Financial incentives and professionalism: another fine mess

In 2007, writing about the Quality and Outcomes Framework (QOF), Mangin and Toop drew attention to the risk of using financial incentives to drive clinical behaviour.¹ Asking plaintively, ‘What have you done to yourselves?’, the authors expressed concerns that a combination of chasing financial reward, prioritising population health over individual patient care, and a reduction of clinical autonomy were all contributing to an erosion of professional values. QOF was transforming the landscape in general practice, enabling practices to demonstrate levels of achievement that were higher than many had predicted² and yet, like most quality improvement interventions, the unintended consequences could not be ignored.

Early versions of QOF were criticised in particular for the large number of indicators that were not based on rigorous research evidence. Subsequent refinements have started to take this criticism on board. The National Institute for Health and Care Excellence now plays a central role in assessing the scientific properties of the indicators, drawing more explicitly on the science of improvement to link practice and evidence. Critics continue to voice concerns about the principle of incentivising clinical behaviours, the preoccupation with economic drivers for change over educational ones, and the disproportionate size of the financial incentives in comparison with other elements of practice income, but a decade after its introduction QOF has become embedded as part of the identity of primary care in the UK.

DIRECT ENHANCED SERVICES

As the design and implementation of QOF adapts to take into account some of the legitimate criticisms, it is now being replaced by new incentive schemes, which again beg an answer to Mangin and Toop’s provocative question about professionalism. In 2014–2015, there were six national so-called ‘direct enhanced services’ (DESs), which incentivised specific clinical behaviours and were available to all practices in England: unplanned admissions, dementia, learning disability, alcohol, extended hours, and patient participation. Two of these have been discontinued for 2015–2016 (alcohol and patient participation).³ Each is labour intensive, template driven, with payment linked to template completion.

QOF was the product of detailed negotiation between professional leaders and policymakers but in contrast the DESs feel like they are being imposed on a reluctant workforce. Where professional leaders have had a say, their stance appears to be driven more by a desire to recoup the practice income lost though the slimming down of QOF than by any sense of vision about how general practice can best meet the needs of patients. Like QOF, the DESs are dominating what happens in general practice, contributing substantially to both GP income and to workload. It is difficult to escape the impression that primary care policy is being influenced more by an unhealthy combination of ideology and political pragmatism, than by the research evidence of what works.

THE UNPLANNED ADMISSION DES

The unplanned admission DES exemplifies the approach that this policy initiative promotes. It requires each practice to use a risk-stratification tool in order to identify the 2% of registered patients deemed to be at greatest risk of hospitalisation in the coming year. Each identified patient then has to be contacted, consent obtained, and a detailed care plan drawn up and agreed with the patient or their carer. A series of home visits and admission avoidance strategies are being implemented.

It is unlikely that this DES costing £160 million will achieve its aim of reducing unplanned admissions.³ There is good evidence that despite their intuitive appeal most of the proposed interventions designed to prevent hospital admission are ineffective.^4–7 Indeed, two large controlled studies of admission avoidance found that overall hospital admissions increased during the study period, in spite of anecdotal evidence during the course of the studies to the contrary.⁴ The Evercare scheme introduced in 2003 was the first large-scale implementation of case-management in the UK and introduced community matrons as a key intervention to prevent hospital admission. Hailed as the answer to rising levels of unplanned hospital admissions, this scheme was reported to have achieved 50% reductions in admissions of the frail elderly in the US. In the UK the intervention was highly valued by carers and patients but resulted in a non-statistically significant increase in hospital admissions, in part because the implementation of case-management revealed a large amount of unmet need in the community.^5,6 However challenging this evidence may have been for policy makers, the commitment to use pilots and formal evaluation to understand the impact of the Evercare initiative contrasts markedly with that seen for the DESs.

A MISMATCH BETWEEN POLICY, PRACTICE, AND ACADEMIC EXPERTISE

Few would argue with the intent of the unplanned admissions DES. Preventable hospital admissions are a major problem for patients, a pressure on over-stretched hospitals, a waste of precious resources, and reflect poorly on the quality of primary and community services. The fragmentation of the current health system needs to be addressed and quickly. But good intent does not justify poor policy making and poor implementation. Policy making is not credible if it ignores established scientific evidence about the relative effectiveness of financial incentives in comparison with other interventions, the risk of unintended consequences and the opportunity costs, the most appropriate size of incentives as a proportion of overall income, and the likelihood of the policy intervention to achieve its stated aims.

Poor policy making leads inevitably to poor implementation. The workload implications of implementing the DES as originally designed, at a time when the work force is already under massive stress, seems not to have been thought through, although the rapidly-introduced changes to the implementation guidelines suggest that they were quickly recognised in retrospect. The time allocation required to generate the required care plans and the opportunity costs are substantial. Since data gathering is the trigger to payment, there is a risk that it becomes an end in itself. It is hard to see how professional values are not further being eroded by the process.

PROFESSIONAL VALUES AND THE DES

It is easy to criticise the technical underpinning of the policy development and implementation of the unplanned admissions DES, but the real problem lies deeper. Fundamentally, the DES is not adequately aligned to the professional values of GPs and this is problematic from both a theoretical and an empirical perspective. The improvement science literature tells us that improvement efforts that are not valued by those responsible for implementing them are less likely to be effective and more likely to have unintended consequences.⁸ While GPs are demonstrating their commitment to greater accountability, they also value their professional autonomy because it helps them to provide care that is tailored to the needs of individual patients. They do not value micromanagement or excessive bureaucracy. They value building personal relationships with their patients and their professional colleagues over time, and understand the impact of local context on the quality of these relationships. They do not value interactions mandated through contractual obligations that ignore local practices, history, and culture. They value a formative approach to education and learning, focusing on internal drivers for change and based on social theory. They do not value external drivers for change based on economic theory. Established models of professionalism are sometimes self-serving and need to be challenged but they also serve a positive purpose. Ignoring what matters to front-line staff is unhelpful and self-defeating.

Based on these values, it would not be difficult to design a programme with similar aims to the unplanned admissions DES but one that is more likely to achieve its policy objectives. For example, a DES might promote educational approaches to exploring the causes of unplanned admissions, seeking sophisticated solutions to a complex problem. Or it might encourage regular meetings between groups of practices, community and social care staff, or meetings between secondary and primary care teams to discuss unplanned admissions. Or it could purposefully encourage the development of new knowledge about how to address the problem of unplanned admissions through collaboration between practitioners and academics. All of these approaches would be more likely to get the best out of professionalism and less likely to promote a passive and sometimes cynical set of compliance behaviours. Policy makers should focus on engaging the hearts and minds rather than the pockets of GPs.

Incentives to reduce unplanned admissions are more likely to be effective if complemented by an infrastructure that rewards the search for shared solutions between primary and secondary care. At present, health care in England is configured in a way that has resulted in competitive pressure between primary and secondary care. A focus on population need rather than the priorities of traditional sectors is more likely to manage the inevitable perverse incentives that are generated by the current system.

CONCLUSION

Unless new policy initiatives are aligned to the professional values of general practice, they are unlikely to realise their full potential. We propose that quality improvement is achievable, including admission avoidance, but only through co-creation between primary care policy makers, researchers, commissioners, and educationalists.

Provenance

Freely submitted; not externally peer reviewed.