Can J Diabetes 39 (2015) 83e87
Contents lists available at ScienceDirect
Canadian Journal of Diabetes journal homepage: www.canadianjournalofdiabetes.com
Review
Performance-Based Financial Incentives for Diabetes Care: An Effective Strategy? Lesley P. Latham MSc, Emily Gard Marshall PhD * Dalhousie University, Halifax, Nova Scotia, Canada
a r t i c l e i n f o
a b s t r a c t
Article history: Received 10 December 2013 Received in revised form 10 April 2014 Accepted 11 June 2014
The use of financial incentives provided to primary care physicians who achieve target management or clinical outcomes has been advocated to support the fulfillment of care recommendations for patients with diabetes. This article explores the characteristics of incentive models implemented in the context of universal healthcare systems in the United Kingdom, Australia, Taiwan and Canada; the extent to which these interventions have been successful in improving diabetes outcomes; and the key challenges and concerns around implementing incentive models. Research in the effect of incentives in the United Kingdom demonstrates some improvements in process outcomes and achievement of cholesterol, blood pressure and glycated hemoglobin (A1C) targets. Evidence of the efficacy of programs implemented outside of the United Kingdom is very limited but suggests that physicians participating in these enhanced billing incentive programs were already completing the guideline-recommended care prior to the introduction of the incentive. A shift to a pay-for-performance programs may have important implications for professionalism and patient-centred care. In the absence of definitive evidence that financial incentives drive the quality of diabetes management at the level of primary care, policy makers should proceed with caution. It is important to look beyond simply modifying physicians’ behaviours and address the factors and systemic barriers that make it challenging for patients and physicians to manage diabetes in partnership. Ó 2015 Canadian Diabetes Association
Keywords: diabetes mellitus primary healthcare reimbursement incentive
r é s u m é Mots clés : diabète sucré prime de remboursement soins de santé primaires
L’utilisation d’incitatifs financiers chez les médecins de premier recours qui atteignent leur cible de prise en charge ou les résultats cliniques a été recommandée pour favoriser la réalisation des recommandations pour les soins des patients diabétiques. Cet article explore les caractéristiques des modèles incitatifs mis en place dans le contexte des systèmes de soins de santé universels au Royaume-Uni, en Australie, à Taïwan et au Canada; dans quelle mesure ces interventions ont réussi à améliorer les résultats liés au diabète; et les défis et les préoccupations principales concernant la mise en place des modèles incitatifs. La recherche sur l’effet des mesures incitatives au Royaume-Uni démontre certaines améliorations dans les résultats liés au processus et l’atteinte des valeurs cibles de cholestérol, de pression artérielle et d’hémoglobine glyquée (A1c). Les données probantes sur l’efficacité des programmes mis en place à l’extérieur du Royaume-Uni sont très limitées, mais suggèrent que les médecins participants à ces programmes d’incitatifs à une meilleure performance appliquaient déjà les soins recommandés par les lignes directrices avant l’introduction des mesures incitatives. Le fait d’opter pour les programmes de rémunération en fonction de la performance peut avoir d’importantes répercussions sur le professionnalisme et les soins centrés sur le patient. En l’absence de preuves concluantes sur le fait que les incitatifs financiers promeuvent la qualité de la prise en charge du diabète aux soins primaires, les responsables de l’élaboration des politiques devraient procéder avec prudence. Il est important de dépasser la simple modification des comportements des médecins et de se pencher sur les facteurs et les obstacles systémiques qui font qu’il est difficile pour les patients et les médecins de collaborer ensemble à la prise en charge du diabète. Ó 2015 Canadian Diabetes Association
* Address for correspondence: Emily Gard Marshall, PhD, Dalhousie Family Medicine, AJLB 8 QEII HSC, 5909 Veteran’s Memorial Lane, Halifax, Nova Scotia B3H 2E2, Canada. E-mail address: [email protected] 1499-2671/$ e see front matter Ó 2015 Canadian Diabetes Association http://dx.doi.org/10.1016/j.jcjd.2014.06.002
84
L.P. Latham, E.G. Marshall / Can J Diabetes 39 (2015) 83e87
Introduction Type 2 diabetes is a cause of substantial morbidity and mortality in Canada. Clinical management, according to recommended guidelines at the level of primary care, can improve outcomes for patients with diabetes (1). However, analysis of national health surveys suggests that there is a gap between practices recommended by current guidelines and the care patients with diabetes receive (2). One method of increasing the fulfillment of recommended care is the use of financial incentives provided to primary care physicians who achieve target processes or clinical outcomes in their patients. This approach is sometimes termed pay-for-performance, where performance refers to the meeting of preestablished targets for the delivery of healthcare services. Founded on the presumption that guideline-based care will generate savings by reducing demands on costly hospital-based services over the long term, incentive programs aimed at improving diabetes outcomes are increasing in popularity around the world. In Canada, for example, enhanced billing for completion of diabetes care according to the accepted guidelines has been used in several provinces as part of larger initiatives to improve the quality and outcomes of care for patients who have chronic disease. However, it remains unclear whether these incentives are effective in improving the quality of diabetes care. This article provides a short narrative exploration of the characteristics of incentive models from the United Kingdom, Australia, Taiwan and Canada, the extent to which these interventions have demonstrated success in improving diabetes outcomes, and an overview of key challenges and concerns related to the use of incentives to improve quality of care. Key literature on the topic of performance-based incentives in primary care, including systematic reviews (3,4), have focused on incentives initiated in the United States (US). However, key differences in the way in which US programs are implemented and evaluated do not allow for suitable comparison. Characteristics of Pay-for-Performance Systems The United Kingdom (UK) is home to the most ambitious example of pay-for-performance in diabetes care. In connection with an initiative to improve chronic disease care and outcomes, the UK government introduced a pay-for-performance model called the Quality Outcomes Framework (QOF) as part of the General Practitioner contract in 2004. Under QOF, up to one-quarter of general practice income is linked to performance according to a complex set of quality indicators. Practices can earn up to 1000 points in several domains, including clinical outcomes (661 points); organizational outcomes (262 points); patient experience (33 points) and additional services (44 points). Diabetes indicators account for approximately 14% of QOF clinical domain points (5). Although participation is voluntary, coverage is estimated to be above 99% (5). Under QOF, quality care is defined by the framework developed by Campbell and colleagues (6); it employs process (monitoring, prescribing and counselling); intermediate clinical levels (glycated hemoglobin [A1C], cholesterol and blood pressure); and patientreported indicators (quality of life and satisfaction with care) to evaluate performance. Quality scores are heavily weighted toward clinical and physiologic indicators. It is important to note that patients may be removed from reporting practices (exception reporting) so that physicians are not financially penalized for factors outside of their control (e.g. missed appointments). Australia introduced financial incentives related to quality of care for patients with chronic diseases for primary care providers in 1999 in the form of the Practice Incentive Program (PIP) (7). Under PIP, primary care physicians can elect to participate in a blended
payment model, which includes additional incentives for each patient who completes a cycle of care (described as a complete sequence of visits recommended by clinical guidelines) for asthma, diabetes, mental health or cervical cancer screening. Taiwan introduced a voluntary incentive program (DM-P4P) for diabetes care in 2001. The design by Taiwan’s National Health Insurance initially involved case management fees provided to physicians for achieving process-based outcomes, such as follow up and annual evaluation visits. Service items required for each of these visits include nutrition education, physical examination and laboratory testing (8). Participating physicians are required to complete specialized training in diabetes management (Diabetes Shared Care Program) and are permitted to select which patients are eligible for the program. In Canada, several provinces have introduced incentive programs in the form of enhanced billing or condition-based payments. Physicians can receive additional payments for each patient with diabetes who is managed in accordance with practice guidelines (9). For example, Ontario’s Diabetes Management Incentive provides a $60 annual, per-patient payment to physicians who complete and document the required elements of care recommended by the Canadian Diabetes Association guidelines (10). Family physicians in Nova Scotia are eligible for additional payments through a Chronic Disease Management program if they complete and document assessment of diabetes risk factors and annual monitoring (11). Similarly, in British Columbia, family physicians are offered chronic disease management payments for a number of chronic conditions in accordance with the Full Service Family Practice Condition Payments introduced in 2003. Physicians may receive a $120 payment (per patient) for providing care according to the British Columbia clinical guideline recommendations for diabetes mellitus over the preceding 12 months (12).
