Abstract

Objective: To determine whether risk of death differed by demographic or other characteristics in a population-based cohort of patients with childhood-onset diabetes.

Methods: The Chicago Childhood Diabetes Registry is an ethnically diverse cohort of patients with diabetes onset between ages 0-17. Cases accumulated from 1/1/1985, through 12/31/2000 (n=1238) were assessed for vital status using death certificates, family reports, and two large administrative databases (Social Security Death Index and National Death Index). Mortality was compared between subgroups using Poisson and Cox proportional hazards regression.

Results: Thirty subjects died, with mean follow-up time of 7.75 years, yielding a crude case-fatality rate of 2.4%. Six subjects died of diabetic ketoacidosis (DKA) at initial presentation. Onset age was the predominant risk factor, driven by a substantially higher rate among those diagnosed at age 17 (13.36/1000 person-years) compared to other ages, and by moderately higher rates for those diagnosed in early adolescence, ages 10-13 (3.49/1000 person-years), compared to children diagnosed before age 10 (0.89/1000 person-years) or at ages 14-16 (0.81/1000 person-years). Mortality did not differ significantly by other factors examined; data for Whites were insufficient to allow comparisons with that group. In addition to diabetes, frequent causes of death were trauma, infectious disease, and cardiovascular disease.

Conclusions: This study provides short to medium-term follow-up in a diverse cohort of patients. DKA remains a significant cause of death in young people with diabetes. Young people diagnosed at the threshold of adulthood are at increased risk for mortality.