Abstract
Higher levels of educational preparation for nurses are associated with lower mortality rates in both medical and surgical wards. In mental health inpatient wards, few studies have examined whether specialist mental health nurse training has any impact on patient outcomes. The aim of this retrospective observational study was to establish the feasibility of extracting and linking nurse education and inpatient outcome data from hospital administrative sources to inform the design of future mental health nursing skill mix studies. Study participants were people experiencing mental ill-health and admitted to psychiatric inpatient care for at least 24 h. The exposure was the ratio of mental health nurses to comprehensive nurses for each patient for each day of their admission. The outcome was readmission for psychiatric inpatient care within 12 months of discharge from the index admission. Confounders were patient demographic (age, gender) and clinical characteristics (diagnosis, legal status, community follow-up). Forty-four patients included in the study were inpatients for a total of 595 days. The median hospital stay was 12 days (IQR = 7-17). In total, 11 (25%) patients were readmitted. In the readmitted and not readmitted groups, the median skill mix ratio was 5 (IQR = 5-7) and 5 (1-6), respectively. It was feasible to extract and code patient and nurse data from hospital databases and link them together. However, a substantial amount of manual post hoc recoding was required to enable us to calculate the exposure (mental health to comprehensive nurse ratio) in a precise way. It may be realistic to automate our methodology in an appropriately powered mental health nursing skill mix study. Australian and New Zealand clinical trial registry: ACTRN12619001337167p.