Health incentives: don’t shoot the behavioural economists!

Pedro Rey

The debate on the allocation of (economic) incentives associated with the performance of health professionals has been relevant for a long time. A few months ago, VOX has published an editorial with a very clear title: Paying Doctors Bonuses for Better Health Outcomes Makes Sense in Theory. But it doesn’t Work. The article cites numerous recent studies, such as this one on the United Kingdom or the United States, that show no effect of giving incentives to physicians either in their clinical practice patterns (inputs) or in health outcomes (outputs). A systematic review of studies on "pay for performance" (P4P) in the Annals of Internal Medicine reaches similar negative conclusions. However, a marked anti-economist tone of these articles emerges that I believe comes from a confusion. According to these articles, economists only know how to prescribe the use of incentives, without evaluating their effects, and also the incentives that we advise are only monetary and don’t appeal to the many different motivations, not only monetary, that can affect doctors decisions.

My response to these criticisms is that they generalize and caricature the profession of economist. It’s obvious that there are many economists with media success or with political influence who talk rubbish, but we should stop listening to them (another day I’ll talk about how to discriminate between one type of economist and another). Not even the best economists would be inclined to recommend incentives based only on theoretical models without a previous empirical evaluation, ideally with randomized experiments. All economists have an extremely limited view of the human being (and the doctor) as someone for whom only financial incentives matter and therefore motivate them. The area of ​ Behavioural Economics, which a growing group of economists are working on, studies precisely the multiplicity of human motivations, as well as our cognitive difficulties to make "correct" decisions in situations of pressure. Both the multiple motivations and the fact of having to make decisions, literally of life or death under stressful conditions, are common characteristics of the day to day of the health professional. Designing incentives that really motivate and guide the doctor to make good decisions is therefore a crucial part of behavioural economists’ work, and not only of the so-called health economists. In the article "When and Why Incentives (Don’t) Work to Modify Behaviour", we review the use of incentives in very similar areas such as education or the creation of healthy habits. Perhaps some of these lessons can be applied to incentives for health professionals.

The problem in the indiscriminate use of monetary incentives among professionals is that at first glance, it seems to respond to that logic of domestic economy that our president, Mariano Rajoy, likes so much: "the more I get paid to do something, the harder I will try and be better at my job". Let's analyze the problem of these three phrases applied to the health field:

1. "The more I get paid": it’s not very clear that any health system, moreover a public one, can offer monetary incentives high enough to compete with the pharmaceutical and health technology industries. In an environment in which these institutions combine financial capacity with very clear and biased interests so that the physician can make decisions that favours them, it doesn’t seem very effective to start an upward race to see who can guide the price to a higher level so that the most convenient decisions are made. As the title of one of the articles by one of the key economists of Behavioural Economics,  Uri Gneezy says, "Pay enough or don’t Pay at All". The problem is that we may never be able to pay "enough" to offset the incentives offered by institutions with other interests. But, beware: the possible motivations of doctors are much broader than simply monetary ones. If these motivations are diverse (prestige and professional stability, the altruistic feeling of improving the patient's health, the good image that a doctor projects towards others ...), so should the possible incentives that, properly designed, can be given to get a better performance from the professional. The health manager should realize that many other tools at his disposal also involve powerful incentive mechanisms for health professionals and can help, if used in a systematic and thoughtful way, to greatly improve performance: job stability, flexible scheduling, greater independence in decision making depending on results, possibility of training in other centres and other techniques, facilitating the ongoing contact with the treated and cured patients and greater transparency on the individual contribution of each doctor. In this recent study by Social Science and Medicine it’s shown that not even doctors who receive monetary incentives have greater job satisfaction than those who do not, so money doesn’t seem to be the only thing that motivates them. In addition, there is the classic problem of the possible negative interaction between different types of incentives: by giving a monetary incentive one can substitute the original motivation for which a person did his job well, for the economic motivation, which can have perverse effects and anyway, it makes it very difficult that once the incentive is withdrawn, the original motivation for "doing things right" is retained. For example, would you still be a blood donor if you were paid to do so from now on?
2. "The harder I try": It doesn’t seem that the primary problem of the health professional is the of lack of effort. On the contrary, it’s not so clear what is translated "to try harder" when there are multiple treatment guidelines. Does the interventional surgeon make more effort than the one who decides on a more conservative treatment? Obviously, one can try to define the "appropriate" treatment, for example, through clinical practice guidelines and then study whether the doctors who receive incentives apply this method more, but even in such clear cases, the existing evidence clearly shows that Incentives work to change behavioural patterns of specialists with highly trained opinions on how to treat their patients. In many cases, resources are squandered by encouraging treatment guidelines that already exist, or the doctor who doesn’t follow the indicated pattern is penalized even in cases in which he should not do so simply because of the patient's individual character.
3. "The better I do my job": The problem is even greater when we want to measure the effectiveness of incentives through results, such as health outcomes. In these cases, the problem is not so much in choosing the variable by which to measure the result, but in interpreting causally that this improvement has really been due to the change in behaviour of a doctor who receives an incentive. Separating historical effects, such as the evolution of a disease, is something that good economists could contribute with their econometric techniques as long as they had sufficiently rich data, ideally designed with an experimental approach, to do so. 

Therefore, the debate between managers and health professionals on the use of incentives should remain open, but should leave behind prejudices and listening better to what seems to motivate professionals. Neither economists only talk about "monetary" incentives, nor should doctors only worry about receiving more money. Let's be smarter by designing incentives that really appeal to the true motivations of the professional, to then evaluate their effectiveness (or lack thereof) according to the correctly defined objectives.