Homeless adolescent mothers: a systematic review of the literature

Abstracts

Pouco se sabe sobre mães adolescentes que passam por gravidez ou maternidade fora de casa. Este estudo é uma revisão da literatura epidemiológica sobre esse fenômeno crescente. Os artigos foram identificados através de pesquisa nas seguintes bases de dados eletrônicas: MEDLINE, Lilacs, SciELO, PsychINFO, CINHAL, ERIC e Sociological Abstracts; 19 referências foram recuperadas. Os estudos mostram ampla heterogeneidade dos objetivos e dos métodos e revelam principalmente que as adolescentes têm elevadas taxas de abuso de substâncias, transtornos mentais, falta de apoio social, comportamento sexual, violência física e sexual; gravidez e interação mãe-criança problemáticas. Poucos artigos resultaram da revisão metodológica realizada sobre essa população. As mães adolescentes de rua têm grande exposição à violência, abuso de drogas e risco de problemas de saúde física e mental. Mais estudos centrados no estigma e em métodos de intervenção para esse grupo de mulheres são necessários nesta área, especialmente em populações culturalmente distintas.

Adolescente de rua; gravidez na adolescência; maternidade

REVIEW ARTICLE

Homeless adolescent mothers: a systematic review of the literature

Anne Lise Sandoval Silveira Scappaticci^I; Sergio Luis Blay^II

^IDoutora, Ciências da Saúde. Professora, Departamento de Psiquiatria, Universidade Federal de São Paulo - Escola Paulista de Medicina (UNIFESP-EPM), São Paulo, SP, Brazil.

^IIDoutor, Psiquiatria. Professor associado, Departamento de Psiquiatria, UNIFESP-EPM.

Correspondence

ABSTRACT

Little is known about adolescent mothers who are homeless during pregnancy and motherhood. This study is a review of the epidemiological literature about this increasing phenomenon. Articles were identified by searching the following electronic databases: MEDLINE, Lilacs, SciELO, PsychINFO, CINHAL, ERIC and Sociological Abstracts; 19 studies were retrieved. The studies showed a wide heterogeneity of objectives and methods, and mainly revealed that the teenagers have high rates of substance abuse, mental disorders, lack of social support, sexual behavior, physical and sexual violence, and troublesome pregnancy and mother-child interaction. Our methodological review found few articles about this population. Homeless adolescent mothers are extensively exposed to violence, drug abuse, and risk of physical and mental health problems. Further studies are needed about this topic, mainly involving culturally different populations, focusing on stigma and intervention methods for this group of women.

Keywords: Homeless adolescent, teenage pregnancy, motherhood.

INTRODUÇÃO

Teenage pregnancy has been a focus of health research. The current literature reveals much about the life contexts that lead teenagers to run away from home: victimization, domestic violence, sexual abuse, substance abuse, mental health issues, poverty, limited social contacts, and residential instability are often precursors to episodes of homelessness among teenagers.¹ On the other hand, although teenage families with children are one of the fastest growing homeless subpopulations, relatively few studies have focused on pregnancy or parenting among homeless teenagers.^2,3

Family homelessness has been attributed to a variety of psychological and interpersonal risk factors, including mental illness,⁴ histories of physical or sexual abuse,^5,6 domestic violence, and drug abuse.⁷ In addition, research on homeless and low-income mothers focusing on social conditions and poverty has revealed that homeless mothers had less network support than housed mothers.⁸ There are also some studies indicating developmental delays among such children^4,9,10 or impairment of the mother-child relationship due to drug abuse among mothers living on the streets.¹⁰

A large study on this population was carried out in Worcester, MA, USA (Worcester Family Research Project, WFRP). This recent epidemiological study investigated factors that might be protective against family homelessness. Using multivariate modeling, the protective factors included housing subsidies, graduation from high school, having more people in one's social network, and having fewer conflicting relationships. Factors that reduced a family's economic and/or social capital were also associated with homelessness. For example, mental hospitalization within the last 2 years and frequent use of alcohol or heroin were risk factors, although they were uncommon in the sample.¹¹

In Brazil, adolescent families with children are of great concern.^12-14 However, to our knowledge, there is no investigation summarizing this literature about young women experiencing motherhood or pregnancy in temporary shelters. The objective of the present study was to provide a review of epidemiological studies on pregnancy and motherhood among teenagers living in temporary shelters. Specifically, our objectives were to: 1) consolidate the available findings in the recent literature; and 2) identify the main characteristics of mental health and behavior associated with the phenomenon among these teenagers.