Linking Incentives and Quality in Diabetes Care The evidence suggests that in the UK, incentive models have spurred some improvements in process outcomes and achievement of cholesterol, blood pressure and A1C targets (13e15). A retrospective analysis focusing on glycemic control found that between 2006 and 2008, the percentage of adults who achieved A1C levels within the recommended guidelines rose from 37.9% to 52.1% (16). Still, the evidence is mixed, and its interpretation is hampered by methodologic challenges and confounders. For example, broad implementation and uptake of QOF means there is no adequate control group. Interpreting trends in quality improvement is also complicated by the fact that the quality of diabetes care was improving prior to the introduction of QOF in 2004 (17). Several studies have attempted to separate clinical improvements resulting from QOF from background trends. One longitudinal study found that quality of care improved over and above the pre-incentive trend in the first year after the introduction of QOF by about 14.2%. However, during the second and third years, this difference diminished, and there was increasing variability according to patient demographic characteristics (17). An interrupted, time series analysis of a subset of family practices found that the introduction of QOF accelerated improvements in diabetes quality of care between 2003 and 2005, but this trend had slowed by 2007. The same study found that continuity of care was reduced after the introduction of QOF, while no changes in reported access to care were observed (18). With regard to longer term outcomes such as hospitalizations, one ecologic, cross-sectional study comparing hospital admission rates with quality-of-care scores found a 10-fold variation in diabetes-related admissions despite uniformly high-quality scores (19). Notably, in this study, low neighborhood socioeconomic status
Minimal improvements to management outcomes Poor uptake and self-selection for physicians already adhering to management guidelines In selected provinces with limited uptake Process outcomes only (compliance with Canadian Diabetes Association guidelines) Enhanced billing/conditionbased payments Implemented in some provinces (e.g. Ontario’s Diabetes Management Incentive) Canada
Suggested increased adherence to recommended guidelines Reduced hospitalization rates after 3 years for enrolled patients (limited evidence) Nationally with limited uptake Process outcomes (case management incentives) and intermediate clinical outcomes Enhanced billing/conditionbased payments Implemented nationally (DM-P4P) Taiwan
Suggested increased likelihood of physician ordering appropriate monitoring tests (limited evidence) Nationally with limited uptake Process outcomes only (completion of “cycles of care”) Enhanced billing/conditionbased payments Implemented nationally (Practice Incentive Program) Australia
Some improvements in intermediate clinical outcomes (A1C and cholesterol) Trend of improvement in clinical targets levels-off after the first year of the program Nationally with broad coverage/ uptake
Implementation
Complex set of quality indicators (process outcomes, intermediate clinical outcomes, patient-reported indicators) Based on defined framework
Quality outcomes on which incentives are based
Proportion of primary care provider income is linked to quality outcomes Integrated into contract (Quality Outcomes Framework)
Model for provision of financial incentives Country
A shift to pay-for-performance programs may have important implications for professionalism and patient-centred caredeffects that are difficult to measure. For example, a focus on clinical outcomes excludes the complexities of diabetes and social factors that may make it difficult for patients to meet their targets, and it deprioritizes important facets of care, such as continuity and selfmanagement (26). There is a concern that patients who fail to achieve clinical targets may be viewed as noncompliant, or physicians may be subtly influenced to select patients on the basis of ease of management, which may increase health disparities. For example, there was concern following the introduction of QOF that physicians may inappropriately exclude patients from quality measures to avoid being financially penalized. Doran et al originally reported that a minority of practices may be achieving targets by excluding large numbers of patients (27). However, subsequent studies of exception reporting found little evidence of “gaming” (28). It is unclear whether incentives have influenced how physicians select and retain patients outside of the UK. Although it was not specifically designed to do so, the introduction of the QOF was expected to reduce health disparities through standardization of care (29). Although 2 studies suggest the benefits of the pay-for-performance model are similar to those with high and low income and social status (30,31), findings from several recent studies suggest that QOF does not address ethnic disparities in diabetes care adequately (13,32,33). Concerns have also emerged that patients from disadvantaged and vulnerable populations may be disproportionately excepted from QOF because their diabetes may be more challenging to manage (15,34). Dalton
85
UK
Challenges and Concerns Related to the Use of Incentives
Table Summary of the characteristics of performance-based incentive programs for diabetes care and the evidence for their efficacy
was more closely associated with hospitalization rates than were clinical indicators. There are few studies examining the effects of diabetes incentive programs outside of the UK. An Australian retrospective analysis of physician encounters with patients with diabetes concluded that participation in PIP increased the likelihood of a physician’s ordering an A1C test (20). Several cross-sectional studies have indicated that patients enrolled in Taiwan’s incentive program were more likely to receive guideline-recommended tests and examinations (21,22). A longitudinal study of the effect of the Taiwanese program on hospitalization rates found that patients enrolled in the incentive program were less likely to be hospitalized after 3 years of care compared with non-enrolled patients (23). Another longitudinal study found increased cost and health service use by patients enrolled in the first year of the program, owing to increased physician visits and examinations. The authors were unable to determine whether the costs of the program were offset by the savings related to long-term declines in service use (8). Concerns raised about the Taiwanese program include the absence of riskadjusted clinical targets as well as the lack of specific criteria for exclusion of patients. Accordingly, 1 study has demonstrated that older patients and those with higher comorbidity and severity of disease are more likely to be excluded from this program (24). In Ontario, researchers studying completion of recommended diabetes management practices before and after the introduction of the incentive billing code found minimal improvements to monitoring practices (25). In less than 25% of the patients studied, incentive codes had been billed by their physicians. Longitudinal analysis suggested that patients who received the recommended testing following the introduction of the incentive were the same patients who were receiving all recommended tests before it had been introduced. A summary of the characteristics of performance-based incentive programs and the evidence of their efficacy in improving the quality of diabetes care is presented in the Table.
