG 3.19 3.19 0.1 (–2.0 to 2.3) modality, Charlson Comorbidity Index category (ie, 0, CT 10.40 7.07 –29.3 (–42.1 to –13.5) .01 1, 2, and3), and health system specific time trends. MR 7.67 7.37 –3.4 (–8.3 to 1.8) Models include hierarchical random effects for clinic Image finding type (intercept and treatment) and primary care Likely CI 9.26 8.83 –4.2 (–9.0 to 0.9) professional (intercept only). P values for subgroup LIRE finding without likely CI 3.60 3.58 –0.4 (–2.6 to 1.9) models (ie, index imaging type and image finding type) Neither finding type 2.35 2.36 0.3 (–2.7 to 3.3) are for Wald tests for effect modification. CI indicates –30 –20 –10 0 10 clinically important, CT, computed tomography; RG, Difference (95% CI), % radiograph; and MR, magnetic resonance. The prespecified subgroup analysis of whether the intervention differentially affected spine- related RVUs by imaging modality revealed that the small number of patients (943 [0.4%]) of patients who had computed tomography as the index imaging had markedly lower subsequent median RVUs if exposed to the intervention (mean difference, −29.3%; 95% CI,−42.1% to −13.5%). The nearly 20% of patients (45 508 [19.1%]) who had magnetic resonance imaging had lower subsequent RVUs in the intervention group (difference, −3.4%; 95% CI, −8.3% to 1.8%), although this was not statistically significant (Figure 2). The second prespecified subgroup analysis that examined whether image finding type differentially affected spine-related RVUs revealed no differences in subsequent median RVUs in the intervention compared with the control group (Figure 2). In a post hoc subgroup analysis, the adjusted proportion of control patients without a prior opioid prescription who received an opioid prescription from a LIRE PCP within 1 year following index imaging was 25% compared with 72% for control patients with a prior opioid prescription. However, there was no intervention effect modification by prior opioid prescription status (test for effect modification, P = .58) (Figure 3). When we included prescriptions from non-LIRE PCPs who were not exposed to the intervention in the 1-year opioid outcome, the intervention effect was attenuated JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 8/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults (adjusted opioid proportion, 47.1% intervention vs 47.5% control; OR, 0.98; 95% CI, 0.94-1.02; P = .27) (Figure 3). Discussion The LIRE intervention did not reduce subsequent spine-related RVUs for the population as a whole. However, patients in the intervention group were less likely (OR, 0.95; 95% CI, 0.90-0.99; P =.02) to receive a subsequent opioid prescription compared with patients not receiving the intervention. The intervention also reduced subsequent spine-related RVUs for the small proportion of patients with CT as the index imaging. Pragmatic trials must be simple to implement and the populations relatively unselected. Thus, a 25-27 negative primary result is not unusual. This suggests the likely importance of heterogeneous intervention effects, prespecified subgroup analyses, and prespecified secondary outcomes. An explanation for the differential effect by imaging modality is that patients undergoing CT for their index imaging were more likely to receive back pain interventions than patients receiving other modalities, and thus, the intervention was more effective at reducing subsequent interventions in patients who were most likely to receive those interventions in the first place (eAppendix 7 in Supplement 1). Our finding of no greater subsequent emergency department visits and deaths in the intervention group provides reassurance that the intervention did not cause deleterious undertreatment. Given the climate of overdiagnosis and overtreatment of back pain in the United Figure 3. Model Results for Opioid Prescriptions Within 12 months Adjusted Adjusted opioid rate opioid rate Favors Favors Population control, % intervention, % OR (95% CI) intervention control P value Whole cohort 37.0 36.2 0.95 (0.91-1.00) .04 Prior opioid prescription No 24.8 24.1 0.96 (0.91-1.01) NA Yes 72.2 70.8 0.93 (0.85-1.02) NA Alternative time definition Spline 1 knot 37.0 36.2 0.95 (0.91-1.00) .04 Spline 2 knots 37.0 36.2 0.95 (0.91-1.00) .03 Alternative source definition Prescription from any source 47.5 47.1 0.98 (0.94-1.02) .27 Prescription from non-LIRE source 23.4 23.4 1.00 (0.96-1.04) .96 –30 –20 –10 0 10 OR (95% CI) All models adjust for health system, clinic size, age range (ie, 18-39, 40-60, and61 the 120 days prior to index imaging. A Lumbar Imaging with Reporting of Epidemiology years), sex, imaging modality, Charlson Comorbidity Index category (ie, 0, 1, 2, and3), (LIRE) source is any health care professional who ordered an index lumbar spine image prior opioid use, and health system specific time trends. Models include hierarchical for 1 or more participants in the LIRE trial. It need not be the same individual who ordered random effects for clinic (intercept and treatment) and primary care professional the patient’s index image. A non-LIRE source is any other health care professional. Any (intercept only). Prior opioid prescription is defined as having 1 or more prescriptions in source includes both LIRE and non-LIRE clinicians. NA indicates not applicable. Figure 4. Safety Outcomes Adjusted rate Adjusted rate Favors Favors Outcome control, % intervention, % OR (95% CI) intervention control ED visit within 90 days 11.3 11.1 0.98 (0.94-1.01) 6-month mortality 11.3 0.81 1.03 (0.88-1.20) 0.90 0.95 1.00 1.05 1.10 OR (95% CI) All models adjust for health system, clinic size, age range (ie, 18-39, 40-60, and61 visit model includes hierarchical random effects for clinic (intercept and treatment) and years), sex, imaging modality, Charlson Comorbidity Index category (ie, 0, 1, 2, and3), primary care professional (intercept only). The mortality model uses general estimating seasonality, and health system specific time trends. The emergency department (ED) equations with clustering on clinic. JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 9/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults States, undertreatment may be less likely to occur in the United States than elsewhere. Our intervention provided an opportunity to increase the knowledge of patients and health care professionals. Because we did not detect any harm of the intervention and we did detect a possible benefit, including the intervention should safely allow patients and health care professionals to make better informed decisions. Finally, our primary null result may have been different if we had studied different health systems. For example, if we had enrolled clinics with higher baseline utilization of tests for back pain patients, we may have found a positive result. Limitations This study has limitations. Opioid prescribing decreased in the United States during our study. Although we made multiple efforts to account for this potential confounding in our modeling, residual confounding may exist. Because we did not collect patient-reported outcomes, we cannot comment on outcomes such as functional status, pain, or psychosocial functioning. The decision not to collect patient-reported data was deliberate, based on the recognition that it could jeopardize the feasibility of this large pragmatic trial of more than 250 000 patients. We also did not capture patient care not included in the EMRs. However, we found similar results to those of our primary analysis when we examined subsequent RVUs from patients less likely to seek outside care (eAppendix 5 in Supplement 1). Previous studies have shown high degrees of accuracy when EMR data were validated by manual medical record reviews. All of our participating health systems were integrated delivery systems and nonprofit. There is evidence that nonprofit hospitals may be less responsive to the type of intervention that we tested than for-profit hospitals. However, this conservative bias emphasizes the robustness of the positive impact that we observed with respect to opioid prescribing. Our findings may also not be generalizable to systems having greater restrictions on advanced imaging. We do not know the indication for imaging, including whether the patient had a red flag, so we cannot comment on the appropriateness. Conclusions In this study, adding benchmark prevalence information for spine imaging findings did not reduce subsequent spine-related RVUs, but it slightly reduced the likelihood of subsequent opioid prescribing, an important prespecified secondary outcome. Reporting benchmark information is a fundamental change to the imaging reporting paradigm that may be relevant for other conditions and could easily be applied to other diagnostic tests (eg, other imaging tests, genetic testing). Finally, unmeasured benefits of the intervention may result from patients and health care professionals having a better understanding of the clinical meaning of imaging findings. ARTICLE INFORMATION Accepted for Publication: June 12, 2020. Published: September 4, 2020. doi:10.1001/jamanetworkopen.2020.15713 Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Jarvik JG et al. JAMA Network Open. Corresponding Author: Jeffrey G. Jarvik, MD, MPH, Department of Radiology, University of Washington, 325 Ninth Ave, Box 359728, Seattle, WA 98104-2499 (jarvikj@uw.edu). Author Affiliations: Department of Radiology, University of Washington, Seattle (Jarvik, James, Gold, Bresnahan); Department of Neurological Surgery, University of Washington, Seattle (Jarvik); Department of Health Services, University of Washington, Seattle (Jarvik, Kessler); Comparative Effectiveness, Cost, and Outcomes Research Center, University of Washington, Seattle (Jarvik, James, Gold, Rundell, Friedly, Turner, Bresnahan); Department JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 10/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults of Biostatistics, University of Washington, Seattle (Meier, Tan, Comstock, Heagerty); Center for Biomedical Statistics, University of Washington, Seattle (Meier, Tan, Comstock, Heagerty); Flatiron Health, New York, New York (Tan); Rehabilitation Care Services, VA Puget Sound Health Care System, Seattle, Washington (Suri); Department of Rehabilitation Medicine, University of Washington, Seattle (Suri, Rundell, Friedly, Turner); Department of Radiology, Mayo Clinic, Rochester, Minnesota (Kallmes, Luetmer); Kaiser Permanente Washington, Seattle (Cherkin, Sherman); Departments of Family Medicine and Internal Medicine, Oregon Health and Science University, Portland (Deyo); Department of Radiology, Henry Ford Hospital, Detroit, Michigan (Halabi, Griffith); Department of Radiology, Stanford University School of Medicine, Palo Alto, California (Halabi); Division of Research, Kaiser Permanente Northern California, Oakland, California (Avins); Surgical Outcomes Research Center, University of Washington, Seattle (Lavallee); Department of Psychiatry and Behavioral Sciences, University of Washington, Seattle (Stephens, Turner). Author Contributions: Dr Jarvik had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: Jarvik, Kallmes, Deyo, Comstock, Turner, Bresnahan, Heagerty. Acquisition, analysis, or interpretation of data: Jarvik, Meier, James, Gold, Tan, Kessler, Suri, Cherkin, Sherman, Halabi, Comstock, Luetmer, Avins, Rundell, Griffith, Friedly, Lavallee, Stephens, Turner, Bresnahan, Heagerty. Drafting of the manuscript: Jarvik, Meier, Gold, Suri, Lavallee, Stephens, Turner, Heagerty. Critical revision of the manuscript for important intellectual content: Jarvik, Meier, James, Gold, Tan, Kessler, Suri, Kallmes, Cherkin, Deyo, Sherman, Halabi, Comstock, Luetmer, Avins, Rundell, Griffith, Friedly, Turner, Bresnahan, Heagerty. Statistical analysis: Jarvik, Meier, Gold, Tan, Kessler, Comstock, Stephens, Heagerty. Obtained funding: Jarvik, Comstock, Avins, Heagerty. Administrative, technical, or material support: Jarvik, James, Tan, Suri, Cherkin, Deyo, Halabi, Luetmer, Lavallee, Stephens, Heagerty. Supervision: Jarvik, James, Kessler, Kallmes, Luetmer, Lavallee, Heagerty. Conflict of Interest Disclosures: Dr Jarvik reported receiving royalties from Springer Publishing and Wolters/ Kluwer/UpToDate and receiving travel reimbursement from GE–Association of University Radiologists outside the submitted work. Dr Tan reported being employed by Flatiron Health and owning stock in Roche outside the submitted work. Dr Deyo reported having an endowed professorship funded by Kaiser Permanente and receiving personal fees from UpToDate outside the submitted work. Dr Rundell reported receiving grants from the Institute of Translational Health Sciences, the Center for Large Data Research and Data Sharing in Rehabilitation, the Center on Health Services Training and Research, and the Scan Design Foundation; having a contract with the Agency for Healthcare Research and Quality; and receiving personal fees from the Department of Defense and ATI Physical Therapy outside the submitted work. Dr Friedly reported receiving grants from the Department of Defense and salary support from the American Academy of Physical Medicine and Rehabilitation for serving as editor in chief outside the submitted work. No other disclosures were reported. Funding/Support: This work was supported within the National Institutes of Health (NIH) Health Care Systems Re- search Collaboratory by the NIH Common Fund through cooperative agreement U24AT009676 from the Office of Strategic Coordination within the Office of the NIH Director and cooperative agreements UH2AT007766 and UH3AR066795 from the National Institute of Arthritis and Musculoskeletal and Skin Diseases. It was also supported by the University of Washington Clinical Learning, Evidence, And Research Center for Musculoskeletal Disorders funded by NIH/National Institute of Arthritis and Musculoskeletal and Skin Diseases grant P30AR072572. Role of the Funder/Sponsor: The NIH Healthcare Systems Collaboratory provided oversight of the design and conduct of the study, the collection, management, analysis, and interpretation of the data. They did not participate in the preparation, review, or approval of the manuscript, nor the decision to submit the manuscript for publication. Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Data Sharing Statement: See Supplement 3. REFERENCES 1. Jarvik JJ, Hollingworth W, Heagerty P, Haynor DR, Deyo RA. The Longitudinal Assessment of Imaging and Disability of the Back (LAIDBack) Study: baseline data. Spine (Phila Pa 1976). 2001;26(10):1158-1166. doi:10.1097/ 00007632-200105150-00014 2. Brinjikji W, Luetmer PH, Comstock B, et al. Systematic literature review of imaging features of spinal degeneration in asymptomatic populations. AJNR Am J Neuroradiol. 2015;36(4):811-816. JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 11/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults 3. Jarvik JG, Gold LS, Comstock BA, et al. Association of early imaging for back pain with clinical outcomes in older adults. JAMA. 2015;313(11):1143-1153. doi:10.1001/jama.2015.1871 4. Graves JM, Fulton-Kehoe D, Jarvik JG, Franklin GM. Health care utilization and costs associated with adherence to clinical practice guidelines for early magnetic resonance imaging among workers with acute occupational low back pain. Health Serv Res. 2014;49(2):645-665. doi:10.1111/1475-6773.12098 5. Roland M, van Tulder M. Should radiologists change the way they report plain radiography of the spine? Lancet. 1998;352(9123):229-230. doi:10.1016/S0140-6736(97)11499-4 6. McCullough BJ, Johnson GR, Martin BI, Jarvik JG. Lumbar MR imaging and reporting epidemiology: do epidemiologic data in reports affect clinical management? Radiology. 2012;262(3):941-946. doi:10.1148/radiol. 7. Fried JG, Andrew AS, Ring NY, Pastel DA. Changes in primary care health care utilization after inclusion of epidemiologic data in lumbar spine MR imaging reports for uncomplicated low back pain. Radiology. 2018;287(2): 563-569. doi:10.1148/radiol.2017170722 8. Karran EL, Yau YH, Hillier SL, Moseley GL. The reassuring potential of spinal imaging results: development and testing of a brief, psycho-education intervention for patients attending secondary care. Eur Spine J. 2018;27(1): 101-108. doi:10.1007/s00586-017-5389-8 9. Medalian Y, Moseley GL, Karran EL. An online investigation into the impact of adding epidemiological information to imaging reports for low back pain. Scand J Pain. 2019;19(3):629-633. doi:10.1515/sjpain-2019-0023 10. Hsiao WC, Braun P, Dunn D, Becker ER. Resource-based relative values: an overview. JAMA. 1988;260(16): 2347-2353. doi:10.1001/jama.1988.03410160021004 11. Hsiao WC, Braun P, Yntema D, Becker ER. Estimating physicians’ work for a resource-based relative-value scale. N Engl J Med. 1988;319(13):835-841. doi:10.1056/NEJM198809293191305 12. American Medical Association. CPT (Current Procedural Terminology). Accessed August 4, 2020. https://www. ama-assn.org/amaone/cpt-current-procedural-terminology 13. Jarvik JG, Comstock BA, James KT, et al. Lumbar Imaging With Reporting Of Epidemiology (LIRE)—protocol for a pragmatic cluster randomized trial. Contemp Clin Trials. 2015;45(Pt B):157-163. doi:10.1016/j.cct.2015.10.003 14. Johnson KE, Neta G, Dember LM, et al. Use of PRECIS ratings in the National Institutes of Health (NIH) Health Care Systems Research Collaboratory. Trials. 2016;17:32. doi:10.1186/s13063-016-1158-y 15. Brinjikji W, Diehn FE, Jarvik JG, et al. MRI findings of disc degeneration are more prevalent in adults with low back pain than in asymptomatic controls: a systematic review and meta-analysis. AJNR Am J Neuroradiol. 2015;36 (12):2394-2399. doi:10.3174/ajnr.A4498 16. World Health Organization. International Classification of Diseases, Ninth Revision (ICD-9). World Health Organization; 1977. 17. World Health Organization. International Statistical Classification of Diseases, Tenth Revision (ICD-10). World Health Organization; 1992. 18. Martin B, Mirza SK, Lurie JD, Tosteson ANA, Deyo RA. Validation of an administrative coding algorithm to identify back-related degenerative diagnoses. Paper presented at: International Society for the Study of the Lumbar Spine (ISSLS); May 14, 2013; Scottsdale, AZ. 19. Martin BI, Mirza SK, Franklin GM, Lurie JD, MacKenzie TA, Deyo RA. Hospital and surgeon variation in complications and repeat surgery following incident lumbar fusion for common degenerative diagnoses. Health Serv Res. 2013;48(1):1-25. doi:10.1111/j.1475-6773.2012.01434.x 20. Tan WK, Hassanpour S, Heagerty PJ, et al. Comparison of natural language processing rules-based and machine-learning systems to identify lumbar spine imaging findings related to low back pain. Acad Radiol. 2018;25 (11):1422-1432. doi:10.1016/j.acra.2018.03.008 21. Jarvik JG, Hollingworth W, Heagerty PJ, Haynor DR, Boyko EJ, Deyo RA. Three-year incidence of low back pain in an initially asymptomatic cohort: clinical and imaging risk factors. Spine (Phila Pa 1976). 2005;30(13):1541-1548. doi:10.1097/01.brs.0000167536.60002.87 22. Turner JA, Shortreed SM, Saunders KW, LeResche L, Von Korff M. Association of levels of opioid use with pain and activity interference among patients initiating chronic opioid therapy: a longitudinal study. Pain. 2016;157(4): 849-857. doi:10.1097/j.pain.0000000000000452 23. Charlson M, Szatrowski TP, Peterson J, Gold J. Validation of a combined comorbidity index. J Clin Epidemiol. 1994;47(11):1245-1251. doi:10.1016/0895-4356(94)90129-5 JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 12/13 JAMA Network Open | Imaging Effect of Benchmark Prevalence Data in Spine Image Reports on Health Care Utilization Among Adults 24. Quan H, Li B, Couris CM, et al. Updating and validating the Charlson comorbidity index and score for risk adjustment in hospital discharge abstracts using data from 6 countries. Am J Epidemiol. 2011;173(6):676-682. doi: 10.1093/aje/kwq433 25. Dember LM, Lacson E Jr, Brunelli SM, et al. The TiME trial: a fully embedded, cluster-randomized, pragmatic trial of hemodialysis session duration. J Am Soc Nephrol. 2019;30(5):890-903. doi:10.1681/ASN.2018090945 26. Huang SS, Septimus E, Kleinman K, et al; ABATE Infection Trial Team. Chlorhexidine versus routine bathing to prevent multidrug-resistant organisms and all-cause bloodstream infections in general medical and surgical units (ABATE Infection trial): a cluster-randomised trial. Lancet. 2019;393(10177):1205-1215. doi:10.1016/S0140-6736 (18)32593-5 27. Coronado GD, Petrik AF, Vollmer WM, et al. Effectiveness of a mailed colorectal cancer screening outreach program in community health clinics: the STOP CRC cluster randomized clinical trial. JAMA Intern Med. 2018;178 (9):1174-1181. doi:10.1001/jamainternmed.2018.3629 28. Pezalla EJ, Rosen D, Erensen JG, Haddox JD, Mayne TJ. Secular trends in opioid prescribing in the USA. J Pain Res. 2017;10:383-387. doi:10.2147/JPR.S129553 29. Patel NK, Moses RA, Martin BI, Lurie JD, Mirza SK. Validation of using claims data to measure safety of lumbar fusion surgery. Spine (Phila Pa 1976). 2017;42(9):682-691. doi:10.1097/BRS.0000000000001879 30. Horwitz JR. Making profits and providing care: comparing nonprofit, for-profit, and government hospitals. Health Aff (Millwood). 2005;24(3):790-801. doi:10.1377/hlthaff.24.3.790 SUPPLEMENT 1. eAppendix 1. Outcome Sources and Definitions eAppendix 2. PRECIS-2 Diagram eAppendix 3. Intervention Text eAppendix 4. Imaging Findings eAppendix 5. Sensitivity Analyses for Spine-Related Relative Value Units (RVUs) Outcome eAppendix 6. Opioid Prescription Within 90 Days eAppendix 7. Patient Characteristics and Outcomes by Index Modality eReferences. SUPPLEMENT 2. Trial Protocol and Statistical Analysis Plan SUPPLEMENT 3. Data Sharing Statement JAMA Network Open. 2020;3(9):e2015713. doi:10.1001/jamanetworkopen.2020.15713 (Reprinted) September 4, 2020 13/13 Lumbar Imaging Reporting with Epidemiology (LIRE) Study Protocol Version 3.0 ! