Mannion, R. and Davies, H.T.O. (2008), "Incentives in health systems: developing theory, investigating practice", Journal of Health Organization and Management, Vol. 22 No. 1. https://doi.org/10.1108/jhom.2008.02522aaa.001Download as .RIS
Emerald Group Publishing Limited
Copyright © 2008, Emerald Group Publishing Limited
Incentives in health systems: developing theory, investigating practice
Article Type: Guest editorial From: Journal of Health Organization and Management, Volume 22, Issue 1.
Incentives in health systems: developing theory, investigating practice
Introducing this Special Issue
Improving the performance and quality of health services has become a prime objective of policy reform worldwide. Reliance on innate professionalism to deliver high quality, safe and efficient services has come to be seen as curiously old fashioned. Modern health systems are managed like never before (Davies et al., 2007).
In particular, ideas loosely grouped under the rubric “new public management” (Ferlie et al., 2005) have led to an explosion of schemes aimed at measuring and shaping the performance of health care delivery. Central to these are ideas of agency, motivation and incentives (Le Grand, 2007; Mannion et al., 2007), and arguments as to the extent to which health care and doctors in particular can be subject to market and bureaucratic controls (Davies and Harrison, 2003).
Regardless of the debates and contested terrain, policy makers and service managers have, over the past decade, designed and implemented a wide range of implicit and explicit incentive schemes tied to routinised measurement as a means of driving performance and quality improvement in provider organisations (Mannion et al., 2005a). But what are the design principles of such schemes? How are they implemented? And what are their effects? It is to these core concerns that the papers in this Special Issue of the Journal of Health Organisation and Management are addressed, drawing on a wide range of UK and international experience.
In simple terms an incentive can be defined as a reward (or sanction) associated with a particular aspect of performance. An incentive can be purposive (in the sense that it has been designed specifically to induce beneficial behaviour) or accidental (perhaps resulting as a by-product of other managerial actions). In health care, incentives can operate at a number of levels, including the individual, the clinical team, and the hospital, and can take a variety of forms (see following list). The sanctions side of incentives may be less visible, less invoked and under-examined, but includes such aspects as: enforcing requirements to undertake re-training and professional development; threats of professional deregistration or dismissal; and even although rarely criminal proceedings. The varying types of incentives include:
intrinsic reward (the feeling that a job has been done well);
visible approbation and enhancement of peer esteem;
career advancement and development opportunities;
personal and/or collective (e.g. team- or organisation-based) financial rewards;
enhanced local budget for service development;
enlarged scale and scope of service lines;
time allowance to pursue alternative activities, including research; and
increased autonomy and reduced levels of inspection.
One of the most popular approaches to the use of (indirect) incentives over recent years in both publicly funded and private payer health systems has been the public release of comparative performance information through “report” cards, comparative league tables and “star rating systems” (Marshall et al., 2000; Majeed et al., 2007). There is a growing body of evidence to suggest that public dissemination of performance data can stimulate provider organisations to improve internal data collection systems and processes and lever beneficial changes in staff behaviour (Davies, 2001). There is also growing evidence to suggest that in addition to driving beneficial changes, public disclosure of performance data can also induce a range of unintended and dysfunctional consequences for organisations and patients (Smith, 1995; Mannion et al., 2005a).
In recent years increasing attention has been devoted to the creation of explicit and finely tuned financial incentives to promote performance improvements. Pay for performance programmes (sometimes abbreviated to P4P) are financial arrangements in which a portion of the payment is based on performance assessed against one or more defined measures (Hahn, 2006; Mannion and Davies, 2007). P4P schemes have seen most significant development in the US where there are currently over 100 private and Federal Medicare reward and incentive programmes in operation. However, the recent general practitioner contract in the UK (the Quality and Outcomes Framework, or QOF) is perhaps the most ambitious P4P scheme anywhere in the world (Shekelle, 2003).
Experience, to date, suggests that P4P schemes are potentially powerful but unpredictable devices, suggesting that they need to be designed and implemented with great care. P4P schemes operate in many different ways, with a wide variety of common and fundamental decisions intrinsic to their design (see list, abstracted from Mannion and Davies, 2007). Many of these design elements involve important trade-offs, the relative impacts of which are currently unknown the evidence base has failed to keep pace with policy and managerial developments on the ground (Mannion and Davies, 2007). Moreover, because of the diversity of potential schemes, and hampered by a lack of coherent established theory with predictive validity, great care must be exercised in transferring inferences from one scheme and context to another. The key design elements of a P4P system are as follows:
Objectives. Although the majority of schemes focus on health care quality, performance objectives could cover a wide range of variables including: volume, equity, patient satisfaction, patient safety and cost-effectiveness.
Unit of assessment. P4P schemes vary in terms of who is subject to evaluation and financial incentives. For example rewards could be targeted at individual clinicians, clinical teams, or larger organisational aggregates.
Performance measures. A fundamental design issue intrinsic to the design of a P4P system is the choice of measures used to assess that performance. When assessing quality, the key decision is whether to focus on process or outcome measures of care.
