An Exploratory Investigation into Institutional and Staff Performance: Evidence from Public Hospitals in Gauteng

Abstract

There are various challenges facing the South African health system, especially in Gauteng, the most densely populated province in South Africa with the largest public sector hospital network. This research was undertaken to understand how hospital performance and staff performance is measured in Gauteng public hospitals and to determine if there is a correlation between hospital performance and staff performance. Panel data from 2 administrative datasets for 34 hospitals was analysed over an 8- year period from 2013/14 to 2020/21. Hospitals were classified according to level of care into 5 categories (central, tertiary, regional, district and specialised) and data for 11 hospital performance indicators and 5 staff category indicators were analysed. Trend analysis was conducted for hospital and staff performance, PCA was used to explore the relationship between hospital performance indicators and investigate the construction of a composite index of hospital performance, and regression analysis was used to investigate the relationship between hospital performance and staff performance. Overall, capacity within the Gauteng hospitals has not increased and output only slightly increased over the 8-year period. Trends suggest that the majority of inpatient and outpatient care were provided by central hospitals where the costs of care were the highest. Regional hospitals followed closely on most indicators except emergency headcount and expenditure per PDE. Emergency care was provided more by regional and district hospitals, but it costed more to treat a patient at a district hospital than a regional hospital. Specialised hospitals tended to be outliers on most of the indicators assessed. Staff performance had consistent scores for all 5 staff categories with doctors being scored lower than other staff categories. PCA analysis determined that patient day equivalent, bed utilisation rate, average length of stay and inpatient crude death rate could form a composite indicator of hospital performance. However, it is not straightforward to capture hospital performance in a single indicator as each of these indicators captures a different element of performance. iv The key finding of this research was that there was no correlation between hospital performance and staff performance over the 8-year period based on both individual indicators and principal components. This emphasises the need for staff performance to be linked to hospital performance indicators from an output, process, quality and outcome perspective in order for Gauteng hospitals and staff to provide health care in an effective and efficient way.