In this essay, Richard Disney analyses the economic incentives offered to GPs over where and how to work, and uses economic theory to propose improvements.
The quality of provision of medical services by GPs has become an issue, highlighted by COVID and the expansion of online consultation rather than face-to-face appointments. There have also been suggestions that, over the longer term, patients have suffered from a reduction in the number of GPs and in the hours that they work. To understand the factors underlying provision of GP services, however, it is important to understand how GP services are provided. Like other private businesses, GPs have economic incentives to offer services in a particular way and, like most people, offer those services with a combination of maximising their economic ‘utility’ (i.e. a combination of income and leisure) and altruistic intent.
Most of the workforce in the NHS is salaried and works to the terms of employment contracts which specify salaries, hours and terms of service. For example, many NHS professional employees are paid according to the Agenda for Change system, by which workers are allocated to pay grades according to the nature and skill of the tasks they undertake, as determined by an explicit job evaluation scheme.
Payment of GPs is entirely different. From the early-Twentieth Century, GPs were independent contractors to local insurance committees. In order to treat illness, GPs were paid a sum based on a fixed capitation fee multiplied by the number of their potential patients covered by each practice. With the introduction of the National Health Service from 1946, proposals to make GPs salaried professional staff within the NHS were rejected by doctors’ representatives. Thereafter, until 2004, each doctor had a contract with the local NHS to provide health services to the patients registered to that individual doctor.
In 2004, a new General Medical Services contract was introduced. The aim of this contract was to incentivise GPs to improve the quality of their service. In exchange GPs had the potential to increase their income significantly. Each GP practice would now receive a basic sum of money according to a formula which adjusted the capitation fee for the age and sex-mix of their patients, whether the practice was rural and an adjustment for “market forces” meaning the cost of employing ancillary staff. (That this formula has become somewhat outdated in the light of demographic changes is a pertinent issue but is not discussed here.) However, the major innovation of the 2004 contract was the introduction of the Quality and Outcomes Framework (QOF) by which practices could receive extra money by achieving certain targets.
The QOF initially contained 146 indicators, for example concerning health checks and treatment of heart problems, high blood pressure, asthma and so on, as well as criteria such as patient satisfaction. Achieving each indicator rewarded the practice with points on which additional income could be obtained. A typical target indicator (for example, treating high blood pressure) would have a lower threshold (above 0%) and an upper threshold (typically well below 100% and between 50% and 90%) of the eligible patients treated between which additional income-generating points could be awarded. Below and above the threshold, no extra income would be received. Over time, the indicators have changed and the points structure amended; for example in the recent QOF, vaccination rates have achieved a higher priority and other targets have been downgraded.
In addition to the intention to increase productivity, the QOF was designed to increase the attractiveness of GP work, reflecting concern over a falling number of GPs. Indeed, estimates suggest that, as a result of QOF, there was an immediate jump in GPs’ incomes of some 25-30%. However, as any microeconomist would note, a system of this kind has both an incentive effect and a wealth effect – the former inducing a greater supply of GPs, the latter a reduction in hours of work. Both outcomes occurred. In addition, the introduction of QOF accelerated a trend towards a greater variety of GP modes of provision: from traditional single practitioners and partnerships towards private profit-maximising companies which employed GPs as salaried professionals. It should be noted, however, that whatever the mode of GP provision, the same GP contract applied.
What is the impact of incentivising GPs practices in this way? The evidence strongly suggests that, in exchange for increased income, higher levels of targets were achieved; indeed the success rate seems to have exceeded expectations in the Department of Health and Social Care with a consequent greater cost to the Treasury than was anticipated. However there are also side-effects of incentives structures of this kind. The first in this context is that QOF potentially diverts activity towards areas of provision for which there are indicator points to be gained at the expense of activities for which there are no points to be gained. There appears to be no strong evidence that this took place, although data are limited. However, to avoid this potential problem, constant tinkering with indicators has taken place to ensure that official public health priorities are retained.
