Physical Activity Interventions in Older Adults: A Systematic Review of Reviews

Introduction: The proportion of people meeting recommended physical activity guidelines declines with age. We conducted a systematic review of existing reviews to assess evidence on interventions to encourage older people to be more active.

Methods: Six electronic databases were searched in June 2015 in line with our protocol (PROSPERO: CRD42013004333). Older people were defined as those aged 50+. Two reviewers independently screened title, abstracts and reference lists, extracted data from included studies, and assessed the quality of reviews using the AMSTAR and Progress-Plus checklists.

Results: Seventeen papers reporting on 15 reviews including between 6 and 38 studies were included. Eleven focussed specifically on older adults (mean age range: 60 to 77 years) and 4 reported sub-group analyses for older adults. Eleven included various types of interventions (e.g. supporting lifestyle change, walking groups, exercise classes); the remainder focussed specifically on walking, remote, text-message and pedometer-based interventions. Five meta-analyses reported small (e.g. d=0.14) to medium (e.g. d=0.52) effects on physical activity, with moderate to high heterogeneity. Narrative reviews reported mixed findings. Some behaviour change theories (e.g. social cognitive theory) and techniques (e.g. problem solving) were associated with more successful interventions across multiple reviews. Ascertaining other intervention and sample characteristics linked to effectiveness was hampered by limited reporting and some contradictory findings.

Conclusions: Our findings will be informative to researchers, intervention designers, and policy-makers but need to be interpreted in light of the generally poor quality of the reviews themselves and the quantity, quality and diversity of their included studies.