In high-income countries, 41 % of men and 48 % of women have an inactive lifestyle, based on the World Health Organisation (WHO) Global physical activity guidelines [1, 2]. According to the WHO, physical inactivity is defined as not adhering to physical activity guidelines, thus spending less than 150 min of moderate-intensity aerobic physical activity throughout the week, or less than 75 min on vigorous-intensity aerobic physical activity throughout the week or less than an equivalent combination of moderate—and vigorous-intensity activity [2]. Physical inactivity has negative consequences for people’s health, as it is the fourth leading risk factor for mortality worldwide and it increases the risk of cardiovascular diseases, obesity and diabetes [1–3]. Physical activity can reduce the risk of several chronic conditions, such as diabetes and cardiovascular diseases. Moreover, it is associated with more favorable outcomes in the course of disease. If people would achieve the recommended level of activity, an all-cause mortality risk reduction of almost 30 % would be possible [4]. Still, a substantial proportion of the high-income population is insufficiently active. It is therefore important to find ways to improve physical activity levels, particularly among those who are the least active. However, behavior such as physical activity is complex and therefore difficult to change, implying a serious challenge concerning program adherence and maintaining results after program completion [5, 6].

One setting from which physical activity programs are initiated is the healthcare setting. Many people with (a high risk of) a chronic disease are already within the healthcare setting for treatment of their condition. For these people being physically active to a sufficient extent may be important to prevent a deterioration of their condition. At the same time, healthcare providers can play an important role in motivating patients to participate in a physical activity program [7]. However, research shows that long-term adherence varies greatly between 10 % and 80 % in therapeutic exercise interventions for diabetes patients [8]. There are many reasons that people find it difficult to adhere to exercise schemes, one of which is motivation One of many ways to address motivation is to include financial incentives in the intervention.

Financial incentives provide economic encouragement for people to show desired behavior, such as increasing their physical activity level [9]. Incentives can be either positive or negative. Positive incentives reward individuals either for participation or for when they fulfill the desired outcome of certain health behavior. Negative incentives or disincentives penalize individuals if they do not participate, or if they do not meet the required outcomes established [10].

Financial incentives have the potential to affect both participation rates and program adherence [11, 12]. An important point to address however when studying and discussing effectiveness of financial incentives on behavioral change, is the general notion that a financial incentive constitutes an external motivation for changing behavior. According to the health promotion literature, people need skills and knowledge (intrinsic motivation) to change their lifestyle behavior and simply giving them a financial incentive is not expected to teach them these skills [10, 13, 14]. Building intrinsic motivation takes time and needs work, but financial incentives may help, for instance to increase program adherence to an intervention that teaches these skills and knowledge. Financial incentives can be provided on many levels in healthcare, for example incentives for insurers to promote the financing of exercise programs, for healthcare providers to incorporate physical activity in treatment and rehabilitation, for employers to establish training facilities at work places, or for patients to participate. The providers of the incentives also vary, depending on the healthcare system in a country. Incentives can be provided by the government, insurers, employers or non-profit organizations. The government may have an interest in this, if the benefits to society and/or the government budget (in terms of potential for saved healthcare spending in the long run) exceed the cost of providing the incentive. Similar rationales may apply for insurer—and employer-financed incentive schemes.

Hypotheses on the effectiveness of direct financial incentives to improve physical activity levels vary. One opinion is that offering rewards may be counterproductive in the sense that this extrinsic motivation may crowd out the intrinsic motivation already present. Therefore any increase in physical activity during the time of the intervention, as well as part of the activity level present before the intervention started, will disappear after the incentives are removed [15–17]. A competing hypothesis states that getting people interested in physical activity by giving financial incentives may very well contribute to habit formation. This theory assumes that if exercising is a form of habitual behavior, giving financial incentives to motivate people to exercise for a certain period, may increase future utility from exercising [15, 18]. Previous studies on the effect of financial incentives to change relatively simple health-related behaviors, such as attending appointments at clinics and take up of child immunization, indicate that financial incentives are effective [10, 15]. Systematic reviews on effectiveness of financial incentives to increase physical activity showed positive results in both community- and school setting, particularly in the short term [11, 12]. No such systematic review has been carried out for the healthcare setting. The objective of this study was to systematically review the literature with respect to the effectiveness of direct financial incentives used to promote physical activity in the healthcare setting.

The objective of this systematic review was to provide an insight in the effectiveness of financial incentives used for physical activity promotion in the healthcare setting. The search revealed only three eligible studies (two RCTs among adults and one among children) that specifically studied the effect of a financial incentive on improving physical activity measured by physical activity outcomes or weight loss [19–21]. Two of the three studies combined a financial incentive with other methods, such as motivational interviewing or nutrition classes [19, 20]. Despite short-term differences between intervention groups in one study, no differences were found between the control and intervention group over a longer period of time (12 months) in these studies [19, 20]. The study of Islam measured only short term effects and found almost no significant improvements in the intervention group [21]. The included studies do not indicate that financial incentives stimulate physical activity in the healthcare setting.

Two studies included in this review found no long-term effects of the financial incentive. The third study did not measure long-term effects, but did not find important effects in the short term [21]. Harland et al. found some short-term effects. Possibly, the duration and/or intensity of intervention activities in these studies were not enough to alter behavior, since effects regardless of the incentive were small or absent. A well-known physical activity intervention strategy in the healthcare setting is exercise on prescription, which is usually integrated into multidisciplinary combined lifestyle interventions. Such programs tend to include physical activity promotion, improvement of diet, and reduction of psychological barriers using motivational interviewing [22]. Two studies included in this review did not consist of a strong and structured physical activity component, which might have caused participants to focus on other aspects of the intervention than actually becoming physically active [19, 20]. The study of Islam had a structured physical activity component, but the duration was just six weeks [21].

