- Email: [email protected]
Childhood Abuse as a Risk Factor for Adolescent Pregnancy in El Salvador
Journal of Adolescent Health 42 (2008) 580 –586
Original article
Childhood Abuse as a Risk Factor for Adolescent Pregnancy in El Salvador Christina C. Pallitto, Ph.D.a,*, and Victoria Murillo, B.A.b a Independent Consultant, Geneva, Switzerland University of Professional Arts, Rotterdam, The Netherlands Manuscript received September 1, 2007; manuscript accepted November 29, 2007 b
Abstract
Purpose: To determine whether having been abused as a child increases the risk of adolescent pregnancy in El Salvador and whether intimate partner violence during adolescence affects the association. Methods: Using data from 3753 women between the ages of 15 and 24 from a nationally representative household health survey of Salvadoran women (FESAL 2002/2003), the association between history of childhood abuse (emotional, physical, or sexual abuse, and witnessing abuse of one’s mother) and adolescent pregnancy was explored using multiple logistic regression analyses. The effect of intimate partner violence during adolescence on the relationship was explored among a subgroup of 15–19-year-olds. Results: The risk of adolescent pregnancy was significantly higher among women abused as children. Women who were sexually abused, physically abused, or who experienced any type of abuse had a 48%, 42%, and 31% higher risk, respectively, of adolescent pregnancy than those without a history of abuse, after adjusting for confounding factors. Intimate partner violence during adolescence was also strongly and significantly linked with adolescent pregnancy risk. Conclusions: This is the first study from a Latin American country to demonstrate a relationship between childhood abuse and adolescent pregnancy. Greater efforts are needed to promote detection of abuse, expand knowledge about sexual and reproductive health, protect vulnerable youth, and to advocate for greater rights and social protections to Salvadoran children and adolescents. © 2008 Society for Adolescent Medicine. All rights reserved.
Keywords:
Adolescent pregnancy; Child abuse; El Salvador; Intimate partner violence
Rates of adolescent pregnancy in El Salvador are alarmingly high, with 40% of women interviewed in the national household health survey (FESAL 2002/2003) reporting that their first birth occurred before they were 20 years old [1]. Childbearing during adolescence is associated with serious consequences for young mothers, their infants, and society in general. Adolescent mothers have a higher risk of pregnancy complications, which can lead to maternal morbidity and mortality, as well as adverse perinatal and infant outcomes [2,3]. They are
*Address correspondence to: Christina C. Pallitto, Ph.D., Chemin des Marais 23A, 1291 Commugny, Vaud, Switzerland. E-mail address: [email protected]
also more likely to drop out of school, to have fewer employment opportunities, to have fewer socioeconomic resources, to be single mothers, to have higher fertility rates, and to have fewer options to remove themselves from poverty [4,5]. Understanding the risk factors for adolescent pregnancy is an important step in preventing its occurrence and the associated consequences. Previous research has explored a variety of risk factors for adolescent pregnancy, including the role of childhood abuse on the risk of adolescent pregnancy, and some researchers have found a significant association, although the evidence is inconclusive [6]. With the exception of a recent study, in which researchers found that the odds of having an adolescent pregnancy increased with
1054-139X/08/$ – see front matter © 2008 Society for Adolescent Medicine. All rights reserved. doi:10.1016/j.jadohealth.2007.11.148
C.C. Pallitto and V. Murillo / Journal of Adolescent Health 42 (2008) 580 –586
greater exposure to adverse childhood experiences, even after adjusting for ethnicity, education, and age [7], most studies have found that the associations between physical and/or sexual abuse and adolescent pregnancy no longer remained significant after controlling for confounding factors [8], or they did not control for confounding factors [9 –11]. Some researchers have studied school-based samples [12], excluding those who dropped out of school and who might have been at higher risk for adolescent pregnancy, whereas other researchers studied pregnant teens without comparing them to a comparison group [13–17]. In addition, most of the research has been conducted in the United States and Western Europe, and little evidence is available from Latin America. Existing theoretical frameworks provide a basis for understanding the mechanisms through which different forms of child abuse can lead to adolescent pregnancy. For example, some researchers have found that adolescents who were sexually abused had more sexual experiences due to their identity being shaped around early sexualization [8,9,14]. Either they sought affection through sexual relationships, felt unable to refuse unwanted sexual advances, [9,14], or developed a premature interest in sexual behavior [18]. These reactions can lead to a greater number of sexual partners and higher risk of adolescent pregnancy [9,14,19 –21]. Similarly, physically and sexually abused girls may wish to escape from an abusive childhood home by creating their own family and intentionally becoming pregnant prematurely [9,14], or they might suffer adverse consequences on their cognitive and emotional development that affect decision making and self-esteem, leading to repeated victimization experiences [1,14,22]. In addition, increased substance use and depression resulting from childhood trauma could adversely affect sexual risk taking [23]. The study described here makes a unique contribution to the understanding of the association between child abuse and adolescent pregnancy by exploring the association among a nationally representative sample of women in El Salvador using the national household health dataset (2002/ 2003 FESAL). Multivariate regression models are used to control for the effect of confounding factors, making this the first study from Latin America, to our knowledge, to do so. Multiple forms of childhood abuse have been operationalized, including emotional abuse, physical abuse, sexual abuse, and witnessing of abuse. Studies exploring all of these forms of childhood abuse are rare in the existing literature [7]. This study also explores the role of partner abuse during adolescence as an intermediate variable in the pathway between childhood abuse and adolescent pregnancy, which is a factor that has not been adequately explored in relation to childhood abuse and adolescent pregnancy.
581
Methodology The analyses presented here are based on the Salvadoran family health survey (FESAL 2002/2003), a national survey that used a multistage, probabilistic, random selection process to interview 10,689 women between 15 and 49 years old in their homes, achieving a 91% response rate [1]. A standard questionnaire was used to ask participants about their sexual and reproductive health, contraceptive use, experiences of family violence, children’s health, children’s education, and knowledge, attitudes, and practices surrounding HIV/AIDS. The study was conducted by the Salvadoran Demographic Association (ADS) with the participation of a committee of national and international institutions. Although the Centers for Disease Control and Prevention (CDC) provided technical oversight, the study was exempted from review by CDC’s institutional review board because it was categorized as public health surveillance. However, all participants gave informed consent verbally and the dataset contained no personal identifiers. This study focuses on a subgroup of 3753 Salvadoran women between the ages of 15 and 24 years old for whom information on age of first pregnancy was available. Prior to conducting the analyses, the subgroup was compared to the group not included in the analysis (women 25 to 49 years old) on selected sociodemographic and fertility-related characteristics to show how representative the subgroup was of the entire population of Salvadoran women interviewed. A woman was considered to have had an adolescent pregnancy if she answered positively that she had been pregnant or found out she was pregnant at the age of 19 or earlier. Several forms of child abuse were studied—sexual, physical, emotional, or witnessing of abuse. The sexual abuse variable includes sexual abuse both with and without penetration before the age of 18 years, before any pregnancy, and by someone other than her spouse, partner, boyfriend, or ex-boyfriend (partner abuse will be discussed shortly). Physical abuse was classified as physical punishment before the age of 18 years old, including having been hit with a belt, stick, or rope; burned or injured in some part of the body; and/or tied up or locked up. Emotional abuse was classified as punishment before the age of 18 years old, including having been locked up, left alone, not allowed to leave or interact with others, threatened, yelled at, humiliated, or ridiculed in front of others. A woman was classified as having witnessed abuse if she had witnessed a woman being abused by a man in the home where she lived before the age of 18 years old. Socioeconomic status and education level were also controlled for in the multivariate analyses based on number of household goods and services, and whether a woman had at least some secondary school education or not. To further investigate the relationship between childhood abuse and adolescent pregnancy, and a possible mechanism through which the factors could be related, a subanalysis
582
C.C. Pallitto and V. Murillo / Journal of Adolescent Health 42 (2008) 580 –586
Table 1 Comparison of 15–24-year-olds with those 25 and older on selected characteristics
Education*** None Primary Secondary Superior SESa* Lives in rural area* Sexually abused as child Physically abused as child*** Emotionally abused as child* Witnessed abuse as child*** Any abuse as child*** Current contraceptive use*** Ever contraceptive use*** Knows where to get contraceptives***
Subgroup (15–24 years old)
Others (⬎24 years old)
0.45% 66.3% 25.4% 7.9% 3.7 items 48.8% 7.1% 30.7% 8.7% 15.8% 45.2% 26.5% 40.3%
1.1% 69.6% 17.3% 12.0% 3.8 items 46.1% 7.0% 40.3% 7.2% 21.1% 50.3% 59.1% 77.0%
72.3%
88.6%
Note: Differences significant at * p ⬍ .05, *** p ⬍ .001. a Mean number of household goods and services out of a possible seven items.
