[Reproducibility of a German scale for assessing the need for inpatient treatment in surgery]

During the past years, the assessment of the appropriateness of hospital utilization has become increasingly important in the German health care system. Previous evaluations by regional review organizations in several states demonstrated the need for a standardized, reliable, and valid instrument to evaluate the appropriateness of inpatient care. Objective of the study is to test the reliability of a German adaptation of the "Appropriateness Evaluation Protocol" (AEP). Among all 2672 admissions from the department of surgery of a regional medical center during one calendar year, 54 patients were randomly selected to evaluate the inter-rater reliability and 51 patients to test intra-rater reliability. Overall agreement, specific agreement and Kappa statistics were estimated for every hospital admissions and all consecutive hospital days. The German AEP showed an inter-rater agreement of 74% (62-86%) for hospital admissions (Kappa = 0.44) and 84% (79%-88%) for all hospital days (K = 0.55). Intra-rater reliability was 88% (79%-97%) for hospital admissions (K = 0.60) and 88% (85%-92%) for all hospital days (K = 0.70). The observed agreement is independent of length of hospital stay and proportion of appropriate days. A standardized instrument with known metric properties is essential for quality management in hospitals to prepare for an increasingly consolidating health care market in Germany. The German AEP is a reliable instrument, which will allow to identify inefficiencies in the management of surgical inpatients.