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

Learn More →

Incidence of Treatment for Opioid Use Disorder Following Nonfatal Overdose in Commercially Insured Patients

Incidence of Treatment for Opioid Use Disorder Following Nonfatal Overdose in Commercially... Key Points Question How often do commercially IMPORTANCE Timely initiation and referral to treatment for patients with opioid use disorder seen insured patients obtain follow-up in the emergency department is associated with reduced mortality. It is not known how often treatment for opioid use disorder after a commercially insured adults obtain follow-up treatment after nonfatal opioid overdose. nonfatal opioid overdose? Findings In this cohort study of national OBJECTIVE To investigate the incidence of follow-up treatment following emergency department commercial insurance claims for 6451 discharge after nonfatal opioid overdose and patient characteristics associated with receipt of patients, 16.6% of patients obtained follow-up treatment. follow-up treatment after a nonfatal opioid overdose. Among those who had DESIGN, SETTING, AND PARTICIPANTS A retrospective cohort study was conducted using an not received treatment for opioid use administrative claims database for a large US commercial insurer, from October 1, 2011, to September disorder before the overdose, patients 30, 2016. Data analysis was performed from May 1, 2019, to September 26, 2019. Adult patients of older age, female sex, black race, and discharged from the emergency department after an index opioid overdose (no overdose in the Hispanic ethnicity were less likely to preceding 90 days) were included. Patients with cancer and without continuous insurance obtain follow-up. enrollment were excluded. Meaning Timely treatment for opioid MAIN OUTCOMES AND MEASURES The primary outcome was follow-up treatment in the 90 days use disorder following overdose appears following overdose, defined as a combined outcome of claims for treatment encounters or to be low among commercially insured medications for opioid use disorder (buprenorphine and naltrexone). Analysis was stratified by patients, with race/ethnicity, sex, and whether patients received treatment for opioid use disorder in the 90 days before the overdose. age disparities. Logistic regression models were used to identify patient characteristics associated with receipt of follow-up treatment. Marginal effects were used to report the average adjusted probability and Invited Commentary absolute risk differences (ARDs) in follow-up for different patient characteristics. Supplemental content RESULTS A total of 6451 patients were identified with nonfatal opioid overdose; the mean (SD) age Author affiliations and article information are was 45.0 (19.3) years, 3267 were women (50.6%), and 4676 patients (72.5%) reported their race as listed at the end of this article. non-Hispanic white. A total of 1069 patients (16.6%; 95% CI, 15.7%-17.5%) obtained follow-up treatment within 90 days after the overdose. In adjusted analysis of patients who did not receive treatment before the overdose, black patients were half as likely to obtain follow-up compared with non-Hispanic white patients (ARD, −5.9%; 95% CI, −8.6% to −3.6%). Women (ARD, −1.7%; 95% CI, −3.3% to −0.5%) and Hispanic patients (ARD, −3.5%; 95% CI, −6.1% to −0.9%) were also less likely to obtain follow-up. For each additional year of age, patients were 0.2% less likely to obtain follow-up (95% CI, −0.3% to −0.1%). CONCLUSIONS AND RELEVANCE Efforts to improve the low rate of timely follow-up treatment following opioid overdose may seek to address sex, race/ethnicity, and age disparities. JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 Open Access. This is an open access article distributed under the terms of the CC-BY License. JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 1/13 JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose Introduction Each year, the emergency department (ED) provides care for an increasing number of patients who 1-3 present with opioid overdose as well as medical complications of opioid use disorder (OUD). The 4-7 ED serves as an essential touchpoint for patients seeking care for withdrawal and addiction. Akey strategy in secondary prevention of opioid overdose deaths is the engagement of patients with OUD 8-11 in treatment following discharge. 12-14 However, few patients successfully transition to treatment following nonfatal overdose. In evidence from 2 states, less than 5% of Medicaid patients initiated treatment with medication for 13,14 opioid use disorder (MOUD) following overdose. For patients who are ready to engage in 4,9 treatment, care coordination can help to overcome barriers to access. Yet hospitals have few 5,8,15-18 incentives and capacity to provide resource-intensive care navigation after ED visits. 19,20 Patients have high risk of death in the days immediately following opioid overdose. The initiation of MOUD during or after emergency care is associated with improvements in a variety of patient outcomes, including all-cause mortality and engagement in outpatient treatment, and other 12,21-24 hospital-based interventions have been developed. As a consequence, policy makers have identified the transition of patients from emergency care to sustained treatment (termed warm 25-28 handoffs) as an urgent priority. In this study, we sought to examine the rate of follow-up treatment after discharge from the ED following overdose in a national population of commercially insured adults. Previous studies have 12-14,29 focused on single states, the Medicaid population, and MOUD treatment. To our knowledge, no previous studies have included the full scope of treatment services available to patients. We also sought to examine patient-level characteristics associated with timely receipt of follow-up care. Evidence suggests that significant treatment disparities on the basis of race, sex, and geography have emerged as the opioid epidemic has evolved, possibly owing to differences in health 30-37 insurance coverage. We hypothesized that these treatment disparities by race and sex would persist within a commercially insured population. Methods Data Sources, Study Population, and Outcomes We conducted a retrospective cohort study of adult patients who were discharged from the ED following treatment for opioid overdose between October 1, 2011, and September 30, 2016. We used 38,39 an administrative claims database, the Optum Clinformatics Data Mart (Optum). The Optum database includes all inpatient, ED, outpatient, and pharmacy claims from a large national health insurance company that enrolled between 15 million and 18 million unique patients each year during the study period. Data analysis was performed from May 1, 2019, to September 26, 2019. The institutional review board at the University of Pennsylvania determined that this study was exempt from review because data are deidentified. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cohort studies. Selection of Patient Cohort We identified ED encounters for opioid overdose in the study period for patients with commercial insurance coverage (eFigure 1 in the Supplement). To do so, we used previously validated International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) and International Statistical Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) 41-44 diagnosis codes before and after October 1, 2015, respectively (eTable 1 in the Supplement). We used Current Procedural Terminology codes to specifically identify ED encounters (eTable 1 in the Supplement). We excluded encounters for patients who did not have continuous insurance enrollment for 90 days before and after the date of the overdose, to provide a sufficient window to measure patient JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 2/13 JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose exposures and outcomes and exclude fatal overdoses. We excluded patients with age younger than 18 years. We then limited the cohort to encounters for an index opioid overdose, defined as an encounter for opioid overdose with no ED encounter or hospital admission for opioid overdose in the preceding 90 days. We excluded encounters resulting in inpatient hospital admission to obtain a cohort of patients stable for ED discharge and likely to not have disability or sequelae from the overdose. In addition, we excluded encounters for patients with diagnosis of cancer based on treatment claims 12,46,47 ICD-9-CM and ICD-10 diagnosis codes in the preceding 90 days (eTable 1 in the Supplement). Patients with pain related to active cancer diagnoses represent a separate population and may be prescribed high doses of prescription opioids. Of the remaining encounters, we included only the first index opioid overdose for any individual patient during the study period (eFigure 1 in the Supplement). Outcomes The primary outcome was whether the patient obtained follow-up treatment in the 90 days following the index opioid overdose. We defined follow-up treatment as the presence of either 1 pharmacy claim for MOUD or 1 medical claim for an outpatient or inpatient opioid treatment encounter. For pharmacy claims, we identified National Drug Codes for all formulations of 48-50 buprenorphine, buprenorphine with naloxone, or naltrexone (eTable 2 in the Supplement). Methadone maintenance therapy was not covered by insurance for this population during the study period and was not included in this study. Medical claims for treatment encounters had an ICD-9-CM or ICD-10-CM diagnosis code for opioid use disorder in any position (eTable 3 in the Supplement) and Current Procedural Terminology or Healthcare Common Procedure Coding System codes for a variety of services including outpatient clinic visits, psychiatric services, inpatient and outpatient behavioral health services, outpatient treatment programs, and case management (eTable 3 in the Supplement). Repeated ED or inpatient hospital encounters were not included as follow-up treatment. Supplemental analyses were performed for the purpose of hypothesis generation. These included secondary outcomes that were the receipt of MOUD independently from treatment encounters within 90 days of the index overdose. We also examined the number of days from the index overdose to follow-up treatment. To address the absence of mortality data, we determined the date of service for the last insurance claim for all patients in the cohort. We performed a sensitivity analysis excluding patients for whom there was no claim beyond the 90-day follow-up period. Although the absence of claims does not indicate death, we could not ensure survival to the end of the follow-up period for those patients. Covariates We examined patient-level characteristics as covariates that we hypothesized could be associated with access to follow-up treatment, including patient age, sex, and race/ethnicity. Optum uses data on race/ethnicity that is self-reported or derived from administrative data sources. We also included geographic location, according to 4 United States Census Regions (Northeast, South, Midwest, West). Year of the index overdose was included given the increasing overdose incidence over the study period. We examined the type of overdose (heroin or prescription opioid) based on diagnosis codes. Prescription opioid refers to medications available by prescription but does not mean that the patient received a prescription for the medication. We also included exposures to treatment for behavioral health conditions in the 90 days preceding the index overdose. We included the presence of claims for anxiety or depression based on ICD-9-CM or ICD-10-CM diagnosis codes (eTable 1 in the Supplement) due to potential association with overdose. We also included claims for prescription opioid medications and benzodiazepines in the 90 days preceding the index overdose using American Hospital Formulary Service Pharmacologic-Therapeutic Classification codes. In addition, we determined whether patients had JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 3/13 JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose pharmacy claims for MOUD or medical claims for treatment encounters in the 90 days preceding the index overdose. Statistical Analysis First, we described the patient cohort, stratified by overdose type. We used 2-sided χ tests and t tests to describe differences in the cohort between overdose type. Next, we summarized patient outcomes, stratified by overdose type and treatment for OUD in the 90 days preceding the overdose. We then used multivariable logistic regression models to examine the association between patient characteristics, as described in the first paragraph of the Covariates section, and the binary primary outcome. Given that patients were hypothesized to more likely access follow-up treatment if they had received recent treatment before the overdose, we stratified the analyses based on whether patients had received OUD treatment in the 90 days before the overdose. For ease of interpretation, we used predictive margins to report average adjusted probability and absolute risk 54,55 differences (ARDs), with 95% CIs. For categorical variables, ARD represents the difference in adjusted probability of follow-up treatment between patients with a given characteristic and the reference value. In addition to the primary analysis, we investigated potential interactions between race/ ethnicity and overdose type by including an interaction term in the logistic regression model. Also, we used multivariable logistic regression models to examine the association between patient characteristics and the secondary outcome of MOUD treatment alone. In addition, we used Kaplan- Meier failure analysis to examine days to receipt of follow-up treatment, stratified by overdose type. Data analysis was conducted from June 1, 2019, to September 1, 2019. Analyses were performed using Stata software, version 15.1 (StataCorp LP). Results The total cohort consisted of 6451 patients, of whom 1896 (29.4%) overdosed from heroin and 4555 (70.6%) overdosed from prescription opioids (Table 1). Further delineation of the type of opioid overdose is reported in eTable 7 in the Supplement. The mean (SD) age was 45.0 (19.3) years and there were 3267 (50.6%) women. A total of 4676 patients (72.5%) reported their race as non-Hispanic white, 601 patients (9.3%) reported their race as black, and 536 patients (8.3%) who reported Hispanic ethnicity. Only 682 patients (10.6%) received treatment for opioid use disorder in the 90 days preceding the overdose, including 320 (5.0%) with pharmacy claims for MOUD. Patients with heroin overdose significantly differed across all patient characteristics compared with those with prescription opioid overdose. Primary Analysis For all patients in the study cohort, 1069 individuals (16.6%; 95% CI, 15.7%-17.5%) obtained follow-up treatment in the 90 days following overdose (Figure 1;eTable8inthe Supplement). Among the 5769 patients who did not receive treatment for OUD in the 90 days before the overdose, 643 (11.1%; 95% CI, 10.3%-12.0%) obtained follow-up treatment. Among the 682 patients who received treatment before the overdose, 426 individuals (62.5%; 95% CI, 58.7%-66.1%) patients obtained follow-up. In the adjusted analysis for patients who did not receive treatment before the overdose, patients with prescription opioid overdose were less likely to obtain follow-up compared with heroin overdose (Table 2) (ARD, −8.8%; 95% CI, −11.2% to −6.5%). Compared with patients of non-Hispanic white race, black (ARD, −5.9%; 95% CI, −8.6% to −3.6%) and Hispanic (ARD, −3.5%; 95% CI, −6.1% to −0.9%) patients were less likely to obtain follow-up. Women were less likely to obtain follow-up than men (ARD, −1.7%; 95% CI, −3.3% to −0.5%). For each additional year of age, patients were 0.2% less likely to obtain follow-up (95% CI, −0.3% to −0.1%). However, patients with recent treatment JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 4/13 JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose for anxiety, including a treatment encounter for anxiety (ARD, 3.4%, 95% CI, 1.1%-5.8%) or prescription for a benzodiazepine (ARD, 2.8%; 95% CI, 0.7%-5.0%), were more likely to obtain follow-up. In this adjusted analysis, there was no statistically significant change with regard to the rate of patients obtaining follow-up treatment over the 5 years of the study (Figure 2). These associations were not present for patients who received treatment in the 90 days before overdose, apart from a decreased rate of follow-up for patients in the South (ARD, −15.0%; 95% CI, −25.9% to −4.1% and the West (ARD, −20.1%; 95% CI, −32.