A Systematic Review of the Effect of Payment Mechanisms on Family Physicians Service Provision and Referral Rate Behavior

Health care delivery through family medicine and primary care is undergoing reforms in many countries around the world. Family physicians have a considerable role in health system as gatekeepers [1, 2]. In fact, some evidence shows that the lack of access to family physicians not only adversely affects the health and well-being of the population but also have substantial cost to the society and patients. Conversely, universal access to family physician services is a crucial element of an efficient and effective publicly funded health care system [3, 4]. Health care policy makers generally decide about how much and through which method healthcare providers should be paid and thus policy makers have the leverage to create powerful incentives that can influence the actions of organizations and individuals within the health care system [4, 5]. There is an emerging interest in using financial incentives. Several countries are in the process of redesigning their health care systems in response to current and expected future needs of their population, and long-term allocation of limited resources. Changing payment method for health care providers is one of the important interventions that are being studied to improve health systems outcomes and to achieve health policy objectives [6, 7]. Payment system includes all payment mechanisms, such as contracting, accountability mechanisms that accompany the payment method and management information systems [8]. Typically, payment schemes can be classified according to the method used to reimburse healthcare providers such as fee-for-service (FFS), capitation, and salary systems. In FFS system, healthcare providers are reimbursed for each service provided. In capitation system, the payment for all services is bundled depending on diagnosis/procedure. Capitation system also includes payments for providing comprehensive care to a patient throughout a defined period of time irrespective of the amount and intensity of services rendered [6, 9, 10]. On the other hand, because of wide medical knowledge gaps between physicians and patients, physicians can positively or negatively manage and affect user’s health service demands [5]. Empirical evidence consistently shows that financial incentives are one of the most important incentives that influence family physician behavior [11]. Based on the incentive schemes, salaried and capitation-based paid physician may respond to incentives by reducing costs and by under-treatment of patients whereas FFS physicians are incentivized for over-treatment [12]. Hickson (1987) found that physicians at FFS payment system see more patients than physicians paid through salary [13]. The effects of financial incentives depend directly on the structure, socioeconomic and cultural context of the health care system [14]. Experience gained and results obtained with financial incentives in one country and may not be implementable in another country and may require country-specific modifications [15]. Physician and hospital reimbursement methods have been the subject of much debate over the past many years [16]. Significant theoretical and empirical studies that examined the effect of contract mechanism on physician’s behavior are rare and show variable results [17]. Thereby, the purpose of this literature review is to examine the impact of different methods of payments to family physicians and general practitioners on the quantity of the services provided and on the referral behavior.

Data sources and search strategies: We performed a systematic literature search using five online electronic databases: PubMed, Science Direct and Emerald, Wiley Inter Science, Springer Link and ANNFAMMED.  We selected articles that addressed financial-incentive programs. We also reviewed the references of the final selected articles to identify articles that might have been missed in electronic database search. To search articles for review, we used English and Persian language and three themes that were connected with Boolean connectors: (payment, remuneration, reimbursement, compensation method), (service provision, service production, referral rate, referral to specialist) and (family physician, general physician, GP, general practitioner). In some databases, search was performed with methodological filters according to the method of study. Each full-text article was reviewed by two reviewers independently and disagreements between reviewers were resolved by mutual consensus.

Selection criteria: Articles were included if:

• Published between 1985 to2011
• Assessed the effect of three basic types of payment mechanisms (salary, FFS, capitation) on physician behavior
• Addressed the confounding factors (by adjusting for these factors)
• Had adequate response rate (at least 60%)
• Used valid data sources
• Reported quantitative results, effects, or impacts of payment mechanism on family physician program and general practitioners behavior

Study design:

• Prospective or retrospective cohort studies
• Randomized controlled trials
• Controlled before-and-after studies
• Comparative studies

The majority of the included studies were cohort studies due to the lack of randomized controlled trial studies. We excluded program evaluation studies if they attempted to increase or decrease the number of patient referrals to specialist and rate of service production. Reviews, commentaries, editorials, news and policy briefs were also excluded. Finally, 11 articles that matched with our inclusion criteria were included in our systematic review. We extracted data from selected articles and performed qualitative analyses for the type of intervention. If there were any data missing from a study, this was explicitly stated. Due to different study methods, settings, and objectives, quantitative comparison and pooling of the study results was not possible.

