Prevalence and predictors of dropout from high-intensity interval training in sedentary individuals: A meta-analysis

Recent evidence suggests that high-intensity interval training (HIIT) is an effective method to improve fitness and various health markers. However, the tolerability and acceptability of HIIT among sedentary individuals is currently controversially discussed. Therefore, our objective was to investigate the prevalence and predictors of dropout among sedentary individuals in HIIT-based exercise interventions. MEDLINE/PubMed, SPORTDiscus, and Web of Science were searched systematically for relevant articles until 06/2018. Studies included were required to (a) be written in English, (b) include sedentary healthy adults, (c) use some form of HIIT without any complementary intervention, (d) last ≥4 weeks, (e) report detailed description of the applied HIIT protocol, and (f) report data that allow calculation of a dropout rate. Fifty-five studies reporting results from 67 HIIT interventions with 1318 participants met the eligibility criteria. The trim and fill adjusted pooled dropout rate across all interventions was 17.6% (95% confidence interval 14.2-21.5%). Dropout rates were significantly lower in cycling-based interventions compared with studies using running/walking as exercise modality (P < 0.001). Longer session time (β = 0.02, P < 0.05), higher time effort/week (β = 0.005, P < 0.05), and overall time effort/intervention (β = 0.0003, P < 0.05) predicted greater dropout. Exercise intensity was not related to dropout. Our data suggest that HIIT-based interventions are tolerable and acceptable for previously sedentary individuals, exhibiting generally lower dropout rates than commonly reported for traditional exercise programs. Given the association between HIIT volume and dropouts, future studies should further focus on identifying the minimally effective dose of practical HIIT for improving health status. Such efforts would be important to increase implementation and public health impact of HIIT.