Maternal diabetes linked to a slight increase in ADHD risk in children

By Tarun Sai LomteApr 9 2024Reviewed by Susha Cheriyedath, M.Sc.

In a recent study published in the journal Nature Medicine, researchers evaluated associations between maternal diabetes mellitus (MDM) and the risk of attention-deficit/hyperactivity disorder (ADHD) in the offspring.

Sixteen percent of pregnant individuals experience hyperglycemia worldwide. The prevalence of MDM has increased globally, which is associated with the advancing maternal age, the growing obesity epidemic, and improved MDM diagnostic approaches. Animal studies have shown the adverse effects of hyperglycemia in pregnancy on intrauterine oxidative stress, inflammatory response, and epigenetic mechanisms, which might lead to poor neurodevelopment in the offspring.

ADHD is a neurodevelopmental disorder characterized by impulsivity, hyperactivity, and inattentiveness. Globally, 2% to 7% of children are affected by ADHD. Besides, ADHD can have a substantial burden on families of affected individuals and society. Evidence suggests that gestational diabetes mellitus (GDM) and pre-GDM (PGDM) are associated with ADHD. A meta-analysis revealed that the offspring of diabetic mothers had a 40% increased risk of ADHD.

Study: Maternal diabetes and risk of attention-deficit/hyperactivity disorder in offspring in a multinational cohort of 3.6 million mother–child pairs. Image Credit: Pixel-Shot / Shutterstock

About the study

In the present study, researchers assessed the association between MDM and ADHD risk in the offspring. This population-based cohort study used healthcare data from New Zealand, Hong Kong, Taiwan, and Nordic countries (Iceland, Sweden, Norway, and Finland). The study included children from live births within site-specific periods. Mother-child pairs were linked using exact deterministic linkage.

Children without complete birth information, six years of follow-up, and valid linkage were excluded. Follow-up commenced at birth and continued until outcome occurrence, death, or end of study period. The primary exposure was MDM, including PGDM and GDM. MDM was stratified into (unmedicated and medicated) GDM and (types 1 and 2) PGDM.

ADHD was defined using site-specific diagnosis and medication codes. The primary analyses compared ADHD status in children born to mothers with any type of diabetes with those born to non-diabetic mothers. Sibling-matched analyses compared the ADHD status in children of the same mother but with discordant GDM status.

In secondary analyses, ADHD status was compared between children born to mothers with different subtypes of diabetes. The researchers computed hazard ratios of average treatment effect and 95% confidence intervals to examine associations between MDM status and ADHD using Cox proportional hazard regression models.

Covariates included demographic factors, socioeconomic status, birth year, multifetal pregnancies, body mass index (BMI), use of psychotropic medication, alcohol and smoking status, neurologic and psychiatric conditions, and other chronic conditions. Several sensitivity analyses were also performed to evaluate the robustness and validity of the findings.

Findings

The study included more than 3.6 million mother-child pairs. About 6.6%, 8%, 4.1%, and 13.7% of children had mothers with diabetes in the Nordic countries, Hong Kong, New Zealand, and Taiwan, respectively. ADHD risk was higher among children whose mothers had any type of diabetes than in those born to non-diabetic mothers. ADHD risk was higher in children born to mothers with PGDM, type 1 PGDM, type 2 PGDM, or GDM compared to those born to non-diabetic mothers.

Sibling-matched analyses did not find differential risks of ADHD. ADHD risk was similar among children whose mothers were diagnosed with GDM at different trimesters in Hong Kong. However, ADHD risk was the highest in children whose mothers had GDM diagnosis in the first trimester in New Zealand and Taiwan. Further, children whose mothers had GDM might have a lower ADHD risk than those born to mothers with PGDM.

ADHD risks did not differ between children born to mothers with type 1 and type 2 PGDM. Children whose mothers had GDM and received anti-diabetic medications had a similar risk of ADHD compared to those born to mothers with GDM who were unmedicated. Sensitivity analyses produced similar results as primary analyses.

Conclusions

In sum, the findings revealed that MDM, PGDM, and GDM were associated with a small/moderate risk of ADHD in the offspring. ADHD risks did not differ between siblings with discordant GDM status in pregnancy, suggesting potential confounding by unmeasured, shared familial or genetic factors. Moreover, ADHD risk estimates were smaller compared to a previous meta-analysis. Future studies should reevaluate the specific roles of hyperglycemia and genetic factors in the relationship between MDM and ADHD.