Of the 11 studies included in the systematic review, 3 were from Norway, 3 from Canada and rest were from the United States, England, South Africa, Denmark and Uruguay. Majority of the studies were prospective or retrospective cohort studies.

Six articles compared FFS method with salary, 4 articles compared FFS with capitation and one of them compared salary with capitation/FFS payment methods on physician behavior. The articles retrieved from the initial search from the 5 databases are included in Table 1. As Table 2 shows, in Sorensen study that compared FFS with salary payment, physicians with a FFS contract produced a higher number of consultations and other patient contacts than physicians with a fixed salary. This difference was mostly due to longer working hours, but time efficiency is greater as well. Moreover, a part of the difference is due to a selection effect: salaried physicians prefer shorter working hours and prefer to work less intensively. Kristiansen et al found that doctors paid on a FFS basis tended to choose home visits more often than salaried doctors and study indicated that financial incentives may be used to change behavior and encourage home visiting. Similarly, Krasnik et al conclude that introducing a partial FFS system seemed to stimulate the provision of services by general practitioners, resulting in reduced referral rates. Another study found a higher probability of caesarian section in women without risk factors who were treated in private hospitals (25%) than women in public hospitals (11) and authors concluded that the remuneration system explained an important part of this difference. Godsen et al compared salary payment and FFS and found that salaried GPs tended to provide shorter consultations compared with standard contract GPs, prescribed fewer consultations, but referral rates were similar. Lee et al conducted a retrospective study and found that none of the variables showed any statistically significant association between patients who were treated in one or the other model. Budget et al compared FFS and capitation payment system and showed the average number of physician visits was similar for both groups (4.47/year in the capitated program; 5.09/year in the FFS system). However, the average number of hospital admissions per recipient (0.11 versus 0.22 per year), and average

number of hospital days per 1,000 recipients (461 versus 909 per year) were 5% to 60% lower in the capitated group than in the FFS group. Broomberg et al also demonstrated that providers working in the FFS system are likely to increase the supply of services compared with providers who are salaried.

As this literature review shows, each payment method for family physicians and health professionals creates a particular set of risks and incentive for physicians and they respond to the different payment methods in a different but predictable way. The type of payment method has both positive and negative effects on service provision and referral rate behavior of providers. Our study indicates that salary payment method is associated with low service provision and higher referral rate compared with fee-for-service and capitation methods, because in this method physicians receive wage regardless of the number of provided services [21, 25, 28]. Godsen in his recent study about the effect of salary payment method on physician behavior showed similar results; authors found an association between salary payments and reduced number of services per patient, reduced volume of patients per physician and greater degrees of preventive care compared to fee-for-service [31]. Studies included in our review show that FFS payment method cause higher service production and induced more services to the patient. In comparison to salary and capitation, FFS leads to lower referral rate because physicians may want to treat patients and increase their incomes by producing more services [24, 28, 25, 21]. Grytten published his results in 2001 that were inconsistent with some of the more recent literature in this field. Authors concluded that Norwegian physicians prefer the professional norms to financial mechanisms [20]. Sørensen et al (2003) indicated that physicians paid on a FFS basis produced a higher number of visits, other patient contacts and diagnostic services than salaried physicians, and concluded that a change in physician payment schemes from salary to FFS would increase service production in the range of 20–40% [21]. Due to the knowledge gap between physicians and patients, physicians can positively or negatively affect demand by advising patients on when to come back for another visit, what drugs to take, what specialists to see, and what laboratory tests or surgical treatments to undergo [5]. Compared with FFS, capitation payment decreased service production due to budget limitations [17, 26, 30]. Our study shows that referral rate to hospital and specialist increased in capitation compared with FFS. Some studies examined the effect of capitation payment on provider behavior. Iversen et al (2000) evaluated the impact of capitation on Norwegian GPs’ referral decisions. They found a 42% increase in the rate of referral from general practitioners to specialists after Norway introduced capitation based remunerating system [32]. As countries are governed within highly variable economic, cultural, political, demographic, and epidemiological contexts, forming a firm conclusion is difficult. One difficulty in interpreting the effects of financial incentives is  that other factor can also affect health objectives hereby the analysis of the effect of financial incentives cannot be separated from the general context of health care financing system [29]. In order to use available evidence on the effect of financial incentives on physician behavior, future studies should evaluate programs from a more diverse set of countries, in particular from developing countries. In these studies, researchers should attempt to control selection biases as rigorously as possible, using selection models in observational studies and randomized controlled trials where fund providers and policy makers are willing to support such experiments [34]. From the policy maker’s point of view, financial incentives are not sufficient to alleviate the family physician system problems. Broader changes in organizational structures, educational systems, and policy must be combined with changes in remuneration methods to improve physician service provision, collaboration, care continuity, and care quality. We cannot expect to rely on payment method per se to resolve health policy problems that require system-wide approaches [35]. In addition, to make the payment system effective, development of an appropriate information system such as electronic medical records is essential for monitoring data as a supporting system [35]. John and colleagues published a book in 2009 about designing provider payment system and indicated that payment systems can help to achieve health policy objectives by encouraging access to necessary health services for patients, high quality of care, and improved equity, while promoting the effective and efficient use of resources and cost containment [8]. The actual impact of payment mechanisms on health care system depends on the context of the system and the combination with other relevant control nubs. Non-financial incentives are social control mechanisms, such as performance monitoring, peer review, practice ownership, education, professional ethics and audits, all affect physicians behavior. These factors can mitigate or even reverse purely financial incentive mechanisms [9, 36]. Non-financial incentives are largely created by the organizational structures of health care. Decreases in professional flexibility, monitoring and regulation all act as motivators towards provision of appropriate care, but might be met with resentment from physicians [33].

