TY - JOUR
AU - Stroebel, Robert J.
AB - Abstract Background Screening mammography is recommended for early detection of breast cancer but screening rates remain suboptimal. Methods A primary care portal for a large academic primary practice was developed for all preventive services. Another Web-based system (PRECARES [PREventive CAre REminder System]) was developed for appointment secretaries to manage proactive breast cancer screening. Female patients aged 40 to 75 years were randomly assigned to a control group (usual care) and an intervention group. For the intervention group, 2 monthly letters inviting patients to undergo mammography were sent starting 3 months before they were due for annual screening, followed by a telephone call to nonresponding patients. A subgroup of women employees was further randomized to receive a reminder by either US mail or e-mail. Results Of the total eligible population of 6665 women identified as having consented to participate in research, 3339 were randomly assigned to the control group and 3326 to the intervention group. The screening rate for annual mammography was 64.3% for the intervention group and 55.3% for the control group (P <.001). There were no significant differences between the 2 groups for any of the other adult preventive services. For the employee subgroup, the screening rate was 57.5% for the control group, 68.1% for the US mail group, and 72.2% for the e-mail group (intervention vs control, P <.001; e-mail vs US mail; P = .24). Conclusion The breast cancer screening rate improved significantly with the practice redesign of having appointment secretaries proactively manage breast cancer screening needs. Breast cancer is the most common nondermatologic cancer in women in the United States. In 2005, an estimated 211 240 new cases were diagnosed and 40 410 women died of the disease.1 The US Preventive Services Task Force recommends screening mammography with or without a clinical breast examination every 1 or 2 years for women 40 years and older.2 Early detection of breast cancer with screening programs has been reported to decrease mortality from breast cancer by as much as 32%.3,4 Despite the availability of evidenced-based guidelines for screening, a large number of women do not undergo screening mammography. A recent study5 reported that only 47% of eligible Medicare patients underwent screening mammography in the preceding 2 years, and other studies6-10 have reported similar poor performance for this preventive service. Rankings of the performance of this quality indicator show large variation across the United States.11 The delivery rate of mammography is higher in patients of large practice groups than for those of smaller groups, especially if information technology is available to provide physicians reminders at the point of care.5,11-13 Many women depend on an annual physical examination or other visit to a primary care physician for breast cancer screening. In an era of increasing health care costs, it would be beneficial for women to undergo mammography without requiring an otherwise unnecessary office visit. Increasingly, primary care practices are being measured on rates of preventive service delivery to their patients. One aspect of practice consistently designated for measurement is the delivery of preventive services. The challenge for primary care providers is to develop cost-effective processes for delivering mammography and other preventive services to patients in the context of a busy practice.14 In the past, primary care providers attributed the inability to provide recommended preventive services to lack of time.15 It has been reported that primary care physicians would be required to devote most of their workday to address the recommended preventive services for their patients if they were to provide all the preventive services recommended by the US Preventive Services Task Force.16 Using other allied health members of the practice team has been recommended to overcome this barrier and help the primary care physician enhance delivery of evidence-based preventive services.17,18 The primary purposes of our study were as follows: to ascertain the feasibility of developing a Web-based information system used by allied health staff, mainly appointment secretaries, to manage mammography needs for a large patient population of a primary care practice, and to determine the effectiveness of this intervention in improving mammography rates. Secondary goals were to determine the effect of a proactive patient reminder system for scheduling mammography on the utilization rates of annual physical examinations in this population of patients and to compare the effectiveness of reminders sent by e-mail and US mail. Methods Practice setting and patient population Mayo Clinic Rochester is a large multispecialty group practice in Rochester, Minn. The Division of Primary Care Internal Medicine (PCIM) consists of 38 internists who provide care for 32 000 adult patients from the population of approximately 120 000 in Rochester. Of these primary care patients, 7183 are women between the ages of 40 and 75 years. Study design The study was a randomized controlled trial. All women between the ages of 40 and 75 years who were patients of PCIM were eligible for the study, and all of them who were due for annual mammography in the next 3 months were identified from the Primary Care Physician Portal. These eligible