Background: Although physical activity (PA) interventions are frequently reported to be effective, there are criticisms that outcomes are often self-reported (and as a result unreliable) and are measured only in the short term (up to 12months). To influence health in the long term, there is a need to identify longer-term (and objectively valid) physical activity behaviour changes. Purpose: To examine the long-term effects of physical activity on health using objective measures.

Relevance: This review brought insight in the effects of physical activity measured objectively in the management and prevention of chronic conditions.

Methods: A Systematic review using Cochrane recommended reviewing methods. Electronic databases were searched to 2^nd March 2019. The inclusion criteria were Randomised control trials (RCTs), adults from and above 18 years at high risk of chronic diseases, Lifestyle PA interventions, objective measures of PA, follow-up from 24 months and beyond. Exclusion criteria were studies which do not meet inclusion criteria. Risk of Bias was assessed by 2 independent reviewers using revised RoB2. Data extracted included study Characteristics, Outcomes measures (moderate to vigorous physical activity (MVPA) in minutes/week and Steps/day) at different time points and examined intervention characteristics to identify components associated with effectiveness.

Analysis: A meta-analysis was completed in Review Manager version 5.3, and the results were displayed from fixed effects and effects size in Standard Mean Differences (SMD 95% CI).

Results: Six studies examined physical activity intervention on daily steps 24 months post-baseline and the meta-analysis indicates a small significant effect (0.17 95% CI 0.09 to 0.25). Four trials extended further (from 36 to 48 months) and similar results were found (0.17 95% CI 0.07 to 0.27). Five trials assessed PA intervention on time spent in MVPA 24 months post-baseline and the meta-analysis showed a small significant effect (0.19 95% CI 0.11 to 0.28). MVPA at 36 to 48 months remained similar effect size was small (0.19 95% CI 0.12 to 0.25). Statistical results converted in clinical value were (up to 56min/week and 579steps/day). Methods used to measure outcomes, large sample size and additional support during follow-up were associated with the effectiveness of PA in both outcomes (daily steps and MVPA). The methodological quality assessment was moderate to high.

Conclusion: This review shows that physical activity interventions measured objectively were post-baseline for promoting physical activity in the long-term, particularly if intervention programmes included the components highlighted above.

Implications: Statistical results showed small effects sizes and clinically meaningful. Evidence supports that relatively modest changes in PA (30-60min/week of moderate-intensity) modify key cardiovascular risk factors (e.g. cholesterol, blood pressure) to a clinically meaningful extent. Furthermore, difference in walking activity of 40 min/week is associated with around 10% reduced risk of non-communicable diseases. Future research is needed to investigate the factors associated with increased physical activity level to meet guideline recommendations.

Keywords: Physical activity intervention, objective outcome measure, adult, randomised control trials

Funding Source No funding received

Ethical approval Not applicable