Effectiveness of Non-pharmacologic Interventions on Device-measured Physical Activity in Adults With Cancer, and Methodology Used for Assessment: A Systematic Review and Meta-analysis

Abstract

Objective: To investigate the effectiveness of different types of interventions aimed at enhancing device-measured physical activity (PA) and summarize the devices and methodologies used to measure PA in adults with cancer. Data Sources: A systematic review was prospectively registered on PROSPERO (CRD42020199466). The search was conducted in PubMed, The Cochrane Library, EMBASE (via Ovid), and PEDro from 2005 onward. Study Selection: Prospective interventional studies (randomized controlled trials [RCTs], non-randomized controlled trials, and single-group trials), that included adults within 12 months from cancer diagnosis, and device-measured PA before and after commencement of an intervention were included. Studies were excluded if PA was measured at a single time point. Two independent reviewers screened 3033 records and 30 articles met the inclusion criteria. Data Extraction: Two reviewers independently extracted the data. PEDro scale and GRADE approach were used to assess methodological quality of RCTs and overall quality of evidence, respectively. A meta-analysis of relevant RCTs was conducted. Data Synthesis: Thirty studies were identified, mainly including adults with multiple cancer types. Interventions were behavior change interventions (n=15), exercise training (n=13), neuromuscular electrostimulation (n=1), or a nutritional program (n=1). The meta-analysis showed improvements on moderate-to-vigorous intensity PA (MVPA) in the experimental group (8 studies; standardized mean difference (SMD)=0.23; 95% CI 0.06-0.39); with subgroup analysis showing that findings were mainly driven by behavior change interventions (5 studies; SMD=0.23, 95% CI 0.05-0.41). An uncertain effect on sedentary behavior, daily steps, and light intensity PA was found. PA was measured with medical devices and commercial wearables, quality of the methodology was variable. Conclusions: Behavior change interventions increased device-derived MVPA in adult cancer patients who underwent the intervention within 12 months of the cancer diagnosis. Various devices and methodologies were used to assess PA, which limits comparisons across the studies.