High Incidence of Childhood Type 1 Diabetes in the Avalon Peninsula, Newfoundland, Canada

The Memorial University of Newfoundland and Health Care Corporation of St. John’s Ethics Committees approved this study.

From 1987 to 2002, all newly diagnosed children with type 1 diabetes were admitted to the Janeway Hospital and were subsequently followed at the Hospital Clinic. Subjects included in this study were ascertained from three sources. First, subjects were ascertained from the diabetes register for type 1 diabetes, which was kept at the Janeway Hospital by the diabetes nurse educator from 1987 onward. Next, research nurses carried out a search of the Hospital Medical Records Department using the diagnostic index code for insulin-dependent diabetes from 1987 to 2002. Subjects were also ascertained from the office records of pediatricians who cared for patients with diabetes. Using the previous three sources, a master list of all patients with type 1 diabetes was established. The research nurse reviewed all charts obtained from the various sources described. Using a data abstraction form, basic demographic details were reported, including a child’s name and sex, date of birth, date of diagnosis, and address at diagnosis. The geographical address of the child at the time of diagnosis was recorded and confirmed to be located on the Avalon Peninsula.

Finally, a registry for the Provincial Diabetes Camp was obtained from years 1987 to 2002 from the provincial diabetes camp director. All camp participants with addresses on the Avalon Peninsula were identified and matched to those from the master list compiled by the JCHCC research nurse. One hundred percent ascertainment was confirmed using the capture-recapture method.

The diagnosis of type 1 diabetes was confirmed based on current guidelines from the Canadian Diabetes Association classification of diabetes and diagnostic criteria (9). Patients included were those <15 years of age at the time of diagnosis. Patients excluded were those with type 2 diabetes, maturity-onset diabetes of youth, transient hyperglycemia, and diabetes caused by chemotherapy or cystic fibrosis. A small percentage of patients with diabetes diagnosed at <15 years of age have type 2 diabetes. Those patients are normally identified by the pediatric endocrinologist or diabetologist. Patients with suspected and confirmed type 2 diabetes were excluded. Although there are some First Nations patients followed at the JCHCC diabetes clinic from Labrador, there were none included in this study because they were not living on the Avalon Peninsula.

The denominator for the analysis was children <15 years of age with residency in the study area that is defined geographically to correspond with census boundaries. The overall size of the population was obtained from census data (8), which are published by Statistics Canada, a department within the government of Canada. A national census performed every 5 years was undertaken in 1991, 1996, and 2001.

The incidence was calculated as the number of newly diagnosed subjects per 100,000 individuals per year in the age-group 0–14 years and in 5-year age-groups (0–4, 5–9, and 10–14 years), and 95% CIs were calculated. Comparisons were made between boys and girls.

A total of 294 new cases of type 1 diabetes was identified among children ages 0–14 years during the study period. The overall incidence per 100,000 individuals per year over the period 1987–2002 was 35.93 (95% CI 31.82–40.03) for the Avalon Peninsula, Newfoundland, Canada. See Table 1 for the mean incidences for the various age-groups between boys and girls.

There was no significant difference between the incidence of 36.15 for boys and 35.69 for girls during this 16-year period with a P value of 0.752. The corresponding Z score was −0.316. The incidence per year for the age-group 0–14 years from 1998 to 2002 has remained >40/100,000 (Table 1). The linear regression model fits well for the data of the incidence of boys and girls on years with P < 0.01. The value of the correlation coefficient between the total incidence and years is 0.7. The estimated rate at which the incidence is increasing per year in this population over the period 1987–2002 using the linear additive regression model is 1.25. The incidence for the 0- to 4-year age-group was 24.95, 5- to 9-year age-group was 37.01, and 10- to 14-year age-group was 43.62 per 100,000 (1987–2002) (Fig. 1).

This current study represents an analysis on the population from the Avalon Peninsula of Newfoundland, confirming the high and increasing incidence over the 16-year study period. The incidence has an increasing trend in all age-groups for both boys and girls (Table 2).

Worldwide, there is a wide variation in the incidence among various populations. In addition, there is a general increase in the incidence of type 1 diabetes in many European (10) and Middle-Eastern countries (4), which is confirmed in our population.

The current Newfoundland population is composed mainly of descendants from approximately 20,000 English and Irish immigrants who settled there in the mid-1700s (11). The cod fishery spurred the settlement of Newfoundland, which occurred particularly in the late 18th and early 19th centuries. Immigrants came from primarily two main areas, Southwest England and Southeast Ireland. By the mid-1830s, the major migrations had concluded, and the population of Newfoundland was approximately 75,000. After this, natural increase became the mechanism for population growth. Other factors, including geographic isolation, lack of roads, segregation by religion, and limited immigration kept related families together. In 1982, 50% of the population lived in communities <2,500 and 41% in communities of <1,000. The Newfoundland population can be considered to have relatively homogeneous origins as a result of how the population was settled and expanded (12). Founder effects have been identified in several other diseases in this population (13,14), and it is hypothesized that this may also be the case for type 1 diabetes, perhaps accounting for the high incidence in the Avalon Peninsula.

The incidence of type 1 diabetes in children aged <15 years on the Avalon Peninsula has been higher than in most areas of the world over the study duration. The incidence of type 1 diabetes has increased, especially over the last 5-year period. Environmental factors are suspected; however, these remain elusive. Although the etiology of type 1 diabetes is unknown, it is proposed that environmental factors may be at play in genetically susceptible individuals, triggering an immune response that leads to the destruction of the pancreatic β-cell. Environmental factors hypothesized in the etiology of type 1 diabetes have included viruses, toxins, stressful life events, and dietary factors (15–20).

The early infant diet may be important. Breast-feeding for >3 months may be protective against type 1 diabetes (21). Breast-feeding has protective effects that reduce enteric infections early in life. Early introduction of cow’s milk protein into an infants diet may possibly be harmful (18). Newfoundland has low rates of breast-feeding as compared with the rest of Canada; however, the rate of breast-feeding initiation has increased gradually over the same study period from 39 (1992) to 55.5% (2001) (22).

In conclusion, the Avalon Peninsula in Newfoundland appears to have the highest incidence of childhood type 1 diabetes in North America, with recent incidence approaching those of Finland and Sardinia. The high incidence is likely due to a combination of genetic and environmental factors. We are currently collecting incidence data from all diabetes centers across the province in a prospective manner. Continued research to determine the reasons for such a high incidence is warranted, and further study into this unique population may contribute to a better understanding of the etiology and pathogenesis of type 1 diabetes.

This research was supported by grants from the Janeway Children’s Hospital Foundation, General Hospital Foundation, Banting Research Foundation, and the Juvenile Diabetes Research Foundation. A.D.P. is a recipient of Canadian Research Chair in Genetics of Complex Diseases (Tier II).

We thank Mary Pelley and Debbie Pantin for secretarial support and Dr. V. Gadag for statistical expertise.