Type 2 Diabetes in the Washington State Twin Registry

BackgroundOnce considered a disease of older age, type 2 diabetes (T2D) is now prevalent across both younger and older age and all demographic groups in the US. The reasons for this prevalent health problem span biologic to... [ view full abstract ]

Background

Once considered a disease of older age, type 2 diabetes (T2D) is now prevalent across both younger and older age and all demographic groups in the US. The reasons for this prevalent health problem span biologic to policy-level factors. In adults 45 years and older, the DISCOTWIN consortium showed a moderate to high contribution of genetic factors in structural equation modelling of T2D in twins from European and Australian cohorts, with a pooled heritability estimate of 72%. The purpose of this study was to investigate the contributions of genetic and environmental factors from structural equation modeling of T2D in twins 45 years and older in a US twin cohort and compare the estimates with those reported in the DISCOTWIN consortium, and to determine how these factors differ in older and younger (above and below 45 years) US twins.

Methods

Data was obtained from 6,881 same-sex monozygotic (MZ) and dizygotic (DZ) twin pairs from the Washington State Twin Registry that responded to the question “Has a doctor ever diagnosed you with (type 2) diabetes?” Twins missing a response were excluded. In 43% of the sample, diabetes type was not specified, whereas in the remainder the question specifically queried on type 2 diabetes. We combined responses to both questions to ascertain T2D prevalence because T2D accounts for 90% of all diabetes cases in the Registry. Twin similarity in T2D among those over and under 45 was analyzed using both tetrachoric correlations and structural equation modeling. Twin pairs discordant for being above and below 45 years were excluded, unless the younger twin was within 6 months of turning 45, in which case they were included in the above 45 group.

Results

There were 4,187 pairs (70% MZ, 64% female) under and 2,694 pairs (64% MZ, 64% female) over 45 in the sample. Overall, 9.3% of MZs over 45 and 1.6% of MZs under 45 were discordant for T2D. In both MZ and DZ twins discordant for T2D, the twin with T2D had a significantly higher BMI than the nonaffected cotwin. Correlations were higher in MZ (r_MZ=0.75, 95% CI=0.68-0.81) than DZ (r_DZ=0.51, 95% CI=0.38-0.62) pairs over 45. However, differences in the MZ-DZ correlations were attenuated in twin pairs under 45 (r_MZ=0.65, 95% CI=0.45-0.79; r_DZ=0.67, 95% CI=0.46-0.80). In twins over 45, the best fitting model only included additive genetics and unique environment, with a heritability of 77%, consistent with the DISCOTWIN findings. However, in twins under 45, the best fitting model included only the shared and unique environment, with the shared environment contributing 66% of the variance.

Conclusion

Consistent with findings from European and Australian twin cohorts, heredity is the strongest influence on T2D in US twins over 45. On the other hand, the shared environment is the strongest influence on T2D in US twins under 45.