was conducted to explore the role of intimate partner violence. Because women were asked retrospectively about partner violence without specifying when it had occurred, the subanalysis was limited to women currently in their adolescent years to ensure that reports of violence had occurred during adolescence. First, the prevalence of each type of partner abuse was calculated among the subgroup. Next, the association between partner abuse during adolescence and adolescent pregnancy as well as the association between childhood abuse and partner abuse during adolescence was explored using chi-squared tests. Finally, the association between partner abuse and adolescent pregnancy was explored while stratifying by child abuse status and controlling for confounding factors. Partner abuse questions were administered to women who currently or previously had had a boyfriend, spouse, or partner. Those women who responded affirmatively that their current or past partner had ever hit her with his hands or feet, had injured her or fractured a bone, had burned some part of her body, or had threatened her with death, prior to the age of 18 and prior to her pregnancy, were classified as physically abused. Emotional abuse included threats to remove economic support, to throw her out of the house, to take away the children, or to leave or abandon her; as well as destruction of furniture, household items or her personal items; shouting at her, insulting her or humiliating her in front of others; as well as prohibiting her from having friendships, interacting with her family, or dressing as she pleased. Finally, a woman was considered sexually abused if a current or past partner had forced her to have sexual relations. Women who had reported any of these types of abuse were considered to have experienced adolescent part-
ner abuse, as long as the abuse had occurred prior to any pregnancy and before 18 years old. Results The main analysis was limited to 3753 women between the ages of 15 and 24, because certain modules of the questionnaire were administered only to that subgroup. Table 1 shows the mean values for some of the key descriptive variables and the comparison of the group of 15- to 24-yearolds with those over 24 years old. The majority of the subgroup of 15- to 24-year-olds (66.3%) had a primary school education, and one-third completed secondary or above. However, many of them were still studying at the time of the survey because of their age. Women in the subgroup had slightly but significantly lower socioeconomic status than those in the older group. Almost half lived in rural areas, and significantly more women in the subgroup lived in rural areas compared to women over the age of 24. The groups differed on child abuse history as well. More women in the group of women over 24 years old had experienced some type of abuse as a child compared to the women between the ages of 15 to 24 (50.3% vs. 45.2%). Although rates of sexual abuse were similar in both groups, and the younger women had experienced slightly more emotional abuse, significantly more older women had experienced physical abuse or had witnessed abuse as a child. In addition, women in the younger subgroup had significantly lower rates of current and ever use of contraceptives, with 27% of the younger subgroup currently using as opposed to 59% of the other group, and 40% versus 77% ever using contraceptives, respectively. Some of the difference can clearly be explained by differences in sexual behavior and fertility desires of the two groups because of age. It is important to note that the differences between the two groups can affect the generalizability of the findings to the population of Salvadoran women and that lower overall rates of childhood abuse among the study population likely underestimate the actual association between childhood abuse and adolescent pregnancy. Table 2 shows how many women experienced each type of abuse, regardless of her experience of other forms of abuse. The most common form of childhood abuse was physical abuse, with 31% of respondents reporting a history
Table 2 Women experiencing each type of childhood abuse*
Sexual abuse Physical abuse Emotional abuse Witnessing of abuse ANY abuse
Number
Percentage
267 1153 328 594 1698
7.1% 30.7% 8.7% 15.8% 45%
* Rates of each type of abuse regardless of other abuse experiences.
C.C. Pallitto and V. Murillo / Journal of Adolescent Health 42 (2008) 580 –586
of physical abuse as a child. Sixteen percent witnessed abuse of their mothers during their childhood, 7% were sexually abused, and 9% were emotionally abused. Many respondents experienced multiple forms of abuse, and almost half (45%) experienced at least one of these forms of abuse. The direct relationship between childhood abuse and adolescent pregnancy, prior to adjustment for confounding factors, is seen in Figure 1. The more types of abuse experienced, the greater the risk of adolescent pregnancy, although the effect of two or three types of childhood abuse on adolescent pregnancy risk is similar. Five separate multiple logistic regression models were created, each of them testing the effect of one form of childhood abuse in relation to the dependent variable of adolescent pregnancy, while controlling for confounding factors (Table 3). Model 5 tests the relationship between having experienced any type of childhood abuse and adolescent pregnancy. Having experienced sexual abuse as a child increased a woman’s risk of adolescent pregnancy by 48%, after controlling for confounding factors such as age, education, socioeconomic status, and living in a rural area. Similarly, physical abuse increased a woman’s risk of adolescent pregnancy by 42%, and having experienced any type of abuse increased a woman’s risk by 31%, after controlling for these same confounding factors. The association between emotional abuse and adolescent pregnancy did not increase a woman’s risk of adolescent pregnancy, and having witnessed abuse as a child increased a woman’s risk but not significantly. Next, a subanalysis was conducted to explore the effect of intimate partner abuse during adolescence on the association between child abuse and adolescent pregnancy among a subgroup of adolescent women who were between 15 and 19 years old at the time of the survey. Of these 1839 women, only 6% reported any type of abuse by their part80 70
Percentage
60 50 40
583
Table 3 Adjusted odds ratios of adolescent pregnancy risk for each type of child abuse Model 1 Model 2 Model 3 Model 4 Model 5 Sexual Physical Emotional Witnessed Any abuse abuse abuse abuse abuse Sexual Physical Emotional Witness Any abuse
1.48** 1.42*** .90 1.14 1.31***
Note: All models control for education, age, socioeconomic status, and living in rural area. ** p ⬍ .005, *** p ⬍ .001.
