Heterogeneity in Women’s Adherence and Its Role in Optimal Breast Cancer Screening Policies

Abstract

Most major health institutions recommend women to undergo repeat mammography screening for early diagnosis of breast cancer, the leading cause of cancer deaths among women worldwide. Although the proportion of women who ever had a mammogram is increasing, a majority of women do not get repeat mammograms. This paper analyzes the role of heterogeneity in women’s adherence on optimal mammography screening recommendations. We develop a dynamic modeling framework that considers imperfect and heterogeneous adherence to screening recommendations. We carefully calibrate our model and solve it using real data. Unlike the existing breast cancer screening guidelines, our results suggest that adherence and heterogeneity in women’s adherence behaviors should be explicitly considered. In particular, we find that when screening strategies are optimized assuming average adherence for everyone, the effect on patients with already low adherence would be relatively small, but patients with high adherence would be adversely affected. Considering imperfect and heterogeneous adherence in the population, our model suggests (1) given the current low adherence rates, an aggressive screening policy such as annual screening between the ages of 40 and 79 should be promoted to the general population; (2) screening strategies may be adjusted in clinical practice based on women’s adherence, and screening intervals can be extended to two years for women with a history of high adherence; and (3) if the screening patterns change in the long run and most regular screeners adopt the most recent U.S. Preventive Services Task Force guidelines, then improving overall mammography adherence in society becomes more critical. This paper was accepted by Rakesh Sarin, decision analysis.