Modelling Primary Health Care Use: A Panel Zero Inflated Interval Regression Approach

Sarah Brown, Mark N. Harris, Jennifer Roberts, Karl Taylor

Abstract

We introduce the (panel) zero-inflated interval regression (ZIIR) model, to investigate GP visits using individual-level data from the British Household Panel Survey. The ZIIR is particularly suitable for this application as it jointly estimates the probability of visiting the GP and then, conditional on visiting, the frequency of visits (defined by given numerical intervals in the data). The results show that different socioeconomic factors influence the probability of visiting the GP and the frequency of visits.