Recall Bias in Population-Based Case-Control Studies of Ovarian Cancer and Genital Talc Use

HESI Global’s Environmental Epidemiology Committee is hosting a webinar series in collaboration with the QBA Special Interest Group of ISEE  to illustrate the critical role epidemiology can play in the field of quantitative risk assessment. The purpose of these presentations is to provide insight on how to best realize the full potential of human studies in risk assessment and regulatory decision making, and to feature ongoing efforts in this space.

Please, join us for our next webinar, presented by Dr. Igor Burstyn, Drexel University.

Title: Recall Bias in Population-Based Case-Control Studies of Ovarian Cancer and Genital Talc Use

Abstract: We examine the potential role of differential exposure misclassification that results from recall bias, based on published results of population-based case-control studies on the relationship between genital use (ever/never) of talcum powder and ovarian cancer. The case-control studies yielded crude odds ratio of 1.3 (95%CI 1.2-1.4). Bias parameters that capture recall bias were derived from published expert elicitation conducted by the International Agency for Research on Cancer as part of their monograph program's evaluation of carcinogenicity of talc. We combined these priors with available to data to conduct Bayesian adjustments for recall bias. The posterior mean of the bias-adjusted odds ratio was 1.1 (95% credible interval: 0.7 – 1.8), demonstrating that evidence of the association based on case-control studies is far less certain that implied by the analyses that ignore presence of the recall bias.

Wu D, Burstyn I, Thompson WJ, Qian J, Mundt KA: Recall bias in population-based case-control studies of ovarian cancer and genital talc use: potential impact and quantitative bias analysis. 28 October 2025, PREPRINT (Version 1) available at Research Square (https://doi.org/10.21203/rs.3.rs-7924814/v1)

Speaker Bio: Dr. Burstyn is an educator and researcher in the field of environmental and occupational health, with training in both epidemiology and occupational hygiene (BSc in Microbiology from the University of British Columbia, MSc in Occupational Hygiene from the University of British Columbia, PhD in Environmental & Occupational Health from Utrecht University). He is also successfully impersonating a statistician. Dr. Burstyn is a tenured faculty at Drexel University, PA. His methodological expertise lies in the areas of measurement error, misclassification, and confounding as they apply to observational data in epidemiology. His achievements have been recognized by Canadian Journal of Statistics Award (2022) for contribution to a paper on dose-response modelling and Outstanding Contribution to Epidemiology Award (2019) from the American College of Epidemiology. Dr. Burstyn is a founding member of the Global Epidemiology Institute (https://globalepiinstitute.com/) and Editor-in-Chief of Global Epidemiology (https://www.sciencedirect.com/journal/global-epidemiology). He always wants to declare more conflicts of interest than is strictly required and advised.