Abstract
INTRODUCTION: Routine data from statutory health insurance (SHI) are to be used for the surveillance of non-communicable diseases (NCDs) at the Robert Koch Institute (RKI), including for the estimation of administrative incidences of mental disorders. The aim of the study is to identify a methodology that is as sensitive and practicable as possible. METHODS: Using data available from 2005 onwards from the BARMER health insurance fund, with which approximately 10% of the German population was insured, cumulative incidences were determined for the years 2006 to 2022 after gradual exclusion of persons with previous diagnoses in 1 to 17 previous years for depression, anxiety disorders, schizophrenia spectrum disorders, and mental disorders with uniform population-based standardization by gender, age, and region of residence. RESULTS: With an extension of the prior observation period, a similar relative decline in the incidence estimates can be observed for all four diagnostic groups in different observation years. From a prior observation period of two years, the values of the incidence estimators differ with further extension of the prior observation period, but relative changes and thus trends are depicted almost identically. A large proportion of the persons identified as incident already had diagnoses of other mental disorders in previous years. DISCUSSION: For the estimation of administrative incidence rates, the exclusion of persons with a previous diagnosis in two previous years appears to be sufficient to depict trends. Comparisons and interpretations of incidence estimates should only ever be made with similar prior observation and methodology. Further methodological aspects and results for the four diagnosis groups are discussed.