Can physical activity measurement alone improve objectively-measured physical activity in primary care? A systematic review and meta-analysis

There is evidence that simply measuring physical activity alone can increase self-reported physical activity behaviour. The aim of this review was to describe changes in objectively-measured physical activity within control groups in primary care physical activity intervention studies. Five electronic databases (PubMed, MEDLINE, SPORTDiscus, PsychINFO and CINAHL) were searched from inception to February 2019. Physical activity controlled intervention studies objectively measuring physical activity in primary care with adults were included and meta-analyses were completed. Thirty studies were eligible and 22 studies were included in the meta-analysis. No statistically significant change in steps.day⁻¹, counts.day⁻¹ and counts.minute^-1 were found in the meta-analyses within control groups. Moderate-to-vigorous physical activity minutes.day⁻¹ significantly decreased (-3.97; 95% CI −6.31 to −1.64; P < 0.001). Sub-analyses revealed there was a trend for steps.day⁻¹ to increase in participants < 50 years old (504; 95% CI −20 to 1029; P = 0.06). Noteworthy increases (≥10%) in objectively-measured physical activity within control groups were found in 17% of studies. Noteworthy increases were reported in studies with younger participants, one-third of the pedometer studies, one-third of studies with participants at risk of chronic disease and in studies with a shorter duration between measurements. No control group improvements were found in participants with chronic disease. Overall, no significant improvements in objectively-measured physical activity were found within control groups in primary care. Further investigation of noteworthy increases in control group physical activity levels is indicated, particularly in certain sub-groups of participants as this may effect physical activity research and interventions in these populations.