% to −7.6%), compared with the Northeast (eTable 4 in the Supplement). Supplemental Analyses In supplemental analyses, differences in the adjusted probability of follow-up rate persisted across overdose type for black patients compared with non-Hispanic white patients (Figure 3). Among patients who did not receive treatment before overdose, black patients were less likely to obtain follow-up treatment than non-Hispanic white patients whether the index overdose was due to heroin (ARD, −8.8%; 95% CI, −11.5% to −6.1%) or prescription opioids (ARD, −4.7%; 95% CI, −5.7% to −3.7%). For Hispanic patients compared with patients of non-Hispanic white race, the difference in Table 1. Characteristics of Patient Cohort, Stratified by Overdose Type No. (%) Overdose Characteristic All patients (n = 6451) Heroin (n = 1896) Prescription opioid (n = 4555) Age, mean (SD), y 45.0 (19.3) 31.0 (13.2) 50.8 (18.4) Sex Male 3184 (49.4) 1291 (68.1) 1893 (41.6) Female 3267 (50.6) 605 (31.9) 2662 (58.4) Race/ethnicity Non-Hispanic white 4676 (72.5) 1450 (76.5) 3226 (70.8) Black 601 (9.3) 148 (7.8) 453 (9.9) Hispanic 536 (8.3) 135 (7.1) 401 (8.8) Asian 78 (1.2) 9 (0.5) 69 (1.5) Unknown 560 (8.7) 154 (8.1) 406 (8.9) Year 2011, quarter 4 229 (3.5) 40 (2.1) 189 (4.1) 2012 1099 (17.1) 239 (12.6) 860 (18.9) 2013 1164 (18.1) 276 (14.6) 888 (19.5) 2014 1248 (19.3) 362 (19.1) 886 (19.5) 2015 1387 (21.5) 475 (25.1) 912 (20.0) 2016, quarters 1-3 1324 (20.5) 504 (26.6) 820 (18.0) Region Northeast 659 (10.2) 316 (16.7) 343 (7.5) South 2627 (40.7) 617 (32.5) 2010 (44.1) Midwest 1619 (25.1) 703 (37.1) 916 (20.1) West 1546 (24.0) 260 (13.7) 1286 (28.2) 90 d Before overdose Anxiety treatment 1625 (25.2) 403 (21.3) 1222 (26.8) Depression treatment 1416 (22.0) 322 (17.0) 1094 (24.0) Prescription opioid claim 3266 (50.6) 317 (16.7) 2949 (64.7) Benzodiazepine claim 2009 (31.1) 373 (19.7) 1636 (35.9) MOUD claim 320 (5.0) 201 (10.6) 119 (2.6) Abbreviations: MOUD, medication for opioid use Buprenorphine 278 (4.3) 168 (8.9) 110 (2.4) disorder; OUD, opioid use disorder. Naltrexone 42 (0.7) 33 (1.7) 9 (0.2) a 2 Two-sided t test and χ tests were performed; Treatment encounter for OUD 539 (8.4) 347 (18.3) 192 (4.2) P < .001 for all patient characteristics. JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 5/13 JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose adjusted follow-up rate was significant only for patients with prescription opioid overdose (ARD, −4.0%; 95% CI, −5.% to 2.8%). We investigated the secondary outcome of MOUD treatment alone. Among the 6131 patients who did not file an MOUD claim in the 90 days before the index overdose, 280 individuals (4.6%) had a claim for MOUD following the overdose. In adjusted analyses, patients who were older, women, black race, and experienced a prescription opioid overdose were less likely to obtain MOUD treatment, while patients with a prescription for a benzodiazepine or treatment encounters for OUD were more likely (eTable 5 in the Supplement). We examined the timing of follow-up treatment following the index overdose, with results of the Kaplan-Meier failure analysis shown in eFigure 2 in the Supplement. Among all 1069 patients who obtained follow-up treatment, 318 individuals (29.7%) did so in 7 or fewer days after the overdose. In addition, we performed a sensitivity analysis excluding 233 patients (3.6%) who did not have claims beyond the 90-day follow-up period, which demonstrated equivalent outcomes to the primary analysis (eTable 6 in the Supplement). Discussion We analyzed commercial insurance claims to determine how often patients obtained treatment for OUD in the 90 days following ED presentation for nonfatal opioid overdose. Most had not received OUD treatment immediately preceding the overdose. Among that group, we found that only 11.1% of patients obtained follow-up treatment through an encounter in the outpatient setting, inpatient treatment, or filled prescriptions for a buprenorphine or naltrexone. The few patients that recently received treatment had a higher incidence of follow-up treatment. Despite the increasing number of overdoses across the years of this study, there was no significant change in the proportion of patients receiving follow-up treatment. Given that patients with commercial insurance likely have a superior ability to access care compared with patients who have public insurance, this persistently low rate suggests an opportunity for improvement. Disparities in the receipt of follow-up treatment with regard to race/ethnicity, age, and age persisted within this cohort. In particular, black patients were half as likely to obtain treatment following overdose compared with non-Hispanic white patients. This disparity was present regardless of whether the overdose was due to heroin or prescription opioids. To our knowledge, these disparities in treatment following opioid overdose have not been previously documented. However, our findings are consistent with emerging evidence that there are disparities in Figure 1. Patient Outcomes Stratified by Overdose Type and Treatment Status Before Overdose A All patients B Heroin overdose C Prescription overdose 100 100 100 Both MOUD and treatment encounter, 90 d after index opioid overdose 80 80 80 Only MOUD claim, 90 d after index opioid overdose Only treatment encounter, 60 90 d after index opioid overdose 60 60 40 40 40 20 20 20 0 0 0 All Patients No preceding MOUD Preceding MOUD No preceding MOUD Preceding MOUD or treatment or treatment or treatment or treatment Data shown for status at 90 days before overdose for all patients (A), heroin overdose (B), and prescription opioid overdose (C). MOUD indicates medication for opioid use disorder. JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 6/13 Proportion of Patients, % Proportion of Patients, % Proportion of Patients, % JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose 30,32,33,36 buprenorphine treatment with regard to race/ethnicity and sex. Although this study cannot determine whether these disparities are associated with patient preferences, barriers to access, implicit or explicit bias, or other causes, it is important to better understand and account for these factors when designing systems that seek to improve engagement and equity in treatment. Previous studies have examined changes in treatment rates before and after opioid overdose 12-14 using data from individual states. These studies primarily focused on medication treatment, with only one study including a limited range of treatment encounters. Our study included a range of possible treatments, from outpatient clinic visits to inpatient residential treatment. In general, we found that fewer than half of patients who obtained follow-up treatment received medication. Treatment with opioid agonists has been associated with reduced risk of relapse by 50% compared Table 2. Adjusted Probability of Follow-up Treatment After Opioid Overdose, for Patients Not Treated Before Overdose b c Patient characteristics Average adjusted probability, % (95% CI) P value Overdose type Prescription opioid 8.3 (7.3- 9.2) [Reference] Heroin 17.1 (15.1-19.2) <.001 Age, at mean, y 9.9 (9.1-10.7) <.001 Sex Male 11.9 (10.9-13.0) [Reference] Female 10.1 (9.1-11.3) .04 Race/ethnicity Non-Hispanic white 12.1 (11.1-13.0) [Reference] Black 6.1 (4.0-8.3) <.001 Hispanic 8.5 (6.1-11.0) .009 Asian 10.2 (2.8-17.5) .62 Unknown 10.1 (7.4-12.8) .18 Year 2011, quarter 4 12.2 (7.9-16.6) [Reference] 2012 9.3 (7.6-11.3) .22 2013 11.5 (9.6-13.5) .75 2014 10.0 (8.3-11.7) .32 2015 12.9 (11.1-14.6) .82 2016, quarters 1-3 11.1 (9.5-13.0) .64 Region Northeast 14.0 (11.6-16.6) [Reference] Results are given for patients who did not receive South 10.4 (9.1-11.4) .01 treatment for 90 days before the index opioid Midwest 11.1 (9.7-12.7) .07 overdose, defined as either a pharmacy claim for West 11.0 (9.3-12.8) .06 medication for opioid use disorder or medical claim 90 d Before overdose for opioid use disorder treatment encounter. Anxiety treatment Estimated with logistic regression model using No 10.3 (9.4-11.2) [Reference] predictive margins. Average adjusted probability is the adjusted rate, holding covariates at their actual Yes 13.8 (11.7-15.8) .004 values, at which patients obtain follow-up treatment Depression treatment within 90 days after the index opioid overdose, No 10.9 (10.1-11.9) [Reference] defined as either a pharmacy claim for medication for Yes 11.6 (9.7-13.5) .64 opioid use disorder or medical claim for opioid use Prescription opioid claim disorder treatment encounter. No 11.0 (9.9-12.1) [Reference] P values are given for average marginal effects, which represent the difference in adjusted probability Yes 11.2 (9.8-12.7) .84 between a given characteristic and the Benzodiazepine claim reference group. No 10.3 (9.4-11.2) [Reference] Average adjusted probability for continuous variable Yes 13.2 (11.4-15.0) .009 (age) is given for the mean patient age (46.3 years). JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 7/13 JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose with behavioral treatment alone. Better understanding of current treatment and referral patterns 57,58 may help inform efforts to expand evidence-based practices. We hypothesized that the rate of follow-up treatment would be higher for patients with commercial insurance, given potentially greater resources and access to care. While we cannot directly compare across studies, the rate of OUD treatment in this cohort did not appear to be appreciably higher in this cohort than that described in other populations. Not all patients can be expected to engage in treatment after overdose. Higher rates of treatment engagement have been observed in experimental settings, often with screening of patients for substance use 22,59-62 disorder. While the optimal rate of follow-up treatment may be difficult to estimate, there is still need for widely implemented interventions that may help patients overcome the many pervasive 4,15 barriers to accessing care. We intentionally examined outcomes for a short time following the overdose. Recent evidence suggests that risk of death is high immediately following overdose, with nearly 5% of deaths occurring within 2 days of discharge from the ED. In a secondary analysis, only 30% of patients who obtained follow-up did so within 7 days. Patients may benefit from rapid linkage to treatment, Figure 2. Proportion of Index Opioid Overdoses by Quarter, Stratified by Overdose Type and Receipt of Follow-up Treatment Heroin overdose, follow-up treatment Heroin overdose, no follow-up treatment Prescription opioid overdose, follow-up treatment Prescription opioid overdose, no follow-up treatment 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 2011 2012 2013 2014 2015 2016 Quarter Figure 3. Average Adjusted Probability of Follow-up Treatment After Opioid Overdose, by Overdose Type and Race/Ethnicity Race/ethnicity Non-Hispanic white Hispanic Black Estimated from logistic regression model with interaction term for overdose type and race/ethnicity. Error bars denote 95% confidence intervals for average adjusted probability. Results shown only for patients who had not received treatment for opioid use disorder in the 90 days before the index opioid overdose. Race/ethnicity was self-reported or derived Heroin overdose Prescription opioid overdose from other administrative data sources. JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 8/13 Opioid overdoses, % Adjusted probability of follow-up treatment, % JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose potentially through recovery specialists who can provide navigation and harm reduction counseling 3-5 regardless of the client’s willingness to engage in treatment. Limitations This study has several limitations. First, we cannot account for patients who pay for OUD treatment out-of-pocket. Although treatment services, including MOUD, were covered by the insurer during the study period, some patients may have elected to pursue alternative options. Second, this study did not include patients who obtain methadone maintenance therapy. Methadone is an important treatment modality for many patients with opioid use disorder. However, methadone was not covered for this indication by the insurer during the study period. It is possible that patients in this cohort obtained methadone through self-pay or other mechanisms, although this rate cannot be 63-65 estimated from these data and is difficult to extrapolate from other sources. Third, these data do not specifically account for patient deaths in the days following the index overdose. However, additional analysis that only included patients known to have survived to the end of the follow-up period showed similar results. Fourth, the use of administrative claims data in this study limits our ability to ascertain the reasons that patients obtain or do not obtain follow-up treatment. It is not known whether patients do not receive appropriate referrals, lack treatment facilities in their communities, or may be unwilling to engage in treatment. A corollary limitation is that patients may have received prescriptions for MOUD but not filled those prescriptions. Fifth, this cohort likely includes patients who may not have OUD, which may explain differential rates in follow-up treatment for patients with heroin and prescription opioid overdose. Regardless, patients with accidental prescription opioid overdose also should obtain timely follow-up for reevaluation, medication adjustment, and discussion of the long-term risks associated with opioid use. Conclusions Engagement of patients into treatment following opioid overdose is necessary to prevent subsequent opioid overdose death and other harm. Among commercially insured patients who were not receiving active addiction treatment, only 11.1% received follow-up treatment after an overdose. We showed apparent disparities in treatment with regard to race/ethnicity (eg, black patients were half as likely to obtain follow-up compared with non-Hispanic white patients), sex, and age. Research is needed to better understand the mechanisms behind these disparities. As health professionals adopt evidence-based practices for initiating medications for treatment of OUD and linking patients to sustained treatment, payers and policy makers should implement strategies to overcome systemic barriers to ensure that patients are given the best opportunity to access timely treatment. These interventions must account for disparities to ensure expanded and equitable access to life-saving treatment following overdose. ARTICLE INFORMATION Accepted for Publication: March 22, 2020. Published: May 27, 2020. doi:10.1001/jamanetworkopen.2020.5852 Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Kilaru AS et al. JAMA Network Open. Corresponding Author: Austin S. Kilaru, MD, Center for Emergency Care Policy and Research, Perelman School of Medicine, Department of Emergency Medicine, University of Pennsylvania, 421 Guardian Dr, 1303 Blockley Hall, Philadelphia, PA 19104 (austin.kilaru@pennmedicine.upenn.edu). Author Affiliations: National Clinician Scholars Program, Corporal Michael J. Crescenz Veterans Affairs Medical Center, University of Pennsylvania, Philadelphia (Kilaru, Lowenstein, Khatri); Center for Emergency Care Policy and Research, Perelman School of Medicine, Department of Emergency Medicine, University of Pennsylvania, JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 9/13 JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose Philadelphia (Kilaru, Meisel, Perrone, Khatri, Delgado); Leonard Davis Institute of Health Economics, University of Pennsylvania, Philadelphia (Kilaru, Xiong, Lowenstein, Meisel, Perrone, Khatri, Delgado); Penn Injury Science Center, Philadelphia, Pennsylvania (Kilaru, Lowenstein, Meisel, Khatri, Delgado); Perelman School of Medicine, Department of Biostatistics and Epidemiology, University of Pennsylvania, Philadelphia (Mitra, Delgado). Author Contributions: Dr Kilaru 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: Kilaru, Lowenstein, Meisel, Perrone, Khatri, Delgado. Acquisition, analysis, or interpretation of data: Kilaru, Xiong, Meisel, Mitra, Delgado. Drafting of the manuscript: Kilaru, Xiong. Critical revision of the manuscript for important intellectual content: Kilaru, Lowenstein, Meisel, Perrone, Khatri, Mitra, Delgado. Statistical analysis: Kilaru, Xiong, Mitra. Obtained funding: Kilaru, Delgado. Administrative, technical, or material support: Meisel, Khatri, Delgado. Supervision: Kilaru, Meisel, Delgado. Conflict of Interest Disclosures: Dr Delgado reported an honorarium from United Health Group outside the submitted work. No other disclosures were reported. Funding/Support: This study was supported by a pilot grant from the Leonard Davis Institute of Health Economics at the University of Pennsylvania (Dr Kilaru). Role of the Funder/Sponsor: The Leonard Davis Institute of Health Economics at the University of Pennsylvania had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Disclaimer: The contents do not represent the views of the US Department of Veteran Affairs or the US government. REFERENCES 1. Vivolo-Kantor AM, Seth P, Gladden RM, et al. Vital signs: trends in emergency department visits for suspected opioid overdoses—United States, July 2016-September 2017. MMWR Morb Mortal Wkly Rep. 2018;67(9):279-285. doi:10.15585/mmwr.mm6709e1 2. Tedesco D, Asch SM, Curtin C, et al. Opioid abuse and poisoning: trends in inpatient and emergency department discharges. Health Aff (Millwood). 