Talc, ovarian cancer, and recall bias in the sister study

Abstract

O'Brien et al. [6] reported on the patterns and reliability of self-reported talc use in the Sister Study, a US-based prospective cohort study of women aged 35–74 who had a sister with a history of breast cancer. They found that among certain groups of women, reported use of talc was different at baseline and follow-up. O'Brien et al. [7] evaluated the association between talc and ovarian cancer in this cohort and conducted a quantitative bias analysis (QBA), reporting evidence for differential recall of talc use at baseline and follow-up, which likely increased the magnitude of risk estimates based on recall at follow-up. Additional analyses (e.g., using distributions of recall bias rather than fixed point estimates) may allow for a more complete characterization of the potential impact of recall bias, including a better characterization of the uncertainty around the bias-corrected effect estimates. Future analyses that evaluate recall and other biases more comprehensively, particularly with respect to more fully addressing uncertainty, will contribute to a better understanding of the magnitude of the impact of differential recall on estimated risks.