團隊

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 real-world 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.

Problem definition: Scant empirical research studies the impact of drug shortages on the quality of medical care in hospitals. We study the causal relationship between drug shortages and medication errors using a natural experiment: hurricane damage to factories that produce heparin, an essential medication used frequently in hospitals. Methodology/results: We collect data on medication errors from the U.S. Food and Drug Administration’s Adverse Event Reporting System and drug sales from IQVIA’s National Sales Perspective. Applying the synthetic control method, we find that hurricane-related heparin supply disruptions increased medication error rates by 152%. In addition, we find significant spillover effects. The disruption increased medication error rates of a substitute drug, enoxaparin, by about 114%. Managerial implications: Our study uses an exogenous event to show that medication supply chain disruptions may negatively impact hospitals’ quality of care. We contribute to the literature by empirically linking the effects of supply chain disruptions to downstream service quality. Our results show that commonly used measures to mitigate the impact of drug shortages, such as substituting medications, may be unsafe. We discuss several measures that hospital managers may consider implementing to mitigate the potentially harmful effects of drug shortages.