Payment for performance in the Family Health Programme : lessons from the UK Quality and Outcomes Framework Pagamento para desempenho no Programa Saúde da Família : lições do Quality and Outcomes Framework da Inglaterra

OBJECTIVE: Payment for performance fi nancial incentive schemes reward doctors based on the quality and the outcomes of their treatment. In Brazil, the Ministry of Health is looking to scale up its use in public hospitals and some municipalities are developing payment for performance schemes even for the Family Health Programme. In this article the Quality and Outcomes Framework used in the UK since 2004 is discussed, as well as its experience to elaborate some important lessons that Brazilian municipalities should consider before embarking on payment for performance scheme in primary care settings. Descriptors: Physicians, Primary Care. Family Health Program. Motivation. Reward. Quality Assurance, Health Care.

Although there is much variation in its defi nition and approach, several developing country health systems have experimented with payment for performance (P4P) schemes. 6These are fi nancial incentive schemes that reward doctors based on the quality and the outcomes of their treatment.The better the quality, the more they are paid.
The Brazilian Ministry of Health is looking to scale up its use for civil servants and in public hospitals a and some municipalities are developing P4P schemes even for the Family Health Programme (e.g.Prefeitura Municipal do Belo Horizonte; Prefeitura Municipal do Rio de Janeiro).The recent 11 th Congress for Family and Community Medicine had an entire panel discussion dedicated to the subject.b This article aimed to discuss payment for performance, by drawing on the eight years of experience that the UK has had with their system of payment for the Quality and Outcomes Framework (QOF) performance.If P4P is to be used in the Family Health Programme, it would be worthwhile to learn from the pitfalls and challenges that have been experienced in the UK and potentially avoid replicating mistakes.

QOF AND LESSONS FROM THE UK P4P SCHEME
The payment of quality of care incentives to general practitioners (GPs) in the UK is one of the great revolutions in primary health care in the last 20 years.First implemented in 2004, GPs receive fi nancial rewards on top of their salaries if they manage to reach certain targets in healthcare quality, process and outcome.There are 134 different targets that the GPs are paid to achieve in the QOF.c The targets cover clinical, organizational, patient experience and additional service domains (Table ).Each target is allocated a number of points depending on how challenging it is and GP practices are rewarded up to 1,000 points each year for reaching the targets.Every point attained equates to a payment of £126.00 to the GP.

INTRODUCTION
a Ministério da Saúde (BR).Portaria no1.537, de 4 de Julho 2011.Altera a Portaria no 3.627/GM/MS, de 19 de novembro de 2010, que fi xa os critérios e procedimentos específi cos de avaliação de desempenho individual e institucional para efeito de pagamento da Gratifi cação de Desempenho da Carreira do Plano Geral de Cargos do Poder Executivo (GDPGPE), da Gratifi cação de Desempenho da Previdência, da Saúde e do Trabalho (GDPST) e da Gratifi cação de Desempenho de Pesquisa e Investigação Biomédica em Saúde Pública (GDPIB), devida aos servidores do quadro de pessoal do Ministério da Saúde pertencentes ao Plano Geral Selecting indicators is a complex and dynamic process that is the responsibility of four organizations in the UK QOF.The National Institute for Clinical Excellence (NICE) identifi es whether reaching defi ned targets in clinical care quality could result in signifi cant public health impact.The University of York calculates the cost-benefi t of paying for doctors to reach the target.Two National Health Service (NHS) departments (the NHS Information Centre and NHS Evidence) consider whether systems are in place for GPs to collect data on progress towards the target and the implications, in terms of workload, in doing so.
For example, if the blood pressure of diabetic patients is well controlled, this can lead to better health outcomes, avoid hospitalizations, and complications.d,e One of the new diabetes targets in the QOF, replacing the DM12 indicator, is the indicator DM30 and is worth 8 points.f A GP should maintain the blood pressure in his/her diabetic patients at under 150/90.If the GP reaches the defi ned target of 71% i.e. 71% of the GP's diabetic patients have a record of successfully maintaining their blood pressure at less than this level, then the GP practice will receive £1,000.00 (8 points multiplied by £126.00).
The fi nancial rewards are only for GP practices, not secondary care specialists, and so the indicators are chosen in areas that are sensitive to primary care services, i.e., where primary care has a direct and clear responsibility for the continuous care and management of the condition; in areas where there is evidence that good care at the primary care level results in positive and benefi cial outcomes; and in areas where the disease is a public health priority.The UK has invested over £1billion in the QOF so far.GPs income has risen by 60% in the last eight years and 20% of GPs' salaries are directly derived from the fi nancial payments made from the incentive scheme.GP performance, as identifi ed through the QOF, is published annually and is publically accessible.This adds considerable pressure for GPs to perform to a high standard, reaching the set targets, because patients are able to choose which GP they are registered with in the UK.

