The validity of routine data on quality assurance. A qualitative systematic review

The assessment of the quality of medical practice is a legitimate requirement by society. Reliable methods for measurement of the quality of performance are sought worldwide. Quality is often quantified by using administrative data and in Germany this method has been implemented by the health insurance company AOK. (1) How is the AOK quality system rated by senior consultant surgeons? (2) How valid are quality statements derived from administrative data? This article was compiled following the PRISMA (i.e. preferred reporting items for systematic reviews and meta-analyses) statement for qualitative systematic reviews. In order to answer the first question the Professional Association of German Surgeons (Berufsverband der Deutschen Chirurgen) initiated two surveys and to answer the second question a structured literature search following the PICO (i.e. patient problem or population, intervention, comparison control or comparator and outcomes) format was initiated. In addition numerous websites were contacted. Of the responding senior consultant surgeons 95 % considered that the AOK method of quality measurement by administrative data is not objective. One third was definitely wrongly classified. The literature search revealed that no validation data exist for the AOK indicators, including the Elixhauser comorbidity risk score. Altogether, the sensitivity of indicators is poor when good sensitivity is defined by the Institute for Applied Quality Improvement and Research in Health Care (AQUA Institute) as a parts per thousand yenaEuro parts per thousand 80 < 90 %. Quality statements resulting from administrative data alone are unreliable.