Health Technology Research in Brazil
Anais Online do "Health Technology Research in Brazil: Challenges for the New Decade"

Introduction: This study aims to evaluate the quality of systematic reviews (SR) regarding the effect of COVID-19 on the worsening and/or mortality of cardiovascular patients. To the best of our knowledge, this is the first study to examine this subject. Given the relevance of cardiovascular disease to COVID-19 and the speed with which new related information emerges, understanding the quality of the evidence that is being summarized on this topic is necessary to assess the confidence in its results. Methods: A cross-sectional methodological survey of SRs performed in November 2020. An electronic search was conducted in the following databases: PubMed, Embase and CINAHL, based on the descriptors and terms ‘Systematic review’, ‘COVID-19’, ‘COVID’, ‘Coronavirus’ and ‘Cardiovascular Disease’. SRs published in peer-reviewed journals that met the following conditions were included: 1) Address the worsening outcome and/or mortality of cardiovascular patients with COVID-19, 2) Carry out a systematic search in at least two databases. Two reviewers screened for study eligibility and for data extraction. The following information was extracted: bibliometric data, participants characteristics and results of the AMSTAR 2 checklist. Results: The searches resulted in 2,170 references, of which 2,152 were excluded during the reading of titles and abstracts and 8 in the full-text reading. All the 10 SRs included in the final sample were published in 2020, in English and involved only observational studies. Of the total, four were published in the USA, two in China and Iran and the others in European countries. None was a Cochrane review and only two presented a protocol record. The AMSTAR 2 mean percentage score was 43.7% (± 25). The overall confidence in the results of all SRs was rated as ‘Critically low’. 10% had two critical domains, 40% three and the rest had four or more critical domains. SRs performed poorly with respect to selection of study designs and source of funding. Only two domains were satisfactory in 90% of the SRs. Regarding the explanatory predictors (country of publication, funding and number of participants), no association was found with the quality of SRs. The relationship between impact factor and AMSTAR 2 scores was positive, but weak. Conclusion: Methodological quality of SRs regarding the effects of COVID-19 on cardiovascular patients is very critical. Based on that, results of these SRs must be interpreted with caution and raises concerns about the flood of substandard research amid the rush to publish during the COVID-19 pandemic.