Financial Incentives for Providers

Pay-for-performance is ubiquitous and is evolving (or mutating) into value-based purchasing programs.  Are these programs always effective–do they improve health outcomes?  What are the characteristics of programs that are good versus those that appear ineffective?  Researchers have created a checklist based on the available research regarding these programs and published it in the British Medical Journal.   (BMJ Article)   The authors first note that even outside health care there is mixed evidence about the effect of financial incentives in meeting targeted objectives.  Within health care, there is little evidence from the existing research that financial incentives improve outcomes, that they seem generally ineffective in improving compliance with guidelines, that they seem to improve processes of care, and perhaps some other aspects of health care delivery.  Most research looks at the effect of individual incentives, not group ones, but doctors increasingly practice in groups.  Factors which might affect the effectiveness of incentives are poorly studied and understood and unintended consequences such as attention shifting, gaming or loss of motivation have not been well examined.

The authors have six questions in their checklist; if the answer to any is no, they suggest postponing implementation of a program.  The questions are does the desired clinical action improve patient outcomes; will undesirable clinical behavior persist with the financial incentive; are there valid, reliable and easy to use measures of the clinical behavior being incented; are the barriers and enablers to the desired behavior well-understood; will financial incentives work, and work better than other methods, to change the behavior and why would they work; and will benefits outweigh potential harms with an acceptable cost for the program?  If the answer to all the questions is yes, the authors have three succeeding questions relating to the design of a successful program:  are systems and structures in place to facilitate the change; do you know how much should be paid, for how long and to which providers or members of a care team; and how will the incentives be delivered, from a practical perspective.  It is obvious that strong tools for assessing an organization, the desired behavior, the design of the incentive and the actual effects, good or bad, of the incentive are mandatory.  Not many organizations have this capability.