2017;36(10):1748-1753. doi:10.1377/hlthaff.2017.0260 3. Martin A, Mitchell A, Wakeman S, White B, Raja A. Emergency department treatment of opioid addiction: an opportunity to lead. Acad Emerg Med. 2018;25(5):601-604. doi:10.1111/acem.13367 4. Doran KM, Raja AS, Samuels EA. Opioid overdose protocols in the emergency department: are we asking the right questions? Ann Emerg Med. 2018;72(1):12-15. doi:10.1016/j.annemergmed.2018.05.024 5. Samuels EA, D’Onofrio G, Huntley K, et al. A quality framework for emergency department treatment of opioid use disorder. Ann Emerg Med. 2019;73(3):237-247. doi:10.1016/j.annemergmed.2018.08.439 6. Herring AA, Perrone J, Nelson LS. Managing opioid withdrawal in the emergency department with buprenorphine. Ann Emerg Med. 2019;73(5):481-487. doi:10.1016/j.annemergmed.2018.11.032 7. Larochelle MR, Bernstein R, Bernson D, et al. Touchpoints—opportunities to predict and prevent opioid overdose: a cohort study. Drug Alcohol Depend. 2019;204:107537. doi:10.1016/j.drugalcdep.2019.06.039 8. D’Onofrio G, McCormack RP, Hawk K. Emergency departments: a 24/7/365 option for combating the opioid crisis. N Engl J Med. 2018;379(26):2487-2490. doi:10.1056/NEJMp1811988 9. Houry DE, Haegerich TM, Vivolo-Kantor A. Opportunities for prevention and intervention of opioid overdose in the emergency department. Ann Emerg Med. 2018;71(6):688-690. doi:10.1016/j.annemergmed.2018.01.052 10. Winetsky D, Weinrieb RM, Perrone J. Expanding treatment opportunities for hospitalized patients with opioid use disorders. JHospMed. 2018;13(1):62-64. doi:10.12788/jhm.2861 11. Naeger S, Mutter R, Ali MM, Mark T, Hughey L. Post-discharge treatment engagement among patients with an opioid-use disorder. J Subst Abuse Treat. 2016;69:64-71. doi:10.1016/j.jsat.2016.07.004 12. Larochelle MR, Bernson D, Land T, et al. Medication for opioid use disorder after nonfatal opioid overdose and association with mortality: a cohort study. Ann Intern Med. 2018;169(3):137-145. doi:10.7326/M17-3107 13. Frazier W, Cochran G, Lo-Ciganic WH, et al. Medication-assisted treatment and opioid use before and after overdose in Pennsylvania Medicaid. JAMA. 2017;318(8):750-752. doi:10.1001/jama.2017.7818 JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 10/13 JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose 14. Koyawala N, Landis R, Barry CL, Stein BD, Saloner B. Changes in outpatient services and medication use following a non-fatal opioid overdose in the West Virginia Medicaid program. J Gen Intern Med. 2019;34(6): 789-791. doi:10.1007/s11606-018-4817-8 15. D’Onofrio G, Edelman EJ, Hawk KF, et al. Implementation facilitation to promote emergency department- initiated buprenorphine for opioid use disorder: protocol for a hybrid type III effectiveness-implementation study (Project ED HEALTH). Implement Sci. 2019;14(1):48. doi:10.1186/s13012-019-0891-5 16. Katz EB, Carrier ER, Umscheid CA, Pines JM. Comparative effectiveness of care coordination interventions in the emergency department: a systematic review. Ann Emerg Med. 2012;60(1):12-23.e1. doi:10.1016/j. annemergmed.2012.02.025 17. Carrier EYT, Holzwart RA. Coordination between Emergency and Primary Care Physicians. National Institute for Health Care Reform; February 2011. 18. Medford-Davis L, Marcozzi D, Agrawal S, Carr BG, Carrier E. Value-based approaches for emergency care in a new era. Ann Emerg Med. 2017;69(6):675-683. doi:10.1016/j.annemergmed.2016.10.031 19. Weiner SG, Baker O, Bernson D, Schuur JD. One-year mortality of patients after emergency department treatment for nonfatal opioid overdose. Ann Emerg Med. 2020;75(1):13-17. doi:10.1016/j.annemergmed.2019. 04.020 20. Olfson M, Crystal S, Wall M, Wang S, Liu SM, Blanco C. Causes of death after nonfatal opioid overdose. JAMA Psychiatry. 2018;75(8):820-827. doi:10.1001/jamapsychiatry.2018.1471 21. D’Onofrio G, O’Connor PG, Pantalon MV, et al. Emergency department–initiated buprenorphine/naloxone treatment for opioid dependence: a randomized clinical trial. JAMA. 2015;313(16):1636-1644. doi:10.1001/jama. 2015.3474 22. D’Onofrio G, Chawarski MC, O’Connor PG, et al. Emergency department–initiated buprenorphine for opioid dependence with continuation in primary care: outcomes during and after intervention. J Gen Intern Med.2017;32 (6):660-666. doi:10.1007/s11606-017-3993-2 23. Busch SH, Fiellin DA, Chawarski MC, et al. Cost-effectiveness of emergency department-initiated treatment for opioid dependence. Addiction. 2017;112(11):2002-2010. doi:10.1111/add.13900 24. Hu T, Snider-Adler M, Nijmeh L, Pyle A. Buprenorphine/naloxone induction in a Canadian emergency department with rapid access to community-based addictions providers. CJEM. 2019;21(4):492-498. doi:10.1017/ cem.2019.24 25. Kilaru AS, Perrone J, Kelley D, et al. Participation in a hospital incentive program for follow-up treatment for opioid use disorder. JAMA Netw Open. 2020;3(1):e1918511. doi:10.1001/jamanetworkopen.2019.18511 26. Pitt AL, Humphreys K, Brandeau ML. Modeling health benefits and harms of public policy responses to the US opioid epidemic. Am J Public Health. 2018;108(10):1394-1400. doi:10.2105/AJPH.2018.304590 27. Ahmed OM, Mao JA, Holt SR, et al. A scalable, automated warm handoff from the emergency department to community sites offering continued medication for opioid use disorder: lessons learned from the EMBED trial stakeholders. J Subst Abuse Treat. 2019;102:47-52. doi:10.1016/j.jsat.2019.05.006 28. Governor Baker Signs Second Major Piece of Legislation to Address Opioid Epidemic in Massachusetts. Published August 14, 2018. Accessed September 1, 2018. https://www.mass.gov/news/governor-baker-signs- second-major-piece-of-legislation-to-address-opioid-epidemic-in 29. Larochelle MR, Liebschutz JM, Zhang F, Ross-Degnan D, Wharam JF. Opioid prescribing after nonfatal overdose and association with repeated overdose: a cohort study. Ann Intern Med. 2016;164(1):1-9. doi:10.7326/ M15-0038 30. Lagisetty PA, Ross R, Bohnert A, Clay M, Maust DT. Buprenorphine treatment divide by race/ethnicity and payment. JAMA Psychiatry. 2019;76(9):979-981. doi:10.1001/jamapsychiatry.2019.0876 31. Allen B, Nolan ML, Kunins HV, Paone D. Racial differences in opioid overdose deaths in New York City, 2017. JAMA Intern Med. 2019;179(4):576-578. doi:10.1001/jamainternmed.2018.7700 32. Hadland SE, Wharam JF, Schuster MA, Zhang F, Samet JH, Larochelle MR. Trends in receipt of buprenorphine and naltrexone for opioid use disorder among adolescents and young adults, 2001-2014. JAMA Pediatr. 2017;171 (8):747-755. doi:10.1001/jamapediatrics.2017.0745 33. Shiels MS, Freedman ND, Thomas D, Berrington de Gonzalez A. Trends in US drug overdose deaths in non-Hispanic black, Hispanic, and non-Hispanic white persons, 2000-2015. Ann Intern Med. 2018;168(6): 453-455. doi:10.7326/M17-1812 34. Saloner B, Karthikeyan S. Changes in substance abuse treatment use among individuals with opioid use disorders in the United States, 2004-2013. JAMA. 2015;314(14):1515-1517. doi:10.1001/jama.2015.10345 JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 11/13 JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose 35. Santoro TN, Santoro JD. Racial bias in the US opioid epidemic: a review of the history of systemic bias and implications for care. Cureus. 2018;10(12):e3733. doi:10.7759/cureus.3733 36. Krawczyk N, Feder KA, Fingerhood MI, Saloner B. Racial and ethnic differences in opioid agonist treatment for opioid use disorder in a US national sample. Drug Alcohol Depend. 2017;178:512-518. doi:10.1016/j.drugalcdep. 2017.06.009 37. Haffajee RL, Lin LA, Bohnert ASB, Goldstick JE. Characteristics of US counties with high opioid overdose mortality and low capacity to deliver medications for opioid use disorder. JAMA Netw Open. 2019;2(6):e196373. doi:10.1001/jamanetworkopen.2019.6373 38. Optum Claims Data. Accessed May 1, 2019. https://www.optum.com/solutions/data-analytics/data/real-world- data-analytics-a-cpl/claims-data.html 39. Sanghavi DAA, Hane C, Bleicher P. Optum Opioid Data, Health Affairs Blog.pdf. Health Affairs Blog; 2017. 40. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP; STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Ann Intern Med. 2007;147(8):573-577. doi:10.7326/0003-4819-147-8- 200710160-00010 41. Safe States Injury Surveillance Workgroup (ISW7). Consensus recommendations for national and state poisoning surveillance. Published April 2012. Accessed May 1, 2019. https://cdn.ymaws.com/www.safestates.org/ resource/resmgr/imported/ISW7%20Full%20Report_3.pdf 42. Green CA, Perrin NA, Janoff SL, Campbell CI, Chilcoat HD, Coplan PM. Assessing the accuracy of opioid overdose and poisoning codes in diagnostic information from electronic health records, claims data, and death records. Pharmacoepidemiol Drug Saf. 2017;26(5):509-517. doi:10.1002/pds.4157 43. Reardon JM, Harmon KJ, Schult GC, Staton CA, Waller AE. Use of diagnosis codes for detection of clinically significant opioid poisoning in the emergency department: a retrospective analysis of a surveillance case definition. BMC Emerg Med. 2016;16:11. doi:10.1186/s12873-016-0075-4 44. Rowe C, Vittinghoff E, Santos GM, Behar E, Turner C, Coffin PO. Performance measures of diagnostic codes for detecting opioid overdose in the emergency department. Acad Emerg Med. 2017;24(4):475-483. doi:10.1111/ acem.13121 45. American Medical Association. CPT (Current Procedural Terminology). 2019. Accessed May 1, 2019. https:// www.ama-assn.org/amaone/cpt-current-procedural-terminology 46. Centers for Disease Control and Prevention. International Classification of Diseases, Tenth Revision, Clinical Modification. Accessed May 1, 2019. https://www.cdc.gov/nchs/icd/icd10cm.htm 47. Centers for Disease Control and Prevention. ICD-9-CM Addenda, Conversion Table, and Guidelines. Accessed May1,2019. https://www.cdc.gov/nchs/icd/icd9cm_addenda_guidelines.htm 48. Saloner B, Levin J, Chang HY, Jones C, Alexander GC. Changes in buprenorphine-naloxone and opioid pain reliever prescriptions after the Affordable Care Act Medicaid Expansion. JAMA Netw Open. 2018;1(4):e181588. doi:10.1001/jamanetworkopen.2018.1588 49. Centers for Disease Control and Prevention. Analyzing prescription data and morphine milligram equivalents (MME). Updated October 23, 2019. Accessed August 1, 2019. https://www.cdc.gov/drugoverdose/resources/data.html 50. Centers for Medicare and Medicaid Services. Chronic Conditions Data Warehouse—opioid use disorder. 2019. Accessed May 1, 2019. https://www2.ccwdata.org/web/guest/condition-categories 51. United States Census Bureau. Geography program. Accessed May 1, 2019. https://www.census.gov/programs- surveys/geography.html 52. National Institute on Drug Abuse. Overdose death rates. Accessed February 28, 2020. https://www.drugabuse. gov/related-topics/trends-statistics/overdose-death-rates 53. AHFS Clinical Drug Information. AHFS pharmacologic therapeutic classification. Accessed May 1, 2019. https:// www.ahfsdruginformation.com/ahfs-pharmacologic-therapeutic-classification/ 54. Norton EC, Dowd BE, Maciejewski ML. Odds ratios—current best practice and use. JAMA. 2018;320(1):84-85. doi:10.1001/jama.2018.6971 55. Norton EC, Dowd BE, Maciejewski ML. Marginal effects—quantifying the effect of changes in risk factors in logistic regression models. JAMA. 2019;321(13):1304-1305. doi:10.1001/jama.2019.1954 56. Clark RE, Baxter JD, Aweh G, O’Connell E, Fisher WH, Barton BA. Risk factors for relapse and higher costs among Medicaid members with opioid dependence or abuse: opioid agonists, comorbidities, and treatment history. J Subst Abuse Treat. 2015;57:75-80. doi:10.1016/j.jsat.2015.05.001 JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 12/13 JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose 57. Mojtabai R, Mauro C, Wall MM, Barry CL, Olfson M. Medication treatment for opioid use disorders in substance use treatment facilities. Health Aff (Millwood). 2019;38(1):14-23. doi:10.1377/hlthaff.2018.05162 58. Sharfstein J, Meisel ZF. Low-value treatment for opioid addiction: what is to be done? JAMA Forum. Published July 25, 2019. Accessed November 7, 2019. https://newsatjama.jama.com/2019/07/25/jama-forum-low-value- treatment-for-opioid-addiction-what-is-to-be-done/ 59. Hawk K, D’Onofrio G. Emergency department screening and interventions for substance use disorders. Addict Sci Clin Pract. 2018;13(1):18. doi:10.1186/s13722-018-0117-1 60. D’Onofrio G, Degutis LC. Integrating Project ASSERT: a screening, intervention, and referral to treatment program for unhealthy alcohol and drug use into an urban emergency department. Acad Emerg Med. 2010;17(8): 903-911. doi:10.1111/j.1553-2712.2010.00824.x 61. Bogenschutz MP, Donovan DM, Mandler RN, et al. Brief intervention for patients with problematic drug use presenting in emergency departments: a randomized clinical trial. JAMA Intern Med. 2014;174(11):1736-1745. doi: 10.1001/jamainternmed.2014.4052 62. Edwards FJ, Wicelinski R, Gallagher N, McKinzie A, White R, Domingos A. Treating opioid withdrawal with buprenorphine in a community hospital emergency department: an outreach program. Ann Emerg Med. 2020;75 (1):49-56. doi:10.1016/j.annemergmed.2019.08.420 63. Polsky D, Arsenault S, Azocar F. Private Coverage of methadone in outpatient treatment programs. Psychiatr Serv. 2020;71(3):303-306. doi:10.1176/appi.ps.201900373 64. Reif S, Creedon TB, Horgan CM, Stewart MT, Garnick DW. Commercial health plan coverage of selected treatments for opioid use disorders from 2003 to 2014. J Psychoactive Drugs. 