ners, but having been abused by a partner significantly increased an adolescent woman’s chance of getting pregnancy during her adolescence. Of all adolescents who had a pregnancy, 80% had been abused by a partner physically, sexually, or emotionally. Likewise, the odds of adolescent pregnancy were 11 times higher for abused versus nonabused adolescents (odds ratio: 11.1, 95% confidence interval 7.0, 17.7), prior to adjusting. A strong relationship was also found between child abuse and partner abuse during adolescence. Experience of child abuse increased a woman’s risk of partner abuse by more than four times (odds ratio of 4.3, 95% confidence interval of 2.8, 6.7), prior to adjusting for confounding factors. The effect of partner abuse on adolescent pregnancy risk was analyzed in stratified analyses among this subgroup, while controlling for socioeconomic status, education, and locality (Table 4). Partner abuse increased the risk of adolescent pregnancy by almost 11 times among all adolescents, as seen in Model 1. Among adolescents abused as children, the effect was similar and significant (Model 2), while the effect was greatest among adolescents who were not abused as children (Model 3). These models show the importance of the partner abuse variable in the association and the fact that partner abuse is an even greater risk factor than childhood abuse on adolescent pregnancy risk. Although the analysis is limited to women in their adolescence, it is suspected that the relationship would be similar among women of older
30 20 10 0 0
1
2
3
4
Number of forms of child abuse
Figure 1. Percentage of women with adolescent pregnancy by number of forms of childhood abuse. Note: 2054 women experienced no abuse; 1177 women experiences one form of abuse, 410 women experienced two forms, 99 women experienced three forms, and 12 women experienced four forms.
Table 4 Odds ratios of adolescent pregnancy by intimate partner violence experience during adolescence (stratified by child abuse history) Model 1 All adolescents
Model 2 Adolescents abused as children
Model 3 Adolescents not abused as children
10.83 (6.43–18.24)
10.63 (5.80–19.49)
12.19 (4.18–35.56)
Note: All models control for age, education, socioeconomic status, living in rural area. Model 1 controls for child abuse.
584
C.C. Pallitto and V. Murillo / Journal of Adolescent Health 42 (2008) 580 –586
age groups as well, if it had been possible to measure adolescent partner abuse among women who were no longer adolescents. Discussion The study found strong support for the hypothesis that abuse during childhood is associated with an increased risk of adolescent pregnancy among a large sample of young women in the general population of El Salvador. Although previous research from the United States and other developed countries has also shown that childhood abuse increased the risk of adolescent pregnancy even after controlling for confounding factors [24,25], such an association has not been previously identified in a Latin American setting. Presumably, these findings would be relevant in other Latin American countries that also have high rates of adolescent pregnancy and similar social and cultural norms. Countries such as Honduras, Guatemala, and Nicaragua have adolescent pregnancy rates of 93, 107, and 113 births per 1000 women aged 15 to 19, respectively [26]. These rates surpass El Salvador’s rate of 81, which is considerably higher than the Latin American average of 71 and the average for developed countries worldwide of 25 births per 1000 women aged 15 to 19 [26]. The findings of this study, although informative to neighboring countries, are not generalizable to these other settings, however, and the association should be analyzed there as well. Unlike other studies with more limited definitions of child abuse, this study considered multiple types of child abuse experiences and found that each additional type of abuse experienced increased one’s risk of adolescent pregnancy. Physical and sexual abuse increased the odds of adolescent pregnancy significantly. Having been a witness to childhood abuse increased the odds of adolescent pregnancy but not significantly, whereas emotional abuse had a paradoxical protective effect on adolescent pregnancy status, which could be because of the inclusion of less abusive forms of discipline as emotional abuse or the fact that the gravity of emotional abuse alone was not sufficient to increase one’s risk of adolescent pregnancy. However, having experienced at least one of the multiple forms of child abuse measured, increased the risk of adolescent pregnancy later in life. This study has demonstrated the importance of considering the role of intimate partner violence on adolescent pregnancy risk. In the few studies to include this variable, some researchers found high rates of partner abuse among pregnant adolescents [27] or greater risk of pregnancy among adolescents experiencing dating violence [28], whereas others found that childhood sexual abuse was a risk factor for abuse during adolescence and adulthood [29]. The connections between these three factors— child abuse, partner abuse during adolescence, and adolescent pregnancy—and the intergenerational patterns of adolescent childbearing [30] and abuse [13] show how the cycles of abuse and adolescent childbearing make it
difficult for the next generation of children born to adolescent mothers to break the cycle. To interpret the findings, it is necessary to consider the status of children’s rights and sexual and reproductive health in El Salvador. El Salvador ratified the Convention on the Rights of the Child, an international treaty that promotes the human rights of children, in 1990. Despite the ratification of the treaty, Salvadoran efforts to promote human rights of children and adolescents have been inadequate, partially because of the lack of human and material resources dedicated to responding to abuse, as well as the lack of coordination in programmatic and policy efforts among public and private agencies working on children’s rights. In addition, many girls, adolescents, and adult women fear that if they report abuse or seek help they will face discrimination or social stigma. These fears are well grounded in a setting in which girls and adolescent women, especially those from impoverished communities in El Salvador, may be more vulnerable to discrimination and victimization because they are young, female, and of a low socioeconomic status. Changing deep-rooted beliefs that foster discrimination and abuse requires a focused multisectoral effort not only at the societal level but also at the family and community levels. In the absence of such changes, individual girls and adolescents will continue to face short- and long-term risks of abuse, low self-esteem, revictimization, and adverse results such as adolescent pregnancy. Issues surrounding education on sexual and reproductive health, including pregnancy prevention, have proven extremely controversial in the country, despite El Salvador’s high adolescent pregnancy rate. Although advocates have promoted school-based and nonschool-based education efforts on sexual and reproductive health in the country, strong resistance among conservative religious and political factions have prevented a comprehensive school-based sexual and reproductive health curriculum from being carried out. As a result, many adolescents find themselves in sexual situations in which they are unprepared and without contraceptives. Rather than feeling capable of negotiating sexual decision making and protecting themselves from pregnancy, HIV/AIDS, and sexually transmitted infections, these sexual encounters put them at risk for pregnancy and other health risks. As long as reproductive health remains a taboo topic and adolescents fail to receive appropriate information and services, adolescents will be unprepared to prevent premature pregnancies, as evidenced by the epidemic proportions at which adolescent pregnancy is occurring in the country. Despite these important findings, a few limitations in the analysis must be recognized. The sample was limited to a subgroup of women between the ages of 15 to 24 years old, because only this age group was asked about age of first pregnancy. The subgroup differed significantly from the older women interviewed in terms of child abuse history, contraceptive use, education level, and where they lived. In addition, the cross-sectional nature of the study limits the conclusions about