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Table!of!Contents! 1.0 Project!Summary!…………………………………………………………………………………………………………………4! ! 1.1!!!!!!!Project!Organization! Schema'1:'Overall'Organization'Chart' Schema'2:'Site'Organization'Chart' Personnel'Directory' 1.2!! Interaction!with!Collaboratory!Coordinating!Center! 2.0 Background!and!Rationale………………………………………………………………………………………..…………13! Figure'1:'MRI 'of'spine'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''' Figure'2:'Epidemiologic'Statement'Included'In'Lumbar'Spine'MR'Imaging'Reports '''''''''''' Table'1:'Outcomes'Of'Patients'Whose'Imaging'Did'And'Did'Not'Include'A'Statement' Containing'Epidemiological'Benchmarks' 3.0! Specific!Aims…………………………………………………………………………………………………………………..…16! 4.0! Study!Details……………………………………………………………………………………………………………………..17! 4.1!Eligibility!Criteria! 4.2!Consent!procedure! 4.3!Inclusion!and!Exclusion!Criteria! 5.0! Research!Design!and!Methods……………………………………………………………………………………..……18! 5.1!Clinic/Practitioner/Patient!Identification! ! 5.2!Randomization! Figure'3:'Proposed'Randomization'Schedule' 5.3!Clinic/Patient!Enrollment! Table'2:'Lumbar'Examinations'to'be'Included'in'Pragmatic'Trial' 5.4!Data!Collection! 5.5!Aims!for!UH2!Phase! Table'3:'AgeMspecific'Rates'of'Lumbar'Spine'Imaging'Findings'' Figure'4:'Comparison'of'Parallel,'Crossover'and'Stepped'Wedge'Designs'' 5.6!Working!Groups! Table'4:'Milestones'for'UH2'Planning'Year'Needed'to'Transition'to'UH3'Impleme ntation' 5.7!Aims!for!UH3!Phase! Table'5:'Sample'of'RVUs'and'CMS Mbased'payment'amounts'for'lumbar'imaging' Table'6:''Milestones'for'UH3'(Implementation'Phase) M'Years'2M5' Table'7:'Timeline'for'the'UH3'Phase'(Years'2M5)' 2013409426/kj! 2! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! 6.0! Statistical!Considerations…………………………………………………………………………………………………31! 7.0! Human!Subjects!………………………………………………………………………………………………………………31! 7.1!Human!Subjects!Involvement!and!Characteristics!! 7.2!Research!Data!!! 7.3!Potential!risks! 7.4!Adequacy!of!Protection!Against!Risks! 7.5!Potential!Benefits!of!the!Proposed!Research!to!the!Subjects!! 7.6!Importance!o!the!Knowledge!to!be!Gained! 8.0! Data!and!Safety!Monitoring……………………………………………………………………………………….……33! 9.0! References………………………………………………………………………………………………………………………34! 10.0! Appendices! ! ' Appendix'A:'Article'Extraction'form'(Working'Group'1)' Appendix'B:'Literature'Search'and'Articles'used'in'Intervention'text'(Working'Group'1)' Appendix'C:'Pilot'Implementation'Site'Checklist'(Working'Group'2)' Appendix'D:'Literature'Search'for'RVUMbased'Assessment'(Working'Group'3)' ! ! 2013409426/kj! 3! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! 1.!Project!Summary! Low!back!pain,!an!Institute!of!Medicine!priority!condition!for!comparative!effectiveness!research,!is!of! major! public! health! importance.! It! is! one! of! the! most! common! reasons! for! physician! visits! and! an! important!cause!of!functional!limitation!and!disability.!Imaging!is!frequently!performed!as!part!of!the! diagnostic!evaluation!and!is!an!important!contributor!to!the!cost!of!back!pain!care,!which!totaled!more! than!$86!billion!in!2005.!It!is!well!known!that,!even!without!back!pain,!magnetic!resonance!(MR)!imaging! of! the! lumbar! spine! frequently! reveals! findings! such! as! disc! desiccation! or! bulging.! Patients! and! their! providers! may! attribute! greater! importance! to! these! findings,! which! are! often! age\related,! than! they! should,!because!they!do!not!have!an!appropriate!frame!of!reference!in!which!to!interpret!the!findings.! These!“incidental”!findings!may!initiate!a!cascade!of!events!leading!possibly!even!to!surgery,!without! improving!patient!outcomes.!! The! overall! goal! of! the! Lumbar! Image! Reporting! with! Epidemiology! (LIRE)! trial,! is! to! perform! a! large,! pragmatic,!randomized!controlled!trial!to!determine!the!effectiveness!of!a!simple,!inexpensive!and!easy! to!deploy!intervention!–!of!inserting!epidemiological!benchmarks!into!lumbar!spine!imaging!reports!–!at! reducing! subsequent! tests! and! treatments.! The! long\term! public! health! significance! is! that! our! intervention!has!the!potential!to!substantially!reduce!unnecessary!and!expensive!care!not!only!for!back! pain,!but!also!for!a!wide!range!of!other!conditions,!since!it!could!easily!be!applied!to!other!diagnostic! tests! (e.g.! other! imaging! tests,! laboratory! tests,! genetic! testing).! If! our! study! is! positive,! adding! epidemiologic! benchmarks! to! diagnostic! test! reporting! could! become! the! dominant! paradigm! for! communicating!all!diagnostic!information.! We! propose! an! efficient,! novel,! cluster! randomized! design! referred! to! as! a! “stepped! wedge”! design,! permitting!longitudinal!comparisons!while!controlling!for!temporal!trends.!We!plan!to!passively!collect! primary! outcome! measures! of! healthcare! utilization! both! pre\! and! post\intervention,! using! robust! electronic!medical!records!at!the!participating!sites.!We!hypothesize!that!for!patients!of!primary!care! providers,!inserting!epidemiological!benchmarks!in!lumbar!spine!imaging!reports!will!reduce!subsequent! diagnostic!and!therapeutic!interventions,!including!MR!and!CT,!opioid!prescriptions,!spinal!injections!and! surgery.! The! rationale! is! that! the! epidemiologic! data! may! provide! a! context! for! both! physicians! and! patients!to!better!interpret!imaging!findings.!! The!University!of!Washington!will!serve!as!the!over!data!coordination!center!(DCC)!for!the!project!that! will!take!place!at!four!performance!sites:!!Group!Health!Cooperative,!Kaiser!Permanente!of!Northern! California,!Henry!Ford!Health!Systems,!and!Mayo!Clinic!Health!Systems.!!The!role!of!the!DCC!is!to! coordinate!study!efforts,!thus!overseeing!the!technical!implementation!of!the!intervention!across!the! sites.!!The!DCC!will!oversee!the!transfer!and!storage!of!study!data,!provide!biostatistical!and!analysis! expertise,!as!well!as!lead!manuscript!writing!efforts.!!Each!performance!site!is!tasked!with!the! implementation!of!the!randomized!intervention!at!the!primary!care!clinics!within!their!system,!as!well!as! the!technical!abstraction!(and!transfer)!of!electronic!medical!record!(EMR)!and!administrative!data!from! their!system!to!the!DCC.!!See!Schemas!1!and!2!below!for!organizational!overview.! 2013409426/kj! 4! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! 1.1!Project!Organization:!! Schema!1:!Overall!Organization!Chart! Schema!2:!Site!Organization!Chart! 2013409426/kj! 5! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Personnel!Directory! University!of!Washington ! Data!Coordination!Center! Jeffrey!G.!Jarvik!MD,!MPH! Principal'Investigator' Professor,!Radiology!and!Neurological!Surgery! Adjunct!Professor,!Health!Services! Director,!Comparative!Effectiveness,!Cost!and!Outcomes!Research!Center!(CECORC)! University!of!Washington! Box!359455! 4333!Brooklyn!Ave!NE! Seattle,!WA!98195! Tel:!(206)!616\2941! ! Fax:!(206)!616\3135! Email:!jarvikj@uw.edu! Patrick!Heagerty!PhD! CoMInvestigator' Professor,!Biostatistics! Director,!Center!for!Biomedical!Statistics! University!of!Washington! Box!357232! Seattle,!WA!!98104! Tel:!(206)!616\2720! ! ! ! ! Fax:!(206)!543\3286! Email:!heagerty@uw.edu! Judith!Turner!PhD! CoMInvestigator' Professor,!Psychiatry!and!Behavioral!Sciences,!Rehabilitation!Medicine! Adjunct!Professor,!Anesthesiology!and!Pain!Medicine! University!of!Washington! Box!356560! Seattle,!WA!!98195! Tel:!(206)!543\3997! ! Fax:!(206)!685\1139! Email:!jturner@uw.edu! Brian!Bresnahan!PhD! Health'Economist' Research!Assistant!Professor,!Radiology! University!of!Washington! Box!359736! 2013409426/kj! 6! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! 325!Ninth!Ave.! Seattle,!WA!!98104! Tel:!(206)!744\1805! ! ! ! ! Fax:!(206)!744\9962! Email:!bres@uw.edu! Bryan!Comstock,!MS! Biostatistician' Operations!Director,!Center!for!Biomedical!Statistics! University!of!Washington! Box!359461! Seattle,!WA!!98195! Tel:!(206)!543\1882! ! ! Fax:!(206)!543\5881! Email:!bac4@uw.edu! Kathryn!James!PAWC,!MPH! Overall'Project'Director' Administrative!&!Operations!Director,!Comparative!Effectiveness,!Cost!&!Outcomes!Research! Center!(CECORC)! Box!359455! 4333!Brooklyn!Ave!NE! Seattle,!WA!98195! Tel:!(206)!221\7444! ! Fax:!(206)!616\3135! Email:!jarvikj@uw.edu! UW!Grant!Contact :!My\Hanh!Tong!\!mhtt@uw.edu! Oregon!Health!Sciences!University! Subcontracted!Site!for!Consultant! Richard!A.!Deyo!MD,!MPH! CoMInvestigator' Kaiser!Permanente!Professor!of!Evidence\Based!Family!Medicine! Director,!KL2!Multidisciplinary!Clinical!Research!Career!Development!Program! Director,!OCTRI!Community!and!Practice\based!Research!Program! Departments!of!Family!Medicine!and!Internal!Medicine! Oregon!Health!and!Science!University! 3181!SW!Sam!Jackson!Park!Rd.! Portland,!Oregon!97239\3098! Tel:!(503)!494\1694! ! ! ! ! Fax:!(503)!494\2746! Email:!deyor@ohsu.edu! OHSU!Grant!Contact:!Connie!YuW!yuco@ohsu.edu! 2013409426/kj! 7! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! University!of!CaliforniaW!Davis! Subcontracted!Site!for!Consultant! ! Nicolas!Anderson,!PhD! ! CoMInvestigator' ' Nick!Anderson,!Ph.D.! Robert!D.!Cardiff!Professor!of!Informatics! Director!of!Informatics!Research! Department!of!Pathology!and!Laboratory!Medicine! University!of!California,!Davis! Tel:!(916)!703!6976! ! UC!Davis!Grant!Contract:!Kate!Marie\!kate.marie@ucdmc.ucdavis.edu! Group!Health!Cooperative! Performance!Site! Dan!Cherkin,!PhD! Site'Principal'Investigator' Group!Health!Research!Institute! 1730!Minor!Ave,!Ste!1600! Seattle,!WA!!98101! Email:!!cherkin.d@ghc.org! Heidi!Berthoud!MPH! Project'Manager' Group!Health!Research!Institute! 1730!Minor!Ave,!Ste!1600! Seattle,!WA!!98101' Email:!berthoud.h@ghc.org! GHC!Grant!Contact:!David!Hawkes\!hawkes.d@ghc.org! Henry!Ford!Health!System! Performance!Site! Safwan!Halabi!MD! Site'Principal'Investigator' Associate!Professor,!Radiology! Director,!Imaging!Informatics! 2799!West!Grand!Boulevard!! Detroit,!MI!48202' Email:!safwanh@rad.hfhs.edu! 2013409426/kj! 8! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! David!Nerenz!PhD! Site'CoMInvestigator! Director,!Outcomes!Research! ! Neuroscience!Institute! Department!of!Neurosurgery! Henry!Ford!Hospital! 2799!West!Grand!Boulevard,!K\11,!W\1136! Detroit,!MI!!48202\2689! Tel:!(313)!916\5454! ! ! ! ! Fax:!(313)!874\7137! Email:!dnerenz1@hfhs.org! Brooke!Wessman! Project'Manager' Tel:!(313)!916\0829' Email:!brookew@rad.hfhs.org! HF!Grant!Contact:!Kim!Sadlocha!\ksadloc1@hfhs.org! Kaiser!Permanente!of!Northern!California!! Performance!Site! Andrew!Avins!MD,!MPH!!! Site'Principal'Investigator! Clinical!Professor,!Medicine! Adjunct!Professor,!Epidemiology!and!Biostatistics! Kaiser!Permanente!Division!of!Research! 2000!Broadway! Oakland,!CA!!94612\!2304! Office:!(510)!891\3557! ! ! ! Fax:!(510)!891\3606!! ! ! Cell:!(415)!302\5986! Email:!andrew.avins@ucsf.edu! Luisa!M.!Hamilton! Project'Manager ,!KPNC!Division!of!Research! 2000!Broadway! Oakland,!CA!!94612!\2304! Tel:!(510)!891\3712! ! ! ! Fax:!(510)!891\3802! Email:!luisa.M.Hamilton@kp.org! KPNC!Grant!Contact:!Anna!DelaneyWHeathW!delaney@kp.org! 2013409426/kj! 9! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Mayo!Clinic!Health!System! Performance!Site! David!F.!Kallmes!MD! Site'Principal'Investigator' Professor,!Radiology! Adjunct!Professor,!Neurological!Surgery! 200!First!St.!SW! Rochester,!MN!55905! Tel:!507\266\3350! Email:!kallmes.david@mayo.edu! Administrative!Contact:!! Kimberly!Collins:!!Collins.kimberly@mayo.edu! Patrick!Leutmer!MD! Site'CoMInvestigator' Assistant!Professor,!Radiology! 200!First!St.!SW! Rochester,!MN!55905! Tel:!507\284\2097! Email:!leutmer.patrick@mayo.edu! Jyotishman!Pathak!PhD! Site'IT'Consultant' Associate!Professor,'Biomedical!Statistics!and!Informatics' 200!First!St.!SW! Rochester,!MN!55905! Tel:!507\284\5541! Fax:!507\284\0460' Email:!!pathak.jyotishman@mayo.edu! Administrative!Contact:! Stacy!Tapp:!!tapp.stacy2@mayo.edu! Beth!Connelly! Project'Manager,'' Associate!Clinical!Research!Coordinator,!Department!of!Radiology! Tel:!507\538\3928! Pager:!507\293\4510! Email:!connelly.beth@mayo.edu! 2013409426/kj! 10! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Kristina!Schmidtknecht!! Site'IRB'Contact' Protocol!Development!Coordinator,!!Department!of!Radiology!!! Tel:!507!266\2082!! ! ! ! ! ! Fax:!507\284\8249!!! Pager:!507!538\2477!!! Email:!!!schmidtknecht.kristina@mayo.edu! Mayo!Grant!Contact:!Tracey!AndersonW!anderson.tracey@mayo.edu! 2013409426/kj! 11! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Sponsor!Information:! National!Institutes!of!Health!(NIH)!! Funding!Award:!1UH2AT007766\01! !! Health!Care!Systems!Research!Collaboratory:! UH2/UH3!mechanism!with!UH2!as!a!planning!year!and!UH3!a!separate!award!for!Yr!2\5!awarded!after! competitive!review!based!on!progress!against!UH2!milestones.!! NIH!press!release!regarding!award:!http://nccam.nih.gov/news/2012/092512!! Collaboratory!Website:!!www.nihcollaboratory.org! !! Timeline:! Budget!Period:!!09/30/2012!\!12/31/2013!! Project!Period:!!01/01/2014!\!12/31/2017!! !!! Participating!Institutions:! Data!Coordinating!Center!(DCC)!and!Prime!Awardee:!!! University!of!Washington\!Seattle,!WA!! Principal!Investigator:!Jeffrey!G.!Jarvik,!MD,!MPH!! Performance!Sites:! Group!Health!Cooperative!(GHC)!and!Group!Health!Research!Institute:!!Site!PI:!Dan!Cherkin,!PhD!! Henry!Ford!Health!System!(HFHS)!Site!PI:!Safwan!Halabi,!MD!! Kaiser!Permanente!of!Northern!California!(KPNC)!Site!PI:!Andy!Avins,!MD,!MPH!! Mayo!Clinic!Health!System!(MCHS)!Site!PI:!!David!Kallmes,!MD!! Subcontracted!Sites:! Oregon!Health!Sciences!University!(OHSU):!Rick!Deyo,!MD,!MPH! University!of!California\!Davis!(UCD):!Nicholas!Anderson,!PhD!(subcontract!in!UH2!phase!only,!contractor! in!UH3)! 1.2!LIRE!Interaction!with!Collaboratory!Coordinating!Center!at!Duke!Clinical!Research!Institute:! The!Collaboratory!Coordinating!Center!has!several!“Cores”!aimed!at!organizing!topic\specific!working! groups!across!the!seven!demonstration!projects.!!LIRE!is!contributing!to!the!Collaboratory!by!assigning! key!team!members!to!these!Cores!in!the!following!ways:! ¥ Electronic!health!records\!Anderson,!Jarvik,!Comstock,!and!James! ¥ Provider!Health!Systems!Interactions\!Jarvik!and!James! ¥ Regulatory/Ethics\!James! ¥ Biostatistics/Study!design\!Heagerty!and!Comstock! ¥ Stakeholder!engagement\!Jarvik!and!James! ¥ Pheontype!and!Data!Standards\!Anderson! 2013409426/kj! 12! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! 2.!Background!and!Rationale! Summary'of rationale:!A!common!problem!with!many!diagnostic!tests!is!the!discovery!of!incidental! findings!unrelated!to!patient!symptoms!or!complaints.!Such!findings!can!lead!to!wasteful!subsequent! testing!and!intervention,!sometimes!with!avoidable!complications.!Our!overall!goal!is!to!test!a!strategy! for!mitigating!these!“cascade!effects”!of!incidental!findings.!We!focus!on!the!example!of!lumbar!spine! imaging,!where!incidental!findings!are!extremely! common.!We!propose!a!pragmatic!randomized! trial!of!the!strategy!of!inserting!epidemiological! evidence!into!routine!spine!imaging!reports. If!the!study!is!positive,!the!method!is likely!to!be! generalizable!to!many!other!conditions!and!to! other!kinds!of!testing!(eg,!laboratory!tests).!!So! while!back!pain,!especially!the!back!pain!that! primary!care!providers!see!and!treat,!is!incredibly! important,!our!project!can!also!be!viewed!as!a! "proof!of!concept"!study!that!could!open!the! doors!to!many!similar!interventions.!!Moreover,! the!potential!cost/effectiveness!of!this! intervention,!if!successful,!is!enormous.!!The!cost! of!the!intervention!itself!is!minimal,!yet! substantial!clinical!and!financial!benefits!could! result.!Few!medical!interventions!can!make!that! claim.! Back!pain!is!one!of!the!most!important!causes!of! functional!limitation!and!disability!worldwide!and! is!an!Institute!of!Medicine!priority!condition.(1,2)! It!is!one!of!the!most!common!reasons!for! physician!visits.(3)!The!American!College!of!Physicians!(ACP)!instituted!a!program!in!2011!called!High\ Value,!Cost\Conscious!Health!Care!(HVCCHC).(4)!The!purpose!of!the!program!is!“…!to!help!physicians! and!patients!understand!the!benefits,!harms,!and!costs!of!an!intervention!and!whether!it!provides!good! value,!and!to!slow!the!unsustainable!rate!of!health!care!costs!while!preserving!high\value,!high\quality! care.”!The!importance!of!back!pain!is!highlighted!by!the!first!recommendations!of!the!program!being! focused!on!the!appropriate!use!of!spine!imaging.!(5)!In!April!2012,!the!ACP!in!combination!with!the! ABIM!Foundation!released!their!5!top!“Things!that!Patients!and!Physicians!Should!Question.”!!Number! two!on!the!list!was!“Don’t!obtain!imaging!studies!in!patients!with!non\specific!low!back!pain”!(6).!! Luo!and!colleagues!estimate!that!the!1998!direct!costs!of!low!back!pain!in!the!U.S.!were!over!$26!billion.! More!recently,!Martin!et!al,!estimated!that!the!2005!direct!costs!were!over!$86!billion.!(7)!Diagnostic! imaging!is!a!critical!step!in!the!work\up!of!back!pain.!It!can!quickly!lead!to!a!precise!and!actionable! diagnosis,!such!as!severe!central!spinal!stenosis!with!cauda!equina!compression!that!may!require!rapid! surgical!consultation.!But!imaging!examinations!of!the!lumbar!spine!frequently!reveal!numerous!findings,! including!disk!desiccation,!height!loss,!or!bulging,!with!questionable!relevance!to!patient!symptoms.! 2013409426/kj! 13! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Figure!1,!from!our!study,!Longitudinal!Assessment!of!Imaging!and!Disability!of!the!Back!(LAIDBack)!in! 2001!demonstrates!a!lumbar!spine!MR!without!degenerative!changes!(Figure!1A)!and!a!subject!with! extensive!degenerative!changes!(Figure!1B).!Neither!of!these!subjects!had!low!back!pain.!(8)!These! findings!are!common!in!asymptomatic!adults,!with!prevalences!in!this!group!as!high!as!90%.!(8\10)! Moreover,!multiple!studies!have!failed!to!demonstrate!clinical!benefit!with!the!use!of!early!MR!imaging! for!low!back!pain!(LBP)!compared!with!radiographs!alone!or!no!imaging!at!all;!furthermore,!the!imaging! results!may!negatively!affect!patients’!sense!of!well\being!(7,13,14).!But!diagnostic!imaging!of!the! lumbar!spine!can!also!lead!to!a!cascade!of!subsequent!tests!and!treatments!that!may!have!little! beneficial!impact!on!a!patient’s!outcome!and!may!even!be!deleterious.!(11,!12)!Because!incidental! findings!are!nearly!ubiquitous!with!spine!imaging,!it!is!important!to!have!a!good!understanding!of!the! prevalence!of!various!findings!in!asymptomatic!patients.!! While!spine!imaging!may!be!one!of!the!most!common!examples!of!incidental!findings!on!diagnostic! testing!resulting!in!a!cascade!of!subsequent!tests!and!treatments,!this!situation!is!by!no!means!limited!to! spinal!diagnosis.!Lung!cancer!screening!with!CT!was!recently!shown!to!be!beneficial!in!a!high\risk! population,!but!one!of!the!concerns!with!such!screening!are!the!frequent!benign!nodules!that!are! discovered.!(13)!Adrenal!nodules!seen!on!body!CTs!(14),!thyroid!nodules!seen!on!neck!and!chest!CTs!(14),! sinus!mucosal!thickening!seen!on!head!MR!(15)!and!CTs!(16)!could!all!lead!to!subsequent!diagnostic!and! therapeutic!interventions!if!their!prevalence!in!patients!without!disease!was!not!well\understood.! While!spine!specialists!are!well!aware!of!these!prevalence!data,!non\specialists!such!as!family! practitioners!and!general!internists!may!not!know!that!a!finding!such!as!an!annular!fissure!is!seen!in! about!one\third!of!asymptomatic!patients!and!if!present,!is!likely!not!related!to!a!patient’s!pain.! Several!years!ago!our!group!at!the!University!of!Washington!implemented!into!our!clinical!practice,!the! recommendation!of!Roland!and!van!Tulder!(17)!to!include!epidemiologic!information!in!the!radiology! report!to!help!physicians!interpret!findings!frequently!seen!on!lumbar!spine!imaging!(Figure!2).!By! providing!a!context!for!these!common!findings,!we!hoped!to!mitigate concern!and!dampen!any! subsequent!cascade!of!inappropriate!testing!and!treatment.! ! FIGURE 2: EPIDEMIOLOGIC STATEMENT INCLUDED IN LUMBAR SPINE MR IMAGING REPORTS Multiple!randomized!controlled!trials! have!shown!that!the!early!use!of!imaging! for!LBP!is!not!associated!with!improved! outcomes!and!may!be!harmful!to!the! patient!(11,!18\23).!The!American!College! of!Physicians!recently!re\issued!guidelines! for!imaging!patients!with!LBP!emphasizing! not!only!the!inefficiencies!of!early!imaging! but!also!the!potential!harms!(24).! Furthermore,!as!rates!of!MR!imaging!of! the!lumbar!spine!have!increased,!so!too! have!treatments;!including!narcotics! prescriptions,!lumbosacral!injections,!and! spinal!surgery,!often!without!benefit!(25\32).!Not!only!do!these!treatments!result!in!increased! expenditures!(7,!32,!33),!but,!more!importantly,!they!pose!serious!risks!to!the!health!of!the!patient.! Narcotics!are!associated!with!multiple!side!effects,!including!respiratory!depression,!cognitive! 2013409426/kj! 14! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! impairment,!constipation,!and!even!death,!as!well!as!the!development!of!tolerance!and!dependency!(34,! 35).!Complications!from!spinal!surgeries,!especially!more!invasive!fusions,!include!wound!complications,! major!medical!complications,!and!death!