Predictors of adherence, contamination and dropout in home-based walking by lung and oesophageal cancer patients from two randomised control trials: An exploratory study

Objectives: In this study, we examined predictors of exercise adherence, contamination and dropout in lung and oesophageal cancer patients who participated in two randomised controlled trials. Methods: We used data on 188 lung and oesophageal cancer patients from two previous studies (intervention: moderate-intensity walking for 12 weeks). Baseline measurements included demographic variables, disease characteristics, Hospital Anxiety and Depression Scale and Bouchard 3-day physical activity (PA) record. We used multiple linear and logistic regressions to analyse predictors of exercise adherence in the walking group, contamination in the control group and dropout in both groups. Results: Pre-intervention exercise habits and baseline depression scores predicted adherence, with an explanatory power of 16.7% (p < 0.0001). Pre-intervention exercise habits (odds ratio [OR] 19.65, 95% confidence interval [CI] 2.76–139.97), baseline moderate PA (min/day) (OR 1.03, 95% CI 1.01–1.05) and baseline vigorous PA (min/day) (OR 1.09, 95% CI 1.01–1.18) predicted contamination. Baseline mild PA (10 min/day) (OR 0.94, 95% CI 0.89–0.99) predicted dropout. Conclusions: Pre-intervention exercise habits and baseline depression levels predicted exercise adherence in the walking group. In the control group, pre-intervention exercise habits and baseline moderate and vigorous PA predicted contamination. Baseline mild PA predicted dropout rates in both groups.