Brief summary of evidence
L.P. Latham, E.G. Marshall / Can J Diabetes 39 (2015) 83e87
86
L.P. Latham, E.G. Marshall / Can J Diabetes 39 (2015) 83e87
et al examined the pattern of exclusions in 23 practices in northwest London (34). They found excepted patients were less likely to achieve treatment targets for A1C, blood pressure and cholesterol than were patients included in QOF. Patients with longstanding diabetes or multiple comorbidities were also more likely to be excluded from the A1C indicator. The same study found that black and South Asian patients were more likely to be excluded from the A1C indicator than were white patients. Financial rewards may encourage inappropriate (and potentially costly) target-oriented care by physicians and may divert attention from addressing patient-centered values and goals of treatment. A qualitative study examining the concerns of physicians and nurses following the introduction of QOF in the UK found that providers felt that incentives and quality measures brought needed attention to appropriate clinical targets (35). There was, however, a concern that the need to address QOF quality measures during office visits took away from the patients’ agendas, effectively making the encounter more physician centred than patient centred (35). Similarly, a qualitative study of UK physicians reported that data collection of performance measures during office visits was potentially distracting and changed the nature of the office visit (36). In another qualitative study conducted in the UK, primary care nurses working in chronic disease management reported feeling as though QOF encouraged regarding patients as conditions, rather than people, although they supported the overall goals of QOF, feeling that its structure helped them to provide high-quality care to patients (37). Despite concerns about distraction from patientcentred care, physicians reported that they did not pursue target incentives if they felt doing so would be detrimental to patients (18,35). Other qualitative studies describe lower patient satisfaction ratings for continuity of care following the introduction of QOF (38). In a system where interprofessional collaboration is actively promoted, selectively rewarding primary care physicians may discourage teamwork and coordinated care with other members of the heathcare team (35,39). For example, although nurses were integral in the care of patients with chronic disease following the introduction of QOF, increased income generated by achieving targets was directed largely to physicians because they were the employers and owners of the practices (35). Conclusion Although we have not fully discussed studies of pay-forperformance from the US, they generally provide similar perspectives on the efficacy of incentive in the improvement of diabetes care. For example, a retrospective cohort study of 334 primary care physicians in New York found no statistically significant improvements in process of care outcomes for diabetes management following the introduction of an incentive program (40). Notably, the implementation of performance-based incentives for diabetes care in the US differs from that in Canada and other countries in important ways. First, incentive programs in the US are typically not government sponsored but are administered through particular health plans or health maintenance organizations. Incentives within this model would be of potential benefit only for patients who had access to private insurance through these plans. Second, rather than providing incentives directly to physicians or practices, incentives are provided to health groups or institutions. Programs are commonly overseen separately by quality-improvement organizations, and data collection is performed by third parties, rather than by the healthcare providers themselves. For these reasons, we have chosen to focus this short review on programs that, although diverse, were relatively similar in context to incentive programs implemented in Canada. In fact, one of the main challenges in interpreting the evidence around financial incentives for the improvement of diabetes care is
the fact that programs have varied characteristics. The importance of the characteristics of the incentive program, such as the nature and magnitude of the incentive, the involvement of physicians in the development process, and the available infrastructure, have been discussed in detail elsewhere (3). In brief, the evidence suggests that a nationally implemented incentive program coupled with integration of pay-for-performance elements into primary care physicians’ salaries may be effective in improving the quality of diabetes care. However, even broadly implemented incentive programs such as QOF have demonstrated mainly the effects on process and intermediate clinical diabetes outcomes. More evidence is required to understand whether these improvements are sustained and translate into better long-term outcomes such as reduced hospitalizations for diabetes-related complications (26). It is important to note that evidence is lacking as to whether the investment required to initiate and maintain these performancebased incentive programs will be offset by downstream savings associated with reduced diabetes-related health service usage and hospitalizations (8,14,19). Many of the possible negative consequences of financial incentives have not been explored in the literature. A systematic method of evaluating the potential benefits and unintended consequences of implementing financial incentives may be of benefit. Glasziou et al have developed a checklist to guide decision making as to where and how financial incentives should be used (41); however, the lack of available evidence concerning the costs, benefits and harms of specific interventions may limit the applicability of such tools. In the absence of definitive evidence that financial incentives drive the quality of diabetes management at the level of primary care, policy makers advocating the use of such incentives in Canada should proceed with caution. A thoughtful and holistic approach to diabetes and chronic disease heath policy will look beyond simply modifying physician behaviours and will address the factors and systemic barriers that make it challenging for patients and physicians to manage diabetes in partnership. Improvements could include providing adequate billing codes to care appropriately for patients with complex issues; avoiding one-issue-only practices; addressing accessibility to primary care so that patients can develop longer-term relationships with providers, thus allowing for continuity and encouraging more interprofessional collaborative care models to help patients change lifestyle risk factors; and addressing chronic disease prevention in a more proactive way in primary care. Competing Interests None.
