(161) Bias in Observational Studies on The Effectiveness of Hydroxychloroquine in COVID-19

Sunday, August 28, 2022

8:00 AM – 1:45 PM CEST

Location: Hall C

Publication Number: 1206

Presenting Author(s)

Mirjam Hempenius
– University of Utrecht, Utrecht, Utrecht, Netherlands

Co-Author(s)

Rolf H.H. Groenwold
– Department of Clinical Epidemiology, Leiden University Medical Center, Netherlands
Anthonius de Boer
– Division of Pharmacoepidemiology and Clinical Pharmacology, Utrecht Institute for Pharmaceutical Sciences, Utrecht University, Netherlands
Olaf H. Klungel, PharmD, PhD, FISPE
– Professor of pharmacoepidemiology, University of Utrecht, Utrecht, Netherlands
Helga Gardarsdottir, PhD, FISPE
– Associate Professor, University of Utrecht, Utrecht, Netherlands

Background: During the first waves of the coronavirus pandemic, evidence was urgently needed about potential effective treatments. Results from observational studies on the effectiveness of hydroxychloroquine (HCQ) were conflicting, potentially due to biases.

Objectives: We aimed to assess the quality of observational studies on HCQ and its relation to effect sizes.

Methods: Observational studies on the effectiveness of in-hospital use of HCQ in COVID-19 patients were reviewed. Study quality was assessed regarding seven items, based on the ROBINS-I tool: confounding, selection bias, misclassification of interventions and of outcomes, deviation from intended intervention, missing data, and reporting. Effects sizes found in observational studies were compared to those from RCTs, and differences were related to apparent study quality.

Results: None of the 33 included observational studies were free of risk of bias, most commonly related to confounding (n=26, 79%) and misclassification of interventions (n=22, 67%). Observational studies with estimates closer to those of RCTs appeared less often at risk of bias than studies with more diverging estimates (p=0.02).

Conclusions: Overall, the quality of observational HCQ studies was poor, and studies reporting more extreme estimates appeared of lower quality than studies reporting estimates closer to those of RCTs. Synthesis of evidence of effectiveness of HCQ in COVID-19 should focus on RCTs and carefully consider the added value and quality of observational evidence.