2017;49(2):102-110. doi:10.1080/ 02791072.2017.1300360 65. Fullerton CA, Kim M, Thomas CP, et al. Medication-assisted treatment with methadone: assessing the evidence. Psychiatr Serv. 2014;65(2):146-157. doi:10.1176/appi.ps.201300235 SUPPLEMENT. eFigure 1. Flowchart for Selection of Patient Cohort eTable 1. ICD-9-CM, ICD-10, CPT, and AHFS Codes for Selection of Patient Cohort and Patient Characteristics eTable 2. National Drug Codes for Medications for Opioid Use Disorder eTable 3. CPT, HCPCS, ICD-9-CM, and ICD-10-CM Codes for Treatment Encounters eTable 4. Adjusted Probability of Follow-up Treatment After Opioid Overdose for Patients Treated Prior to Overdose eTable 5. Adjusted Probability of MOUD Treatment After Opioid Overdose, Stratified by Treatment Status Prior to Overdose eFigure 2. Kaplan-Meier Failure Curve for Days to First Follow up Treatment Following Index ED Overdose eTable 6. Adjusted Probability of Follow-up Treatment After Opioid Overdose, Excluding Patients Without Known Claims Beyond 90-Day Follow-up Period (Sensitivity Analysis to Address Potential Mortality During Follow-up Period) eTable 7. Index Opioid Overdoses by specific ICD-9 or ICD-10 Diagnosis Code, With Number and Frequency for Each Diagnosis Code eTable 8. Patient Cohort and Unadjusted Outcomes, Stratified by Overdose Type and Treatment Status Before Overdose JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 13/13 Supplementary Online Content Kilaru AS, Xiong A, Lowenstein M, et al. Incidence of treatment for opioid use disorder following nonfatal overdose in commercially insured patients. JAMA Netw Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 eFigure 1. Flowchart for Selection of Patient Cohort eTable 1. ICD-9-CM, ICD-10, CPT, and AHFS Codes for Selection of Patient Cohort and Patient Characteristics eTable 2. National Drug Codes for Medications for Opioid Use Disorder eTable 3. CPT, HCPCS, ICD-9-CM, and ICD-10-CM Codes for Treatment Encounters eTable 4. Adjusted Probability of Follow-up Treatment After Opioid Overdose for Patients Treated Prior to Overdose eTable 5. Adjusted Probability of MOUD Treatment After Opioid Overdose, Stratified by Treatment Status Prior to Overdose eFigure 2. Kaplan-Meier Failure Curve for Days to First Follow up Treatment Following Index ED Overdose eTable 6. Adjusted Probability of Follow-up Treatment After Opioid Overdose, Excluding Patients Without Known Claims Beyond 90-Day Follow-up Period (Sensitivity Analysis to Address Potential Mortality During Follow-up Period) eTable 7. Index Opioid Overdoses by specific ICD-9 or ICD-10 Diagnosis Code, With Number and Frequency for Each Diagnosis Code eTable 8. Patient Cohort and Unadjusted Outcomes, Stratified by Overdose Type and Treatment Status Before Overdose This supplementary material has been provided by the authors to give readers additional information about their work. © 2020 Kilaru AS et al. JAMA Network Open. 1. Flowchart for Selection of Patient Cohort © 2020 Kilaru AS et al. JAMA Network Open. ICD-9-CM, ICD-10, CPT, and AHFS Codes for Selection of Patient Cohort and Patient Characteristics a a ICD-9-CM Diagnosis Codes ICD-10 Diagnosis Codes CPT Codes AHFS Pharmacologic- Therapeutic Codes Opioid Overdose Heroin: Heroin: 965.01, E850.0 T40.1X Prescription: Prescription: 965.00, 965.02, 965.09, T40.0X, T40.2-4X, T40.6X E.850.1, E.850.2 Emergency 99281, Department 99282, Encounters 99283, 99284, 99285 Cancer (Malignant 140.X – 208.X C00X – C97X Neoplasm) Diagnosis 209.0 – 209.3 V10.X 293.84, 300.00, 300.01, 300.02, F06.4, F40.00, F40.01, F40.02, F40.10, Anxiety Diagnosis 300.09, 300.10, 300.20, 300.21, F40.11, F40.210, F40.218, F40.220, 300.22, 300.23, 300.29, 300.3, F40.228, F40.230, F40.231, F40.232, 300.5, 300.89, 300.9, 308.0, F40.233, F40.240, F40.241, F40.242, 308.1, 308.2, 308.3, 308.4, 308.9, F40.243, F40.248, F40.290, F40.291, 309.81, 313.0, 313.1, 313.21, F40.298, F40.8, F40.9, F41.0, F41.1, 313.22, 313.3, 313.82, 313.83 F41.3, F41.8, F41.9, F42, F42.2, F42.3, F42.4, F42.8, F42.9, F43.0, F43.10, F43.11, F43.12, F44.9, F45.8, F48.8, F48.9, F93.8, F99, R45.2, R45.5, R45.6, R45.7 296.20, 296.21, 296.22, 296.23, F32.0, F32.1, F32.2, F32.3, F32.4, Depression Diagnosis 296.24, 296.25, 296.26, 296.30, F32.5, F32.89, F32.9, F33.0, F33.1, 296.31, 296.32, 296.33, 296.34, F33.2, F33.3, F33.8, F33.40, F33.41, 296.35, 296.36, 300.4, 311, V79.0 F33.42, F33.9, F34.1 Prescription Opioid 280808 Full Agonist Benzodiazepine 281208, 282408 © 2020 Kilaru AS et al. JAMA Network Open. ICD-9-CM diagnosis codes are used for any claim prior to October 1 2015. ICD-10 diagnosis codes are used for claims on that date or after. Benign neoplasms or neoplasms of uncertain behavior were excluded © 2020 Kilaru AS et al. JAMA Network Open. . National Drug Codes for Medications for Opioid Use Disorder National Drug Codes (NDC) 54017613, 54017713, 54018813, 54018913, 74201201, 74201232, 93360021, 93360040, 93360121, Buprenorphine and 93360140, 93360221, 93360240, 93360321, 93360340, 93537856, 93537956, 93572056, 93572156, Buprenorphine- 149075701, 228315303, 228315403, 228315473, 228315503, 228315567, 228315573, 228315603, Naloxone 378092393, 378092493, 406192303, 406192403, 406802003, 409201203, 409201232, 490005100, 490005130, 490005160, 490005190, 12496010001, 12496010002, 12496010005, 12496030001, 12496030002, 12496030005, 12496075701, 12496075705, 12496120201, 12496120203, 12496120401, 12496120403, 12496120801, 12496120803, 12496121201, 12496121203, 12496127802, 12496128302, 12496130602, 12496131002, 16590066605, 16590066630, 16590066705, 16590066730, 16590066790, 21695051510, 23490927003, 23490927006, 23490927009, 35356000407, 35356000430, 35356055530, 35356055630, 35356060504, 35356060704, 38779088800, 38779088801, 38779088803, 38779088805, 38779088806, 38779088809, 40042001001, 42023017901, 42023017905, 42291017430, 42291017530, 42858035340, 42858049340, 42858050103, 42858050203, 42858058640, 42858075040, 42858083940, 43063018407, 43063018430, 43063066706, 43063075306, 49452129201, 49452129202, 49452129203, 49452825301, 49452825302, 49452825303, 49999039507, 49999039515, 49999039530, 49999063830, 49999063930, 50268014411, 50268014415, 50268014511, 50268014515, 50383028793, 50383029493, 50383092493, 50383093093, 51552076501, 51552076502, 51552076505, 51552076506, 51552076509, 51552076510, 51552076550, 51927101200, 52959030430, 52959074930, 53217013830, 54123011430, 54123090730, 54123091430, 54123092930, 54123095730, 54123098630, 54569141600, 54569141601, 54569549600, 54569573900, 54569573901, 54569573902, 54569632500, 54569632600, 54569639900, 54569640800, 54569657800, 54868570700, 54868570701, 54868570702, 54868570703, 54868570704, 54868575000, 55045378403, 55390010010, 55700014730, 55700018430, 55700030230, 55700030330, 55700057904, 58284010014, 59011075004, 59011075104, 59011075204, 59011075704, 59011075804, 59385001201, 59385001230, 59385001401, 59385001430, 59385001601, 59385001630, 59385002101, 59385002160, 59385002201, 59385002260, 59385002301, 59385002360, 59385002401, 59385002460, 59385002501, 59385002560, 59385002601, 59385002660, 59385002701, 59385002760, 60429058611, 60429058630, 60429058633, 60429058711, 60429058730, 60429058733, 62756045983, 62756046083, 62756096983, 62756097083, 62991158301, 62991158302, 62991158303, 62991158304, 62991158306, 62991158307, 62991158308, 63275992201, 63275992202, 63275992203, 63275992204, 63275992205, 63275992207, 63370090506, 63370090509, 63370090510, 63370090515, 63459030042, 63481016101, 63481016160, 63481020701, 63481020760, 63481034801, 63481034860, 63481051901, 63481051960, 63481068501, 63481068560, 63481082001, 63481082060, 63481095201, 63481095260, 63629402801, 63629403401, 63629403402, 63629403403, 63629409201, 63874108403, 63874108503, 63874117303, 65162041503, 65162041603, 65757030001, 65757030202, 66336001630, 68071138003, 68071151003, 68258299103, 68258299903, 68308020230, 68308020830, 35356060604, 53217024630, 55887031204, 55887031215, 63874117403, 66336001530, 406192309, 406192409, 406800503, 50090157100, 55700056804, 60846097003, 60846097103, 62175045232, 62175045832, 62756045964, 62756046064, 62756096964, 62756097064, 63629409202, 63629507401, 63629712501, 63629712502, 63629712503, 63629712504, 63629712505, 63629712506, 63629712507, 63629712601, 63629712602, 63629712603, 63629712604, 63629712605, 63629712606, 63629712607, 63629712608, 63629727001, 63629727002, 64725093003, 64725093004, 64725192403, 64725192404, 65162041509, 65162041609, 71335035301, 71335035302, 71335035303, 71335035304, 71335035305, 71335035306, 71335035307, 76519117000, 76519117001, 76519117002, 76519117003, © 2020 Kilaru AS et al. JAMA Network Open. 43063059115, 47335032683, 47335032688, 50436010501, 51224020630, 51224020650, Naltrexone 51285027501, 51285027502, 52152010502, 52152010504, 52152010530, 54868557400, 63459030042, 65694010003, 65694010010, 65757030001, 65757030202, 68084029111, 68084029121, 68094085362, 68115068030, 56001122, 56001130, 56001170, 56007950, 56008050, 185003901, 185003930, 406009201, 406009203, 406117001, 406117003, 555090201, 555090202, 16729008101, 16729008110, 42291063230 © 2020 Kilaru AS et al. JAMA Network Open. CPT, HCPCS, ICD-9-CM, and ICD-10-CM Codes for Treatment Encounters Current Procedure Healthcare Common ICD-9-CM Diagnosis ICD-10-CM Diagnosis Codes Terminology (CPT) Codes Procedure Coding System Codes (claim on or after October 1 (HCPCS) Codes (claim prior to October 1 2015) 2015) Office or Outpatient Visit: Drug, Alcohol, and Behavioral 304.00-304.03 Opioid use, abuse, 99201-99205 Health Services (Outpatient 305.50-305.53 dependence: 99211-99215 and Inpatient): 304.70-304.73 F1190, F11920-F11922, H0001, H0002 965.00-965.02, 965.09 F11929 Psychiatric Diagnosis: H0004-H0019 E85.00-E85.02 F1193, F1194, F11950- 90791-90792 H0031-H0040 E93.50-E93.51 F11951, F11959, F11981, G0396-G0397 F11982, F11988 Psychotherapy Services: F1199, F1110, F11120- 90832-90839 Mental Health Services NOS: F11122, 90853 H0046-H0047 F1129, F1114, F11150- 90863 F11151, F11159, F11181, 90875-90876 Halfway House / Treatment F11182, F11188, F1119, 90801-90815 Program: F1120, F11220-F11222, 90824 H2034-H2036 F11229, F1123, F1124, 90862 F11250, F11251, F11259, Clinic Visit / Case F11281, F11282, F11288, Screening, Brief Intervention, Management: F1129 and Referral to Treatment T1015-T1017 (SBIRT): T1001 Poisoning: 99408-99409 T400X1*, T400X2*, MOUD Codes: T400X3*,T400X4* J0571-J0575 T401X1*, T401X2*, T401X3*, J1230, J2315 T401X4* T403X1*, T403X2*, T403X3*, Excluded T403X4* Methadone maintenance T402X1*, T402X2*, T402X3*, therapy: T402X4* H0020 T404X1*, T404X2*, T404X3*, T404X4* © 2020 Kilaru AS et al. JAMA Network Open. T40601*, T40602*, T40603*, T40604* T40605*, T40691*, T40692*, T40693*, T40694*, T40695*, T403X5* * Position can include A, D, or © 2020 Kilaru AS et al. JAMA Network Open. Adjusted probability of follow-up treatment after opioid overdose for patients treated prior to overdose Average Adjusted Prediction (95% CI), % Overdose Type Prescription 64.2 (57.5 to 70.7) -- Heroin 61.3 (56.0 to 66.5) .55 Age (years), mean (SD) 62.5 (59.0 to 66.1) .21 Sex Male 62.2 (57.5 to 70.0) -- Female 62.7 (57.4 to 68.2) .89 Race/Ethnicity White 62.8 (58.7 to 66.9) -- Black 75.9 (63.4 to 88.4) .06 Hispanic 61.3 (47.3 to 75.3) .84 59.6 (13.8 to Asian .89 105.3) Unknown 51.7 (39.1 to 64.1) .10 Year 2011 Q4 48.3 (22.2 to 74.5) -- 2012 66.0 (56.1 to 75.8) .22 2013 66.9 (58.9 to 75.8) .19 2014 60.9 (52.8 to 69.0) .37 2015 54.8 (46.9 to 62.7) .65 2016 Q1-3 66.4 (59.2 to 73.6) .19 Region Northeast 73.3 (64.7 to 82.0) South 58.3 (51.9 to 64.1) .007 Midwest 67.3 (61.0 to 73.6) .26 West 53.2 (44.4 to 62.0) .002 Anxiety treatment, No 62.3 (57.6 to 67.1) 90 d prior to overdose Yes 62.6 (56.8 to 68.5) .94 Depression treatment, No 63.3 (59.0 to 67.6) 90 d prior to overdose Yes 60.7 (54.1 to 67.2) .53 Prescription opioid claim, No 65.1 (60.4 to 69.8) 90 d prior to overdose Yes 56.9 (48.9 to 64.8) .11 Benzodiazepine claim, 90 No 60.1 (55.4 to 64.7) d prior to overdose Yes 66.7 (60.7 to 72.6) .10 © 2020 Kilaru AS et al. JAMA Network Open. Adjusted probability of MOUD treatment after opioid overdose, stratified by treatment status prior to overdose n = 6131 n = 320 P P (95% CI), % (95% CI), % Prescription 3.2 (2.6 to 3.8) -- 55.8 (45.6 to 66.0) -- Heroin 6.8 (5.6 to 8.1) < .001 53.5 (46.2 to 60.9) .75 A at mean 3.6 (3.0 to 4.1) < .001 54.4 (49.2 to 59.6) .22 Male 5.1 (4.5 to 5.7) -- 54.6 (47.9 to 61.3) -- Female 3.9 (3.2 to 4.6) .03 54.0 (45.0 to 62.9) .92 White 4.8 (4.3 to 5.3) -- 53.2 (47.1 to 59.3) -- Black 3.0 (1.6 to 4.4) .03 71.8 (50.8 to 92.7) .10 Hispanic 4.1 (2.3 to 5.9) .47 70.4 (50.8 to 90.0) .10 Asian 5.9 (-.19 to 12.1) .71 62.0 (2.6 to 126.7) .79 Unknown 4.3 (2.6 to 6.0) .60 40.5 (20.7 to 60.1) .23 2011 Q4 6.0 (2.9 to 9.2) -- 27.3 (-3.9 to 58.6) -- 2012 4.2 (3.0 to 5.5) .29 48.8 (34.6 to 63.0) .22 2013 6.4 (4.9 to 7.8) .85 61.7 (50.4 to 72.9) .05 2014 3.4 (2.4 to 4.4) .12 58.1 (45.1 to 71.0) .07 2015 4.4 (3.4 to 5.5) .35 50.4 (37.8 to 63.1) .18 2016 Q1-3 4.4 (3.3 to 5.5) .33 54.2 (42.6 to 65.9) .11 Northeast 4.4 (3.0 to 5.9) -- 55.3 (40.0 to 70.7) -- South 5.1 (4.2 to 6.0) .46 51.9 (43.4 to 60.4) .70 Midwest 3.9 (3.0 to 4.8) .52 58.7 (48.7 to 68.8) .72 West 4.8 (3.6 to 5.9) .75 52.6 (40.0 to 65.4) .79 No 4.3 (3.7 to 4.9) -- 54.5 (47.3 to 61.7) -- 90 d prior to overdose Yes 5.4 (4.2 to 6.6) .13 54.1 (45.2 to 63.0) .95 No 4.4 (3.8 to 5.0) -- 59.0 (52.7 to 65.3) -- 90 d prior to overdose Yes 5.0 (3.8 to 6.3) .39 42.9 (32.5 to 53.2) .01 No 4.3 (3.6 to 4.9) -- 57.0 (50.4 to 63.4) -- 90 d prior to overdose Yes 5.1 (4.0 to 6.2) .22 47.3 (35.2 to 59.3) .20 , 90 d prior No 4.0 (3.4 to 4.6) -- 53.9 (46.8 to 61.0) -- to overdose Yes 6.2 (2.9 to 7.5) .003 55.0 (46.3 to 63.8) .86 No 4.0 (3.5 to 4.5) -- 50.9 (42.6 to 59.1) -- d prior to overdose Yes 10.1 (7.5 to 12.7) < .001 57.2 (50.0 to 64.5) .27 © 2020 Kilaru AS et al. JAMA Network Open. Kaplan-Meier Failure Curve for Days to First Follow Up Treatment Following Index ED Overdose Follow up treatment includes claim for OUD treatment encounter or pharmacy claim for MOUD © 2020 Kilaru AS et al. JAMA Network Open. Adjusted probability of follow-up treatment after opioid overdose, excluding patients without known claims beyond 90-day follow-up period (sensitivity analysis to address potential mortality during follow-up period). n = 6131 n = 320 P P (95% CI), % (95% CI), % Prescription 8.5 (7.5 to 9.4) -- 64.2 (57.5 to 70.1) -- Heroin 18.0 (15.8 to 20.2) < .001 62.2 (56.8 to 67.6) .69 A mean (SD) 10.1 (9.3 to 11.0) < .001 63.1 (59.5 to 66.7) .22 Male 12.4 (11.3 to 13.7) -- 62.5 (57.7 to 67.3) -- Female 10.4 (9.2 to 11.5) .01 63.7 (58.2 to 69.1) .76 White 12.5 (11.6 to 13.4) -- 63.3 (59.2 to 67.4) -- Black 6.3 (4.1 to 8.6) < 0.001 75.8 (63.2 to 88.4) .06 Hispanic 9.1 (6.8 to 11.4) .02 60.8 (46.6 to 74.9) .73 Asian 11.1 (3.2 to 19.1) .75 60.0 (14.4 to 105.6) .89 Unknown 10.3 (7.6 to 13.1) .15 53.0 (40.1 to 65.9) .14 2011 Q4 12.4 (8.0 to 16.7) -- 48.2 (22.0 to 74.2) -- 2012 9.5 (7.7 to 11.4) .24 66.2 (56.3 to 76.0) .21 2013 11.7 (9.7 to 13.6) .77 66.6 (58.6 to 74.7) .19 2014 10.2 (8.4 to 11.9) .36 61.0 (52.9 to 69.1) .36 2015 13.5 (11.7 to 15.4) .62 56.7 (48.7 to 64.7) .54 2016 Q1-3 11.9 (10.1 to 13.8) .86 67.0 (59.7 to 74.2) .17 Northeast 14.3 (11.9 to 16.8) -- 73.2 (64.5 to 81.8) -- South 10.7 (9.5 to 11.8) .02 58.9 (52.3 to 65.4) .01 Midwest 11.4 (9.9 to 13.0) .06 68.3 (62.0 to 74.7) .37 West 11.6 (9.8 to 13.4) .09 53.5 (44.6 to 62.3) .002 No 10.7 (9.8 to 11.6) -- 63.1 (58.2 to 67.9) -- 90 d prior to overdose Yes 14.0 (11.9 to 16.1) .01 63.0 (57.1 to 68.9) .98 No 11.4 (10.5 to 12.3) -- 63.9 (59.6 to 68.3) -- 90 d prior to overdose Yes 11.8 (9.9 to 13.8) .70 61.0 (57.0 to 68.9) .49 No 11.5 (10.4 to 12.7) -- 65.8 (61.0 to 70.6) -- 90 d prior to overdose Yes 11.4 (10.0 to 12.9) .93 57.3 (49.4 to 65.3) .11 , 90 d prior No 10.8 (9.8 to 11.7) -- 60.5 (55.8 to 65.1) -- to overdose Yes 13.3 (11.5 to 15.1) .02 67.4 (61.4 to 73.4) .08 © 2020 Kilaru AS et al. JAMA Network Open. Index Opioid Overdoses by specific ICD-9 or ICD-10 Diagnosis Code, with number and frequency for each diagnosis code Diagnosis Code Definition Number Frequency (%) Poisoning by opium (unspecified) 965.00 1135 17.59 Poisoning by heroin 965.01 1209 18.74 Poisoning by methadone 965.02 139 2.15 Poisoning by other opiates and related narcotics 965.09 2243 34.77 Accidental poisoning by heroin E.850.0 4 0.06 Accidental poisoning by methadone E.850.1 2 0.03 Accidental poisoning by other opiates and related narcotics E.850.2 21 0.33 ICD-10 [Starting October 1 2015] Poisoning by opium, accidental (unintentional), initial T400X1A 28 0.43 encounter Poisoning by opium, intentional self-harm, initial encounter T400X2A 2 0.03 Poisoning by opium, undetermined, initial encounter T400X4A 3 0.05 Poisoning by heroin, accidental (unintentional), initial T401X1A 535 8.29 encounter Poisoning by heroin, intentional self-harm, initial encounter T401X2A 38 0.59 Poisoning by heroin, undetermined, initial encounter T401X4A 51 0.79 Poisoning by other opioids, accidental (unintentional), T402X1A 579 8.98 initial encounter Poisoning by other opioids, accidental (unintentional), T402X1D 2 0.03 subseequent encounter Poisoning by other opioids, accidental (unintentional), T402X1S 1 0.02 sequelae Poisoning by other opioids, intentional self-harm, initial T402X2A 160 2.48 enounter Poisoning by other opioids, undetermined, initial encounter T402X4A 62 0.96 Poisoning by methadone, accidental (unintentional), initial T403X1A 40 0.62 encounter Poisoning by methadone, intentional self-harm, initial T403X2A 6 0.09 encounter Poisoning by methadone, undetermined T403X4A 1 0.02 Poisoning by other synthetic narcotics, accidental T404X1A 119 1.84 (unintentional), initial encounter Poisoning by other synthetic narcotics, intentional self- T404X2A 51 0.79 harm, initial encounter Poisoning by other synthetic narcotics, undetermined, initial T404X4A 20 0.31 encounter TOTAL 6451 100.00 © 2020 Kilaru AS et al. JAMA Network Open. eTable 8 Patient cohort and unadjusted outcomes, stratified by overdose type and treatment status before overdose © 2020 Kilaru AS et al. JAMA Network Open. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png JAMA Network Open American Medical Association

