Sex-specific and sex-independent steroid-related biomarkers in early second trimester maternal serum associated with autism

Table 4 Association between total testosterone and odds of offspring developing autism

Cohort	Mean	Range	SD	Crude models^a			Adjusted models^a,b
Cohort	Mean	Range	SD	OR	95% CI	P Value	AOR	95% CI	P Value
Overall (N = 135)	0.9	0.22–3.50	0.47	0.79	0.38–1.64	0.52	1.01	0.41–2.50	0.979
By sex
Males (n = 81)	0.9	0.34–3.50	0.5	0.6	0.24–1.51	0.275	1.02	0.35–2.97	0.967
Females (n = 54)	0.89	0.22–2.10	0.42	1.34	0.35–5.10	0.672	0.36	0.05–2.45	0.293
By gestational age category^c
Term (n = 97)	0.87	0.30–2.31	0.39	1.05	0.38–2.95	0.921	1.21	0.34–4.30	0.768
Preterm (n = 38)	0.96	0.22–3.50	0.62	0.6	0.15–2.35	0.466	0.43	0.10–1.87	0.262

^aOR and AOR are calculated for every 1 ng/ml increase in total testosterone
^bAdjusted for Principal Component Factor 1 (gestational age, birthweight), Principal Component Factor 2 (maternal age, paternal age, maternal education duration, paternal education duration), pre-pregnancy body mass index (BMI), gestational weight gain, newborn sex. When stratified by sex, sex was removed as a covariate. When stratified by gestational age category, principal component factor 1 was replaced with birthweight
^cPreterm < 37 weeks gestation; term ≥ 37 weeks gestation

ISSN: 2040-2392