USE OF AMSTAR 2 IN THE META-ANALYSES OF AN UMBRELLA REVIEW ON NON-PHARMACOLOGICAL MANAGEMENT OF KNEE OSTEOARTHRITIS: A SECTIONAL STUDY

Abstract

Knee osteoarthritis is a disease that considerably affects quality of life. The treatment with the use of drugs or invasive measures causes harm to patients, highlighting the importance of non-pharmacological and non-surgical treatment. In this scenario, the search for better management drives the growing number of systematic reviews on the subject. However, a low quality of these reviews is also observed. Therefore, this study sought to identify possible predictors of quality and to describe the performance of the meta-analyses used in the umbrella review "Effect of physical exercises on pain and functional capacity in patients with knee osteoarthritis: an umbrella review" using the AMSTAR 2 instrument. This cross-sectional study was conducted using the AMSTAR 2 instrument and MATLAB software.  The studies were classified into high, moderate, low or critically low quality according to the fulfillment of the critical and non-critical domains of the instrument and in sequence a percentage score of the contemplation of its items was constructed. The results were disappointing: of the 23 reviews analyzed, 22 presented as critically low and one as low. Among the predictive factors, Cocrhane reviews, open access and external funding were relevant. Thus, it is clear that authors need to master with greater skill the key points for the elaboration of a result with a lower risk of bias and the critical domains of the instrument. That said, the conclusions found in the umbrella review do not present significant relevance, lacking further studies in the area.