Maternal Use of Prenatal Nutritional Supplements and Risk of Autism in the Stockholm Youth Cohort

Background:  Recent epidemiological studies have suggested that prenatal use of folic acid (FA) or iron may reduce risk of autism spectrum disorders (ASD). These findings require verification.

Objectives:  To examine associations of prenatal use of FA, iron, and multivitamins and risk of ASD with and without intellectual disability (ID). 

Methods: Data on 254,405 mother/child pairs born 1996-2007 (1,074 ASD with ID; 3,700 ASD without ID) were drawn from the Stockholm Youth Cohort, a register-based cohort in Stockholm County, Sweden. ASD case status as of December 2011 was ascertained using national and regional registers covering all pathways to ASD diagnosis and care in Stockholm County, and this approach has been previously validated. Supplement use was assessed at first antenatal visit.  The sample was categorized into mutually exclusive user groups: multivitamin or multiple supplement use (N=43,634); FA only use (8,494); iron only use (53,575); and non-use (148,702). Covariates included maternal age, BMI, national origin, family income, antiepileptic use, prior hospitalizations the year before pregnancy, psychiatric history, child gender, and year of birth. 

Results: Supplement users were different from non-users across multiple medical and social characteristics.  Multivitamin users were older, more likely to be born in Sweden, have higher income, and to be healthier overall than non-users (more likely to have normal BMI, fewer hospitalizations, and less psychiatric illness).  Comparatively, folic acid users were older, more likely to be immigrants, have lower income, have normal BMI, have less psychiatric history but more hospitalizations, and greater use of antiepileptic medications than non-users. In adjusted models, compared to risk of ASD with ID (low functioning autism) in non-users, odds ratios (95% CI) were: multivitamin use: 0.79 (0.65-0.95); iron use: 1.10 (0.95, 1.27); FA use: 1.40 (1.04, 1.88). Supplement use was not associated with ASD without ID (high functioning autism). 

Conclusions:  The apparent increased risk of ASD with ID for FA may be because FA is prescribed for women with neurological disorders that increase risk of child ASD. Although results suggest multivitamin use is associated with lower risk of ASD with ID, confounding is likely since users were different from non-users. Even if associations were causal, whether a particular nutrient is protective cannot be identified since formulations include multiple nutrients. Further investigation is required, especially of the heterogeneous results by presence of comorbid intellectual disability.