Primary Care Clinicians' Perspectives on Reducing Low-Value Care in an Integrated Delivery System
Perm J 2016 Winter;20(1):41-46
Context: Perceptions about low-value care (eg, medical tests and procedures that may be unnecessary and/or harmful) among clinicians with capitated salaries are unknown.
Improving quality and enhancing value in health care delivery is a national priority. Many public and private sector initiatives are developing, testing, and deploying strategies to reduce overuse and to promote appropriate use, quality, and optimal health outcomes.1-3 Overtreatment and failures of care delivery and coordination account for nearly 20% and 18%, respectively, of waste of health care expenditures.4 Many public and private sector initiatives, including the Choosing Wisely5 initiative, are developing, testing, and deploying strategies to reduce overuse and to promote the appropriate use of health care services. Started in 2012 by the American Board of Internal Medicine Foundation in Philadelphia, PA, the Choosing Wisely initiative focuses on encouraging physicians, patients, and other health care stakeholders to have conversations about medical tests and procedures that may be unnecessary and, in some instances, may cause harm. As of June 2015, more than 70 specialty society partners have developed lists of at least 5 recommendations for tests and treatments that are potentially overused and do not provide meaningful benefits for patients.5
Care models that focus on clinician accountability, insurance risk protection, and alternative clinician payments have been receiving increasing attention, with health care reform as a possible model to decrease expenditures through the removal of financial incentives for providing more care. Understanding clinicians' perspectives is particularly important because health care professionals play a key role in choosing which health care services are used.
Group Health Cooperative (Group Health) is a mixed-model delivery system with just under 600,000 members in Washington State, 380,000 of whom are cared for in the integrated delivery system by approximately 1000 salaried, multidisciplinary clinicians (320 primary care physicians and pediatricians and 635 specialty physicians) in 25 Medical Centers. Group Health has been implementing a "Resource Stewardship" initiative to bring personalized information to clinicians to inform their clinical decision making on the use of particular services and to reduce the use of low-value care. In October 2012, each Medical Center received a report with clinician-specific practice pattern data from the previous year that included peer comparisons within each clinic. Data included clinician-specific total cost of care for patients and use of high-end imaging (computed tomography and magnetic resonance imaging), laboratory test ordering, prescribing, and specialty referrals. Follow-up data in Spring 2013 provided additional peer comparison data on rates of overuse of Papanicolaou tests and prostate-specific antigen screening in men older than 75 years in the previous 12 months.
The aims of this study were to explore the clinicians' perceived use of and responsibility for reducing low-value care, barriers to decreasing its use, and knowledge and perceived legitimacy of the Choosing Wisely initiative.5 Additionally, we evaluated whether clinicians believed they have decreased their use of low-value care after receiving a peer comparison report on their practice patterns.
Design Overview, Settings, and Participants
We conducted an online survey of 304 primary care clinicians at Group Health in Fall 2013 (Appendix 1 available at: www.thepermanentejournal.org/files/Winter2016/CareSurvey.pdf). An e-mail invitation was sent by the Medical Group President and Medical Director, with up to 3 reminder e-mails spaced 1 week apart. The survey collected information on our specific work in Group Health on the Resource Stewardship initiative and attitudes toward and awareness of low-value care. It also used select questions from a national survey of physicians conducted for the American Board of Internal Medicine Foundation by the Robert Wood Johnson Foundation in Princeton, NJ, to hear physicians' views about the health care system.6 We used several of the same questions in our survey to enable direct comparisons with other US clinician populations. In the survey, we defined low-value care as "unnecessary care, or care that does not improve patient outcomes and can harm patients." To help clinicians better understand our definition, we also included the following examples of low-value care: "prescribing antibiotics for viral infections, recommending screening exams more frequently than needed, and ordering tests without informing patients of their harms and benefits so they can participate in a decision about whether they want an imperfect test (eg, PSA [prostate-specific antigen] screening)." The overall response rate was 62%; the response rate was 50% for physician assistants (n = 23) and 64% for physicians (n = 166).
We compared the characteristics of respondents to nonrespondents to assess representativeness. Clinician characteristics were obtained from employee files. At the end of the survey, respondents were given the opportunity to provide open-ended responses, which were collated and reviewed for themes.
This quality-improvement initiative and subsequent evaluation survey analyses were determined to be exempt from human subjects review in compliance with the US Office for Human Research Protections (45 CFR part 46). All analyses were conducted with Stata 13.0 statistical software (StataCorp, College Station, TX).