METHODS

Literature search

Search sources

The following electronic databases were consulted: MEDLINE, Lilacs (Latin American Caribbean Literature in Health Sciences Information), SciELO (Scientific Electronic Library on line), PsychInfo, CINHAL, ERIC and Sociological Abstracts. In addition to the electronic search, there was also a manual search for references in articles, books, dissertations, and manuscripts submitted for publication. We asked investigators for reprints of studies that we could not find and set three months as a time limit for receiving information.

Search strategy for electronic databases

Our inclusion criteria were deliberately broad due to the difficulty of finding studies regarding this subject. Since homeless women are often associated with sexual abuse and maltreatment, mental disorders, drug abuse, poverty, foster care, pregnancy, and disintegrated family backgrounds, we initially used all keywords associated with these themes, such as: "homeless adolescents" OR "foster care children" OR "adolescent motherhood" OR "pregnancy in adolescence" OR "mother child relations."

Selection of epidemiological studies

Inclusion criteria

a) All studies were original studies in English or Portuguese. We decided to focus this review on the recent literature, evaluating studies conducted in the last 24 years (1985-2009) when this subject became more of a societal concern.

b) Age: based on the WHO criteria,¹⁵ we defined adolescents/teenagers as between 10 and 19 years old. Studies with large homeless adult samples, but including teenagers as a subgroup, were also included.

c) We searched for epidemiological studies with large population samples, or even case series in hospitals, outpatient settings or obstetrics centers, which included teenage females, pregnancy or motherhood, and homeless teenagers living in out-of-home placements.

d) A clear description of the study methodology was required, including the design, sampling strategy, and assessment instruments. Face-to-face interviewing, i.e. questionnaires applied through personal contact, were included in the study.

Exclusion criteria

We excluded theoretical articles, investigations with an unclear or poor description of the methodology used, and manuscripts based on annual statistical reports such as data census and information obtained in an indirect manner through charts or files. We also excluded qualitative studies, which will be addressed in another study. Studies exclusively dealing with drug abuse, HIV, and sexually transmitted diseases were also excluded.

Procedures

The article search strategy was to read through the titles to find articles that investigated homeless teenage mothers. Next, the abstracts were read, and then the full article. When the title and abstract did not make it clear whether the study included a group of teenage mothers, the full text was examined.

When there was doubt about inclusion, the article was read by an independent examiner and then the decision to include or exclude it was made consensually by two independent reviewers (AS, SLB). The reviewers were not blind to the names of authors, institutions or publishing journals and magazines.

Quality assessment and data management

The methodological quality of the selected studies was assessed by two independent reviewers (AS, SLB), who also extracted the data. If the studies failed to describe the size of the teenage sample, the authors of the original studies were contacted for additional information. The quality criterion was defined as A or B. Quality A studies were surveys using valid and reliable questionnaires that were applied in person by trained interviewers, among locally representative (or shelter) randomly selected samples that were defined as runaway teenagers. Studies that did not fulfill one or more criteria for quality A were classified as quality B.

Data extraction and analysis

Country, year of data collection, sociodemographic measurements, sample size, prevalence rates, and methodological characteristics of the studies (such as design, setting, type of interview, runaway teenager definition) were extracted from the selected papers. First, the studies were analyzed qualitatively. Depending on the nature and outcome of the studies, meta-analysis could then be performed.

RESULTS

The search strategy resulted in 8,062 studies. The article titles were then reviewed in order to exclude the ones that were not of interest for this project. This left 3,682 articles, from which we read the abstracts. After excluding manuscripts based on their abstracts, we then examined 278 manuscripts by reading them in full. There was full agreement between the judges. A flow chart of the selection procedure is shown in Table 1.

Finally, 19 articles fulfilled the selection criteria and were included in this review (Table 2). They were all classified as quality B in accordance with the criteria adopted.