Incidence of Treatment for Opioid Use Disorder Following Nonfatal Overdose in Commercially Insured Patients

Download PDF
27 pages

Loading next page...
 
/lp/american-medical-association/incidence-of-treatment-for-opioid-use-disorder-following-nonfatal-20aFI0fhTr

References (136)

Publisher
American Medical Association
Copyright
Copyright 2020 Kilaru AS et al. JAMA Network Open.
eISSN
2574-3805
DOI
10.1001/jamanetworkopen.2020.5852
Publisher site
See Article on Publisher Site

Abstract

Key Points Question How often do commercially IMPORTANCE Timely initiation and referral to treatment for patients with opioid use disorder seen insured patients obtain follow-up in the emergency department is associated with reduced mortality. It is not known how often treatment for opioid use disorder after a commercially insured adults obtain follow-up treatment after nonfatal opioid overdose. nonfatal opioid overdose? Findings In this cohort study of national OBJECTIVE To investigate the incidence of follow-up treatment following emergency department commercial insurance claims for 6451 discharge after nonfatal opioid overdose and patient characteristics associated with receipt of patients, 16.6% of patients obtained follow-up treatment. follow-up treatment after a nonfatal opioid overdose. Among those who had DESIGN, SETTING, AND PARTICIPANTS A retrospective cohort study was conducted using an not received treatment for opioid use administrative claims database for a large US commercial insurer, from October 1, 2011, to September disorder before the overdose, patients 30, 2016. Data analysis was performed from May 1, 2019, to September 26, 2019. Adult patients of older age, female sex, black race, and discharged from the emergency department after an index opioid overdose (no overdose in the Hispanic ethnicity were less likely to preceding 90 days) were included. Patients with cancer and without continuous insurance obtain follow-up. enrollment were excluded. Meaning Timely treatment for opioid MAIN OUTCOMES AND MEASURES The primary outcome was follow-up treatment in the 90 days use disorder following overdose appears following overdose, defined as a combined outcome of claims for treatment encounters or to be low among commercially insured medications for opioid use disorder (buprenorphine and naltrexone). Analysis was stratified by patients, with race/ethnicity, sex, and whether patients received treatment for opioid use disorder in the 90 days before the overdose. age disparities. Logistic regression models were used to identify patient characteristics associated with receipt of follow-up treatment. Marginal effects were used to report the average adjusted probability and Invited Commentary absolute risk differences (ARDs) in follow-up for different patient characteristics. Supplemental content RESULTS A total of 6451 patients were identified with nonfatal opioid overdose; the mean (SD) age Author affiliations and article information are was 45.0 (19.3) years, 3267 were women (50.6%), and 4676 patients (72.5%) reported their race as listed at the end of this article. non-Hispanic white. A total of 1069 patients (16.6%; 95% CI, 15.7%-17.5%) obtained follow-up treatment within 90 days after the overdose. In adjusted analysis of patients who did not receive treatment before the overdose, black patients were half as likely to obtain follow-up compared with non-Hispanic white patients (ARD, −5.9%; 95% CI, −8.6% to −3.6%). Women (ARD, −1.7%; 95% CI, −3.3% to −0.5%) and Hispanic patients (ARD, −3.5%; 95% CI, −6.1% to −0.9%) were also less likely to obtain follow-up. For each additional year of age, patients were 0.2% less likely to obtain follow-up (95% CI, −0.3% to −0.1%). CONCLUSIONS AND RELEVANCE Efforts to improve the low rate of timely follow-up treatment following opioid overdose may seek to address sex, race/ethnicity, and age disparities. JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 Open Access. This is an open access article distributed under the terms of the CC-BY License. JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 1/13 JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose Introduction Each year, the emergency department (ED) provides care for an increasing number of patients who 1-3 present with opioid overdose as well as medical complications of opioid use disorder (OUD). The 4-7 ED serves as an essential touchpoint for patients seeking care for withdrawal and addiction. Akey strategy in secondary prevention of opioid overdose deaths is the engagement of patients with OUD 8-11 in treatment following discharge. 12-14 However, few patients successfully transition to treatment following nonfatal overdose. In evidence from 2 states, less than 5% of Medicaid patients initiated treatment with medication for 13,14 opioid use disorder (MOUD) following overdose. For patients who are ready to engage in 4,9 treatment, care coordination can help to overcome barriers to access. Yet hospitals have few 5,8,15-18 incentives and capacity to provide resource-intensive care navigation after ED visits. 19,20 Patients have high risk of death in the days immediately following opioid overdose. The initiation of MOUD during or after emergency care is associated with improvements in a variety of patient outcomes, including all-cause mortality and engagement in outpatient treatment, and other 12,21-24 hospital-based interventions have been developed. As a consequence, policy makers have identified the transition of patients from emergency care to sustained treatment (termed warm 25-28 handoffs) as an urgent priority. In this study, we sought to examine the rate of follow-up treatment after discharge from the ED following overdose in a national population of commercially insured adults. Previous studies have 12-14,29 focused on single states, the Medicaid population, and MOUD treatment. To our knowledge, no previous studies have included the full scope of treatment services available to patients. We also sought to examine patient-level characteristics associated with timely receipt of follow-up care. Evidence suggests that significant treatment disparities on the basis of race, sex, and geography have emerged as the opioid epidemic has evolved, possibly owing to differences in health 30-37 insurance coverage. We hypothesized that these treatment disparities by race and sex would persist within a commercially insured population. Methods Data Sources, Study Population, and Outcomes We conducted a retrospective cohort study of adult patients who were discharged from the ED following treatment for opioid overdose between October 1, 2011, and September 30, 2016. We used 38,39 an administrative claims database, the Optum Clinformatics Data Mart (Optum). The Optum database includes all inpatient, ED, outpatient, and pharmacy claims from a large national health insurance company that enrolled between 15 million and 18 million unique patients each year during the study period. Data analysis was performed from May 1, 2019, to September 26, 2019. The institutional review board at the University of Pennsylvania determined that this study was exempt from review because data are deidentified. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cohort studies. Selection of Patient Cohort We identified ED encounters for opioid overdose in the study period for patients with commercial insurance coverage (eFigure 1 in the Supplement). To do so, we used previously validated International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) and International Statistical Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) 41-44 diagnosis codes before and after October 1, 2015, respectively (eTable 1 in the Supplement). We used Current Procedural Terminology codes to specifically identify ED encounters (eTable 1 in the Supplement). We excluded encounters for patients who did not have continuous insurance enrollment for 90 days before and after the date of the overdose, to provide a sufficient window to measure patient JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 2/13 JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose exposures and outcomes and exclude fatal overdoses. We excluded patients with age younger than 18 years. We then limited the cohort to encounters for an index opioid overdose, defined as an encounter for opioid overdose with no ED encounter or hospital admission for opioid overdose in the preceding 90 days. We excluded encounters resulting in inpatient hospital admission to obtain a cohort of patients stable for ED discharge and likely to not have disability or sequelae from the overdose. In addition, we excluded encounters for patients with diagnosis of cancer based on treatment claims 12,46,47 ICD-9-CM and ICD-10 diagnosis codes in the preceding 90 days (eTable 1 in the Supplement). Patients with pain related to active cancer diagnoses represent a separate population and may be prescribed high doses of prescription opioids. Of the remaining encounters, we included only the first index opioid overdose for any individual patient during the study period (eFigure 1 in the Supplement). Outcomes The primary outcome was whether the patient obtained follow-up treatment in the 90 days following the index opioid overdose. We defined follow-up treatment as the presence of either 1 pharmacy claim for MOUD or 1 medical claim for an outpatient or inpatient opioid treatment encounter. For pharmacy claims, we identified National Drug Codes for all formulations of 48-50 buprenorphine, buprenorphine with naloxone, or naltrexone (eTable 2 in the Supplement). Methadone maintenance therapy was not covered by insurance for this population during the study period and was not included in this study. Medical claims for treatment encounters had an ICD-9-CM or ICD-10-CM diagnosis code for opioid use disorder in any position (eTable 3 in the Supplement) and Current Procedural Terminology or Healthcare Common Procedure Coding System codes for a variety of services including outpatient clinic visits, psychiatric services, inpatient and outpatient behavioral health services, outpatient treatment programs, and case management (eTable 3 in the Supplement). Repeated ED or inpatient hospital encounters were not included as follow-up treatment. Supplemental analyses were performed for the purpose of hypothesis generation. These included secondary outcomes that were the receipt of MOUD independently from treatment encounters within 90 days of the index overdose. We also examined the number of days from the index overdose to follow-up treatment. To address the absence of mortality data, we determined the date of service for the last insurance claim for all patients in the cohort. We performed a sensitivity analysis excluding patients for whom there was no claim beyond the 90-day follow-up period. Although the absence of claims does not indicate death, we could not ensure survival to the end of the follow-up period for those patients. Covariates We examined patient-level characteristics as covariates that we hypothesized could be associated with access to follow-up treatment, including patient age, sex, and race/ethnicity. Optum uses data on race/ethnicity that is self-reported or derived from administrative data sources. We also included geographic location, according to 4 United States Census Regions (Northeast, South, Midwest, West). Year of the index overdose was included given the increasing overdose incidence over the study period. We examined the type of overdose (heroin or prescription opioid) based on diagnosis codes. Prescription opioid refers to medications available by prescription but does not mean that the patient received a prescription for the medication. We also included exposures to treatment for behavioral health conditions in the 90 days preceding the index overdose. We included the presence of claims for anxiety or depression based on ICD-9-CM or ICD-10-CM diagnosis codes (eTable 1 in the Supplement) due to potential association with overdose. We also included claims for prescription opioid medications and benzodiazepines in the 90 days preceding the index overdose using American Hospital Formulary Service Pharmacologic-Therapeutic Classification codes. In addition, we determined whether patients had JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 3/13 JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose pharmacy claims for MOUD or medical claims for treatment encounters in the 90 days preceding the index overdose. Statistical Analysis First, we described the patient cohort, stratified by overdose type. We used 2-sided χ tests and t tests to describe differences in the cohort between overdose type. Next, we summarized patient outcomes, stratified by overdose type and treatment for OUD in the 90 days preceding the overdose. We then used multivariable logistic regression models to examine the association between patient characteristics, as described in the first paragraph of the Covariates section, and the binary primary outcome. Given that patients were hypothesized to more likely access follow-up treatment if they had received recent treatment before the overdose, we stratified the analyses based on whether patients had received OUD treatment in the 90 days before the overdose. For ease of interpretation, we used predictive margins to report average adjusted probability and absolute risk 54,55 differences (ARDs), with 95% CIs. For categorical variables, ARD represents the difference in adjusted probability of follow-up treatment between patients with a given characteristic and the reference value. In addition to the primary analysis, we investigated potential interactions between race/ ethnicity and overdose type by including an interaction term in the logistic regression model. Also, we used multivariable logistic regression models to examine the association between patient characteristics and the secondary outcome of MOUD treatment alone. In addition, we used Kaplan- Meier failure analysis to examine days to receipt of follow-up treatment, stratified by overdose type. Data analysis was conducted from June 1, 2019, to September 1, 2019. Analyses were performed using Stata software, version 15.1 (StataCorp LP). Results The total cohort consisted of 6451 patients, of whom 1896 (29.4%) overdosed from heroin and 4555 (70.6%) overdosed from prescription opioids (Table 1). Further delineation of the type of opioid overdose is reported in eTable 7 in the Supplement. The mean (SD) age was 45.0 (19.3) years and there were 3267 (50.6%) women. A total of 4676 patients (72.5%) reported their race as non-Hispanic white, 601 patients (9.3%) reported their race as black, and 536 patients (8.3%) who reported Hispanic ethnicity. Only 682 patients (10.6%) received treatment for opioid use disorder in the 90 days preceding the overdose, including 320 (5.0%) with pharmacy claims for MOUD. Patients with heroin overdose significantly differed across all patient characteristics compared with those with prescription opioid overdose. Primary Analysis For all patients in the study cohort, 1069 individuals (16.6%; 95% CI, 15.7%-17.5%) obtained follow-up treatment in the 90 days following overdose (Figure 1;eTable8inthe Supplement). Among the 5769 patients who did not receive treatment for OUD in the 90 days before the overdose, 643 (11.1%; 95% CI, 10.3%-12.0%) obtained follow-up treatment. Among the 682 patients who received treatment before the overdose, 426 individuals (62.5%; 95% CI, 58.7%-66.1%) patients obtained follow-up. In the adjusted analysis for patients who did not receive treatment before the overdose, patients with prescription opioid overdose were less likely to obtain follow-up compared with heroin overdose (Table 2) (ARD, −8.8%; 95% CI, −11.2% to −6.5%). Compared with patients of non-Hispanic white race, black (ARD, −5.9%; 95% CI, −8.6% to −3.6%) and Hispanic (ARD, −3.5%; 95% CI, −6.1% to −0.9%) patients were less likely to obtain follow-up. Women were less likely to obtain follow-up than men (ARD, −1.7%; 95% CI, −3.3% to −0.5%). For each additional year of age, patients were 0.2% less likely to obtain follow-up (95% CI, −0.3% to −0.1%). However, patients with recent treatment JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 4/13 JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose for anxiety, including a treatment encounter for anxiety (ARD, 3.4%, 95% CI, 1.1%-5.8%) or prescription for a benzodiazepine (ARD, 2.8%; 95% CI, 0.7%-5.0%), were more likely to obtain follow-up. In this adjusted analysis, there was no statistically significant change with regard to the rate of patients obtaining follow-up treatment over the 5 years of the study (Figure 2). These associations were not present for patients who received treatment in the 90 days before overdose, apart from a decreased rate of follow-up for patients in the South (ARD, −15.0%; 95% CI, −25.9% to −4.1% and the West (ARD, −20.1%; 95% CI, −32.