(32).!! In!2012!our!group!published!a!pilot!study!demonstrating!insertion!of!such!epidemiological!evidence!was! associated!with!reduced!narcotic!prescriptions!and!a!non\significant!reductions!in!subsequent!MR,!CT! and!physical!therapy!as!well.!Taken!together,!these!findings!suggest!primary!care!providers!were!more! reserved!in!their!management!of!patients!whose!MR!report!included!the!epidemiological!evidence! statement.!(Table!1:!from!McCoullough!et!al,!2012)!(36)!Additionally,!if!patients!learned!about!the! statement,!as!some!undoubtedly!did!since!patients!at!the!study!site!have!direct!access!to!their!medical! records,!knowledge!that!their!spine!findings!are!common!in!patients!without!back!pain!might!alleviate! anxiety,!which!is!known!to!have!an!important!influence!on!pain.!(37)! Table 1: Outcomes of Patients Whose Imaging Did and Did Not Include a Statement Containing Epidemiological Benchmarks (from McCoullough et al, 2012) (36) The!relatively!new!field!of!clinical!genomics!is!on!the!verge!of!a!virtual!explosion!of!genetic!tests!that!will! be!inexpensive!and!readily!available.!(38,!39)!!However,!genetic!testing!faces!the!same!challenges!of! communicating!risk!information!that!more!traditional!diagnostic!testing!has!faced!for!decades.!(39)! Lessons!learned!from!diagnostic!imaging!may!be!applied!to!genomic!testing!and!vice\versa.!! Because!our!intervention!is!simple,!inexpensive!and!can!be!automated,!it!is!easy!to!implement!on!a!large! scale,!making!it!nearly!ideal!to!study!in!the!context!of!a!large,!pragmatic!trial!in!multiple!health!systems.! We!decided!to!confine!our!participating!sites!to!large!health!systems!that!have!sophisticated!electronic! medical!records!allowing!us!to!passively!collect!our!outcomes!through!electronic!queries.!! Our!method!of!random!assignment!is!also!relatively!novel.!We!propose!to!use!a!stepped!wedge!cluster! design,!where!the!order!in!which!clinics!receive!the!intervention!is!determined!at!random!and!by!the! end!of!the!random!allocation,!all!clinics!will!have!received!the!intervention.!(40,!41)! Finally,!given!the!rapid!spread!and!adoption!of!IT!clinical!tools,!like!the!EMR!and!templates!for!radiology! readings,!finding!ways!to!capitalize!on!the!technology!itself!to!positively!influence!the!process!of!care! will!make!the!mammoth!nationwide!clinical!IT!investment!much!more!compelling.!This!project!is!truly! 2013409426/kj! 15! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! emblematic!of!the!kinds!of!innovative!thinking!that!needs!to!be!applied!to!the!clinical!IT!world!to!derive! the!maximum!benefit!of!the!tools!meant!to!deliver!better!and!more!efficient!care.! 3.!Specific!Aims! This!study!is!a!pragmatic!cluster!randomized!controlled!trial,!randomly!assigning!primary!care!clinics!at! four!sites,!to!receive!either!standard!lumbar!spine!imaging!reports!or!reports!containing!epidemiological! benchmarks!for!common!imaging!findings.!Our!primary!outcome!will!be!a!metric!of!back\related! intervention!intensity,!measured!passively!using!the!electronic!medical!record!(EMR).!The!primary! analysis!will!focus!on!clinic\level!changes!by!using!aggregate!patient\level!data.! Aim!1:!To!determine!whether!inserting!a!description!of!age\specific!prevalence!of!imaging!findings! among!asymptomatic!subjects!into!lumbar!spine!imaging!reports!decreases!back\related!interventions! !imaging,!injections,!surgeries,!etc.)!over!the!subsequent!year.! Aim!1a:!To!determine!if!inserting!epidemiological!evidence!reduces!Relative!Value!Units!(RVUs)! attributable!to!spine!interventions!(imaging,!injections,!specialist!referrals,!surgeries,!etc.).! Hypothesis!1a:!After!primary!care!clinics!are!randomly!assigned!to!receive!the!modified!report,!they!will! have!a!lower!average!overall!RVU!(technical!and!professional)!per!imaged\patient!attributable!to!spine! interventions!than!when!clinics!are!not!receiving!the!modified!reports.!Spine!interventions!reflect!visits,! tests,!and!procedures!and!are!patient!centered,!having!both!direct!and!indirect!impacts!on!patients.! Aim!1b:!To!determine!if!inserting!epidemiological!data!decreases!opioid!prescriptions.! Hypothesis!1b:!Time!periods!during!which!clinics!are!randomly!assigned!to!receive!the!modified!imaging! reports!will!have!a!lower!rate!of!subsequent!opioid!prescriptions!than!time!periods!during!which!clinics! do!not!receive!modified!reports.! Aim!1c:!To!determine!if!inserting!epidemiological!evidence!decreases!subsequent!cross\sectional! imaging!!magnetic!resonance!(MR)!and!computed!tomography!(CT).! Hypothesis!1c:!Time!periods!during!which!randomly!assigned!clinics!receive!modified!imaging!reports! will!have!a!lower!rate!of!subsequent!cross\sectional!imaging!than!time!periods!for!which!clinics!do!not.! Aim!1d:!To!explore!whether!adding!epidemiological!evidence!decreases!overall!costs!of!care!for!low! back!pain!based!on!CMS!reimbursement.! Hypothesis!1d:!Clinics!that!are!randomly!assigned!to!receive!the!modified!imaging!reports!will!have! lower!back!pain\related!estimated!payer!costs!than!clinics!whose!patients!do!not!receive!modified! reports.!Costs!are!another!outcome!that!are!highly!relevant!to!both!patients!and!health!systems.! Aim!2:!To!determine!whether!inserting!age\specific!prevalence!of!imaging!findings!in!asymptomatic! subjects!has!a!differential!effect!on!subsequent!back\related!interventions!if!inserted!into!lumbar!spine! MR!and!CT!imaging!reports!compared!with!plain!films.! 2013409426/kj! 16! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Hypothesis!2:!Inserting!epidemiological!information!into!plain!film!reports!will!result!in!a!greater! decrease!in!subsequent!back\related!interventions!than!similar!information!put!into!MR!and!CT!reports.! Given!that!plain!films!are!generally!obtained!earlier!in!the!course!of!back!pain!and!more!frequently!than! MR!and!CT,!the!potential!impact!of!inserting!epidemiological!information!into!plain!film!reports!is!large.! Aim!3:!To!determine!if!specific!imaging!findings!influence!subsequent!interventions.! Hypothesis!3:!Inserting!the!statement!will!result!in!a!greater!decrease!in!subsequent!interventions!for! patients!without!clinically!important!findings!compared!with!patients!who!have!clinically!important! imaging!findings.!Our!work!and!others!have!shown!that!certain!imaging!findings!are!likely!to!be!clinically! more!important!than!others!(e.g.!nerve!root!compression,!moderate!to!severe!central!stenosis,!disc! extrusions).!We!expect!that!patients!without!these!more!important!findings!will!have!a!greater! reduction!in!subsequent!interventions.! 4.!Study!Details! 4.1!Eligibility!Criteria! Because!this!is!a!pragmatic!trial,!we!have!minimized!eligibility!restrictions,!making!the!inclusion!criteria! as!broad!as!possible.!Clinics!will!be!the!primary!unit!of!randomization!and!analysis,!while!the! intervention!will!be!applied!at!the!individual!patient!level.!Thus!two!sets!of!eligibility!criteria!are! necessary:!clinic!and!patient.! The!criteria!for!clinic'eligibility!are!that!the!health!care!providers!are!a!distinct,!readily!identifiable!group! that!has!at!least!a!subgroup!of!primary!care!providers!who!do!not!practice!at!another!clinic!that!will!also! be!part!of!the!trial.!This!requirement!of!being!based!primarily!at!one!site!is!to!minimize!cross\ contamination!(having!the!use!of!epidemiological!benchmarks!at!one!site!influence!another!site!not! receiving!the!benchmarks).! The!criteria!for!patient'eligibility!are!that!they!have!had!an!imaging!study!of!the!lumbar!spine!requested! by!a!primary!care!provider.!We!will!include!all!conventional!lumbar!spine!imaging!(plain!films,!CT,!MR)! ordered!by!primary!caregivers.!! 4.2!Consent!procedure! Because!the!intervention!will!be!administered!at!the!clinic!level,!consent!of!either!individual!patients!or! primary!caregivers!is!neither!feasible!nor!warranted.!Moreover,!the!intervention!is!relatively!benign!(the! insertion!of!additional!epidemiological!information!into!the!radiology!report)!and!poses!minimal!risk!to! caregivers!and!patients.!The!performance!sites!are!enthusiastic!about!incorporating!the!epidemiological! benchmarks!into!their!reports!and!may!well!eventually!adopt!them!regardless!of!the!project,!our!study! simply!allows!for!systematic!study!of!the!effects!of!a!well\controlled!implementation!of!the!insertion!of! the!benchmark!information.!The!randomization!scheme!defines!when!each!clinic!begins!including!the! epidemiological!information!into!the!reports,!with!all!sites!eventually!receiving!the!intervention!of! interest.! 4.3!Inclusion!and!Exclusion!Criteria! We!will!define!a!clinic!as!a!primary!care!clinic!if!a!majority!of!the!practitioners!at!that!clinic!are!providing! 2013409426/kj! 17! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! primary!care.!We!will!include!general!internal!medicine!and!family!practice!physicians!as!primary! caregivers!as!well!as!mid\level!providers!working!with!physicians!such!as!nurse!practitioners!and! physician!assistants.! We!will!include!all!adult!patients!of!eligible!caregivers!who!have!had!a!lumbar!spine!imaging!study!plain! film,!CT!or!MR)!ordered!by!their!primary!care!practitioner.! 5.!Research!Design!and!Methods! 5.1!Clinic/Practitioner/Patient!Identification! The!site!PI!will!identify!eligible!clinics!within!their!health!system,!working!closely!with!their! administrative!and!information!technology!(IT)!staff!to!assure!complete!inclusion!of!primary!care!clinics.! The!site!PIs!will!then!categorize!practitioners!within!each!clinic!by!specialty,!designating!general! internists,!family!practitioners!and!obstetrician/gynecologists!as!primary!care!practitioners.!Mid\level! providers!(e.g.!nurse!practitioners!and!physician!assistants)!working!as!primary!caregivers!will!also!be! classified!as!primary!care!practitioners.!The!health!information!system!will!be!used!to!automatically! identify!when!a!practitioner!from!a!particular!clinic!orders!a!lumbar!spine!imaging!study.! 5.2!Randomization! At!each!site!we!will!identify!the!settings!where!primary!care!is!delivered!and!designate!an!appropriate! unit!that!will!constitute!a!functional!“clinic”!for!randomization!and!analysis.!We!will!randomly!assign!all! predetermined!clinics!at!each!site!to!receive!the!intervention!at!one!of!five!fixed!time\points,!rolling! interventions!out!every!six!months!beginning!at!the!start!of!the!second!quarter!of!Year!2.!Using!cutoffs! determined!in!the!UH2!project!phase,!we!will!sort!clinics!by!number!of!primary!care!providers!into! tertiles!(e.g.!small,!medium,!large!clinics).!From!each!tertile!we!will!randomly!select!clinics!using!urn\ based!randomization!(without!replacement)!stratified!by!site!and!clinic!size!such!that!clinics!of!small,! medium,!and!large!size!are!equally!represented!in!each!randomization!wave.!!For!more!details!regarding! the!Analysis!plan,!please!refer!to!the!UH3!transition!request!proposal!and!accompanying!Appendix!10,! Analysis!plan!that!reflects!modifications!made!to!the!original!plan!we!outlined!for!UH2.! Figure!3:!Proposed!Randomization!Schedule! 2013409426/kj! 18! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! 5.3!Clinic/Patient!Enrollment! Using!the!site!administrative!data!systems,!we!will!identify!all!primary!care!providers!(PCP)!at!a!given! clinic.!When!an!identified!PCP!from!a!randomized!clinic!submits!a!request!for!a!lumbar!spine!imaging! study,!the!report!will!automatically!be!flagged.!The!PCP’s!name!will!be!cross\referenced!with!the! randomization!assignment!and!those!PCPs!who!work!in!clinics!assigned!to!receive!the!intervention!will! have!the!epidemiological!benchmark!information!automatically!inserted!into!their!imaging!reports.! Those!PCPs!who!work!in!clinics!not!yet!scheduled!to!receive!the!benchmark!information!will!get!the! usual!imaging!report!issued!by!their!radiologists.!Since!the!intervention!will!be!applied!at!the!PCP!and! clinic!level,!all!patients!receiving!lumbar!spine!imaging!studies!at!those!clinics!will!be!part!of!the!trial.! The!lumbar!spine!imaging!studies!that!we!plan!to!include!in!the!trial!are!plain!films,!magnetic!resonance!! (MR)!imaging!examinations!and!computerized!tomography!(CT).!Table!2!lists!the!proposed!CPT!codes! that!we!will!flag!for!inclusion.!We!are!currently!not!planning!on!including!nuclear!medicine!studies!(e.g.! bone!scans,!both!planar!and!SPECT)!both!because!they!are!infrequently!ordered!by!primary!care! clinicians!as!well!as!because!there!is!inadequate!benchmarking!information.! Table!2:!Lumbar!Examinations!to!be!Included!in!Pragmatic!Trial! CPT!Code! Examination!Description!!!!!!!!!!!!!!!!!!!!!!!!!!!! 72080!!!!!!!!!!!!THORACOLUMBAR!SPINE,!2!views! 72100! LUMBAR!SPINE!2!VIEWS! 72110! LUMBAR!SPINE!3\4!VIEWS! 72114! LUMBAR!SPINE!5!VIEWS! 72131! CT!L!SPINE!W/O!CONTRAST! 72132! CT!L!SPINE!W/!CONTRAST! 72133! CT!L!SPINE!W/O!&!W/!CONTRAST! 72148! MRI!LUMBAR!SPINE!W/O!CONTRAST! 72149! MRI!LUMBAR!SPINE!W/!CONTRAST! 72158! MRI!L!SPINE!W/!&!W/O!CONTRAST! 5.4!Data!collection! We!will!collect!all!baseline!and!follow\up!data!from!the!electronic!information!systems!which,! depending!on!the!site,!will!include!both!the!electronic!medical!record!(EMR)!as!well!as!administrative! data!systems.! Baseline'Data'Collection:!We!will!include!all!patients!receiving!lumbar!spine!imaging!studies!(plain!films,! MR!and!CT)!in!the!last!quarter!of!Year!1!and!the!first!quarter!of!Year!2!as!a!part!of!a!baseline!accrual! period!to!establish!baseline!parameters!for!the!primary!care!physicians!in!participating!clinics.!Since!the! randomization!will!occur!at!the!clinic!level,!the!baseline!data!will!reflect!clinic!level!ordering!patterns!of! diagnostic!and!therapeutic!interventions.! FollowMup'Data'Collection:!We!will!capture!EMR!data!on!patients!for!a!minimum!of!one!and!up!to!two! years!after!the!index!imaging!test.!All!patients!will!have!a!minimum!of!one\year!follow\up.!Eighty! percent!of!patients!will!have!two\year!follow\up!due!to!the!staggered!implementation!of!the! 2013409426/kj! 19! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! intervention.!The!six\month!length!of!each!patient!accrual!period!is!sufficiently!long!to!account!for! temporal!or!system\level!trends!in!the!measured!outcomes!over!the!course!of!the!study.!! 5.5!Aims!for!UH2!Phase! Our!goal!is!to!use!the!planning!UH2!phase!of!the!grant!to!accomplish!the!following:!First,!we!will!refine! the!epidemiological!benchmarks!that!we!will!insert!into!the!radiology!report.!Second,!we!will!develop! and!test!our!site\specific!deployment!method!for!the!cluster!randomization.!Third,!we!will!develop!a! metric!that!reflects!the!intensity!of!interventions!for!back!pain\related!care!and!develop!CMS\based! standardized!cost!estimates!associated!with!resource!use!intensity!that!can!be!applied!uniformly!among! health!systems.!We!will!validate!this!metric!using!data!from!the!health!care!systems!electronic!medical! record.!Fourth,!we!will!develop!and!validate!our!methods!for!extracting!outcome!data!from!the! electronic!medical!record.!Fifth,!once!we!have!defined!the!above,!we!will!obtain!Institutional!Review! Board!approval!for!the!implementation!phase!of!the!study.!We!will!also!use!this!time!to!assemble! subcontracts!for!administrative!review!at!each!site.! Aim!1:!Refine!the!information!to!be!included!in!the!radiology!report!so!that!it!is!specific!for!imaging! modality!and!patient!age.!! In!our!original!implementation,!we!only!inserted!the!epidemiological!benchmarks!into!reports!of!lumbar! spine!MRs!whereas!in!the!current!project!we!propose!to!insert!the!information!into!reports!of!MR,!CT! and!plain!films.!Moreover,!we!used!epidemiological!data!from!a!single!study!published!by!our!group! (Table!3).(8)!Other!groups!have!published!similar!data!for!MR!as!well!as!other!modalities.!(9,!10,!42\59)! In!addition!to!updating!the!epidemiological!benchmarks!and!expanding!them!to!other!modalities,!we! will!also!gather!data!regarding!age\specific!rates!for!various!imaging!findings.!While!eventually!we!would! envision!a!decision!support!tool!that!could!recognize!specific!patient!attributes,!such!as!age!or!the! presence!of!a!particular!finding,!and!insert!customized!benchmark!data!for!that!individual,!such!a!system! is!beyond!the!scope!of!this!project.!Instead!we!plan!to!insert!benchmarks!that!are!simply!stratified!by! age!ranges.!We!will!perform!a!systematic!review!of!the!literature!so!that!we!are!inserting!the!most! recent!and!complete!epidemiological!evidence!into!the!radiology!report.! Table!3:!Age\specific!Rates!of!Lumbar!Spine!Imaging!Findings!(from!Jarvik!et!al,!2001)! 2013409426/kj! 20! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Aim!2:!Develop!siteWspecific!deployment!methods!for!the!stepped!wedge,!cluster!randomization! scheme.!! The!unit!of! Figure!4:!Comparison!of!Parallel,!Crossover!and!Stepped!Wedge! randomization!for! Designs!(from!Hussey!et!al.,!2007)! this!project!will!be!at! the!level!of!the!clinic.! The!stepped!wedge! design!is!a!one\way! cluster,!crossover! design!that! temporally!spaces!the! intervention!and! assures!that!each! participating!clinic! will!eventually! receive!the! intervention.!Figure!4,!from!Hussey!et!al,!(41)!compares!the!stepped!design!with!parallel!and!crossover! designs:! Our!experience!with!the!BOLD!project!and!other!multicenter!studies!informs!us!that!the!procedures!and! hurdles!at!each!site!will!be!different!so!that!our!approach!for!implementing!the!cluster!randomization! must!be!customized!at!least!to!some!extent.!In!the!UH2!phase!of!the!project,!we!will!work!closely!with! the!informatics!groups!at!each!health!system!to!develop!the!schedule!of!clinics!to!be!randomized!to! receive!the!insertion!of!epidemiological!data!into!the!relevant!reports.!A!key!component!of!deploying! the!intervention!is!that!it!be!automated!and!not!require!a!radiologist!to!actively!insert!benchmark! statements!into!the!report.!We!also!plan!to!create!a!process!for!notifying!sites!of!intervention! deployment,!being!careful!to!minimize!opportunities!for!internal!or!external!sources!of!contamination.! Aim!3:!Develop!and!validate!a!composite!measure!of!spine!intervention!intensity!that!combines!into!a! single!metric!the!overall!intensity!of!resource!utilization!for!back!pain!care.!! Relative!value!units!(RVUs)!are!a!measure!of!work!effort!associated!with!a!particular!medical!service.! Although!there!are!potential!drawbacks!to!RVUs,!including!overvaluing!certain!services!relative!to!others,! it!is!a!widely!used!metric!and!one!potential!method!for!measuring!the!intensity!of!services!provided!(or! resources!utilized)!for!back!pain!treatment.!During!the!planning!year,!our!group!would!need!to! specifically!identify!services!that!would!comprise!the!back!care!intensity!metric.!! Once!defined,!we!could!then!attempt!to!validate!the!metric!using!the!World!Health!Organization’s! stepped!care!approach!to!pain!treatment,!which!amounts!to!an!escalating!ladder!of!treatment! intensiveness.!! Aim!4:!Develop!and!validate!electronic!data!methods!and!tools!to!capture!the!outcomes!of!interest! (subsequent!diagnostic!testing,!opioid!prescriptions,!spinal!injections,!spine!surgeries).!! 2013409426/kj! 21! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! While!all!of!the!health!care!systems!that!we!propose!to!involve!in!our!project!have!sophisticated!and! comprehensive!electronic!medical!records!systems,!we!know!from!experience!that!accessing!the! relevant!data!and!transmitting!it!to!the!project’s!data!coordinating!center!will!require!careful!planning! and!individualized!approaches!for!each!health!system.!We!have!already!successfully!implemented! methods!for!collecting!some!of!this!information!from!2!of!our!proposed!sites!(Kaiser!Northern!California! and!Henry!Ford!Health!System,!Detroit,!MI)!and!would!refine!and!deploy!these!methods!at!two!new! sites,!Group!Health!Cooperative!in!Seattle,!WA!and!the!Mayo!Clinic!Health!System!in!Minnesota!and! Wisconsin.! 5.6!Working!Groups ! The!planning!UH2!phase!of!the!grant!will!be!used!to!accomplish!the!following:!First,!we!will!refine!the! epidemiological!benchmarks!that!we!will!insert!into!the!radiology!report.!Second,!we!will!develop!and! test!our!site\specific!deployment!method!for!the!cluster!randomization.!Third,!we!will!develop!a!metric! that!reflects!the!intensity!of!interventions!for!back!pain\related!care!and!develop!Medicare\based! standardized!cost!estimates!associated!with!resource!use!intensity!that!can!be!applied!uniformly!among! health!systems.!We!will!validate!this!metric!using!data!from!BOLD!and!Medicare.!Fourth,!we!will!develop! and!validate!our!methods!for!extracting!outcome!data!from!the!electronic!medical!record.! The!scope!of!work!for!the!UH2!planning!year!has!been!outlined!and!assigned!to!one!of!four!working! groups!that!align!with!the!aims!described!above.! !! Working!Group!1 !will!focus!on!refining!the!information!to!insert!into!the!radiological!report.!They!will! lead!a!critical!review!of!the!literature!focusing!on!the!age\specific!prevalence!of!common!findings!seen! on!plain!films,!CT!and!MR!in!people!without!low!back!pain.!We!will!use!meta\analytic!methods!to! combine!the!prevalence!estimates!from!multiple!sources,!weighting!by!study!quality!and!relevance!to! the!LIRE!population.!We!will!summarize!the!epidemiological!information!so!that!it!can!be!inserted!into! plain!film,!MR!and!CT!reports!of!lumbar!spine!imaging.!Dr.!Jarvik!will!lead!this!working!group!and!work! closely!with!Dr.!Kallmes!at!the!Mayo!Clinic!Health!System.! Working!Group!1!progress :!!Members!of!this!working!group!developed!and!pilot!tested!the!intervention! text!and!took!several!steps!in!this!process.! 