Determinants of Nosocomial Infection

The data entry was performed using Epi-Data version 4.6 and then exported into STATA 16 for analysis. Descriptive statistics like mean, median, frequency, and percentage were used to present variables using texts, tables, and graphs. A p-value of ≤ 0.05 was used as statistically significant. To identify the determinants of nosocomial infection, Cox proportional hazard model was used. To identify statistically significant variables, Cox regression (Bivariable and multivariable) was performed using a p value ≤ 0.2 in the univariable Cox regression analysis to identify candidate variables for multivariable Cox regression. To declare statistically significant variables, an adjusted hazard ratio with 95% CI was used, based on p value < 0.05 in the multivariable Cox regression analysis. The goodness of fit of the model was tested by using the Cox-Snell residuals together with Nelson Aalen's cumulative hazard function.