Disaggregating Subgroup Differences in Safe Sleep Practices of Non-Hispanic Black and Hispanic Mothers with a Recent Birth — Massachusetts, 2007–2012

BACKGROUND:  Placing infants supine to sleep can reduce the incidence of sudden unexpected infant death.  In Massachusetts (MA), non-Hispanic black and Hispanic mothers reported lower prevalence of supine sleep (58.1% and 63.8%, respectively) compared with non-Hispanic white mothers (83.4%). To better understand these disparities, we examined frequency of supine sleep position by ethnic subgroups among MA mothers self-identifying as non-Hispanic black mothers and among Hispanic mothers.

METHODS: Data from 2007-2012 MA Pregnancy Risk Assessment Monitoring System (N= 9,041 mothers) were analyzed regarding infant sleep position (supine, prone, side or combination) and maternal characteristics.  Mothers who identified as non-Hispanic black (n=1,718) [Cape Verdean, Haitian, Caribbean Islander, Ghanaian, Nigerian, Ugandan, Kenyan, Somalian, Ethiopian or other non-Hispanic black (mothers without further self-identification)] or Hispanic (n=1,940) [Puerto Rican, Dominican, Salvadoran, Guatemalan, Mexican, Honduran, Colombian, Cuban or other Hispanic] were included. We used multivariate logistic regression to examine adjusted relative risk (aRR) of infant supine sleeping among ethnic subcategories of Hispanic or non-Hispanic black, using the largest subgroup as referent and controlling for maternal education, age, nativity, and marital status. 

RESULTS: Among non-Hispanic black mothers, infant supine sleep position ranged from 54.6% (95% CI: 48.1―61.0%) for Haitian mothers to 81.6% (95% CI: 60.1―92.9) for Ethiopian mothers.  Compared with other non-Hispanic black mothers (n=915), Haitians were significantly less likely to place their infant supine [aRR=0.81(95% CI: 0.68―0.96)], while Ethiopians were more likely to place their infant supine [aRR=1.36 (95% CI: 1.04―1.77)]. 

Among Hispanic mothers, infant supine sleep position ranged from 43.5% (95% CI: 38.5―48.6) for Dominican mothers to 82.5% (95% CI: 62.2―93.1) for Colombian mothers. Compared with Puerto Rican mothers (n=739), Dominicans were significantly less likely to place their infant supine [aRR=0.68 (95% CI: 0.08―0.79)], while Salvadorans, Guatemalans, Mexicans, Cubans and other Hispanics were more likely to place their infant supine [aRR=1.15 (95% CI: 1.03―1.29), aRR=1.16 (95% CI: 1.03―1.31), aRR=1.24 (95% CI: 1.09―1.41), aRR=1.35 (95%CI: 1.06―1.72) and aRR=1.19 (95% CI: 1.02―1.40), respectively].

CONCLUSIONS: Substantial variation in infant supine sleep was observed by ethnic subgroup among mothers who identified as Hispanic and non-Hispanic black. A potential limitation is selection bias, since PRAMS includes only mothers reachable by mail or phone for survey administration.  Infant safe sleep educational programs may need to be tailored for specific maternal ethnic subgroups within Hispanic and non-Hispanic black communities for best infant health outcomes.