Predicting adherence of adults to a 12-month exercise intervention.

Background: The purpose of this study was to (a) examine demographic, psychosocial, and physiological predictors of exercise adherence in a yearlong exercise intervention and (b) describe the trajectory of adherence over time.

Methods: Participants were 51 men and 49 women aged 40 to 75 years. The supervised and home-based intervention consisted of 60 minutes/day, 6 days/ week of moderate-to-vigorous intensity exercise. Three adherence measures were used: (1) minutes/week, (2) MET-hours/week, and (3) change in cardiopulmonary fitness (VO2max). Predictors of adherence were determined separately by sex using mixed models and multivariable regression.

Results: Participants performed 287 ± 98 minutes/week of moderate-to-vigorous activity with 71% adhering to at least 80% (288 minutes/week) of the prescription. Men adhered better than women (P < .001). Among women, adiposity-related variables were significantly related to poorer adherence on all 3 measures (P < .05). A less consistent pattern was observed among men but in follow-up analyses, adiposity was associated with fewer MET-hours/week of exercise. Social support, pain, and perceived benefits were predictive in some models. Men and nonobese women experienced peak adherence at 4 to 6 months, while obese women peaked during months 0 to 3.

Conclusions: When provided with supervision and support, previously sedentary men and women can achieve and maintain high levels of aerobic activity.