Paying doctors by salary: a controlled study of general practitioner behaviour in England
Introduction
The majority of general practitioners (GPs) in England are ‘independent contractors’ who supply general medical services (GMS) to the national health service (NHS) under a standard national contract. Remuneration is based mainly on capitation supplemented by allowances, fees for items of service, and payments for achieving target levels of service provision (called target payments). The NHS (Primary Care) Act of 1997 introduced a new voluntary scheme, which freed GPs from the constraints of the standard contractual arrangements. Instead, local health service organisations (mostly general practices and community NHS trusts) are able to propose local service contracts designed to meet the needs of local populations. GPs can be employed on a salaried basis to provide ‘personal medical services’ (PMS) alongside other services needed in the locality. These salaried schemes were intended to improve GP recruitment and retention, and the quality of care particularly in under served areas [1].
By May 1999, of the original 87 first wave PMS pilots that went live in England in April 1998 there were 46 sites (54%) with a salaried GP, although the situation continued to fluctuate, with salaried sites and posts being added or dropped [1].
Predicting the outcome of this experimental introduction of salaried contracts is complex since there has been no empirical work comparing GPs’ working practices under salaried and mixed capitation payment in the UK and very little from other countries [2], [3]. Cross sectional studies comparing salary and capitation systems have found that consultations are longer under salary payment [4], [5] but there are few other differences [6]. However, this evidence may be misleading to policy-makers because of the problems of generalising from one country to another and determining causality from cross sectional study designs. Also there are few quantitative studies that directly compare salary and capitation payment systems in terms of their relative impacts on the quality of care. We took the opportunity offered by this policy change in the UK to carry out a controlled study of the effects of salaried payment on GP working practices and quality of care to address this gap in knowledge.
Standard contract GPs (GMS) are financially rewarded for increasing patient list size (through capitation payments) and for providing specific types of services (through target payments). In the absence of such incentives, we would expect salaried GPs to reduce patient list size and provide fewer of the targeted services. Salaried GPs may also have the opportunity to reduce their out-of-hours responsibilities [7]. Salaried contracts in PMS pilots specifically free GPs from administrative responsibilities so increasing the time available for direct patient care. This, together with lower list sizes, might be expected to improve access to GP services and longer consultations may lead to better technical, interpersonal and communication aspects of care, thereby resulting in an improvement in the overall quality of patient care [8], [9].
Section snippets
Methods
We used a controlled before-and-after study design to evaluate the effects of salary payment because the voluntary nature of the PMS initiative precluded the use of randomisation. From our study group of 46 salaried pilot sites we sampled ten practices in which GPs had switched from standard GMS contracts to salaried contracts without moving practice (PMS practices). A sample of ten practices owned and staffed by standard contract GPs (GMS practices) was matched to the PMS practices in terms of
Results
By design, most of the GMS and PMS practice pairs were well matched at the outset in terms of numbers of GPs and deprivation in the patient population (Table 2). Over the course of the study, however, doctors left and joined practices, locums were sometimes appointed, and list sizes changed. Moreover doctors in GMS and PMS practices differed in their age and gender distributions, which also changed over time largely due to attrition (Table 2). All but one salaried GP and two control group GPs
Discussion
The findings of this study were only partly consistent with our expectations. Many of the changes in GP behaviour, following the move to salaried status, were small and statistically insignificant. List sizes increased at a lower rate in PMS as compared with GMS practices but, by virtue of their staffing policies, PMS practices ended with higher list sizes per WTE GP. Doctors in PMS practices spent more time doing out-of-hours work and on direct patient care but less time on practice
Acknowledgements
We would like to thank all the GPs and practice staff involved in this study for their contribution and support. We are grateful to Professor Douglas Fleming at the Royal College of General Practitioners Research Unit in Birmingham for his help in designing the workload diary, and Dr Aneez Esmail for helping to pilot the diary. Thanks also go to Sylvia Wright, Dianne Oliver, Michele Bohan at NPCRDC, for their help in collecting the data. Thanks must also go to colleagues at the University of
References (18)
- et al.
Defining quality of care
Social Science and Medicine
(2000) - et al.
Re-organising primary medical care in Finland: the personal doctor program
Health Policy
(1989) - et al.
Setting out. Piloting innovations in primary care
(1999) - et al.
How should we pay doctors? A systematic review of salary payments and their effect on doctor behaviour
Quarterly Journal of Medicine
(1999) - et al.
Capitation, salaried, fee for service and mixed systems of payment: effects on the behaviour of Primary Care Physicians
- et al.
Spending time with patients. The impact of organizational structure on medical practice
Medical Care
(1982) - et al.
The organization of medical practice and primary care physician income
American Journal of Public Health
(1983) Workload and provision of care in general practice
(1998)- et al.
A profile of PMS salaried GP contracts and their impact on recruitment
Family Practice
(2001)
Cited by (40)
Do treatment decisions depend on physicians’ financial incentives?
2017, Journal of Public EconomicsCitation Excerpt :A higher supply of medical services under fee-for-service is also confirmed experimentally by Hennig-Schmidt et al. (2011). On the other hand, Sarma et al. (2010) find no effect of fee-for-service on total hours worked by physicians in Canada, and, based on a field experiment in the UK, Gosden et al. (2003) find little or no significant effects of salaried contracts versus fee-for-service. This literature, and our paper, also relates more broadly to the huge literature on supplier-induced demand (SID).
Rapid HIV test in family practice
2015, Medecine et Maladies InfectieusesCitation Excerpt :Changing the payment mode could improve the feasibility of screening activities. Compensation by salary could be associated to longer consultations allowing the family physician to better adapt his care management [18,19,20]. Another possibility would be to specifically pay the physician for screening [21].
Research on financing and distribution of family physician contracted service fees and related issues
2022, Chinese Journal of General PractitionersPayment methods for healthcare providers working in outpatient healthcare settings
2021, Cochrane Database of Systematic ReviewsRationalizing health personnel financing schemes for evidence-informed policy reforms: Policy analysis
2020, Acta Medica Philippina