Psych Paper

10 | International Journal of Childbirth Education | Volume 29 Number 1 January 2014

Introduction The financial costs of teenage pregnancy were estimated

at over $11 billion dollars in the United States in 2008 (Cen- ters for Disease Control and Prevention, 2012). In addition to societal costs, teenage pregnancy is associated with a host of negative consequences for both these mothers and their children. These sequelae are far-reaching and encompass medical, psychological, and economic domains. Adolescent pregnancy carries increased risk of specific obstetric com- plications, including anemia and hypertension (Kramer & Lancaster, 2010), unsafe abortion, postpartum hemorrhage, depression, and maternal death (World Health Organization [WHO], 2013). Independent of known confounding factors such as socioeconomic status, neonatal outcomes are poorer in adolescent pregnancies with higher rates of neonatal death, premature birth, low birth weight, and low APGAR scores (Chen et al., 2007).

Children of teenage mothers are at increased risk for adverse developmental outcomes, although the relation- ship may not be causal (Levine, Emery, & Pollack, 2007). In Cornelius et al.’s (2009) study, 6-year-old children of teenage mothers had significantly higher body mass indices, smaller head circumferences, and lower child IQ scores than children of older mothers. De Genna, Larkby, and Cornelius (2011) noted that children of teenage mothers are at an increased risk of becoming a young parent, particularly if they experi- ence early puberty onset. In De Genna et al.’s (2011) study of 254 14-year-old offspring of teenage mothers, one in four children had engaged in sexual intercourse. Francesconi (2008) examined data from the British Household Panel Survey and found that adults who were children of teenage mothers experienced fewer academic successes, limited eco- nomic prosperity, and increased odds of becoming a teenage parent.

Primary Research

Post Trauma Moderators in the Childhood Sexual Abuse–Teenage Pregnancy/Teenage Childbirth Relationships by Amy E. Sickel, PhD, Dana M. Dillard, MS, Penelope K. Trickett, PhD, Frank W. Putnam, MD, Jennie G. Noll, PhD

Abstract: The relationships between Childhood Sexual Abuse (CSA) and both teenage pregnancy and childbearing were examined using data from 71 sexually abused and 80 comparison females post abuse. Logistic regression was used to test the relationships between CSA and both teenage pregnancy and childbearing and whether participants’ non-abusing caregiver (i.e., mothers) perceptions of family environment moderated either reproductive outcome. Abused females reported significantly more teenage pregnancies and teenage childbearing. Maternal perceptions of decreased organization in abuse family environments predicted teenage pregnancy, and maternal perceptions of decreased control in abuse family environments predicted teenage childbearing.

Keywords: child sexual abuse/csa, teenage pregnancy/childbearing, moderators, family environment, maternal perceptions

This research was partially supported by grants from the National Institute of Mental Health (MH48330, MH01284); National Institute of Child Health and Human Development (HD41402, HD045346, HD 060604); National Center on Child Abuse and Neglect (90-CA-1549, 90-CA-1686); and private foun- dations including the W.T. Grant Foundation, Smith Richardson Foundation, and John Templeton Foundation (ID5119).

continued on next page

Volume 29 Number 1 January 2014 | International Journal of Childbirth Education | 11

Further complicating these challenges, adolescent parents are at substantial risk for not obtaining a high school diploma or its equivalent as compared to non-parent peers (Mollborn, 2010), which may result in lasting economic disparities. Because of the poor outcomes associated with early childbirth, the issue of teenage childbearing remains of particular concern for subgroups of the adolescent popula- tion for whom early onset of sexual activity, increased sexual activity, and poor contraceptive practices are known to occur. One such subset includes women with histories of childhood sexual abuse (CSA).

There is considerable support for the direct association between abuse and both teenage pregnancy and teenage childbearing. In their review of 14 studies, Noll, Shenk, and Putman (2009) found that females were twice as likely to become pregnant in adolescence if they experienced CSA. Noll and Shenk (2013) further found that both CSA and neglect act as unique predictors of teen childbirth. Francisco et al. (2008) also reviewed this literature and found that the majority of studies did support a direct CSA-teenage preg- nancy relationship. Hillis et al. (2004) too found that 60% of women sampled with CSA history were likely to have teenage pregnancies. In a large state wide survey of high school students, Saewyc, Magee, and Pettingell (2004) found that students with a history of sexual abuse were more likely to have a history of teen pregnancy and to engage in current risky sexual behavior. Pallitto and Murillo (2008) found international support for this link in their sample of El Salva- dorian females. The authors found greater risk of adolescent pregnancy as associated with sexual abuse relative to physical abuse or experiencing violence more generally.

