Zusammenhang zwischen der regionalen Siedlungsstruktur und der Krankenhausinzidenz, Therapieform und Mortalität von nicht-rupturierten abdominalen Aortenaneurysmen. Sekundärdatenanalyse der deutschen DRG-Statistik von 2005–2014

Abstract

INTRODUCTION

In Germany, the regional settlement structure is heterogenous, ranging from densely populated cities with a tight network of vascular health care to large regions in which access to health care is limited in terms of space and time. Therefore, the aim of this secondary data analysis was to investigate the association between the settlement structure of the patient's home district (KT), and the hospital incidence, type of therapy, and mortality of non-ruptured abdominal aortic aneurysms (nrAAA).

METHODS

The microdata of the DRG statistics of the Federal Statistical Office for the years 2005-2014 were evaluated. All patients with nrAAA (ICD-10 Code I71.4) who were admitted to a German hospital and treated by open surgery and endovascular repair were included. Classification of treatment was based on the German Operation and Procedure Code. Patients were grouped according to the settlement structure of their home district defined by the Federal Institute for Research on Building, Urban Affairs and Spatial Development (KT1 independent city, KT2 urban district, KT3 rural district, KT4 sparsely populated region). The age-, sex- and risk-adjusted association between the type of settlement structure and in-hospital mortality was analysed using a multivariable multi-level regression model. The Elixhauser co-morbidity score validated for administrative data was used for risk adjustment.

RESULTS

Of 95,452 cases included, 88 % were men. Mean age was 72 years. There were 28,970 (30 %) patients in KT1, 37,759 (40 %) in KT2, 14,442 (15 %) in KT3 and 14,281 (15 %) in KT4. The hospital incidence was 12.4 per 100,000 inhabitants in KT1, 11.8 in KT2, 10.8 in KT3 and 11.2 in KT4 (p <0.001, falling trend). The proportion of EVAR treatment was 56 % in KT1, 54 % in KT2, 57 % KT3, and 59 % in KT4 (p <0.001, increasing trend). The raw hospital mortality of patients from KT1 to KT4 was 3.4 %, 3.4 %, 3.2 % and 3.6 %, respectively (p=0.553 for trend). The multivariable regression analysis revealed no statistically significant association between the KT and hospital mortality (KT1=reference, RR KT2=0.97 [95% CI 0.79-1.15], RR KT3=0.98 [0.81-1.14], RR KT4=0.98 [0.86-1.11]).

CONCLUSIONS

The study shows that both the hospital incidence and the type of therapy (endovascular vs. open) differed between the settlement structural district types, but there is no urban-rural gap regarding in-hospital mortality of treated nrAAA.