The Impact of the QOF
Despite this investment in GP performance, there are divergent views as to whether the QOF has had any meaningful impact on public health. 4,12In part, this is because of limitations in research study design and the fact that the QOF was implemented throughout the country -there are no control settings. 12Nonetheless, the wide variation in quality of care by GP practices experienced prior to the implementation of the QOF has decreased across all QOF domains.This has disproportionately benefi tted poorer areas and so by implication health inequalities have actually improved as a result of the QOF. 5,12The QOF led to overall improvements in clinical care quality specifi cally for asthma and diabetes in the fi rst two years of its use. 2 Conversely, these improvements levelled off once targets had been reached and there was a worsening of continuity of care measures. 2Despite the QOF, ethnic disparities still persist in the management of cardiovascular disease 9 and it has had a minimal impact on hospitalizations. 1 Improvements in some clinical outcomes, such as hypertension management, had been occurring even before the incentive scheme was introduce. 10

Lessons from the QOF
Paying doctors extra for achieving a higher standard of care for their patients could be an interesting approach for the Family Health Programme as it might reduce the current variation in quality experienced across the country.However, there are important pitfalls and challenges in the experience of P4P in the UK that should be considered before implementing these schemes.

Developing the indicator set
Choosing a good indicator requires a trade off between several features: the scope of the indicator, the target to be reached, its value in terms of fi nancial reward and when it will be retired or replaced.This process is a politico-institutional one in the QOF and whilst NICE, York University, NHS Information and NHS Evidence work hard to create fair yet relevant indicators, based on the best available evidence, professional lobbying groups infl uence which indicators get adopted and which do not.Care needs to be taken to ensure that there is consensus around indicator development.The legitimacy of P4P The percentage of patients with diabetes whose notes record body mass index in the previous 15 months.

DM 5
The percentage of patients with diabetes who have a record of HbA1c or equivalent in the previous 15 months DM 9  The percentage of patients with diabetes with a record of the presence or absence of peripheral pulses in the previous 15 months.

DM 10
The percentage of patients with diabetes with a record of neuropathy testing in the previous 15 months.

DM 11
The percentage of patients with diabetes who have a record of the blood pressure in the previous 15 months.

DM 12
The percentage of patients with diabetes in whom the last blood pressure is 145/85 or less.

DM 13
The percentage of patients with diabetes who have a record of micro-albuminuria testing in the previous 15 months (exception reporting for patients with proteinuria).

DM 15
The percentage of patients with diabetes with a diagnosis of proteinuria or micro-albuminuria who are treated with angiotensin-converting-enzyme inhibitors (or A2 antagonists).

DM 16
The percentage of patients with diabetes who have a record of total cholesterol in the previous 15 months.

DM 17
The percentage of patients with diabetes whose last measured total cholesterol within the previous 15 months is 5mmol/l or less.

DM 18
The percentage of patients with diabetes who have had infl uenza immunisation in the preceding 1 September to 31 March.

DM 19
The practice can produce a register of all patients aged 17 years and over with diabetes mellitus, which specifi es whether the patient has Type 1 or Type 2 diabetes.

DM 21
The percentage of patients with diabetes who have a record of retinal screening in the previous 15 months.

DM 22
The percentage of patients with diabetes who have a record of estimated glomerular fi ltration rate or serum creatinine testing in the previous 15 months.

DM 23
The percentage of patients with diabetes in whom the last HbA1c is 7 or less (or equivalent test/reference range depending on local laboratory) in the previous 15 months.

DM 24
The percentage of patients with diabetes in whom the last HbA1c is 8 or less (or equivalent test/reference range depending on local laboratory) in the previous 15 months.

DM 25
The percentage of patients with diabetes in whom the last HbA1c is 9 or less (or equivalent test/reference range depending on local laboratory) in the previous 15 months.
targets can be undermined if there is not suffi cient buy-in from clinicians.Setting the target too high and the doctors will not get rewarded.Too low, and it will not discriminate between the high and low performers.None of the 134 targets are set at 100% and the GP will receive a fi nancial reward for most of the indicators even if only 40% of the target is met.
There is an ethical question: should doctors be rewarded for providing high quality care for only some and not all of their patients?All indicators reach a plateau eventually: their ability to infl uence and motivate better standards of care levels off over time.It is necessary to consider when to retire some indicators and when to create new ones to keep the fi nancial reward scheme relevant.
A choice has to be made between emphasizing process, outcome and impact indicators.Processes, e.g.measurement of HbA1c in diabetic patients, are easy to capture, however they do not automatically lead to desired outcomes.Outcomes, e.g.HbA1c levels below 7%, are of interest, however case mix needs to be taken into account for targets to be considered fair.Impact, e.g.number of hospitalizations due to diabetes, is of ultimate interest, however it is diffi cult to associate the activities of the Family Health Programme teams directly with these broader changes in health.Often a blend of all three might be required.Macinko et al 7 (2007) have developed a validated primary care evaluation tool for the Family Health Programme that could be a useful starting point for the standardisation of performance monitoring across different Family Health Programme teams.

