Abstract
To ensure optimal health outcomes for various populations and to promote fair healthcare distribution, it is imperative to do research on the spatial accessibility of public healthcare facilities. In order to guarantee that communities have sufficient access to vital services, an understanding of spatial accessibility is helpful for both long-term health planning and emergency response and preparation. To promote health equity, improve healthcare efficiency, and advance the general well-being of communities, it is essential to look into the spatial accessibility of public healthcare facilities. Spatial accessibility plays a significant role in obtaining primary healthcare that is fundamental for ensuring the wellbeing of a populace. Poor and unequal access to healthcare facilities, particularly in developing country’s cities, continues to be a major concern, endangering prompt and effective healthcare system. Healthcare system planning should therefore take into consideration the location of healthcare facilities in relation to the community they are planned to serve. This study aimed to assess the physical accessibility of public healthcare facilities in the City of Tshwane, South Africa. The specific objectives were (1) to determine physical accessibility of public healthcare facilities in relation to communities they are intended to serve, (2) to identify communities with geographic barriers in accessing public healthcare facilities and (3) to determine socioeconomic factors that affect access to healthcare facilities.
For the first two objectives, the study utilized healthcare facilities data sourced from the Department of Health and 2021 population estimates data sourced from GeoTerraImage. These objectives were achieved through the application of kernel density and Two-step Floating Catchment Area (2SFCA) approaches. Kernel density findings showed that in certain parts of the City of Tshwane such as Soshanguve, Attridgeville, Mamelodi, and Bronkhorspruit, there is a decline in the spatial accessibility to public healthcare. The 2SFCA results indicated that majority of Bronkhorstspruit, Vlakfontein and some areas in Wonderboom, Soshanguve, and Refilwe regions fell into the highest spatial accessibility index category or band of 0.011157 to 0.025196.
The third objective was achieved using data from the Gauteng City-Region Observatory (GCRO) quality of life survey (2020/2021). Multivariate logistic regression was applied to analyse relationships between multiple independent variables (socioeconomic factors and other related factors such as age, health status, insurance, gender, dwelling type, health services satisfaction and neighbourhood length of stay) and a dependent variable (access to healthcare facilities). As multivariate logistic regression is regarded as a global regression model,
ii
Geographical Weighted Regression (GWR) which is a local regression model, was also used to determine local variations and spatial patterns in the influence of socioeconomic factors and related variables (age, health status, insurance, gender, dwelling type, health services satisfaction and neighbourhood length of stay) on the accessibility of public healthcare facilities at varying spatial areas. The findings revealed that 66.3% of the respondents reported that they had access to public healthcare facilities within their area. The results of multivariate logistic regression revealed that those who lived in informal houses were significantly (Odds Ratio = 0.55, 95% CI [0.37–0.80], p < 0.01) less likely to report that they had access to public healthcare facilities in their area compared to those who lived in formal houses. The GWR results showed deviance residuals that are generally evenly distributed, lying between -2.67 and 1.83. This shows the goodness of the model in predicting the number of people who have access to healthcare facilities using the socio-economic explanatory variables used in the study. Certain locations in the east and west areas of Soshanguve, Centurion, Bronkhorstspruit, and the eastern areas of Pretoria CBD showed high negative deviance residuals, ranging from -2.67 to -1.28. Conversely, high positive deviance residuals (0.74 - 1.83) were noted in the western, southern, and northern regions of the Pretoria CBD, Atteridgeville, Bronkhorstspruit, and in the northern and southern areas of Soshanguve and Mamelodi. showed that all variables directly relate to access and positive correlation exists between all the variables.
The results indicate a notable spatial imbalance in the availability of public healthcare facilities. Certain areas experience a deficit in service provision due to a combination of insufficient facilities, heightened population needs and substantial geographical distances. To ensure that public healthcare facilities are distributed fairly and are positioned strategically to serve a range of populations, a complete strategy should be devised. Maintaining spatial equality requires monitoring population growth, demographic changes, and shifts in healthcare demand. Prioritizing spatial equity in policy decisions necessitates the integration of spatial considerations into healthcare policies and planning. This includes taking into account the geographic distribution of healthcare facilities and the distinct issues that each region faces.
Keywords: Spatial accessibility, Public healthcare access, Two-step floating catchment area, Socioeconomic factors, Multivariate logistic regression, Geographically weighted regression, Local bivariate relationship, City of Tshwane, South Africa.