Table 3.

Odds of developmental delay by BSID-III domain at 24 months according to HIV exposure

	Infants with delay, n/N (%)		Unadjusted		Adjusted*
	HIV-exposed	HIV-unexposed	OR (95% CI)	p value	OR (95% CI)	p value
Cognitive	18/167 (11%)	52/562 (9%)	1·18 (0·67 to 2·09)	0·56	1·01 (0·55 to 1·85)	0·97
Receptive language	23/165 (14%)	40/556 (7%)	2·09 (1·21 to 3·61)	0·0081	1·96 (1·09 to 3·52)	0·025
Expressive language	18/158 (11%)	31/542 (6%)	2·12 (1·15 to 3·90)	0·016	2·14 (1·11 to 4·15)	0·024
Fine motor	6/166 (4%)	12/562 (2%)	1·72 (0·64 to 4·65)	0·29	1·53 (0·53 to 4·42)	0·44
Gross motor	6/159 (4%)	19/535 (4%)	1·07 (0·42 to 2·71)	0·90	1·23 (0·44 to 3·43)	0·69

We did a complete case analysis by outcome. ORs greater than 1 indicate that HIV exposure was associated with higher risk of delay in that BSID-III domain (ie, poorer outcomes). Total number of participants assessed (N) for each domain was the same for both unadjusted and adjusted models; no covariates had missing data. BSID-III=Bayley Scales of Infant and Toddler Development, third edition. OR=odds ratio.

^*Adjusted for child age, child sex, maternal education, household income, and maternal age.