C.C. Pallitto and V. Murillo / Journal of Adolescent Health 42 (2008) 580 –586
causality, although we attempted to limit errors of reverse causality in a few ways— by only categorizing adolescents as having been abused if the abuse preceded the pregnancy and by categorizing sexual abuse by a family member as having occurred prior to age 18 (a similar option was not possible for the physical and emotional abuse during childhood variables). Likewise, with the partner abuse and pregnancy association, the order in which these events occurred is not known. Although we hypothesize that the abuse preceded the pregnancy, it is also plausible that pregnancy preceded abuse by intimate partners. In addition, because the study deals with sensitive issues of childhood abuse, partner abuse, and adolescent sexual behavior, it is possible that there was underreporting for these questions, which could result in underestimates of the association studied here. Conclusions Clearly, the time has come to break the cycle of abuse and adolescent pregnancy in El Salvador and other countries in Latin America. Efforts to reduce child abuse and adolescent pregnancy should include the following: awareness raising about the rights of children and adolescents to live without violence from their families, partners, or communities; education on existing mechanisms for reporting abuse; expansion of the response of the judicial and social service sectors in cases of suspected and confirmed abuse; training of medical personnel to detect and report abuse more proactively, to respect the integrity of children and adolescents, and to provide services in a nonjudgmental manner; work with young women to promote self-esteem, learn about sexual and reproductive health, and encourage more healthy negotiation regarding sexual decision making; promote male involvement in pregnancy prevention, responsible parenting, gender equality, and conflict resolution among adolescent and adult men; provision of ageappropriate education about sexual and reproductive health; prevention of school desertion among pregnant and nonpregnant teens, and promotion of mechanisms of support for adolescent mothers. All efforts must be coordinated between governmental institutions and civil society at both the national and community levels, involving policy changes as well as community-based efforts for reaching a broader base of vulnerable youth and their families more directly before the consequences to Salvadoran society are even greater. Acknowledgments Funding for the study was provided by the United States Agency for International Development, through the POLICY II Project, implemented by Futures Group. The authors were independent researchers living in El Salvador at the time they were awarded funds for this study. The authors wish to thank Verónica Simán de Betancourt of the POLICY II Project and José Mario Cáceres of the Asociación Demográ-
585
fica Salvadoreña for their insights and feedback on the project documents. References [1] Encuesta Nacional de Salud Familiar—FESAL 2002/3. Informe Final 2004. San Salvador, El Salvador: Asociación Demográfica Salvadoreña. [2] Mayor S. Pregnancy and childbirth are leading causes of death in teenage girls in developing countries. BMJ 2004;328:1152. [3] Alan Guttmacher Institute. Risks and Realities of Early Childbearing Worldwide. [Issues in Brief]. New York: The Institute, 1996. [4] Moore KA, Myers DR, Morrison C, et al. Age at first childbirth and later poverty. J Res Adolesc 1993;3(4):393– 422. [5] Grogger J, Bronars S. The socioeconomic consequences of teenage childbearing: findings from a natural experiment. Fam Plann Perspect 1993;25(4):156 – 61. [6] Blinn-Pike L, Berger T, Dixon D, et al. Is there a causal link between maltreatment and adolescent pregnancy? A literature review. Perspect Sex Reprod Health 2002;34(2):68 –72. [7] Hillis SD, Anda RF, Dube SR, et al. The association between adverse childhood experiences and adolescent pregnancy, long-term psychosocial consequences, and fetal death. Pediatrics 2004;113:320 –7. [8] Stock JL, Bell MA, Boyer DK, et al. Adolescent pregnancy and sexual risk-taking among sexually abused girls. Fam Plann Perspect 1997;29(5):200 –203, 277. [9] Rainey DY, Stevens-Simon, Kaplan DW. Are adolescents who report prior sexual abuse at higher risk for pregnancy? Child Abuse Neglect 1995;19(10):1283– 8. [10] Guijarro S, Naranjo J, Padilla M, et al. Family risk factors associated with adolescent pregnancy: study of a group of adolescent girls and their families in Ecuador. J Adolesc Health 1999;25(2):166 –72. [11] Kenney JW, Reinholtz C, Angelini PJ. Ethnic differences in childhood and adolescent sexual abuse and teenage pregnancy. J Adolesc Health 1997;21(1):3–10. [12] Raj A, Silverman JG, Amaro H. The relationship between sexual abuse and sexual risk among high school students: findings from the 1997 Massachusetts youth risk behavior survey. Matern Child Health J 2000;4(2):125–34. [13] Boyer D, Fine D. Sexual abuse as a factor in adolescent pregnancy and child maltreatment. Fam Plann Perspect 1992;24(1):4 –11, 19. [14] Butler JR, Burton L. Rethinking teenage childbearing: is sexual abuse a missing link? Fam Relat 1990;39(1):73– 80. [15] Esperat MC, Esparza DV. Minority adolescent mothers who reported childhood sexual abuse and those who did not: perceptions of themselves and their relationships. Issues Mental Health Nurs 1997;18(3):229 – 46. [16] Fiscella K, Kitzman HJ, Cole RE, Sidora KJ, Olds D. Does child abuse predict adolescent pregnancy? Pediatrics 1998;101(4):620 – 4. [17] Kellogg ND, Hoffman TJ, Taylor ER. Early sexual experiences among pregnant and parenting adolescents. Adolescence 1999;34(134): 293–303. [18] Noll JG, Trickett PK, Putnam FW. A prospective investigation of the impact of childhood sexual abuse on the development of sexuality. J Consult Clin Psychol 2003;71(3):575– 86. [19] Luster T, Small SA. Sexual abuse history and number of sex partners among female adolescents. Fam Plann Perspect 1997;29(5):204 –11. [20] Cinq-Mars C, Wright J, Cyr M, McDuff P. Sexual at-risk behaviors of sexually abused adolescent girls. J Child Sex Abuse 2003;12(2):1–18. [21] Steel JL, Herlitz CA. The association between childhood and adolescent sexual abuse and proxies for sexual risk behavior: a random sample of the general population of Sweden. Child Abuse Neglect 2005;29(10):1141–53. [22] DeBellis MD. Developmental traumatology: the psychobiological development of maltreated children and its implications for research, treatment and policy. Dev Psychopathol 2001;13(3):539 – 64.
586
C.C. Pallitto and V. Murillo / Journal of Adolescent Health 42 (2008) 580 –586
[23] Miller M. A model to explain the relationship between sexual abuse and HIV risk among women. AIDS Care 1999;11:3–20. [24] Saewyc EM, Magee LL, Pettingell SE. Teenage pregnancy and associated risk behaviors among sexually abused adolescents. Perspect Sex Reprod Health 2004;36(3):98 –105. [25] Raj A, Silverman JG, Amaro H. The relationship between sexual abuse and sexual risk among high school students: findings from the 1997 Massachusetts youth risk behavior survey. Matern Child Health J 2000;4(2):125–34. [26] United Nations Population Fund (UNFPA). State of the World Population 2007: Unleashing the Potential of Urban Growth. New York: UNFPA, 2007.
[27] Parker B, McFarlane J, Soeken K. Abuse during pregnancy: effects on maternal complications and birth weight in adult and teenage women. Obstet Gynecol 1994;84(3):323– 8. [28] Silverman JG, Raj A, Clements K. Dating violence and associated sexual risk and pregnancy among adolescent girls in the United States. Pediatrics 2004;114(2):e220 –5. [29] Fergusson DM, Horwood LJ, Lynskey MT. Childhood sexual abuse, adolescent sexual behaviors and sexual revictimization. Child Abuse Neglect 1997;21(8):789 – 803. [30] Bonnell C, Allen E, Strange V, et al. Influence of family type and parenting behaviours on teenage sexual behaviour and conceptions. J Epidemiol Community Health 2006;60(6):502– 6.
