Outcomes of Continuous Ambulatory Peritoneal Dialysis at Charlotte Maxeke Johannesburg Academic Hospital: impact of demographic and socioeconomic factors

Ralise, Mantsebo Elizabeth
Journal Title
Journal ISSN
Volume Title
Background: Chronic kidney disease and end stage kidney disease are becoming a huge health challenge. The optimal treatment is renal transplantation but due to low rates of transplantation most patients who are enrolled in the chronic renal replacement programme are on dialysis. This study aimed at investigating demographic and social factors that are associated with the outcomes of peritoneal dialysis (PD). The study also investigates how co-morbidity contributed to the outcomes of PD. Methods: The study makes use of retrospective analysis of demographic data (age, marital status, residential area, race) and socioeconomic status, level of education, family support, poor access to health care system as well as co-morbidities and underlying cause of ESKD, obtained from 167 patients who were enrolled on continuous ambulatory peritoneal dialysis (CAPD) over the period of 2008 - 2012 at the Charlotte Maxeke Academic Johannesburg Hospital (CMJAH). The data analysis for the present study was conducted using STATA version 14.0. To describe the demographic characteristics of the patients, frequency tables were computed for all categorical variables. For continuous variables, the Shapiro Wilk test for normality was used to assess the distribution of the data to report the appropriate central tendency measure i.e. mean±SD or median (IQR). To assess the contribution of demographic factors to the overall outcome of CAPD, a Fisher’s exact test of comparison was used to assess the difference between the proportions for each demographic factor and CAPD outcome. The Fisher exact test was used because the proportions for each frequency table included a proportion below five (<5). To assess how the relationship of demographic and co-morbid disease affects the outcome of the treatment, a multivariate logistic regression model was fitted adjusting for co-morbid disease for each of the demographic factors. Findings and interpretation: Of 167 patients enrolled, the majority were black with low levels of education, living in townships and 56% were on subsisting on disability grants. PD failure occurred in 53.3% of patients over the study period and 46.7% were successful on PD. Of the variables tested, age was statistically significant for CAPD outcomes [Fisher exact test (p= 0.004)], indicating a significant difference in the proportion of CAPD outcomes among different age categories). The univariate and multivariate logistic regression analysis did not show significant association with CAPD outcomes. Adherence also significantly impacted on outcomes in both univariate and multivariate analyses, showing that non-compliant patients were less likely to have successful outcomes on PD. In addition, the Fisher exact test showed no significant difference in the distribution of CAPD outcome with marital status while the multivariate analysis showed that single patients were three times more likely to succeed with PD compared to married patients. These could be due to chance, because of the small sample size, and require further investigation. Conclusions: Prospective studies are needed to fully understand the extent that demographic and socioeconomic factors impact on the outcomes of PD. This will assist in formulating comprehensive recommendations and ways to improve PD utilization and outcomes.
A research report submitted in partial fulfilment of the requirements of the Master of Medicine at the University of Witwatersrand. Johannesburg, 16 April 2018.