References 1. Dagogo-Jack S. Preventing diabetes-related morbidity and mortality in the primary care setting. J Natl Med Assoc 2002;94:549e60. 2. Canadian Institute for Health Information. Diabetes care gaps and disparities in Canada. December, 2009. 3. Van Herck P, De Smedt D, Annemans L, et al. Systematic review: Effects, design choices, and context of pay-for-performance in health care. BMC Health Serv Res 2010;10:247. 4. Scott A, Sivey P, Ait Ouakrim D, et al. The effect of financial incentives on the quality of health care provided by primary care physicians. Cochrane Database Syst Rev 2011;7(9):CD008451. 5. Health and Social Care Information Centre, Prescribing and Primary Care Team., Quality and outcomes framework: Achievement, prevalence and exceptions data 2011/12. Author; 2012. 6. Campbell SM, Roland MO, Buetow SA. Defining quality of care. Soc Sci Med 2000;51:1611e25. 7. Scott A, Naccarella L, Furler J, et al. Financial incentives and the quality of primary care in Australia Report. Australian Primary Health Care Research Institute (APHCRI). 8. Cheng SH, Lee TT, Chen CC. A longitudinal examination of a pay-forperformance program for diabetes care: Evidence from a natural experiment. Med Care 2012;50:109e16.
L.P. Latham, E.G. Marshall / Can J Diabetes 39 (2015) 83e87 9. The Canadian Diabetes Association. Diabetes: Canada at the tipping point: charting a new path. Author; 2010. 10. Government of Ontario. Guide to physician compensation. 2009. 11. Nova Scotia Medical Services Insurance. Physician’s Manual. 2012. 12. General Practice Services Committee. Chronic disease management incentives billing guide. 2010. 13. Alshamsan R, Lee JT, Majeed A, et al. Effect of a UK pay-for-performance program on ethnic disparities in diabetes outcomes: Interrupted time series analysis. Ann Fam Med 2012;10:228e34. 14. Alshamsan R, Millett C, Majeed A, Khunti K. Has pay for performance improved the management of diabetes in the United Kingdom? Prim Care Diabetes 2010; 4:73e8. 15. Calvert M, Shankar A, McManus RJ, et al. Effect of the quality and outcomes framework on diabetes care in the United Kingdom: Retrospective cohort study. BMJ 2009;338:b1870. 16. Oluwatowoju I, Abu E, Wild SH, Byrne CD. Improvements in glycaemic control and cholesterol concentrations associated with the Quality and Outcomes Framework: A regional 2-year audit of diabetes care in the UK. Diabet Med 2010;27:354e9. 17. Kontopantelis E, Reeves D, Valderas JM, et al. Recorded quality of primary care for patients with diabetes in England before and after the introduction of a financial incentive scheme: A longitudinal observational study. BMJ Qual Saf 2013;22:53e64. 18. Campbell SM, Reeves D, Kontopantelis E, et al. Effects of pay for performance on the quality of primary care in England. N Engl J Med 2009;23(361):368e78. 19. Bottle A, Millett C, Xie Y, et al. Quality of primary care and hospital admissions for diabetes mellitus in England. J Ambul Care Manage 2008;31:226e38. 20. Scott A, Schurer S, Jensen PH, Sivey P. The effects of an incentive program on quality of care in diabetes management. Health Econ 2009;18:1091e108. 21. Lai CL, Hou YH. The association of clinical guideline adherence and pay-forperformance among patients with diabetes. J Chin Med Assoc 2013;76:102e7. 22. Lee TT, Cheng SH, Chen CC, Lai MS. A pay-for-performance program for diabetes care in Taiwan: A preliminary assessment. Am J Manag Care 2010;16:65e9. 23. Chen JY, Tian H, Taira Juarez D, et al. The effect of a PPO pay-for-performance program on patients with diabetes. Am J Manag Care 2010;16:e11e9. 24. Chen TT, Chung KP, Lin IC, Lai MS. The unintended consequence of diabetes mellitus pay-for-performance (P4P) program in Taiwan: Are patients with more comorbidities or more severe conditions likely to be excluded from the P4P program? Health Serv Res 2011;46(Pt 1):47e60. 25. Kiran T, Victor JC, Kopp A, et al. The relationship between financial incentives and quality of diabetes care in Ontario. Canada. Diabetes Care 2012;35: 1038e46.
87
26. Gillam SJ, Siriwardena AN, Steel N. Pay-for-performance in the United Kingdom: impact of the quality and outcomes framework: A systematic review. Ann Fam Med 2012;10:461e8. 27. Doran T, Fullwood C, Gravelle H, et al. Pay-for-performance programs in family practices in the United Kingdom. N Engl J Med 2006;355:375e84. 28. Doran T, Fullwood C, Reeves D, et al. Exclusion of patients from pay-forperformance targets by English physicians. N Engl J Med 2008;17(359): 274e84. 29. Department of Health. The NHS Improvement Plan: putting people at the heart of public services. London, UK: 2004. 30. Hamilton FL, Bottle A, Vamos EP, et al. Impact of a pay-for-performance incentive scheme on age, sex, and socioeconomic disparities in diabetes management in UK primary care. J Ambul Care Manage 2010;33:336e49. 31. Crawley D, Ng A, Mainous 3rd AG, et al. Impact of pay for performance on quality of chronic disease management by social class group in England. J R Soc Med 2009;102:103e7. 32. Millett C, Netuveli G, Saxena S, Majeed A. Impact of pay for performance on ethnic disparities in intermediate outcomes for diabetes: A longitudinal study. Diabetes Care 2009;32:404e9. 33. Millett C, Gray J, Saxena S, et al. Ethnic disparities in diabetes management and pay-for-performance in the UK: The Wandsworth Prospective Diabetes Study. PLoS Med 2007;4:e191. 34. Dalton AR, Alshamsan R, Majeed A, Millett C. Exclusion of patients from quality measurement of diabetes care in the UK pay-for-performance programme. Diabet Med 2011;28:525e31. 35. Campbell SM, McDonald R, Lester H. The experience of pay for performance in English family practice: A qualitative study. Ann Fam Med 2008;6:228e34. 36. McDonald R, Roland M. Pay for performance in primary care in England and California: comparison of unintended consequences. Ann Fam Med 2009;7: 121e7. 37. McDonald R, Harrison S, Checkland K, et al. Impact of financial incentives on clinical autonomy and internal motivation in primary care: Ethnographic study. BMJ 2007;334:1357. 38. Campbell SM, Kontopantelis E, Reeves D, et al. Changes in patient experiences of primary care during health service reforms in England between 2003 and 2007. Ann Fam Med 2010;8:499e506. 39. Qaseem A, Snow V, Gosfield A, et al. Pay for performance through the lens of medical professionalism. Ann Intern Med 2010;152:366e9. 40. Young GJ, Meterko M, Beckman H, et al. Effects of paying physicians based on their relative performance for quality. J Gen Intern Med 2007;22:872e6. 41. Glasziou PP, Buchan H, Del Mar C, et al. When financial incentives do more good than harm: A checklist. BMJ 2012;345:e5047.