Original article
Childhood Abuse as a Risk Factor for Adolescent Pregnancy in El Salvador Christina C. Pallitto, Ph.D.a,*, and Victoria Murillo, B.A.b a Independent Consultant, Geneva, Switzerland University of Professional Arts, Rotterdam, The Netherlands Manuscript received September 1, 2007; manuscript accepted November 29, 2007 b
Abstract
Purpose: To determine whether having been abused as a child increases the risk of adolescent pregnancy in El Salvador and whether intimate partner violence during adolescence affects the association. Methods: Using data from 3753 women between the ages of 15 and 24 from a nationally representative household health survey of Salvadoran women (FESAL 2002/2003), the association between history of childhood abuse (emotional, physical, or sexual abuse, and witnessing abuse of one’s mother) and adolescent pregnancy was explored using multiple logistic regression analyses. The effect of intimate partner violence during adolescence on the relationship was explored among a subgroup of 15–19-year-olds. Results: The risk of adolescent pregnancy was significantly higher among women abused as children. Women who were sexually abused, physically abused, or who experienced any type of abuse had a 48%, 42%, and 31% higher risk, respectively, of adolescent pregnancy than those without a history of abuse, after adjusting for confounding factors. Intimate partner violence during adolescence was also strongly and significantly linked with adolescent pregnancy risk. Conclusions: This is the first study from a Latin American country to demonstrate a relationship between childhood abuse and adolescent pregnancy. Greater efforts are needed to promote detection of abuse, expand knowledge about sexual and reproductive health, protect vulnerable youth, and to advocate for greater rights and social protections to Salvadoran children and adolescents. © 2008 Society for Adolescent Medicine. All rights reserved.
Keywords:
Adolescent pregnancy; Child abuse; El Salvador; Intimate partner violence
Rates of adolescent pregnancy in El Salvador are alarmingly high, with 40% of women interviewed in the national household health survey (FESAL 2002/2003) reporting that their first birth occurred before they were 20 years old [1]. Childbearing during adolescence is associated with serious consequences for young mothers, their infants, and society in general. Adolescent mothers have a higher risk of pregnancy complications, which can lead to maternal morbidity and mortality, as well as adverse perinatal and infant outcomes [2,3]. They are
*Address correspondence to: Christina C. Pallitto, Ph.D., Chemin des Marais 23A, 1291 Commugny, Vaud, Switzerland. E-mail address: [email protected]
also more likely to drop out of school, to have fewer employment opportunities, to have fewer socioeconomic resources, to be single mothers, to have higher fertility rates, and to have fewer options to remove themselves from poverty [4,5]. Understanding the risk factors for adolescent pregnancy is an important step in preventing its occurrence and the associated consequences. Previous research has explored a variety of risk factors for adolescent pregnancy, including the role of childhood abuse on the risk of adolescent pregnancy, and some researchers have found a significant association, although the evidence is inconclusive [6]. With the exception of a recent study, in which researchers found that the odds of having an adolescent pregnancy increased with
1054-139X/08/$ – see front matter © 2008 Society for Adolescent Medicine. All rights reserved. doi:10.1016/j.jadohealth.2007.11.148
C.C. Pallitto and V. Murillo / Journal of Adolescent Health 42 (2008) 580 –586
greater exposure to adverse childhood experiences, even after adjusting for ethnicity, education, and age [7], most studies have found that the associations between physical and/or sexual abuse and adolescent pregnancy no longer remained significant after controlling for confounding factors [8], or they did not control for confounding factors [9 –11]. Some researchers have studied school-based samples [12], excluding those who dropped out of school and who might have been at higher risk for adolescent pregnancy, whereas other researchers studied pregnant teens without comparing them to a comparison group [13–17]. In addition, most of the research has been conducted in the United States and Western Europe, and little evidence is available from Latin America. Existing theoretical frameworks provide a basis for understanding the mechanisms through which different forms of child abuse can lead to adolescent pregnancy. For example, some researchers have found that adolescents who were sexually abused had more sexual experiences due to their identity being shaped around early sexualization [8,9,14]. Either they sought affection through sexual relationships, felt unable to refuse unwanted sexual advances, [9,14], or developed a premature interest in sexual behavior [18]. These reactions can lead to a greater number of sexual partners and higher risk of adolescent pregnancy [9,14,19 –21]. Similarly, physically and sexually abused girls may wish to escape from an abusive childhood home by creating their own family and intentionally becoming pregnant prematurely [9,14], or they might suffer adverse consequences on their cognitive and emotional development that affect decision making and self-esteem, leading to repeated victimization experiences [1,14,22]. In addition, increased substance use and depression resulting from childhood trauma could adversely affect sexual risk taking [23]. The study described here makes a unique contribution to the understanding of the association between child abuse and adolescent pregnancy by exploring the association among a nationally representative sample of women in El Salvador using the national household health dataset (2002/ 2003 FESAL). Multivariate regression models are used to control for the effect of confounding factors, making this the first study from Latin America, to our knowledge, to do so. Multiple forms of childhood abuse have been operationalized, including emotional abuse, physical abuse, sexual abuse, and witnessing of abuse. Studies exploring all of these forms of childhood abuse are rare in the existing literature [7]. This study also explores the role of partner abuse during adolescence as an intermediate variable in the pathway between childhood abuse and adolescent pregnancy, which is a factor that has not been adequately explored in relation to childhood abuse and adolescent pregnancy.
581
Methodology The analyses presented here are based on the Salvadoran family health survey (FESAL 2002/2003), a national survey that used a multistage, probabilistic, random selection process to interview 10,689 women between 15 and 49 years old in their homes, achieving a 91% response rate [1]. A standard questionnaire was used to ask participants about their sexual and reproductive health, contraceptive use, experiences of family violence, children’s health, children’s education, and knowledge, attitudes, and practices surrounding HIV/AIDS. The study was conducted by the Salvadoran Demographic Association (ADS) with the participation of a committee of national and international institutions. Although the Centers for Disease Control and Prevention (CDC) provided technical oversight, the study was exempted from review by CDC’s institutional review board because it was categorized as public health surveillance. However, all participants gave informed consent verbally and the dataset contained no personal identifiers. This study focuses on a subgroup of 3753 Salvadoran women between the ages of 15 and 24 years old for whom information on age of first pregnancy was available. Prior to conducting the analyses, the subgroup was compared to the group not included in the analysis (women 25 to 49 years old) on selected sociodemographic and fertility-related characteristics to show how representative the subgroup was of the entire population of Salvadoran women interviewed. A woman was considered to have had an adolescent pregnancy if she answered positively that she had been pregnant or found out she was pregnant at the age of 19 or earlier. Several forms of child abuse were studied—sexual, physical, emotional, or witnessing of abuse. The sexual abuse variable includes sexual abuse both with and without penetration before the age of 18 years, before any pregnancy, and by someone other than her spouse, partner, boyfriend, or ex-boyfriend (partner abuse will be discussed shortly). Physical abuse was classified as physical punishment before the age of 18 years old, including having been hit with a belt, stick, or rope; burned or injured in some part of the body; and/or tied up or locked up. Emotional abuse was classified as punishment before the age of 18 years old, including having been locked up, left alone, not allowed to leave or interact with others, threatened, yelled at, humiliated, or ridiculed in front of others. A woman was classified as having witnessed abuse if she had witnessed a woman being abused by a man in the home where she lived before the age of 18 years old. Socioeconomic status and education level were also controlled for in the multivariate analyses based on number of household goods and services, and whether a woman had at least some secondary school education or not. To further investigate the relationship between childhood abuse and adolescent pregnancy, and a possible mechanism through which the factors could be related, a subanalysis
582
C.C. Pallitto and V. Murillo / Journal of Adolescent Health 42 (2008) 580 –586
Table 1 Comparison of 15–24-year-olds with those 25 and older on selected characteristics
Education*** None Primary Secondary Superior SESa* Lives in rural area* Sexually abused as child Physically abused as child*** Emotionally abused as child* Witnessed abuse as child*** Any abuse as child*** Current contraceptive use*** Ever contraceptive use*** Knows where to get contraceptives***
Subgroup (15–24 years old)
Others (⬎24 years old)
0.45% 66.3% 25.4% 7.9% 3.7 items 48.8% 7.1% 30.7% 8.7% 15.8% 45.2% 26.5% 40.3%
1.1% 69.6% 17.3% 12.0% 3.8 items 46.1% 7.0% 40.3% 7.2% 21.1% 50.3% 59.1% 77.0%
72.3%
88.6%
Note: Differences significant at * p ⬍ .05, *** p ⬍ .001. a Mean number of household goods and services out of a possible seven items.