1. Comprehensive!literature!review!for!relevant!articles!regarding!radiologic!findings! !2.!!Two!independent!reviews!and!data!extraction!of!findings!from!relevant!articles.!See!! Appendices!1!and!2!for!the!Abstraction!form!used!and!list!of!final!articles!contributing!data!to!the! prevalence!rates!included!in!the!intervention!text.!! !3.!!!Data!cleaning!and!compilation! !4.!!!Data!analysis!and!modeling!to!determine!a)!which!findings!had!sufficient!data!to!report!on! and!b)!age!cut!points!! !7.!!!Consult!with!the!Program!for!Readability!In!Science!&!Medicine!(PRISM)!scientific!writing! group!at!Group!Health.!!! !8.!!!Key!informant!interviews!with!two!patient!advisors!for!feedback!on!format!and!readability!! !9.!!!On\line!survey!with!over!20!patient!advisors!for!feedback!and!comment! Participating!patient!representatives!were!provided!with!a!sample!radiology!report!and!four! different!versions!of!the!intervention!macro.!!They!were!asked!the!following!questions:! 2013409426/kj! 22! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! ¥ What!is!your!age?! ¥ Have!you!ever!sought!medical!care!for!low!back!pain?! ¥ Which%format%option%did%you%prefer?! ¥ Why!did#you#prefer#that#format#option#over#the#others?#! ¥ Based&on&this&information,&would&you&say&disc&degeneration&is&common&or&rare?! ¥ What%is%clear!about&the&information&presented?&! ¥ What%is%confusing)about&the&information&presented?&! ¥ How$would$you$use$the$information(presented(in(the(option(that(you(preferred?(! ¥ If#you#received#a#radiology#report#with#the#information#in#the#option#that#you#preferred,# would&you&feel&less&or&more&concerned&about&having&the&imaging&finding(s)&that&are& common%in%people%without%back!pain?&!For$example,$would$you$feel$less$or$more$ concerned'about'having'a'degenerated'disk?! !10.!!On\line!survey!with!seven!primary!care!physicians!for!feedback!and!comment.!!Participating! providers!were!provided!with!a!sample!radiology!report!and!three!different!versions!of!the! intervention!macro.!!They!were!asked!the!following!questions:! ¥ What!proportion!of!your!patient!visits!are!related!to!low!back!pain?!!! ¥ Which%format%option%do%you%prefer?! ¥ Why!did#you#prefer#that#format#option#over#the#others?#! ¥ Based&on!this%information,%would%you%say%disc%degeneration%is%common%or%rare?! ¥ What%could%be%improved%about%your!preferred&option?&! ¥ If#you#received#a#radiology#report#with#the#information#in#the#option#that#you#preferred,# would&you&be&less&or&more&concerned&about&a!given&patient's&imaging&finding?&!For$ example,(would(you(feel(less(or(more(concerned(about(a(patient(having(a(degenerated( disk?! ¥ How$would$the$information$presented$in$the$option$that$you$preferred$inform$your$ clinical&decision&making?! The!final!text!is!a!product!of!the!efforts!described!above!and!will!serve!as!the!final!intervention.!! Different!wording!is!offered!based!on!1)!Age!(<40,!40\60,!and!>60!years!of!age)!and!2)!Modality! (plain!film,!CT,!MR).! Working!Group!2 !will!focus!on!methods!to!practically!deploy!the!stepped!wedge!cluster!randomization! scheme.!This!will!require!working!closely!with!the!informatics!group!at!each!of!the!four!sites!to! determine!the!optimal!method!for!inserting!the!intervention!text!into!their!reports.!This!might!occur!at! the!level!of!the!radiology!reporting!software!(RRS),!the!radiology!information!system!(RIS)!or!the! hospital!information!system!(HIS).!Sites!must!demonstrate!the!ability!to!selectively!insert!the! intervention!text!only!into!reports!where!the!clinic!is!randomly!allocated!to!receive!the!intervention.!Dr.! Heagerty!and!Mr.!Comstock!will!lead!this!working!group!out!the!Center!for!Biomedical!Statistics!(CBS)! that!is!part!of!the!Institute!for!Translational!Health!Sciences!(ITHS),!the!University!of!Washington’s! Clinical!and!Translational!Science!Award!(CTSA).!Dr.!Heagerty!is!Professor!and!Associate!Chair!of! Biostatistics!and!the!Director!of!the!CBS.! Working!Group!2! progress:!!A!site\readiness!tool!has!been!developed!to!guide!efforts!for!pilot!testing! the!insertion!of!the!intervention!macro!across!the!clinics!at!a!given!site!(see!Appendix!3!for!an!example! of!this!pilot!testing!document!and!site!checklist).!!At!the!conclusion!of!this!pilot,!the!number!and!size!of! 2013409426/kj! 23! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! participating!clinics!at!each!site!will!be!verified!as!well!a!final!determination!regarding!technical! feasibility!of!clinic!randomization!and!intervention!text!insertion.! Working!Group!3 !will!develop!and!validate!the!composite!measure!of!spine!intervention!intensity,!likely! to!be!based!on!units!and!types!of!services!used.!We!will!validate!this!metric!using!data!from!a!variety!of! sources!including!the!Back!pain!Outcomes!using!Longitudinal!Data!(BOLD)!registry.!This!will!be!an!on\ going!effort!through!much!of!the!planning!year!as!we!review!the!literature!and!obtain!expert!opinion!to! optimize!the!factors!comprising!the!composite!measure,!and!validate!the!measure.!Drs.!Deyo!and! Bresnahan!will!co\lead!this!effort.!Working!Group!3!will!also!develop!cost!estimates!to!apply!to!overall! spine!intervention!intensity!and!to!individual!services.! Working!Group!3!progress: !!A!comprehensive!review!of!the!literature!was!conducted!searching!for! articles!pertaining!to!RVU\related!assessment!(see!Appendix!4!for!a!complete!list!of!articles).!Mapping!of! CPT!codes!to!relative!value!units!has!been!completed!using!the!BOLD!Registry!data.!!A!manuscript!of!this! work!is!currently!being!drafted.!Mapping!of!codes!determined!to!be!“spine\related”!has!been!completed! utilizing!previous!work!performed!by!a!colleague!currently!at!Dartmouth,!Brook!Martin,!PhD,!MPH.!! Working!Group!4 !will!develop!and!validate!the!methods!to!extract!the!necessary!data!to!passively! measure!outcomes!from!each!site’s!EMR.!The!group!will!perform!test!data!pulls!from!each!site!of!the! key!variables!identified!by!Working!Group!3.!Anonymized!data!will!be!transmitted!to!the!Data! Coordinating!Center!at!the!UW,!housed!at!the!CBS.!UW!would!only!receive!limited!data!sets!without! protected!health!information.!!Dr.!Nick!Anderson!will!lead!this!group,!working!closely!with!informatics! experts!at!each!of!the!sites.!Dr.!Anderson!was!Associate!Director!of!the!Bioinformatics!Core!at!the!UW! ITHS!at!the!start!of!the!project!and!has!since!taken!a!position!with!University!of!California\!Davis!but!will! remain!involved!in!the!project.! Working!Group!4!progress:!!An!overall!plan!has!been!developed!to!leveraging!PopMedNet!!to!virtually! connect!the!various!implementation!sites!to!UW!who!is!to!serve!as!the!DCC!for!this!project!in! anticipation!of!the!data!exchange!which!will!take!place!in!the!UH3!phase.!!For!EHR!extraction,!we! anticipate!utilizing!the!Virtual!Data!Warehouse!at!the!three!HMORN!sites!as!much!as!possible,! developing!customized!programming!pieces!that!are!necessary!beyond!that!to!further!capture!all!the! necessary!data!elements.!!Mayo!clinic!does!not!use!the!VDW!so!all!data!extraction!programming!will! have!to!be!customized!at!that!site.! The! working! group! activities! as! well! as! general! project! coordination! will! utilize! the! web\based! tool,! Basecamp!(www.BaseCamp.com)!to!streamline!efforts.!!All!faculty!and!staff!at!the!DCC,!as!well!as!Site! PI’s!and!study!personnel!at!each!site,!will!be!given!access!to!Basecamp.!In!addition,!each!working!group! has! a! dedicated! “project”! within! Basecamp! that! will! be! used! to! facilitate! discussions! among! group! members,!organize!files,!keep!notes,!and!centralize!study\related!documents.! In!order!to!coordinate!efforts!at!each!of!the!individual!sites!and!assure!the!UH2!milestones!are!met!in!a! timely! fashion,! site! milestones! have! been! outlined! and! site! PIs! and! their! respective! research! and! technical!teams!are!working!towards!these!goals.! 1.“Radiology!buy\in”!! Assignee:!Site!PI/!Jarvik! 2013409426/kj! 24! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Objective:!Facilitate!consensus!among!Radiologists!that!they!are!willing!to!have!the!intervention! used! Deliverable!#1:!!Letter!of!support!from!Radiology!chair!(template!provided)! Due!Date:!July!31,!2013! 2.“IRB!approval”! Assignee:!!Site!PI/!James! Objective:!Coordinate!IRB!review!such!that!waiver!of!consent!and!HIPAA!for!patients!and!waiver! of!consent!for!physicians!is!in!in!place! Deliverable!#2:!!Final!IRB!approval!documentation! Due!Date:!!Final!approval!of!application!August!31,!2013!!! 3.!“Randomization!Pilot”! Assignee:!!Site!PI/!Comstock! Objective:!Demonstrate!ability!to!insert!template!into!radiology!report!on!schedule,!randomized! by!clinic! Deadline:!July!31,!2013! 4.!“EMR!data!extraction”! Assignee:!Site!PI/!Anderson! IT!resource!identified!who!Installs!and!authenticates!the!PopMedNet!Client! Deadline:!July!31,!2013! Identify!a!programming!resource!and!validate!a!"starter!set"!VDW!query!that!has!been!mapped! against!the!LIRE!data!set! Deadline:!!August!31,!2013! 2013409426/kj! 25! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Table 4: Milestones for UH2 Planning Year Needed to Transition to UH3 Implementation Suitability for Importance to UH3 UH2 Phase Milestone Timeline Assessing UH2 Success Success Aim 1:Complete critical literature Critically important: Highly suitable: review to refine the information to be Development and Necessary to develop included in the radiology report so refinement of the text describing age- that it is specific for imaging modality months intervention text is specific epidemiologic and patient age. necessary for the timely benchmarks for plain Jarvik, Bresnahan, Deyo, Halabi, implementation of the films, CT and MR. Kallmes, Turner, Luetmer, Avins intervention. Aim 2: Develop site-specific Critically important: Sites deployment methods for the stepped Highly suitable: Will must have proven ability wedge, cluster randomization scheme need detailed to selectively implement Comstock, Anderson, Avins, 3 implementation intervention text in the Cherkin, Halabi, Heagerty, James, months protocols prior to radiologic reports Jarvik, Pathak, Murphy, Ciarelli, implementation of generated for providers at Needed: KP technical resource intervention. a given primary care Needed: GHC technical resource clinic. Important: The Aim 3: Develop and validate a achievement is important Suitable: Definition of composite measure of spine but not critical. While intervention intensity that combines composite measure other metrics could be to be used as primary into a single metric the overall used successfully, intensity of resource utilization for outcome and including single months back pain care and develop cost validation using parameters, a composite existing BOLD data. estimates associated with units of measure enables a more resource used and intensity of use. comprehensive estimate Bresnahan, Deyo, James, Jarvik of overall care received for back pain. Aim 4:Develop and validate electronic data methods and tools to Highly suitable: Data capture the outcomes of interest Critically Important: dictionary and (subsequent diagnostic testing, opioid Feasibility must be 6 protocol to query the prescriptions, spinal injections, demonstrated in order to months electronic medical specialist visits, spine surgeries, etc.) passively collect record (EMR) Anderson, Comstock, James, outcomes using the EMR. necessary Jarvik, Turner Highly suitable: Drafting of DSMP, Additional Goals: Data Safety Critically Important: A submission to Monitoring Plan (DSMP) formulation 6 safety officer will need to IRBs,funding agency, and designation of safety officer months review and approve the and designation of Heagerty, Comstock, James, Jarvik study DSMP. Safety Officer needed prior to study initiation Highly suitable: Draft Critically Important: The Additional Goals: Draft study the study protocol so study protocol will need protocol it incorporates the to be reviewed by the months James, Comstock, Jarvik decisions made for IRBs before the study Aims 1-4 receives final approval. 2013409426/kj! 26! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Additional Goals: IRB approval Highly suitable: Critically Important: IRB James, Avins, Cherkin, Halabi, Conditional IRB 9-12 approval is required Jarvik, Kallmes, Project Managers approval at multiple months before study procedures from each site (Hamilton, Connelly, sites always can be initiated. Wessman, Hawkes) challenging Additional Goals: Establish subcontracts with sites Draft and submit Critically Important: 9-12 James, Avins, Cherkin, Halabi, subcontracts for the Subcontracting with sites months Jarvik, Kallmes, Post-award UH3 phase at all sites is a required process. personnel at each site 5.7!Aims!for!UH3!Phase! Aim!1:!To!determine!whether!inserting!a!description!of!ageWspecific!prevalence!of!imaging!findings! among!asymptomatic!subjects!into!lumbar!spine!imaging!reports!decreases!backWrelated!interventions! during!the!subsequent!year.! Aim!1a:!!To!determine!if!inserting!epidemiological!evidence!reduces!RVUs!attributable!to!spine! interventions.! Hypothesis'1a:'After'primary'care'clinics'are'randomly'assigned'to'receive'the'modified'report,'they'will' have'a'lower'average'overall'RVU'(technical'and'professional)'per'imagedMpatient'attributable'to'spine' interventions'than'clinics'that'are'not'receiving'the'modified'reports.' We!will!calculate!an!overall!spine\related!RVU!for!each!patient!in!the!study!by!summing!all!RVUs! attributable!to!spine\interventions!within!one!and!two!years!after!the!date!of!return!of!the!index!image! report!(the!plain!film,!MR!or!CT!of!the!lumbar!spine!imaging!study!whose!report!either!does!or!does!not! contain!the!epidemiological!benchmark!data).!For!each!patient!accrual!period!in!Figure!3,!we!will! calculate!a!total!spine\related!RVU!per!primary!care!provider!who!orders!at!least!one!lumbar!imaging! exam!in!Table!2.!We!will!aggregate!spine\related!RVUs!across!the!study\eligible!patient!panel.!The! calculated!primary!care!provider!RVU!will!serve!as!the!primary!outcome!measure!of!this!study.!As!noted,! we!will!also!apply!standardized!CMS\based!costs!to!RVU!calculations.! We!will!use!generalized!linear!mixed!models!with!jackknifed!standard!to!model!the!change!in!post\ intervention!RVU!from!pre\intervention!RVU.!We!will!use!random!effects!for!the!clinic!and!for!the!effect! of!intervention!defined!by!the!indicator!of!exposure!to!the!LIRE!intervention.!We!will!use!a!three\month! pre\intervention!washout!period!where!patients!with!index!visits!in!this!window!will!be!excluded!from! the!analysis.!We!will!adjust!the!model!for!period!of!time!(period!0!through!period!5)!as!a!fixed!effect!to! adjust!for!general!calendar!trends!in!RVUs.!We!will!also!adjust!the!model!for!the!type!of!image!ordered! at!the!initial!index!visit!as!a!categorical!variable!(plain!film,!CT,!MR).! Aim!1b:!To!determine!if!inserting!epidemiological!evidence!decreases!subsequent!opioid!prescriptions.!!! Hypothesis'1b:'Clinics'that'are'randomly'assigned'to'receive'the'modified'imaging'reports'will'have'a' lower'rate'of'subsequent'opioid'prescriptions'than'clinics'whose'patients'do'not'receive'modified'reports.' 2013409426/kj! 27! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Our!pilot!work!suggested!that!including!epidemiological!evidence!was!associated!with!a!nearly!3.5\fold! reduction!in!opioid!prescriptions.!We!feel!that!it!is!important!to!test!this!hypothesis!given!the!growing! recognition!that!the!overuse!of!opioids!is!an!important!issue!in!the!management!of!back!pain!patients.! We!will!calculate!a!binomial!outcome!of!the!number!of!patients!(out!of!the!eligible!patient!panel)!with!a! prescription!for!opioids!that!occurred!after!the!lumbar!spine!imaging!report!was!finalized.!We!will! examine!this!for!each!primary!care!provider!across!each!of!the!six!patient!accrual!periods.!We!will!utilize! a!generalized!linear!mixed!model!to!assess!the!impact!of!the!LIRE!intervention!on!subsequent!written! opioid!prescription!rates.!Since!some!patients!will!already!have!an!active!prescription!for!opioids!at!the! time!of!the!index!visit,!we!will!also!conduct!a!subgroup!analysis!by!conducting!analyses!separately!for! those!patients!with!and!without!an!active!prescription!at!baseline.!As!described!earlier,!we!will!convert! all!prescriptions!into!MEDs.!This!will!allow!us!to!examine!temporal!trends!in!not!only!prescriptions!but! also!dose.! ! ! Aim!1c:!To!determine!if!inserting!epidemiological!evidence!decreases!subsequent!crossWsectional! imaging!(magnetic!resonance!(MR)!and!computed!tomography).! Hypothesis'1c:'Clinics'that'are'randomly'assigned'to'receive'the'modified'imaging'reports'will'have'a' lower'rate'of'subsequent'crossMsectional'imaging'than'clinics'whose'patients'do'not'receive'modified' reports.' The!number!of!patients!who!receive!cross\sectional!re\imaging!within!1!and!2!years!(out!of!the!eligible! patient!panel)!will!be!calculated!as!a!binomial!outcome!measure!for!each!primary!care!provider!across! each!of!the!six!patient!accrual!periods.!We!will!again!utilize!a!generalized!linear!mixed!model!to!assess! the!impact!of!the!LIRE!intervention!on!subsequent!rates!of!cross\sectional!reimaging,!including!random! effects!for!the!baseline!rate!of!cross\sectional!re\imaging!and!for!an!indicator!of!exposure!to!the!LIRE! intervention.!We!will!adjust!the!model!for!period!of!time!(period!0!through!period!5)!as!a!fixed!effect!to! adjust!for!general!longitudinal!trends!in!re\imaging!rates.!We!will!also!adjust!the!model!for!the!type!of! image!ordered!at!the!initial!index!visit!as!a!categorical!variable!(plain!film,!CT,!MR).! Aim!1d:!To!explore!whether!adding!epidemiological!evidence!decreases!overall!costs!of!care!for!low! back!pain!based!on!CMS!reimbursement.! Hypothesis'1d:'Clinics'that'are'randomly'assigned'to'receive'the'modified'imaging'reports'will'have'lower' back'painMrelated'estimated'payer'costs'than'clinics'whose'patients'do'not'receive'modified'reports.' We!will!apply!CMS\based!reimbursement!amounts!to!RVU!calculations!in!order!to!standardize!unit!cost! estimation!among!our!sites.!We!will!use!CPT!and!diagnostic!codes!to!determine!whether!interventions! are!associated!with!back!pain,!and!multiply!the!total!back\treatment!related!RVUs!at!sites!by!the!unit! price!payment!amounts!for!respective!RVU!calculations.! Table!5:!Sample!of!RVUs!and!CMSWbased!payment!amounts!for!lumbar!imaging!(US$!2012)!! CPT!! Diagnostic!Imaging!Exam! Hospital!(facility)! !! !! Professional!! Hosp!!!!! Hosp! Pro! Pro! Pro! !! !!!! RVU! payment! !! wRVU! tRVU! payment! 72100! 2!view!x\ray!exam!lower! 0.6399! $48.22! !! 0.22! 0.34! !!!$12.08! 2013409426/kj! 28! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! spine! 4!view!x\ray!exam!lower! 72120! spine! 0.6399! $48.22! !! 0.22! 0.35! $12.47! ! ! ! ! ! ! ! ! 72131! CT!lumbar!spine!w/o!dye! 2.746! $206.54! !! 1! 1.44! $51.57! 72132! CT!lumbar!spine!w/dye! 4.2918! $323.42! !! 1.22! 1.75! $62.58! CT!lumbar!spine!w/o!&! 72133! w/dye! 4.7716! $359.59! !! 1.27! 1.82! $65.10! ! ! ! ! ! ! ! ! 72148! MRI!lumbar!spine!w/o!dye! 4.8333! $364.24! !! 1.48! 2.14! $76.44! 72149! MRI!lumbar!spine!w/dye! 6.2248! $469.10! !! 1.78! 2.57! $91.88! MRI!lumbar!spine!w/o!&! 72158! w/dye! 7.6273! $574.79! !! 2.36! 3.41! $121.79! Aim!2:!To!determine!whether!inserting!ageWspecific!prevalence!of!imaging!findings!in!asymptomatic! subjects!has!a!differential!effect!on!subsequent!backWrelated!interventions!if!inserted!into!lumbar! spine!magnetic!resonance!(MR)!and!computed!tomography!(CT)!imaging!reports!compared!with!plain! films.!! Hypothesis'2:'Inserting'epidemiological'information'into'plain'film'reports'will'result'in'a'greater' decrease'in'subsequent'backMrelated'interventions'than'similar'information'put'into'MR'and'CT'reports.' Our!pilot!work!examined!only!MR.!However,!given!that!plain!films!are!generally!obtained!earlier!in!the! course!of!back!pain!and!more!frequently!than!cross\sectional!imaging,!the!potential!impact!of!inserting! epidemiological!information!into!plain!film!reports!is!large.!In!each!separate!analysis!of!Aim!1!outcomes,! we!will!add!an!indicator!of!imaging!modality!(plain!film!versus!MR!versus!CT)!and!an!imaging!modality!by! treatment!interaction!term!in!the!model.!Primary!inference!will!be!on!the!interaction!term,!where!we! expect!that!patients!with!more!advanced!imaging!will!have!a!significantly!greater!reduction!in! subsequent!interventions!than!those!who!receive!a!plain!film!image.! Aim!3:!To!determine!whether!the!presence!of!certain!imaging!findings!influence!subsequent! interventions!!! Hypothesis'3:'Inserting'epidemiological'information'will'result'in'a'greater'decrease'in'subsequent' interventions'for'patients'without'findings'that'are'clearly'clinically'important'compared'with'patients' who'have'clinically'important'imaging'findings.' Our!work!and!others!have!shown!that!certain!imaging!findings!are!likely!clinically!more! important!than!others!(e.g.!nerve!root!compression,!moderate!to!severe!central!stenosis,!disc! extrusions).!To!address!this!hypothesis,!we!will!use!the!returned!result!from!radiology!to!categorize! imaging!findings!into!clinically!important!versus!not!clinically!important.!!We!have!identified!central! canal!stenosis,!nerve!root!compression!and!disc!extrusion!(a!type!of!herniation)!as!the!clinically!most! important!imaging!findings.!!This!is!in!contrast!to!findings!that!are!less!clinically!important!(disc!bulge,! disc!narrowing,!Modic!change,!annular!fissure,!etc).! In!each!separate!analysis!of!Aim!1!outcomes,!we!will!add!an!indicator!of!clinical!importance!and!a! clinical!importance!by!treatment!interaction!term!in!the!model.!Similar!to!Aim!2,!primary!inference!will! 2013409426/kj! 29! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! again!be!on!the!interaction!term,!where!we!expect!that!patients!without!these!more!important!findings! will!have!a!significantly!greater!reduction!in!subsequent!interventions!compared!to!patients!with! important!findings.!!! Table!6:!Milestones!for!UH3!