Background: Type 2 diabetes is associated with low birthweight followed by obesity in adulthood. Persons who develop the disease may therefore have a particular pattern of growth from birth through childhood.
Objective: To examine the relation of type 2 diabetes to size at birth and childhood growth.
Design: Cohort study.
Setting: Helsinki, Finland.
Participants: Men (n = 3639) and women (n = 3447) who were born at the Helsinki University Central Hospital between 1924 and 1933, who went to school in Helsinki, and who still lived in Finland in 1971. Detailed birth and school health records were available for all 7086 participants. We identified 471 men and women who developed type 2 diabetes by using the national Social Insurance Institution's register of all persons in Finland who are receiving long-term therapy with medication.
Measurements: Incidence of diabetes ascertained from a national register. The main explanatory measurements were size at birth and childhood growth in terms of height, weight, and body mass index.
Results: The cumulative incidence of type 2 diabetes was 7.9% (n = 286) in men and 5.4% (n = 185) in women. The incidence increased with decreasing birthweight, birth length, ponderal index (birthweight/length 3), and placental weight. The odds ratio for type 2 diabetes was 1.38 (95% CI, 1.15 to 1.66; P < 0.001) for each 1-kg decrease in birthweight. The mean weights and heights of the children at 7 years of age who later developed type 2 diabetes were about average. Thereafter, their growth in weight and height was accelerated until 15 years of age. The odds ratio for development of type 2 diabetes was 1.39 (CI, 1.21 to 1.61; P < 0.001) for each standard deviation increase in weight between 7 and 15 years of age. The odds ratio became 1.83 (CI, 1.37 to 2.45; P < 0.001) in an analysis restricted to persons whose birthweights were below 3000 g. Children of both sexes whose mothers had a high body mass index in pregnancy had more rapid growth during childhood and an increased incidence of type 2 diabetes.
Conclusions: These findings are consistent with the hypothesis that type 2 diabetes is programmed in utero in association with low rates of fetal growth. The increased risk for type 2 diabetes associated with small size at birth is further increased by high growth rates after 7 years of age.