was conducted to explore the role of intimate partner violence. Because women were asked retrospectively about partner violence without specifying when it had occurred, the subanalysis was limited to women currently in their adolescent years to ensure that reports of violence had occurred during adolescence. First, the prevalence of each type of partner abuse was calculated among the subgroup. Next, the association between partner abuse during adolescence and adolescent pregnancy as well as the association between childhood abuse and partner abuse during adolescence was explored using chi-squared tests. Finally, the association between partner abuse and adolescent pregnancy was explored while stratifying by child abuse status and controlling for confounding factors. Partner abuse questions were administered to women who currently or previously had had a boyfriend, spouse, or partner. Those women who responded affirmatively that their current or past partner had ever hit her with his hands or feet, had injured her or fractured a bone, had burned some part of her body, or had threatened her with death, prior to the age of 18 and prior to her pregnancy, were classified as physically abused. Emotional abuse included threats to remove economic support, to throw her out of the house, to take away the children, or to leave or abandon her; as well as destruction of furniture, household items or her personal items; shouting at her, insulting her or humiliating her in front of others; as well as prohibiting her from having friendships, interacting with her family, or dressing as she pleased. Finally, a woman was considered sexually abused if a current or past partner had forced her to have sexual relations. Women who had reported any of these types of abuse were considered to have experienced adolescent part-
ner abuse, as long as the abuse had occurred prior to any pregnancy and before 18 years old. Results The main analysis was limited to 3753 women between the ages of 15 and 24, because certain modules of the questionnaire were administered only to that subgroup. Table 1 shows the mean values for some of the key descriptive variables and the comparison of the group of 15- to 24-yearolds with those over 24 years old. The majority of the subgroup of 15- to 24-year-olds (66.3%) had a primary school education, and one-third completed secondary or above. However, many of them were still studying at the time of the survey because of their age. Women in the subgroup had slightly but significantly lower socioeconomic status than those in the older group. Almost half lived in rural areas, and significantly more women in the subgroup lived in rural areas compared to women over the age of 24. The groups differed on child abuse history as well. More women in the group of women over 24 years old had experienced some type of abuse as a child compared to the women between the ages of 15 to 24 (50.3% vs. 45.2%). Although rates of sexual abuse were similar in both groups, and the younger women had experienced slightly more emotional abuse, significantly more older women had experienced physical abuse or had witnessed abuse as a child. In addition, women in the younger subgroup had significantly lower rates of current and ever use of contraceptives, with 27% of the younger subgroup currently using as opposed to 59% of the other group, and 40% versus 77% ever using contraceptives, respectively. Some of the difference can clearly be explained by differences in sexual behavior and fertility desires of the two groups because of age. It is important to note that the differences between the two groups can affect the generalizability of the findings to the population of Salvadoran women and that lower overall rates of childhood abuse among the study population likely underestimate the actual association between childhood abuse and adolescent pregnancy. Table 2 shows how many women experienced each type of abuse, regardless of her experience of other forms of abuse. The most common form of childhood abuse was physical abuse, with 31% of respondents reporting a history
Table 2 Women experiencing each type of childhood abuse*
Sexual abuse Physical abuse Emotional abuse Witnessing of abuse ANY abuse
Number
Percentage
267 1153 328 594 1698
7.1% 30.7% 8.7% 15.8% 45%
* Rates of each type of abuse regardless of other abuse experiences.
C.C. Pallitto and V. Murillo / Journal of Adolescent Health 42 (2008) 580 –586
of physical abuse as a child. Sixteen percent witnessed abuse of their mothers during their childhood, 7% were sexually abused, and 9% were emotionally abused. Many respondents experienced multiple forms of abuse, and almost half (45%) experienced at least one of these forms of abuse. The direct relationship between childhood abuse and adolescent pregnancy, prior to adjustment for confounding factors, is seen in Figure 1. The more types of abuse experienced, the greater the risk of adolescent pregnancy, although the effect of two or three types of childhood abuse on adolescent pregnancy risk is similar. Five separate multiple logistic regression models were created, each of them testing the effect of one form of childhood abuse in relation to the dependent variable of adolescent pregnancy, while controlling for confounding factors (Table 3). Model 5 tests the relationship between having experienced any type of childhood abuse and adolescent pregnancy. Having experienced sexual abuse as a child increased a woman’s risk of adolescent pregnancy by 48%, after controlling for confounding factors such as age, education, socioeconomic status, and living in a rural area. Similarly, physical abuse increased a woman’s risk of adolescent pregnancy by 42%, and having experienced any type of abuse increased a woman’s risk by 31%, after controlling for these same confounding factors. The association between emotional abuse and adolescent pregnancy did not increase a woman’s risk of adolescent pregnancy, and having witnessed abuse as a child increased a woman’s risk but not significantly. Next, a subanalysis was conducted to explore the effect of intimate partner abuse during adolescence on the association between child abuse and adolescent pregnancy among a subgroup of adolescent women who were between 15 and 19 years old at the time of the survey. Of these 1839 women, only 6% reported any type of abuse by their part80 70
Percentage
60 50 40
583
Table 3 Adjusted odds ratios of adolescent pregnancy risk for each type of child abuse Model 1 Model 2 Model 3 Model 4 Model 5 Sexual Physical Emotional Witnessed Any abuse abuse abuse abuse abuse Sexual Physical Emotional Witness Any abuse
1.48** 1.42*** .90 1.14 1.31***
Note: All models control for education, age, socioeconomic status, and living in rural area. ** p ⬍ .005, *** p ⬍ .001.