(Implementation!Phase)W!Years!2W5! Year of UH3 Milestone Comment Project Year 2 To be completed before Wave 1 of Final testing of intervention deployment randomization scheduled for April, 2014 Planned staggered implementation using Intervention implemented at 40% of clinical sites stepped wedge design will require close monitoring of progress. This will continue the work started as Algorithm finalized for electronic medical record UH2 Milestone #4. Each site will require extraction and tested at all sites a customized algorithm- hence the need for site-specific development and testing We will prepare a manuscript describing Protocol paper submitted for publication our study protocol and procedures. Year 3 Planned staggered implementation using Randomized intervention implemented at 80% of stepped wedge design will require close clinical sites monitoring of progress. Medical record extraction complete for 12mo Data extraction ongoing for duration of outcomes on randomization waves 1-2 project for 12 and 24mo time-points. Comparison of abstraction methods for radiology reports (natural language processing vs. Amazon Turk) Year 4 All clinical sites randomized to Intervention implementation completed intervention by this time. Medical record extraction complete for 12mo outcomes on randomization waves 3-4 and 24mo outcomes on randomization waves 1-2 Abstraction of radiology reports through 12mo for waves 1-4 using preferred method from yr 3 Year 5 Medical record extraction including imaging reports complete for 12mo outcomes on randomization wave 5 and 24 mo outcomes on randomization waves 3-4 2013409426/kj! 30! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Data analysis, manuscript writing & dissemination Manuscripts for 12mo outcomes and of results at national meetings 24mo outcomes submitted for publication Table!7:!Timeline!for!the!UH3!Phase!(Years!2W5)! YEAR222(2014) YEAR232(2015) YEAR242(2016) YEAR252(2017) J F MA M J J A S O N D J F MA M J J A S O N D J F MA M J J A S O N D J F MA M J J A S O N D Randomization2Wave* Wave%#1 Wave%#2 Wave#3 Wave%#4 Wave%#5 Validation2datasets** Validation%sets%transferred%from%sites%to%DCC ✖ ✖ ✖ ✖ ✖ ✖ Comprehensive2datasets*** Comprehensive%datasets%transferred%from%sites%to%DCC%% ✖ ✖ ✖ ✖ ✖ ✖ Data2Quality2Assessment Quality%Assessment%on%set%#1 ✖ Quality%Assessment%between%set%#1%and%#2 ✖ Data2Safety2&2Monitoring2Reports ! ! ! ! ! Reports%every%six%monthes *Insertion%of%Intervention%text%,%randomized%at%clinic%level **Index%file%with%CPT%codes,%LIRE%IDs,%radiology%image%reports%verifying%insertion%of%intervention%text ***Electronic%Medical%Record%and%administrative/billing%data%for%12mo%and%24mo%outcomes 6.!Statistical!Considerations! For!details!regarding!sample!size!and!power!calculations!as!well!as!all!other!statistical!considerations,! please!refer!to!the!LIRE!Statistical!Analysis!Plan.!! 7.!Human!Subjects! Because!the!intervention!will!be!administered!at!the!clinic!level,!consent!of!either!individual!patients!or! primary!caregivers!is!neither!feasible!nor!warranted.!!Moreover,!the!intervention!is!relatively!benign! (the!insertion!of!additional!epidemiological!information!into!the!radiology!report)!and!poses!minimal! risk!to!caregivers!and!patients.!Because!leadership!at!the!Healthcare!Systems!making!up!the! performance!sites!are!enthusiastic!about!incorporating!the!epidemiological!benchmarks!into!their! reports!and!may!well!eventually!adopt!them!regardless!of!the!project,!our!study!simply!allows!us!to! systematically!study!the!effects!of!a!well\controlled!implementation!of!the!insertion!of!the!benchmark! information.!The!randomization!scheme!defines!when!each!clinic!begins!including!the!epidemiological! information!into!the!reports,!with!all!sites!eventually!receiving!the!intervention!of!interest.!! !! 7.1!Human!Subjects!Involvement!and!Characteristics!! Eligibility!criteria:!A!patient!will!be!eligible!for!inclusion!in!the!study!if!they!are!at!least!18!years!old!and! referred!by!their!primary!care!provider!for!plain!films,!CT!or!MR!of!the!lumbar!spine!to!evaluate!low! back!or!leg!pain.!!We!will!access!patient!medical!records!6!months!prior!to!the!index!image!and!for!two! years!after!the!index!image!in!order!to!track!patient!outcomes!before!and!after!the!intervention.! Subjects!will!receive!usual!care,!and!neither!their!diagnostic!evaluation!nor!their!therapy!will!be! 2013409426/kj! 31! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! constrained!by!study!considerations.!We!anticipate!enrolling!~100,000,!patients!who!underwent!lumbar! spine!imaging!examinations!across!four!different!health!systems.!!! 7.2!Research!Data!!! Research!data!will!consist!of!individual!subjects’!medical!record!data!and!information!on!clinics!and! providers.!We!will!collect!all!data!passively!with!automated!data!extractions.!!Data!extracted!from!the! medical!record!will!include!demographic!data,!variables!related!to!imaging,!pharmacy,!procedures,! hospitalizations,!and!other!factors!related!to!healthcare!utilization.!We!will!not!collect!patient!reported! outcomes!unless!they!are!part!of!the!medical!record.!!! We!will!collect!demographic!data!on!primary!care!providers!and!will!code!the!data!in!such!a!way!that!an! individual!practitioner!is!not!identifiable.!!We!will!use!patient!data!to!derive!pre!and!post!randomization! rates!of!spine!related!interventions!(diagnostic!imaging,!opioid!prescriptions,!spine!related!procedures,! physical!therapy!etc.)!among!a!provider’s!patient!panel.!!!!!!! We!will!code!with!a!unique!study!identification!number,!without!reference!to!patient!or!provider! identity.!!The!code!key!will!be!kept!secured!at!the!recruitment!site,!separate!from!the!data.!!Only!the! site!researchers!will!have!access!to!the!code!key!(not!the!researchers!at!the!DCC).!!! 7.3!Potential!risks! The!research!activities!in!this!trial!are!very!low!risk.!Perhaps!the!most!important!risk!is!a!breach!of! confidentiality!of!clinical!information.!!! !! Individual!subjects!will!not!be!contacted!or!consented!for!this!project.!!No!patient!reported!outcomes! are!being!collected!and!so!no!patient!interviews!will!be!performed.!!The!intervention!is!being! administered!at!the!clinic!level;!therefore,!consent!of!either!individual!patients!or!providers!is!neither! feasible!nor!warranted.!Moreover,!the!intervention!is!relatively!benign!(the!insertion!of!epidemiological! benchmark!data!into!the!radiology!report)!and!poses!virtually!no!risk!to!either!providers!or!patients.!!We! will!not!constrain!the!choice!of!tests!or!treatments!offered!to!subjects.!!!! !! The!main!risk!associated!with!this!project!will!be!loss!of!confidentiality!as!medical!record!access!will!be! necessary!in!order!to!assess!the!impact!of!the!intervention.!We!will!make!extensive!efforts!to!assure! that!records!are!kept!in!locked!files!and!are!not!identifiable!to!anyone!but!the!investigators.!All!PHI!will! be!stored!securely!locally.!Non\PHI!data!will!be!uploaded!via!a!web\based!system!to!the!Data! Coordinating!Center!at!the!Center!for!Biomedical!Informatics!and!Biomedical!Statistics!at!the!University! of!Washington.!Anonymized!data!will!be!stored!on!a!server!located!at!Biomedical!Informatics,!where!no! names!or!hospital!numbers!are!included!and!only!study!numbers!will!be!attached!to!the!data!files.!Data! will!be!kept!on!a!server!that!requires!a!password!for!entry!and!in!a!locked!office.!! !! As!the!identities!and!clinical!information!gathered!on!patients!will!be!guarded,!so!too,!will!the!identities! and!data!collected!on!clinic!providers.!!All!identifying!information!will!be!stored!securely!at!the!local! recruitment!sites.!!Only!coded,!limited!data!set!will!be!transferred!to!the!DCC!such!that!an!individual! provider!from!a!given!clinic!within!a!health!system;!cannot!be!identified.!!!! 7.4!Adequacy!of!Protection!Against!Risks! We!anticipate!that!each!site!will!work!within!their!own!health!system!to!identify!primary!care!clinics,! 2013409426/kj! 32! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! primary!care!providers,!and!operationalize!the!technical!aspects!regarding!the!intervention.!!The! intervention!itself!is!the!addition!of!epidemiologic!data!relevant!to!the!imaging!modality!and!age!range! (in!deciles)!of!a!given!patient!for!whom!a!radiologic!image!was!ordered.!!This!data!will!be!automatically! added!to!existing!template!radiology!reports!in!the!intervention!group.!!The!randomization!schedule!will! be!allocated!at!the!clinic!site!rather!than!at!the!individual!patient!or!provider!level.!!Only!group!results! will!be!reported.!! Per!Health!and!Human!Services!Policy!for!Protection!of!Human!Research!Subjects,!Section!46.102.i:!! “Minimal!risk!means!that!the!probability!and!magnitude!of!harm!or!discomfort!anticipated!in!the! research!are!not!greater!in!and!of!themselves!than!those!ordinarily!encountered!in!daily!life!or!during! the!performance!of!routine!physical!or!psychological!examinations!or!tests”.!! !! We!will!seek!a!waiver!of!consent!from!the!IRB’s!at!each!of!the!participating!health!care!systems!since!the! risk!to!individuals!is!minimal,!the!intervention!is!relatively!benign!and!consent!of!patients!and!providers! is!not!practical.!! 7.5!Potential!Benefits!of!the!Proposed!Research!to!the!Subjects!! We!believe!that!the!risks!to!subjects!are!minimal!and!that!the!relevant!knowledge!gains!may!be!great.!! Individual!subjects!in!this!study!are!not!likely!to!benefit!immediately!from!this!new!knowledge,!although! it!could!influence!their!subsequent!treatment,!and!may!influence!the!treatment!of!others!with!a!similar! condition.!!Knowledge!of!benefit!or!lack!thereof)!will!inform!providers!and!patients!in!the!future!about! the!usefulness!providing!epidemiologic!context!to!radiologic!results!in!the!management!of!low!back!pain.!! !! 7.6!Importance!o!the!Knowledge!to!be!Gained! This!study!will!assess!the!impact!epidemiologic!data!tailored!to!radiologic!modality!and!age!range!of!a! given!patient)!has!on!treatment!outcomes!among!those!with!low!back!pain!in!the!primary!care!setting.!! Low!back!pain!is!prevalent,!imaging!is!routinely!used!in!its!assessment!and!evaluation,!and!radiologic! results!can!heavily!inform!providers’!clinical!decision!making.!!Since!the!risks!to!research!subjects!are! minor!and!there!is!the!potential!for!improved!patient!management,!this!research!should!be!pursued.!! All!implementation!sites!now!have!IRB!approval!in!place.!!Mayo!Clinic!and!Henry!Ford!each!went! through!their!own!IRB!approval!process!for!minimal!risk!applications!and!received!approval.!!Group! Health!Cooperative!has!agreed!to!the!IRB!of!record!for!the!study!and!UW!and!Kaiser’s!IRB’s!both!ceded! authority!to!them!for!monitoring!the!study!moving!forward.! 8.0!Data!and!Safety!Monitoring!! We!have!drafted!a!data!safety!and!monitoring!plan!(DSMP)!and!have!designated!two!Safety!Officers,! Steven!Atlas,!MD!and!Constantine!Gatsonis,!PhD who!has!agreed!to!review!study!data!at!regular! intervals!for!safety!concerns.!! 2013409426/kj! 33! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! References! 1.! Lidgren!L.!The!bone!and!joint!decade!2000\2010.!Bull!World!Health!Organ.!2003;81(9):629.! PMCID:!2572548.! 2.! Initial!National!Priorities!for!Comparative!Effectiveness!Research.!Washington!DC:!Institute!of! Medicine;!2009!June!30,!2009!Contract!No.:!Document!Number|.! 3.! Schappert!SM,!Rechtsteiner!EA.!Ambulatory!medical!care!utilization!estimates!for!2006.!Natl! Health!Stat!Report.!2008(8):1\29.! 4.! Owens!DK,!Qaseem!A,!Chou!R,!Shekelle!P.!High\value,!cost\conscious!health!care:!concepts!for! clinicians!to!evaluate!the!benefits,!harms,!and!costs!of!medical!interventions.!Ann!Intern! Med.154(3):174\80.! 5.! Chou!R,!Qaseem!A,!Owens!DK,!Shekelle!P.!Diagnostic!imaging!for!low!back!pain:!advice!for!high\ value!health!care!from!the!American!College!of!Physicians.!Ann!Intern!Med.154(3):181\9.! 6.! Foundation!ABoIM,!Physicians!ACo.!Choosing!Wisely!Campaign.!2012![updated!2012;!cited];! Available!from:!choosingwisely.org/.! 7.! Martin!BI,!Deyo!RA,!Mirza!SK,!Turner!JA,!Comstock!BA,!Hollingworth!W,!et!al.!Expenditures!and! health!status!among!adults!with!back!and!neck!problems.!JAMA.!2008;299(6):656\64.! 8.! Jarvik!JJ,!Hollingworth!W,!Heagerty!P,!Haynor!DR,!Deyo!RA.!The!Longitudinal!Assessment!of! Imaging!and!Disability!of!the!Back!(LAIDBack)!Study:!baseline!data.!Spine.!2001;26(10):1158\66.! 9.! Boden!SD,!Davis!DO,!Dina!TS,!Patronas!NJ,!Wiesel!SW.!Abnormal!magnetic\resonance!scans!of! the!lumbar!spine!in!asymptomatic!subjects.!A!prospective!investigation.!J!Bone!Joint!Surg![Am].! 1990;72(3):403\8.! 10.! Jensen!MC,!Brant\Zawadzki!MN,!Obuchowski!N,!Modic!MT,!Malkasian!D,!Ross!JS.!Magnetic! resonance!imaging!of!the!lumbar!spine!in!people!without!back!pain![see!comments].!N!Engl!J! Med.!1994;331(2):69\73.! 11.! Jarvik!JG,!Hollingworth!W,!Martin!B,!Emerson!SS,!Gray!DT,!Overman!S,!et!al.!Rapid!magnetic! resonance!imaging!vs!radiographs!for!patients!with!low!back!pain:!a!randomized!controlled!trial.! JAMA.!2003;289(21):2810\8.! 12.! Lurie!JD,!Birkmeyer!NJ,!Weinstein!JN.!Rates!of!advanced!spinal!imaging!and!spine!surgery.!Spine! (Phila!Pa!1976).!2003;28(6):616\20.! 13.! Aberle!DR,!Adams!AM,!Berg!CD,!Black!WC,!Clapp!JD,!Fagerstrom!RM,!et!al.!Reduced!lung\cancer! mortality!with!low\dose!computed!tomographic!screening.!N!Engl!J!Med.!2011;365(5):395\409.! 14.! Johnson!PT,!Horton!KM,!Megibow!AJ,!Jeffrey!RB,!Fishman!EK.!Common!incidental!findings!on! MDCT:!survey!of!radiologist!recommendations!for!patient!management.!J!Am!Coll!Radiol.! 2011;8(11):762\7.! 15.! Iwabuchi!Y,!Hanamure!Y,!Ueno!K,!Fukuda!K,!Furuta!S.!Clinical!significance!of!asymptomatic!sinus! abnormalities!on!magnetic!resonance!imaging.!Arch!Otolaryngol!Head!Neck!Surg.! 1997;123(6):602\4.! 16.! Havas!TE,!Motbey!JA,!Gullane!PJ.!Prevalence!of!incidental!abnormalities!on!computed! tomographic!scans!of!the!paranasal!sinuses.!Arch!Otolaryngol!Head!Neck!Surg.!1988;114(8):856\ 9.! 17.! Roland!M,!van!Tulder!M.!Should!radiologists!change!the!way!they!report!plain!radiography!of!the! spine?!Lancet.!1998;352(9123):229\30.! 2013409426/kj! 34! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! 18.! Modic!MT,!Obuchowski!NA,!Ross!JS,!Brant\Zawadzki!MN,!Grooff!PN,!Mazanec!DJ,!et!al.!Acute!low! back!pain!and!radiculopathy:!MR!imaging!findings!and!their!prognostic!role!and!effect!on! outcome.!Radiology.!2005;237(2):597\604.! 19.! Kendrick!D,!Fielding!K,!Bentley!E,!Kerslake!R,!Miller!P,!Pringle!M.!Radiography!of!the!lumbar!spine! in!primary!care!patients!with!low!back!pain:!randomised!controlled!trial.!BMJ.! 2001;322(7283):400\5.!PMCID:!26570.! 20.! Kerry!S,!Hilton!S,!Dundas!D,!Rink!E,!Oakeshott!P.!Radiography!for!low!back!pain:!a!randomised! controlled!trial!and!observational!study!in!primary!care.!Br!J!Gen!Pract.!2002;52(479):469\74.! PMCID:!1314322.! 21.! Deyo!RA,!Diehl!AK,!Rosenthal!M.!Reducing!roentgenography!use.!Can!patient!expectations!be! altered?!Arch!Intern!Med.!1987;147(1):141\5.! 22.! Gilbert!FJ,!Grant!AM,!Gillan!MG,!Vale!LD,!Campbell!MK,!Scott!NW,!et!al.!Low!back!pain:!influence! of!early!MR!imaging!or!CT!on!treatment!and!outcome\\multicenter!randomized!trial.!Radiology.! 2004;231(2):343\51.! 23.! Ash!LM,!Modic!MT,!Obuchowski!NA,!Ross!JS,!Brant\Zawadzki!MN,!Grooff!PN.!Effects!of!diagnostic! information,!per!se,!on!patient!outcomes!in!acute!radiculopathy!and!low!back!pain.!AJNR!Am!J! Neuroradiol.!2008;29(6):1098\103.! 24.! Chou!R,!Qaseem!A,!Owens!DK,!Shekelle!P.!Diagnostic!imaging!for!low!back!pain:!advice!for!high\ value!health!care!from!the!American!College!of!Physicians.!Ann!Intern!Med.!2011;154(3):181\9.! 25.! Deyo!RA,!Mirza!SK,!Turner!JA,!Martin!BI.!Overtreating!chronic!back!pain:!time!to!back!off?!J!Am! Board!Fam!Med.!2009;22(1):62\8.!PMCID:!2729142.! 26.! Friedly!J,!Chan!L,!Deyo!R.!Increases!in!lumbosacral!injections!in!the!Medicare!population:!1994!to! 2001.!Spine!(Phila!Pa!1976).!2007;32(16):1754\60.! 27.! Luo!X,!Pietrobon!R,!Hey!L.!Patterns!and!trends!in!opioid!use!among!individuals!with!back!pain!in! the!United!States.!Spine!(Phila!Pa!1976).!2004;29(8):884\90;!discussion!91.! 28.! Deshpande!A,!Furlan!A,!Mailis\Gagnon!A,!Atlas!S,!Turk!D.!Opioids!for!chronic!low\back!pain.! Cochrane!Database!Syst!Rev.!2007(3):CD004959.! 29.! Airaksinen!O,!Brox!JI,!Cedraschi!C,!Hildebrandt!J,!Klaber\Moffett!J,!Kovacs!F,!et!al.!Chapter!4.! European!guidelines!for!the!management!of!chronic!nonspecific!low!back!pain.!Eur!Spine!J.! 2006;15!Suppl!2:S192\300.! 30.! Deyo!RA,!Gray!DT,!Kreuter!W,!Mirza!S,!Martin!BI.!United!States!trends!in!lumbar!fusion!surgery! for!degenerative!conditions.!Spine!(Phila!Pa!1976).!2005;30(12):1441\5;!discussion!6\7.! 31.! Mirza!SK,!Deyo!RA.!Systematic!review!of!randomized!trials!comparing!lumbar!fusion!surgery!to! nonoperative!care!for!treatment!of!chronic!back!pain.!Spine!(Phila!Pa!1976).!2007;32(7):816\23.! 32.! Deyo!RA,!Mirza!SK,!Martin!BI,!Kreuter!W,!Goodman!DC,!Jarvik!JG.!Trends,!major!medical! complications,!and!charges!associated!with!surgery!for!lumbar!spinal!stenosis!in!older!adults.! JAMA.!2010;303(13):1259\65.!PMCID:!2885954.! 33.! Martin!BI,!Turner!JA,!Mirza!SK,!Lee!MJ,!Comstock!BA,!Deyo!RA.!Trends!in!health!care! expenditures,!utilization,!and!health!status!among!US!adults!with!spine!problems,!1997\2006.! Spine!(Phila!Pa!1976).!2009;34(19):2077\84.! 34.! Bartleson!JD.!Evidence!for!and!against!the!use!of!opioid!analgesics!for!chronic!nonmalignant!low! back!pain:!a!review.!Pain!Med.!2002;3(3):260\71.! 35.! Paulozzi!LJ,!Ryan!GW.!Opioid!analgesics!and!rates!of!fatal!drug!poisoning!in!the!United!States.!Am! J!Prev!Med.!2006;31(6):506\11.! 2013409426/kj! 35! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! 36.! McCullough!BJ,!Johnson!GR,!Martin!BI,!Jarvik!JG.!Lumbar!MR!imaging!and!reporting! epidemiology:!do!epidemiologic!data!in!reports!affect!clinical!management?!Radiology.! 2012;262(3):941\6.!PMCID:!3285226.! 37.! Scholich!SL,!Hallner!D,!Wittenberg!RH,!Hasenbring!MI,!Rusu!AC.!The!relationship!between!pain,! disability,!quality!of!life!and!cognitive\behavioural!factors!in!chronic!back!pain.!Disabil!Rehabil.! 38.! Bick!D,!Dimmock!D.!Whole!exome!and!whole!genome!sequencing.!Curr!Opin!Pediatr.23(6):594\ 600.! 39.! Majewski!J,!Schwartzentruber!J,!Lalonde!E,!Montpetit!A,!Jabado!N.!What!can!exome!sequencing! do!for!you?!J!Med!Genet.48(9):580\9.! 40.! Brown!CA,!Lilford!RJ.!The!stepped!wedge!trial!design:!a!systematic!review.!BMC!Med!Res! Methodol.!2006;6:54.!PMCID:!1636652.! 41.! Hussey!MA,!Hughes!JP.!Design!and!analysis!of!stepped!wedge!cluster!randomized!trials.!Contemp! Clin!Trials.!2007;28(2):182\91.! 42.! Coventry!MB,!Ghormley!RK,!Kernohan!JW.!Intervertebral!disc;!its!microscopic!anatomy!and! pathology;!anatomy,!development,!and!physiology.!J!Bone!Joint!Surg.!1945;27:105.! 43.! McRae!DL.!Asymptomatic!intervertebral!disc!protrusions.!Acta!Radiologica.!1956;46:9\27.! 44.! Hitselberger!WE,!Witten!RM.!Abnormal!myelograms!in!asymptomatic!patients.!J!Neurosurg.! 1968;28(3):204\6.! 45.! Wiesel!S,!Tsourmas!N,!Feffer!H,!Citrin!C,!Patronas!N.!A!study!of!computer\assisted!tomography.!I.! The!incidence!of!positive!CAT!scans!in!an!asymptomatic!group!of!patients.!Spine.!1984;9:549\51.! 46.! Powell!MC,!Wilson!M,!Szypryt!P,!Symonds!EM,!Worthington!BS.!Prevalence!of!lumbar!disc! degeneration!observed!by!magnetic!resonance!in!symptomless!women.!Lancet.! 1986;2(8520):1366\7.! 47.! Weinreb!JC,!Wolbarsht!LB,!Cohen!JM,!Brown!CE,!Maravilla!KR.!Prevalence!of!lumbosacral! intervertebral!disk!abnormalities!on!MR!images!in!pregnant!and!asymptomatic!nonpregnant! women.!Radiology.!1989;170(1!Pt!1):125\8.! 48.! Szypryt!EP,!Twining!P,!Mulholland!RC,!Worthington!BS.!The!prevalence!of!disc!degeneration! associated!with!neural!arch!defects!of!the!lumbar!spine!assessed!by!magnetic!resonance!imaging.! Spine.!1989;14(9):977\81.! 49.! Tertti!MO,!Salminen!JJ,!Paajanen!HE,!Terho!PH,!Kormano!MJ.!Low\back!pain!and!disk! degeneration!in!children:!a!case\control!MR!imaging!study.!Radiology.!1991;180(2):503\7.! 50.! Bartolozzi!C,!Caramella!D,!Zampa!V,!Dal!Pozzo!G,!Tinacci!E,!Balducci!F.![The!incidence!of!disk! changes!in!volleyball!players.!The!magnetic!resonance!findings].!Radiol!Med!(Torino).! 1991;82(6):757\60.! 51.! Buirski!G,!Silberstein!M.!The!symptomatic!lumbar!disc!in!patients!with!low\back!pain.!Magnetic! resonance!imaging!appearances!in!both!a!symptomatic!and!control!population.!Spine.! 1993;18(13):1808\11.! 52.! Parkkola!R,!Rytokoski!U,!Kormano!M.!Magnetic!resonance!imaging!of!the!discs!and!trunk!muscles! in!patients!with!chronic!low!back!pain!and!healthy!control!subjects.!Spine.!1993;18(7):830\6.! 53.! Boos!N,!Rieder!R,!Schade!V,!Spratt!KF,!Semmer!N,!Aebi!M.!1995!Volvo!Award!in!clinical!sciences.! The!diagnostic!accuracy!of!magnetic!resonance!imaging,!work!perception,!and!psychosocial! factors!in!identifying!symptomatic!disc!herniations.!Spine.!1995;20(24):2613\25.! 54.! Salminen!JJ,!Erkintalo!M,!Laine!M,!Pentti!J.!Low!back!pain!in!the!young.!A!prospective!three\year! follow\up!study!of!subjects!with!and!without!low!back!pain.!Spine.!1995;20(19):2101\7;! discussion!8.! 2013409426/kj! 36! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! 55.! Burns!JW,!Loecker!TH,!Fischer!JR,!Jr.,!Bauer!DH.!Prevalence!and!significance!of!spinal!disc! abnormalities!in!an!asymptomatic!acceleration!subject!panel.!Aviat!Space!Environ!Med.! 1996;67(9):849\53.! 56.! Paajanen!H,!Erkintalo!M,!Parkkola!R,!Salminen!J,!Kormano!M.!Age\dependent!correlation!of!low\ back!pain!and!lumbar!disc!regeneration.!Arch!Orthop!Trauma!Surg.!1997;116(1\2):106\7.! 57.! Savage!RA,!Whitehouse!GH,!Roberts!N.!The!relationship!between!the!magnetic!resonance! imaging!appearance!of!the!lumbar!spine!and!low!back!pain,!age!and!occupation!in!males.!Eur! Spine!J.!1997;6(2):106\14.! 