patients were randomly assigned to either the control group who received usual care or the intervention group who, in addition, received a personalized letter via US mail from the patient's physician indicating the need for screening mammography, together with a brochure that explained all adult preventive services. The letter advised the patient to call the PCIM appointment desk and schedule an appointment for mammography. Women in the intervention group who were Mayo Clinic employees were further assigned randomly to receive a reminder by either US mail or e-mail through their work e-mail account sent on behalf of their primary provider. Patients who did not respond to the initial reminder received a second e-mail or letter 1 month after the first reminder. Those who did not respond to the second reminder received a scripted telephone call from an appointment secretary at PCIM 1 month after the second reminder. Other than the letter or telephone call reminders, there was no difference in the care provided to the patients in the 2 groups. The study was approved by the institutional review board of the Mayo Foundation. All patients in the intervention group were sent a detailed research authorization form used by our institution and were requested to complete and return it to the study coordinator. To comply with the recent regulations, a Health Insurance Portability and Accountability Act form was similarly sent to patients. Primary care physician portal The Primary Care Physician Portal is a Web-based information system at Mayo Clinic Rochester that consolidates data from institutional operational, clinical, and administrative data sources to assist primary care physicians in the management of their assigned patient population. Among many other functions, this system defines patient eligibility and completion for preventive services including mammography. Eligible population All of the 7183 eligible patients were randomly assigned to the intervention group or the control group (usual care). From the eligible patient population, 6665 women were identified to have consented to participate in this specific study or had documented authorization allowing access to their medical records for general research. Of these women, 3339 composed the control group and 3326 composed the intervention group. Two hundred eighty-six women in the control group and 232 in the intervention group either declined to participate in the study or their records indicated general denial for research authorization. Of the 3326 women in the intervention group, 2927 received reminders by US mail and 399 (employee patients) by e-mail. Proactive web-based intervention process A Web-based system (PRECARES [PREventive CAre REminder System]) was developed for appointment secretaries to manage proactive mammography for the patients in the intervention group. Each month, a list of patients due to undergo mammography in the next 3 months was retrieved from this application by an appointment secretary at PCIM. The secretary assessed the appointment system to determine whether a mammography appointment was already scheduled. If mammography was appropriate for the patient, the appointment secretary sent a letter inviting the patient to call to schedule mammography. When the patient called for the appointment, the appointment secretary scheduled the test and recorded the response in PRECARES. If the patient refused mammography or indicated that mammography had been performed elsewhere, this also was recorded. Patients who were previously scheduled for or had refused mammography or who had undergone screening elsewhere were excluded from reminder lists for the next 12 months. Two appointment secretaries from the existing staff were trained to manage the patient population for mammography; a 0.2 full-time equivalent was allocated to manage the practice of 38 physicians for this purpose. No additional staff were recruited. Usual care for breast cancer screening There are many different avenues for patients in the control group (usual care) to undergo mammography. Many undergo mammography as part of an office visit for an annual examination or any other visit. In our practice, every patient who comes for any appointment is assessed for the need for all preventive services. Some patients also call to request breast cancer screening, in which case the appointment secretaries forward the request to the nurse and physician team. The team then reviews the record and sends a message to the appointment secretaries to schedule the screening test. Study duration The study was conducted from October 1, 2003, to October 31, 2004. The study began as a pilot study in October 2003 withpatients of 1 primary care physician used to assess the feasibility of the process of proactively contacting patients for scheduling mammography. In January 2004, all female patients of PCIM were enrolled. Each month, patients were randomly assigned to the intervention group as they became due for mammography; that is, they had not undergone mammography in the past 9 months. The last group of patients was included in the trial in October 2004. Measurement The assessment of mammography rates for analysis was a cross-sectional measurement of annual screening rates using their status as of December 31, 2004, for each group. The rates of all other preventive screening tests were also assessed for both groups for their recommended time frames. The