ners, but having been abused by a partner significantly increased an adolescent woman’s chance of getting pregnancy during her adolescence. Of all adolescents who had a pregnancy, 80% had been abused by a partner physically, sexually, or emotionally. Likewise, the odds of adolescent pregnancy were 11 times higher for abused versus nonabused adolescents (odds ratio: 11.1, 95% confidence interval 7.0, 17.7), prior to adjusting. A strong relationship was also found between child abuse and partner abuse during adolescence. Experience of child abuse increased a woman’s risk of partner abuse by more than four times (odds ratio of 4.3, 95% confidence interval of 2.8, 6.7), prior to adjusting for confounding factors. The effect of partner abuse on adolescent pregnancy risk was analyzed in stratified analyses among this subgroup, while controlling for socioeconomic status, education, and locality (Table 4). Partner abuse increased the risk of adolescent pregnancy by almost 11 times among all adolescents, as seen in Model 1. Among adolescents abused as children, the effect was similar and significant (Model 2), while the effect was greatest among adolescents who were not abused as children (Model 3). These models show the importance of the partner abuse variable in the association and the fact that partner abuse is an even greater risk factor than childhood abuse on adolescent pregnancy risk. Although the analysis is limited to women in their adolescence, it is suspected that the relationship would be similar among women of older
30 20 10 0 0
1
2
3
4
Number of forms of child abuse
Figure 1. Percentage of women with adolescent pregnancy by number of forms of childhood abuse. Note: 2054 women experienced no abuse; 1177 women experiences one form of abuse, 410 women experienced two forms, 99 women experienced three forms, and 12 women experienced four forms.
Table 4 Odds ratios of adolescent pregnancy by intimate partner violence experience during adolescence (stratified by child abuse history) Model 1 All adolescents
Model 2 Adolescents abused as children
Model 3 Adolescents not abused as children
10.83 (6.43–18.24)
10.63 (5.80–19.49)
12.19 (4.18–35.56)
Note: All models control for age, education, socioeconomic status, living in rural area. Model 1 controls for child abuse.
584
C.C. Pallitto and V. Murillo / Journal of Adolescent Health 42 (2008) 580 –586
age groups as well, if it had been possible to measure adolescent partner abuse among women who were no longer adolescents. Discussion The study found strong support for the hypothesis that abuse during childhood is associated with an increased risk of adolescent pregnancy among a large sample of young women in the general population of El Salvador. Although previous research from the United States and other developed countries has also shown that childhood abuse increased the risk of adolescent pregnancy even after controlling for confounding factors [24,25], such an association has not been previously identified in a Latin American setting. Presumably, these findings would be relevant in other Latin American countries that also have high rates of adolescent pregnancy and similar social and cultural norms. Countries such as Honduras, Guatemala, and Nicaragua have adolescent pregnancy rates of 93, 107, and 113 births per 1000 women aged 15 to 19, respectively [26]. These rates surpass El Salvador’s rate of 81, which is considerably higher than the Latin American average of 71 and the average for developed countries worldwide of 25 births per 1000 women aged 15 to 19 [26]. The findings of this study, although informative to neighboring countries, are not generalizable to these other settings, however, and the association should be analyzed there as well. Unlike other studies with more limited definitions of child abuse, this study considered multiple types of child abuse experiences and found that each additional type of abuse experienced increased one’s risk of adolescent pregnancy. Physical and sexual abuse increased the odds of adolescent pregnancy significantly. Having been a witness to childhood abuse increased the odds of adolescent pregnancy but not significantly, whereas emotional abuse had a paradoxical protective effect on adolescent pregnancy status, which could be because of the inclusion of less abusive forms of discipline as emotional abuse or the fact that the gravity of emotional abuse alone was not sufficient to increase one’s risk of adolescent pregnancy. However, having experienced at least one of the multiple forms of child abuse measured, increased the risk of adolescent pregnancy later in life. This study has demonstrated the importance of considering the role of intimate partner violence on adolescent pregnancy risk. In the few studies to include this variable, some researchers found high rates of partner abuse among pregnant adolescents [27] or greater risk of pregnancy among adolescents experiencing dating violence [28], whereas others found that childhood sexual abuse was a risk factor for abuse during adolescence and adulthood [29]. The connections between these three factors— child abuse, partner abuse during adolescence, and adolescent pregnancy—and the intergenerational patterns of adolescent childbearing [30] and abuse [13] show how the cycles of abuse and adolescent childbearing make it
difficult for the next generation of children born to adolescent mothers to break the cycle. To interpret the findings, it is necessary to consider the status of children’s rights and sexual and reproductive health in El Salvador. El Salvador ratified the Convention on the Rights of the Child, an international treaty that promotes the human rights of children, in 1990. Despite the ratification of the treaty, Salvadoran efforts to promote human rights of children and adolescents have been inadequate, partially because of the lack of human and material resources dedicated to responding to abuse, as well as the lack of coordination in programmatic and policy efforts among public and private agencies working on children’s rights. In addition, many girls, adolescents, and adult women fear that if they report abuse or seek help they will face discrimination or social stigma. These fears are well grounded in a setting in which girls and adolescent women, especially those from impoverished communities in El Salvador, may be more vulnerable to discrimination and victimization because they are young, female, and of a low socioeconomic status. Changing deep-rooted beliefs that foster discrimination and abuse requires a focused multisectoral effort not only at the societal level but also at the family and community levels. In the absence of such changes, individual girls and adolescents will continue to face short- and long-term risks of abuse, low self-esteem, revictimization, and adverse results such as adolescent pregnancy. Issues surrounding education on sexual and reproductive health, including pregnancy prevention, have proven extremely controversial in the country, despite El Salvador’s high adolescent pregnancy rate. Although advocates have promoted school-based and nonschool-based education efforts on sexual and reproductive health in the country, strong resistance among conservative religious and political factions have prevented a comprehensive school-based sexual and reproductive health curriculum from being carried out. As a result, many adolescents find themselves in sexual situations in which they are unprepared and without contraceptives. Rather than feeling capable of negotiating sexual decision making and protecting themselves from pregnancy, HIV/AIDS, and sexually transmitted infections, these sexual encounters put them at risk for pregnancy and other health risks. As long as reproductive health remains a taboo topic and adolescents fail to receive appropriate information and services, adolescents will be unprepared to prevent premature pregnancies, as evidenced by the epidemic proportions at which adolescent pregnancy is occurring in the country. Despite these important findings, a few limitations in the analysis must be recognized. The sample was limited to a subgroup of women between the ages of 15 to 24 years old, because only this age group was asked about age of first pregnancy. The subgroup differed significantly from the older women interviewed in terms of child abuse history, contraceptive use, education level, and where they lived. In addition, the cross-sectional nature of the study limits the conclusions about