Analysis and interpretation of performance data. The analysis of performance data in health care poses significant problems of attribution. In highly interdependent health systems many external factors serve to weaken the link between actual performance and measured performance. P4P schemes vary widely in the sophistication of the data analysis that underpins them.
Performance standards. All P4P schemes require performance criteria or thresholds against which to determine whether payments are triggered. These may be absolute benchmarks or performance relative to peers, or be based on changes over time, within a given unit of assessment.
Financial rewards. The size of the financial pay-off for those participating in a P4P scheme is a central issue. If the rewards are small, they may be insufficient to garner attention; conversely, large incentives may lead to major, potentially unpredictable, changes, and as rewards rise the risk of serious unintended consequences becomes more severe.
In this Special Issue we present a collection of papers that draw on broad experience and detailed study across diverse and international settings. Taken together, the papers advance new theory and evidence on the development and implementation of managerial incentives particularly P4P schemes across a range of health systems and contexts. They are both descriptive and analytic in intent, and they range from accounts of policy design and implementation (sometimes refined over many iterations across a decade or more) to fine-grained ethnographic explorations of micro-change in organisational and professional dynamics. Collectively, they help articulate the scale and scope of study required if we are to begin to understand and evaluate the changing incentive environment for health care delivery.
The opening paper by Lega and Vendramini (this issue) traces the history and development of performance measurement and management systems in the Italian National Health Service over almost 20 years. They demonstrate the growth in interest and increasing sophistication of such schemes, and explain such growth both in isomorphic terms and as a result of changing methods of reimbursement, clinical governance and human resource management. The authors end their paper with a plea for reduced isolationism and increased international comparison with concomitant evaluative effort. We need to learn from all these natural experiments that are unfolding around us.
Gross et al. (this issue) to some extent pick up this challenge. They first of all analyse the P4P programmes that have evolved in Israel, providing an insightful account of an unusual example of nation-wide, long-term implementation. While reporting significant improvements in quality measures, the authors none-the-less note that attribution is difficult because of other wide-spread contemporaneous trends. Such difficulties bedevil evaluations of P4P programmes worldwide, leaving estimates of impact through randomised study a relative rarity (Rosenthal et al., 2006; Peterson et al., 2006). As such, accounts such as this from Israel serve more to uncover the practical and operational difficulties encountered, and highlight the myriad ways in which schemes might unfold differently as a result of diverse local contingencies. The account does however begin to rise to the challenge of Lega and Vendramini (this issue) by comparing the Israeli experience with that of another country, in this case New Zealand.
Stephen Buetow (this issue) adds further flesh to these bones in his detailed account of how New Zealand is beginning to reward primary healthcare organisations (PHOs) for their performance. In doing so, he contrasts some of the New Zealand experiences with that in the UK under QOF (studied in detail in two later papers). Buetow notes that it is unclear whether there is good alignment between those values expressed through payments and the actual values and goals of health care providers. At the same time, the payment of bonuses to the overarching PHO rather than practices or practitioners themselves blunts the impact of what are, in any case, small and marginal payments. In this sense, the approach taken in New Zealand represents cautious experimentation around the margin compared to practice-focused UK QOF where up to a quarter of practice income might (theoretically) be at risk.
Two complementary papers (McDonald et al. and Huby et al., both this issue) take a broadly ethnographic approach to examining the micro-details of change in English and Scottish general practices as implementation of the QOF unfolded. The object of these studies was in no way to assess the benefits of QOF on care, quality or performance, but to explore instead the perceptions, meanings and consequences of QOF on day-to-day practice, local organisational arrangements and local sense-making.
McDonald et al. (this issue) studied two practices in England and noted the increase in surveillance by some practice staff of their clinical colleagues. Traditionally, the clinician-patient interaction was “beyond the gaze of the outside” but this had now been “opened up to scrutiny and influence” as practices sought to respond to the demands of external P4P. As such, new strata of staff were created, including those trusted (to deliver clinical practice in terms of QOF diktats) and those not; those chased (to change practice) and those doing the chasing. While surface accounts from practitioners suggest QOF simply recognised work that was being done anyway and downplayed the extent of any change, the work of McDonald et al. highlights more subtle dynamics and documents the emergence of new tensions within and between existing professional groupings.
Huby et al. (this issue) similarly set out to explore the apparent contradiction between QOF as the biggest and most ambitious P4P scheme in health care and protestations from (especially) general practitioners that it was “business as usual”. Drawing on ideas of “sense-making” (Weick, 1995), they show that whereas each practice (of the four studied) had a dominant and distinctive story about itself in terms of its approach to practice work and the values underpinning this approach, such diversity was belied by very significant convergence of organisational forms in response to QOF. Decision-making had become concentrated in fewer hands in each of the practices, and the emergence of these decision-making elites had led to new tensions. Thus, divergent practice narratives masked a key and common trend towards corporatisation and the strengthening of a managerialist agenda.