58.! Stadnik!TW,!Lee!RR,!Coen!HL,!Neirynck!EC,!Buisseret!TS,!Osteaux!MJ.!Annular!tears!and!disk! herniation:!prevalence!and!contrast!enhancement!on!MR!images!in!the!absence!of!low!back!pain! or!sciatica.!Radiology.!1998;206(1):49\55.! 59.! Weishaupt!D,!Zanetti!M,!Hodler!J,!Boos!N.!MR!imaging!of!the!lumbar!spine:!prevalence!of! intervertebral!disk!extrusion!and!sequestration,!nerve!root!compression,!end!plate!abnormalities,! and!osteoarthritis!of!the!facet!joints!in!asymptomatic!volunteers.!Radiology.!1998;209(3):661\6.! 60.! Rao!VM,!Levin!DC,!Parker!L,!Frangos!AJ,!Sunshine!JH.!Trends!in!utilization!rates!of!the!various! imaging!modalities!in!emergency!departments:!nationwide!Medicare!data!from!2000!to!2008.!J! Am!Coll!Radiol.!2011;8(10):706\9.! 61.! Picano!E.!Sustainability!of!medical!imaging.!BMJ.!2004;328:578\80.! 62.! Fineberg!H.!Computerized!cranial!tomography:!effect!on!diagnostic!and!therapeutic!plans.!JAMA.! 1977;38:224\7.! 63.! Fryback!D,!Thornbury!J.!The!efficacy!of!diagnostic!imaging.!Med!Decis!Making.!1991;11:88\94.! 64.! Sox!!HJ,!Margulies!I,!Sox!C.!Psychologically!mediated!effects!of!diagnostic!tests.!Ann!Intern!Med.! 1981;95:680\5.! 65.! Franklin!GM,!Stover!BD,!Turner!JA,!Fulton\Kehoe!D,!Wickizer!TM.!Early!opioid!prescription!and! subsequent!disability!among!workers!with!back!injuries:!the!Disability!Risk!Identification!Study! Cohort.!Spine!(Phila!Pa!1976).!2008;33(2):199\204.! 66.! Franklin!GM,!Mai!J,!Turner!J,!Sullivan!M,!Wickizer!T,!Fulton\Kehoe!D.!Bending!the!prescription! opioid!dosing!and!mortality!curves:!impact!of!the!Washington!State!opioid!dosing!guideline.!Am!J! Ind!Med.55(4):325\31.! 67.! Kallmes!DF,!Comstock!BA,!Heagerty!PJ,!Turner!JA,!Wilson!DJ,!Diamond!TH,!et!al.!A!randomized! trial!of!vertebroplasty!for!osteoporotic!spinal!fractures.!N!Engl!J!Med.!2009;361(6):569\79.! 68.! Jarvik!JG,!Comstock!BA,!Kliot!M,!Turner!JA,!Chan!L,!Heagerty!PJ,!et!al.!Surgery!versus!non\surgical! therapy!for!carpal!tunnel!syndrome:!a!randomised!parallel\group!trial.!Lancet.! 2009;374(9695):1074\81.! 69.! Wei!LJ,!Lachin!JM.!Properties!of!the!urn!randomization!in!clinical!trials.!Control!Clin!Trials.! 1988;9(4):345\64.! 70.! Health!WK.!Drug!facts!and!compaarisons;!2004!Contract!No.:!Document!Number|.! 71.! Efron!B,!Stein!C.!The!jackknife!estimate!of!variance.!Annals!of!Statistics.!1981;9(3):586\96.! 72.! Briggs!A,!Claxton!K,!Sculpher!M.!Decision!modeling!for!health!economic!evaluation.!Oxford:! Oxford!University!Press;!2008.! 2013409426/kj! 37! Version!3! LIRE! ! ! ! ! ! !!!!!!!!!!!!!!Study!Protocol!! Appendices! Appendix!A:!Article!Extraction!form!(Working!Group!1) ! Appendix!B:!Literature!Search!and!Articles!used!in!Intervention!text!(Working!Group!1) ! Appendix!C:!Pilot!Implementation!Site!Checklist!(Working!Group!2) ! Appendix!D:!Literature!Search!for!RVUWbased!Assessment!(Working!Group!3) ! 2013409426/kj! 38! Version!3! ! LIRE%Article%Abstraction%Form% Abstractor!Name:! ! ! ! Lead!author:! ! ! Journal:! ! Year:! ! 1.!Review!Article!__ __yes!(do!not!abstract!prevalence!data)!!!!!!!!__ __no! 2.!Clearly!asymptomatic!LBP!population:! no!(do!not!abstract)!!!! yes!!!!! ambiguous! 3.!Article!not!relevant!for!other!reason:!! yes!(do!not!abstract)!! no! Prevalence!of!Imaging!Findings!in!People!Without!Low!Back!Pain!(include!point! estimate!and!if!available!CI)! Finding! Overall!! Age! Age! Age! Age! Age! Age! Mean! range!! range! range! range! range! range! Age! ! ! ! ! ! ! range! #!subjects! ! ! ! ! ! ! ! without!LBP! Disc! ! ! ! ! ! ! ! degeneration! Disc!signal! ! ! ! ! ! ! ! loss! Disc!height! ! ! ! ! ! ! ! loss! ! ! ! ! ! ! ! Disc!bulge! Disc! ! ! ! ! ! ! ! protrusion! Annular! ! ! ! ! ! ! ! fissure! ! ! ! ! ! ! ! Modic!1! change! Modic!2! ! ! ! ! ! ! ! change! Facet! ! ! ! ! ! ! ! degeneration! (any)! Facet! ! ! ! ! ! ! ! degeneration! (modSsev)! Other!1! ! ! ! ! ! ! ! ! ! ! ! ! ! ! Other!2! Comments!(e.g.!population!characteristics,!potential!problems!using!data):! ! ! Last!modified!12/5/12!2:25pm! LIRE literature search strategy for Intervention Text 1. The LIRE WG1 team worked with University of Washington librarians to develop our search strategy and list of search terms. 2. Using PubMed, we used the following terms for our initial WG1 literature search: ("Morbidity"[Mesh] OR ("epidemiology"[Subheading] OR "epidemiology"[All Fields] OR "prevalence"[All Fields] OR "prevalence"[MeSH Terms]) OR "Epidemiology"[Mesh] OR "epidemiology"[Subheading] OR "Epidemiologic Factors"[Mesh] OR "Incidental Findings"[Mesh] OR incidental[All Fields] OR "Asymptomatic Diseases"[Mesh] OR asymptomatic[All Fields] OR "Unnecessary Procedures"[MeSH Terms]) AND ("Lumbar Vertebrae"[Mesh] OR "Low Back Pain"[Mesh] OR "Intervertebral Disc Displacement"[Mesh] OR "Sciatica"[Mesh] OR "Spinal Stenosis"[Mesh] OR "Synovial Cyst"[Mesh] OR "Scoliosis"[Mesh] OR "Spondylolysis"[Mesh] OR "Spinal Osteochondrosis"[Mesh] OR "disc degeneration"[All Fields] OR "disc height loss"[All Fields] OR "disc bulge"[All Fields] OR "disc protrusion"[All Fields] OR annular[All Fields] OR anular[All Fields] OR "high intensity zone"[All Fields] OR anulus[All Fields] OR annulus[All Fields] OR listhesis[All Fields] OR (disc[All Fields] AND ("desiccation"[MeSH Terms] OR "desiccation"[All Fields] OR "dessication"[All Fields])) OR "disc dehydration"[All Fields] OR (modic[All Fields] AND endplate[All Fields] AND ("Change"[Journal] OR "change"[All Fields])) OR "nerve root displacement"[All Fields] OR "nerve root compression"[All Fields] OR "disc sequestration"[All Fields] OR "intravertebral herniations"[All Fields] OR (intravertebral[All Fields] AND ("hernia"[MeSH Terms] OR "hernia"[All Fields])) OR (intradiscal[All Fields] AND ("hernia"[MeSH Terms] OR "hernia"[All Fields])) OR (intradiscal[All Fields] AND ("hernia"[MeSH Terms] OR "hernia"[All Fields] OR "herniation"[All Fields])) OR "intradural herniation"[All Fields] OR (intravertebral[All Fields] AND ("hernia"[MeSH Terms] OR "hernia"[All Fields])) OR (intravertebral[All Fields] AND ("hernia"[MeSH Terms] OR "hernia"[All Fields] OR "herniation"[All Fields])) OR "prolapsed disc"[All Fields] OR "radial fissure"[All Fields]) AND ("Magnetic Resonance Imaging"[Mesh] OR "Tomography, X-Ray Computed"[Mesh] OR "Radiography"[Mesh]) 3. Using PubMed advance search tool, the above search criteria generates N=2957 identified articles. LIRE WG1 used the following reviewers and inclusion criteria to evaluate the abstracts. Start End reviewer 1 reviewer 2 1 330 avins bresnahan 331 660 bresnahan chen 661 990 chen deyo 991 1320 deyo halabi 1321 1650 halabi jarvik 1651 1980 jarvik kallmes 1981 2310 kallmes luetmer 2311 2640 luetmer turner 2641 2957 turner avins Inclusion criteria for articles: 1. article included subjects without low back pain (LBP) 2. listed prevalence of imaging finding in patients without LBP 3. subjects were >=18 (exclude series that were strictly peds) 4. subjects were human and alive (no cadaver or animal studies) 5. imaging study prevalence data was for either MR, CT or plain film 6. prevalence for at least one of the following was included : 4. spinal stenosis, 5. disc bulge, 6. disc protrusion, 7. disc extrusion, 8. disc herniation, 9. disc degeneration, 10. disc dessication (or dehydration), 11. disc height loss, 12. nerve root involvement (contact, displacement or compression), 13. anular fissure (or anular tear or HIZ), 14. spondylolysis, 15. spondylolisthesis, 16. modic change, 17. SchmorlÕs node, 18. synovial cyst, 19. osteochondrosis References(for(LIRE(Intervention(Text( 1. Boden(SD,(Davis(DO,(Dina(TS,(Patronas(NJ,(Wiesel(SWAbno .( rmal(magnetic> resonance(scans(of(the(lumbar(spine(in(asymptomatic(subjects.(A(prospective( investigation.(J(Bone(Joint(Surg(Am.(1990(Mar;72(3):403>8.( 2. Boden(SD,(Riew(KD,(Yamaguchi(K,(Branch(TP,(Schellinger(D,(Wiesel(SW.( Orientation(of(the(lumbar(facet(joints:(association(with(degenerative(disc( disease.(J(Bone(Joint(Surg(Am.(1996(Mar;78(3):40>311.( 3. Boos(N,(Rieder(R,(Schade(V,(Spratt(KF,(Semmer(N,(Aebi(M. (1995(Volvo(Award(in( clinical(sciences.(The(diagnostic(accuracy(of(magnetic(resonance(imaging,(work( perception,(and(psychosocial(factors(in(identifying(symptomatic(disc(herniations.( Spine((Phila(Pa(1976).(1995(Dec(15;20(24):2613>25.( 4. Capel(A.(et(al.(Magnetic(resonance(study(of(lumbar(disks(in(female(dancers.(Am(J( Sports(Med.(2009(Jun;37(6):1208>13.(Epub(2009(May(5(. 5. Carragee(E,(Alamin(T,(Cheng(I,(Franklin(T,(van(den(Haak(E,(Hurwitz(E.(Are(fir> st time(episodes(of(serious(LBP(associated(with(new(MRI(findings?(Spine(J.(2006( Nov>Dec;6(6):624>35.(Epub(2006(Oct(11(. 6. Danielson(B(et(al.(Axially(loaded(magnetic(resonance(image(of(the(lumbar(spine( in(asymptomatic(individuals.(Spine((Phila(Pa(1976).(2001(Dec(1;26(23):260>16.( 7. Dora(C(et(al.(The(significance(of(spinal(canal(dimensions(in(discriminating( symptomatic(from(asymptomatic(disc(herniations.(Eur(Spine(J.(2002( Dec;11(6):575>81.(Epub(2002(Aug(28(. 8. Edmondston(et(al.(MRI(evaluation(of(lumbar(spine(flexion(and(extension(in( asymptomatic(individuals.(Manual(Therapy((2000)(5(3),(158>164( 9. Feng(T,(Zhao(P,(Liang(G.(Clinical(significance(on(protruded(nucleus(pulposus:(a( comparative(study(of(44(patients(with(lumbar(intervertebral(disc(protrusion(and( 73(asymptomatic(control(in(tridimentional(computed(tomography.Z(hongguo( Zhong(Xi(Yi(Jie(He(Za(Zhi([Chinese(Journal(of(Integrated(Traditiona l(and(Western( Medicine].(2000(May;20(5):347>9.( 10. Greenberg(JO,(Schnell(RG.(Magnetic(resonance(imaging(of(the(lumbar(spine(in( asymptomatic(adults.(Cooperative(study>>American(Society(of(Neuroimaging.(J( Neuroimaging.(1991(Feb;1(1):2>7.( ( 11. Hamanishi(C,(Kawabata(T,(Yosii(T,(Tanaka(S.(Schmorl's(nodes(on(magnetic( resonance(imaging.(Their(incidence(and(clinical(relevance.(Spine((Phila(Pa(1976).( 1994(Feb(15;19(4):450>3.( 12. Healy(JF,(Healy(BB,(Wong(WH,(Olson(EM.(Cervical(and(lumbar(MRI(in( asymptomatic(older(male(lifelong(athletes:(frequency(of(degenerative(findings.(J( Comput(Assist(Tomogr.(1996(Jan>Feb;20(1):107>12.( 13. ( (Jarvik(JJ,(Hollingworth(W,(Heagerty(P,(Haynor(DR,(Deyo(RA.(The(Longitudinal( Assessment(of(Imaging(and(Disability(of(the(Back((LAIDBack) (Study:(baseline( data.(Spine((Phila(Pa(1976).(2001(May(15;26(10):115>866.( 14. Jensen(MC,(Brant>Zawadzki(MN,(Obuchowski(N,(Modic(MT,(Malkasian(D,(Ross(JS.( Magnetic(resonance(imaging(of(the(lumbar(spine(in(people(without(back(pain.(N( Engl(J(Med.(1994(Jul(14;331(2):69>73.( 15. ( Kalichman,(L,(Li(L,(Hunter(DJ,(Been(E.(Association(between(computed( tomography>evaluated(lumbar(lordosis(and(features(of(spinal(degeneration,( evaluated(in(supine(position.(Spine(J.(2010(Apr;11(4);(308>15.( 16. Kanayama(M,(Togawa(D,(Takahashi(C,(Terai(T,(Hashimoto(T.(Cross>sectional( magnetic(resonance(imaging(study(of(lumbar(disc(degeneration(in ( 200(healthy(individuals.(J(Neurosurg(Spine.(2009(Oct;11(4):50>17.( 17. Karakida(O,(Ueda(H,(Ueda(M,(Miyasaka (T.(Diurnal(T2(value(changes(in(the(lumbar( intervertebral(discs.(Clin(Radiol.(2003(May;58(5):389>92.( 18. Kjaer(P,(Leboeuf>Yde(C,(Korsholm(L,(Sorensen(JS,(Bendix(TMag .( netic(resonance( imaging(and(low(back(pain(in(adults:(a(diagnostic(imaging(study(of(40>year>old( men(and(women.(Spine((Phila(Pa(1976).(2005(May(15;30(10):1173>80.( 19. Ranson(et(al.(Magnetic(resonance(imaging(of(the(lumbar(spine(in(asymptomatic( professional(fast(bowlers(in(cricket.(J(Bone(Joint(Surg([Br](2005;8>7B:1111>16.( 20. Savage(RA,(Whitehouse(GH,(Roberts(N.(The(relationship(between(the(magnetic( resonance(imaging(appearance(of(the(lumbars(pine(and(low(back(pain,(age(and( occupation(in(males.(Eur(Spine(J.(1997;6(2):106>14.( 21. Silcox(DH(3rd,(Horton(WC,(Silverstein(AM.(MRI(of(lumbar(intervertebral(discs.( Diurnal(variations(in(signal(intensities.(Spine((Phila(Pa(1976).(1995(Apr( 1;20(7):807>11;(discussion(811>2.( 22. Stadnik(TW,(Lee(RR,(Coen(HL,(Neirynck(EC,(Buisseret(TS,(Osteaux(MJAnnular .( ( tears(and(disk(herniation:(prevalence(and(contrast(enhancement(on(MR(images( in(the(absence(of(low(back(pain(or(sciatica. (Radiology.(1998(Jan;206(1):49>55.( ( 23. Szypryt(EP,(Twining(P,(Mulholland(RC,(Worthington(BS.(The(prevalence(of(disc( degeneration(associated(with(neural(arch(defects(of(th(elumbar(spine(assessed( by(magnetic(resonance(imaging.(Spine((Phila(Pa(1976).(1989(Sep;14(9):977>81.( 24. Takatalo(J(et(al.(Association(of(modic(changes,(Schmorl’s (nodes,(spondylolytic( defects,(high>intensity(zone(lesions,(disc(herniations,(and(radial(tears(with(low( back(symptom(severity(among(young(Finnish(adults.((Spine.(2012(June( 15;37(14)1231>9.( 25. Weinreb(JC,(Wolbarsht(LB,(Cohen(JM,(Brown(CE,(Maravilla(KR. Pr(evalence(of( lumbosacral(intervertebral(disk(abnormalities(on(MR(images(in(pregnant(and( asymptomatic(nonpregnant(women.(Radiology.(1989(Jan;170(1(Pt(1):125>8.( 26. Weishaupt(D,(Zanetti(M,(Hodler(J,(Boos(N.MR( ( imaging(of(the(lumbar(spine:( prevalence(of(intervertebral(disk(extrusion(and(sequestration,(nerve(root( compression,(end(plate(abnormalities,(and(osteoarthritis(of(the(facet(joints(in( asymptomatic(volunteers.(Radiology.(1998(Dec;209(3):661( 27. Zobel(BB(et(al.(A(Magnetic(resonance(imaging(quantification(of(early(lumbar( intervertebral(disc(degeneration(in(healthy(your(adults.((Spine.(2012(June( 15;37(14)1224>30.( ( LIRE%Projec% t % % % % Intervention%Pilot%Testing%P%lan% % April% ;% May%201% 3 1.#LIRE#study#design#overview# LIRE%is% % clu a ster%randomized%trial%to%study%the%impact%of%i texnser t% detsicng ri%a ptio% n% of%ag;especific%prevalence%of%imaging%findings%among%asymptomatic,% %su into% bjects lumbar%spine%imaging%rep.%We%a orts im%to%st % suubdy sequent% back;related% interventions%(narcotic%prescriptions,%subsequent%imaging,%injections,%surgeries,%etc)% over%thefo%llowing% 1%and%2%y.e% W ares%are%using%a%stepped%wedge%design%and%will% randomize%each%clinic%to%begin%implementing%the%intervention%textp%a re;t%one%of%5% specified%dates% :% April%1,%2014 % October%1,%20% 14 April%1,%2015 % October%1,%20% 15 April%1,%2016 % At%clinics%randomized%to%receive%the%intervention,%the%text%will%be%inserted%into%the% radiology%report%whenever%one%of%the%follow coindge% s% CP isT%%generate7d2:1 %00,% 72110,%72114,%72131,1 %3722,%72133,%72148,%72149,%72158,%72080%for% %1p8a% tients and%older.%Once%the%intervention%is%implemented%at%a%given%clinic,%it%will%remain%“on”% indefinitely%unless%the%study%stopped%earlys% onfos% r%orf% easafety%or%the%health% %system wants%to%turn%it%“off”%after%the%study%period%(April%2018)%although%this%is%not% anticipated.% 2.#Pilot#test#implementation# The%goal%of%this%pilot%implementation%is%to%verify%clinic%eligibility%for%the%LIRE%study% and%to%demonstrate%successful%insert% tiohn%o e%LIfRE%text%into%radiology%reports.%We% are%capturing%data%in%Table%2%of%this%document%that%will%be%included%in%the%UH2% progress%report%to%NIH%in%our%request%foor% %tfruanndsiintigo% n% t to%the%UH3%phase.% (So,%this%is%important% …!)% Your%approach%towardse%p %thilot%implementation%needs%to%include%the%f %ollowing: ! Assess%clinic%eligibility:%for%each%clinic%identified%in%Table%2,%verify%the% questions%in%Table%1%and%mark%in%Table%2%whether%the%site%meets%or%does%not% meet%the%inclusion%criteria%or%if%you%a e.%P re%u lease%p nsur rovide%comments%if% necessary.% ! Proof%of%implementation%of%intervention%text%in%Section%3%(Options%1%or%2):% using%dummy%records,%successfully%demonstrate%that%for%an%eligible%CPT%code% (and%patient%a>g% e1% 8)%the%text%has%been%inserted%intoo% logthye%%rreapdoirt.% An%example%printout%of%the%text%would%suffice%to%show%this.% ! Complete%pilot%implementation%befMoare% y#31,#2013.%This%is%a%hard%deadline% as%the%transition%report%must%be%written%and%submitted%to%the%NIH%soon%after% this.% ! Since%we%do%not%hav% aep% pIrRoB val,%this%test%needs%to%be%conducted%%in%a%test environment%or%by%using%a%dummy%case.% # LIRE%Projec% t % % % % Intervention%Pilot%Testing%P%lan% % April% ;% May%201% 3 Table#1.#LIRE#clinic#inclusion#criteria# Intervention implementation inclusion criteria to verify in pilot test Required Inclusion Criteria 1. Can the intervention text be delivered based upon a specific CPT code (Xray: 72100, 72110, 72114, 72080; CT: 72131, 72132, 72133; MR: 72148, 72149, 72158)? 2. Can modality-specific (Xray, CT, MR) intervention text be inserted? 3. Can the intervention text be delivered based upon patient age (patients 18 and older)? 4. Can the intervention text be delivered to clinics on a scheduled basis at the 5 pre- specified dates listed above? 5. Through an electronic medical record or radiology information system data pull, can you verify that the text was inserted into a patientÕs record with an eligible imaging CPT code? Must meet one of the following two criteria: 6.1 Can age range-specific (Section 3, Option 1) text be displayed in the radiology report depending on patient age? 6.2 Can tabular information by age (Section 3, Option 2) be displayed in the radiology report? 3.#Intervention#Pilot#Testing#Text# One%of%the%following%options%would%be%inserted%specific%to%imaging%modality% indicated%by%CPT%c.% ode Option'1:'Age-specific'intervention'text' “Comment:%The%following%findings%are%so%common%in%people%without%low%back%pain% that%while%we%report%their%presence,%they%must%be%interpreted%with%caution%and%in% the%context%of%the%clinical%situation%(R–% efJearrevnikc% e% et%al,%Spine%% 2001) Findings:%(prevalence%in%patientsX%Xa-YgY e% % without%low%back%pain),%Disk% degeneration%(decreased%T2%signal,%height%lgoes)s% ,%(9 b1 u% l ),%Disk% –% T si2g% nal%loss% (83%),%Disk%height%loss%(56%),%Disk%bulge% sk(6 %4p% ro),t% ruDsiion%(32%)n,% nuAlar% fissure%(38%)% ” Option'2:'Age-tabulated'intervention'text' The following MRI findings are so common in people without low back pain that while we report their presence, they must be interpreted with caution and in the context of the clinical situation. Disk Disk T2 Disk Disk Disk Annular Degeneration Signal Height Bulge Protrusion Fissure Loss Loss 21-30 A % A % A % A % A % A % 1 2 3 4 5 6 31-40 B % B % B % B % B % B % 1 2 3 4 5 6 41-50 C % C % C % C % C % C % 1 2 3 4 5 6 51-60 D % D % D % D % D % D % 1 2 3 4 5 6 61-70 E % E % E % E % E % E % 1 2 3 4 5 6 71-80 F % F % F % F % F % F % 1 2 3 4 5 6 81-90 G % G % G % G % G % G % 1 2 3 4 5 6 >91 H % H % H % H % H % H % 1 2 3 4 5 6 LIRE%Projec% t % % % % Intervention%Pilot%Testing%P%lan% % April% ;% May%201% 3 4.#Group#Health#clinics#identified#for#the#LIRE#project.#Please#verify#that#the# clinic#meets#the#inclusions#criteria#listed#in#Table#1.%% Table#2.#Site#eligibility#evaluation.# Meets Meets Inclusion Inclusion Criteria Criteria Reasons # Group Health Clinic Name #PCPs 1 - 5 6.1 or 6.2 for Failure 1 Bellevue Medical Center 12 2 Burien Medical Center 15 3 Capitol Hill Campus 32 4 Downtown Seattle Medical Center 7 5 Everett Medical Center 20 6 Factoria Medical Center 12 7 Federal Way Medical Center 14 8 Kent Medical Center 7 Spokane-Lidgerwood Medical 9 Center 13 10 Northgate Medical Center 31 11 Northshore Medical Center 8 12 Olympia Medical Center 41 13 Port Orchard Medical Center 18 14 Poulsbo Medical Center 8 15 Puyallup Medical Center 11 16 Rainier Medical Center 8 17 Redmond Medical Center 10 18 Renton Medical Center 13 Spokane-Riverfront Medical 19 Center 17 20 Silverdale Medical Center 16 21 Spokane-South Hill Medical Center 5 22 Tacoma Medical Center 10 23 Tacoma South Medical Center 15 24 Spokane-Veradale Medical Center 8 25 Lynnwood Medical Center 15 Additional#comments#from#pilot#test#implementation# <comments here> % LIRE Working Group 3 Appendix of Relevant Articles Using RVU-based Assessment 1. Adelgais KM, Grossman DC, Langer SG, Mann FA. Use of helical computed tomography for imaging the pediatric cervical spine. Academic emergency medicine : official journal of the Society for Academic Emergency Medicine. Mar 2004;11(3):228-236. 2. Allen B, Jr., Donovan WD, McGinty G, et al. Professional component payment reductions for diagnostic imaging examinations when more than one service is rendered by the same provider in the same session: an analysis of relevant payment policy. Journal of the American College of Radiology : JACR. Sep 2011;8(9):610-616. 3. Appleton P, Chacko A, Rodriguez EK. Financial implications of nonoperative fracture care at an academic trauma center. Journal of orthopaedic trauma. Nov 2012;26(11): 617-619. 4. Archila AM, Jarecke AJ, Damiani TN, Boorady J. Tri-service optometry productivity findings and benchmark recommendation. Military medicine. Apr 2007;172(4):418-420. 5. Arenson RL, Lu Y, Elliott SC, Jovais C, Avrin DE. Measuring the academic radiologist's clinical productivity: applying RVU adjustment factors. Academic radiology. Jun 2001;8(6):533-540. 6. Barnes SL, Robinson BR, Richards JT, et al. The devil is in the details: maximizing revenue for daily trauma care. Surgery. Oct 2008;144(4):670-675; discussion 675-676. 7. Becker ER, Hall K. Physician services in an academic neurology department: using the resource-based relative-value scale to examine physician activities. Journal of health care finance. Summer 2001;27(4):79-91. 8. Becker ER, Mauldin PD, Culler SD, Kosinski AS, Weintraub WS, King SB. Applying the resource-based relative value scale to the Emory angioplasty versus surgery trial. The American journal of cardiology. Mar 15 2000;85(6):685-691. 9. Begaz T, Decker MC, Treat R, Tews M. No relationship between measures of clinical efficiency and teaching effectiveness for emergency medicine faculty. Emergency medicine journal : EMJ. Jan 2011;28(1):37-39. 10. Bender JR, Johnson AJ, Schenk TW. The economic value of automotive occupational health services: business metrics for performance management. Journal of occupational and environmental medicine / American College of Occupational and Environmental Medicine. Feb 2008;50(2):138-145. 