Indirect links between CSA and teenage pregnancy have also been identified. Early onset of or an increase in risky sexual behavior predicts teenage pregnancy. Coping strate- gies that increase the possibility of risky sexual behaviors may emerge as the result of CSA, such as increased substance abuse, prostitution, and running away (Francisco et al., 2008; Raj, Silverman, & Amaro, 2000; Saewyc et al., 2004). Hillis et al. (2004) noted that adverse childhood events are significantly correlated with teenage pregnancy. One chal- lenge with this literature is that when teenage pregnancy is assessed, it is often assumed to result in teenage childbear- ing. In fact, data from national surveys do not differentiate

between teen pregnancies that result in live births, miscar- riages, or abortions (Office of Adolescent Health, 2013). While childbirth may appear to be an obvious outcome, childbirth and pregnancy may also have unique predictors. In addition, results can be seen as varied to the extent that researchers fail to simultaneously study both teenage preg- nancy and teenage childbearing as potential outcomes.

Family environment may be a central influence in the CSA-teenage pregnancy and CSA-teenage childbearing relationships. Abramovic (2005) asserted that victims of CSA tended to come from disruptive or broken homes, a finding supported by Hillis et al. (2004). McClure, Chavez, Agars, Peacock, and Matosian (2008) expanded upon the role of family conflict in CSA in their investigation of resiliency in survivors. McClure et al. found that higher levels of self- reported family cohesion corresponded with greater levels of resiliency, whereas greater family conflict corresponded with reduced resiliency. Related to family environment, many victims of CSA experience unresolved/disorganized attachment, which may subsequently and adversely affect mother-child attachments for the adolescent mother (Bailey, Moran, & Pederson, 2007). Given existing research docu- menting the impact of CSA family environment variables in particular along with an increasing awareness of the potential impact of maternal attachment and attitudes on adolescent sexual behavior in the general population (Dittus & Jaccard, 2000; Tyano, Keren, Herman & Cox, 2010), it is reasonable that maternal perceptions of family environment may offer unique insight into CSA females’ reproductive behavior.

Considering our knowledge of both the prevalence of teenage pregnancy and childbearing in the United States and the complicated and far reaching effects of CSA, a more pre- cise understanding of the causal mechanisms specific to the CSA-teenage pregnancy/teenage childbearing relationships is warranted. The current analysis asks the following four questions: 1) What is the strength and nature of the direct relationship between CSA and teenage pregnancy? 2) What is the strength and nature of the direct relationship between CSA and teenage childbearing? 3) If the CSA-teenage preg- nancy relationship is significant, do perceptions of family environment influence the likelihood that the CSA female will have a teenage pregnancy? 4) If the CSA-teenage child-

Post Trauma Moderators continued from previous page

continued on next page

12 | International Journal of Childbirth Education | Volume 29 Number 1 January 2014

bearing relationship is significant, do perceptions of family environment influence the likelihood that the CSA female will have a teenage childbirth? First, it is hypothesized that there will be a significant and positive relationship between CSA and teenage pregnancy. Second, it is hypothesized that there will be a significant and positive relationship between CSA and teenage childbearing. Third, it is hypothesized that mothers of CSA females will report a significantly negative relationship between perceived family environment and teenage pregnancy. Fourth, it is hypothesized that mothers of CSA females will report a significantly negative relationship between perceived family environment and teenage child- bearing.

Method Participants

Participants for the current analysis were part of a long-term study on the psychobiological impact of CSA on female development in which participants were assessed over six time points (T1 – T6). Abused females were referred by protective service agencies in the metropolitan Washington, D.C. area. Eligibility for study inclusion required: 1) the victim be female, age 6-16; 2) disclosure of referring abuse had occurred within 6 months; 3) sexual abuse involved genital contact and/or penetration; 4) the perpetrator was a family member including parent, stepparent, sibling, uncle or mother’s live-in boyfriend; and 5) a non-abusing parent/ guardian (usually the child’s mother) was willing to par- ticipate in the project. Comparison females were recruited via community advertising and were similar to the abused participants in terms of ethnic group, age, pre-disclosure socioeconomic status (SES), family constellation (one vs. two parent families), and history of other non-sexual traumas. All families ranged from low to middle SES, with mean Hol- lingshead (1975) scores of approximately 35 (defined as ‘blue collar’ or working class).

