144 / 2019-11-14 17:01:05

The methodological quality of systematic reviews and meta-analysis on treatments for asthma: a cross-sectional study

AMSTAR-2; Methodological quality; Asthma; Systematic reviews and meta-analysis; Cross-sectional study

Background
It is well recognized that systematic review (SRs) will provide the best evidence regarding treatment effects. However, flows in SRs may affect the trustworthy of the results and yield biased conclusions. This study is conducted to assess the methodological quality of an up-to-date sample of SRs on asthma treatments.
Methods
A cross-sectional study was conducted to identify asthma treatment SRs published between 2013 and 2019 by systematic searching of the Cochrane Database for Systematic Review, EMBASE, MEDLINE, and PsycINFO databases. SRs included at least one meta-analysis on asthma treatments were considered eligible. A pre-developed questionnaire was used to collect the bibliographical data of included SRs and the methodological quality was assessed with the AMSTAR 2 (A MeaSurement Tool to Assess systematic Reviews). Associations between bibliographical characteristics and methodological quality were investigated with multivariate regression analysis.
Results
A total of 136 SRs were included, with the majority of them being non-Cochrane reviews (71.3%) and focusing on pharmacological treatments (76.5%). Regarding the overall methodological quality, only 12 (8.8%) were of high quality, 9 (6.6%) were of moderate quality, 32 (23.5%) were of low quality, and over half of them (83, 61.0%) were of critically low quality. To be specified, 133 SRs (97.8%) did not provide an explanation for the selection of the study design in the eligibility criteria, 99 SRs (72.8%) did not report sources of funding for the included studies, and 82 SRs (60.3%) did not provide a list of excluded studies with justifications. Multi-ordinal regression analysis revealed that Cochrane reviews [adjusted odds ratio (AOR): 64.7, 95% confidence interval (CI) 16.28-257.09] and SRs published more recently (AOR: 1.96, 95%CI: 1.44-2.66) positively associated with the overall methodological quality of SRs.
Conclusion
The overall methodological quality of SRs on asthma treatments are suboptimal. Cochrane reviews and SRs that published more recently had better methodological quality. SRs authors, journal editors, and reviewers are suggested to make their efforts in improving the quality of future SRs in this field. Special attention should be paid to conducting a comprehensive literature search, providing a list of excluded studies with justifications, and reporting sources of funding for the included studies.