Cancer Screening Outreach Guided by Machine Learning: The Benefits of Proactive Care

Abstract

Problem definition: With the advance of data analytics, many disease prediction models have been developed with the intent of detecting diseases earlier and improving patient outcomes through earlier treatment. The operationalization of interventions and care based on these predictive models is critical to attaining these goals. We study the realworld effects of a machine learning-guided colorectal cancer screening outreach program deployed at a health system in Pennsylvania. Methodology/results: Using a regression discontinuity design based on the predicted risk score for having cancer, we find that the program increases the likelihood of colonoscopy uptake in three and six months by 6.0 percentage points (214% increase relative to the control sample within the bandwidth) and 6.9 percentage points (117% increase), respectively. Importantly, we also find significant effects on mortality. We estimate that the program decreases two-year mortality by 6.2 percentage points (43% decrease). Managerial implications: Our finding suggests that a proactive cancer screening outreach program where individuals are selected for intervention based on a machine learning algorithm could significantly improve patient outcomes in addition to achieving higher disease detection rates. Our analysis demonstrates an analytical framework for rigorously evaluating machine learning-aided outreach programs for other cancers and diseases. Establishing unbiased estimates of the impact of machine learning-aided screening outreach is critical for capacity planning of screening resources, such as colonoscopies.