Participants in the original Time 1 (T1) sample included sexually abused females (n=84) and comparison females (n=82) for a total T1 sample size of 166. During the inter- views subsequent to T1, 14 comparison females revealed some form of childhood sexual abuse occurring after T1 and

were dropped from the comparison group and delegated to a new group named Group 3 or the ‘noncriterion abuse’ group. These 14 participants are not utilized in the present set of analyses. At T4, 21 new comparison females were recruited to augment the longitudinal sample bringing the fortified sample size to n=173. The average participant ages for the six data collections were: T1 M= 10.67, SD=3.04; T2 M=11.86, SD=3.04; T3 M=13.02, SD=3.04; T4 M=17.59, SD=3.39; T5 M=19.99, SD=3.40 and T6 M=23.98, SD=3.23.

The current analysis is twofold. First, it measures the di- rect relationship between CSA and both teenage pregnancy and teenage childbearing. Second, it considers the post- trauma/T1data as potential moderators in the CSA-teenage pregnancy and CSA-teenage childbearing relationships. The current analysis examines perceptions of family environment by mothers of CSA females (T1); demographic variables including Age, Race/Ethnicity (T1), and socioeconomic status (SES; T1 or closest time measured), as well as history of teen- age pregnancy and teenage childbearing (T5/T6).

Measures Socioeconomic Status (SES). SES is measured using

the Hollingshead Four Factor Index of Social Status (Hol- lingshead, 1975). This measure produces a single score for an individual’s SES rating which is a product of both occupation and education. Reliability on the Hollingshead is reportedly acceptable and Gottfriend reports 95% agreement in inter- rater reliability (Gottfried, 1985).

Race/Ethnicity. Race/Ethnicity is a variable in which participants were coded as either Caucasian/European/Non- Hispanic ethnicity, or being from a non-Caucasian/Euro- pean/Non-Hispanic ethnicity.

Age. Age was reported in years, months, and days for the longitudinal study and is rounded to the current year in the present analyses.

Teenage Pregnancy & Teenage Childbearing. Teenage pregnancy was coded if the participant reported a history of pregnancy, regardless of pregnancy outcome, between the ages of 13 and 19 years. Teenage childbearing was coded if the participant reported a history of giving birth as a teenager between the ages of 13 and 19. These variables were taken from the Sexual Activities & Attitudes Questionnaire (SAAQ), computerized administration. The SAAQ is a validated measure. Reliabilities for the factor analyzed scale items of the SAAQ range from .94 to .37 with the majority of alphas rang- ing from .61-.94 (Noll, Trickett, & Putnam, 2003).

Post Trauma Moderators continued from previous page

continued on next page

Volume 29 Number 1 January 2014 | International Journal of Childbirth Education | 13

Family Environment. Perceptions of family environ- ment were reported from participating CSA female mothers (FESM) using Moos and Moos’ (1981) Family Environment Scale. Form ‘R’ of the FES was used and was designed to measure an individual’s perceptions of their nuclear family. This is a ninety-item scale measuring ten separate subscale or aspects of the family environment. Each composite consists of nine items. According to the scale’s authors, the subscales are designed to measure the following constructs: 1) cohesion as the degree of commitment, help, and support family members provide for one another; 2) expressiveness as the extent to which family members are encouraged to act openly and to express their feelings directly; 3) conflict as the amount of openly expressed anger, aggression and conflict among family members; 4) independence as the extent to which family members are assertive, self-sufficient and make their own decisions; 5) achievement as the extent to which activities (such as school and work) are cast into an achieve- ment oriented/competitive framework; 6) culture as the degree of interest in political, social, intellectual and cultural activities; 7) recreation as the extent of participation in social and recreational activities; 8) religion as the degree of em- phasis on ethical and religious issues and values; 9) organiza- tion as the degree of importance of clear organization and structure in planning family activities/responsibilities; and 10) control as the extent to which set rules and procedures are used to run family life. Higher scores indicate stronger construct endorsement (e.g. higher control score indicates greater perceived control in the family environment). Test- retest reliabilities for the FES subscales ranged from .61 to .78 (Moos & Moos, 1981).

Procedure Study participants were interviewed individually at a

central location. Most measures for this analysis were com- pleted in a face-to-face, semi-structured interview format with a trained researcher as part of a larger four to five hour protocol. The SAAQ was administered via computer. Follow- ing debriefing, participants were thanked and reimbursed monetarily for their participation.

