Assessment of methodological quality of systematic reviews published in the Cochrane Library about physical activity for chronic noncommunicable diseases

Article type
Authors
Carvalho APV1, Grande AJ1, Silva V1, Martimbianco ALC1, Puga MEDS1, Melnik T1, Atallah AN1
1Brazilian Cochrane Centre
Abstract
Background: Chronic Noncommunicable Diseases (NCDs) is the leading cause of mortality worldwide (36 millions of deaths or 63% of all deaths). The number of systematic reviews (SRs) about physical activity for treating and preventing NCDs has increased, and there is a need to assess the methodological quality presented in SRs.

Objective: To analyze the quality of all SRs published in the Cochrane Library about physical activity for treating or preventing NCDs, and to assess the methods used for risk of bias analysis.

Methods: The search strategy was developed for the Cochrane Library to identify all systematic reviews related to the theme, all references selected were screened, and the full text were assessed. The AMSTAR tool was used to assess the methodological quality of SRs. Agreement proportion between authors were independently assessed by two authors (AJG and APVC). SRs were evaluated by only one author (AJG).

Results: Were found 91 references in the first screening (23 protocol; 18 registered title; two SRs were withdraw); 48 SRs fulfilled the inclusion criteria.

Conclusions: The methodological quality of systematic reviews about physical activity for NCDs published in the Cochrane Library demonstrated high quality. Most of characteristics represented in the questions considered in the tool were contemplated in the review analyzed. Although all SRs have evaluated the risk of bias in the included studies, the tool used differed among the reviews, and it could indicate lack of standardization.
Images