The Worcester Family Research Project, which was a large study of homeless and low-income families conducted in Worcester, MA, USA, during the 1990s, produced a large amount of data. Manuscripts based on this study^9,16 were included in this review. With the exception of one study done in Australia⁷ and another in Brazil,¹⁷ all of this research was carried out in the USA.

Methodological appraisal

The studies were conducted with different methodological focuses. The characteristics of the study designs are described next.

Objectives

The studies had varying objectives. There were investigations about: 1) drugs in pregnancy and motherhood^7,10; 2) pregnancy and its impact^18-21; 3) mother-child interaction^9,23; 4) psychosocial characteristics of homeless women^{4,6,8,11,20,24}; 5) risky sexual behavior among homeless teenagers or those living in shelters^17,21,26; 6) homelessness and mental health.^11,16,27

Settings

The investigations were carried out in different settings: 1) population samples^9,16,17,21; 2) hospital and clinical studies^7,24,27; 3) studies in shelters.^{9,10,18,19,21-23}

Age of subject

With regard to the subjects' ages, seven studies included exclusively teenagers between 7 and 19 years of age,^{7-17,22-24,27} and the rest of the studies had teenagers as a subgroup of a larger sample of adult mothers.^{4,9,16,19-20,26}

Main characteristics of mental health and behavior

Given the heterogeneity of objectives in each investigation, the results were grouped according to the main areas of investigation: substance abuse, mental disorders, lack of social support, sexual behavior, physical and sexual violence, pregnancy, and mother-child interaction.

Substance abuse

Many studies considered substance abuse to be one of the main risk factors for the homeless population, either as a main outcome^7,10 or as a finding parallel to the main objectives of the study.^4-26 The two studies in which drug abuse was the main concern^7,10 placed their results within a broader context. Quinlivan and Evans⁷ correlated a no-drug group with a multidrug group with significantly greater incidence of social isolation, homelessness, and domestic violence. They specifically found that teenagers using marijuana were significantly more likely to be socially isolated, homeless or victims of domestic violence than those in the no-drug group. Zlotnick et al.¹⁰ considered the broad impact of substance use on women's lives and concluded that only use within the last 30 days contributed towards loss of child custody. Mothers making use of drugs for longer periods did not have the outcome anticipated by the authors. Other factors seemed to influence the maintenance of the mother's relationship with the child, such as the time for which they remained homeless: those who did not lose the child remained homeless for shorter times. Homelessness also brought other problems that can be considered as parental neglect, such as lack of schooling for the child. The authors concluded that there was a circular relationship between family stability, poverty, and income: "loss of family stability is associated with loss of child custody (legal guardianship); when child custody is lost, income may be reduced; and income loss further destabilizes the integrity of the family." In both studies, illegal drug use was associated with increased incidence of concurrent alcohol use.

Mental disorders

Mental disorders are a commonly described health problem of out-of-home teenage mothers in these studies. The assessment instruments used to evaluate mental problems were: the Structured Clinical Interview for the DSM-Non-Patient edition (SCID-NP), the Structured Clinical Interview for DSM-III-R, non-patient version (DSM-III-R), the Symptom Checklist-90 (SCL-90), and the Center for Epidemiologic Studies Depression Scale (CES-D ). Some researchers did not explain how their evaluations of mental disorders were done. Pennbridge et al.²⁴ revealed that the homeless sample in their study presented several mental health profiles, especially regarding suicide attempts and depression (30%). Similarly, the subjects were twice as likely to be diagnosed with a major mental disorder. In an investigation of mother-child bonding comparing homeless and low-income mothers, Easterbrooks and Graham⁹ found no association between mental health and housing status. Mental problems were evenly distributed in the two groups of the study. Of the total sample, two thirds (n = 74) of the mothers reported current depressive symptoms severe enough to be categorized within the clinical range. In a study on the prevalence of mental health and substance use disorders among homeless and low-income housed mothers, Weinreb et al.²⁶ found that both groups had high rates of major depression: 45% for the homeless group and 42.8% for the low-income housed group. The groups also presented with posttraumatic stress disorder (PTSD), 36.1% and 34.1%, respectively.

