Naihui Zhou
Iowa State University, Columbia University, New York City Health and Hospitals Corporation
38 – Applications in Biological and Health Data
Effects of Beta Blockers on Hospital Admission Rates and Cost
Naihui Zhou
Iowa State University, Columbia University, New York City Health and Hospitals Corporation
Ronald Low
New York City Health and Hospitals Corporation, MetroPlus, New York University
Shunsuke Ito
New York City Health
Van Dunn
MetroPlus
Objective: We study the effect of beta blocker use by New York City MetroPlus health plan patients on hospital admissions and cost. Methods: To model admissions, we used a zero-inflated negative binomial regression model. We modeled total medical costs with a log linear regression model with Gamma distribution. Medical costs were calculated by summing charges for hospital admissions, clinic and emergency department visits, professional charges, and drug prescriptions paid by insurance. Explanatory variables include comorbidities, gender, age, socioeconomic factors and drug usage. Results: Beta1-specific blockers with odds ratio of 2.58 (2.13-3.14) and nonselective beta blockers of 1.69 (1.21-2.36) are positively associated with the occurrence of hospital admission (zero inflation component). We did not find interaction effects with COPD or depression. Nor did we find a relationship between beta blocker usage rate of hospital admission regression component). A cost ratio of 1.5376 (1.4197, 1.6652) for beta1 blocker takers and non-takers is found for the COPD population. Increased cost of beta blocker therapy is also significant in depression patients. Conclusions: We found statistically significant relationship between beta blocker use and lack of hospital admission (zero inflation component), but not with the hospital admission rate among COPD and depression patients. Beta blocker use is associated with higher total cost.