New mothers who had been homeless at some point during the year before their child’s birth are less likely than others to have received early prenatal care or to have taken their infant for a well-baby checkup, according to data from a multistate surveillance system.¹ Women who had been homeless around the time of their pregnancy had had fewer prenatal visits than other women (mean, 10.0 vs. 11.6) and had higher odds of not having had a visit during the first trimester (odds ratio, 2.0). Homelessness was positively associated with not having taken the infant for any well-baby checkups within 2–4 months of delivery (2.4) and negatively associated with having breast-fed for at least eight weeks (0.7). In addition, mothers who had been homeless had an elevated likelihood of having had a low-birth-weight or very low birth weight infant.

At any given time, more than 600,000 U.S. residents are homeless, which may impede their ability to obtain prenatal and other medical care. To examine the prevalence and outcomes of births to homeless women, researchers examined data from the Pregnancy Risk Assessment Monitoring System (PRAMS), a program that collects information about prenatal, neonatal and postnatal outcomes from women who had delivered 2–4 months earlier. The investigators analyzed data collected in 31 states that participated in PRAMS for one or more years between 2000 and 2007. Women were sampled systematically from birth certificates; more than 11,000 homeless women and 257,000 nonhomeless women were surveyed, representing weighted populations of 442,000 homeless women and 10.2 million nonhomeless women.

Women were classified as homeless if they reported having been homeless at any time during the 12 months before delivery. Demographic measures used in the analysis included age, race and ethnicity, education, marital status and whether the woman had health insurance or Medicaid or had received government aid in the past year. A range of maternal and child health measures were also assessed, including prepregnancy body mass index, cigarette and alcohol use, number of prenatal care visits and well-baby checkups, breast-feeding duration, birth weight and whether the infant had been treated in the neonatal intensive care unit. In addition to compiling descriptive statistics, the researchers performed multiple logistic regression analyses that adjusted for maternal age, race and ethnicity, and religion to identify behaviors and birth outcomes associated with homelessness.

Overall, 4% of PRAMS respondents had been homeless in the 12 months before the birth of their infant; the proportion was highest in Illinois (7%), Oregon and Washington (6% each). Homeless women were more likely than other women to be black rather than white (odds ratio, 1.9), Hispanic rather than non-Hispanic (7.7) and unmarried rather than married (2.3). In addition, homeless women were less likely than nonhomeless women to have health insurance and to have taken vitamins during the preconception period, and more likely to have been underweight before pregnancy, to have had class III obesity (body mass index of 40 or more) before pregnancy and to have smoked at least 100 cigarettes in the past two years.

In general, maternal and infant health outcomes were less positive among homeless women than among other respondents. While 78% of nonhomeless women had had a prenatal visit during their first trimester, only 57% of homeless women had had one; in the 2–4 months after delivery, 95% of nonhomeless women, but only 88% of homeless women, had obtained a well-baby checkup for their infant. In multiple regression analyses, homeless women had higher odds than other women of not having had a prenatal visit in the first trimester (odds ratio, 2.0), not having had at least one well-baby checkup (2.4) and not having initiated breast-feeding after delivery (1.4). They were less likely than other women to have had an infant who did not require a stay in the intensive care unit (0.8) and to have breast-fed for at least eight weeks rather than for less than one week (0.7).

Although gestational age did not differ between homeless and other women after adjustment for maternal age, race and ethnicity, and region, homelessness was associated with fewer prenatal care visits (mean, 10.0 vs. 11.6) and lower mean infant birth weight (3,242 vs. 3,311 g). Compared with other women, homeless women had higher proportions of low-birth-weight infants (7% vs. 6%) and very low birth weight infants (1.6% vs. 1.3%).

The researchers note that the study’s limitations include the simplicity of the PRAMS homelessness measure, which did not determine timing or duration; the adverse effects of homelessness and the accuracy of respondents’ recall might be expected to vary according to these factors. Moreover, because PRAMS is administered only to women with known contact information, the survey may underestimate the prevalence of homelessness. Nonetheless, the findings suggest that homelessness "can adversely affect maternal health behaviors during pregnancy and subsequent infant health outcomes." The investigators suggest that future studies examine the impact of homelessness on factors that influence pregnancy weight gain and infant feeding practices and, in turn, affect other aspects of maternal and infant health.—P. Doskoch