% to −7.6%), compared with the Northeast (eTable 4 in the Supplement). Supplemental Analyses In supplemental analyses, differences in the adjusted probability of follow-up rate persisted across overdose type for black patients compared with non-Hispanic white patients (Figure 3). Among patients who did not receive treatment before overdose, black patients were less likely to obtain follow-up treatment than non-Hispanic white patients whether the index overdose was due to heroin (ARD, −8.8%; 95% CI, −11.5% to −6.1%) or prescription opioids (ARD, −4.7%; 95% CI, −5.7% to −3.7%). For Hispanic patients compared with patients of non-Hispanic white race, the difference in Table 1. Characteristics of Patient Cohort, Stratified by Overdose Type No. (%) Overdose Characteristic All patients (n = 6451) Heroin (n = 1896) Prescription opioid (n = 4555) Age, mean (SD), y 45.0 (19.3) 31.0 (13.2) 50.8 (18.4) Sex Male 3184 (49.4) 1291 (68.1) 1893 (41.6) Female 3267 (50.6) 605 (31.9) 2662 (58.4) Race/ethnicity Non-Hispanic white 4676 (72.5) 1450 (76.5) 3226 (70.8) Black 601 (9.3) 148 (7.8) 453 (9.9) Hispanic 536 (8.3) 135 (7.1) 401 (8.8) Asian 78 (1.2) 9 (0.5) 69 (1.5) Unknown 560 (8.7) 154 (8.1) 406 (8.9) Year 2011, quarter 4 229 (3.5) 40 (2.1) 189 (4.1) 2012 1099 (17.1) 239 (12.6) 860 (18.9) 2013 1164 (18.1) 276 (14.6) 888 (19.5) 2014 1248 (19.3) 362 (19.1) 886 (19.5) 2015 1387 (21.5) 475 (25.1) 912 (20.0) 2016, quarters 1-3 1324 (20.5) 504 (26.6) 820 (18.0) Region Northeast 659 (10.2) 316 (16.7) 343 (7.5) South 2627 (40.7) 617 (32.5) 2010 (44.1) Midwest 1619 (25.1) 703 (37.1) 916 (20.1) West 1546 (24.0) 260 (13.7) 1286 (28.2) 90 d Before overdose Anxiety treatment 1625 (25.2) 403 (21.3) 1222 (26.8) Depression treatment 1416 (22.0) 322 (17.0) 1094 (24.0) Prescription opioid claim 3266 (50.6) 317 (16.7) 2949 (64.7) Benzodiazepine claim 2009 (31.1) 373 (19.7) 1636 (35.9) MOUD claim 320 (5.0) 201 (10.6) 119 (2.6) Abbreviations: MOUD, medication for opioid use Buprenorphine 278 (4.3) 168 (8.9) 110 (2.4) disorder; OUD, opioid use disorder. Naltrexone 42 (0.7) 33 (1.7) 9 (0.2) a 2 Two-sided t test and χ tests were performed; Treatment encounter for OUD 539 (8.4) 347 (18.3) 192 (4.2) P < .001 for all patient characteristics. JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 5/13 JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose adjusted follow-up rate was significant only for patients with prescription opioid overdose (ARD, −4.0%; 95% CI, −5.% to 2.8%). We investigated the secondary outcome of MOUD treatment alone. Among the 6131 patients who did not file an MOUD claim in the 90 days before the index overdose, 280 individuals (4.6%) had a claim for MOUD following the overdose. In adjusted analyses, patients who were older, women, black race, and experienced a prescription opioid overdose were less likely to obtain MOUD treatment, while patients with a prescription for a benzodiazepine or treatment encounters for OUD were more likely (eTable 5 in the Supplement). We examined the timing of follow-up treatment following the index overdose, with results of the Kaplan-Meier failure analysis shown in eFigure 2 in the Supplement. Among all 1069 patients who obtained follow-up treatment, 318 individuals (29.7%) did so in 7 or fewer days after the overdose. In addition, we performed a sensitivity analysis excluding 233 patients (3.6%) who did not have claims beyond the 90-day follow-up period, which demonstrated equivalent outcomes to the primary analysis (eTable 6 in the Supplement). Discussion We analyzed commercial insurance claims to determine how often patients obtained treatment for OUD in the 90 days following ED presentation for nonfatal opioid overdose. Most had not received OUD treatment immediately preceding the overdose. Among that group, we found that only 11.1% of patients obtained follow-up treatment through an encounter in the outpatient setting, inpatient treatment, or filled prescriptions for a buprenorphine or naltrexone. The few patients that recently received treatment had a higher incidence of follow-up treatment. Despite the increasing number of overdoses across the years of this study, there was no significant change in the proportion of patients receiving follow-up treatment. Given that patients with commercial insurance likely have a superior ability to access care compared with patients who have public insurance, this persistently low rate suggests an opportunity for improvement. Disparities in the receipt of follow-up treatment with regard to race/ethnicity, age, and age persisted within this cohort. In particular, black patients were half as likely to obtain treatment following overdose compared with non-Hispanic white patients. This disparity was present regardless of whether the overdose was due to heroin or prescription opioids. To our knowledge, these disparities in treatment following opioid overdose have not been previously documented. However, our findings are consistent with emerging evidence that there are disparities in Figure 1. Patient Outcomes Stratified by Overdose Type and Treatment Status Before Overdose A All patients B Heroin overdose C Prescription overdose 100 100 100 Both MOUD and treatment encounter, 90 d after index opioid overdose 80 80 80 Only MOUD claim, 90 d after index opioid overdose Only treatment encounter, 60 90 d after index opioid overdose 60 60 40 40 40 20 20 20 0 0 0 All Patients No preceding MOUD Preceding MOUD No preceding MOUD Preceding MOUD or treatment or treatment or treatment or treatment Data shown for status at 90 days before overdose for all patients (A), heroin overdose (B), and prescription opioid overdose (C). MOUD indicates medication for opioid use disorder. JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 6/13 Proportion of Patients, % Proportion of Patients, % Proportion of Patients, % JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose 30,32,33,36 buprenorphine treatment with regard to race/ethnicity and sex. Although this study cannot determine whether these disparities are associated with patient preferences, barriers to access, implicit or explicit bias, or other causes, it is important to better understand and account for these factors when designing systems that seek to improve engagement and equity in treatment. Previous studies have examined changes in treatment rates before and after opioid overdose 12-14 using data from individual states. These studies primarily focused on medication treatment, with only one study including a limited range of treatment encounters. Our study included a range of possible treatments, from outpatient clinic visits to inpatient residential treatment. In general, we found that fewer than half of patients who obtained follow-up treatment received medication. Treatment with opioid agonists has been associated with reduced risk of relapse by 50% compared Table 2. Adjusted Probability of Follow-up Treatment After Opioid Overdose, for Patients Not Treated Before Overdose b c Patient characteristics Average adjusted probability, % (95% CI) P value Overdose type Prescription opioid 8.3 (7.3- 9.2) [Reference] Heroin 17.1 (15.1-19.2) <.001 Age, at mean, y 9.9 (9.1-10.7) <.001 Sex Male 11.9 (10.9-13.0) [Reference] Female 10.1 (9.1-11.3) .04 Race/ethnicity Non-Hispanic white 12.1 (11.1-13.0) [Reference] Black 6.1 (4.0-8.3) <.001 Hispanic 8.5 (6.1-11.0) .009 Asian 10.2 (2.8-17.5) .62 Unknown 10.1 (7.4-12.8) .18 Year 2011, quarter 4 12.2 (7.9-16.6) [Reference] 2012 9.3 (7.6-11.3) .22 2013 11.5 (9.6-13.5) .75 2014 10.0 (8.3-11.7) .32 2015 12.9 (11.1-14.6) .82 2016, quarters 1-3 11.1 (9.5-13.0) .64 Region Northeast 14.0 (11.6-16.6) [Reference] Results are given for patients who did not receive South 10.4 (9.1-11.4) .01 treatment for 90 days before the index opioid Midwest 11.1 (9.7-12.7) .07 overdose, defined as either a pharmacy claim for West 11.0 (9.3-12.8) .06 medication for opioid use disorder or medical claim 90 d Before overdose for opioid use disorder treatment encounter. Anxiety treatment Estimated with logistic regression model using No 10.3 (9.4-11.2) [Reference] predictive margins. Average adjusted probability is the adjusted rate, holding covariates at their actual Yes 13.8 (11.7-15.8) .004 values, at which patients obtain follow-up treatment Depression treatment within 90 days after the index opioid overdose, No 10.9 (10.1-11.9) [Reference] defined as either a pharmacy claim for medication for Yes 11.6 (9.7-13.5) .64 opioid use disorder or medical claim for opioid use Prescription opioid claim disorder treatment encounter. No 11.0 (9.9-12.1) [Reference] P values are given for average marginal effects, which represent the difference in adjusted probability Yes 11.2 (9.8-12.7) .84 between a given characteristic and the Benzodiazepine claim reference group. No 10.3 (9.4-11.2) [Reference] Average adjusted probability for continuous variable Yes 13.2 (11.4-15.0) .009 (age) is given for the mean patient age (46.3 years). JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 7/13 JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose with behavioral treatment alone. Better understanding of current treatment and referral patterns 57,58 may help inform efforts to expand evidence-based practices. We hypothesized that the rate of follow-up treatment would be higher for patients with commercial insurance, given potentially greater resources and access to care. While we cannot directly compare across studies, the rate of OUD treatment in this cohort did not appear to be appreciably higher in this cohort than that described in other populations. Not all patients can be expected to engage in treatment after overdose. Higher rates of treatment engagement have been observed in experimental settings, often with screening of patients for substance use 22,59-62 disorder. While the optimal rate of follow-up treatment may be difficult to estimate, there is still need for widely implemented interventions that may help patients overcome the many pervasive 4,15 barriers to accessing care. We intentionally examined outcomes for a short time following the overdose. Recent evidence suggests that risk of death is high immediately following overdose, with nearly 5% of deaths occurring within 2 days of discharge from the ED. In a secondary analysis, only 30% of patients who obtained follow-up did so within 7 days. Patients may benefit from rapid linkage to treatment, Figure 2. Proportion of Index Opioid Overdoses by Quarter, Stratified by Overdose Type and Receipt of Follow-up Treatment Heroin overdose, follow-up treatment Heroin overdose, no follow-up treatment Prescription opioid overdose, follow-up treatment Prescription opioid overdose, no follow-up treatment 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 2011 2012 2013 2014 2015 2016 Quarter Figure 3. Average Adjusted Probability of Follow-up Treatment After Opioid Overdose, by Overdose Type and Race/Ethnicity Race/ethnicity Non-Hispanic white Hispanic Black Estimated from logistic regression model with interaction term for overdose type and race/ethnicity. Error bars denote 95% confidence intervals for average adjusted probability. Results shown only for patients who had not received treatment for opioid use disorder in the 90 days before the index opioid overdose. Race/ethnicity was self-reported or derived Heroin overdose Prescription opioid overdose from other administrative data sources. JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 8/13 Opioid overdoses, % Adjusted probability of follow-up treatment, % JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose potentially through recovery specialists who can provide navigation and harm reduction counseling 3-5 regardless of the client’s willingness to engage in treatment. Limitations This study has several limitations. First, we cannot account for patients who pay for OUD treatment out-of-pocket. Although treatment services, including MOUD, were covered by the insurer during the study period, some patients may have elected to pursue alternative options. Second, this study did not include patients who obtain methadone maintenance therapy. Methadone is an important treatment modality for many patients with opioid use disorder. However, methadone was not covered for this indication by the insurer during the study period. It is possible that patients in this cohort obtained methadone through self-pay or other mechanisms, although this rate cannot be 63-65 estimated from these data and is difficult to extrapolate from other sources. Third, these data do not specifically account for patient deaths in the days following the index overdose. However, additional analysis that only included patients known to have survived to the end of the follow-up period showed similar results. Fourth, the use of administrative claims data in this study limits our ability to ascertain the reasons that patients obtain or do not obtain follow-up treatment. It is not known whether patients do not receive appropriate referrals, lack treatment facilities in their communities, or may be unwilling to engage in treatment. A corollary limitation is that patients may have received prescriptions for MOUD but not filled those prescriptions. Fifth, this cohort likely includes patients who may not have OUD, which may explain differential rates in follow-up treatment for patients with heroin and prescription opioid overdose. Regardless, patients with accidental prescription opioid overdose also should obtain timely follow-up for reevaluation, medication adjustment, and discussion of the long-term risks associated with opioid use. Conclusions Engagement of patients into treatment following opioid overdose is necessary to prevent subsequent opioid overdose death and other harm. Among commercially insured patients who were not receiving active addiction treatment, only 11.1% received follow-up treatment after an overdose. We showed apparent disparities in treatment with regard to race/ethnicity (eg, black patients were half as likely to obtain follow-up compared with non-Hispanic white patients), sex, and age. Research is needed to better understand the mechanisms behind these disparities. As health professionals adopt evidence-based practices for initiating medications for treatment of OUD and linking patients to sustained treatment, payers and policy makers should implement strategies to overcome systemic barriers to ensure that patients are given the best opportunity to access timely treatment. These interventions must account for disparities to ensure expanded and equitable access to life-saving treatment following overdose. ARTICLE INFORMATION Accepted for Publication: March 22, 2020. Published: May 27, 2020. doi:10.1001/jamanetworkopen.2020.5852 Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Kilaru AS et al. JAMA Network Open. Corresponding Author: Austin S. Kilaru, MD, Center for Emergency Care Policy and Research, Perelman School of Medicine, Department of Emergency Medicine, University of Pennsylvania, 421 Guardian Dr, 1303 Blockley Hall, Philadelphia, PA 19104 (austin.kilaru@pennmedicine.upenn.edu). Author Affiliations: National Clinician Scholars Program, Corporal Michael J. Crescenz Veterans Affairs Medical Center, University of Pennsylvania, Philadelphia (Kilaru, Lowenstein, Khatri); Center for Emergency Care Policy and Research, Perelman School of Medicine, Department of Emergency Medicine, University of Pennsylvania, JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 9/13 JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose Philadelphia (Kilaru, Meisel, Perrone, Khatri, Delgado); Leonard Davis Institute of Health Economics, University of Pennsylvania, Philadelphia (Kilaru, Xiong, Lowenstein, Meisel, Perrone, Khatri, Delgado); Penn Injury Science Center, Philadelphia, Pennsylvania (Kilaru, Lowenstein, Meisel, Khatri, Delgado); Perelman School of Medicine, Department of Biostatistics and Epidemiology, University of Pennsylvania, Philadelphia (Mitra, Delgado). Author Contributions: Dr Kilaru 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: Kilaru, Lowenstein, Meisel, Perrone, Khatri, Delgado. Acquisition, analysis, or interpretation of data: Kilaru, Xiong, Meisel, Mitra, Delgado. Drafting of the manuscript: Kilaru, Xiong. Critical revision of the manuscript for important intellectual content: Kilaru, Lowenstein, Meisel, Perrone, Khatri, Mitra, Delgado. Statistical analysis: Kilaru, Xiong, Mitra. Obtained funding: Kilaru, Delgado. Administrative, technical, or material support: Meisel, Khatri, Delgado. Supervision: Kilaru, Meisel, Delgado. Conflict of Interest Disclosures: Dr Delgado reported an honorarium from United Health Group outside the submitted work. No other disclosures were reported. Funding/Support: This study was supported by a pilot grant from the Leonard Davis Institute of Health Economics at the University of Pennsylvania (Dr Kilaru). Role of the Funder/Sponsor: The Leonard Davis Institute of Health Economics at the University of Pennsylvania had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Disclaimer: The contents do not represent the views of the US Department of Veteran Affairs or the US government. REFERENCES 1. Vivolo-Kantor AM, Seth P, Gladden RM, et al. Vital signs: trends in emergency department visits for suspected opioid overdoses—United States, July 2016-September 2017. MMWR Morb Mortal Wkly Rep. 2018;67(9):279-285. doi:10.15585/mmwr.mm6709e1 2. Tedesco D, Asch SM, Curtin C, et al. Opioid abuse and poisoning: trends in inpatient and emergency department discharges. Health Aff (Millwood). 