Does the density of gastroenterologists and the county type have an influence on the health care of patients with inflammatory bowel diseases? A claims data analysis of a statutory health insurance fund

• Applicant: Dr. Anne Prenzler
• Subject Area: Public Health, Healthcare Research, Social and Occupational Medicine
• Term: from 2012 to 2014
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 231909582

The two main idiopathic inflammatory bowel diseases (IBD) are ulcerative colitis (UC) and Crohn's disease (CD). These diseases are very relevant from a health services research perspective. Features of an optimal health care of IBD-patients were defined in the course of the development of evidence-based and consented IBD-pathways. However, there is a lack of studies, which map the actual health care situation. Especially cross-sectional, longitudinal and cross-regional studies are missing, whereas the consideration of regional structures seems to be important due to the high scientific and political importance of nearby health care. The empirical evidence for the connection between regional structures and the health care of patients is insufficient.Hence, the main research question is: Does the density of gastroenterologists and the county type have an influence on the health care of patients with inflammatory bowel diseases (IBD) and on the application of the consented IBD-pathways? The density of gastroenterologists and the county type, which depends of the population density, will be combined with claims data of IBD-patients, who are insured by a nationwide operating health insurance fund. Besides socio-demographic parameters, the claims data provide information on specialist visits, hospitalizations and medication prescriptions. Hence, the claims data can be used to evaluate the application of aspects of the IBDpathways.The analysis takes place in two steps. First, variables in the claims data like number IBD-related hospitalizations will be analysed descriptively and stratified by subgroups, e.g. sex, county type and disease (CD, UC). Second, potential health care differences will be analysed via Poisson and probit regression models. For this, four hypotheses were defined, which can be tested on the basis of claims data. Potential interactions between physician densities and county type will be considered.The results of this study can scientifically contribute to the discussion, if the health care of patients is related to physician densities and county type. Finally, the analyses can support the application and modification process of the IBD-pathways.

• DFG Programme: Research Grants
• Participating Persons: Professor Dr. Roland Linder; Privatdozent Dr. Jan Zeidler