use of annual physical examinations in the preceding 12 months was also determined from the administrative data. Demographic information, including age, insurance payer type, and geographic region, was obtained from the administrative data. Statistical analysis The percentage of patients current with mammography screening recommendations was compared between groups using the χ2 test. Comparisons between the groups for other preventive screening rates were also based on χ2 tests. Similar tests were performed to compare the Mayo Clinic employees who received reminders via US mail with those who received e-mail reminders. Logistic regression was performed on mammography status, with adjustment for the clustering of patients within primary care provider, as well as patient age, payer type, distance from home to Rochester, and study arm. A generalized linear model accounting for the correlation of patients clustered by physician was used to analyze the number of annual physical examination visits per patient. This model included patient age, payer type, distance from home to Rochester, and study arm. Results The postintervention rates for all eligible preventive services for the study patients are listed in Table 1. Of the women in the intervention group, 64.3% underwent annual mammography compared with 55.3% in the control group (P <.001). The rates for all other adult preventive services were not statistically different between the 2 groups. The lipid screening rate was slightly higher in the control group, with 85.6% of patients current for lipid screening compared with 84.1% in the intervention group (P = .08). The screening rates for all preventive services for the Mayo Clinic employee subpopulation are listed in Table 2 and Table 3. For annual mammography, 70.0% in the intervention group and 57.5% in the control group underwent annual screening. For employees receiving e-mail reminders, the annual mammography rate was 72.2%. Overall, all screening rates were higher for the Mayo employee subpopulation than for non–Mayo Clinic employees. The logistic regression models identified several factors significantly related to mammography status, including patient age (P <.001), payer status (P <.001), geographic region (P = .03), and study arm (P <.001). Patients who were older, had Mayo Clinic–provided insurance, lived in southeastern Minnesota, or received proactive reminders were more likely to be current for mammography, and women who had government-provided insurance (Medicaid) or who paid out of pocket were less likely to be current. Odds ratio estimates and 95% confidence intervals for the logistic model are given in Table 4. In a logistic model of Mayo Clinic employees who received a reminder, age was significant (P = .02), but there was no difference between women who received proactive US mail reminders and those who received e-mail reminders (P = .24). The generalized linear model accounting for the patient clustering within physician practice found no significant difference between groups for adjusted annual physical examination visits during our study (P = .38). The parameter estimates and P values are listed in Table 5. Comment Our study demonstrates the feasibility and effectiveness of a practice redesign to manage and improve breast cancer screening for a large population of primary care patients. During the 12-month study, appointment coordinators used an administrative database to send timely reminders to 3346 patients due for mammography. As a result of this intervention, a significantly higher completion rate for annual mammography was noted among the women receiving the reminders (intervention group) compared with those receiving usual care (control group). The use of appointment secretaries to proactively manage and schedule breast cancer screening for our patients was a practical solution for our large group practice. This is consistent with other published reports. Practice type had a pronounced effect on the delivery of preventive services.5 Larger practices have been found to deliver higher quality of care.19-21 This is attributed to the ability of large practices to apply resources, management systems, and office staff to help physicians provide preventive services for their patients.21-23 The results of previous studies of the effect of information technology on delivery of preventive services have been mixed, with some studies showing enhancement and others showing no effect on delivery rates.12,13,19 It seems that having enhanced information technology alone is insufficient to consistently affect the delivery of preventive services; practice redesign must also occur. A process in which the entire practice team, including allied health staff working at the maximum of their licensure, can significantly improve the performance of the delivery of preventive services.17 Physician or nursing staff input had no important role in the delivery of this service in our practice. In our employee population, the women who received e-mail reminders had an annual mammography rate of 72.2% compared with 68.1% for those who received letters, a statistically insignificant difference. Our study was not powered to detect a difference of this size as small as 4%. Nineteen hundred fifty-five patients would be needed in each group to declare the same difference as statistically significant with 80% power and with P = .05. An electronic notification system was