Respondents were significantly older than nonrespondents (mean = 49 vs 46 years), and had more years practicing (mean = 21 vs 17 years; Table 1). Respondents were also significantly more likely than nonrespondents to have been at Group Health 10 or more years (59% vs 38%). There were no statistically significant differences between respondents and nonrespondents in panel size, sex, race, provider type, and primary department.
Attitudes and Awareness of Low-Value Care
Nearly all clinicians who responded to the survey (> 90%) replied that the cost of care they personally deliver is important to different stakeholders and believe it is fair to ask clinicians to be cost-conscious (Table 2). In contrast, clinicians were split about trading off patients' best interests and cost; 71% reported feeling pressure from patients to order more tests. Less than 15% of clinicians reported that low-value care was provided by themselves or by their colleagues sometimes, frequently, or almost always; 91% said they provided low-value care only rarely or occasionally.
Almost two-thirds of respondents said they had changed their practice to reduce low-value care in the prior 12 months, and approximately 40% had reduced low-value care a significant or moderate amount in the 2 areas targeted by resource use reports: overscreening for cervical and prostate cancers. Receipt of transparent resource use reports resulted in having conversations (74% of those who remember receiving the report reported participating in a discussion with their clinic chiefs; data not shown). Most clinicians found the clinic-level reports useful for understanding their own practice patterns and for provoking discussion among their peers. One-third of respondents found the information on their resource use surprising (data not shown).
One-third of clinicians were unaware of the Choosing Wisely campaign; among the 66% who were aware, 97% reported they considered it a legitimate source of information on unnecessary tests and procedures (data not shown).
Low-Value Care in Practice
Almost two-thirds (62%) of respondents said patients request tests or procedures that the clinician considered unnecessary at least several times per week (Table 3). Most (88%) indicated they were comfortable discussing avoiding low-value care with their patients, and 56% reported having such discussions 5 or more times in the past 30 days. In contrast, fewer physicians reported being comfortable speaking with colleagues about low-value care (70%); only 9% had conversations with colleagues 5 or more times in the past 30 days. Clinicians reported that patients follow their advice about avoiding unnecessary tests or procedures always or almost always (16%), often (45%), and about half the time (36%) when conversations with patients happen. However, when patients are insistent about getting a test or procedure without clear clinical indication, 80% of clinicians reported that they order the test or procedure but provide patients their reasons for advising against it.
Perceived Barriers to Reducing Low-Value Care
Notable barriers included time constraints (45%), challenges overcoming patient preferences and values (44%), community standards of care (43%), fear of patients' dissatisfaction (41%), patients' knowledge about harms of having low-value care (38%), availability of tools to support relevant shared decision making (37%), and fear of litigation (31%), as shown in Figure 1. Relationships with patients, tolerance of uncertainty, and support and prioritization from leaders and clinical colleagues notably were not perceived as important barriers to reducing low-value care.
Nearly 40% of clinicians (n = 70) provided open-ended responses about ways to improve efforts to decrease low-value care. Feedback ranged from ideas on reasons underlying the provision of low-value care (both upstream and downstream from primary care) to potentially useful toolkits (Table 4). The most frequent feedback was that clinicians felt a tension between being responsive to patients to promote patient satisfaction and restricting the use of procedures with little or no value. For example, physicians reported being torn by the conflicting institutional messages and incentives to simultaneously reduce low-value care and maintain patient satisfaction after patients request these tests or procedures. A number of clinicians expressed the need for more information for patients and for their own use. Additionally, there was a desire for public awareness campaigns to raise the profile of the importance of reducing low-value care. Several clinicians described the complexity of measuring low-value care because procedures and test decisions are often not clear and decision making may be nuanced for reporting. Despite this, there was unprompted support for receiving transparent, comparative performance information.
Salaried primary care clinicians in an integrated delivery system want to provide high-value care, but they identify the need for additional support to manage patient requests for unnecessary and low-value care and for tools to provide direct and timely feedback on their use of low-value care. Our salaried clinicians identified important barriers to reducing low-value care, many of which are actionable and reasonable targets for change. Some of the identified barriers can be addressed through public awareness and education strategies (eg, changes in community standards of care, patients' knowledge about the harms of having low-value care, and availability of tools to help efficiently share benefits and harms of low-value care to patients), whereas others rely on changes and enhancements in delivery system infrastructure (eg, longer appointment times). With additional information made available directly to patients and communities through a variety of channels, individuals may be able to engage in more informed discussions with providers about why more care is not always better and how more care can actually be harmful in certain situations.