2017;36(10):1748-1753. doi:10.1377/hlthaff.2017.0260 3. Martin A, Mitchell A, Wakeman S, White B, Raja A. Emergency department treatment of opioid addiction: an opportunity to lead. Acad Emerg Med. 2018;25(5):601-604. doi:10.1111/acem.13367 4. Doran KM, Raja AS, Samuels EA. Opioid overdose protocols in the emergency department: are we asking the right questions? Ann Emerg Med. 2018;72(1):12-15. doi:10.1016/j.annemergmed.2018.05.024 5. Samuels EA, D’Onofrio G, Huntley K, et al. A quality framework for emergency department treatment of opioid use disorder. Ann Emerg Med. 2019;73(3):237-247. doi:10.1016/j.annemergmed.2018.08.439 6. Herring AA, Perrone J, Nelson LS. Managing opioid withdrawal in the emergency department with buprenorphine. Ann Emerg Med. 2019;73(5):481-487. doi:10.1016/j.annemergmed.2018.11.032 7. Larochelle MR, Bernstein R, Bernson D, et al. Touchpoints—opportunities to predict and prevent opioid overdose: a cohort study. Drug Alcohol Depend. 2019;204:107537. doi:10.1016/j.drugalcdep.2019.06.039 8. D’Onofrio G, McCormack RP, Hawk K. Emergency departments: a 24/7/365 option for combating the opioid crisis. N Engl J Med. 2018;379(26):2487-2490. doi:10.1056/NEJMp1811988 9. Houry DE, Haegerich TM, Vivolo-Kantor A. Opportunities for prevention and intervention of opioid overdose in the emergency department. Ann Emerg Med. 2018;71(6):688-690. doi:10.1016/j.annemergmed.2018.01.052 10. Winetsky D, Weinrieb RM, Perrone J. Expanding treatment opportunities for hospitalized patients with opioid use disorders. JHospMed. 2018;13(1):62-64. doi:10.12788/jhm.2861 11. Naeger S, Mutter R, Ali MM, Mark T, Hughey L. Post-discharge treatment engagement among patients with an opioid-use disorder. J Subst Abuse Treat. 2016;69:64-71. doi:10.1016/j.jsat.2016.07.004 12. Larochelle MR, Bernson D, Land T, et al. Medication for opioid use disorder after nonfatal opioid overdose and association with mortality: a cohort study. Ann Intern Med. 2018;169(3):137-145. doi:10.7326/M17-3107 13. Frazier W, Cochran G, Lo-Ciganic WH, et al. Medication-assisted treatment and opioid use before and after overdose in Pennsylvania Medicaid. JAMA. 2017;318(8):750-752. doi:10.1001/jama.2017.7818 JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 10/13 JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose 14. Koyawala N, Landis R, Barry CL, Stein BD, Saloner B. Changes in outpatient services and medication use following a non-fatal opioid overdose in the West Virginia Medicaid program. J Gen Intern Med. 2019;34(6): 789-791. doi:10.1007/s11606-018-4817-8 15. D’Onofrio G, Edelman EJ, Hawk KF, et al. Implementation facilitation to promote emergency department- initiated buprenorphine for opioid use disorder: protocol for a hybrid type III effectiveness-implementation study (Project ED HEALTH). Implement Sci. 2019;14(1):48. doi:10.1186/s13012-019-0891-5 16. Katz EB, Carrier ER, Umscheid CA, Pines JM. Comparative effectiveness of care coordination interventions in the emergency department: a systematic review. Ann Emerg Med. 2012;60(1):12-23.e1. doi:10.1016/j. annemergmed.2012.02.025 17. Carrier EYT, Holzwart RA. Coordination between Emergency and Primary Care Physicians. National Institute for Health Care Reform; February 2011. 18. Medford-Davis L, Marcozzi D, Agrawal S, Carr BG, Carrier E. Value-based approaches for emergency care in a new era. Ann Emerg Med. 2017;69(6):675-683. doi:10.1016/j.annemergmed.2016.10.031 19. Weiner SG, Baker O, Bernson D, Schuur JD. One-year mortality of patients after emergency department treatment for nonfatal opioid overdose. Ann Emerg Med. 2020;75(1):13-17. doi:10.1016/j.annemergmed.2019. 04.020 20. Olfson M, Crystal S, Wall M, Wang S, Liu SM, Blanco C. Causes of death after nonfatal opioid overdose. JAMA Psychiatry. 2018;75(8):820-827. doi:10.1001/jamapsychiatry.2018.1471 21. D’Onofrio G, O’Connor PG, Pantalon MV, et al. Emergency department–initiated buprenorphine/naloxone treatment for opioid dependence: a randomized clinical trial. JAMA. 2015;313(16):1636-1644. doi:10.1001/jama. 2015.3474 22. D’Onofrio G, Chawarski MC, O’Connor PG, et al. Emergency department–initiated buprenorphine for opioid dependence with continuation in primary care: outcomes during and after intervention. J Gen Intern Med.2017;32 (6):660-666. doi:10.1007/s11606-017-3993-2 23. Busch SH, Fiellin DA, Chawarski MC, et al. Cost-effectiveness of emergency department-initiated treatment for opioid dependence. Addiction. 2017;112(11):2002-2010. doi:10.1111/add.13900 24. Hu T, Snider-Adler M, Nijmeh L, Pyle A. Buprenorphine/naloxone induction in a Canadian emergency department with rapid access to community-based addictions providers. CJEM. 2019;21(4):492-498. doi:10.1017/ cem.2019.24 25. Kilaru AS, Perrone J, Kelley D, et al. Participation in a hospital incentive program for follow-up treatment for opioid use disorder. JAMA Netw Open. 2020;3(1):e1918511. doi:10.1001/jamanetworkopen.2019.18511 26. Pitt AL, Humphreys K, Brandeau ML. Modeling health benefits and harms of public policy responses to the US opioid epidemic. Am J Public Health. 2018;108(10):1394-1400. doi:10.2105/AJPH.2018.304590 27. Ahmed OM, Mao JA, Holt SR, et al. A scalable, automated warm handoff from the emergency department to community sites offering continued medication for opioid use disorder: lessons learned from the EMBED trial stakeholders. J Subst Abuse Treat. 2019;102:47-52. doi:10.1016/j.jsat.2019.05.006 28. Governor Baker Signs Second Major Piece of Legislation to Address Opioid Epidemic in Massachusetts. Published August 14, 2018. Accessed September 1, 2018. https://www.mass.gov/news/governor-baker-signs- second-major-piece-of-legislation-to-address-opioid-epidemic-in 29. Larochelle MR, Liebschutz JM, Zhang F, Ross-Degnan D, Wharam JF. Opioid prescribing after nonfatal overdose and association with repeated overdose: a cohort study. Ann Intern Med. 2016;164(1):1-9. doi:10.7326/ M15-0038 30. Lagisetty PA, Ross R, Bohnert A, Clay M, Maust DT. Buprenorphine treatment divide by race/ethnicity and payment. JAMA Psychiatry. 2019;76(9):979-981. doi:10.1001/jamapsychiatry.2019.0876 31. Allen B, Nolan ML, Kunins HV, Paone D. Racial differences in opioid overdose deaths in New York City, 2017. JAMA Intern Med. 2019;179(4):576-578. doi:10.1001/jamainternmed.2018.7700 32. Hadland SE, Wharam JF, Schuster MA, Zhang F, Samet JH, Larochelle MR. Trends in receipt of buprenorphine and naltrexone for opioid use disorder among adolescents and young adults, 2001-2014. JAMA Pediatr. 2017;171 (8):747-755. doi:10.1001/jamapediatrics.2017.0745 33. Shiels MS, Freedman ND, Thomas D, Berrington de Gonzalez A. Trends in US drug overdose deaths in non-Hispanic black, Hispanic, and non-Hispanic white persons, 2000-2015. Ann Intern Med. 2018;168(6): 453-455. doi:10.7326/M17-1812 34. Saloner B, Karthikeyan S. Changes in substance abuse treatment use among individuals with opioid use disorders in the United States, 2004-2013. JAMA. 2015;314(14):1515-1517. doi:10.1001/jama.2015.10345 JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 11/13 JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose 35. Santoro TN, Santoro JD. Racial bias in the US opioid epidemic: a review of the history of systemic bias and implications for care. Cureus. 2018;10(12):e3733. doi:10.7759/cureus.3733 36. Krawczyk N, Feder KA, Fingerhood MI, Saloner B. Racial and ethnic differences in opioid agonist treatment for opioid use disorder in a US national sample. Drug Alcohol Depend. 2017;178:512-518. doi:10.1016/j.drugalcdep. 2017.06.009 37. Haffajee RL, Lin LA, Bohnert ASB, Goldstick JE. Characteristics of US counties with high opioid overdose mortality and low capacity to deliver medications for opioid use disorder. JAMA Netw Open. 2019;2(6):e196373. doi:10.1001/jamanetworkopen.2019.6373 38. Optum Claims Data. Accessed May 1, 2019. https://www.optum.com/solutions/data-analytics/data/real-world- data-analytics-a-cpl/claims-data.html 39. Sanghavi DAA, Hane C, Bleicher P. Optum Opioid Data, Health Affairs Blog.pdf. Health Affairs Blog; 2017. 40. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP; STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Ann Intern Med. 2007;147(8):573-577. doi:10.7326/0003-4819-147-8- 200710160-00010 41. Safe States Injury Surveillance Workgroup (ISW7). Consensus recommendations for national and state poisoning surveillance. Published April 2012. Accessed May 1, 2019. https://cdn.ymaws.com/www.safestates.org/ resource/resmgr/imported/ISW7%20Full%20Report_3.pdf 42. Green CA, Perrin NA, Janoff SL, Campbell CI, Chilcoat HD, Coplan PM. Assessing the accuracy of opioid overdose and poisoning codes in diagnostic information from electronic health records, claims data, and death records. Pharmacoepidemiol Drug Saf. 2017;26(5):509-517. doi:10.1002/pds.4157 43. Reardon JM, Harmon KJ, Schult GC, Staton CA, Waller AE. Use of diagnosis codes for detection of clinically significant opioid poisoning in the emergency department: a retrospective analysis of a surveillance case definition. BMC Emerg Med. 2016;16:11. doi:10.1186/s12873-016-0075-4 44. Rowe C, Vittinghoff E, Santos GM, Behar E, Turner C, Coffin PO. Performance measures of diagnostic codes for detecting opioid overdose in the emergency department. Acad Emerg Med. 2017;24(4):475-483. doi:10.1111/ acem.13121 45. American Medical Association. CPT (Current Procedural Terminology). 2019. Accessed May 1, 2019. https:// www.ama-assn.org/amaone/cpt-current-procedural-terminology 46. Centers for Disease Control and Prevention. International Classification of Diseases, Tenth Revision, Clinical Modification. Accessed May 1, 2019. https://www.cdc.gov/nchs/icd/icd10cm.htm 47. Centers for Disease Control and Prevention. ICD-9-CM Addenda, Conversion Table, and Guidelines. Accessed May1,2019. https://www.cdc.gov/nchs/icd/icd9cm_addenda_guidelines.htm 48. Saloner B, Levin J, Chang HY, Jones C, Alexander GC. Changes in buprenorphine-naloxone and opioid pain reliever prescriptions after the Affordable Care Act Medicaid Expansion. JAMA Netw Open. 2018;1(4):e181588. doi:10.1001/jamanetworkopen.2018.1588 49. Centers for Disease Control and Prevention. Analyzing prescription data and morphine milligram equivalents (MME). Updated October 23, 2019. Accessed August 1, 2019. https://www.cdc.gov/drugoverdose/resources/data.html 50. Centers for Medicare and Medicaid Services. Chronic Conditions Data Warehouse—opioid use disorder. 2019. Accessed May 1, 2019. https://www2.ccwdata.org/web/guest/condition-categories 51. United States Census Bureau. Geography program. Accessed May 1, 2019. https://www.census.gov/programs- surveys/geography.html 52. National Institute on Drug Abuse. Overdose death rates. Accessed February 28, 2020. https://www.drugabuse. gov/related-topics/trends-statistics/overdose-death-rates 53. AHFS Clinical Drug Information. AHFS pharmacologic therapeutic classification. Accessed May 1, 2019. https:// www.ahfsdruginformation.com/ahfs-pharmacologic-therapeutic-classification/ 54. Norton EC, Dowd BE, Maciejewski ML. Odds ratios—current best practice and use. JAMA. 2018;320(1):84-85. doi:10.1001/jama.2018.6971 55. Norton EC, Dowd BE, Maciejewski ML. Marginal effects—quantifying the effect of changes in risk factors in logistic regression models. JAMA. 2019;321(13):1304-1305. doi:10.1001/jama.2019.1954 56. Clark RE, Baxter JD, Aweh G, O’Connell E, Fisher WH, Barton BA. Risk factors for relapse and higher costs among Medicaid members with opioid dependence or abuse: opioid agonists, comorbidities, and treatment history. J Subst Abuse Treat. 2015;57:75-80. doi:10.1016/j.jsat.2015.05.001 JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 12/13 JAMA Network Open | Substance Use and Addiction Incidence of Treatment for Opioid Use Disorder Following Overdose 57. Mojtabai R, Mauro C, Wall MM, Barry CL, Olfson M. Medication treatment for opioid use disorders in substance use treatment facilities. Health Aff (Millwood). 2019;38(1):14-23. doi:10.1377/hlthaff.2018.05162 58. Sharfstein J, Meisel ZF. Low-value treatment for opioid addiction: what is to be done? JAMA Forum. Published July 25, 2019. Accessed November 7, 2019. https://newsatjama.jama.com/2019/07/25/jama-forum-low-value- treatment-for-opioid-addiction-what-is-to-be-done/ 59. Hawk K, D’Onofrio G. Emergency department screening and interventions for substance use disorders. Addict Sci Clin Pract. 2018;13(1):18. doi:10.1186/s13722-018-0117-1 60. D’Onofrio G, Degutis LC. Integrating Project ASSERT: a screening, intervention, and referral to treatment program for unhealthy alcohol and drug use into an urban emergency department. Acad Emerg Med. 2010;17(8): 903-911. doi:10.1111/j.1553-2712.2010.00824.x 61. Bogenschutz MP, Donovan DM, Mandler RN, et al. Brief intervention for patients with problematic drug use presenting in emergency departments: a randomized clinical trial. JAMA Intern Med. 2014;174(11):1736-1745. doi: 10.1001/jamainternmed.2014.4052 62. Edwards FJ, Wicelinski R, Gallagher N, McKinzie A, White R, Domingos A. Treating opioid withdrawal with buprenorphine in a community hospital emergency department: an outreach program. Ann Emerg Med. 2020;75 (1):49-56. doi:10.1016/j.annemergmed.2019.08.420 63. Polsky D, Arsenault S, Azocar F. Private Coverage of methadone in outpatient treatment programs. Psychiatr Serv. 2020;71(3):303-306. doi:10.1176/appi.ps.201900373 64. Reif S, Creedon TB, Horgan CM, Stewart MT, Garnick DW. Commercial health plan coverage of selected treatments for opioid use disorders from 2003 to 2014. J Psychoactive Drugs. 