Variation in definition and measurement of primary outcomes in Cochrane systematic reviews of pharmacological treatment of stable COPD

Article type
Authors
Maekawa R1, Castillo M1, Huayanay L1
1Universidad Peruana Cayetano Heredia
Abstract
Background: an important element in defining the efficacy of a treatment is the appropriate selection of the primary outcome(s). Previous studies have demonstrated the heterogeneity of selection of primary outcome(s) in Cochrane systematic reviews (SRs) that occur between systematic reviews and between clinical trials of the same systematic review.
In 2008, the Task Force of American Thoracic Society and European Respiratory Society (ATS/ERS) published outcomes for pharmacological trials, that is considered as the basis for the core outcome set (cos) of pharmacological treatment of chronic obstructive pulmonary disease (COPD) according to the COMET (Core Outcome Measures in Effectiveness Trials) initiative. There is no previous study that compares the application of these recommendations in the Cochrane systematic reviews.

Objectives:
1) To evaluate the concordance of definitions and measurements of the most frequent main results among systematic reviews on pharmacological treatment of stable COPD.
2) To evaluate whether the most relevant outcomes follow the recommendations of the ATS/ERS task force.

Methods: we selected Cochrane SRs on stable COPD pharmacological treatment published between 2008 and 2019. Two investigators extracted the most frequent major outcomes from the Cochrane SRs methods sections. For each primary outcome, we extracted the definition and measurement method according to the SR. We evaluated the concordance of these variables among systematic reviews. We checked whether they met the guidelines of the ATS/ERS task force.

Results: we identified 25 Cochrane SRs on pharmacological treatment of stable COPD. The most frequent principal outcomes were: exacerbation of COPD 20/25, quality of life 15/25, pulmonary function tests 8/25, and mortality 8/25. We considered the most relevant outcome to be the exacerbation of COPD. For COPD exacerbation, the SRs specified similar concepts and measurements. But in each SR the trials were grouped with heterogeneous definitions of exacerbation. There were also discrepancies between the recommendations of the ATS/ERS working group for the definition and measurement of COPD exacerbation. More details will be given at the Colloquium.

Conclusions: the primary outcome most frequently measured in the SRs was exacerbation of COPD, the definitions were similar among the SRs, but the concordance of the definitions among the trials of a SR was heterogeneous. The editorial group could suggest that the authors describe the criteria for selecting the outcomes of the trials that were grouped in SR more clearly.

Patient or healthcare consumer involvement: none