Eline van den Broek Altenburg, PhD

In an ideal world, everyone who is eligible for cancer screening would be screened and everyone who needs imaging would have easy access to this diagnostic service. Research has found, however, that some groups have lower utilization of key screening services, including mammography and CT imaging. Health disparities have typically been measured by observed differences between the most advantaged group in a category (income, race, etc.) and disadvantaged groups. Standard analysis of a “representative” group yields average effects and reporting across the entire population. But a subtle yet significant problem is how to model differences in unobserved preferences and knowledge — the things we do not know about the individual patient — when looking at their motivation for not getting recommended health care services. I am developing research methodology to create stepwise value assessment that better incorporates the factors we cannot directly observe, like individual emotions, perceptions, and beliefs. This methodology can subsequently be used to develop incentives and reimbursement models that better incorporate value of health care services to individual patients. We will do this with a pilot project analyzing data regarding preferences for diagnostic imaging services.