Mental health of mothers and children

Women's mental health is a motif in the studies by Bassuk et al.^4,5,8,11,16 A comparison of homeless and low-income populations showed that a mother's emotional status, in addition to various stressors, strongly predicts a child's negative outcome for both child behavior checklist (CBCL) scales.¹¹

Lack of social support

Almost all studies emphasized the vulnerability of teenagers who are pregnant or mothers, the vulnerability of the homeless population, and need for innovative assistance programs for this population. For example, comparing homeless teenagers with low-income teenagers living with their families,^4,16,26 it was stressed that both homeless and low-income housed families had fragmented support networks, while homeless individuals were significantly more socially isolated than housed individuals. The study by Bassuk et al.⁸ pointed out the economic fragility of the homeless population and its housing instability. Pennbridge et al.²⁴ argued that the homeless pregnant teenagers not only were at higher risk of a wide variety of problems but they also had far fewer social supports than their housed peers. The main finding from a study by Rich²³ was that there were no significant differences between teenagers living in shelters under the care of an assistance program for teenage mothers in New Jersey and those in the general population, thus emphasizing the importance of social support in women's shelters for nourishing a relationship between the young mother and her child. Based on interviews, Bassuk et al.⁴ found that almost two-thirds of their sample lacked or had minimal supportive relationships and a quarter mentioned their child as the major support. This study correlated lack of social support for the family: most of the families were headed by women; most had histories of residential instability; and most received aid for families with dependent children. The search for housing facilities had led the interviewed families to move 6.6 times over the 5 years prior to the first homeless episode. During the preceding 5 years, 85% had doubled up and more than 50% had been in other emergency housing facilities.

Sexual behavior

Confirming the information in the literature,²⁹ the authors of the studies analyzed here^17,18 regarded homeless teenagers as more exposed to high-risk behavior such as drug use, early pregnancy,¹⁹ survival sex,²⁴ and absence of birth control.^6,17 Comparing teenagers living in their own homes with those in foster homes, Polit et al.⁶ concluded that teenagers living at home were more likely to have intercourse and were more sexually active than those in foster homes. A large number of teenagers living in foster care had been abused and whether the teenager had ever had intercourse voluntarily was strongly and positively related to the number of foster care placements. Teenagers in foster care were less likely to have used a contraceptive during their first intercourse and were substantially less likely to have used it in the most recent episode of sexual intercourse (45 vs. 64%). Analyses comparing welfare clients and the national sample have suggested that child welfare teenagers are not only more at risk of premarital intercourse than their peers but also less informed about birth control and human sexuality.

Homelessness and its duration, along with previous pregnancy,²² came out in these studies as risk factors for another pregnancy. Concerns regarding sexual abuse¹⁸ and non-intact families were voiced by these authors as a backdrop. Homeless teenagers were more involved in prostitution or survival sex, presented poorer mental health profiles and more medical problems, reported much higher percentages of experiences of abuse and had had their first sexual experience earlier than their non-homeless peers.²⁴ In addition, in every drug category from intravenous drugs to cigarettes, pregnant homeless young women reported more drug use than non-homeless teenagers. In a study on homeless teenagers in a large Brazilian city, Raffaelli et al.¹⁷ stressed that the responses revealed that teenagers living on the streets were in a risky sexual environment.

Physical and sexual violence

We could not find any study investigating violence and homeless teenage mothers as its main objective. The studies examined covered a wide range of topics and when the authors investigated the terms "abuse" or "maltreatment," the type or severity of abuse experienced by the respondents or the relationships between the perpetrators and victims were unclear. When describing their samples, most studies revealed that low-income and homeless adult women and teenagers suffered physical violence or sexual abuse.^{4,16,18,20,22} Easterbrooks and Graham⁹ stated that homeless women reported more physical or sexual victimization than housed mothers. Comparing groups of teenagers who had ever and never been pregnant, Sheaff and Talashek²² found that rape (sexual assault by an unknown person) was reported by 20% of the group who had ever been pregnant, compared with 7% of those who had never been pregnant. Overall, they found that there had been high levels of violence, with more than 40% (n = 136) reporting physical abuse. Polit et al.⁶ stated that their study confirmed a tendency for the victims of childhood sexual abuse to show greater risk of early sexual behavior, a factor that presumably accounted for at least some of the elevated risk among child welfare clients. Investigating physical and sexual violence in groups of homeless mothers living in shelters and low-income housed mothers, Bassuk et al.⁸ concluded that for the majority of the respondents, physical and sexual abuse and assault had began before adulthood (40%). During adult life, nearly two thirds of the overall sample had suffered severe physical assault by an intimate partner. The studies analyzed confirmed the tendency brought out in the literature concerning the teenage or adult homeless population²⁸: domestic violence was significantly associated with drug use and pregnancy.

