OBJECTIVES: The aim of this study is to evaluate the use of preprints in the elaboration of Health Technology Assessment (HTA) reports on treatments for severe COVID-19, and how many of these preprints were finally published in a scientific journal after peer review and its impact on the results of the HTA reports

METHODS: A search was performed in Pubmed and in databases of the INAHTA, CADTH, NICE, EUnetHTA among others. The inclusion criteria for the HTA reports were that they should analyze the efficacy and safety of treatments for severe COVID-19 and be published during the year 2020.

RESULTS: Forty-two HTA reports that met the criteria were included, of which 16 included preprints within their body of evidence and 26 did not include preprints (p<0,001). 14% of the individual studies included in the HTA reports were preprints (52/372). The most common type of design of preprint was the cohort study (23 of 52 preprints) and the mean time to final publication in scientific journal was 124 days. Twenty-seven percent of the preprints (14/52) had not been published in a scientific journal one year after publication of the preprint version compared to 62% that were published (p=0,02). In only one case out of 52 of the preprints analyzed was there a change in the conclusions with respect to the version published in a scientific journal.

CONCLUSIONS: The preprints were an important part of the body of evidence used to evaluate the efficacy and safety of treatments for severe COVID-19. They provided earlier evidences that were later confirmed. It is to be explored if the preprints and other early publication approaches could be useful for HTA not only in emergency times but also in a routine way.