Factors associated with survival of in-patients with diabetes at Mulago Hospital, Uganda

Abstract

The main objective of the study was to identify the prognostic factors associated with survival of in-patients with diabetes at Mulago Hospital, Uganda. The study was based on a sample of diabetic patients that registered with the diabetic clinic between January 2000 and July 2009 and had been on treatment for at least six months. The Cox regression model (proportional hazards model) was employed to identify the prognostic factors associated with survival. There were 552 patients: 251(45.5%) were male and 301 (54.5%) females. The mean age at onset of diabetes was 40.3years. There were 89 dead cases (33 female and 56 male) of whom 74 (83.1%) were under 70 years, of which 45(60.8%) were female and majority of the dead cases had nephropathy. Of the 36 cases that had records of weight and height, obesity was present in 66.7% of the patients and 13.9% patients were overweight. Generally female patients were fatter than the male. 73.9% had Type 2 diabetes, 90.6% had no family history of diabetes, 52.2% had stayed with diabetes for less than seven years and 59.2% had diastolic blood pressure more than 80mmHg. At admission, 84.1% were hyperglycaemic. Diabetic Ketoacidosis was the commonest complication observed among the type 2 cases and majority of the patients were managed on lente and/or soluble insulin (73.9%). Factors that were significantly associated with survival at 95% confidence level were sex of patient, duration of diabetes, Diabetic Ketoacidosis, neuropathy, diastolic blood pressure, random blood glucose , OHA, mixtard and lente/soluble. Diabetic patients who had stayed longer with the disease were at an increased risk of death as compared to those who had stayed with it for a shorter period. This agrees with the physiological effects of diabetes since it is a metabolic and multisystem disorder. Males had a significantly increased death risk as compared to females and the treatments administered (Oral hypoglycaemic agents, mixtard and lente/soluble insulin) significantly decreased the death risk. This partly explains why patients with high blood sugar level, diastolic blood pressure, Diabetic Ketoacidosis and neuropathy complications tended to survive longer than those whose levels were low. Given the increased exposure to the common risk factors for diabetes, there is urgent need to mobilize the population to undergo screening tests, integrate prevention and care into primary healthcare policies across all government and hospital departments. Otherwise, a timely robust and inevitable result of the growing public health threat of diabetes epidemic deaths across the nation triumphantly awaits, moreover malaria and HIV are still ruling concomitantly.