2017;49(2):102-110. doi:10.1080/ 02791072.2017.1300360 65. Fullerton CA, Kim M, Thomas CP, et al. Medication-assisted treatment with methadone: assessing the evidence. Psychiatr Serv. 2014;65(2):146-157. doi:10.1176/appi.ps.201300235 SUPPLEMENT. eFigure 1. Flowchart for Selection of Patient Cohort eTable 1. ICD-9-CM, ICD-10, CPT, and AHFS Codes for Selection of Patient Cohort and Patient Characteristics eTable 2. National Drug Codes for Medications for Opioid Use Disorder eTable 3. CPT, HCPCS, ICD-9-CM, and ICD-10-CM Codes for Treatment Encounters eTable 4. Adjusted Probability of Follow-up Treatment After Opioid Overdose for Patients Treated Prior to Overdose eTable 5. Adjusted Probability of MOUD Treatment After Opioid Overdose, Stratified by Treatment Status Prior to Overdose eFigure 2. Kaplan-Meier Failure Curve for Days to First Follow up Treatment Following Index ED Overdose eTable 6. Adjusted Probability of Follow-up Treatment After Opioid Overdose, Excluding Patients Without Known Claims Beyond 90-Day Follow-up Period (Sensitivity Analysis to Address Potential Mortality During Follow-up Period) eTable 7. Index Opioid Overdoses by specific ICD-9 or ICD-10 Diagnosis Code, With Number and Frequency for Each Diagnosis Code eTable 8. Patient Cohort and Unadjusted Outcomes, Stratified by Overdose Type and Treatment Status Before Overdose JAMA Network Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 (Reprinted) May 27, 2020 13/13 Supplementary Online Content Kilaru AS, Xiong A, Lowenstein M, et al. Incidence of treatment for opioid use disorder following nonfatal overdose in commercially insured patients. JAMA Netw Open. 2020;3(5):e205852. doi:10.1001/jamanetworkopen.2020.5852 eFigure 1. Flowchart for Selection of Patient Cohort eTable 1. ICD-9-CM, ICD-10, CPT, and AHFS Codes for Selection of Patient Cohort and Patient Characteristics eTable 2. National Drug Codes for Medications for Opioid Use Disorder eTable 3. CPT, HCPCS, ICD-9-CM, and ICD-10-CM Codes for Treatment Encounters eTable 4. Adjusted Probability of Follow-up Treatment After Opioid Overdose for Patients Treated Prior to Overdose eTable 5. Adjusted Probability of MOUD Treatment After Opioid Overdose, Stratified by Treatment Status Prior to Overdose eFigure 2. Kaplan-Meier Failure Curve for Days to First Follow up Treatment Following Index ED Overdose eTable 6. Adjusted Probability of Follow-up Treatment After Opioid Overdose, Excluding Patients Without Known Claims Beyond 90-Day Follow-up Period (Sensitivity Analysis to Address Potential Mortality During Follow-up Period) eTable 7. Index Opioid Overdoses by specific ICD-9 or ICD-10 Diagnosis Code, With Number and Frequency for Each Diagnosis Code eTable 8. Patient Cohort and Unadjusted Outcomes, Stratified by Overdose Type and Treatment Status Before Overdose This supplementary material has been provided by the authors to give readers additional information about their work. © 2020 Kilaru AS et al. JAMA Network Open. 1. Flowchart for Selection of Patient Cohort © 2020 Kilaru AS et al. JAMA Network Open. ICD-9-CM, ICD-10, CPT, and AHFS Codes for Selection of Patient Cohort and Patient Characteristics a a ICD-9-CM Diagnosis Codes ICD-10 Diagnosis Codes CPT Codes AHFS Pharmacologic- Therapeutic Codes Opioid Overdose Heroin: Heroin: 965.01, E850.0 T40.1X Prescription: Prescription: 965.00, 965.02, 965.09, T40.0X, T40.2-4X, T40.6X E.850.1, E.850.2 Emergency 99281, Department 99282, Encounters 99283, 99284, 99285 Cancer (Malignant 140.X – 208.X C00X – C97X Neoplasm) Diagnosis 209.0 – 209.3 V10.X 293.84, 300.00, 300.01, 300.02, F06.4, F40.00, F40.01, F40.02, F40.10, Anxiety Diagnosis 300.09, 300.10, 300.20, 300.21, F40.11, F40.210, F40.218, F40.220, 300.22, 300.23, 300.29, 300.3, F40.228, F40.230, F40.231, F40.232, 300.5, 300.89, 300.9, 308.0, F40.233, F40.240, F40.241, F40.242, 308.1, 308.2, 308.3, 308.4, 308.9, F40.243, F40.248, F40.290, F40.291, 309.81, 313.0, 313.1, 313.21, F40.298, F40.8, F40.9, F41.0, F41.1, 313.22, 313.3, 313.82, 313.83 F41.3, F41.8, F41.9, F42, F42.2, F42.3, F42.4, F42.8, F42.9, F43.0, F43.10, F43.11, F43.12, F44.9, F45.8, F48.8, F48.9, F93.8, F99, R45.2, R45.5, R45.6, R45.7 296.20, 296.21, 296.22, 296.23, F32.0, F32.1, F32.2, F32.3, F32.4, Depression Diagnosis 296.24, 296.25, 296.26, 296.30, F32.5, F32.89, F32.9, F33.0, F33.1, 296.31, 296.32, 296.33, 296.34, F33.2, F33.3, F33.8, F33.40, F33.41, 296.35, 296.36, 300.4, 311, V79.0 F33.42, F33.9, F34.1 Prescription Opioid 280808 Full Agonist Benzodiazepine 281208, 282408 © 2020 Kilaru AS et al. JAMA Network Open. ICD-9-CM diagnosis codes are used for any claim prior to October 1 2015. ICD-10 diagnosis codes are used for claims on that date or after. Benign neoplasms or neoplasms of uncertain behavior were excluded © 2020 Kilaru AS et al. JAMA Network Open. . National Drug Codes for Medications for Opioid Use Disorder National Drug Codes (NDC) 54017613, 54017713, 54018813, 54018913, 74201201, 74201232, 93360021, 93360040, 93360121, Buprenorphine and 93360140, 93360221, 93360240, 93360321, 93360340, 93537856, 93537956, 93572056, 93572156, Buprenorphine- 149075701, 228315303, 228315403, 228315473, 228315503, 228315567, 228315573, 228315603, Naloxone 378092393, 378092493, 406192303, 406192403, 406802003, 409201203, 409201232, 490005100, 490005130, 490005160, 490005190, 12496010001, 12496010002, 12496010005, 12496030001, 12496030002, 12496030005, 12496075701, 12496075705, 12496120201, 12496120203, 12496120401, 12496120403, 12496120801, 12496120803, 12496121201, 12496121203, 12496127802, 12496128302, 12496130602, 12496131002, 16590066605, 16590066630, 16590066705, 16590066730, 16590066790, 21695051510, 23490927003, 23490927006, 23490927009, 35356000407, 35356000430, 35356055530, 35356055630, 35356060504, 35356060704, 38779088800, 38779088801, 38779088803, 38779088805, 38779088806, 38779088809, 40042001001, 42023017901, 42023017905, 42291017430, 42291017530, 42858035340, 42858049340, 42858050103, 42858050203, 42858058640, 42858075040, 42858083940, 43063018407, 43063018430, 43063066706, 43063075306, 49452129201, 49452129202, 49452129203, 49452825301, 49452825302, 49452825303, 49999039507, 49999039515, 49999039530, 49999063830, 49999063930, 50268014411, 50268014415, 50268014511, 50268014515, 50383028793, 50383029493, 50383092493, 50383093093, 51552076501, 51552076502, 51552076505, 51552076506, 51552076509, 51552076510, 51552076550, 51927101200, 52959030430, 52959074930, 53217013830, 54123011430, 54123090730, 54123091430, 54123092930, 54123095730, 54123098630, 54569141600, 54569141601, 54569549600, 54569573900, 54569573901, 54569573902, 54569632500, 54569632600, 54569639900, 54569640800, 54569657800, 54868570700, 54868570701, 54868570702, 54868570703, 54868570704, 54868575000, 55045378403, 55390010010, 55700014730, 55700018430, 55700030230, 55700030330, 55700057904, 58284010014, 59011075004, 59011075104, 59011075204, 59011075704, 59011075804, 59385001201, 59385001230, 59385001401, 59385001430, 59385001601, 59385001630, 59385002101, 59385002160, 59385002201, 59385002260, 59385002301, 59385002360, 59385002401, 59385002460, 59385002501, 59385002560, 59385002601, 59385002660, 59385002701, 59385002760, 60429058611, 60429058630, 60429058633, 60429058711, 60429058730, 60429058733, 62756045983, 62756046083, 62756096983, 62756097083, 62991158301, 62991158302, 62991158303, 62991158304, 62991158306, 62991158307, 62991158308, 63275992201, 63275992202, 63275992203, 63275992204, 63275992205, 63275992207, 63370090506, 63370090509, 63370090510, 63370090515, 63459030042, 63481016101, 63481016160, 63481020701, 63481020760, 63481034801, 63481034860, 63481051901, 63481051960, 63481068501, 63481068560, 63481082001, 63481082060, 63481095201, 63481095260, 63629402801, 63629403401, 63629403402, 63629403403, 63629409201, 63874108403, 63874108503, 63874117303, 65162041503, 65162041603, 65757030001, 65757030202, 66336001630, 68071138003, 68071151003, 68258299103, 68258299903, 68308020230, 68308020830, 35356060604, 53217024630, 55887031204, 55887031215, 63874117403, 66336001530, 406192309, 406192409, 406800503, 50090157100, 55700056804, 60846097003, 60846097103, 62175045232, 62175045832, 62756045964, 62756046064, 62756096964, 62756097064, 63629409202, 63629507401, 63629712501, 63629712502, 63629712503, 63629712504, 63629712505, 63629712506, 63629712507, 63629712601, 63629712602, 63629712603, 63629712604, 63629712605, 63629712606, 63629712607, 63629712608, 63629727001, 63629727002, 64725093003, 64725093004, 64725192403, 64725192404, 65162041509, 65162041609, 71335035301, 71335035302, 71335035303, 71335035304, 71335035305, 71335035306, 71335035307, 76519117000, 76519117001, 76519117002, 76519117003, © 2020 Kilaru AS et al. JAMA Network Open. 43063059115, 47335032683, 47335032688, 50436010501, 51224020630, 51224020650, Naltrexone 51285027501, 51285027502, 52152010502, 52152010504, 52152010530, 54868557400, 63459030042, 65694010003, 65694010010, 65757030001, 65757030202, 68084029111, 68084029121, 68094085362, 68115068030, 56001122, 56001130, 56001170, 56007950, 56008050, 185003901, 185003930, 406009201, 406009203, 406117001, 406117003, 555090201, 555090202, 16729008101, 16729008110, 42291063230 © 2020 Kilaru AS et al. JAMA Network Open. CPT, HCPCS, ICD-9-CM, and ICD-10-CM Codes for Treatment Encounters Current Procedure Healthcare Common ICD-9-CM Diagnosis ICD-10-CM Diagnosis Codes Terminology (CPT) Codes Procedure Coding System Codes (claim on or after October 1 (HCPCS) Codes (claim prior to October 1 2015) 2015) Office or Outpatient Visit: Drug, Alcohol, and Behavioral 304.00-304.03 Opioid use, abuse, 99201-99205 Health Services (Outpatient 305.50-305.53 dependence: 99211-99215 and Inpatient): 304.70-304.73 F1190, F11920-F11922, H0001, H0002 965.00-965.02, 965.09 F11929 Psychiatric Diagnosis: H0004-H0019 E85.00-E85.02 F1193, F1194, F11950- 90791-90792 H0031-H0040 E93.50-E93.51 F11951, F11959, F11981, G0396-G0397 F11982, F11988 Psychotherapy Services: F1199, F1110, F11120- 90832-90839 Mental Health Services NOS: F11122, 90853 H0046-H0047 F1129, F1114, F11150- 90863 F11151, F11159, F11181, 90875-90876 Halfway House / Treatment F11182, F11188, F1119, 90801-90815 Program: F1120, F11220-F11222, 90824 H2034-H2036 F11229, F1123, F1124, 90862 F11250, F11251, F11259, Clinic Visit / Case F11281, F11282, F11288, Screening, Brief Intervention, Management: F1129 and Referral to Treatment T1015-T1017 (SBIRT): T1001 Poisoning: 99408-99409 T400X1*, T400X2*, MOUD Codes: T400X3*,T400X4* J0571-J0575 T401X1*, T401X2*, T401X3*, J1230, J2315 T401X4* T403X1*, T403X2*, T403X3*, Excluded T403X4* Methadone maintenance T402X1*, T402X2*, T402X3*, therapy: T402X4* H0020 T404X1*, T404X2*, T404X3*, T404X4* © 2020 Kilaru AS et al. JAMA Network Open. T40601*, T40602*, T40603*, T40604* T40605*, T40691*, T40692*, T40693*, T40694*, T40695*, T403X5* * Position can include A, D, or © 2020 Kilaru AS et al. JAMA Network Open. Adjusted probability of follow-up treatment after opioid overdose for patients treated prior to overdose Average Adjusted Prediction (95% CI), % Overdose Type Prescription 64.2 (57.5 to 70.7) -- Heroin 61.3 (56.0 to 66.5) .55 Age (years), mean (SD) 62.5 (59.0 to 66.1) .21 Sex Male 62.2 (57.5 to 70.0) -- Female 62.7 (57.4 to 68.2) .89 Race/Ethnicity White 62.8 (58.7 to 66.9) -- Black 75.9 (63.4 to 88.4) .06 Hispanic 61.3 (47.3 to 75.3) .84 59.6 (13.8 to Asian .89 105.3) Unknown 51.7 (39.1 to 64.1) .10 Year 2011 Q4 48.3 (22.2 to 74.5) -- 2012 66.0 (56.1 to 75.8) .22 2013 66.9 (58.9 to 75.8) .19 2014 60.9 (52.8 to 69.0) .37 2015 54.8 (46.9 to 62.7) .65 2016 Q1-3 66.4 (59.2 to 73.6) .19 Region Northeast 73.3 (64.7 to 82.0) South 58.3 (51.9 to 64.1) .007 Midwest 67.3 (61.0 to 73.6) .26 West 53.2 (44.4 to 62.0) .002 Anxiety treatment, No 62.3 (57.6 to 67.1) 90 d prior to overdose Yes 62.6 (56.8 to 68.5) .94 Depression treatment, No 63.3 (59.0 to 67.6) 90 d prior to overdose Yes 60.7 (54.1 to 67.2) .53 Prescription opioid claim, No 65.1 (60.4 to 69.8) 90 d prior to overdose Yes 56.9 (48.9 to 64.8) .11 Benzodiazepine claim, 90 No 60.1 (55.4 to 64.7) d prior to overdose Yes 66.7 (60.7 to 72.6) .10 © 2020 Kilaru AS et al. JAMA Network Open. Adjusted probability of MOUD treatment after opioid overdose, stratified by treatment status prior to overdose n = 6131 n = 320 P P (95% CI), % (95% CI), % Prescription 3.2 (2.6 to 3.8) -- 55.8 (45.6 to 66.0) -- Heroin 6.8 (5.6 to 8.1) < .001 53.5 (46.2 to 60.9) .75 A at mean 3.6 (3.0 to 4.1) < .001 54.4 (49.2 to 59.6) .22 Male 5.1 (4.5 to 5.7) -- 54.6 (47.9 to 61.3) -- Female 3.9 (3.2 to 4.6) .03 54.0 (45.0 to 62.9) .92 White 4.8 (4.3 to 5.3) -- 53.2 (47.1 to 59.3) -- Black 3.0 (1.6 to 4.4) .03 71.8 (50.8 to 92.7) .10 Hispanic 4.1 (2.3 to 5.9) .47 70.4 (50.8 to 90.0) .10 Asian 5.9 (-.19 to 12.1) .71 62.0 (2.6 to 126.7) .79 Unknown 4.3 (2.6 to 6.0) .60 40.5 (20.7 to 60.1) .23 2011 Q4 6.0 (2.9 to 9.2) -- 27.3 (-3.9 to 58.6) -- 2012 4.2 (3.0 to 5.5) .29 48.8 (34.6 to 63.0) .22 2013 6.4 (4.9 to 7.8) .85 61.7 (50.4 to 72.9) .05 2014 3.4 (2.4 to 4.4) .12 58.1 (45.1 to 71.0) .07 2015 4.4 (3.4 to 5.5) .35 50.4 (37.8 to 63.1) .18 2016 Q1-3 4.4 (3.3 to 5.5) .33 54.2 (42.6 to 65.9) .11 Northeast 4.4 (3.0 to 5.9) -- 55.3 (40.0 to 70.7) -- South 5.1 (4.2 to 6.0) .46 51.9 (43.4 to 60.4) .70 Midwest 3.9 (3.0 to 4.8) .52 58.7 (48.7 to 68.8) .72 West 4.8 (3.6 to 5.9) .75 52.6 (40.0 to 65.4) .79 No 4.3 (3.7 to 4.9) -- 54.5 (47.3 to 61.7) -- 90 d prior to overdose Yes 5.4 (4.2 to 6.6) .13 54.1 (45.2 to 63.0) .95 No 4.4 (3.8 to 5.0) -- 59.0 (52.7 to 65.3) -- 90 d prior to overdose Yes 5.0 (3.8 to 6.3) .39 42.9 (32.5 to 53.2) .01 No 4.3 (3.6 to 4.9) -- 57.0 (50.4 to 63.4) -- 90 d prior to overdose Yes 5.1 (4.0 to 6.2) .22 47.3 (35.2 to 59.3) .20 , 90 d prior No 4.0 (3.4 to 4.6) -- 53.9 (46.8 to 61.0) -- to overdose Yes 6.2 (2.9 to 7.5) .003 55.0 (46.3 to 63.8) .86 No 4.0 (3.5 to 4.5) -- 50.9 (42.6 to 59.1) -- d prior to overdose Yes 10.1 (7.5 to 12.7) < .001 57.2 (50.0 to 64.5) .27 © 2020 Kilaru AS et al. JAMA Network Open. Kaplan-Meier Failure Curve for Days to First Follow Up Treatment Following Index ED Overdose Follow up treatment includes claim for OUD treatment encounter or pharmacy claim for MOUD © 2020 Kilaru AS et al. JAMA Network Open. Adjusted probability of follow-up treatment after opioid overdose, excluding patients without known claims beyond 90-day follow-up period (sensitivity analysis to address potential mortality during follow-up period). n = 6131 n = 320 P P (95% CI), % (95% CI), % Prescription 8.5 (7.5 to 9.4) -- 64.2 (57.5 to 70.1) -- Heroin 18.0 (15.8 to 20.2) < .001 62.2 (56.8 to 67.6) .69 A mean (SD) 10.1 (9.3 to 11.0) < .001 63.1 (59.5 to 66.7) .22 Male 12.4 (11.3 to 13.7) -- 62.5 (57.7 to 67.3) -- Female 10.4 (9.2 to 11.5) .01 63.7 (58.2 to 69.1) .76 White 12.5 (11.6 to 13.4) -- 63.3 (59.2 to 67.4) -- Black 6.3 (4.1 to 8.6) < 0.001 75.8 (63.2 to 88.4) .06 Hispanic 9.1 (6.8 to 11.4) .02 60.8 (46.6 to 74.9) .73 Asian 11.1 (3.2 to 19.1) .75 60.0 (14.4 to 105.6) .89 Unknown 10.3 (7.6 to 13.1) .15 53.0 (40.1 to 65.9) .14 2011 Q4 12.4 (8.0 to 16.7) -- 48.2 (22.0 to 74.2) -- 2012 9.5 (7.7 to 11.4) .24 66.2 (56.3 to 76.0) .21 2013 11.7 (9.7 to 13.6) .77 66.6 (58.6 to 74.7) .19 2014 10.2 (8.4 to 11.9) .36 61.0 (52.9 to 69.1) .36 2015 13.5 (11.7 to 15.4) .62 56.7 (48.7 to 64.7) .54 2016 Q1-3 11.9 (10.1 to 13.8) .86 67.0 (59.7 to 74.2) .17 Northeast 14.3 (11.9 to 16.8) -- 73.2 (64.5 to 81.8) -- South 10.7 (9.5 to 11.8) .02 58.9 (52.3 to 65.4) .01 Midwest 11.4 (9.9 to 13.0) .06 68.3 (62.0 to 74.7) .37 West 11.6 (9.8 to 13.4) .09 53.5 (44.6 to 62.3) .002 No 10.7 (9.8 to 11.6) -- 63.1 (58.2 to 67.9) -- 90 d prior to overdose Yes 14.0 (11.9 to 16.1) .01 63.0 (57.1 to 68.9) .98 No 11.4 (10.5 to 12.3) -- 63.9 (59.6 to 68.3) -- 90 d prior to overdose Yes 11.8 (9.9 to 13.8) .70 61.0 (57.0 to 68.9) .49 No 11.5 (10.4 to 12.7) -- 65.8 (61.0 to 70.6) -- 90 d prior to overdose Yes 11.4 (10.0 to 12.9) .93 57.3 (49.4 to 65.3) .11 , 90 d prior No 10.8 (9.8 to 11.7) -- 60.5 (55.8 to 65.1) -- to overdose Yes 13.3 (11.5 to 15.1) .02 67.4 (61.4 to 73.4) .08 © 2020 Kilaru AS et al. JAMA Network Open. Index Opioid Overdoses by specific ICD-9 or ICD-10 Diagnosis Code, with number and frequency for each diagnosis code Diagnosis Code Definition Number Frequency (%) Poisoning by opium (unspecified) 965.00 1135 17.59 Poisoning by heroin 965.01 1209 18.74 Poisoning by methadone 965.02 139 2.15 Poisoning by other opiates and related narcotics 965.09 2243 34.77 Accidental poisoning by heroin E.850.0 4 0.06 Accidental poisoning by methadone E.850.1 2 0.03 Accidental poisoning by other opiates and related narcotics E.850.2 21 0.33 ICD-10 [Starting October 1 2015] Poisoning by opium, accidental (unintentional), initial T400X1A 28 0.43 encounter Poisoning by opium, intentional self-harm, initial encounter T400X2A 2 0.03 Poisoning by opium, undetermined, initial encounter T400X4A 3 0.05 Poisoning by heroin, accidental (unintentional), initial T401X1A 535 8.29 encounter Poisoning by heroin, intentional self-harm, initial encounter T401X2A 38 0.59 Poisoning by heroin, undetermined, initial encounter T401X4A 51 0.79 Poisoning by other opioids, accidental (unintentional), T402X1A 579 8.98 initial encounter Poisoning by other opioids, accidental (unintentional), T402X1D 2 0.03 subseequent encounter Poisoning by other opioids, accidental (unintentional), T402X1S 1 0.02 sequelae Poisoning by other opioids, intentional self-harm, initial T402X2A 160 2.48 enounter Poisoning by other opioids, undetermined, initial encounter T402X4A 62 0.96 Poisoning by methadone, accidental (unintentional), initial T403X1A 40 0.62 encounter Poisoning by methadone, intentional self-harm, initial T403X2A 6 0.09 encounter Poisoning by methadone, undetermined T403X4A 1 0.02 Poisoning by other synthetic narcotics, accidental T404X1A 119 1.84 (unintentional), initial encounter Poisoning by other synthetic narcotics, intentional self- T404X2A 51 0.79 harm, initial encounter Poisoning by other synthetic narcotics, undetermined, initial T404X4A 20 0.31 encounter TOTAL 6451 100.00 © 2020 Kilaru AS et al. JAMA Network Open. eTable 8 Patient cohort and unadjusted outcomes, stratified by overdose type and treatment status before overdose © 2020 Kilaru AS et al. JAMA Network Open.

Journal

JAMA Network Open – American Medical Association

Published: May 27, 2020

There are no references for this article.