SMU SOE Seminar Series (March 6, 2024, 10am-11.30am): Selecting the Patients Who Benefit the Most: Evidence from Marginal Patients in Health Checks

Please click here if you are unable to view this page.

ABSTRACT There is little evidence of the benefits of health checks. We study the effect of receiving a hyperlipidemia diagnosis from health checks on health and medical utilization using the administrative data from 6 million health checks in Taiwan. Our regression discontinuity design exploits the hyperlipidemia diagnostic threshold. We find that a diagnosis increases the likelihood of having related outpatient visits and drug prescriptions by 85.3 and 15.8 basis point (one in ten thousand) in the following year. The likelihood of having stroke hospitalizations decreases by 9.7 basis points. The causal forest estimator reveals substantial treatment effect heterogeneity: Compared to the average, the effect on stroke hospitalization is 16.0 times stronger for the top 20% quantile and 4.784 times stronger for the oldest 20%. A reference range based solely on cholesterol levels may result in unnecessary diagnoses or missed patients. Incorporating age into the diagnostic criteria could be beneficial. Keywords: Health Checks, Hyperlipidemia, Causal Forest. Click here to view the CV. Click here to view the paper.
		PRESENTER Kuan-Ming Chen National Taiwan University
		RESEARCH FIELDS Labor Economics Applied Econometrics
		DATE: 6 March 2024 (Wednesday)
		TIME: 10am - 11.30am
		VENUE: Meeting Room 5.1, Level 5 School of Economics Singapore Management University 90 Stamford Road Singapore 178903