Pregnancy

The studies mentioned here^17-19,22 seem to agree that age and duration of homelessness were key factors in predicting pregnancy. Greene and Ringwalt¹⁸ concluded that "older youths were significantly more likely to have been pregnant than younger subjects, blacks were more likely than whites. Young women who had been away from home for longer periods of time were the most likely to report having been pregnant. Street youths were the most likely to have ever been pregnant, followed by shelter youths and then youths in households." Sheaff and Talashek²² added that ever-pregnant teens had: "significantly greater histories of rape and voluntary sexual activity than never-pregnant teens" and concluded: "Teens in this sample reported twice as many pregnancies as do teenagers in the general population. They also have few resources to call on when pregnancy occurs: limited family resources, education and cognitive skills for problem solving." Comparing 704 homeless public assistance families in New York City with 524 housed families, Weitzman¹⁹ stated that 26% of the homeless women interviewed had had babies in the previous year, compared to 6% and 11% of the housed women. Another interesting finding that emerges in these studies is pregnancy as a form of protection for the teenager and for the baby. Teenagers participating in the Capable Adolescent Mothers program were those who decided to raise their babies, who showed readiness to take care of themselves, and who were not currently abusing drugs or alcohol.²³ This association between pregnancy and opting to leave the street and drug abuse was also found by Quinlivan and Evans,⁷ who reported that 40% (181) of the study participants ceased using illegal drugs immediately before or during early pregnancy. It has been suggested that the factors motivating a teenager to have a baby also motivate her to quit illicit drug use.

Mother-child interaction

The results of the study by Bassuk et al.⁸ involving homeless and low-income populations were surprising: both homeless and low-income children experienced significant adversity in their lives, with homeless children facing more stress. In addition to various stressors, a mother's emotional status strongly predicted a child's negative outcome on both CBCL subscales. Both groups had fragmented support networks, with homeless individuals significantly more socially isolated than housed individuals. Weinreb et al.²⁶ examined the determinants of health and service use patterns among 627 homeless and low-income housed children. In addition to the previously reported findings such as the fact that homeless children were significantly younger and more likely to have moved in the past year, mothers of homeless children were more likely to report their children as being in fair or poor health and had a higher frequency of outpatient and emergency room visits compared with their housed counterparts. Easterbrooks and Graham⁹ reported that both homeless and housed infants in families with low economic resources were at risk of insecure attachment behavior. These authors suggested that manifestations of ambivalent insecurity might be an infant's way of coping with inconsistent maternal availability. For infants living in chaotic, crowded or unpredictable environments (including homeless shelters and informal doubling up in homes for mothers) this may be somewhat adaptive. In this sense, Rich²³ found that teenagers could interact adequately with their children in a context that provided continuity. Like Rich²³, Zlotnick et al.²⁰ emphasized that parent-child interaction might be problematic in family shelters where privacy is rare; the impact of this variable on mother-child interaction has not been adequately studied by any of the authors. Zlotnick et al.²⁰ concluded that homelessness threatened family integrity. They emphasized that programs targeting homeless families needed to address these problems in their interventions, as well as considering the challenges posed by the environments of homeless shelters, where parents' actions and conversations were continuously under the scrutiny of other parents, children, and shelter personnel.

DISCUSSION

Since the present study dealt with a current phenomenon, we sought to conduct a systematic review of epidemiological studies covering the topic of homeless mothers or pregnant teenagers. The search yielded 19 articles. Not all of them directly addressed the problem, although they contributed towards our understanding of it by revealing characteristics and facets of this population that, when seen in their contextual complexity, seemed interconnected: substance abuse, mental health, lack of social support, physical and sexual violence, pregnancy, mother-child interaction, and sexual behavior.

