Table 2. Characteristics of studies that associate exposure to agricultural pesticides with autism by maternal biomarkers – studies cited according to the year of publication.

Author	Study population	Exposure and outcome	Results
Cheslack-Postava et al. 13	Children born in Finland between 1991 and 2000 (n=75 case-control pairs)	Maternal serum biomarkers (collected during the first trimester of pregnancy) analyzed by chromatography. Finnish registry with the diagnosis of ICD10 F84.0 and subsequent validation with ADI-R.	All samples, except DDT and BDE-47, were measured at levels above the LOQ. Total PCBs at or above the 90th percentile were associated with an OR=1.91 for autism.
Lyall et al. 20	Children born in three California counties between 2000 and 2003 with ASD (n=545), ID (n=181), and GP (n=418).	Samples of serum and maternal blood cell sediment (collected during 15–19 weeks of gestation) analyzed by chromatography. Previous registration and revised by MADDSP protocol. ASD classified by DSM-IV ed. With or without ID (functional cognitive test <70).	In the ASD group, the levels of PCBs and trans-nanochlorines were higher than in the CG. Increased risk of ASD mainly for PCBs 138/158, 153, 170, and 180 (aOR>1.5). The sum of PCB congeners demonstrated a moderately high risk of ASD for the upper quartile (OR=1.39). The trend tests were significant only for PCB 138/158 (p=0.03) and PCB153 (p=0.04). Increased risk of ASD for PCBs was not observed, either crude or adjusted.
Brown et al. 19	Children born in Finland between 1987 and 2005 (n=778 case-control pairs).	Maternal serum biomarkers (collected between 2 and 4 months of pregnancy) analyzed by chromatography. Hospital record with diagnosis of ICD10 F84.0 and validation with the Autism Diagnostic Interview.	p,p’-DDE was measured in an amount greater than the LOQ in all samples. Among similar PCBs, it was measured above the LOQ between 95% and 100% of the samples. Increased chances of autism in offspring when maternal levels of p,p’-DDE in the 75th percentile (OR=1.32 and p=0.03), male (OR=1.35 and p=0.04), and also individuals with intellectual disabilities (OR=2.21 and p=0.002). No association was observed for PCB 138 or PCB 153 as well as the other total PCBs, congeners, and autism.

ASD: autism spectrum disorders; ID: intellectual disability; OR: Odds Ratio; aOR: adjusted Odds Ratio; GP: general population; p,p’-DDE: dichlorodiphenyldichloroethylene; PCB: polychlorinated biphenyls; NEQ: neurotoxic equivalent; LOQ: limit of quantification; HCB: hexachlorobenzene; BDE-47: 2,2’4,4’-tetrabromodiphenyl ether; ICD10 F84.0: International Classification of Disease for Autistic Disorder; DSM-IV: Diagnostic and Statistical Manual of Mental Disorders, 4th edition; ADI-R: Autism Diagnostic Interview – Revised; MADDSP: Metropolitan Atlanta Developmental Disabilities Surveillance Program.