Rounding off this collection of papers, Mannion et al. (this issue) turn to the hospital sector. Again, reporting from the English NHS, through case studies and theoretical exploration, the authors show how Payment by Results (PbR) provides strong incentives for producers to increase activity, potentially beyond affordable levels. They also highlight the potential for unwanted effects, such as unplanned hospital readmissions and a worsening of joint planning. As such, the authors reiterate the urgent need for careful monitoring of intended and unintended effects, and provide some worthwhile clues as to what these unintended effects might be.
Overall then, these papers attest to the proliferation of incentive schemes worldwide and begin to articulate a complex and comprehensive research agenda. Crucially, we need to move beyond case accounts of on-the-surface-successful implementation of P4P schemes to more theoretically driven and analytic evaluations of such schemes in all their diversity. In particular, we must be alert to the ways in which incentive schemes communicate values across the system, and cognisant of a sometimes lack of congruence between those values communicated and those held by important stakeholders (staff, yes, but where too are the patients and service users in all of this?) We should also be wary of falling into the trap of believing that only those behaviours which are the target of incentives will be shaped, and instead look as some of these authors have done for the wider fallout and sequelae of changes in the incentive context. Such changes may be subtle, masked, and even denied, but none-the-less important for their longer-term implications. They may also be as much about identity, values and sense-making as observable practice change.
Managing through incentives is likely to see a burgeoning of interest in coming years as data systems grow over-more sophisticated and “purchasers” (of all forms) ever-more demanding. As the work presented here shows, we have barely scratched the surface of the possibilities, and we are still a long-way distant from any comprehensive theoretical and empirical accounts. But while the evidence and understanding lags behind, proponents of P4P and other incentive mechanisms will continue to advocate unabashed. This places great import on concomitant analytic effort to which we hope this Issue contributes.
Further Reading Mannion, R. and Davies, M.M. (2005), Cultures for Performance in Health Care, Open University Press, Buckingham
About the journal: readership statistics
In 2007 the Journal of Health Organization and Management attracted high readership: over 110,000 articles were downloaded in 2007. While around half of downloads are from the UK, the other half is taken up by readers in other countries, with Australia, Malaysia, USA, South Africa, Canada and Iran, being in the top ten download markets.
In terms of numbers of downloads, the most popular paper has now been downloaded over 8,800 times in total and, in terms of “immediacy” of impact, the highest impact paper achieved over 900 downloads in the six months following publication.
Russell Mannion, Huw T. O. DaviesGuest Editors
Davies, H.T.O. (2001), “Public release of performance data and quality improvement: internal responses to external data by US health care providers”, Quality in Health Care, Vol. 10, pp. 10410
Davies, H.T.O. and Harrison, S. (2003), “Trends in doctor-manager relationships”, British Medical Journal, Vol. 326, pp. 6469
Davies, H.T.O., Mannion, R., Jacobs, R., Powell, A. and Marshall, M. (2007), “Exploring the relationship between senior management team culture and hospital performance”, Medical Care Research and Review, Vol. 64, pp. 4665
Felie, E., Lynne, L. and Pollit, C. (2005), The Oxford Handbook of Public Management, Oxford University Press, Oxford
Hahn, J. (2006), Pay for-Performance in Health Care, Congressional Research Service The Library of Congress, Washington, DC (order code RL33713, available at:www.crs.gov)
Le Grand, J. (2007), The Other Invisible Hand: Delivering Public Services Through Choice and Competition, Princeton University Press, Princeton, NJ
Majeed, A., Lester, H. and Bindman, A. (2007), “Improving the quality of care with performance”, British Medical Journal, Vol. 335, pp. 9168
Mannion, R., Davies, H. and Marshall, M. (2005a), “Impact of star performance ratings in English acute hospital trusts”, Journal of Health Services Research and Policy, Vol. 10 No. 1, pp. 1824
Mannion, R. and Davies, H.T.O. (2007), “Payment for performance (P4P) in health care”, British Medical Journal, Vol. 64 No. 1, pp. 4665
Mannion, R., Goddard, M. and Bate, A. (2007), “Aligning incentives and motivations in health care”, Financial Accountability and Management, Vol. 23 No. 4, pp. 40120
Marshall, M., Shekelle, P., Leatherman, S. and Brook, R. (2000), “The public release of performance data. What do we expect to gain? A review of the evidence”, Journal of the American Medical Association, Vol. 283, pp. 186674
Rosenthal, M.B., Frank, R.G., Li, Z. and Epstein, A.M. (2006), “Early experience with pay-for-performance: from concept to practice”, JAMA, Vol. 294, pp. 7881793
Shekelle, P. (2003), “New contract for general practitioners”, British Medical Journal, Vol. 326, pp. 4578
Smith, P.C. (1995), “On the unintended consequences of publishing performance data in the public sector”, International Journal of Public Administration, Vol. 18, pp. 277310
Peterson, L.A., Woodard, L.D., Urech, T., Daw, C. and Sookanan, S. (2006), “Does pay for performance improve the quality of care?”, Annals of Internal Medicine, Vol. 145, pp. 26572
Weick, K.E. (1995), Sense-Making in Organisations, Sage Foundations for Organisational Science Series, Sage, Newbury Park, CA