Methodological quality for systematic reviews of adverse events with surgical interventions: a cross-sectional survey

The confidence in the results of physiotherapy systematic reviews in the musculoskeletal field is not increasing over time: a meta-epidemiological study using AMSTAR 2 tool

OBJECTIVES

To assess the confidence in the results of systematic reviews on the effectiveness of physiotherapy for musculoskeletal conditions in the past 10 years and to analyze trends and factors associated.

METHODS

This is a metaepidemiological study on systematic reviews (SRs) with meta-analysis of randomized controlled trials (RCTs). MEDLINE, Cochrane Database of Systematic Reviews, CINAHL, and PEDro were searched for SRs of RCT on physiotherapy interventions for musculoskeletal disorders from December 2012 to December 2022. Two researchers independently screened the records based on the inclusion criteria; a random sample of 100 studies was selected, and each journal, author, and study variable was extracted. The methodological quality of SRs was independently assessed with the AMSTAR 2 tool. Any disagreement was solved by consensus.

RESULTS

The confidence in SRs results was critically low in 90% of the studies, and it did not increase over time. Cochrane reviews are predominantly represented in the higher AMSTAR 2 confidence levels, with a statistically significant difference compared to non-Cochrane reviews. The last author's H-index is the only predictor of higher confidence among the variables analyzed (OR 1.04; 95% CI: 1.01, 1.06).

CONCLUSION

The confidence in SRs results is unacceptably low. Given the relevance of musculoskeletal disorders and the impact of evidence synthesis on the clinical decision-making process, there is an urgent need to improve the quality of secondary research by adopting more rigorous methods.

Quality of systematic reviews in African emergency medicine: a cross-sectional methodological study

Introduction

Reliable systematic reviews are essential to inform clinical practice guidelines, policies and further research priorities in Africa. For systematic review findings to be trustworthy, they need to be conducted with methodological rigour and reported transparently. We assessed the methodological quality of systematic reviews published in African emergency medicine journals, comparing them to those published in international emergency medicine journals. Additionally, we describe the types of review literature published in the African journals.

Methods

We performed a cross-sectional methodological study of systematic reviews published in selected African and international emergency medicine journals from 2012 to 2021. Studies were eligible if they were i) a systematic review on an emergency medicine topic, ii) published in one of the top five emergency medicine journals in the African region or internationally and iii) published between January 2012 and December 2021 in English or French. We searched PubMed, Web of Science and Scopus databases and hand-searched selected journals. Two authors screened titles, abstracts and full texts independently and in duplicate. Data extraction was performed by one reviewer, using a standardised form, after completing a calibration exercise. We described the characteristics of systematic reviews and assessed methodological quality using AMSTAR II.

Results

We identified 34 (37%) African and 511 (54%) international systematic reviews from 92 and 948 review articles respectively across 10 journals. We included all 34 African and a random sample of 100 international systematic reviews. Methodological quality was low or critically low for all the African systematic reviews (n=34, 100%) and all but three international systematic reviews (n=97, 97%). The median number of critical domain weaknesses was 4 (IQR 4;5) and 2 (IQR 2;4) for African and international systematic reviews respectively. The most common weaknesses across both African and international systematic reviews were i) not establishing a priori review protocols, ii) unclear selection of study designs iii) not providing a list of excluded studies and iv) unclear reporting on funding sources for included studies.

Conclusion

Emergency medicine systematic reviews published in African and international journals are lacking in methodological quality. Reporting an a priori protocol, developing a comprehensive search strategy, appropriate evidence synthesis and adequate assessment of risk of bias, heterogeneity and evidence certainty may improve the quality of systematic reviews.

Inaccuracy in the Scientific Record and Open Postpublication Critique

There is growing evidence that the published psychological literature is marred by multiple errors and inaccuracies and often fails to reflect the changing nature of the knowledge base. At least four types of error are common-citation error, methodological error, statistical error, and interpretation error. In the face of the apparent inevitability of these inaccuracies, core scientific values such as openness and transparency require that correction mechanisms are readily available. In this article, I reviewed standard mechanisms in psychology journals and found them to have limitations. The effects of more widely enabling open postpublication critique in the same journal in addition to conventional peer review are considered. This mechanism is well established in medicine and the life sciences but rare in psychology and may assist psychological science to correct itself.