A neighborhood’s level of physical degradation and availability of walking spaces are associated with residents’ maternal health behaviors and pregnancy outcomes, according to findings from a cohort study in North Carolina.1 After maternal covariates were adjusted for, neighborhood level of physical degradation was positively associated with white women’s odds of smoking during pregnancy, gaining an inadequate or an excessive amount of weight during pregnancy, having pregnancy-induced hypertension, and bearing a low-birth-weight or preterm infant; it was positively associated with smoking and inadequate weight gain during pregnancy among black women, but not with any pregnancy outcomes. For both white and black women, the greater the availability of walking spaces, the lower the likelihood of smoking or gaining inadequate weight during pregnancy.
Researchers conducted a geographically defined cohort study using birth records from 2001–2005 and directly observed street-level data collected in 2005 and 2006. Women giving birth before 22 weeks’ gestation or after 42 weeks’ were excluded, as were those with infants weighing less than 500 g or more than 6,000 g at birth, and those for whom a stillbirth outcome was unclear; in all, 23,304 women with singleton births were included. The maternal risk factors examined were inadequate and excessive weight gain (defined as less than 15 pounds and more than 40 pounds, respectively) and smoking; the pregnancy outcomes were pregnancy-induced hypertension (including eclampsia) and having a low-birth-weight baby (one delivered at 37–42 weeks and weighing less than 2,500 g) or a preterm birth (delivery of an infant weighing less than 3,888 g before 37 weeks’ gestation).
Five neighborhood indices were constructed, comprising 3–7 variables each. The indices assessed neighborhood degradation, availability of social spaces, walkability, borders (physical signs of property boundaries) and features for pedestrian and traffic safety. To determine the associations between neighborhood variables (categorized by quartile) and pregnancy-related behaviors and outcomes, the researchers estimated multilevel logistic models, stratified by race. The analyses were adjusted for maternal age and education, marital status and previous pregnancies.
Three-fifths of women in the sample were white, and the rest were black. The largest proportion of white mothers were aged 30–34 (36%), while the largest proportion of black mothers were aged 20–24 (29%). A large proportion of whites had completed more than a high school education (78%); fewer than half of blacks had done so (48%). Eighty-six percent of white women and 39% of blacks were married; 39% and 30%, respectively, reported no previous pregnancies. In each racial group, approximately 10% reported smoking during pregnancy, 25% had excessive weight gain and 5% had pregnancy-induced hypertension. Among blacks, 12% had inadequate weight gain, 13% had preterm births and 4% had low-birth-weight infants; among whites, the proportions were 5%, 7% and 1%, respectively.
The largest proportion of white women lived in neighborhoods with the lowest levels of degradation (34%), while the largest proportion of blacks lived in neighborhoods with the highest levels (38%); in each group, participants were equally distributed across the social spaces quartiles. A greater proportion of whites resided in the most walkable neighborhoods than in the least walkable ones; black women were evenly distributed by walkability quartile. Most women in both groups lived in neighborhoods with well delineated property boundaries. Most black women lived in areas with high levels of traffic safety features; most white women lived in areas with lower levels.
After adjustment for maternal covariates, level of neighborhood degradation was positively associated with every adverse maternal health behavior and pregnancy outcome studied among white women, but it had fewer associations among blacks. White women living in areas with the most degradation had odds of smoking that were twice those for whites living in areas with the least degradation; differences for other outcomes were smaller (odds ratios, 1.2–1.9). For black women, the greater the level of neighborhood degradation, the higher the likelihood of smoking and inadequate weight gain; the odds of these outcomes were, respectively, 65% and 41% higher for women in the most degraded areas than for those in the least degraded areas.
Among white women, the odds of smoking and inadequate weight gain during pregnancy increased with the availability of social spaces; the odds of these behaviors were, respectively, 82% and 40% higher in areas with the most such spaces than in areas with the fewest. Availability of social spaces was not significant among blacks.
Within each racial group, the more walkable the neighborhood, the lower the likelihood of smoking and inadequate weight gain during pregnancy. Compared with women in the least walkable neighborhoods, those in the most walkable neighborhoods had 40–44% lower odds of smoking and 26–29% lower odds of inadequate weight gain. In addition, white women in these areas had reduced odds of having a preterm birth (odds ratio, 0.8).
The presence of property boundaries was associated only with an increasing likelihood of inadequate pregnancy weight gain among blacks. Black women in neighborhoods with the most property boundaries had 28% higher odds of this outcome than blacks in areas with the fewest boundaries.
As the level of traffic safety features increased, the odds of smoking decreased among white women; whites living in areas with the most traffic safety features had odds of smoking that were half those for whites in areas with the least traffic safety features. For black women, as the level of traffic safety features increased, the odds of pregnancy-induced hypertension decreased; black women in areas with the most traffic safety features had 27% lower odds of pregnancy-induced hypertension than those in neighborhoods with the least traffic safety features.
The researchers acknowledge several shortcomings of their study, including its limited generalizability, the lack of data on races other than white and black, and the imprecision of measures on pregnancy weight gain and preterm birth. Despite these limitations, the researchers conclude that their results highlight "the associations between neighborhood environment and maternal health behaviors, illustrating a potential pathway to negative birth outcomes."—L. Melhado