Contents lists available at ScienceDirect
Canadian Journal of Diabetes journal homepage: www.canadianjournalofdiabetes.com
Review
Performance-Based Financial Incentives for Diabetes Care: An Effective Strategy? Lesley P. Latham MSc, Emily Gard Marshall PhD * Dalhousie University, Halifax, Nova Scotia, Canada
a r t i c l e i n f o
a b s t r a c t
Article history: Received 10 December 2013 Received in revised form 10 April 2014 Accepted 11 June 2014
The use of financial incentives provided to primary care physicians who achieve target management or clinical outcomes has been advocated to support the fulfillment of care recommendations for patients with diabetes. This article explores the characteristics of incentive models implemented in the context of universal healthcare systems in the United Kingdom, Australia, Taiwan and Canada; the extent to which these interventions have been successful in improving diabetes outcomes; and the key challenges and concerns around implementing incentive models. Research in the effect of incentives in the United Kingdom demonstrates some improvements in process outcomes and achievement of cholesterol, blood pressure and glycated hemoglobin (A1C) targets. Evidence of the efficacy of programs implemented outside of the United Kingdom is very limited but suggests that physicians participating in these enhanced billing incentive programs were already completing the guideline-recommended care prior to the introduction of the incentive. A shift to a pay-for-performance programs may have important implications for professionalism and patient-centred care. In the absence of definitive evidence that financial incentives drive the quality of diabetes management at the level of primary care, policy makers should proceed with caution. It is important to look beyond simply modifying physicians’ behaviours and address the factors and systemic barriers that make it challenging for patients and physicians to manage diabetes in partnership. Ó 2015 Canadian Diabetes Association
Keywords: diabetes mellitus primary healthcare reimbursement incentive
r é s u m é Mots clés : diabète sucré prime de remboursement soins de santé primaires
L’utilisation d’incitatifs financiers chez les médecins de premier recours qui atteignent leur cible de prise en charge ou les résultats cliniques a été recommandée pour favoriser la réalisation des recommandations pour les soins des patients diabétiques. Cet article explore les caractéristiques des modèles incitatifs mis en place dans le contexte des systèmes de soins de santé universels au Royaume-Uni, en Australie, à Taïwan et au Canada; dans quelle mesure ces interventions ont réussi à améliorer les résultats liés au diabète; et les défis et les préoccupations principales concernant la mise en place des modèles incitatifs. La recherche sur l’effet des mesures incitatives au Royaume-Uni démontre certaines améliorations dans les résultats liés au processus et l’atteinte des valeurs cibles de cholestérol, de pression artérielle et d’hémoglobine glyquée (A1c). Les données probantes sur l’efficacité des programmes mis en place à l’extérieur du Royaume-Uni sont très limitées, mais suggèrent que les médecins participants à ces programmes d’incitatifs à une meilleure performance appliquaient déjà les soins recommandés par les lignes directrices avant l’introduction des mesures incitatives. Le fait d’opter pour les programmes de rémunération en fonction de la performance peut avoir d’importantes répercussions sur le professionnalisme et les soins centrés sur le patient. En l’absence de preuves concluantes sur le fait que les incitatifs financiers promeuvent la qualité de la prise en charge du diabète aux soins primaires, les responsables de l’élaboration des politiques devraient procéder avec prudence. Il est important de dépasser la simple modification des comportements des médecins et de se pencher sur les facteurs et les obstacles systémiques qui font qu’il est difficile pour les patients et les médecins de collaborer ensemble à la prise en charge du diabète. Ó 2015 Canadian Diabetes Association
* Address for correspondence: Emily Gard Marshall, PhD, Dalhousie Family Medicine, AJLB 8 QEII HSC, 5909 Veteran’s Memorial Lane, Halifax, Nova Scotia B3H 2E2, Canada. E-mail address: [email protected] 1499-2671/$ e see front matter Ó 2015 Canadian Diabetes Association http://dx.doi.org/10.1016/j.jcjd.2014.06.002
84
L.P. Latham, E.G. Marshall / Can J Diabetes 39 (2015) 83e87
Introduction Type 2 diabetes is a cause of substantial morbidity and mortality in Canada. Clinical management, according to recommended guidelines at the level of primary care, can improve outcomes for patients with diabetes (1). However, analysis of national health surveys suggests that there is a gap between practices recommended by current guidelines and the care patients with diabetes receive (2). One method of increasing the fulfillment of recommended care is the use of financial incentives provided to primary care physicians who achieve target processes or clinical outcomes in their patients. This approach is sometimes termed pay-for-performance, where performance refers to the meeting of preestablished targets for the delivery of healthcare services. Founded on the presumption that guideline-based care will generate savings by reducing demands on costly hospital-based services over the long term, incentive programs aimed at improving diabetes outcomes are increasing in popularity around the world. In Canada, for example, enhanced billing for completion of diabetes care according to the accepted guidelines has been used in several provinces as part of larger initiatives to improve the quality and outcomes of care for patients who have chronic disease. However, it remains unclear whether these incentives are effective in improving the quality of diabetes care. This article provides a short narrative exploration of the characteristics of incentive models from the United Kingdom, Australia, Taiwan and Canada, the extent to which these interventions have demonstrated success in improving diabetes outcomes, and an overview of key challenges and concerns related to the use of incentives to improve quality of care. Key literature on the topic of performance-based incentives in primary care, including systematic reviews (3,4), have focused on incentives initiated in the United States (US). However, key differences in the way in which US programs are implemented and evaluated do not allow for suitable comparison. Characteristics of Pay-for-Performance Systems The United Kingdom (UK) is home to the most ambitious example of pay-for-performance in diabetes care. In connection with an initiative to improve chronic disease care and outcomes, the UK government introduced a pay-for-performance model called the Quality Outcomes Framework (QOF) as part of the General Practitioner contract in 2004. Under QOF, up to one-quarter of general practice income is linked to performance according to a complex set of quality indicators. Practices can earn up to 1000 points in several domains, including clinical outcomes (661 points); organizational outcomes (262 points); patient experience (33 points) and additional services (44 points). Diabetes indicators account for approximately 14% of QOF clinical domain points (5). Although participation is voluntary, coverage is estimated to be above 99% (5). Under QOF, quality care is defined by the framework developed by Campbell and colleagues (6); it employs process (monitoring, prescribing and counselling); intermediate clinical levels (glycated hemoglobin [A1C], cholesterol and blood pressure); and patientreported indicators (quality of life and satisfaction with care) to evaluate performance. Quality scores are heavily weighted toward clinical and physiologic indicators. It is important to note that patients may be removed from reporting practices (exception reporting) so that physicians are not financially penalized for factors outside of their control (e.g. missed appointments). Australia introduced financial incentives related to quality of care for patients with chronic diseases for primary care providers in 1999 in the form of the Practice Incentive Program (PIP) (7). Under PIP, primary care physicians can elect to participate in a blended
payment model, which includes additional incentives for each patient who completes a cycle of care (described as a complete sequence of visits recommended by clinical guidelines) for asthma, diabetes, mental health or cervical cancer screening. Taiwan introduced a voluntary incentive program (DM-P4P) for diabetes care in 2001. The design by Taiwan’s National Health Insurance initially involved case management fees provided to physicians for achieving process-based outcomes, such as follow up and annual evaluation visits. Service items required for each of these visits include nutrition education, physical examination and laboratory testing (8). Participating physicians are required to complete specialized training in diabetes management (Diabetes Shared Care Program) and are permitted to select which patients are eligible for the program. In Canada, several provinces have introduced incentive programs in the form of enhanced billing or condition-based payments. Physicians can receive additional payments for each patient with diabetes who is managed in accordance with practice guidelines (9). For example, Ontario’s Diabetes Management Incentive provides a $60 annual, per-patient payment to physicians who complete and document the required elements of care recommended by the Canadian Diabetes Association guidelines (10). Family physicians in Nova Scotia are eligible for additional payments through a Chronic Disease Management program if they complete and document assessment of diabetes risk factors and annual monitoring (11). Similarly, in British Columbia, family physicians are offered chronic disease management payments for a number of chronic conditions in accordance with the Full Service Family Practice Condition Payments introduced in 2003. Physicians may receive a $120 payment (per patient) for providing care according to the British Columbia clinical guideline recommendations for diabetes mellitus over the preceding 12 months (12).
