Data Analysis
Categorical variables are reported as whole numbers with proportions, and continuous variables are reported as the median and interquartile range (IQR), unless indicated otherwise. The Life-time table method was used to compute the cumulative survival rates and mortality density. Cox proportional hazard regression models were used to identify the effectors of mortality and calculate the hazard ratio (HR). Univariable Cox proportional hazard regression models were implemented first, and variables with level of significance = 0.10 were included in multivariable analysis. The results of these models were expressed as the HR and 95% confidence interval (CI). Statistical analyses were performed by using SPSS 19.0 (SPSS Inc., Chicago, IL, USA).
The variable with a level of significance = 0.05, identified using multivariable analysis, served as the explanatory variable for time-to-death after initiation of second-line ART. The Kaplan-Meier model was used to estimate the survival and log-rank test were used to compare survival curves of different groups classified according to the explanatory variable. Figures showing the survival curves were prepared using R software (version 3.6.1). Statistical analyses employed two-sided P <0.05 to indicate a significant difference.