Mortality of young patients with diabetes in Kinshasa, DR Congo

AIMS: As data on mortality of young patients with diabetes is not available in the Democratic Republic of Congo (DRC), we studied mortality rates, the influence of some determinants and causes of death. METHODS: A retrospective review of standardized medical records of all patients with diabetes aged<or=30 years at diagnosis who were enrolled between 1994 and 2004 in a large, integrated healthcare network in Kinshasa, DRC. Diabetes was diagnosed clinically. As resources for classification according to type of diabetes were not available, the study sample was a mixed-type cohort. Death was established by review of medical records and by community interview. Mortality was recorded until 2007. Overall mortality rates and 95% confidence intervals (CI) were calculated. Kaplan-Meier survival and Cox regression analyses were performed to study the influence of determinants. RESULTS: Of the patients, 17.4% (159/915) died, mostly during the first 5 years of follow-up. Mean yearly mortality was 3.62/100 patient-years (95% CI 3.1-4.2). Independent predictors were male gender [hazard ratio (HR) 0.66 (95% CI 0.5-0.9) for females vs. males] and age at diagnosis (HR 0.97 (95% CI 0.94-0.99) per increase of 1 year]. Major causes of death were ketoacidosis (38%) and infection (12%). Cause of death was unknown in 31% of cases. CONCLUSIONS: One out of six patients died, most within 5 years after diagnosis. Death occurred more frequently at home, in male patients and in subjects aged