Among homeless or runaway teenage women who seek short-term shelter services, those with difficult family situations have an increased risk of being pregnant, according to an analysis of a national sample of this population.1 Women who felt abandoned by their families or who had been emotionally abused by their mothers had about 50% higher odds of being pregnant than those not reporting these situations; those who lived in two-parent families had reduced odds of being pregnant. Teenagers’ likelihood of being pregnant was also linked to the amount of time they had been away from home, their school enrollment status and their age, among other characteristics.

Even though homeless or runaway youth often lack familial and community support, and tend to engage in high-risk behaviors, little research has examined pregnancy among this population at the national level. To examine possible risk factors for pregnancy among this group and possible differences between young women who are pregnant and those who are not, researchers used data from the 1997 Runaway/Homeless Youth Management Information System, the only source of national data available on young people who obtained services at youth emergency shelters. The study was limited to 12–18-year-olds who lived away from home at least overnight without permission or supervision. The researchers identified young women who were pregnant when they arrived at the shelter and randomly selected an equal number of female adolescents, matched by age, who were not pregnant to serve as a comparison group; the final sample consisted of 476 pregnant and 475 nonpregnant teenagers.

On average, women in the sample were 17 years old; 59% were white, 23% were black and 13% were Hispanic. One-quarter had dropped out of school. Substantial proportions had used alcohol (75%) or illicit drugs (61%). Twenty-seven percent of the young women had lived in a two-parent household; some 16–32% had been physically abused, emotionally abused or neglected by at least one parent.

Chi-square and t tests revealed significant differences between pregnant and nonpregnant teenagers on a wide array of individual and family-level characteristics. For example, a higher proportion of pregnant adolescents were black (27% vs. 20%) or Hispanic (16% vs. 10%). Roughly twice as many pregnant as nonpregnant young women felt abandoned by their family (30% vs. 16%) and reported a variety of high-risk behaviors: having an STD (6% vs. 3%), having dropped out of school (33% vs. 17%) and being on probation (7% vs. 4%). Some 33% of young women who were not pregnant had lived with two parents, while 22% of those who were pregnant had. Though a smaller proportion of pregnant adolescents than of others had physically abusive mothers (15% vs. 20%), the reverse was true regarding emotionally abusive mothers (35% vs. 29%). Seventy-three percent of pregnant teenagers had stayed away from home for more than two days, compared with 63% of nonpregnant teenagers. A few high-risk behaviors, including drug use and attempted suicide, were equally common for both groups.

In a logistic regression analysis, most of these relationships remained significant. Teenagers who had been emotionally abused by their mothers or who felt abandoned by their family had higher odds of being pregnant than those who did not report these situations (odds ratio, 1.5 for each). The odds of pregnancy were also elevated for young women who had lived away from home for two or more days (1.4), had dropped out of school (2.2) or had an STD (2.2). Interestingly, those who reported having physically abusive mothers had lower odds of being pregnant than those who did not (0.5). White youth and those who had lived with two parents also had reduced odds of being pregnant (0.6 and 0.8, respectively), and the likelihood of pregnancy fell with increasing age (0.9).

While the results of this study indicate that “long-term family difficulties” and a lack of “positive experiences in educational settings” are associated with homeless or runaway young women’s risk of pregnancy, the investigators acknowledge that the findings should be viewed “with caution” because of a few important limitations. Specifically, causality cannot be determined since the study was cross-sectional, and the questions and definitions that shelters used in data collection were not standardized. In addition, the risk of pregnancy may be even higher for homeless or runaway youth not included in this sample, such as those who do not access shelter services. However, these findings point toward the types of support needed by these young women, according to the researchers, who suggest that shelters could serve as a gateway to services like teenage parenting programs. Beyond this, they encourage service providers to “help the pregnant teen to recognize the complexity of her problems and to take steps to overcome them rather than simply running away.”


1. Thompson SJ et al., Runaway and pregnant: risk factors associated with pregnancy in a national sample of runaway/homeless female adolescents, Journal of Adolescent Health, 2008, 43(2):125–132.