Missing Data Imputation. In order to maximize the number of participants used for analyses, missing data were imputed for those items missing from larger, established inventories (e.g. FESM) and information from present items

was used to impute missing items. The NORM (Schafer & Graham, 2002) software program was used to impute data. NORM is a multiple imputation (MI) missing data program, which utilizes the EM algorithm to generate start values. MI is a Monte Carlo technique in which the missing values are replaced by multiple simulations (Rubin, 1987). Not more than 1% of the data was missing. Missingness did not cor- relate significantly with the outcome variable, teenage preg- nancy; therefore, missing-at-random assumptions (Rubin, 1987) were not violated.

Results Descriptive Analyses

There were eight females who did not participate in either T5 or T6, thus the attrition rate for this current set of analyses is 4.6%. Because the current analysis intends to study family environment perceptions as post-CSA modera- tors which occur prior to teenage pregnancy and teenage childbearing and which were measured using both T5 and T6, one final criteria for inclusion is that all females in the current analysis are either over 20 years of age at T5/T6 so as not to include teenagers who may still have a teenage pregnancy; or, be females at T5/T6 who are teenagers but who have already had a teenage pregnancy. There were 14 participants who did not meet this criterion. As a result, participants in the current analysis (n=151; 173-8-14=151). At T1 these participants were approximately 11 years old (M=10.94, SD=2.975) and had an average or middle class SES (M=36.66, SD=12.663). There were 71 abused females and 80 comparison females meeting inclusion criteria for this analysis. Race/Ethnicity indicated 76 participants to be Cau- casian/European/Non-Hispanic with 75 participants being non-Caucasian/European/Non-Hispanic or minority status at time of interview. Regarding having a history of teenage pregnancy, 77 participants had a teenage pregnancy regard- less of pregnancy outcome while 66 participants had no history of teenage pregnancy. Last, 42 participants reported a history of teenage childbearing while 59 reported no teenage childbearing. As seen in Table 1, SES and Race/Ethnicity significantly correlate with several of the study predictor variables. Given the potential for confounding influences, SES, and Race/Ethnicity were controlled in all multivariate analyses.

Internal Consistency. In order to review the internal consistency of the FESM items for this population, Cron-

Post Trauma Moderators continued from previous page

continued on next page

14 | International Journal of Childbirth Education | Volume 29 Number 1 January 2014

bach’s alphas were run on the ten FESM subscales, which yielded these reliabilities: Control (r=.45), Cohesion (r=.76), Expression (r=.35), Conflict (r=.71), Independence (r=.49), Achievement (r=.54), Culture (r=.65), Recreation (r=.73), Religion (r=.62), and Organization (r=.61).

Inferential Analyses To exact some control over the experimental-wise alpha,

an omnibus F tests was run on the family environment vari- ables (CSA female mother perceptions). The omnibus F was significant (p=.038), and individual comparisons of a priori hypothesized group differences revealed were significant or marked trends in significance, including mother perceptions of family cohesion (p=.004), culture (p=.028), expression (p=.024), recreation (p=.016) and organization (p=.098)

Logistic Regression Results. To test the first hypothesis regarding the relationship between CSA history and teenage pregnancy, a logistic regression was run in which Race/Eth- nicity and SES were entered as covariates. This relationship was significant (Odds Ratio = 2.666, p=.007), indicating that for every one unit increase to abuse status, abused females were over two and a half times more likely to have had a teenage pregnancy.

To test the second hypothesis regarding the relationship between CSA history and teenage childbearing, a logistic regression was run in which Race/Ethnicity and SES were entered as covariates. This relationship was significant (Odds Ratio = 5.067, p=.000) indicating that for every one unit increase to group status, abuse females were approximately five times more likely than comparison females to have had a teenage childbirth.

Due to the significant findings of the first two hypoth- eses tests, additional logistic regressions were run to examine maternal family perceptions acting as moderator in the CSA-teenage pregnancy and CSA-teenage childbearing rela- tionships. Demographic variables, again excluding age, were entered as covariates. The results of the test of interaction between CSA and mother perceptions of family environment are presented in Table 2.