Linking Incentives and Quality in Diabetes Care The evidence suggests that in the UK, incentive models have spurred some improvements in process outcomes and achievement of cholesterol, blood pressure and A1C targets (13e15). A retrospective analysis focusing on glycemic control found that between 2006 and 2008, the percentage of adults who achieved A1C levels within the recommended guidelines rose from 37.9% to 52.1% (16). Still, the evidence is mixed, and its interpretation is hampered by methodologic challenges and confounders. For example, broad implementation and uptake of QOF means there is no adequate control group. Interpreting trends in quality improvement is also complicated by the fact that the quality of diabetes care was improving prior to the introduction of QOF in 2004 (17). Several studies have attempted to separate clinical improvements resulting from QOF from background trends. One longitudinal study found that quality of care improved over and above the pre-incentive trend in the first year after the introduction of QOF by about 14.2%. However, during the second and third years, this difference diminished, and there was increasing variability according to patient demographic characteristics (17). An interrupted, time series analysis of a subset of family practices found that the introduction of QOF accelerated improvements in diabetes quality of care between 2003 and 2005, but this trend had slowed by 2007. The same study found that continuity of care was reduced after the introduction of QOF, while no changes in reported access to care were observed (18). With regard to longer term outcomes such as hospitalizations, one ecologic, cross-sectional study comparing hospital admission rates with quality-of-care scores found a 10-fold variation in diabetes-related admissions despite uniformly high-quality scores (19). Notably, in this study, low neighborhood socioeconomic status
Minimal improvements to management outcomes Poor uptake and self-selection for physicians already adhering to management guidelines In selected provinces with limited uptake Process outcomes only (compliance with Canadian Diabetes Association guidelines) Enhanced billing/conditionbased payments Implemented in some provinces (e.g. Ontario’s Diabetes Management Incentive) Canada
Suggested increased adherence to recommended guidelines Reduced hospitalization rates after 3 years for enrolled patients (limited evidence) Nationally with limited uptake Process outcomes (case management incentives) and intermediate clinical outcomes Enhanced billing/conditionbased payments Implemented nationally (DM-P4P) Taiwan
Suggested increased likelihood of physician ordering appropriate monitoring tests (limited evidence) Nationally with limited uptake Process outcomes only (completion of “cycles of care”) Enhanced billing/conditionbased payments Implemented nationally (Practice Incentive Program) Australia
Some improvements in intermediate clinical outcomes (A1C and cholesterol) Trend of improvement in clinical targets levels-off after the first year of the program Nationally with broad coverage/ uptake
Implementation
Complex set of quality indicators (process outcomes, intermediate clinical outcomes, patient-reported indicators) Based on defined framework
Quality outcomes on which incentives are based
Proportion of primary care provider income is linked to quality outcomes Integrated into contract (Quality Outcomes Framework)
Model for provision of financial incentives Country
A shift to pay-for-performance programs may have important implications for professionalism and patient-centred caredeffects that are difficult to measure. For example, a focus on clinical outcomes excludes the complexities of diabetes and social factors that may make it difficult for patients to meet their targets, and it deprioritizes important facets of care, such as continuity and selfmanagement (26). There is a concern that patients who fail to achieve clinical targets may be viewed as noncompliant, or physicians may be subtly influenced to select patients on the basis of ease of management, which may increase health disparities. For example, there was concern following the introduction of QOF that physicians may inappropriately exclude patients from quality measures to avoid being financially penalized. Doran et al originally reported that a minority of practices may be achieving targets by excluding large numbers of patients (27). However, subsequent studies of exception reporting found little evidence of “gaming” (28). It is unclear whether incentives have influenced how physicians select and retain patients outside of the UK. Although it was not specifically designed to do so, the introduction of the QOF was expected to reduce health disparities through standardization of care (29). Although 2 studies suggest the benefits of the pay-for-performance model are similar to those with high and low income and social status (30,31), findings from several recent studies suggest that QOF does not address ethnic disparities in diabetes care adequately (13,32,33). Concerns have also emerged that patients from disadvantaged and vulnerable populations may be disproportionately excepted from QOF because their diabetes may be more challenging to manage (15,34). Dalton
85
UK
Challenges and Concerns Related to the Use of Incentives
Table Summary of the characteristics of performance-based incentive programs for diabetes care and the evidence for their efficacy
was more closely associated with hospitalization rates than were clinical indicators. There are few studies examining the effects of diabetes incentive programs outside of the UK. An Australian retrospective analysis of physician encounters with patients with diabetes concluded that participation in PIP increased the likelihood of a physician’s ordering an A1C test (20). Several cross-sectional studies have indicated that patients enrolled in Taiwan’s incentive program were more likely to receive guideline-recommended tests and examinations (21,22). A longitudinal study of the effect of the Taiwanese program on hospitalization rates found that patients enrolled in the incentive program were less likely to be hospitalized after 3 years of care compared with non-enrolled patients (23). Another longitudinal study found increased cost and health service use by patients enrolled in the first year of the program, owing to increased physician visits and examinations. The authors were unable to determine whether the costs of the program were offset by the savings related to long-term declines in service use (8). Concerns raised about the Taiwanese program include the absence of riskadjusted clinical targets as well as the lack of specific criteria for exclusion of patients. Accordingly, 1 study has demonstrated that older patients and those with higher comorbidity and severity of disease are more likely to be excluded from this program (24). In Ontario, researchers studying completion of recommended diabetes management practices before and after the introduction of the incentive billing code found minimal improvements to monitoring practices (25). In less than 25% of the patients studied, incentive codes had been billed by their physicians. Longitudinal analysis suggested that patients who received the recommended testing following the introduction of the incentive were the same patients who were receiving all recommended tests before it had been introduced. A summary of the characteristics of performance-based incentive programs and the evidence of their efficacy in improving the quality of diabetes care is presented in the Table.
