The impact of financial incentives on physical activity: a systematic review and meta-analysis

Objective: To evaluate the effects of financial incentives on physical activity (PA). Data Sources: MEDLINE, Embase, 7 other databases, and 2 trial registries until July 17, 2019. Study Inclusion and Exclusion Criteria: Randomized controlled trials with adults aged ≥18 years assessing the effect of financial incentives on PA. Any comparator was eligible provided the only difference between groups was the incentive strategy. Data Extraction: Two independent reviewers extracted data and assessed study quality. Of 5765 records identified, 57 records (51 unique trials; n = 17 773 participants) were included. Data Synthesis: Random-effects models pooling data for each of the 5 PA domains. Results: Financial incentives increase leisure time PA (gym or class attendance; standardized mean difference [95% CI], 0.46 [0.28-0.63], n = 5057) and walking behavior (steps walked; 0.25 [0.13-0.36], n = 3254). No change in total minutes of PA (0.52 [−0.09 to 1.12], n = 968), kilocalories expended (0.19 [−0.06 to 0.44], n = 247), or the proportion of participants meeting PA guidelines (risk ratio [95% CI] 1.53 [0.53-4.44], n = 650) postintervention was observed. After intervention has ceased, incentives sustain a slight increase in leisure time PA (0.10 [0.02-0.18], n = 2678) and walking behavior (0.11 [0.00-0.22], n = 2425). Conclusions: Incentives probably improve leisure time PA and walking at intervention end, and small improvements may be sustained over time once incentives have ceased. They lead to little or no difference in kilocalories expended or minutes of PA. It is uncertain whether incentives change the likelihood of meeting PA guidelines because the certainty of the evidence is low.