The Geriatric Minimum Data Set (Gemidas) as a quality assurance instrumentin clinical geriatrics

Background: Geriatric medicine in Germany is faced with an increasing deman d for continuous documentation and evaluation of its effectiveness and effi ciency. Hence, the Federal Association (FA) of Clinical Geriatric Departmen ts (Bundesarbeits-gemeinschaft der Klinisch-Geriatrischen Einrichtungen e.V .) has funded a working group on improving quality management in geriatrics by developing criteria for quality standards. Methods: In 1996, the FA working group achieved consensus on the definition of the Geriatric Minimum Data Set (Gemidas) which covered (i) core informa tion about a patient's age, sex, living arrangement, and (ii) basic charact eristics of the hospital course such as location prior to admission and pas t discharge,leading and accompanying diagnoses, newly prescribed technical aids, objective functional status on admission and at discharge (e.g., Bart hel Index (BI), Timed Up & Go (TUG), and intensity of professional care (PP R)),as well as subjectively evaluated attainment of treatment goals. This i nitial report describes the instrument and presents analyses of its feasibi lity for routine clinical practice and data consistency. Results: Twenty out of 27 hospitals (74 %) integrated Gemidas successfully in daily routine, 75 % of which (15 hospitals, total n = 10,567 patients) i nstantaneously collected data on constant numbers of patients per month. Mu ltivariate regression analyses used to decompose variances of the instrumen t's central indicators (e.g., BI, TUG, PPR) revealed a satisfactory dimensi onality and high consistency (e.g., covering 59 % of variance in BZ with 53 % of variance uniquely attributable to patient characteristics), as well a s sensitivity to differences between hospitals (e.g., 12 % of variance in d uration of stay uniquely attributable to hospital differences after control ling for patients' characteristics). Conclusion: Gemidas appears to be a feasible quality assurance instrument i n geriatrics, suitable for compiling its data into a central registry datab ase, which may then be used for analyses across and between hospitals. Howe ver, some modifications are still necessary and more detailed analyses need ed, before final recommendations can be made.