These data indicated that many of the young women heading these homeless families now have difficulty establishing themselves as autonomous adults holding jobs; they will remain dependent on agencies over the long term to face their motherhood without appropriate support systems.

Main characteristics of mental health and behavior

There are studies addressing the issue of resiliency, in which the homeless population survives in poverty, with fewer economic resources and social supports than among other population segments. The lack of family support and the violence present in all the samples have been recognized as a substratum of family homelessness. A high number of the women studied had been physically or sexually assaulted during their lifespan. Thus, it is not surprising that PTSD, substance use disorders and major depression were disproportionately high, with lifetime rates of PTSD three times greater than in the general female population.¹⁶

One point that appears in discussions in the published literature is that pregnancy, drug abuse, and other behavioral patterns appear to be strategies for overcoming a chaotic family context. For example, Quinlivan and Evans⁷ argued that "in some cases drug use is an escape mechanism for a teenager who is confronted by an unsupportive and abusive environment." Easterbrooks et al.⁹ pointed out that "for infants living in chaotic, crowded, or unpredictable environments (including homeless shelters and informal doubling up in other's homes), manifestations of ambivalent insecurity may be an infant's way of coping."

Some studies indicate that homeless adolescents living in temporary shelters have been physically or sexually abused, or both.²² Greene and Ringwalt¹⁸ argued that the phenomenon of sexual abuse might be a key factor, along with rape and the economic difficulties faced by the homeless population, in leading these individuals to the use of drugs, survival sex, and other risky behavioral patterns. The connection between violence (rape) and homeless teenage sexual behavior was highlighted in a study by Raffaelli et al.,¹⁷ in which rape was shown to be a coercive means for establishing hierarchy in street gangs, as well as a form of violence, especially towards women. Although Raffaelli's¹⁷ study was part of a project investigating HIV seroprevalence and risk factors that carried out interventions to reduce the risk of HIV infection among children and adolescents living or working on the streets of Belo Horizonte, the fourth largest city in Brazil, the study presents information about sexuality and high-risk behavior among teenagers living on the streets. This study included a pilot study investigating pregnancy and contraception, which is directly related to the present review.

Other studies covering drug use^7,10 were included in the present review in accordance with the same criteria, considering that they did not exclusively treat substance abuse but brought to light important points for understanding homeless pregnant teenagers and teenage mothers. Most studies pointed out the need to gather these mothers in shelters with teams prepared to receive mother and child, together with social policies that could provide support and attenuate this problem. This temporary shelter situation seems to be necessary for a period of time when it contributes towards forming a bond between young mothers and their children. Several authors who have worked with such populations have stressed this concern.^30,31 Rich³² mentioned a girl in her sample, thus: "Rubin holds that a woman may deliver a child but that she does not become a mother in isolation. She needs a social support system to help her achieve the task of motherhood." This author also pointed out these girls' difficulty in accepting staff authority and their penchant for running away from their situations.³² In a study comparing outcomes and cost-effectiveness between comprehensive interdisciplinary teenager-centered prenatal care clinics and "traditional" adult-centered obstetric services, Bensussen-Walls and Saewyc³¹ found that the clients at teenager clinics missed fewer appointments and were more likely to have long-term follow-up and to breastfeed their babies than those at adult clinics. The costs were lower for teenager clinic-based outcomes, while the costs relating to preterm labor were similar.

Methodological appraisal

As mentioned in the Results section, we found that all studies were classified as B with regard to the quality of the methodology used. Furthermore, we observed that there was great heterogeneity in relation to the methods used. The investigations had different objectives, settings, and age groupings. Assessment instruments were not always mentioned nor was it stated how they were administered or which the qualifications of the researchers applying the questionnaires were. In the case of domestic violence, the investigators did not state whether the subject was the victim or the aggressor. Similarly, when considering violence, studies often did not make it clear whether the focus was physical or sexual assault. Furthermore, if this information was in fact present, the study did not state whether the abuse took place during childhood or in adulthood. Thus, although the studies confirm the recognized correlation in the literature between homeless poor populations, violence, and mental disorders, generalization about these issues is difficult. The great methodological heterogeneity between the studies limited the possibilities of quantitative analysis of the results. Because of these findings, investigations within this field would be greatly benefited if the groups interested in this topic could standardize the evaluation instruments used.

