Abstract

Although administrative data on health care in Germany are part of legal quality assurance in hospital care, they are not part of quality assessment in long-term care. However, claims data of German statutory health and long-term care insurance provide valuable information on outcome quality in nursing homes. Claims data-based quality measurement in nursing homes has hardly been researched and basic work in secondary data analysis is required. This involves the claims data linkage of both statutory health and long-term care insurance as well as new ways of operationalization for quality indicators and their risk adjustment for fair facility comparisons. Using the example of pressure ulcer (PU) occurrence in nursing homes, this study develops a claims data-based quality indicator and discusses potentials and methodological challenges. The analysis is based on administrative data from eleven statutory health and long-term care insurance funds (AOK, 2015). The dataset covers 31% of German nursing homes. The operationalisation of PU acquired within the facility included ICD-10 diagnoses, and prescriptions on dressings. Relevance and validity of claims data on PU-specific aids were also checked in this context. Our risk adjustment strategy followed the one already established by the claims data-based QSR (Quality assurance of inpatient health-care). The Standardized Morbidity Ratio was based on logistic regression with robust standard errors. In 2015, 7.2% of the nursing home residents had at least one PU incident within the facility. The outcome quality considerably varied between facilities. Overall, claims data-based measurement of PU occurrence as outcome quality indicator is feasible for inpatient long-term care and can contribute to transparency and evaluation of care in nursing homes. Information derived from an assessment of care dependency as well as within the amended legal quality assurance system for long-term care may offer new opportunities for routine data-based quality indicators in nursing homes.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call