as effective as the US mail system and could provide an efficient, cost-effective system for delivery of reminders of preventive services to primary care patients. In the future, e-mail reminders may be an important way to notify patients, who may prefer these reminders to letters. However, issues of e-mail security need to be further addressed; for example, the Mayo Clinic e-mail policy allows for internal communication about appointment reminders only and not about test or procedure results. Our intervention did not result in a significant decrease in the number of annual physical examinations performed. Patients in the intervention group visited their physicians for in-office annual physical examinations as frequently as those in the control group. In our study, we did not observe any improvement in completion rates for other preventive services in the intervention group. We had anticipated that the preventive service brochure included in the letters would have a positive effect on completion rates. However, it seems that a more active intervention is necessary and that patients need specific recommendations from their physicians about which preventive service they are due for. We also noted that the baseline screening rates for services such as colorectal cancer and cervical cancer were higher than for mammography and influenza vaccinations. Both mammography and influenza vaccinations are performed annually, and longer than a year can easily elapse before patients see their primary care physician. For this reason, interventions as in our study are needed to improve performance. In addition, many external quality-reporting organizations, including the Minnesota Community Measurement Project,24 measure and report mammography rates every 24 months. Within that interval, our screening rate is 84%, which is better than the colorectal cancer screening rate. In 2005, after our study was completed, we implemented the proactive system of scheduling for mammography in our entire patient population, which includes 11 119 women between the ages of 40 and 75 years. The current mammography rate for this population has increased further to 71%. This rate is likely higher than the study rate because we no longer are required to include the confusing research authorization with the patient notification. We have a 0.2 full-time equivalent of our appointment secretary effort to manage the mammography process for the patients of all physicians. We observed that patients with various insurance coverage had significantly different screening rates. Others also have reported that screening rates for breast cancer can vary depending on the patient's health insurance.5 In our practice, the care process for all patients is the same regardless of payer. Although we did not ascertain what accounted for this variation in the screening rate, it seems that different populations may need different strategies for improving the delivery of preventive services. The main limitation of our study was related to patient consent paperwork. The protocol required that paperwork be sent to patients to authorize their participation in the study. This process was complicated and confusing for the patients and seemed to have had a negative effect on the response rate of our patients. We did not ascertain the magnitude of this effect. Another limitation is that both the patient population and available practice resources at the Mayo Clinic may be different from those of other practices, which may raise questions about the generalizability of our study. However, because the baseline screening rates were similar to rates reported nationally,7,8,11 this practice redesign approach for delivering mammography should be similarly beneficial for all primary care practices. Others have advocated similar practice redesign to overcome barriers in addressing the preventive service needs of their patient populations.17 Screening rates for breast cancer and other preventive services are important quality indicators. With pay-for-performance, it will become increasingly important for primary care practices to develop processes to reliably deliver routine preventive screening services to their patients.25 Many preventive screening services can be delivered without involvement of physicians or physician visits, and office staff can manage the preventive service needs of patients, which should also decrease the costs incurred by practices, patients, and insurers. Correspondence: Rajeev Chaudhry, MBBS, MPH, Division of Primary Care Internal Medicine, Mayo Clinic, 200 First St SW, Rochester, MN 55905 (chaudhry.rajeev@mayo.edu). Accepted for Publication: November 25, 2006. Author Contributions:Study concept and design: Chaudhry, Scheitel, Leutink, Cabanela, Naessens, Davis, and Stroebel. Acquisition of data: Chaudhry, McMurtry, Leutink, Cabanela, and Rahman. Analysis and interpretation of data: Chaudhry, Scheitel, McMurtry, Naessens, and Stroebel. Drafting of the manuscript: Chaudhry, Scheitel, McMurtry, Davis, and Stroebel. Critical revision of the manuscript for important intellectual content: Chaudhry, Leutink, Cabanela, Naessens, and Rahman. Statistical analysis: McMurtry, Cabanela, and Naessens. Obtained funding: Chaudhry. Administrative, technical, and material support: Chaudhry, Leutink, Rahman, and Davis. Study supervision: Chaudhry, Cabanela, and Stroebel. Financial Disclosure: None reported. Funding/Support: This study was funded by the Mayo Foundation. Acknowledgment: We thank Mary Pat Anderson for secretarial support and the Section of Scientific Publications, Mayo Clinic, for assistance in preparing the manuscript. References 1. American Cancer Society, Estimated new cancer cases and deaths by sex for all sites, US, 2005 Atlanta, Ga American Cancer Society August15 2006;http://www.cancer.org/downloads/stt/Estimated_New_Cancer_Cases_and_Deaths_by_Sex_for_All_Sites,_US,_2005.pdf 2. US Preventive Services Task Force, Guide to Clinical Preventive Services: Report of the US Preventive Services Task Force. 