ADMINISTRATIVE DATA IN HEALTH CARE: EMPIRICAL APPLICATIONS IN THE ITALIAN CONTEXT

Scientific and economic research based on administrative data has grown considerably over the past two decades. The use of administrative data, in particular hospital discharge data, offered researchers the opportunity to develop new scientific evidence of the determinants of health, as well as evaluate government policies aimed at improving people’s health. This thesis is divided into three chapters. Chapter 1 offers a review of the literature on administrative data, discussing advantages and disadvantages, issues of data quality and the data’s research potential. Administrative data from the SDO database are used in the second chapter to examine the determinants affecting patient choice of distant destinations. Discrete choice models reveal that the factors that are most influential to patient hospital choice are distance, number of beds and hospital case mix. In chapter 3, difference-in-difference models are used to evaluate the effect of an Italian regulatory policy that prohibited the night sale of alcohol in motorway service areas. Combining data provided by the Ministry of Health for the years 2007-2013 on admissions for alcohol intoxication and accidents with information about hospital location and socio-economic characteristics of the area, our results suggest the effectiveness of the policy ban in reducing the number of alcohol related hospitalizations when aggregating data at patient municipality of residence.