Exception reporting
If a GP has been unable to meet a target, it might not have been from lack of effort or poor performance.
The QOF has some fl exibility in this regard called 'Exception Reporting'.GPs are allowed to exclude patients ('exceptions') from the performance target when it is appropriate to do so but they can only do this if the patient fulfi ls one or more of the following criteria: • The indicator is not appropriate in the patient's case • Appropriate services are not available • The patient is already on the maximum tolerated or permitted dosage of a treatment • The patient has an allergy to the medication

• The patient has a co-morbidity of greater severi-tyThe patient refuses treatment
There are considerable differences in Exception Reporting rates between different GP practices and in some municipalities, there are GP practices that Exception Report several times more patients than other GP practices. 11,g Should these practices be rewarded in the same way as the other practices?The GP practice, the patient register and even patients' notes are subject to annual review.However, this is not a particularly well audited or enforced area and the system is susceptible to fraudulent behaviour.In the Family Health Programme, therefore, P4P schemes should allow some fl exibility whilst at the same time ensure proper safeguards against fraud.

Prevalence
The local prevalence of a disease is not considered in the QOF and GPs working in low socioeconomic areas become disadvantaged relative to colleagues in wealthier areas.In areas that have a very high prevalence of a particular disease, reaching the target for an indicator in that disease area will be more challenging and require more resources, time and effort by the doctor than in areas that have a very low prevalence.The incentive payment per patient becomes less than for clinicians in wealthier areas.Some GPs have argued for a review of the payment formula so that health inequalities are not worsened.In the Family Health Programme, therefore, P4P schemes should take into account local epidemiology and provide a weighted reward for performance.

Team working and changing roles
The emphasis on reaching targets has generated a target culture which is inappropriate in primary care.Primary care cannot be easily broken down into small, measurable parts and is complex, messy and dynamic.Not everything that can be measured is of value, and that which cannot be measured may, nonetheless, be of high value.QOF has changed attitudes within GP practices; GPs treat patients more as a collection of measurable indicators than as an individual. 3Patients are called back to the GP practice on multiple occasions so that procedures, examinations and tests can be carried out, the sole purpose of which is to fulfi l and reach desired QOF targets.In the Family Health Programme, P4P schemes should enhance not detract from good quality primary care, and doctors must not lose sight of the patient within this process.

Reward structures
The QOF uses fi nancial payments linked to performance measures to reward GPs and the payments are directed towards the GP practice at the end of the fi nancial year.In the UK, GPs are independent contractors to the State, and although the payment goes to the GP practice as a whole, the GP who owns that practice can decide how the rewards should be best divided and allocated; it bolsters salaries, in particular for the GP.However, primary care performance is stochastic, it is team-based -many professionals are involved and it is diffi cult to determine directly whose effort it was that led to a positive outcome.Without careful attention, P4P schemes can cause dissention and attrition between health professionals of different disciplines.Rewards for the team as a whole would be a better incentive structure.
Furthermore, fi nancial rewards like the QOF attends to one of the very lowest of Maslow's hierarchy of needs and motivation. 8There are many other rewards that can be considered to motivate health professionals other than fi nancial ones.For example, building reputation, training, attending conferences, supporting self-actualization, career promotion, and space to innovate and be creative within the workplace.These are likely to be more effective motivators of good performance in health professionals than simply paying them more.P4P schemes should be creative in the use of different types of incentives in the Family Health Programme and consider rewarding all health professionals, not just the doctors.

CONCLUSION
P4P systems can be useful to bring about an improvement in quality of care and health outcomes in some areas sensitive to primary care activities.However, they are very expensive to develop and maintain, requiring complex information technology systems, and can distort the priorities of healthcare professionals.Financial incentives exist in the Brazilian Unifi ed Health Care System (Sistema Único de Saúde -SUS) but these are for municipalities to create Family Health Programme teams, rather than to improve their performance.The growing interest in P4P schemes in the Family Health Programme is complicated by the relatively weak performance monitoring culture at municipal and state levels h and there are few systematic attempts to analyse routine data sets e.g.(Sistema de Informação da Atenção Básica).Macinko et al 7 (2007:174) noted that a major challenge for the Family Health Programme will be to 'develop and use systems to monitor and improve the quality of care delivered in order to maximize the potential health gains of this innovative approach to integrated primary care delivery'.
Based on the experiences of the QOF in the UK, there are important principles that Brazilian municipalities should consider before embarking on a P4P scheme: • Is there consensus around the indicator set?
• Is there an evidence base for the indicator?
• Are the targets reasonable and an ethical use of resources?
• Can indicator performance be evaluated and can they be retired and renewed as necessary?
• Is the local context and case mix being taken into account in the measurement of doctor performance?
• Can performance data be independently verifi ed and audited on a regular basis?
• Is the P4P having a negative impact on team roles, the doctor-patient relationship and the patient experience?
• Is the reward and incentive structure the most appropriate one for the professionals in the Family Health Programme?
The growing interest in P4P schemes in Brazil should be an opportunity to develop experimental studies so that the impact and value of the incentives can be properly evaluated.

ACKNOWLEDGEMENT
To Dr Luis Pisco of Universidade de Lisboa (Portugal) and to Dr Juliana de Santos of Belo Horizonte Municipal Health Secretariat (Brazil) for comments on an earlier draft.
h Brazilian Court of Audit.Evaluation of the Family Health Programme.Brasilia; 2003.

Table .
Quality and Outcomes Framework indicators for diabetes.United Kingdom, 2009-2010.