11. Bergersen L, Gauvreau K, McElhinney D, et al. Capture of Complexity of Specialty Care in Pediatric Cardiology by Work RVU Measures. Pediatrics. Feb 2013;131(2):258-267. 12. Bernard AM, Hayward RA, Rosevear J, Chun H, McMahon LF. Comparing the hospitalizations of transfer and non-transfer patients in an academic medical center. Academic medicine : journal of the Association of American Medical Colleges. Mar 1996;71(3):262-266. 13. Bittner JGt, Kang R, Stern PJ. Management of flexor tendon sheath ganglions: a cost analysis. The Journal of hand surgery. Jul 2002;27(4):586-590. 14. Bree RL, Kazerooni EA, Katz SJ. Effect of mandatory radiology consultation on inpatient imaging use. A randomized controlled trial. JAMA : the journal of the American Medical Association. Nov 20 1996;276(19):1595-1598. 15. Brennan DF, Silvestri S, Sun JY, Papa L. Progression of emergency medicine resident productivity. Academic emergency medicine : official journal of the Society for Academic Emergency Medicine. Sep 2007;14(9):790-794. 16. Bronstein JM, Adams EK. Rural-urban differences in health risks, resource use and expenditures within three state medicaid programs: implications for medicaid managed care. The Journal of rural health : official journal of the American Rural Health Association and the National Rural Health Care Association. Winter 2002;18(1):38-48. 17. Buntin MJ, Escarce JJ, Goldman D, Kan H, Laugesen MJ, Shekelle P. Increased Medicare expenditures for physicians' services: what are the causes? Inquiry : a journal of medical care organization, provision and financing. Spring 2004;41(1):83-94. 18. Chaudhry R, Kottke TE, Naessens JM, et al. Busy physicians and preventive services for adults. Mayo Clinic proceedings. Mayo Clinic. Feb 2000;75(2):156-162. 19. Cheriff AD, Kapur AG, Qiu M, Cole CL. Physician productivity and the ambulatory EHR in a large academic multi-specialty physician group. International journal of medical informatics. Jul 2010;79(7):492-500. 20. Chin MH, Zhang JX, Merrell K. Specialty differences in the care of older patients with diabetes. Medical care. Feb 2000;38(2):131-140. 21. Chung KC, Walters MR, Greenfield ML, Chernew ME. Endoscopic versus open carpal tunnel release: a cost-effectiveness analysis. Plastic and reconstructive surgery. Sep 1998;102(4):1089-1099. 22. Conoley PM, Vernon SW, Burtram SG. Growth profiles for radiology in the United States, 1973-1989: commentary based upon analysis of relative value units. European journal of radiology. Oct 1992;15(3):264-269. 23. Conoley PM, Vernon SW, Burtram SG. Productivity of radiologists in the United States by imaging technique: a 16-year analysis based upon relative value units. European journal of radiology. Oct 1992;15(3):258-263. 24. Conrad D, Fishman P, Grembowski D, et al. Access intervention in an integrated, prepaid group practice: effects on primary care physician productivity. Health services research. Oct 2008;43(5 Pt 2):1888-1905. 25. Cook SP. Candidate's Thesis: Laryngotracheal separation in neurologically impaired children: long-term results. The Laryngoscope. Feb 2009;119(2):390-395. 26. Cullen EJ, Jr., Lawless ST, Hertzog JH, et al. A model of determining a fair market value for teaching residents: who profits? Pediatrics. Jul 2003;112(1 Pt 1):40-48. 27. Davenport DL, Ferraris VA, Hosokawa P, Henderson WG, Khuri SF, Mentzer RM, Jr. Multivariable predictors of postoperative cardiac adverse events after general and vascular surgery: results from the patient safety in surgery study. Journal of the American College of Surgeons. Jun 2007;204(6):1199-1210. 28. Davis KA, Cabbad NC, Schuster KM, et al. Trauma team oversight improves efficiency of care and augments clinical and economic outcomes. The Journal of trauma. Dec 2008;65(6):1236-1242; discussion 1242-1234. 29. Davis PL. Assessing the potential versus the actual earnings of academic radiologists: effects of unequal duty service assignments. Academic radiology. Aug 2001;8(8): 782-791. 30. Dobrez D, Sasso AL, Holl J, Shalowitz M, Leon S, Budetti P. Estimating the cost of developmental and behavioral screening of preschool children in general pediatric practice. Pediatrics. Oct 2001;108(4):913-922. 31. Dorr DA, Wilcox A, McConnell KJ, Burns L, Brunker CP. Productivity enhancement for primary care providers using multicondition care management. The American journal of managed care. Jan 2007;13(1):22-28. 32. Dudas RA, Monroe D, McColligan Borger M. Community pediatric hospitalists providing care in the emergency department: an analysis of physician productivity and financial performance. Pediatric emergency care. Nov 2011;27(11):1099-1103. 33. Duszak R, Jr., Harris AB. Percutaneous abscess drainage: use of related radiology services and associated economic impact on a radiology practice. Journal of vascular and interventional radiology : JVIR. May 2003;14(5):597-601. 34. Fahy BN, Bold RJ, Schneider PD, Khatri V, Goodnight JE, Jr. Cost-benefit analysis of biopsy methods for suspicious mammographic lesions; discussion 994-5. Archives of surgery (Chicago, Ill. : 1960). Sep 2001;136(9):990-994. 35. Fairchild DG, McLoughlin KS, Gharib S, et al. Productivity, quality, and patient satisfaction: comparison of part-time and full-time primary care physicians. Journal of general internal medicine. Oct 2001;16(10):663-667. 36. Fleming NS, Becker ER, Culler S, Cheng D, McCorkle R, Ballard DJ. Financial performance of primary care physician practices prior to electronic health record implementation. Proceedings (Baylor University. Medical Center). Apr 2009;22(2): 112-118. 37. Grace C, Alston WK, Ramundo M, Polish L, Kirkpatrick B, Huston C. The complexity, relative value, and financial worth of curbside consultations in an academic infectious diseases unit. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. Sep 15 2010;51(6):651-655. 38. Green PH, Neugut AI, Naiyer AJ, Edwards ZC, Gabinelle S, Chinburapa V. Economic benefits of increased diagnosis of celiac disease in a national managed care population in the United States. Journal of insurance medicine (New York, N.Y.). 2008;40(3-4): 218-228. 39. Griffin TC, Hutter JJ, Johnson KK, Moscow JA. A survey of clinical productivity and current procedural terminology (CPT) coding patterns of pediatric hematologist/ oncologists. Pediatric blood & cancer. Aug 2004;43(2):140-147. 40. Henke PK, Kubus J, Englesbe MJ, Harbaugh C, Campbell DA. A statewide consortium of surgical care: a longitudinal investigation of vascular operative procedures at 16 hospitals. Surgery. Oct 2010;148(4):883-889; discussion 889-892. 41. Henley MB, Mann FA, Holt S, Marotta J. Trends in case-mix-adjusted use of radiology resources at an urban level 1 trauma center. AJR. American journal of roentgenology. Apr 2001;176(4):851-854. 42. Hillman BJ, Olson GT, Griffith PE, et al. Physicians' utilization and charges for outpatient diagnostic imaging in a Medicare population. JAMA : the journal of the American Medical Association. Oct 21 1992;268(15):2050-2054. 43. Johnson RG, Arozullah AM, Neumayer L, Henderson WG, Hosokawa P, Khuri SF. Multivariable predictors of postoperative respiratory failure after general and vascular surgery: results from the patient safety in surgery study. Journal of the American College of Surgeons. Jun 2007;204(6):1188-1198. 44. Jordan SW, Mioton LM, Smetona J, et al. Resident Involvement Affects Plastic Surgery Outcomes: An Analysis of 10,356 Patients From the NSQIP Database. Plastic and reconstructive surgery. Dec 17 2012. 45. Katz SJ, McMahon LF, Manning WG. Comparing the use of diagnostic tests in Canadian and US hospitals. Medical care. Feb 1996;34(2):117-125. 46. Khandelwal CM, Meyers MO, Yeh JJ, et al. Relative value unit impact of complex skin closures to academic surgical melanoma practices. American journal of surgery. Sep 2012;204(3):327-331. 47. Khorasani R, Goel PK, Ma'luf NM, Fox LA, Seltzer SE, Bates DW. Trends in the use of radiology with inpatients: what has changed in a decade? AJR. American journal of roentgenology. Apr 1998;170(4):859-861. 48. Kravet SJ, Jones H, Howell EE, Wright SM. Pilot study comparing patients' valuation of health-care services with Medicare's relative value units. Health expectations : an international journal of public participation in health care and health policy. Dec 2008;11(4):391-399. 49. Laditka SB, Mastanduno MP, Laditka JN. Health care use of individuals with diabetes in an employer-based insurance population. Archives of internal medicine. May 28 2001;161(10):1301-1308. 50. Latimer EA, Verrilli D, Welch WP. Utilization of physician services at the end of life: differences between the United States and Canada. Inquiry : a journal of medical care organization, provision and financing. Spring 1999;36(1):90-100. 51. Levin DC, Rao VM, Parker L, Frangos AJ, Sunshine JH. Recent trends in utilization of cardiovascular imaging: how important are they for radiology? Journal of the American College of Radiology : JACR. Sep 2005;2(9):736-739. 52. Lind BK, Gerkovich MM, Cherkin DC, Deyo RA, Sherman KJ, Lafferty WE. Effect of Risk Adjustment Method on Comparisons of Health Care Utilization Between Complementary and Alternative Medicine Users and Nonusers. Journal of alternative and complementary medicine (New York, N.Y.). Oct 4 2012. 53. Lu Y, Zhao S, Chu PW, Arenson RL. An update survey of academic radiologists' clinical productivity. Journal of the American College of Radiology : JACR. Jul 2008;5(7): 817-826. 54. Lynch F. The use of an interactive computerized daily schedule in a busy interventional radiology practice increases efficiency. Journal of the American College of Radiology : JACR. Dec 2004;1(12):965-971. 55. Maitino AJ, Levin DC, Parker L, Rao VM, Sunshine JH. Practice patterns of radiologists and nonradiologists in utilization of noninvasive diagnostic imaging among the Medicare population 1993-1999. Radiology. Sep 2003;228(3):795-801. 56. Maxwell S, Zuckerman S, Berenson RA. Use of physicians' services under Medicare's resource-based payments. The New England journal of medicine. May 3 2007;356(18): 1853-1861. 57. Mortenson MM, Ho HS, Bold RJ. An analysis of cost and clinical outcome in palliation for advanced pancreatic cancer. American journal of surgery. Sep 2005;190(3):406-411. 58. Noll RE, Jr., Tonnessen BH, Kim J, Money SR, Sternbergh WC, 3rd. Long-term postplacement cost comparison of AneuRx and Zenith endografts. Annals of vascular surgery. Nov 2008;22(6):710-715. 59. Noll RE, Jr., Tonnessen BH, Mannava K, Money SR, Sternbergh WC, 3rd. Long-term postplacement cost after endovascular aneurysm repair. Journal of vascular surgery. Jul 2007;46(1):9-15; discussion 15. 60. Phibbs CS, Bhandari A, Yu W, Barnett PG. Estimating the costs of VA ambulatory care. Medical care research and review : MCRR. Sep 2003;60(3 Suppl):54S-73S. 61. Privette AR, Shackford SR, Osler T, Ratliff J, Sartorelli K, Hebert JC. Implementation of resident work hour restrictions is associated with a reduction in mortality and provider- related complications on the surgical service: a concurrent analysis of 14,610 patients. Annals of surgery. Aug 2009;250(2):316-321. 62. Raja AS, Mortele KJ, Hanson R, Sodickson AD, Zane R, Khorasani R. Abdominal imaging utilization in the emergency department: trends over two decades. International journal of emergency medicine. 2011;4:19. 63. Rubinsky AD, Sun H, Blough DK, et al. AUDIT-C alcohol screening results and postoperative inpatient health care use. Journal of the American College of Surgeons. Mar 2012;214(3):296-305 e291. 64. Shine D, Jessen L, Bajaj J, Pencak D, Panush R. Actual and potential effects of medical resident coverage on reimbursement for inpatient visits by attending physicians. Journal of general internal medicine. Jun 2002;17(6):428-434. 65. Showstack J, Lin F, Learman LA, et al. Randomized trial of medical treatment versus hysterectomy for abnormal uterine bleeding: resource use in the Medicine or Surgery (Ms) trial. American journal of obstetrics and gynecology. Feb 2006;194(2):332-338. 66. Siegel JE, Clancy CM. Relative value in healthcare: cost-effectiveness of interventions. Journal of nursing care quality. Apr-Jun 2006;21(2):99-103. 67. Solomon A, Martino S. Relative value units: practical productivity measurement. Radiology management. Winter 1991;13(1):33-37. 68. Soremekun OA, Noble VE, Liteplo AS, Brown DF, Zane RD. Financial impact of emergency department ultrasound. Academic emergency medicine : official journal of the Society for Academic Emergency Medicine. Jul 2009;16(7):674-680. 69. Talley JD, Mauldin PD, Becker ER, Stikovac M, Leesar MA. Cost and therapeutic modification of intracoronary ultrasound-assisted coronary angioplasty. The American journal of cardiology. Jun 15 1996;77(15):1278-1282. 70. Weintraub WS, Culler SD, Kosinski A, et al. Economics, health-related quality of life, and cost-effectiveness methods for the TACTICS (Treat Angina With Aggrastat [tirofiban]] and Determine Cost of Therapy with Invasive or Conservative Strategy)-TIMI 18 trial. The American journal of cardiology. Feb 1 1999;83(3):317-322. 71. Welch WP, Verrilli D, Katz SJ, Latimer E. A detailed comparison of physician services for the elderly in the United States and Canada. JAMA : the journal of the American Medical Association. May 8 1996;275(18):1410-1416. 72. Wiersema MJ, Mergener K. Current procedural terminology, Resource-based Relative Value Scale, and the Center for Medicare and Medicaid Services: overview. Gastrointestinal endoscopy clinics of North America. Oct 2006;16(4):775-787. 73. Williams TR. A geologic survey of the medicare RBRVS system. Journal of the American College of Radiology : JACR. Mar 2004;1(3):192-198. 74. Zuber TJ, Purvis JR. Coding and reimbursement of primary care debridement and excision procedures. The Journal of family practice. Dec 1992;35(6):663-672. 75. Zwolak RM. Resource-Based Relative Value Scale (RBRVS), coding, and Medicare reimbursement. Seminars in vascular surgery. Jun 1997;10(2):119-127. Lumbar Imaging Reporting with Epidemiology (LIRE) STATISTICAL ANALYSIS PLAN Research Aims Aim 1: To determine whether inserting a description of age-specific prevalence of imaging findings among asymptomatic subjects into lumbar spine imaging reports decreases back-related interventions (imaging, injections, surgeries, etc.) over the subsequent year. Aim 1a: To determine if inserting epidemiological evidence reduces Relative Value Units (RVUs) attributable to spine interventions (imaging, injections, specialist referrals, surgeries, etc.). Aim 1b: To determine if inserting epidemiological data decreases opioid prescriptions. Aim 1c: To determine if inserting epidemiological evidence decreases subsequent cross-sectional imaging (magnetic resonance (MR) and computed tomography (CT)). Aim 1d: To explore whether adding epidemiological evidence decreases overall costs of care for low back pain based on CMS reimbursement. Aim 2: To determine whether inserting age-specific prevalence of imaging findings in asymptomatic subjects has a differential effect on subsequent back-related interventions if inserted into lumbar spine MR and CT imaging reports compared with plain films. Aim 3: To determine if specific imaging findings influence subsequent interventions. Intervention: This trial will study two groups of patients within providers, each will have had a lumbar imaging CPT code. Patients and providers at intervention clinics will receive additional prevalence summary data of incidental findings as a part of their radiology report. Control patients and providers will receive the standard imaging report without the LIRE text. Design: Using a stepped wedge cluster randomized design , we will randomly assign all predetermined clinics at each site to receive the intervention at one of five fixed time-points. Interventions will roll out every six months at the start of the second quarter of UH3 Year 2 according to the schedule shown in Figure 1. During the UH2 project phase, Figure 1: Proposed Randomization Schedule we obtained a current and accurate enumeration of PCPs within clinics. Within each recruitment site, we sorted clinics by number of primary care providers into tertiles (e.g. small, medium, large clinics). From each tertile we will randomly select clinics using urn-based randomization (without replacement) stratified by site and clinic size such that clinics of small, medium, and large size are equally represented in each randomization wave. Table 1 Created on 11/11/19 10:31:00 AM Page 1 of 8 Lumbar Imaging Reporting with Epidemiology (LIRE) STATISTICAL ANALYSIS PLAN displays the site-specific strata definitions and size. In total, we will randomize 110 clinics with 1,824 PCPs as units of observation within those clinics. Note that we have chosen to use site-specific definitions for the size of the clinic with the goal of having balance of clinic size within each site. In addition, by balancing randomization on size we will be sure to have comparable time on Table 1. Within-site stratified randomization schedule of clinics by number of PCPs. control and intervention for Units of each clinic size strata. Randomization PCP strata size boundaries (# of PCPs) (# clinics) In the original project application, we assumed 128 Recruitment Site Small Medium Large clinics and 1,898 PCPs would participate in the LIRE Group Health Cooperative 25 (370) 5 to 10 (9) 11 to 15 (8) 16 to 41 (8) project. After input from the Collaboratory Biostatistics Henry Ford Health System 26 (230) 3 to 6 (9) 7 to 9 (9) 10 to 24 (8) Core, we excluded all clinics Kaiser Permanent N. CA 20 (865) 17 to 29 (7) 33 to 39 (5) 43 to 106 (8) with a single PCP (n=18) Mayo Clinic Health System 39 (359) 2 to 4 (15) 5 to 9 (12) 11 to 34 (12) from the primary study and statistical analysis and will Total 110 (1,824) only include clinics with 2 or more PCPs. Primary Outcome: We have devoted substantial effort towards developing and refining the primary outcome measure: a summary back-specific relative value unit (RVU). The back-specific RVU is a composite measure of spine intervention intensity that combines the overall intensity of resource utilization for back pain care into a single metric. To develop the composite RVU measure, we used data from our large cohort of patients with back pain who comprise the Back pain Outcomes using Longitudinal Data (BOLD) Project, Agency for Healthcare Research and Quality (AHRQ)-funded study. During our work with the BOLD Project we developed algorithms to abstract electronic medical record (EMR) data across three health systems (two of which overlap with LIRE): Kaiser Northern California, Henry Ford Health System and Harvard Vanguard/Harvard Pilgrim. For the 5,239 BOLD cohort participants, we obtained extensive EMR data on pharmacy records, healthcare utilization (CPT codes), diagnoses and provider visits (ICD-9 codes), and inpatient hospitalization data. Using the Medicare Physician Fee Schedule (http://www.cms.gov/) we generated and tested a mapping algorithm to assign more than 10,000 unique CPT codes to RVUs. A sample of RVUs from the 2012 CMS file is shown in Table 2. Using the BOLD cohort EMR Table 2. Example spine-related CPT codes and associated RVUs. data, we developed and CPT Code Description RVU tested an algorithm for aggregating individual 72100 X-ray exam of lower spine - 2 views 1.07 RVUs across procedures 97001 PT Evaluation 2.18 over a time interval for a 99214 Detailed office visit 2.26 given patient, as well as 99284 Emergency department visit - high moderate intensity 3.37 across primary care 64483 Epidural injection for lumbar spinal stenosis 3.37 providers or clinics. 