Do doctor bonuses actually improve health care?

Not as much as you might think


Erik K Veland/Flickr

You get what you pay for, right? It’s taken for granted that this holds true when it comes to using financial incentives to improve the quality of physician care. For example, if a GP gets a smoker to quit or a doctor at the hospital treats a heart-attack patient with the best medicine, they’ll be paid extra.

Intuitively, this makes sense: rewarding physicians for providing better care should, theoretically, boost quality and lead to improved health outcomes. That may be why “pay-for-performance” schemes have been touted by policymakers around the world. The Affordable Care Act in the U.S. advocates the use of pay-for-performance programs at hospitals. Britain and Australia have already ushered in these compensation models at the primary-care level. And here in Canada, health-care observers have long argued that rewarding and incenting quality among doctors is the way forward.

But what about the evidence?

Well, it seems the scientific findings aren’t as enthusiastic as some policymakers. A recent article published in New England Journal of Medicine looked at how pay-for-performance schemes in the U.S. impacted patient outcomes, and in particular, whether the folks who went to pay-for-performance hospitals had a lower chance of death within 30 days of hospital admission. The result: there was no evidence that the pay incentives led to better 30-day mortality. “Taken together, these findings are sobering for policymakers who hope to use incentives… to improve patient outcomes,” the researchers stated.

Others have come to similar realizations. The World Health Organization’s review of the literature on the impact of financial incentives to improve the quality of health care found that study results were mixed or that there was not enough evidence to draw conclusions. Similarly, a 2011 Cochrane Systematic Review on this type of compensation in primary care concluded that “despite the popularity of these schemes, there is currently little rigorous evidence of their success in improving the quality of care.”

Dr. Chaim Bell, a professor at the University of Toronto who looked at the issue in Canada, told Science-ish that there are a few problems with pay-for-performance programs that aren’t immediately obvious. “By paying (doctors) for some things and not others, you are inherently putting an emphasis on one group.” So we may reward a conversation with a patient about quitting smoking but ignore cervical cancer screening.

Also, the assumption with pay-for-performance schemes is that it’s up to the doctor to do better. “For some things, that might be true. But maybe that’s not the problem and it’s not that the doctor is not trying or that he or she forgets. The problem may be further down the line.”

Allow Science-ish to complicate matters further by noting that studying this issue is really hard. Much of the research on pay-for-performance schemes involve studies that aren’t randomized, or only look at narrow quality markers, and don’t last long enough to measure the things that matter, like long-term health outcomes with chronic diseases. Plus, it’s difficult to asses whether this kind of compensation is truly cost-effective.

One recent study set in the Philippines, however, showed promise. It measured doctors’ performance at 30 hospitals: Some used individual bonuses for outstanding MDs and others system-level incentives, where whole institutions were rewarded for hitting quality markers. The lead author, Dr. John Peabody, of the University of California Los Angeles and Qure Healthcare, found that “pay per performance worked significantly to not only change performance but improve the health outcomes of children.”

But the story wasn’t quite so simple. “It wasn’t really pay per performance that seemed to drive the behavioural change,” Dr. Peabody continued, “but the measurement of performance.” So it may have been the policy all hospitals had in common—careful tracking and feedback about doctors’ work—that led to improvements in quality, not just the extra cash. Dr. Peabody added that this finding is consistent with non-experimental studies that show only modest or no gains from financial incentives.

Unfortunately, doctors don’t seem to get performance reviews very often. As one Toronto physician told Science-ish, in his some 30 years on the job he received feedback about his care of a patient only once. Maybe, then, better measurement and feedback would lead to better quality—or at least a deeper understanding of where things go wrong. As Dr. Bell said, “People feel if you want something extra, you have to pay for it. I’m not sure that’s the best motivator in this case.”

Science-ish is a joint project of Maclean’s, The Medical Post, and the McMaster Health Forum. Julia Belluz is the associate editor at The Medical Post. Got a tip? Seen something that’s Science-ish? Message her at or on Twitter @juliaoftoronto


Do doctor bonuses actually improve health care?

  1. Doctors should have all patients fill out private customer satisfaction questionnaires at the end of each appointment. These should then be sent to an independant body for review…I think this is the only way towards accountability.

    • an interesting suggestion. i see a few problems however. 

      part of it is the idea that the goal is to increase patient health, not patient satisfaction. Now they may not be mutually exclusive and one can influence the other however not always and most defiantly not the largest variable to achieving better patient health. 

      Two quick different examples, if a patient comes to me and is looking to get healthier then they are more likely to take my medical advice and follow the medical plan. This might mean medication, it might mean lifestyle changes (quitting smoking etc) and so on. As the patient gets healthier their satisfaction would increase (cause its working). This is a good example of your thought process I think.

      The counter example is a patient who presents with symptoms of cardiovascular disease after years with type two diabetes. The doctor can explain lifestyle changes that will improve the patients health, but if the patient doesnt want to quit smoking or doesn’t want to stay compliant with his/her medications then their health generally deteriorates. If that patients glycosylated hemoglobin comes back high, constantly telling the patient the importance of staying compliant with the medication can come off to some as nagging and annoying. and they would have poor ‘customer satisfaction’ and would prefer if their doctor just let do what they want.

      Sadly regardless if a patient wants to get better or not, it is still the physicians job to try to increase their health. If the reward (or punishment) was based on the satisfaction then doctors would be more inclined to tell patients who hate taking their meds to stop taking them, as a result the patients would be much happier, but statistically at a higher risk for a heart attack, stroke, diabetic retinopathy, DVT, PVD, and so on (staying with that example).

      So if the goal is greater patient health (which the article is suggesting), then using satisfaction as the measuring stick might not be the way to go.

      (not that I agree or disagree that bonus would work well, or think that patient satisfaction isn’t important)

    • Yes, I can just see it now…”My doctor is a mean a**hole.  He just told me to lose weight and exercise.  He also advised against doing the tests Oprah recommended on her tv show….. 

  2. Intriguing article — and always its nice to see the usage of some evidence informing journalism. I think however, as always, one needs to be a little bit more careful on which evidence is used. Pay-for-performance for physicians is a little bit different, than the NEJM article was assessing — in fact that article focused more on hospital financing, rather than physician remuneration.

    That said, PfP for physicians is still in its infancy, and infact has shown some mixed results. In mature markets like the InterMountain Health group, in the States, they have used pay-for-performance for physicians in a very effective manner to improve patient outcomes, through better guideline adherence.

    Clearly, there are some lessons to be learnt from such an incentive scheme. After all, physicians also do respond to appropriate incentives, and the struggle is in policy-makers deciding what those right incentives are, so as to prevent perverse outcomes.

    • Why should anyone be paid extra to do a better job?When is earning a good wage not good enough?

      • a good point Dothardy, but i think paying canadian docs a good wage would cost to much, so they are trying to add these cheaper incentives. 

    • very good points. Thanks for pointing out that NEJM article.