A partially observable Markov chain framework to estimate overdiagnosis risk in breast cancer screening: Incorporating uncertainty in patients adherence behaviors

Overdiagnosis is the diagnosis of a medical condition that would never have caused any symptoms or problems in a patient’s lifetime if left undetected. Although overdiagnosis is known to be the major risk inherent in mammography screening; currently there is no way to distinguish between overdiagnosed cancers and the ones that would cause problems over a patient’s lifetime. Therefore, the extent of overdiagnosis risk must be estimated indirectly. This and the limitations in previous studies on quantification of overdiagnosis cause a wide variation in the estimation of this risk. In this study, we develop a stochastic framework, which eliminates some of the limitations of the previous studies, to estimate various measures of overdiagnosis. We introduce different measures of overdiagnosis risk including age and stage-specific overdiagnosis risks, as well as the lifetime overdiagnosis risk. Moreover, overdiagnosis risk significantly depends on a patient’s compliance with screening recommendations. Specifically, we develop two partially observable Markov chains to quantify the risks associated with various screening policies while considering the uncertainty in a patient’s adherence behavior. Our results show that overdiagnosis risk is a function of a patient’s age at diagnosis, as well as the number, frequency, and distribution of screening tests over a patient’s lifetime. Further, the results suggest that detecting breast cancer in early stages poses a higher risk of overdiagnosis than detection in advanced stages. A harm-benefit analysis is also performed to compare the overdiagnosis risk with the benefits that breast cancer screening provides. Our results show that, although overdiagnosis rate is relatively high in breast cancer screening, the benefits of breast cancer mammography screening outweigh the overdiagnosis risk.