A second possible side-effect concerns ratio-based performance targets. Higher payments can be achieved by treating a greater number of eligible patients (the numerator) or by reducing the number of untreated eligible patients (the denominator). This is because some eligible patients, although retained in the practice, could be treated as “exception reported” and deducted from the base. There is a long list of potential “exceptions” for most treatments which ranges from the terminally ill through to those who either refused treatment or were uncontactable. And evidence suggests that some GP practices did indeed achieve the target indicators by adjusting the denominator as well as (or instead of) the numerator.
To give an example of how this incentive works, consider a vaccination programme where doctors are rewarded for vaccinating X% of their patients. Some obvious examples of ‘exception reported’ patients are those in hospital or ‘anti-vaxxers’, but it is apparent that other patients may pose difficulties (being housebound, speak poor English etc) which would tempt a GP to focus on the easier-to-vaccinate until the target is achieved.
At this juncture, it is important to note the evidence that some – perhaps most – GP practices appear to mix these economic responses to incentives with altruistic motives, for example in their willingness to treat eligible patients for conditions above and beyond the maximum threshold for extra payments. Indeed, in the eyes of the public, this is probably what GPs should do, whatever the payment framework in place. And it should be noted that, in other spheres of economic activity, strict economic incentives are not the only motivating factor. For example, the Nobel Prize-winning economist George Akerlof found evidence that, in a factory paying time rates to employees for producing a fixed number of outputs, some employees substantially exceeded their production quota without extra payment. People, including GPs, are motivated by a range of factors.
An issue that has particularly concerned patients is the closure of GPs surgeries at weekends. It is argued that many patients thus had to attend A&E departments of hospitals, exacerbating the already-overcrowded weekends there. And an implication is that the introduction of QOF increased the opportunity for leisure by GPs, which took the form of closing down face-to-face provision at weekends. In the 2012-13 General Medical Services (GMS) revised contract, new QOF indicators were introduced which incentivised GPs to cut down on visits to A&E by patients within the practice. Research in the immediate aftermath of this questioned whether this incentive had any effect, suggesting that the incidence of A&E visits by patients across GP practices was largely driven by variations in their demographic composition. However in 2013 a “7 day opening pilot” was introduced by Clinical Commissioning Groups (CCGs) in Central London for some GPs practices, using extra resources provided from the Prime Minister’s Challenge Fund to fund weekend opening. The object was to investigate whether patients of GPs participating in this scheme were less likely to use A&E than those of non-participating practices. The result was a 9.9% drop in A&E attendance among participants, rising to 17.9% at weekends. Clearly the opening times of GPs do affect A&E attendance.
A final economic issue of relevance to GPs’ economic behaviour is competition. Do practices that are in direct competition with one another, due to spatial proximity, perform better? The evidence suggests that levels of patient satisfaction and, to a lesser extent, quality of treatment are improved by competition among GP practices, although the net effect is quite small (a similar result has been found for hospital quality). But a more general question, which is now at the forefront of public policy, is whether the range of treatments undertaken by providers other than GPs, such as pharmacists and polyclinics provided by hospital trusts, could be extended. There are already services, such as vaccination, where competition among different types of providers already exists. It is likely that this diversification of providers will continue apace.
The Government has now announced plans to enhance the number of face-to-face appointments, threatening a mixture of sticks and carrots to GP practices. Whether these incentives (or disincentives) will be integrated into a new version of the QOF, or added as a supplement to an already complex system of incentives, is as yet unclear. If there really is a shortage of doctors and the existing QOF structure cannot induce the required number of hours among existing GPs, it is not clear what the present proposals will contribute.
A more long-term approach might contain a proposal to simplify the existing system of incentives, to re-examine the parameters of the financial model which allocates funds between GP practices, and to accelerate the diversification of providers for many services that are currently primarily provided by GPs.
Richard Disney is Emeritus Professor of Economics at the University of Sussex, having previously been a full-time Professor of Economics at the Universities of Nottingham, London and Kent. He is a specialist on the economics of labour markets and other areas of applied microeconomics, on which he has published numerous refereed articles and several books. He has been on advisory committees to several UK governments and worked as consultant for international bodies including the World Bank and IMF. He served as a member of the NHS Pay Review Body from 2003-09 and the Senior Salaries Review Body from 2009-14.