Predictors of mortality in a critical care unit in south western Kenya

Abstract

Background: Critical care in developing countries has been neglected in the face of high numbers of communicable and infectious diseases like malaria, tuberculosis, HIV and Critical Care Units continue to be limited to large hospitals in urban areas. There is need to know the type of patients admitted to existing units, common diagnoses and outcomes.

Objective: To determine the demographics, diagnosis and mortality of the patient population admitted to Tenwek Hospital critical care units.

Method: Retrospective observational review of all patients admitted to critical care unit in Tenwek Hospital.

Results: Six hundred and forty four patients admitted over a 7 month period were studied. The patients were young (32.8yrs), male and majority were from surgical service. The leading reason for admission was trauma. The overall mortality was 26.1%. Factors that significantly influenced mortality on univariate analysis were patients age (p <0.001), hospital stay (p <0.001), hospital service (p=0.002) and priority level (p<0.001). On multivariate analysis age and pediatric service were still significantly associated with increased mortality. Increased monitoring was protective OR 0.1 (95% CI 0.1-0.2, p=0.01). Of those who died 40.4% had full resuscitation, 36.8% had no resuscitation and 8.2% had withdrawal of care.

Conclusion: Patients admitted to the critical care units were young, male and mainly from the surgical service with trauma being the commonest diagnosis. Age was an independent predictor of mortality and monitoring was protective.

Keywords: Critical Care, Intensive Care Unit (ICU), High Dependency Unit (HDU)