Prospective study of incidence of juvenile diabetes mellitus over 10 years in Dar es Salaam, Tanzania

BMJ. 1993 Jun 12;306(6892):1570-2. doi: 10.1136/bmj.306.6892.1570.


Objective: To ascertain the annual incidence of diabetes requiring treatment with insulin in children and adolescents aged 0-19 years in Dar es Salaam, Tanzania, during a 10 year period from 1 January 1982 to 31 December 1991.

Design: Prospective registration at a major urban hospital of all patients with newly diagnosed diabetes who were resident in Dar es Salaam.

Setting: Muhimbili Medical Centre, Dar es Salaam, Tanzania.

Patients: 86 patients: 45 male, 41 female.

Results: The annual incidence of juvenile diabetes for both sexes was 1.5 per 100,000 population aged 0-19 years (95% confidence interval 1.3 to 1.7). Incidence per 100,000 population per year increased with age: 0.6 (0.0 to 0.13) in the age group 0-4 years, 0.5 (0.3 to 0.7) at 5-9 years, 2.2 (1.8 to 2.6) at 10-14 years, and 3.4 (2.9 to 3.9) at 15-19 years.

Conclusion: Juvenile diabetes mellitus is fairly rare in sub-Saharan Africa. If environmental factors such as infection and material deprivation were important determinants of insulin dependent diabetes in Africans, as they may be in Europeans, much higher rates would have been expected unless genetic factors possibly exert a protective role. The eightfold greater incidence in African Americans than in Tanzanians may be related to greater genetic admixture in African Americans with people from countries in Europe with a high incidence.

PIP: Between 1982 and 1991 in Tanzania, health workers at the diabetic clinic of Muhimbili Medical Centre in Dar es Salaam registered 86 children and teenagers (0-19 years old) from the indigenous population (Bantu ethnic group) who had insulin-dependent, juvenile diabetes mellitus to determine its incidence in Dar es Salaam. The researchers had taken appropriate steps to assure that they recorded all diagnosed cases of juvenile diabetes in Dar es Salaam. Just 1 child under 5 years old was registered. Most of the children (63) were diagnosed in the first 5 years of the study. The researchers could not explain the decrease in diagnosis in the 2nd half of the 10-year study period. The mean crude annual incidence of diagnosed diabetes over the study period was 1.5/100,000 population. Incidence increased with age (.06 for 0-4 year olds, .5 for 5-9 year olds, 2.2 for 10-14 year olds, and 3.4 for 15-19 year olds). The incidence of juvenile diabetes among African Americans and among black children in the Virgin Islands and Cuba is 8, 4, and 3 times higher, respectively, than the incidence found in Tanzania. These higher incidences suggested a genetic effect because the genetic admixture of the African Americans consisted of 2 ethnic groups, one of which tends to have a low incidence (Bantu) while the other group tends to have a high incidence (northern European), and the genetic admixture of the Cubans consisted of 2 ethnic groups both with a low incidence of juvenile diabetes (Bantu and Hispanic). Genetic factors may have a protective effect because significant determinants of juvenile diabetes found among Europeans (i.e., infection and material deprivation which are present in Tanzania) did not effect an increased risk in the Dar es Salaam population.

MeSH terms

  • Adolescent
  • Adult
  • Age Factors
  • Child
  • Child, Preschool
  • Confidence Intervals
  • Diabetes Mellitus, Type 1 / epidemiology*
  • Female
  • Humans
  • Incidence
  • Infant
  • Infant, Newborn
  • Male
  • Prospective Studies
  • Sex Factors
  • Tanzania / epidemiology