OBJECTIVETo examine the relative importance of maternal preexisting type 1 diabetes (T1D), preexisting type 2 diabetes (T2D), and gestational diabetes mellitus (GDM) on risk of attention deficit/hyperactivity disorder (ADHD) in offspring.
RESEARCH DESIGN AND METHODSThis retrospective birth cohort study included 333,182 singletons born in 1995-2012 within Kaiser Permanente Southern California hospitals. Children were prospectively followed through electronic medical records from age 4 years. Relative risks of ADHD associated with diabetes exposures in utero were estimated by hazard ratios (HRs) using Cox regression with adjustment for potential confounders. For GDM, timing of exposure was evaluated by gestational age at diagnosis and severity was assessed by the need for antidiabetes medication treatment during pregnancy.
RESULTSA total of 37,878 (11.4%) children were exposed to diabetes (522 exposed to T1D, 7,822 T2D, and 29,534 GDM). During a median of 4.9 years (interquartile range 2.2, 9.6) of follow-up after age 4 years, 17,415 (5.2%) children were diagnosed with ADHD. ADHD risk was not associated with GDM taken as a whole (P = 0.50) or with gestational age at GDM diagnosis (P = 0.16). However, the risk was significantly greater for the GDM requiring versus not requiring antidiabetes medications (P < 0.001). Compared with children unexposed to diabetes, the adjusted HRs for ADHD in children were 1.57 (95% CI 1.09-2.25) for exposure to T1D, 1.43 (1.29-1.60) for T2D, 1.26 (1.14-1.41) for GDM requiring antidiabetes medications, and 0.93 (0.86-1.01) for GDM not requiring medications.
CONCLUSIONSThe hierarchy of risks suggests that severity of maternal diabetes (T1D vs. T2D vs. GDM requiring antidiabetes medications) influences the risk of ADHD in offspring of mothers with diabetes.