Multitasking and mixed systems for provider payment

https://doi.org/10.1016/j.jhealeco.2004.09.001Get rights and content

Abstract

The problem of multitasking refers to the challenge of designing incentives to motivate appropriate effort across multiple tasks when the desired outcomes for some tasks are more difficult to measure than others. Multitasking is pervasive in health care. I use a simple model to show that the problem of multitasking further strengthens conventional arguments for mixed payment systems such as partial capitation. When pay-for-performance metrics are imperfect for rewarding service-specific quality efforts, using mixed payment helps to balance incentives for quality effort across services.

Introduction

Providers’ skills, time and attention are critical inputs to effective health care. The recent interest in pay-for-performance highlights the need to reward providers for “doing a better job,” and the difficulty of specifying exactly what that entails. For example, the ambitious new general practitioner contract in the UK includes 146 performance measures across seven areas of practice, affecting about 18% of practice earnings.1 This paper focuses on the question of how financial incentives influence providers’ allocation of effort across the various tasks involved in providing quality care.

The general literature on incentives recognizes that selective payment for performance leads to difficulties when there are many tasks expected of a supplier. The problem of multitasking refers to the challenge of designing incentives to motivate appropriate effort across multiple tasks when the desired outcomes for some tasks are more difficult to measure than others (Holmstrom and Milgrom, 1991). Precise metrics for provider actions that promote quality are notoriously difficult to quantify. Furthermore, there are obviously many dimensions of quality for health care services. As Newhouse (2002) notes, “payment on specific process measures of quality, such as beta-blockers after a heart attack, can distort resource allocation to the measured areas and away from unmeasured areas (the multitasking problem or ‘teaching to the test’). It is therefore hard to know whether on balance patients are better off” (p. 203). Smith and York (2004) identify some “potential risks” of the new UK contract for general practitioners, including potential stinting on services that are not rewarded (e.g., continuity and advocacy) or are only partially rewarded because appropriate measures seemed lacking (e.g., mental health).

Multitasking implies that payers should use pay-for-performance cautiously as long as quality is rewarded only partially or metrics are imperfect. In general, the less precise the measure of performance, the lower-powered pay-for-performance incentives should be. The few US initiatives in this area do currently use only modest financial incentives to reward measured dimensions of physician and hospital performance Rosenthal et al., 2004, Strunk and Hurley, 2004.

The simple model of provider multitasking developed in this paper suggests an additional implication: the problem of multitasking further strengthens the argument for mixed payment systems such as partial capitation. Previous economic arguments for mixed payment center on combating risk selection and quality stinting while maintaining some incentive to control cost (e.g., Ellis and McGuire, 1990, Ma, 1994, Newhouse, 1996, Ma and McGuire, 1997, Pauly, 2000, Eggleston, 2000, Newhouse, 2002). Mixed payment’s advantage of balancing incentives for quality effort across contractible and noncontractible dimensions of quality has not heretofore been highlighted (as far as I am aware). Yet this link deserves attention, especially as more employers and other purchasers include measures of provider performance in payment contracts.

When payers directly compensate for measured quality, quality stinting seems less of a concern. Moreover, risk adjustment may help to mitigate risk selection. Under these circumstances, the conventional argument for mixed payment seems to evaporate. Why not give strong cost-control incentives when paying directly for quality? This reasoning is flawed, however, because it overlooks the problem of multitasking. Since arguably some dimensions of quality will never be contractible, introducing selective pay-for-performance actually heightens the need for mixed payment—to balance incentives for dimensions of quality that complement treatment. The intuition is straightforward. High supply-side cost sharing is similar to high-powered incentives for cost control effort. Since quality increases costs, high supply-side cost sharing discourages quality improvement. When quality metrics used as a basis for pay-for-performance are imperfect – with some important aspects of quality unmeasured and hence unrewarded – pay-for-performance incentives and supply-side cost sharing should both be muted.

This argument provides a theoretical rationale linking two current trends, one away from capitation payment, and the other toward including pay-for-performance in provider contracts. Few if any discussions of the latter pay attention to this interplay between the design of pay-for-performance and the underlying compensation incentives.2

The paper proceeds as follows. The first section uses a relatively general and simple model to illustrate the advantages of mixed payment when what the purchaser seeks is not the same as what the provider contract rewards. The next section develops a model of explicit multitasking in which health service providers choose treatment and quality effort for multiple services. Providers may be ‘benevolent’ in the sense that they care directly about patient benefits from treatment (Chalkley and Malcomson, 1998). The central proposition shows that mixed payment lessens distortions in effort allocation when pay-for-performance is imperfect, thus promoting noncontractible dimensions of quality that complement treatment. The final section concludes with a brief discussion.

Section snippets

Incentive contracts and performance measurement

Consider how a principal (such as an employer or Medicare) should contract with an agent (an insurer or provider group). The agent is asymmetrically better informed (e.g., about how clinical actions impact outcomes). No contract can specify all desired dimensions of performance. This section draws upon the paper by Baker (1992) by the same name to illustrate why in such a setting mixed payment is optimal, as long as performance measures do not perfectly reflect true performance.3

Provider multitasking

In this section, I develop a simple model tailored to the health care setting. A health service provider allocates effort across multiple tasks. The provider also decides how much to spend on treatment for each patient.7 The purchaser wants quality care at a reasonable cost.

Conclusion

To the extent that some dimensions of quality cannot be written into a contract and enforced in court, multitasking problems will always plague provider performance measurement. This dilemma provides a new argument for mixed payment (or one might say an “old” argument with a new twist): mixed payment balances cost control and quality promotion incentives, including mitigating multitasking problems.

The traditional argument for mixed payment runs as follows: since risk adjustment of capitation

Acknowledgements

I am grateful to co-editors Richard Frank and Thomas McGuire, anonymous referees, David Cutler, Albert Ma, Nolan Miller, Joseph Newhouse, Richard Zeckhauser, and participants in the Seventh Northeast Regional Health Economics Research Symposium for helpful comments. All remaining errors are my own.

References (18)

  • M. Chalkley et al.

    Contracting for health services when patient demand does not reflect quality

    Journal of Health Economics

    (1998)
  • R. Ellis et al.

    Optimal payment systems for health services

    Journal of Health Economics

    (1990)
  • T.G. McGuire

    Physician agency

  • M.V. Pauly

    Insurance reimbursement

  • G.P. Baker

    Incentive contracts and performance measurement

    Journal of Political Economy

    (1992)
  • D.M. Cutler

    Your Money or Your Life: Strong Medicine for America’s Health Care System

    (2004)
  • K. Eggleston

    Risk selection and optimal health insurance-provider payment

    Journal of Risk and Insurance

    (2000)
  • Eggleston, K., Yip, W., 2004. Hospital competition under regulated prices: application to urban health sector reforms...
  • B. Holmstrom et al.

    Multitask principal-agent analyses: incentive contracts, asset ownership, and job design

    Journal of Law, Economics and Organization

    (1991)
There are more references available in the full text version of this article.

Cited by (0)

View full text