Incentives to change unhealthy habits: Do they work? Are they ethical?

Pedro Rey

In a previous post I talked about the use of economic incentives to improve the quality of the prescription that physicians make. Today I want to focus on the other side of the problem, that of patients who try to self-stimulate or receive external incentives, to change their life habits towards healthier behaviour. There are two questions of interest for me: 1. Do incentives work? 2. Is it "ethical" to use "economic" incentives to induce changes towards healthier behaviours?

Operationally the evidence is mixed. In this systematic review, it’s argued that in general incentives work better than "doing nothing" but also that their effect is diluted over time and that when they are removed - something that may be necessary because they are not sustainable - the effect tends to disappear. This recent article by Miguel Ángel Máñez also point us to some mobile applications that pay and charge users depending on whether they meet their objectives. In this older article, co-authored with with Uri Gneezy and Stephan Meier, we reviewed the pre existing evidence of programs that reward for going to the gym or for not smoking. Regarding the former, it seems that it’s possible to increase the frequency with which people go to the gym, if paid for each visit, which also has a positive impact in the medium term on health indicators, such as the body mass index. We also show that if the incentives to go to the gym are large enough and maintained for a sufficient period of time, those individuals get hooked to exercise, and thus, once the incentives disappear, it’s possible to keep the individuals continue exercising. However, in this study done with university students, it’s observed that the change in behaviour is very sensitive to minimal environmental changes and, therefore, at the moment when the habit is broken by an external cause (for example, he Easter Break holidays), the treatment group that received the incentives returns to the same sedentary behaviour of the control group.

Even more intricate, and with even weaker results, is paying for something to which one is addicted, such as tobacco. In this case, the addiction can become so strong that, although incentives may have worked relatively well in the short term, they may have little success in creating ex-smokers in the long run. However, sometimes the short term is vital if, for example, with the use of incentives we can get women to quit smoking during pregnancy.

With regard to the question of ethics, it’s convenient to clarify that not all "economic" incentives have to be "monetary" and that there is no other reason than convenience, to offer incentives in the form of money and not other currencies that matter to people. Money is not "unethical" nor is the incentive a bribe. What an incentive does is to add or replace the initial motivation for which someone performs certain behaviour. In many occasions the individual needs an additional motivation because, under the existing conditions, he doesn’t achieve the sought after change of habits. The problems of self-control, lethargy and fear of uncertainty are some of the reasons that prevent them.

Let's start with ambiguity. Faced with a health problem, the patient has to deal with the double difficulty of not knowing at the same time the real consequences of his health problem and the efficacy of the potential behavioural change that he attempts. For example, not all smokers develop lung cancer and not all smoking cessation programs prevent it. Faced with this ambiguity, and given the cost of changing habits, we tend to minimize the importance of the immediate risks to our health from our current behaviour and to distrust the efficiency of the changes. Add to this the fact that, in many cases, the results of changes in behaviour are not visible in the very short term. That is, we incur a certain cost today, perhaps to obtain an uncertain profit tomorrow but on many occasions our "self" of tomorrow, until it arrives, matters less than our "self" of today, with a desire for nicotine, sugar and too lazy to start exercising.

Given these problems, the multiple programs to quit smoking, to lose weight or to exercise propose "incentives" that aim to influence both the uncertainty about the real risk to health (with more precise and individualized information on the specific problem of the patient) or about the effectiveness of changes in habits (for example, showing photos of the "before and now" of patients who followed a treatment), such as bringing the present self closer to the future self. Among these last suggestions are all those programs that either transfer some of the rewards of changing habits to the present moment or increase the costs of not doing so. For example, many programs offer small instant reward for making small changes ("if today you have met your exercise goal, give yourself a reward with a little candy", or some mobile applications that send messages of encouragement or scold you when they detect your undesirable behaviour). Even more interesting are the programs in which the social cost of not changing habits is increased, so that one suffers social pressure when it fails to fulfil commitments acquired publicly in front of their social group.

All these examples are "economic incentives". But what is most talked about lately is explicit monetary incentives for changes in behaviour. For example, tax breaks for the obese who manage to lose weight, higher indirect taxes on tobacco or unhealthy foods or more directly, pay people to lead a healthier life. The key to the ethical problem is not whether money is given or not, but whether the incentive replaces the initial motivation to perform the behaviour. I think that in other environments such as education, the ethical problem is more important, since, when considering whether to "pay children to study", we enter into the debate about whether the goal of education is the acquisition of knowledge or teaching to take responsibility and have a real taste for learning. In the case of health, I consider that the debate on motivations is less important, and what is really crucial, given the cost to public health systems of unhealthy individual behaviour, is whether resources can actually be saved by inducing behavioural changes through the provision of incentives. I realise that I may appear paternalistic, but, as an economist, I prefer for the moment, to continue focusing on my experiments to try to find the incentive that works best for each situation and for those who really ask for help to change their behaviour, and leave for later ethical debate on whether, once found, they should be applied to the population globally.