C.C. Pallitto and V. Murillo / Journal of Adolescent Health 42 (2008) 580 –586
causality, although we attempted to limit errors of reverse causality in a few ways— by only categorizing adolescents as having been abused if the abuse preceded the pregnancy and by categorizing sexual abuse by a family member as having occurred prior to age 18 (a similar option was not possible for the physical and emotional abuse during childhood variables). Likewise, with the partner abuse and pregnancy association, the order in which these events occurred is not known. Although we hypothesize that the abuse preceded the pregnancy, it is also plausible that pregnancy preceded abuse by intimate partners. In addition, because the study deals with sensitive issues of childhood abuse, partner abuse, and adolescent sexual behavior, it is possible that there was underreporting for these questions, which could result in underestimates of the association studied here. Conclusions Clearly, the time has come to break the cycle of abuse and adolescent pregnancy in El Salvador and other countries in Latin America. Efforts to reduce child abuse and adolescent pregnancy should include the following: awareness raising about the rights of children and adolescents to live without violence from their families, partners, or communities; education on existing mechanisms for reporting abuse; expansion of the response of the judicial and social service sectors in cases of suspected and confirmed abuse; training of medical personnel to detect and report abuse more proactively, to respect the integrity of children and adolescents, and to provide services in a nonjudgmental manner; work with young women to promote self-esteem, learn about sexual and reproductive health, and encourage more healthy negotiation regarding sexual decision making; promote male involvement in pregnancy prevention, responsible parenting, gender equality, and conflict resolution among adolescent and adult men; provision of ageappropriate education about sexual and reproductive health; prevention of school desertion among pregnant and nonpregnant teens, and promotion of mechanisms of support for adolescent mothers. All efforts must be coordinated between governmental institutions and civil society at both the national and community levels, involving policy changes as well as community-based efforts for reaching a broader base of vulnerable youth and their families more directly before the consequences to Salvadoran society are even greater. Acknowledgments Funding for the study was provided by the United States Agency for International Development, through the POLICY II Project, implemented by Futures Group. The authors were independent researchers living in El Salvador at the time they were awarded funds for this study. The authors wish to thank Verónica Simán de Betancourt of the POLICY II Project and José Mario Cáceres of the Asociación Demográ-
585
fica Salvadoreña for their insights and feedback on the project documents. References [1] Encuesta Nacional de Salud Familiar—FESAL 2002/3. Informe Final 2004. San Salvador, El Salvador: Asociación Demográfica Salvadoreña. [2] Mayor S. Pregnancy and childbirth are leading causes of death in teenage girls in developing countries. BMJ 2004;328:1152. [3] Alan Guttmacher Institute. Risks and Realities of Early Childbearing Worldwide. [Issues in Brief]. New York: The Institute, 1996. [4] Moore KA, Myers DR, Morrison C, et al. Age at first childbirth and later poverty. J Res Adolesc 1993;3(4):393– 422. [5] Grogger J, Bronars S. The socioeconomic consequences of teenage childbearing: findings from a natural experiment. Fam Plann Perspect 1993;25(4):156 – 61. [6] Blinn-Pike L, Berger T, Dixon D, et al. Is there a causal link between maltreatment and adolescent pregnancy? A literature review. Perspect Sex Reprod Health 2002;34(2):68 –72. [7] Hillis SD, Anda RF, Dube SR, et al. The association between adverse childhood experiences and adolescent pregnancy, long-term psychosocial consequences, and fetal death. Pediatrics 2004;113:320 –7. [8] Stock JL, Bell MA, Boyer DK, et al. Adolescent pregnancy and sexual risk-taking among sexually abused girls. Fam Plann Perspect 1997;29(5):200 –203, 277. [9] Rainey DY, Stevens-Simon, Kaplan DW. Are adolescents who report prior sexual abuse at higher risk for pregnancy? Child Abuse Neglect 1995;19(10):1283– 8. [10] Guijarro S, Naranjo J, Padilla M, et al. Family risk factors associated with adolescent pregnancy: study of a group of adolescent girls and their families in Ecuador. J Adolesc Health 1999;25(2):166 –72. [11] Kenney JW, Reinholtz C, Angelini PJ. Ethnic differences in childhood and adolescent sexual abuse and teenage pregnancy. J Adolesc Health 1997;21(1):3–10. [12] Raj A, Silverman JG, Amaro H. The relationship between sexual abuse and sexual risk among high school students: findings from the 1997 Massachusetts youth risk behavior survey. Matern Child Health J 2000;4(2):125–34. [13] Boyer D, Fine D. Sexual abuse as a factor in adolescent pregnancy and child maltreatment. Fam Plann Perspect 1992;24(1):4 –11, 19. [14] Butler JR, Burton L. Rethinking teenage childbearing: is sexual abuse a missing link? Fam Relat 1990;39(1):73– 80. [15] Esperat MC, Esparza DV. Minority adolescent mothers who reported childhood sexual abuse and those who did not: perceptions of themselves and their relationships. Issues Mental Health Nurs 1997;18(3):229 – 46. [16] Fiscella K, Kitzman HJ, Cole RE, Sidora KJ, Olds D. Does child abuse predict adolescent pregnancy? Pediatrics 1998;101(4):620 – 4. [17] Kellogg ND, Hoffman TJ, Taylor ER. Early sexual experiences among pregnant and parenting adolescents. Adolescence 1999;34(134): 293–303. [18] Noll JG, Trickett PK, Putnam FW. A prospective investigation of the impact of childhood sexual abuse on the development of sexuality. J Consult Clin Psychol 2003;71(3):575– 86. [19] Luster T, Small SA. Sexual abuse history and number of sex partners among female adolescents. Fam Plann Perspect 1997;29(5):204 –11. [20] Cinq-Mars C, Wright J, Cyr M, McDuff P. Sexual at-risk behaviors of sexually abused adolescent girls. J Child Sex Abuse 2003;12(2):1–18. [21] Steel JL, Herlitz CA. The association between childhood and adolescent sexual abuse and proxies for sexual risk behavior: a random sample of the general population of Sweden. Child Abuse Neglect 2005;29(10):1141–53. [22] DeBellis MD. Developmental traumatology: the psychobiological development of maltreated children and its implications for research, treatment and policy. Dev Psychopathol 2001;13(3):539 – 64.
586
C.C. Pallitto and V. Murillo / Journal of Adolescent Health 42 (2008) 580 –586
[23] Miller M. A model to explain the relationship between sexual abuse and HIV risk among women. AIDS Care 1999;11:3–20. [24] Saewyc EM, Magee LL, Pettingell SE. Teenage pregnancy and associated risk behaviors among sexually abused adolescents. Perspect Sex Reprod Health 2004;36(3):98 –105. [25] Raj A, Silverman JG, Amaro H. The relationship between sexual abuse and sexual risk among high school students: findings from the 1997 Massachusetts youth risk behavior survey. Matern Child Health J 2000;4(2):125–34. [26] United Nations Population Fund (UNFPA). State of the World Population 2007: Unleashing the Potential of Urban Growth. New York: UNFPA, 2007.
[27] Parker B, McFarlane J, Soeken K. Abuse during pregnancy: effects on maternal complications and birth weight in adult and teenage women. Obstet Gynecol 1994;84(3):323– 8. [28] Silverman JG, Raj A, Clements K. Dating violence and associated sexual risk and pregnancy among adolescent girls in the United States. Pediatrics 2004;114(2):e220 –5. [29] Fergusson DM, Horwood LJ, Lynskey MT. Childhood sexual abuse, adolescent sexual behaviors and sexual revictimization. Child Abuse Neglect 1997;21(8):789 – 803. [30] Bonnell C, Allen E, Strange V, et al. Influence of family type and parenting behaviours on teenage sexual behaviour and conceptions. J Epidemiol Community Health 2006;60(6):502– 6.