2nd ed. Washington, DC US Dept of Health and Human Services, Office of Public Health and Science, Office of Disease Prevention and Health Promotion1996; 3. Nyström LAndersson IBjurstam NFrisell JNordenskjöld BRutqvist LE Long-term effects of mammography screening: updated overview of the Swedish randomised trials. Lancet 2002;359909- 919[published correction appears in Lancet. 2002;360:724]PubMedGoogle ScholarCrossref 4. Shapiro S Periodic screening for breast cancer: the HIP Randomized Controlled Trial Health Insurance Plan. J Natl Cancer Inst Monogr 1997;2227- 30PubMedGoogle Scholar 5. Pham HHSchrag DHargraves JLBach PB Delivery of preventive services to older adults by primary care physicians. JAMA 2005;294473- 481PubMedGoogle ScholarCrossref 6. Salive MEGuralnik JMBrock D Preventive services for breast and cervical cancer in US office-based practices. Prev Med 1996;25561- 568PubMedGoogle ScholarCrossref 7. McGlynn EAAsch SMAdams J et al. The quality of health care delivered to adults in the United States. N Engl J Med 2003;3482635- 2645PubMedGoogle ScholarCrossref 8. Kerr EAMcGlynn EAAdams JKeesey JAsch SM Profiling the quality of care in twelve communities: results from the CQI study. Health Aff (Millwood) 2004;23247- 256PubMedGoogle ScholarCrossref 9. Freeman JLKlabunde CNSchussler NWarren JLVirnig BACooper GS Measuring breast, colorectal, and prostate cancer screening with Medicare claims data. Med Care 2002;40 ((suppl)) IV-36- IV42Google ScholarCrossref 10. Randolph WMMahnken JDGoodwin JSFreeman JL Using Medicare data to estimate the prevalence of breast cancer screening in older women: comparison of different methods to identify screening mammograms. Health Serv Res 2002;371643- 1657PubMedGoogle ScholarCrossref 11. Jencks SFCuerdon TBurwen DR et al. Quality of medical care delivered to Medicare beneficiaries: a profile at state and national levels. JAMA 2000;2841670- 1676PubMedGoogle ScholarCrossref 12. Dexter PRPerkins SOverhage JMMaharry KKohler RBMcDonald CJ A computerized reminder system to increase the use of preventive care for hospitalized patients. N Engl J Med 2001;345965- 970PubMedGoogle ScholarCrossref 13. Shea SDuMouchel WBahamonde L A meta-analysis of 16 randomized controlled trials to evaluate computer-based clinical reminder systems for preventive care in the ambulatory setting. J Am Med Inform Assoc 1996;3399- 409PubMedGoogle ScholarCrossref 14. Committee on Quality Health Care in America; Institute of Medicine, Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, DC National Academy Press2001; 15. Kottke TEBrekke MLSolberg LI Making “time” for preventive services. Mayo Clin Proc 1993;68785- 791PubMedGoogle ScholarCrossref 16. Yarnall KSPollak KIOstbye TKrause KMMichener JL Primary care: is there enough time for prevention? Am J Public Health 2003;93635- 641PubMedGoogle ScholarCrossref 17. Kottke TESolberg LI STEP(ing)-UP to deliver clinical preventive services. Am J Prev Med 2001;2168- 69PubMedGoogle ScholarCrossref 18. Hung DYRundall TACrabtree BFTallia AFCohen DJHalpin HA Influence of primary care practice and provider attributes on preventive service delivery. Am J Prev Med 2006;30413- 422PubMedGoogle ScholarCrossref 19. Gann PMelville SKLuckmann R Characteristics of primary care office systems as predictors of mammography utilization. Ann Intern Med 1993;118893- 898PubMedGoogle ScholarCrossref 20. Miller RHBovbjerg RR Efforts to improve patient safety in large, capitated medical groups: description and conceptual model. J Health Polit Policy Law 2002;27401- 440PubMedGoogle ScholarCrossref 21. Reschovsky JReed MBlumenthal DLandon B Physicians' assessments of their ability to provide high-quality care in a changing health care system. Med Care 2001;39254- 269PubMedGoogle ScholarCrossref 22. Casalino LPDevers KJLake TKReed MStoddard JJ Benefits of and barriers to large medical group practice in the United States. Arch Intern Med 2003;1631958- 1964PubMedGoogle ScholarCrossref 23. Gillies RRShortell SMCasalino LRobinson JCRundall TG How different is California? a comparison of US physician organizations. Health Aff (Millwood) 2003;22492- 502PubMedGoogle Scholar 24. Community Measurement MN [homepage on the Internet] St Paul MN Community Measurement2005;http://www.mnhealthcare.org. Accessed August 15, 2006 25. Rosenthal MBFernandopulle RSong HRLandon B Paying for quality: providers' incentives for quality improvement. Health Aff (Millwood) 2004;23127- 141PubMedGoogle ScholarCrossref
TI - Web-Based Proactive System to Improve Breast Cancer Screening: A Randomized Controlled Trial
JF - Archives of Internal Medicine
DO - 10.1001/archinte.167.6.606
DA - 2007-03-26
UR - https://www.deepdyve.com/lp/american-medical-association/web-based-proactive-system-to-improve-breast-cancer-screening-a-u7V6ivUCud
SP - 606
EP - 611
VL - 167
IS - 6
DP - DeepDyve
ER -