Systematic review and meta-analysis of maintenance of physical activity behaviour change in cancer survivors

Background Physical activity can improve health and wellbeing after cancer and may reduce cancer recurrence and mortality. To achieve such long-term benefits cancer survivors must be habitually active. This review evaluates the effectiveness of interventions in supporting maintenance of physical activity behaviour change among adults diagnosed with cancer and explores which intervention components and contextual features are associated with effectiveness. Methods Relevant randomised controlled trials (RCTs) were identified by a search of Ovid Medline, Ovid Embase and PsychINFO. Trials including adults diagnosed with cancer, assessed an intervention targeting physical activity and reported physical activity behaviour at baseline and ≥ 3 months post-intervention were included. The behaviour change technique (BCT) taxonomy was used to identify intervention components and the Template for Intervention Description and Replication to capture contextual features. Random effect meta-analysis explored between and within group differences in physical activity behaviour. Standardised mean differences (SMD) describe effect size. Results Twenty seven RCTs were included, 19 were pooled in meta-analyses. Interventions were effective at changing long-term behaviour; SMD in moderate to vigorous physical activity (MVPA) between groups 0.25; 95% CI = 0.16–0.35. Within-group pre-post intervention analysis yielded a mean increase of 27.48 (95% CI = 11.48-43.49) mins/wk. of MVPA in control groups and 65.30 (95% CI = 45.59–85.01) mins/wk. of MVPA in intervention groups. Ineffective interventions tended to include older populations with existing physical limitations, had fewer contacts with participants, were less likely to include a supervised element or the BCTs of ‘action planning’, ‘graded tasks’ and ‘social support (unspecified)’. Included studies were biased towards inclusion of younger, female, well-educated and white populations who were already engaging in some physical activity. Conclusions Existing interventions are effective in achieving modest increases in physical activity at least 3 months post-intervention completion. Small improvements were also evident in control groups suggesting low-intensity interventions may be sufficient in promoting small changes in behaviour that last beyond intervention completion. However, study samples are not representative of typical cancer populations. Interventions should consider a stepped-care approach, providing more intensive support for older people with physical limitations and others less likely to engage in these interventions.