Brief summary of evidence
L.P. Latham, E.G. Marshall / Can J Diabetes 39 (2015) 83e87
86
L.P. Latham, E.G. Marshall / Can J Diabetes 39 (2015) 83e87
et al examined the pattern of exclusions in 23 practices in northwest London (34). They found excepted patients were less likely to achieve treatment targets for A1C, blood pressure and cholesterol than were patients included in QOF. Patients with longstanding diabetes or multiple comorbidities were also more likely to be excluded from the A1C indicator. The same study found that black and South Asian patients were more likely to be excluded from the A1C indicator than were white patients. Financial rewards may encourage inappropriate (and potentially costly) target-oriented care by physicians and may divert attention from addressing patient-centered values and goals of treatment. A qualitative study examining the concerns of physicians and nurses following the introduction of QOF in the UK found that providers felt that incentives and quality measures brought needed attention to appropriate clinical targets (35). There was, however, a concern that the need to address QOF quality measures during office visits took away from the patients’ agendas, effectively making the encounter more physician centred than patient centred (35). Similarly, a qualitative study of UK physicians reported that data collection of performance measures during office visits was potentially distracting and changed the nature of the office visit (36). In another qualitative study conducted in the UK, primary care nurses working in chronic disease management reported feeling as though QOF encouraged regarding patients as conditions, rather than people, although they supported the overall goals of QOF, feeling that its structure helped them to provide high-quality care to patients (37). Despite concerns about distraction from patientcentred care, physicians reported that they did not pursue target incentives if they felt doing so would be detrimental to patients (18,35). Other qualitative studies describe lower patient satisfaction ratings for continuity of care following the introduction of QOF (38). In a system where interprofessional collaboration is actively promoted, selectively rewarding primary care physicians may discourage teamwork and coordinated care with other members of the heathcare team (35,39). For example, although nurses were integral in the care of patients with chronic disease following the introduction of QOF, increased income generated by achieving targets was directed largely to physicians because they were the employers and owners of the practices (35). Conclusion Although we have not fully discussed studies of pay-forperformance from the US, they generally provide similar perspectives on the efficacy of incentive in the improvement of diabetes care. For example, a retrospective cohort study of 334 primary care physicians in New York found no statistically significant improvements in process of care outcomes for diabetes management following the introduction of an incentive program (40). Notably, the implementation of performance-based incentives for diabetes care in the US differs from that in Canada and other countries in important ways. First, incentive programs in the US are typically not government sponsored but are administered through particular health plans or health maintenance organizations. Incentives within this model would be of potential benefit only for patients who had access to private insurance through these plans. Second, rather than providing incentives directly to physicians or practices, incentives are provided to health groups or institutions. Programs are commonly overseen separately by quality-improvement organizations, and data collection is performed by third parties, rather than by the healthcare providers themselves. For these reasons, we have chosen to focus this short review on programs that, although diverse, were relatively similar in context to incentive programs implemented in Canada. In fact, one of the main challenges in interpreting the evidence around financial incentives for the improvement of diabetes care is
the fact that programs have varied characteristics. The importance of the characteristics of the incentive program, such as the nature and magnitude of the incentive, the involvement of physicians in the development process, and the available infrastructure, have been discussed in detail elsewhere (3). In brief, the evidence suggests that a nationally implemented incentive program coupled with integration of pay-for-performance elements into primary care physicians’ salaries may be effective in improving the quality of diabetes care. However, even broadly implemented incentive programs such as QOF have demonstrated mainly the effects on process and intermediate clinical diabetes outcomes. More evidence is required to understand whether these improvements are sustained and translate into better long-term outcomes such as reduced hospitalizations for diabetes-related complications (26). It is important to note that evidence is lacking as to whether the investment required to initiate and maintain these performancebased incentive programs will be offset by downstream savings associated with reduced diabetes-related health service usage and hospitalizations (8,14,19). Many of the possible negative consequences of financial incentives have not been explored in the literature. A systematic method of evaluating the potential benefits and unintended consequences of implementing financial incentives may be of benefit. Glasziou et al have developed a checklist to guide decision making as to where and how financial incentives should be used (41); however, the lack of available evidence concerning the costs, benefits and harms of specific interventions may limit the applicability of such tools. In the absence of definitive evidence that financial incentives drive the quality of diabetes management at the level of primary care, policy makers advocating the use of such incentives in Canada should proceed with caution. A thoughtful and holistic approach to diabetes and chronic disease heath policy will look beyond simply modifying physician behaviours and will address the factors and systemic barriers that make it challenging for patients and physicians to manage diabetes in partnership. Improvements could include providing adequate billing codes to care appropriately for patients with complex issues; avoiding one-issue-only practices; addressing accessibility to primary care so that patients can develop longer-term relationships with providers, thus allowing for continuity and encouraging more interprofessional collaborative care models to help patients change lifestyle risk factors; and addressing chronic disease prevention in a more proactive way in primary care. Competing Interests None.