Areas in need of further research

This review revealed various areas in need for further research. We could not find any studies considering the effects that a shelter or other homeless context has on parenting and child development and the characteristics of parent-child interactions. Topics such as the stigma associated with being a homeless parenting teenager, public policies, longitudinal studies, pregnancy outcomes, maternity, and child development remain obscure and should be the object of future investigations.

CONCLUSION

Our review found few epidemiological studies focusing on this topic. Moreover, these investigations were performed using different methodological structures, thus preventing better data grouping. The data gathered indicate that pregnant teenagers or teenage mothers living on the streets have high rates of mental and behavioral problems and high exposure to violent situations. In addition, the data highlight the potential risks of this context of living on the streets or in shelters on the organization of the relationship between teenage mother and child.

Study limitations

This review has several limitations. First, studies focusing primarily on related areas such as drug abuse, depression, depression in pregnancy or on other population samples may not have been included in this review. This may have happened if the presence of pregnant teenagers or homeless mothers was not reported in the study methods or results or if a study focused exclusively on a topic without touching on the vicissitudes of teenage pregnancy or homeless mothers. A second concern is publication bias. Studies published in non-indexed journals were not included in this review. The majority of the published studies were conducted in the USA; therefore, cross-cultural comparisons were not addressed. Finally, given the stigmatic characteristics of the problem and the sampling difficulties, data may reveal a partial view of the issue. In short, since this is a growing population, more studies are needed to investigate the complex interaction of life, health, and homelessness among teenage mothers.