We believe while nations act differently due to their different health system goals and context, international experience suggests some conditional guidance for policy makers and they must also have a system for ensuring that payment is in compliance with local context such as laws, regulations and tax system.

We are thankful to Dr. Tabrizi and Dr. Janati for their comments and support. We are also thankful to Dr. Pour Asghar for conducting the literature searches.

1. Roberts RG, Hunt VR, Kulie TI, Schmidt W, Schirmer JM, Villanueva T, Wilson CR. Family medicine training the international experience. Med J Aust. 2011; 194:S84-87.
2. Katić M, Juresa V, Bergman-Marković B, Jurković D, Predavec S, Hrastinski M, et al. Preventive work in family medicine–proactive approach. Acta Med Croatica. 2010; 64:443-452.
3. Glazier RH, Moineddin, R, Agha MM, Zagorski B, Hall R, Manuel DG, et al. The Impact of Not Having a Primary Care Physician among People with Chronic Conditions. ICES Investigative Report. Toronto: Institute for Clinical Evaluative Sciences; 2008.
4. Jegers M, Kesteloot K, De Graeve D, Gilles W. A typology for provider payment systems in health care. Health Policy 2002; 60:255–273.
5. Roberts M J, Hsiao  W, Berman P, Reich M R. Getting Health Reform Right: A Guide to Improving Performance and Equity. 4rd ed. USA: Oxford University Press; 2004.
6. Richard B, Melanie H. Note on Physician Compensation and Financial Incentives. Cambridge: Harvard Business Publishing; 1999.
7. Armour BS, Pitts MM, Maclean C, Cangialose M, Kishel M, Imai H, Etchason J. The effect of explicit financial incentives on physician behavior. Arch Intern Med. 2001; 161:1261–1266.
8. John C L, Cheryl C, Sheila O. Designing and implementing health care provider payment systems: how-to manuals. Washington: The World Bank; Volume 434. 2009.
9. Delnoij DMJ, De Bakker DH, Groenewegen PP. Getting what one deserves: a comparison of age-differentiated capitation and a mixed system of remuneration for general practitioners. Medisch Contact. 1994; 49:1109-1111.
10. Anderson G. Implementing practice guideline. Can Med Assoc J. 1993; 148:753-755. Cutler DM, Richard JZ. The Anatomy of Health Insurance (Chapter 5). Handbook of Health Economics, Volume I. Amsterdam: Elsevier, 563-643.