References 1. Dagogo-Jack S. Preventing diabetes-related morbidity and mortality in the primary care setting. J Natl Med Assoc 2002;94:549e60. 2. Canadian Institute for Health Information. Diabetes care gaps and disparities in Canada. December, 2009. 3. Van Herck P, De Smedt D, Annemans L, et al. Systematic review: Effects, design choices, and context of pay-for-performance in health care. BMC Health Serv Res 2010;10:247. 4. Scott A, Sivey P, Ait Ouakrim D, et al. The effect of financial incentives on the quality of health care provided by primary care physicians. Cochrane Database Syst Rev 2011;7(9):CD008451. 5. Health and Social Care Information Centre, Prescribing and Primary Care Team., Quality and outcomes framework: Achievement, prevalence and exceptions data 2011/12. Author; 2012. 6. Campbell SM, Roland MO, Buetow SA. Defining quality of care. Soc Sci Med 2000;51:1611e25. 7. Scott A, Naccarella L, Furler J, et al. Financial incentives and the quality of primary care in Australia Report. Australian Primary Health Care Research Institute (APHCRI). 8. Cheng SH, Lee TT, Chen CC. A longitudinal examination of a pay-forperformance program for diabetes care: Evidence from a natural experiment. Med Care 2012;50:109e16.
L.P. Latham, E.G. Marshall / Can J Diabetes 39 (2015) 83e87 9. The Canadian Diabetes Association. Diabetes: Canada at the tipping point: charting a new path. Author; 2010. 10. Government of Ontario. Guide to physician compensation. 2009. 11. Nova Scotia Medical Services Insurance. Physician’s Manual. 2012. 12. General Practice Services Committee. Chronic disease management incentives billing guide. 2010. 13. Alshamsan R, Lee JT, Majeed A, et al. Effect of a UK pay-for-performance program on ethnic disparities in diabetes outcomes: Interrupted time series analysis. Ann Fam Med 2012;10:228e34. 14. Alshamsan R, Millett C, Majeed A, Khunti K. Has pay for performance improved the management of diabetes in the United Kingdom? Prim Care Diabetes 2010; 4:73e8. 15. Calvert M, Shankar A, McManus RJ, et al. Effect of the quality and outcomes framework on diabetes care in the United Kingdom: Retrospective cohort study. BMJ 2009;338:b1870. 16. Oluwatowoju I, Abu E, Wild SH, Byrne CD. Improvements in glycaemic control and cholesterol concentrations associated with the Quality and Outcomes Framework: A regional 2-year audit of diabetes care in the UK. Diabet Med 2010;27:354e9. 17. Kontopantelis E, Reeves D, Valderas JM, et al. Recorded quality of primary care for patients with diabetes in England before and after the introduction of a financial incentive scheme: A longitudinal observational study. BMJ Qual Saf 2013;22:53e64. 18. Campbell SM, Reeves D, Kontopantelis E, et al. Effects of pay for performance on the quality of primary care in England. N Engl J Med 2009;23(361):368e78. 19. Bottle A, Millett C, Xie Y, et al. Quality of primary care and hospital admissions for diabetes mellitus in England. J Ambul Care Manage 2008;31:226e38. 20. Scott A, Schurer S, Jensen PH, Sivey P. The effects of an incentive program on quality of care in diabetes management. Health Econ 2009;18:1091e108. 21. Lai CL, Hou YH. The association of clinical guideline adherence and pay-forperformance among patients with diabetes. J Chin Med Assoc 2013;76:102e7. 22. Lee TT, Cheng SH, Chen CC, Lai MS. A pay-for-performance program for diabetes care in Taiwan: A preliminary assessment. Am J Manag Care 2010;16:65e9. 23. Chen JY, Tian H, Taira Juarez D, et al. The effect of a PPO pay-for-performance program on patients with diabetes. Am J Manag Care 2010;16:e11e9. 24. Chen TT, Chung KP, Lin IC, Lai MS. The unintended consequence of diabetes mellitus pay-for-performance (P4P) program in Taiwan: Are patients with more comorbidities or more severe conditions likely to be excluded from the P4P program? Health Serv Res 2011;46(Pt 1):47e60. 25. Kiran T, Victor JC, Kopp A, et al. The relationship between financial incentives and quality of diabetes care in Ontario. Canada. Diabetes Care 2012;35: 1038e46.
87
26. Gillam SJ, Siriwardena AN, Steel N. Pay-for-performance in the United Kingdom: impact of the quality and outcomes framework: A systematic review. Ann Fam Med 2012;10:461e8. 27. Doran T, Fullwood C, Gravelle H, et al. Pay-for-performance programs in family practices in the United Kingdom. N Engl J Med 2006;355:375e84. 28. Doran T, Fullwood C, Reeves D, et al. Exclusion of patients from pay-forperformance targets by English physicians. N Engl J Med 2008;17(359): 274e84. 29. Department of Health. The NHS Improvement Plan: putting people at the heart of public services. London, UK: 2004. 30. Hamilton FL, Bottle A, Vamos EP, et al. Impact of a pay-for-performance incentive scheme on age, sex, and socioeconomic disparities in diabetes management in UK primary care. J Ambul Care Manage 2010;33:336e49. 31. Crawley D, Ng A, Mainous 3rd AG, et al. Impact of pay for performance on quality of chronic disease management by social class group in England. J R Soc Med 2009;102:103e7. 32. Millett C, Netuveli G, Saxena S, Majeed A. Impact of pay for performance on ethnic disparities in intermediate outcomes for diabetes: A longitudinal study. Diabetes Care 2009;32:404e9. 33. Millett C, Gray J, Saxena S, et al. Ethnic disparities in diabetes management and pay-for-performance in the UK: The Wandsworth Prospective Diabetes Study. PLoS Med 2007;4:e191. 34. Dalton AR, Alshamsan R, Majeed A, Millett C. Exclusion of patients from quality measurement of diabetes care in the UK pay-for-performance programme. Diabet Med 2011;28:525e31. 35. Campbell SM, McDonald R, Lester H. The experience of pay for performance in English family practice: A qualitative study. Ann Fam Med 2008;6:228e34. 36. McDonald R, Roland M. Pay for performance in primary care in England and California: comparison of unintended consequences. Ann Fam Med 2009;7: 121e7. 37. McDonald R, Harrison S, Checkland K, et al. Impact of financial incentives on clinical autonomy and internal motivation in primary care: Ethnographic study. BMJ 2007;334:1357. 38. Campbell SM, Kontopantelis E, Reeves D, et al. Changes in patient experiences of primary care during health service reforms in England between 2003 and 2007. Ann Fam Med 2010;8:499e506. 39. Qaseem A, Snow V, Gosfield A, et al. Pay for performance through the lens of medical professionalism. Ann Intern Med 2010;152:366e9. 40. Young GJ, Meterko M, Beckman H, et al. Effects of paying physicians based on their relative performance for quality. J Gen Intern Med 2007;22:872e6. 41. Glasziou PP, Buchan H, Del Mar C, et al. When financial incentives do more good than harm: A checklist. BMJ 2012;345:e5047.
