Behavioral effects of screening offers
Principal Investigator
Name
Ethan Lieber
Degrees
Ph.D.
Institution
University of Notre Dame du Lac
Position Title
Associate Professor of Economics and Research Associate (health care program, NBER)
Email
elieber@nd.edu
About this CDAS Project
Study
PLCO
(Learn more about this study)
Project ID
PLCO-2044
Initial CDAS Request Approval
May 29, 2026
Title
Behavioral effects of screening offers
Summary
This project is motivated by nationwide screening guidelines issued by the U.S. Preventive Services Task Force (USPSTF) and legislative changes brought about by the Affordable Care Act (ACA). The ACA made it mandatory for services assigned a letter grade of A or B to be offered on a no-cost-share basis to insured individuals. This study seeks to determine whether screening offers have health consequences beyond the early detection of disease. The findings may help policymakers understand the multidimensional effects of screening policies and identify where supplemental measures, such as counselling, might be beneficial for those undergoing the screening process.
Screening can be viewed as an investment in human capital. Individuals demonstrate a differential willingness to invest in these behaviors even after accounting for price, time, and search costs. Underlying differences in individual-specific health traits, genetics, and environments contribute both directly and indirectly to investments in health behaviors such as exercise, nutrition, sleep, and checkups.
This study addresses three primary gaps in the current academic literature:
First, it remains unclear if the act of screening serves as a catalyst for broader behavioral change. Individuals may improve their diet or exercise habits simply because they were invited to screen—a phenomenon we refer to as the "process effect," which encompasses both anticipatory effects and healthcare provider engagement.
Second, high-risk individuals, such as those with a family history of cancer, face a fundamentally different cost-benefit analysis. However, there is limited data on whether their behavioral response to a screening offer differs from that of the general population.
Third, for those diagnosed, a screening result acts as a major "information shock". Understanding how individuals respond to this shock—whether they "double down" on health investments or disinvest—is crucial for evaluating the total welfare impact of screening programs.
The Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial provides a unique setting to test these questions. Existing literature, such as Einav et al. (2020), uses age-40 mammography recommendations as a natural experiment to show that "compliers"—those who screen only because of the recommendation— are more likely to engage in other preventive behaviors, such as exercise and flu shots. Oster (2020) finds that health behavior selection responds endogenously to recommendations. Kowalski (2023) further advances this literature by developing a framework to separate selection heterogeneity from treatment effect heterogeneity within the Canadian National Breast Screening Study.
While a different strand of literature examines how individuals change their behavior in response to health information, most work focuses on contexts other than cancer screening. The literature on HIV testing provides the closest parallel; for instance, Thornton (2008) finds that HIV-positive individuals who learn their test results are more likely to purchase condoms than those who do not. Angrist and Hull (2023) utilize PLCO data to derive a lower-bound LATE for screening on colorectal cancer incidence; however, they do not use the dataset to assess behavioral effects from these offers.
By analyzing the effects of screening offers, we will contribute to the scarce but crucial policy research on cancer screening behaviors.
Aims
1. Estimate the Intention-to-Treat (ITT) effects of screening offers on long-term health outcomes, specifically focusing on modifications in diet, weight, and exercise.
2. Estimate the effects of screening participation on health outcomes in later life for individuals who took up the treatment, utilizing a Local Average Treatment Effect (LATE) framework.
3. Conduct a heterogeneous treatment effect analysis by stratifying the sample based on baseline characteristics, such as family history and existing health issues. We intend to verify that these sub-samples remain balanced between the treatment (screening offer) and control arms before examining the differential effects of screening offers on subsequent health behaviors.
4. Investigate the impact of "information shocks" by analyzing how learning a positive or negative cancer status influences health-promoting or health-risk behaviors.
5. Disentangle the mechanisms of behavioral change, specifically isolating the anticipatory, healthcare provider engagement, and information effects of screening offers, active screening, and subsequent diagnosis.
References:
Angrist, Joshua D., and Peter Hull. 2023. “Instrumental Variables Methods Reconcile Intention-to-Screen Effects across Pragmatic Cancer Screening Trials.” Proceedings of the National Academy of Sciences 120, no. 51 (December 19, 2023): e2311556120. issn: 0027-8424, 1091-6490, accessed December 5, 2025.
Einav, Liran, Amy Finkelstein, Tamar Oostrom, Abigail Ostriker, and Heidi Williams. 2020. “Screening and Selection: The Case of Mammograms.” American Economic Review 110, no. 12 (December 1, 2020): 3836–3870. issn: 0002-8282, accessed October 17, 2025.
Kowalski, Amanda E. 2023. “Behaviour within a Clinical Trial and Implications for Mammography Guidelines.” The Review of Economic Studies 90, no. 1 (January 12, 2023): 432–462. issn: 0034-6527, 1467-937X, accessed December 5, 2025.
Oster, Emily. 2020. “Health Recommendations and Selection in Health Behaviors. American Economic Review: Insights 2, no. 2 (June 1, 2020): 143–160. issn: 2640-205X, 2640-2068, accessed December 5, 2025.
Thornton, Rebecca. 2008. “The Demand for, and Impact of, Learning HIV Status. American Economic Review 98, no. 5 (November 1, 2008): 1829–1863. issn: 0002-8282, accessed December 5, 2025.
Collaborators
Ethan Lieber University of Notre Dame du Lac
Bhavya Sinha University of Notre Dame du Lac