Analyzing the Social Determinants of Delayed Diagnoses

A life sciences manufacturer sought to understand how sociodemographic factors affect the time lag between disease onset and diagnosis for a rare disease, also called the “diagnostic odyssey.” The disease of interest responds best to early treatment but often presents with obscure or misleading symptoms, resulting in late diagnosis and sometimes misdiagnosis and mistreatment, which can not only delay appropriate treatment but worsen the underlying condition. The client sought to identify social determinants of health associated with mis- and late diagnoses and to measure the clinical and cost implications of delayed diagnosis to inform development of targeted interventions to improve the speed and accuracy of diagnoses for the affected population.

Leveraging managed closed claims and Medicare fee-for-service claims data, we defined “early” and “late” diagnosis cohorts and characterized those cohorts in terms of social determinants of health such as age, sex, race/ethnicity, payer status, marital status, residence area, access to a vehicle, distance from a provider, and primary care shortage area. We also assessed the clinical, healthcare utilization, and healthcare cost outcomes for each cohort. These descriptive analyses were leveraged to create a regression analysis model to identify predictors of early versus late diagnosis. We then applied the model to determine whether and how late diagnosis predicted negative clinical outcomes and high healthcare costs and utilization.

In addition to contributing to the evidence base for social determinants of health in rare disease, our work helped the client understand the modifiable and unmodifiable factors affecting time to diagnosis for the condition of interest and provided information on how late diagnosis translates into negative outcomes across different patient subgroups.