Pre-print studies on COVID-19 take about 6 months to publish

The time between online publication of the preprint and publication in the journal for randomized clinical trials on COVID-19[feminine] subjects can range from 2 weeks to over a year, according to new research.

Data from a team of Canadian researchers showed that preprint research articles on COVID-19 take a median time of more than 6 months to be published in a journal. The long lag in publication—as well as the varied results of publication of preprint articles depending on the size of their test samples and risk of bias—indicate the lack of a uniform strategy or prioritization of review. professional in the latest research on the pandemic virus.

Investigators led by Anthony D. Bai, MD, of the Division of Infectious Diseases, Department of Medicine, Queen’s University, sought to interpret the time to publication in the journal of preprints of randomized clinical trials (RCTs) COVID- 19, as well as differences in research published both as a preprint and possibly as a review article.

Bai and colleagues highlighted the “impulse” to provide RCT data regarding COVID-19 as soon as possible to help clinical care strategies during the pandemic. They cited preprint successes, such as the results of the RECOVERY trial regarding dexamethasone as a beneficial treatment and standard of care for COVID-19.

“However, preprints may not meet the standards required for scientific publication without peer review, they may be subject to change because the manuscript is not yet finalized, and they may contain errors or misinformation,” they wrote. “As a result, preprints may mislead the public.”

Few prior trials have observed differences in pairings of preprinted articles from COVID-19 RCTs and eventual published journal articles; the systematic review and meta-epidemiological analysis of the investigators sought to answer the exact difference between journal articles in publication time, results, analysis, results and conclusions based on the preprint on the publication status of the journal.

“We hypothesized that there would be a significant time lag between the publication of the preprint and the publication of the review and that there would be notable differences between the preprints and the corresponding published review articles,” they said. they wrote.

The team performed a literature search of the World Health Organization (WHO) COVID-19 database and Embase to identify preprint RCT articles published between January 1 and December 31, 2021. The trials eligible included human participants and research questions relating to the treatment or prevention of COVID-19[FEMININE

Pour les prépublications observées, les enquêteurs ont ensuite effectué une recherche documentaire pour localiser l’article de revue correspondant. Les critères de jugement, y compris le risque de biais et le délai de publication, ont été définis par des examinateurs indépendants et le modèle de régression des risques proportionnels de Cox, respectivement. Bai et ses collègues ont décrit de manière indépendante les différences entre les paires d’essais ECR par critères de jugement, analyses, résultats et conclusions.

L’évaluation finale comprenait 152 prépublications, dont 119 (78,3 %) avaient été publiées dans des revues au 1er octobre 2022. Parmi les ECR publiés, 65 (54,6 %) ont été publiés avec des changements dans les critères de jugement, les analyses, les résultats ou les conclusions. Le délai médian de publication était de 186 jours (fourchette, 17 – 407). Seules 2 études (1,7 %) ont été publiées avec des conclusions qui contredisaient la conclusion incluse dans la version préimprimée.

L’équipe a identifié une association entre des tailles d’échantillons d’ECR plus importantes et des risques de biais plus faibles avec les publications dans des revues. Les patients d’ECR ou les tailles d’échantillons de 201 à 1000 étaient 23 % plus susceptibles d’être publiés par rapport à une référence d’ECR avec ≤ 200 échantillons (risque relatif [HR], 1.23; 95% CI, 0.80 – 1.91). Among RCTs with more than 1000 samples, the probability of publication was more than doubled (RR, 2.19; 95% CI, 1.36, 3.53).

Using an RCT with a high risk of bias as a reference, the researchers observed that medium-risk RCTs with few concerns about bias were 77% more likely to be published (HR, 1.77; 95% CI, 1.02-3.09); low-risk RCTs were three times more likely (RR, 3.01; 95% CI, 1.71, 5.30).

Bai and colleagues concluded that the time from preprint to publication in COVID-19 RCTs was “substantial,” as were the differences in outcomes, analyses, sample size, and results from preprint to publication in the majority of RCTs. Among the most significant findings they noted was the need for more relevant context for preprint data, including in RCTs published in journals.

“However, if investigators include preprints, they should exercise caution in synthesizing and interpreting knowledge and should recognize that unpublished preprints are likely to have a higher risk of bias,” they wrote. . “In this study, it was reassuring to find that the primary outcome estimate and conclusion remained generally consistent across the preprints and corresponding review articles, so these updates were unlikely to change the overall conclusion.”

The study, “Comparison of preprint publications of randomized clinical trials on COVID-19 and corresponding published journal articleswas published online in Open JAMA Network.