Type 2 diabetes occurring earlier and earlier

Original article:
Parental history and early-onset type 2 diabetes in African Americans and Latinos in Chicago. Onyemere KU, Lipton RB. J Pediatr 2002; 141: 825–9.

Summary
This report examines the role of parental diabetes in the development of type 2 diabetes among children from two high-risk ethnic groups, African Americans and Latinos, in Chicago. The authors established a population-based dataset of insulin-treated African American and Latino children diagnosed in Chicago between 1985 and 1994 at less than 18 years of age. The database contained information on 1089 probands and the study was based on a subset of 243 participants who later completed a telephone interview. Using well-established criteria, 199 (82%) of the participants were categorized as type 1 and 44 (18%) as early-onset
type 2.
The major findings of the study (Table I) were:
— Mean age at diagnosis in type 2 diabetic patients was 4.5 years older (13.4 vs. 8.9) than in the type 1 group.
— There were more female patients in the early-onset type 2 group compared with the type 1 group (ratio 1/1.9 vs. 1/1.1).
— As expected, type 1 patients were more likely to present with diabetic ketoacidosis (DKA) at onset (75%), although nearly 50% (45.5%) of the type 2 group also presented with DKA.
— Sixty percent of the patients reported a positive family history: 61.4% in the type 2 group (6.8% in both parents, 38.6% in the mother alone and 15.9% in the father alone) and 16.6% in the type 1 group (1.5% in both parents, 9.5% in the mother alone and 5.5% in the father alone). The mean age at onset did not differ between those with and those without a history of parental diabetes.

Table I: Demographic and onset characteristics by phenotype, interviewed children, 1985–2001.


Comment
This population-based study highlights once again the rising prevalence of obesity and type 2 diabetes in children, particularly among certain ethnic groups, and the role of parental history. There are several important limitations to the study that may underestimate the quantitative interpretation of the data. For instance, data were obtained from a registry of insulin-treated patients at diagnosis, which may create a major bias in the selection of patients with type 2 diabetes. In addition, the study analysed only 22% of the registry population, leading to small sample sizes. Finally, a large body of data was obtained from self-reporting, and parents of the probands were still too young for assessment of the family history of a disease that has a prevalence which markedly increases with age.
Notwithstanding these limitations, the study brings significant qualitative messages which confirm and reinforce very important, recently published findings on the prevalence of type 2 diabetes in children and adolescents of different ethnic groups. The prevalence of type 2 diabetes in young people is estimated to vary from 1.3 to 50.9/1000 depending on the study and the ethnic group [1]. In Pima Indians, one of the most studied populations, the prevalence between 1992 and 1996 was 22.3/1000 in the 10- to 14-year-old age group and 50.9/1000 in the 15- to 19-year-old age group. In the American population at large, as assessed by the Third National Health and Nutrition Examination Survey, the prevalence estimate for all types of diabetes was 4.1/1000 in the 12- to 19-year-old age group [2]. In addition, published data indicate that the prevalence of type 2 diabetes in children is rising [3]. In Cincinnati, Ohio, the incidence of type 2 diabetes among 10–19 year olds increased from 0.7/100,000 in 1982 to 7.2/100,000 in 1994. Similarly, between 1988 and 1996, the Indian health service documented a 54% increase in the prevalence of type 2 diabetes among 15- to 19-year-old adolescents; in Tokyo, the prevalence increased from 7.3/100,000 in 1976–1980 to 12.1/100,000 in 1981–1985 and to 13.9/100,000 in 1991–1995 [4]. In some studies, the incidence of type 2 diabetes in children nearly equals that of type 1.
The rising incidence and prevalence of type 2 diabetes mirrors the epidemic of childhood obesity [5]. Type 2 diabetes is associated with overweight or obesity (BMI >27 kg/m2, or greater than the 85th percentile for age and sex) in 80–90% of affected children. In addition to weight gain reflecting decreased exercise and increased caloric intake, one of the most important risk factors is parental history, as stressed by the present study. The authors’ estimate of 61.4% of type 2 diabetic children having at least one diabetic parent is not particularly high, as other series have reported figures of 74–100% in different ethnic groups [1]. Other risk factors are sex and puberty.
What then are the implications of these findings?
— First, to identify and screen children at risk for type 2 diabetes by a fasting blood glucose when risk factors are present [1]:
• overweight;
• family history of type 2 diabetes;
• high-risk race or ethnic group, such as American Indians, African Americans, Hispanic Americans, Asians and South Pacific Islanders, who tend to have insulin resistance from childhood unlike in whites;
• signs of or conditions associated with insulin resistance (acanthosis nigricans, hypertension, dyslipidemia, polycystic ovarian syndrome).
— Screening should lead to lifestyle changes and, in some cases, pharmacological treatment. Most importantly, study results such as these should encourage health care providers, politicians and populations to design strategies to reverse this devastating trend.