Transparent peer comparison data on overall resource use and use of low-value services were reported as important for understanding and promoting discussion about variation in practice patterns with practice groups and colleagues. The provision of peer comparison data through the Resource Stewardship initiative and the awareness and perceived legitimacy of the Choosing Wisely campaign suggest that evidence-based initiatives may be used to motivate clinicians to be conscientious stewards of limited health care resources.
Three other provider surveys have reported US primary and specialty physicians' attitudes toward low-value care, including attitudes and perceptions on the problems and possible solutions.6-8 The first survey7 received a 70% response rate from 627 physicians randomly selected from the American Medical Association Masterfile and reported overuse of unnecessary care within their own patients. In contrast to our salaried clinicians, these respondents identified malpractice concerns (76%), clinical performance measures (52%), and financial incentives (62%) as drivers of unnecessary care use. A national survey commissioned by the American Board of Internal Medicine Foundation reported on perceived problems, causes, and solutions for unnecessary tests and procedures from 600 US primary and specialty physicians in 2014 (no response rate provided).6 Another 2012 survey, the "Physicians, Health Care Costs, and Society," included 2556 physicians (65% response rate) also randomly selected from the American Medical Association Masterfile; this survey focused on physicians' views on controlling health care costs and reported important differences from ours.8 In the 2012 survey, only 36% of physicians responded that they had a major responsibility for helping reduce health care costs; more than half reported that patients (52%), hospitals and health systems (56%), pharmaceutical and device manufacturers (56%), insurance companies (59%), and trial lawyers (60%) had a major responsibility. Unsurprisingly, these opinions appear to be related to the respondents' practice settings and type of compensation; 70% did not support eliminating fee-for-service reimbursement as a strategy. Finally, most of the surveyed clinicians in the 2012 survey agreed that they "should be solely devoted to their patients' best interests, even if that is expensive,"8 compared with less than half of the clinicians in our study who agreed with that statement. Across all 3 surveys,6-8 clinicians indicated they ordered tests because of patient insistence.
The national surveys of physicians6-8 and our survey of salaried clinicians from an integrated delivery system support the notion that patients follow the advice of their clinicians and that most clinicians report talking with patients about reasons to avoid unnecessary tests. Important differences in reported "major" drivers of low-value care were observed across the surveys. In our capitated, integrated delivery system, the major barriers noted by our clinicians were community standards of care, time constraints, and patient satisfaction; fear of litigation was not a notable barrier relative to other recent US physician reports.6-9 Malpractice (52%), safety (36%), and physician reassurance (30%) lead the national survey response, and patient satisfaction (23%) and time (13%) were much less important. Interestingly, only 5% of the national sample reported the fee-for-service system as a major driver.6 Our survey provides an important contrast to reported drivers of low-value care relative to other published national populations.6-8
There are many differences between integrated delivery systems and other care models. However, with the Patient Protection and Accountable Care Act of 201010 (legislation to help increase the quality and affordability of health insurance and health care), clinician groups may become more like integrated practices through accountable care organizations and pay-for-performance/value and quality. Our survey indicated that clinicians are aware of rising health care costs and are willing to be stewards of limited health care resources.
This project was specifically designed to engage frontline primary care clinicians so we could better understand how transparent individual clinician-level practice pattern reports were received and what additional tools are needed to support reducing delivery of low-value care. Clinician engagement and interest expressed in the survey may have resulted from multiple reminder e-mails to respond to the survey and from clinicians being asked to respond to the survey from the Medical Group President and from the Medical Director for Quality Improvement, yielding a greater than 60% response rate. Additionally, we believe the enthusiastic response to the open-ended feedback at the end of the survey suggests that clinicians desire engagement in the discussion about improving value and reducing unnecessary spending.
Reducing the delivery and receipt of low-value care (eg, reducing unnecessary testing) will require continued involvement from clinical teams and patients. Success in reducing the delivery of low-value care will depend on deeper engagement of clinicians in all health care delivery settings. Continued support of materials for patients focused on common low-value care, such as the Choosing Wisely brochures for patients developed by Consumer Reports,11 is an important step to support efficient discussions with health care clinicians.
The author(s) have no conflicts of interest to disclose.
This work was funded by a Partnership for Innovation grant from the Group Health Foundation, Seattle, WA, which had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript.
Kathleen Louden, ELS, of Louden Health Communications provided editorial assistance.
1. Romano MJ, Segal JB, Pollack CE. The association between continuity of care and the overuse of medical procedures. JAMA Intern Med 2015 Jul 1;175(7):1148-54. DOI: http;//dx.doi.org/10.1001/jamainternmed.2015.1340.