72131 CT lumbar spine w/o dye 6.27 To obtain a spine-related 72148 MRI Lumbar Spine w/o Contrast 11.31 summary RVU from CPT 63047 Removal of spinal lamina 32.89 and ICD-9 codes, we used 22804 Fusion of the spine 71.60 an existing algorithm Created on 11/11/19 10:31:00 AM Page 2 of 8 Lumbar Imaging Reporting with Epidemiology (LIRE) STATISTICAL ANALYSIS PLAN 2-4 developed by a colleague at Dartmouth College. !Aggregating across CPT codes identified by this algorithm yields the back-specific RVU. We are currently preparing a manuscript describing this development work as well as a manuscript that directly influences and informs our LIRE UH3 proposal. Using BOLD cohort data, we identified a subset of patients who have had an early lumbar image (MRI/CT or plain film) following an office visit for back pain. Our BOLD cohort manuscript (in preparation) compares the one-year cumulative RVU of early-imaged patients to carefully matched BOLD cohort controls who did not have an early lumbar image. Preliminary results indicate a substantial downstream increase in healthcare utilization for patients who received an early image compared to propensity score matched controls. Patients who underwent a lumbar MRI or CT had a mean one-year RVU of 150 +/- 410, versus 120 +/- 450 for those who had an early plain film, versus 43 +/- 120 for carefully matched controls. Mapping the relative increases of utilization of nearly 80 and 110 RVUs for the plain film and advanced imaging modalities to the example codes shown in Table 2, we see that imaged patients undergo substantially more procedures. Our expectation for the UH3 project is that the insertion of normative prevalence data into lumbar imaging reports will reduce subsequent inappropriate healthcare utilization. Secondary Outcomes: In addition to back-specific RVU, important secondary outcomes will be obtained and derived using electronic medical record data pulls and include: an indicator of opioid prescriptions written within 30 and 90 days after the index image (Aim 1b); subsequent cross-sectional re-imaging within 90 days and 12 months (Aim 1c); and medical costs (Aim 1d). In the BOLD project, we developed mapping algorithms based upon the United States Food and Drug Administration National Drug Codes (NDC) that generate an indicator of whether or not an individual pharmacy record is an opioid analgesic. Similarly, we have enumerated and categorized a listing of CPT codes that indicate cross-sectional lumbar imaging (CT, or MRI). General Analytic Strategy: To evaluate the effectiveness of inserting epidemiologic evidence into an imaging report we will use longitudinal regression methods such as linear mixed effects models (LMMs) or generalized linear mixed models (GLMMs) for all primary and secondary outcome measures. Mixed models provide an efficient method for analysis of longitudinal or multilevel data and will be the basis of our primary analysis approach. However, correct model specification is required to ensure valid results when using LMMs or GLMMs and we will therefore use robust standard errors for our primary analysis. Therefore, we are effectively adopting a “working” correlation structure through the specification of flexible multilevel models (LMM or GLMM) but will rely on non-parametrically valid robust standard errors for inference where we cluster on the clinic. Secondary analysis will directly use generalized estimating equations (GEE) adopting simple exchangeable correlation models at the clinic level to determine whether conclusions appear sensitive to model specification. In each analysis we will also consider a ‘washout period’ in the three months prior to the intervention being activated at a clinic, as determined by the randomization schedule. The rationale for a washout period is to reduce or eliminate within-provider cross-contamination of patient outcomes and utilization in the transition period between control and intervention. Including a washout period reduces the risk of having a patient initially treated in the control time period return to their primary care provider for subsequent care after the primary care provider has been exposed to the intervention through other patients. This reduces the potential bias due to within-provider cross-contamination of outcomes on the estimated inte rvention effect. Primary Analysis: The primary longitudinal model for back pain specific RVUs will use a time-varying intervention status indicator Status (0 = control, 1 = intervention, for clinic k at time t). Use of the time- kt dependent intervention status indicator permits both within-clinic contrasts that inform intervention effects (post- versus pre-intervention) as well as contrasts across clinics with different intervention statuses within each time period. The specific regression model will adopt a functional form given below, with fixed effects for time (linear), age (18-39, 40-59, 60+, using two dummy variables), imaging modality type (plain film, CT, MRI using two dummy variables), and clinic size (small, medium, large, using two dummy variables), and site (Group Health Cooperative, Henry Ford, Kaiser Permanente, Mayo Clinic, using three dummy variables) in addition to random Created on 11/11/19 10:31:00 AM Page 3 of 8 Lumbar Imaging Reporting with Epidemiology (LIRE) STATISTICAL ANALYSIS PLAN effects for provider, clinic, and intervention status: T T T T Y = b + b × Time + b × Age + b × Modality + b × Size + b × Site + l × Status + mean model ijk 0 1 t 2 ijk 3 ijk 4 k 5 k 0 kt b + b × Status + clinic random effects k,0 k,1 kt a + e provider random effects and errors jk,0 ijk We will collect the outcome measure Y on patient i (i= 1,2,…,n ) under primary care provider j (j =1,2,…,n ) ijk j k enrolled in time period t (t = 0,1,2,… ,5) in order to evaluate the overall effect of the intervention at the level of the clinic k (k = 1,2,…,110). Note that we will collect a single outcome measure for each subject recording the total utilization (RVU) over the 12 months after the index imaging event. Given that the random effects structure may contain additional elements (see below) we will use a robust standard error to test the null hypothesis that l = 0. For example, using SAS PROC MIXED we can use the “empirical” option in Figure 2: Hypothetical example of study data showing order to obtain robust standard errors. random clinic intercepts and intervention effects. Each line Alternatively, use of the jackknife (at the clinic shows the expected profile for a specific clinic. Here clinic 1 level) provides a robust standard error initiates intervention at quarter=2 while clinic 2 initiates at estimate (if using R and lmer) that is quarter=4. simple to compute. Key Model Parameters: The primary parameter of interest is l , which represents the average effect of the ● intervention adjusting for temporal trends ● − (Time ), clinic characteristics (Site , Size ), t k k and individual covariates (Age , ijk ● clinic 2 effect Modality ). In order to interpret the ● ijk random effects structure we focus on clinic 1 effect ●● clinic level means removing covariate effects where we have: adjusted mean at clinic k for times prior to intervention = b 0 ● + b ; and the adjusted mean at clinic k k,0 for times after start of the intervention = b 0 1 2 3 4 5 + b + l + b . For clinic-specific means k,0 0 k,1 study.quarter we average over both providers (ajk,0) and patients (e ). Using this representation ijk we interpret b as the pre-intervention adjusted overall mean outcome averaging across all clinics, and b is the 0 k,0 difference between that adjusted overall mean and the pre-intervention (baseline) mean for clinic k. The variance, var(b ), is a measure of the variation in the baseline mean outcome across clinics. The change in the k,0 adjusted mean outcome for clinic k is given by: (post-intervention adjusted mean) – (pre-intervention adjusted mean) = (b + b + l + b ) – (b + b ) = l + b . Here l represents the average intervention effect across all 0 k,0 0 k,1 0 k,0 0 k,1 0 clinics and b represents the difference between that average intervention effect and the intervention effect for k,1 clinic k. The variance, var(b ), is a measure of the variation in the change associated with intervention across k,1 clinics, or a measure of the heterogeneity of the intervention effect. Our primary regression model acknowledges the fundamental multilevel structure of individual-level data collected in health care systems with patients nested within providers, and providers nested within clinics. Although, the basic intervention contrast is the pre-post change associated with the initiation of intervention for each clinic, we do not propose using clinic-level summary measures for inference since the weighting of both patients and providers is not simple when heterogeneity of cluster sizes exists (e.g. PCPs per clinic, and patients per PCP). A proper multilevel model allows for optimal weighting based on the estimated variance components Created on 11/11/19 10:31:00 AM Page 4 of 8 Primary Outcome 0.6 0.7 0.8 0.9 1.0 1.1 1.2 Lumbar Imaging Reporting with Epidemiology (LIRE) STATISTICAL ANALYSIS PLAN (e.g. Gauss-Markov) and yields both an efficient summary of the overall intervention effect, as well an estimate of the variability in the magnitude of effect across clinics. However, we will not rely on the covariance model being correct for statistical inference and will use a robust (empirical) standard error. With greater than 100 total clusters (clinics) we expect valid inference and proper test size and do not anticipate needing to perform any correction such as the jackknife (recommended when the number of clusters is small). In our analysis we effectively assume that individual patients are nested within a single provider. However, in practice a patient may change providers during the follow-up year over which the primary outcome is captured. However, our basic mixed model covariance structure will simply use the assigned primary provider at the index imaging time. Therefore, we do not rely on model-based standard errors since the covariance structure may not match the true within-clinic covariance structure. We will use robust standard errors clustering at the clinic level, and therefore our analysis is valid even if there are changes in patient provider leading to an incorrectly specified covariance structure. Robust standard errors remain valid when a covariance model is not correctly specified. Furthermore, key secondary analysis of the primary outcome will directly use GEE and only cluster at the clinic level and provider level linkages are not used (nor needed) for simple GEE analysis. Secondary Analyses of Primary Outcome: We will conduct additional secondary analyses that evaluate the sensitivity of the multilevel model to the assumed basic random effects structure. We have included in the primary model multilevel random intercepts and a random effect for the clinic-level intervention. However, we will expand the random effects structure to also permit random slopes on time for both clinics and providers. Given the relatively short duration of follow-up with only six (6) total measurement times we do not expect strong heterogeneity across providers or clinics in cluster-specific temporal trends. Figure 2 shows an example of hypothetical data series for two clinics (assuming aggregation of providers to a clinic summary) and illustrates both the staggering of the crossover time and the potential to observe clinic-specific intervention effects. This figure also illustrates the fact that separating random effects of time (linear) from random effects of intervention would be difficult since time and intervention status are correlated give the unidirectional crossover from control to intervention. In addition, we will use GEE as a covariance model robust inference method and therefore can produce valid point estimates and confidence intervals without relying on correct covariance specification. Details of model choice and comparison of alternative models for longitudinal cluster level crossover trials is presented in French and Heagerty (2008) and comparison of alternative approaches is recommended. Models for time and intervention effect: Our primary analysis adopts a linear adjustment for calendar time in order to remove any large-scale temporal trends that may bias estimates of intervention effects. However, our basic regression model assumes a common (adjusted) mean for all times after the initiation of intervention. In practice there may be a delay in the impact of intervention so alternative models will be considered that incorporate a delayed and/or gradual effect of intervention. For example, the basic coding of the time-dependent variable Status takes the value 0 pre-intervention and the value 1 post-intervention. Delay in the impact of kt intervention can be accommodated using alternatives such as: 0 pre-intervention; 0.5 for quarter 1 after intervention; and 1 for all other post-intervention quarters. Such a modified model would allow full impact of the intervention to require two quarters of exposure. We will conduct secondary analyses to explore alternative models for the accumulation or delay of the intervention effect. Secondary Outcome Analysis: We will also analyze the impact of intervention on the rate of opioid prescription using Generalized Linear Mixed Models (GLMMs). For Aim 1b let Y =1 if opioids were prescribed within a given ijk timeframe (e.g. 30 days or 90 days) to patient i (i= 1,2,…,n ) seen by primary care provider j (j =1,2,…,n ) within j k clinic k (k = 1,2,…,110). Analysis for this outcome will use a logistic mixed model given as: T T T T logit(p ) = b + b × Time + b × Age + b × Modality + b × Size + b × Site + l × Status + mean model ijk 0 1 t 2 ijk 3 ijk 4 k 5 k 0 kt b + b × Status + clinic random effects k,0 k,1 kt a provider random effects jk,0 Created on 11/11/19 10:31:00 AM Page 5 of 8 Lumbar Imaging Reporting with Epidemiology (LIRE) STATISTICAL ANALYSIS PLAN where p denotes the probability that Y =1. Our secondary outcome analysis parallels the primary and will be ijk ijk based on a natural multilevel mixed model, with additional robust secondary analysis provided by GEE. For Aim 1c we will use Y =1 if CT or MR imaging occurs within a specified timeframe (e.g. 90 days or 12 months) after ijk the index imaging event. Medical costs (Aim 1d): Spine-related costs of care will be estimated using two approaches. First, we will use the spine-related RVU calculated in Aim 1a and estimate clinical-level, spine-intervention expenditures using the annual Medicare-determined payment amount per RVU (e.g., CY2013 = $34.023 per RVU). (reference: http://www.cms.gov/Outreach-and-Education/Medicare-Learning-Network- LN/MLNProducts/downloads/medcrephysfeeschedfctsht.pdf ) Second, as a proxy for costs of spine care, we will use a standard set of reimbursement amounts, i.e., CMS- based payments, and estimate clinic-level spine-related aggregate expenditures by applying CPT-based payment amounts to specific spine-intervention events (e.g., imaging, office visits, procedures, other). We will present monthly and annual means, medians, and ranges of clinic-level cost estimates, prior to and subsequent to implementing the epidemiological intervention. We will assess the level of right-skewness in the expenditure estimates and use t-tests to compare arithmetic means for clinic-level expenditures. In the case of considerable skewness, we will test for differences in logarithmically transformed mean clinic-level expenditures (before and after implementing intervention). We will also describe categories of prescriptions ordered, when available in the electronic medical records for a health system, and estimate costs for prescribed spine-related medications Analysis for Aim 2: The hypothesis of Aim 2 is that there will be a differential effect of the intervention according to the imaging modality used. In order to test this hypothesis we will analyze patient-level data according to the appropriate LMM or GLMM given above, but including the interactions between Modality indicators (modeled ijk using two indicator variables coding CT and MR, with plain film as the reference) and Status . A test of the kt interaction terms (2 degree of freedom Wald test) will be used to test the null hypothesis that the effect of the intervention does not vary according the imaging modality. Analysis for Aim 3: The hypothesis of Aim 3 is that there will be a differential effect of the intervention according to the results that are found in the imaging report. We will use an additional variable, ImageFinding , that takes ijk the value 1 if a significant image finding is present, and 0 otherwise (see detail regarding variable specification in protocol). We will test the null hypothesis that the interaction between ImageFinding and Status is zero using a ijk kt Wald test. Power Calculations: Our UH2 efforts with respect to sample size and statistical power focused on two key items. First, an important aim of our UH2 Working Group 2 was to obtain an accurate clinic and provider count for each health system. We will now randomize n=110 clinics (1,824 PCPs), which is slightly lower than the n=128 clinics (1,898 PCPs) assumed in our initial project application. However, the majority of the clinics that were dropped were those with only one PCP and therefore would not have contributed much information to the analysis. Second, in our Working Group 3 we sought to develop and characterize a composite RVU summary to be used as the primary outcome measure in this study. In our UH2 project application we discussed statistical power in the context of an important secondary outcome measure, a reduction in subsequent opioid prescription rates. We now present statistical power for the primary outcome measure using data from the BOLD Registry to inform key design parameter estimates. To our knowledge, off-the-shelf calculators do not exist that would adequately characterize statistical power for a stepped wedge cluster randomized trial with a varying number of sampling units between clusters. We therefore utilized simulation methods to generate and analyze data that closely mimics the design characteristics we anticipate for this study. With a simulation approach, we were able to include estimates of both patient and clinic- level variability and implement the proposed primary analysis methods: random intercept linear mixed effects models for RVU outcomes; and generalized linear mixed models for opioid prescription rates. All simulations Created on 11/11/19 10:31:00 AM Page 6 of 8 Lumbar Imaging Reporting with Epidemiology (LIRE) STATISTICAL ANALYSIS PLAN were conducted using R (version 3.0.1) with the lmer and glmmPQL functions implementing mixed model estimation. Power for Primary Outcome: In the BOLD Registry we identified 639 patients in the Kaiser Permanente and Henry Ford health systems that had a qualifying lumbar image within 6 weeks of a PCP visit, the majority (74%) of which occurred within 7 days. As one would expect with a measure of health care utilization intensity, patient- level RVUs are positively skewed. In our simulations and in the future analysis of study data, we therefore utilize an approximately normalizing transformation of log(RVU + 1) but will Figure 2: Statistical power for make interpretations regarding effect size back on the original RVU scale. spine-related RVU With log-transformed BOLD Registry RVU data, we fit a linear mixed effects model adjusting for image type (advanced vs. plain film) and study recruitment site and estimated the variance components for clinic (0.026) and the residual error term (1.230). The observed intra-class correlation coefficient (ICC) across clinics was 0.013 (95% CI: 0.000 to 0.046). In this subset of BOLD data, the number of PCPs with multiple patients was too few to inform the PCP-level variance component and it is therefore conservatively included in the error term variance for power simulations. The numbers of clinics and primary care providers were considered fixed 0% 2% 4% 6% 8% 10% for each simulation and we assumed that each provider would provide Percentage Reduction in Median PCP spine RVU data for all study time periods. For a range of potential RVU effect sizes, we generated 1,000 simulated data sets and performed mixed model estimation with each data set. In Figure 2, we show statistical power for the primary outcome measure of PCP spine-related RVU under the proposed study design. The study has greater than 90% power to detect reductions in the median spine-related RVU of 5.0% or larger. For a patient receiving a lumbar CT, a 5% reduction in spine RVU translates into one fewer additional lumbar CT scan on average compared to a patient unexposed to the LIRE intervention. Power for Secondary Outcome: Using the updated clinic and provider listing, we repeated the UH2 power analyses for a reduction in subsequent opioid prescriptions. We again used an average baseline opioid prescription rate of 22%, suggested from the pilot manuscript to anchor the effect size of percent reduction in the baseline rate. A clinic-specific baseline opioid prescription rate was drawn from a Beta(a=6, b=20) distribution; we used this rate to draw a baseline opioid prescription rate random effect for each primary care provider using a clinic-specific log-normal distribution. We generated 1,000 simulated data sets using the effect size of a 7.5% reduction in the opioid reported in our UH2 application and evaluated each using a GLMM assuming random intercepts. With 110 clinics randomized, the study remains well powered (88.9% power) to detect a reduction in the rate of opioid prescriptions of 7.5% or larger (e.g. 22% down to 20.4%). Statistical Analysis Plan References 1. Hussey MA, Hughes JP. Design and analysis of stepped wedge cluster randomized trials. Contemp Clin Trials 2007;28:182-91. 2. Martin B, Mirza SK, Lurie JD, et al. Validation of an administrative coding algorithm to identify back-related degenerative diagnoses. International Society for the Study of the Lumbar Spine (ISSLS). Scottsdale, AZ, 3. Martin BI, Gerkovich MM, Deyo RA, et al. The association of complementary and alternative medicine use and health care expenditures for back and neck problems. Med Care 2012;50:1029-36. 4. Martin BI, Mirza SK, Franklin GM, et al. Hospital and surgeon variation in complications and repeat surgery following incident lumbar fusion for common degenerative diagnoses. Health Serv Res 2013;48:1-25. 5. http://www.fda.gov/drugs/informationondrugs/ucm142438.htm, accessed September 13, 2013. Created on 11/11/19 10:31:00 AM Page 7 of 8 Statistical Power 0% 20% 40% 60% 80% 100% Lumbar Imaging Reporting with Epidemiology (LIRE) STATISTICAL ANALYSIS PLAN 6. Diggle PJ, Heagerty PJ, Liang KY, Zeger SL. Analysis of Longitudinal Data. Second Edition ed: Oxford University Press; 2002. 7. Efron B, Stein C. The jackknife estimate of variance. Annals of Statistics. 1981;9(3):586-96. 8. French B, Heagerty PJ. Analysis of longitudinal data to evaluate a policy change. Statistics in Medicine 2008; 27(24):5005-25. 9. McCullough BJ, Johnson GR, Martin BI, Jarvik JG. Lumbar MR imaging and reporting epidemiology: do epidemiologic data in reports affect clinical management? Radiology. 2012;262(3):941-6. PMCID: 3285226. Created on 11/11/19 10:31:00 AM Page 8 of 8

Journal

JAMA Network Open – American Medical Association

Published: Sep 4, 2020

There are no references for this article.