12. Woodward RS, Warren –Boulton F. Considering the effects of financial incentives and professional ethics on ‘appropriate’ medical care. J Health Econ. 1984; 3:223-237.
13. Hickson G B, Altemeier W A, Perrin J M. Physician Reimbursement by Salary of Fee-For-Service: Effect on Physician Practice Behavior in a Randomized Prospective Study. Pediatrics 1987; 80:344-350.
14. Pablo G, George S. Health Financing Revisited a Practitioner’s Guide. Washington: The World Bank; 2006.
15. Giuffrida A, Torgerson DJ. Should we pay the patient? Review of financial incentives to enhance patient compliance. Br Med J 1997; 315: 703–707.
16. Sandier S. Health services utilization, physician income trends. In: OECD social services, Health care systems in transition—the search for efficiency. Paris: OECD Press, 1990; 41–56.
17. Hutchison B, Birch S, Hurley J, Lomas J, Stratford-Devai F. Do physician-payment mechanisms affect hospital utilization? A study of Health Service Organizations in Ontario. CMAJ 1996;154:653-661.
18. Ivar S K, Knut H. Effect of the remuneration system on the general practitioner’s choice between surgery consultations and home visits. J Epidemiol Community Health 1993; 47:481-484.
19. Krasnik A, Groenewegen PP, Pedersen PA, von Scholten P, Mooney G, Gottschau A, et al. Changing remuneration systems: effects on activity in general practice. BMJ 1990; 300:698-701.
20. Jostein G, Rune S. Type of contract and supplier-induced demand for primary physicians in Norway. J Health Econ. 2001; 20:379–393.
21. Sørensen RJ, Grytten J. Service production and contract choice in primary physician services. Health Policy 2003; 66:73–93.
22. Gosden T, Sibbald B, Williams J, Petchey R, Leese B. Paying doctors by salary: a controlled study of general practitioner behaviour in England. Health Policy 2003; 64:415-423.
23. Kristiansen I, Holtedahl K. The effect of the remuneration system on the general practitioner’s choice between surgery consultations and home visits. J Epidemiol Community Health 1993; 47:1–4.
24. Lee S, Cowie S, Slobodian P. Payment by salary or fee-for-service Effect on health care resource use in the last year of life. Can Fam Physician 1999; 45:2091-2096.
25. Broomberg J, Price MR. The impact of the fee-for-service reimbursement system on the utilisation of health services. Atr Med J 1990; 78:133-136.
26. Badgett JT, Rabalais GP. Prepaid capitation versus fee-for-service reimbursement in a Medicaid population. Am J Manag Care. 1997; 3:277-282.
27. Devlin RA, Sarma S, Hogg W. Remunerating primary care physicians: emerging directions and policy options for Canada. Healthc Q. 2006; 9(3):34-42.
28. Patricia T, Máximo R. The effect of physicians’ remuneration system on the Caesarean section rate: the Uruguayan case. Int J Health Care Finance Econ 2009; 9:333–345.
29. Chaix-Couturier C, Durand-Zaleski I, Jolly D, Durieux P. Effects of financial incentives on medical practice: results from a systematic review of the literature and methodological issues. Int J Qual Health Care 2000; 12:133-142. Glazier RH, Klein-Geltink J, Kopp A, Sibley LM. Capitation and enhanced fee-for-service models or primary care reform: a population-based evaluation. CMAJ. 2009; 180:E72-81.

31. Gosden T, Forland F, Kristiansen IS, Sutton M, Leese B, Giuffrida A, et al. Impact of Payment Method on Behaviour of Primary Care Physicians: A Systematic Review. J Health Serv Res Policy. 2001; 6:44-55.
32. Iversen, T, Hilde L. The Effect of Capitation on GPs’ Referral Decisions. Health Economics. 2000; 9:199-210.
33. Scott A. Eliciting GPs’ preferences for pecuniary and non-pecuniary job characteristics. J Health Econ. 2001; 20:329-347.
34. Barnighausen T, Bloom DE. Financial incentives for return of service in underserved areas: a systematic review. BMC Health Serv Res. 2009; 29:86.
35. Wranik DW, Durier-Copp M. Physician remuneration methods for family physicians in Canada: expected outcomes and lessons learned. Health Care Anal.2010; 18:35-59.
36. Robinson JC. Theory and practice in the design of physician payment incentives. Milbank Q. 2001; 79:149–77.
37. Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA Group. Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. Ann Intern Med.2009; 151:264-9, W64.