Maternal health around pregnancy and autism risk: a population-based study.

IMPORTANCE: Many maternal diagnoses in pregnancy have been linked with offspring ASD risk. However, pregnant women receive many other diagnoses, most of which have not been evaluated for an association with offspring ASD. OBJECTIVE: Systematically test the associations between maternal diagnoses around pregnancy, and ASD risk in offspring, filtering out potential false positives and accounting for correlation between maternal diagnoses. DESIGN, SETTING AND PARTICIPANTS: This case-cohort study included children born in Israel from January 1, 1999, through December 31, 2008, and followed up until January 26, 2015. We used information on all ICD-9 codes received by their mothers during pregnancy and the preceding year. ASD risk in children whose mothers experienced each of those conditions during the exposure period was compared with ASD risk in unexposed children. Data were analyzed from December 1, 2019, through April 30, 2020. MAIN OUTCOME AND MEASURES: Hazard ratios (HRs) and 95% confidence intervals (CIs) of ASD risk associated with each maternal ICD-9 code, calculated using Cox proportional hazards regression, adjusted for the confounders (birth year, maternal age, socioeconomic status and number of ICD-9 diagnoses during the exposure period). RESULTS: The analytic sample consisted of 80,187 individuals (1,132 cases, 79,055 controls; median [interquartile range] age at the end of follow up was 11.2 [8.8-13.6]; 48.8% female). Of 822 ICD-9 codes tested, 148 were recorded in at least 10 of each case and control mothers. Of those, 22 maternal diagnoses were nominally significantly associated with offspring ASD, and 16 of those survived subsequent filtering steps (permutation testing, multiple testing correction, multiple regression). Among those 16 diagnoses, we recorded increased risk of ASD associated with certain metabolic (e.g. hypertension (ICD-9 401); HR=2.74(1.92-3.90), P=2.43E-8), genitourinary (e.g. noninflammatory disorders of cervix (ICD-9 622); HR=1.88(1.38-2.57), P=7.06E-5) and psychiatric (depressive disorder (ICD-9 311); HR=2.11(1.32-3.35), P=1.70E-3) diagnoses. Meanwhile, mothers of children with ASD were less likely to attend prenatal care appointment (ICD-9 V22; HR=0.62 (0.54-0.71), P=1.80E-11). CONCLUSIONS AND RELEVANCE: We observed an association between ASD and 16 maternal ICD-9 diagnoses, after rigorous filtering out potential false positive associations. Replication in other cohorts and further research to understand the mechanisms underlying the observed associations with ASD are warranted.