Although the effectiveness of financial incentives on increasing physical activity levels and accomplishing weight loss was generally absent in our review, in other settings, such as the community setting, at least short term effects of financial incentives on physical activity behavior were found [11, 12]. The review of Mantzari et al. has evaluated the effect of financial incentives on health-related behavior, which includes for example healthier eating, physical activity, and smoking cessation. In this review it is also acknowledged that effects are not sustained when the incentive is removed [23].

In all three studies included in our systematic review, a motivation was lacking as to why this particular incentive was chosen for the particular population. It is likely that preferences for a certain type of financial incentive differ between target groups. For example, women may be more risk adverse than men so a financial incentive in the form of a lottery might not be as effective for men as for women [24]. If the specific type of incentive does not fit the preferences of the target population, this may partially explain the lack of its effect on behavior. There is research available that elucidates the importance of some attributes of financial incentives. A broader scoped review on the effectiveness of financial incentives on physical activity showed that for an incentive to be effective it should at least be conditional to the targets set in the intervention [25]. Promberger et al. [26] have performed a discrete choice experiment on the acceptability of financial incentives to change health related behavior. They have found that a preference for the type of incentive for smoking cessation is different than the preferred incentive for weight loss [26]. Moreover, the size of the incentive matters [10] and includes an optimum [27]. Therefore, one important recommendation would be to study preferences of the target group to determine a suitable financial incentive before designing and implementing a study.

In a recently published review of reviews the effectiveness of physical activity promotion interventions in the primary care are shown. These interventions seem to have small positive effects [28]. Combining a lifestyle intervention with a financial incentive that is preferred by the target population, might increase the effects on physical activity levels of the individuals. Future research should focus on the most effective combination of the lifestyle intervention and the preferred financial incentive of the target population.

Theoretically, the benefits of the investment in a financial incentive returns to the provider of the incentive, for example in the form of decreased use of healthcare. In national health systems such as in the UK, the provider of the incentive in the healthcare setting is automatically the collector of the benefits. In managed competition systems, insurers might be the provider of incentives with the underlying principle of return on investment, but also gain a competitive advantage in a market with many healthcare insurance providers. It should be acknowledged that financial incentives in the healthcare systems of developing countries might be a bridge too far. The theory of return on investment is a concept that might function as well in healthcare as in the work setting. A review shows that giving incentives in the work setting to employees by providing free wellness programs, and sometimes incentives to increase participation, returns in less healthcare expenditures and less costs for absenteeism [29]. As mentioned before, the present systematic review includes only three studies. We believe however that this is a true reflection of the level of knowledge, despite the fact that the use of financial incentives is fairly common. For example, during many physical activity interventions, participants can freely access sports and/or leisure accommodations or they receive a small reward for participating in the intervention [30, 31]. However only a few studies explicitly address the effectiveness of the incentive given in a separate arm of the study, as was one of the inclusion criteria in our study. There were some studies excluded from the review that stated as their aim to evaluate the effect of changing physical activity behavior by giving financial incentives. A closer look at the study methods revealed that this statement could not be justified because of different reasons. These sub-optimal study designs prevented drawing definite conclusions on the effectiveness of financial incentives on physical activity behavior, because for example the effect of the financial incentive could not be distinguished from the other components of the study or the study did not have a control group [24, 30–32].

We decided not to perform a quality check for the included studies. With a yield of only three very diverse interventions addressing the effect of financial incentives on physical activity our review, although systematic in nature, may be characterized as explorative rather than thoroughly addressing the effectiveness of financial incentives in promoting physical activity from the healthcare setting.

One could argue that extending our search with other databases such as EconLit, Psychlit and Sportsdiscus might have increased the yield of the review. However, if we would have missed a key publication, we would have expected it to be found through reference tracking of the studies already included. The limited set of appropriate study designs is confirmed in other systematic reviews. Two other systematic reviews evaluating the effect of financial incentives on physical activity irrespective of the setting included as few as 10 and 11 studies [11, 12]. Moreover, most of the studies included in these reviews defined ‘attendance’ as the incentivized behavior instead of behavioral change. This could also partly explain why few studies are found to be effective in actually changing physical activity behavior. Perhaps incentives may only offer the particular behavior that has been incentivized.

Few studies have evaluated the effect of a financial incentive on changing physical activity behavior in the healthcare setting. The three studies included in this systematic review did not show effects that could be attributed to the incentive used. However, study designs were not particularly strong and there seems to have been little thought given to whether or not particular incentives suit particular study populations. Nevertheless, based on results in other settings, financial incentives seem promising instruments to increase people’s physical activity.

It is recommended that in future research on the effectiveness of financial incentives on physical activity some basic requirements are met. First, the study protocol should include intervention arms in such a way that effectiveness of incentives can be studied. Second, it is recommended to first study the preferences of the target population with regard to financial incentives to maximize the chance that the incentive will indeed help to increase the intended behavior. Assuming that the control condition will include a program aiming to increase physical activity, it is recommended to consider multidisciplinary combined lifestyle interventions in order to maximize the chance of habit formation and long-term maintenance of behavioral change.

BMI, body mass index; C, control group; EBBS, exercise benefits/barriers scale; ECS, exercise confidence scale; GP, general practitioner; I, intervention group; IPAQ-S, international physical activity questionnaire – short; PA, physical activity; RCT, randomized controlled Trial; YMCA, young men’s christian association; WHO, World Health Organisation; WHR, waist-to-hip ratio