Ask your doctor whether this product is right for you: a Bayesian joint model for patient drug requests and physician prescriptions. (18th April 2018)

• Record Type: Journal Article
• Authors: Pareek, Bhuvanesh; Liu, Qiang; Ghosh, Pulak
• Is Part Of: Journal of the Royal Statistical Society. Volume 182:Number 1(2019)
• Page Start: 197
• Page End: 223
• Publication Date: 2018-04-18
• Subjects: Advertising -- Bayesian joint model -- Direct‐to‐consumer advertising -- Doctor–patient interaction -- Healthcare -- Zero inflation; Social sciences -- Statistical methods -- Periodicals; Statistics -- Periodicals; 300.15195
• DOI: 10.1111/rssa.12365
• Languages: English
• ISSNs: 0964-1998

Abstract

Summary: The goal of this research is to study jointly physician prescription decisions and patient drug request behaviours. We have adopted a binary logit model and a multinomial logit model to study patient drug request data with excessive zero requests and a multinomial logit model to capture physician prescription decisions. These models are further joined by a flexible non‐parametric multivariate distribution for their random effects. We also adopt an analytically consistent expression for interaction effects in our non‐linear and joint modelling framework. We apply our model to a unique physician panel data set from the erectile dysfunction category. Our key empirical findings include that the triggering of drug requests by direct‐to‐consumer advertising (DTCA) is complex with category level DTCA reducing patients' probabilities of making drug requests and drug‐specific DTCA driving drug requests for the drug advertised, patients' characteristics may play a role in both the influence of DTCA on drug requests and the influence of patients' requests on physicians' prescription decisions, patients' drug requests have a positive effect on physicians' prescription decisions and patients can be consistent with physicians in choosing a drug based on their diagnosis levels and some unobserved factors, and there are significant correlations between physician level random effects that drive both patients' drug requests and physicians' prescription decisions, which validate the joint modelling approach.