Intended for healthcare professionals

Feature Public Health

Betting on health

BMJ 2009; 338 doi: (Published 23 April 2009) Cite this as: BMJ 2009;338:b1456
  1. Karen McColl, freelance writer
  1. 1Savoie, France
  1. karen{at}

    New websites in the US are encouraging people to make public commitments to change their behaviour. Karen McColl investigates whether they work and the implications for public health policy

    Increasingly, people who are desperate to change their behaviour are putting their money on the line to motivate them towards success. But can placing a public bet really help people to lose weight, quit smoking, or exercise more? Could this type of commitment contract be a valuable tool for promoting public health?

    Over 20 000 people have publicly signed up to change their behaviour at the online commitment store StickK (pronounced stick—the silent second letter k refers to the legal shorthand for contract) since the website launched in January 2008. Of these, about a third have placed a financial stake—promising to hand over a total of $1.28m (£870 000; €960 000) if they fail to meet their goals.

    It may seem surprising that people chose to put their money at risk in this way, especially since many would already have tried to change their behaviour and failed. But Yale economics professor and StickK cofounder, Dean Karlan, thinks the explanation is clear. “Most people have something in their life that they would like to change. That’s why we make New Year resolutions—yet most people can think back to their previous resolutions and realise that they didn’t succeed. So it makes sense that offering someone a new way to meet their goal should generate very high take-up rates,” he explains.

    The self made contracts at and similar websites such as and are examples of commitment contracts. The idea behind a commitment contract is that people choose a goal for changing an aspect of their behaviour, make a public commitment to change, and in some cases put up a financial stake that they are prepared to forfeit if they fail. Of course, the idea of smokers and dieters placing bets with friends and family to help them achieve their goals is hardly new. But commitment contracts formalise the process to try to improve the chances of success.

    Over half of the contracts on StickK concern weight loss or weight maintenance, with exercise related contracts the second most popular. About a third of users pledge to give money to an individual or an organisation if they don’t meet their goals. Around $25 000 has been earmarked for “anti-charities”—causes that the participant does not want to support. According to the website, 80% of participants who make the contract conditions as stringent as possible—in other words, people who put up a financial stake, who nominate a referee to verify results, and who designate an anti-charity—meet at least half of their targets.

    Behaviour theory

    Professor Karlan, like many exploring the use of commitment devices, is a behavioural economist. This relatively young branch of economics brings psychology into economic theory to explain how humans behave in the real world. Behavioural economics made a big splash in public policy last year with the publication of Nudge by Richard Thaler, considered by many to be the inventor of behavioural economics, and Cass Sunstein.1 The book argued that policy makers need to change the “choice architecture” to nudge people towards making healthy decisions.

    A crucial part of the theory behind commitment contracts is the idea that we tend to devalue the future—in other words, we tend to put a higher value on any immediate costs or benefits than on those which we’ll experience in the future. In effect, the value of the longer term benefits of quitting smoking, resisting a high energy snack, or going for a run are discounted more than the immediate rewards of one last cigarette, a chocolate bar, or postponing the run until tomorrow. Because we knock down the value of future benefits in this way, it can be harder to resist the temptation of a smaller reward here and now.

    And because we also discount future costs, we are enthusiastic about the idea of changing our behaviour in the future. But when tomorrow comes, the costs of this behaviour change have increased and we often change our minds.

    “The commitment contract is a very simple way of making tempting options more expensive so that we do not choose them,” explains Professor Karlan, whose own success in losing weight by placing a hefty bet with a friend inspired him to set up StickK. “I wanted to be able to go into a restaurant and have a nice healthy meal and then to look at the dessert menu and the price, instead of being $20, was effectively $200 or even $1000. That’s what this contract does—it takes the vices which we have and it makes them more expensive.”

    Health policy

    Public health researchers have long experimented with using contracts between health professionals and patients to promote behaviour change. A Cochrane review of 30 studies examining different types of contracts found that there is not enough reliable evidence to recommend their routine use in improving adherence to treatment, prevention, or health promotion activities.2 Some of the contracts studied included financial incentives or rewards, and there is growing interest in the use of incentives to promote behaviour change. Earlier this year, for example, Eastern and Coastal Kent NHS attracted considerable media commentary when they called for participants to take part in a weight loss incentive trial offering rewards of up to £425 for people who lose weight.3

    There are, however, relatively few examples of health policy makers using commitment contracts—where individuals themselves decide on the contract terms and put their own money at risk—to promote health. In one such example, the Swiss federal office of sport and its partners ran a high profile “health bet” initiative in 2002 to encourage physically inactive people to become more active. Participants could set their own goals. Those who succeeded were entered into a lottery for high value prizes and those who did not were encouraged to make a donation to a disabled sports charity.

    Another example is a randomised controlled study among 2000 smokers in the Philippines conducted by Professor Karlan and colleagues.4 They offered smokers a savings account into which they would deposit money. Urine tests for nicotine and cotinine were carried out after six months and 12 months. Those who passed the six months test had their money returned; the money of those who failed was forfeited to a local orphanage. Eighty three smokers took up the offer and 29 passed the six month test.

    Is there any evidence that commitment contracts work? The results of the Swiss health bet initiative were “very disappointing” according to Brian Martin, one of the researchers who implemented the project. Despite extensive publicity through national media, only 55 people volunteered to take part. Furthermore, only 35 applicants met the definition of “physically inactive” to be eligible for inclusion and, of these, only eight succeeded in meeting their activity goals. Martin and colleagues concluded that one reason for the poor take-up might have been the fact that many people who fall within the technical definition of physically inactive simply do not see themselves in that way.

    The results of the Philippines study were more encouraging. The smokers who took up the offer of the savings scheme were around 30 percentage points more likely to stop smoking than they would have been if they had not been offered the scheme.4

    Interest is growing in the use of commitment contracts in developing countries. Development economists are studying the potential of such contracts to help with a wide range of development issues—such as encouraging people to save for the future or to improve take-up and retreatment of insecticide treated bed nets.5 6

    Can commitment contracts have a role in health promotion in richer countries? Tammy Boyce, research fellow at the King’s Fund, has been assessing current NHS practice on promoting behaviour change. She is wary of transplanting initiatives that have been tried in developing countries or in other sectors—such as finance—to health promotion. “Health is very different from something like saving money. Our behaviours in terms of eating, for example, are so much more ingrained on a daily basis,” she said. “It’s really about changing our wider, societal behaviours.”

    It is clear that use of commitment contracts alone won’t be enough to ensure we all make healthy choices all of the time. After all, two thirds of those who had voluntarily signed up to the contract in the Philippines still failed to quit smoking. Behavioural economists are well aware that commitment contracts won’t work for everyone—they are designed for people who are self aware enough to recognise that they need to do something different to be able to change their behaviour. Further work is needed to evaluate how effective commitment contracts are in bringing about long term behaviour change. So, while commitment contracts may prove to be a useful addition to the global health promotion toolbox, wide ranging measures to improve public health—such as tobacco control legislation, changes to the obesogenic environment, and measures to tackle social inequalities—will remain fundamental.


    Cite this as: BMJ 2009;338:b1456



    View Abstract