REFERENCES

• 1. Whitbeck LB, Hoyt DR, Ackley KA. Families of homeless and runaway adolescents: a comparison of parent/caretaker and adolescent perspectives on parenting, family violence, and adolescent conduct. Child Abuse Negl. 1997;21(6):517-28.
• ²
  National Coalition for the Homeless. Homeless families with children. NCH Fact Sheet 1995 [cited 2009 Feb 12]; Available from: http://www.nationalhomeless.org/publications/facts/families.pdf
• ³
  US Conference of Mayors. Hunger and homelessness survey: a status support on hunger and homelessness in America's cities. Hunger and Homelessness Survey 2004 [cited 2009 Feb 12]; Available from: http://www.usmayors.org/uscm/hungersurvey/2004/onlinereport/HungerAndHomelessnessReport2004.pdf
• 4. Bassuk EL, Rubin L, Lauriat AS. Characteristics of sheltered homeless families. Am J Public Health. 1986;76(9):1097-101.
• 5. Bassuk EL, Buckner JC, Perloff JN, Bassuk SS. Prevalence of mental health and substance use disorders among homeless and low-income housed mothers. American J of Psychiatry 1998;155:1561-4.
• 6. Polit DF, Morton TD, White CM. Sex, contraception and pregnancy among adolescents in foster care. Fam Plann Perspect. 1989;21(5):203-8.
• 7. Quinlivan JA, Evans SF. The impact of continuing illegal drug use on teenage pregnancy outcomes--a prospective cohort study. BJOG. 2002;109(10):1148-53.
• 8. Bassuk EL, Weinreb LF, Buckner JC, Browne A, Salomon A, Bassuk SS. The characteristics and needs of sheltered homeless and low-income housed mothers. JAMA. 1996;276(8):640-6.
• 9. Easterbrooks MA, Graham CA. Security of attachment and parenting: homeless and low-income housed mothers and infants. Am J Orthopsychiatry. 1999;69(3):337-46.
• 10. Zlotnick C, Robertson MJ, Tam T. Substance use and separation of homeless mothers from their children. Addict Behav. 2003;28(8):1373-83.
• 11. Bassuk EL, Weinreb LF, Dawson R, Perloff JN, Buckner JC. Determinants of behavior in homeless and low-income housed preschool children. Pediatrics. 1997;100(1):92-100.
• 12. Gama SGN, Szwarcwald CL, Leal MC. Experiência de gravidez na adolescência. Fatores associados e resultados perinatais entre puérperas de baixa renda. Cad Saúde Pública. 2002,18(1):153-61.
• 13. Ribeiro ERO, Barbieri MA, Bettiol H, Silva AAM. comparação entre dois coortes de mães adolescentes em município do Sudeste do Brasil. Rev Saúde Pública. 200;34(2):136-42.
• 14. Freitas GVS, Botega NJ. Gravidez na adolescência. Prevalência de depressão, ansiedade e ideação suicida. Rev Assoc Med Bras. 2002;48(3):245-9.
• 15. WHO. Organization Mundial de La Salud. Necessidades de la salud de los adolescents. Ginebra: OMS; 1977.
• 16. Bassuk EL, Buckner JC, Perloff JN, Bassuk SS. Prevalence of mental health and substance use disorders among homeless and low-income housed mothers. Am J Psychiatry. 1998;155(11):1561-4.
• 17. Raffaelli M, Campos R, Merritt AP, et al. Sexual practices and attitudes of street youth in Belo Horizonte, Brazil. Street Youth Study Group. Soc Sci Med. 1993;37(5):661-70.
• 18. Greene JM, Ringwalt CL. Pregnancy among three national samples of runaway and homeless youth. J Adolesc Health. 1998;23(6):370-7.
• 19. Weitzman BC. Pregnancy and childbirth: risk factors for homelessness? Fam Plann Perspect. 1989;21(4):175-8.
• 20. Zlotnick C, Robertson MJ, Wright MA. The impact of childhood foster care and other out-of-home placement on homeless women and their children. Child Abuse Negl. 1999;23(11):1057-68.
• 21. Greene JM, Ennett ST, Ringwalt CL. Prevalence and correlates of survival sex among runaway and homeless youth. Am J Public Health. 1999.
• 22. Sheaff L, Talashek M. Ever-pregnant and never-pregnant teens in a temporary housing shelter. J Community Health Nurs. 1995;12(1):33-45.
• 23. Rich OJ. Maternal-infant bonding in homeless adolescents and their infants. Matern Child Nurs J. 1990;19(3):195-210.
• 24. Pennbridge J, Mackenzie RG, Swofford A. Risk profile of homeless pregnant adolescents and youth. J Adolesc Health. 1991;12(7):534-8.
• 25. Carpenter SC, Clyman RB, Davidson AJ, Steiner JF. The association of foster care or kinship care with adolescent sexual behavior and first pregnancy. Pediatrics. 2001;108(3):E46.
• 26. Weinreb L, Goldberg R, Perloff J. Health characteristics and medical service use patterns of sheltered homeless and low-income housed mothers. J Gen Intern Med. 1998;13(6):389-97.
• 27. Felice ME, Shragg GP, James M, Hollingsworth DR. Psychosocial aspects of Mexican-American, white, and black teenage pregnancy. J Adolesc Health Care. 1987;8(4):330-5.
• 28. Goodman L. The prevalence of abuse among homeless and housed poor mothers: a comparison study. Am J Orthopsychiatry. 1991;61(4):489-99.
• 29. Kidd SA, Scrimenti K. Evaluating child and youth homelessness. Eval Rev. 2004;28(4):325-41.
• 30. Garrett SC, Tidwell R. Differences between adolescent mothers and nonmothers: an interview study. Adolescence. 1999;34(133):91-105.
• 31. Bensussen-Walls W, Saewyc EM. Teen-focused care versus adult-focused care for the high-risk pregnant adolescent: an outcomes evaluation. Public Health Nurs. 2001;18(6):424-35.
• 32. Rich OJ. Vulnerability of homeless pregnant and parenting adolescents. J Perinat Neonatal Nurs. 1992;6(3):37-46.

*

N.T. Todas as ocorrências de citações diretas de textos originalmente em inglês foram traduzidas livremente com base no texto fornecido pelos autores do artigo.

Publication Dates

• Publication in this collection
  28 July 2010
• Date of issue
  2010

History

• Received
  15 Mar 2009
• Accepted
  12 Oct 2009