Synapse
⌘+K
Synapse
PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
October 17, 2025Open Access

Real-world comparison of counterfactual inference methods for evaluating impact of antidepressants on COVID outcomes

View Full Paper
Ask AI
Bookmark
Share

Authors

SKSonja KleperMHMamoon HabibAVAnugrah Vashnav

Discussion

Loading...

Member takes

Overview

Analysis reveals that counterfactual methods may introduce bias in drug effect estimates for COVID-19 outcomes.

Key Points

  • No antidepressant was found to confidently impact COVID-19 outcomes, indicating significant limitations in existing methods.
  • Empirical simulations revealed biases from unmeasured confounding, complicating the interpretation of observational data.
  • Target trial emulation was utilized to compare 18 antidepressants, but biases hindered reliable conclusions.
  • The N3C database provided extensive health records, which included insights from over 22 million individuals and 8.9 million COVID-19 cases.

Cite This Study

Kleper et al. (2025) studied this question.

synapsesocial.com/papers/68f199ccde32064e504dd20fhttps://doi.org/10.1101/2025.10.13.25337938
View Full Paper
Ask AI
Bookmark
Share