To evaluate patterns of mother-infant sleeping behaviors among US-based mothers who received care from midwives and breastfed their infants the majority of time at 6 weeks postpartum.

Infant sleep locations were reported for 24 915 mother-infant dyads followed through 6 weeks postpartum, following midwife-led singleton births. Using data derived from medical records, we used multinomial logistic regression to identify predictors of sleep location.

The median maternal age was 31 years (IQR, 27-34 years). The majority were White (84.5%), reported having a partner or spouse (95%), had a community birth (87%), and reported bedsharing with their infant for part (13.2%) or most of the night (43.8%). In the adjusted analysis, positive predictors of always bedsharing included increasing maternal age (OR, 1.17; 95% CI, 1.13-1.21; per 5 years), cesarean birth (OR, 1.49; 95% CI, 1.18-1.86), Medicaid eligibility (OR, 1.76; 95% CI, 1.62-1.91), and maternal race/ethnicity (Black OR, 1.40 [95% CI, 1.09-1.79]; Latinx OR, 1.53 [95% CI, 1.35-1.74]; multiracial OR, 1.69 [95% CI, 1.39-2.07]). Negative predictors of bedsharing included having a partner/spouse (OR, 0.66; 95% CI, 0.56-0.77) and birth location in hospitals (OR, 0.56; 95% CI, 0.49-0.64) or birthing centers (OR, 0.48; 95% CI, 0.44-0.51). Partial breastfeeding dyads were less likely to bedshare than those who were exclusively breastfeeding (always bedsharing OR, 0.48 [95% CI, 0.41-0.56]; sometimes bedsharing OR 0.69 [95% CI, 0.56-0.83]).

These data suggest that cosleeping is common among US families who choose community births, most of whom exclusively breastfeed through at least 6 weeks.