RESUMO
AIMS: To examine if the introduction of Diabetes Inpatient Specialist Nurses impacted on length of stay and rates of readmission. DESIGN: Knowledge discovery through data mining as part of a larger realist evaluation of the role. METHODS: Data from January 2017 to January 2019 was extracted and examined. A subset of performance data from July 2017-November 2018 was analysed. This consisted of 7320 records for Hospital Episode Statistics and 272 incident reports (Datix). The data were analysed via Generalised Linear Model regression routines in R. Analysis of readmission rates utilized binary logistic regression, while for the Length of Stay a count regression method was employed. RESULTS: Four trusts were found to have complete and rich data sets. All Trusts that returned complete data were found to have varying decreased length of stay and reduced readmission rates. In two trusts there were significant decreases in patient readmissions and length of stay after the introduction of the Diabetes Inpatient Specialist Nurses. A marked decrease (approximately half) in patient length of stay was found in one London trust after the introduction of the post. Issues with data quality were noted. CONCLUSION: Reduced patient length of stay and rate of readmission were found since introduction of Diabetes Specialist Nurses. Patient safety data was incomplete and varied significantly between trusts. IMPACT: The project sought to understand the impact of employing Diabetes Inpatient Specialist Nurses in hospitals in London. Overall, the specialist nurses helped reduce length of stay and the rate of readmissions. The research will have an impact on the workforce in diabetes and also people with diabetes who need hospital care.