This meta-analysis quantified the effect of FITT (frequency, intensity, time, and type) manipulation on exercise adherence. Databases were searched from inception to 09-2018. Manipulation of intensity resulted in a moderate, significant decrease in adherence (MD: −3.3, 95% CI: −6.1 to −0.5), and for the remaining FITT components, there were trivial-small, non-significant effects. When stratified by chronic disease status, for populations with chronic diseases increasing Intensity resulted in a moderate but non-significant decrease in adherence (MD: −3.6, 95% CI: −7.4, 0.3) and increasing time resulted in a small, significant decrease in exercise adherence (MD: −4.9, 95% CI: −9.4 to −0.4). Additionally, omission of a behavioral model resulted in a moderate but non-significant decrease in exercise adherence (MD: −4.0, CI: −8.3 to 0.2). For healthy populations, FITT manipulation and omission of behavioral model resulted in trivial-small, non-significant effects. In conclusion, manipulation of FITT and use of behavioral models may be of greater importance to exercise adherence in diseased population. In particular, increasing intensity and time and omitting a behavioral model may decrease adherence. The current meta-analysis elucidates gaps in the literature and can assist clinicians and clinical exercise physiologists in determining the importance of FITT manipulation to exercise adherence. K E Y W O R D S adherence, behavioral model, chronic disease, exercise prescription, physical activity | 223 BURNET ET al.