Bridging Disciplinary Boundaries (January 11 - 14, 2007)


Homeless and Pregnant: Factors Associated with Pregnancy in a National Sample of Runaway/Homeless Youth

Kimberly A. Bender, MSW, University of Texas at Austin, Rita Watkins, MSW, University of Texas at Austin, Sanna Thompson, PhD, University of Texas at Austin, and David E. Pollio, PhD, Washington University in Saint Louis.

Purpose: Although the teen pregnancy rates continue to decrease, it still remains an issue of great concern. In 2000, approximately 850,000 U.S. teenagers became pregnant; nearly 475,000 babies were born to these teens. One population at increased risk is homeless youth; research indicates as many as 20% of homeless young women become pregnant. These pregnant, homeless teens are at increased risk for low-birth weight babies and high infant-mortality because of inadequate health care, poor dietary habits, and a lack of prenatal care due to their poor economic and social resources. Their problems are exacerbated by a high risk for psychological problems, often as a result of abusive and/or neglectful relationships, victimization, and housing instability. This study investigates predictors of teen pregnancy among a national sample of runaway/homeless youth.

Methods: The study utilized data from the Runaway/Homeless Youth Management Information System (RHYMIS), a database that is required in all federally-funded youth shelters nationwide. From total data collected (N=14,461), a sub-sample of cases positive for experiencing a pregnancy were identified (n=602). A random, matched sample of cases identified as not pregnant was also drawn for comparison (n=602), resulting in a sample of 1204 respondents. Data included demographics, youth factors (e.g. employment, school status, living situation, mental health, criminality, drug/alcohol use, health service use), and family/parent factors (e.g. parental abuse/neglect of child, child's guardian, employment).

Results: Among youth who experienced pregnancy, 80% were females and 20% were males reporting they had fathered a child. Youth averaged 16 years of age and 90% remained single. Those reporting pregnancy had significantly lower rates of returning home to their parents upon shelter exit as compared to non-pregnant peers (chi-square=14.684, p<.001). Logistic regression analyses (model chi square=143.15, p < .001) demonstrated that youth who were older (OR=1.40), were non-white (OR=.71), had dropped out of school (OR=2.18), or reported gang membership (OR=2.18) were more likely to be pregnant or have fathered a child. However, those who abused (OR=.56) or sold (OR=.55) drugs were less likely to report pregnancy. Among family variables, youth who had been emotionally abused by mothers (OR=1.64) or lived with someone other than parents before shelter admission (OR=.69) were significantly more likely to report a pregnancy.

Implications/conclusions: Pregnancy presents a challenge for any teenager; but for runaway/homeless youth, difficulties are amplified. With housing instability, runaway/homeless pregnant teens often seek shelter services; these youth are likely to have multiple difficulties often impossible to address during short-term shelter admissions, such as academic difficulties, parental abuse, and problems associated with gang involvement. Beyond immediate medical care and basic services, shelter providers must be aware of the various difficulties associated with entry into the shelter system among these highly vulnerable pregnant teens. Discharge from shelter must include attention to housing stability as well as safety for the youth and their child. Over the longer term, findings strongly suggest the need for post-shelter care that provides educational and emotional support to this vulnerable population. In addition, coordination between shelters and teen pregnancy programs appears warranted.