Chasing the Missing Patients: The Impact of Screening Subsidies on the Demand for Health Treatment

This paper provides evidence that the impact of reducing the price of screening on health treatment could be reduced by patients’ self-selection into screening. The Affordable Care Act requires health insurance plans to offer free screening for diabetes, high cholesterol, and hypertension to people at high risk, which should encourage additional patients to be screened. However, we hypothesize that patients who were not previously seeking routine screenings may also be less likely to seek medical care for newly detected conditions, thereby reducing the effect of the policy. To examine this issue, we study the behavior of people who receive low-cost screening to determine if screening motivates them to seek medical treatment. We analyze linked Medicare claims and biomarker data from REGARDS, an epidemiological study that paid older adults to be screened for diabetes, high cholesterol, and hypertension. The study recruited participants using nationwide random phone calls over 2003-2007. Exploiting the rolling recruitment into the study, we compare not-yet-screened participants to already-screened participants to show that screening increased semi-annual doctor visits for previously undiagnosed conditions by only 22 percentage points. Participants with a new diagnosis were less likely to take medications for or control their previously diagnosed conditions prior to REGARDS, implying that newly diagnosed conditions may also be less likely to be controlled. Finally, we find no evidence that participants with undiagnosed conditions switch to health plans with lower cost-sharing after screening, possibly indicating they did not anticipate spending much money on the new conditions. These findings are consistent with our hypothesis that patients whose chronic conditions were not detected prior to screening subsidies seek less medical treatment for their diagnosed conditions. In an economic model, we show that patients’ self-selection into screening based on their idiosyncratic costs of medical care could account for this pattern. We conclude that reducing the price of screening is not enough to engage many of patients with unmet health needs in needed medical care. The findings shed light on possible impacts of reducing patients' out-of-pocket cost of screening on population health and health insurance markets, and raise questions about how to engage patients with newly detected conditions.