Physical Activity Interventions And Chronic Diseases: A Matched-Pair Analysis Comparing Cochrane And Non-Cochrane Systematic Reviews

TY - JOUR

T1 - Physical Activity Interventions And Chronic Diseases: A Matched-Pair Analysis Comparing Cochrane And Non-Cochrane Systematic Reviews

AU - Hacke, Claudia

AU - Mahtani, Kamal

AU - Onakpoya, I

AU - Roberts, Nia

AU - Nunan, David

PY - 2017/6

Y1 - 2017/6

N2 - PURPOSE: Systematic reviews and meta-analyses are important tools used to support evidence-based decisions in health care. A recent study concluded that non-Cochrane reviews report larger effect sizes with less precision when compared with Cochrane reviews. The analysis mainly considered pharmacological interventions. Therefore we explored whether there were similar differences between the two types of reviews that examined the role of physical activity in chronic diseases. Such an analysis is particularly relevant to policy makers in ensuring that recommendations made to populations are based on the most accurate and precise evidence. METHODS: This is a substudy of an overview of 56 Cochrane reviews with RCTs of physical activity interventions for chronic diseases (cancer; cardiovascular, respiratory, renal diseases; metabolic, mental, musculoskeletal disorders). A matched-pair analysis was chosen to contrast Cochrane and non-Cochrane reviews. A systematic literature search was undertaken to identify non-Cochrane reviews that matched based on intervention, condition, outcomes and publication year as in the corresponding Cochrane reviews. Methodological quality of the reviews was assessed using AMSTAR. The pairs will be contrasted in terms of frequency and degree of concordance in their results and whether differences affected the citation rate. RESULTS: Of the 56 Cochrane reviews within our data set, we found 33 non-Cochrane reviews that met the eligibility criteria. The full analysis will be completed by the end of 2016 and final results will be presented fully for the first time at the ACSM meeting. CONCLUSION: This is the first study comparing systematic reviews that were conducted within and outside the Cochrane Collaboration, in particular with respect to physical activity interventions for chronic diseases. Unlike a previous study using meta-analyses on medication therapy within the cardiovascular literature, we applied a more precise methodological approach that concurrently points out how differences can be explained. The findings of our study should make an important contribution to the field of evidence-based physical activity and emphasize that a critical appraisal of systematic reviews reporting physical activity interventions is highly recommended for health professionals.

AB - PURPOSE: Systematic reviews and meta-analyses are important tools used to support evidence-based decisions in health care. A recent study concluded that non-Cochrane reviews report larger effect sizes with less precision when compared with Cochrane reviews. The analysis mainly considered pharmacological interventions. Therefore we explored whether there were similar differences between the two types of reviews that examined the role of physical activity in chronic diseases. Such an analysis is particularly relevant to policy makers in ensuring that recommendations made to populations are based on the most accurate and precise evidence. METHODS: This is a substudy of an overview of 56 Cochrane reviews with RCTs of physical activity interventions for chronic diseases (cancer; cardiovascular, respiratory, renal diseases; metabolic, mental, musculoskeletal disorders). A matched-pair analysis was chosen to contrast Cochrane and non-Cochrane reviews. A systematic literature search was undertaken to identify non-Cochrane reviews that matched based on intervention, condition, outcomes and publication year as in the corresponding Cochrane reviews. Methodological quality of the reviews was assessed using AMSTAR. The pairs will be contrasted in terms of frequency and degree of concordance in their results and whether differences affected the citation rate. RESULTS: Of the 56 Cochrane reviews within our data set, we found 33 non-Cochrane reviews that met the eligibility criteria. The full analysis will be completed by the end of 2016 and final results will be presented fully for the first time at the ACSM meeting. CONCLUSION: This is the first study comparing systematic reviews that were conducted within and outside the Cochrane Collaboration, in particular with respect to physical activity interventions for chronic diseases. Unlike a previous study using meta-analyses on medication therapy within the cardiovascular literature, we applied a more precise methodological approach that concurrently points out how differences can be explained. The findings of our study should make an important contribution to the field of evidence-based physical activity and emphasize that a critical appraisal of systematic reviews reporting physical activity interventions is highly recommended for health professionals.

M3 - Conference abstract in journal

VL - 49

SP - 591

JO - MED SCI SPORT EXER

JF - MED SCI SPORT EXER

SN - 0195-9131

IS - 5S

ER -