The results of the test for hypothesis three are also seen in Table 2. The interaction between abuse history and moth- er perceptions of T1 family organization is significant (Odds Ratio=.629, p=.039). For every one unit decrease in per- ceived organization, abuse females are 36% more likely than

Table 1. Zero-Order Correlations between Demographic, T1 Maternal Perception of Family Environment, Group Status (Abuse/ Comparison Females), Teenage Pregnancy and Teenage Childbearing Variables Variable Age Race/Ethnicity Socioeconomic Status/SES Control .112 .124 -.177* Cohesion -.050 .237*** .019 Expression -.135 -.108 .206 Conflict -.006 -.270*** .247*** Independence .038 .089 .208 Achievement .022 .336*** -.053 Culture .081 .208** .247*** Recreation -.096 .008 .278*** Religion .086 .260*** -.094 Organization -.084 .268*** -.110 Group Status .107 -.140 -.050 Teenage Pregnancy .010 .093 -.212** Teenage Childbearing -.118 .039 -.179*

Note: *p<.10. **p<.05. ***p<.01.

Post Trauma Moderators continued from previous page

continued on next page

Volume 29 Number 1 January 2014 | International Journal of Childbirth Education | 15

control females to have had a teenage pregnancy. Significant at the p<.10 level, the interaction between abuse history and mother perceptions of T1 family expression is significant (Odds Ratio=1.584, p=.083). This indicates that for every one unit decrease in family expression, abused females are 54% more likely to have had a teenage pregnancy.

Also seen in Table 2 are the results of the logistic regressions testing hypothesis four, the interaction of abuse status and maternal perceived family environment in the prediction of teenage childbearing. The interaction between abuse history and mother perceptions of T1 family control is

significant (Odds Ratio=2.036, p=.036) suggesting that for every one unit decrease in perceived family control, abused females are over two times more likely than are comparison females to have given birth to a child as a teenager. Signifi- cant at the p<.10 level, the interaction between abuse history and mother perceptions of T1 family conflict is significant (Odds Ratio=.671, p=.096). This indicates that for every one unit increase in family conflict in perceived abuse family environments, abused females are 60% more likely to have given birth to a child as a teenager.

Table 2. Logistic Regression Results for the Interaction between Maternal Perceptions of Family Environment and Abuse History in Relation to Teenage Pregnancy and Teenage Childbearing. Variable Parameter Estimate Standard Error Odds Ratio p-value

Teenage Pregnancy

Control .186 .232 1.204 .424 Cohesion -.273 .207 .761 .188 Expression .460 .265 1.584 .083* Conflict .044 .197 1.045 .824 Independence .071 .259 1.073 .785 Achievement -.102 .248 .903 .682 Culture -.170 .203 .843 .401 Recreation -.128 .177 .880 .469 Religion -.365 .231 .694 .114 Organization -.463 .225 .629 .039**

Teenage Childbearing

Control .711 .339 2.036 .036** Cohesion -.063 .270 .939 .815 Expression .068 .386 1.071 .860 Conflict -.398 .239 .671 .096* Independence .170 .332 1.185 .609 Achievement .446 .315 1.562 .156 Culture -.212 .250 .809 .396 Recreation -.052 .209 .950 .805 Religion .408 .281 1.504 .147 Organization .405 .299 1.500 .175

Note. *p<.10. **p<.05. ***p<.01, controlling for Race/Ethnicity and Socioeconomic Status/SES.

Post Trauma Moderators continued from previous page

continued on next page

16 | International Journal of Childbirth Education | Volume 29 Number 1 January 2014

Discussion Teenage pregnancy is associated with medical, psycho-

logical and economic sequelae for both the individual and for society at large. While current statistics show an encour- aging decline in teen pregnancy rate (Hamilton, Martin & Ventura, 2013) research with females having a sexual abuse history continues to document the prevalence of teenage pregnancy and teenage childbearing (Noll et al., 2009; Saewyc et al., 2004; Trickett, Noll & Putnam, 2011). The current analysis is one of very few prospective studies of the relationship between sexual abuse history and both teenage pregnancy and teenage childbearing. It is also one of very few known studies which: 1) employs a rigorous definition of CSA, 2) utilizes a well-matched comparison group of non- sexually abused females, and 3) examines maternal percep- tions of family environment as a moderator in the CSA- teenage pregnancy/childbearing relationships. Females with a history of CSA were more likely to have had both a teenage pregnancy (p=.007) and teenage childbearing (p=.000).

CSA and Teenage Pregnancy The results of the current analysis are consistent with

other studies examining both the indirect (Francisco et al., 2008; Raj et al., 2000; Saewyc et al., 2004) and direct relationships between CSA and teenage pregnancy (Noll et al., 2009; Noll & Shenk, 2013; Roberts, O’Connor, Dunn, & Golding, 2004; Saewyc et al., 2004; Trickett, Noll & Putnam, 2011). Specifically, females with a history of CSA were two and a half times more likely to have had a teenage pregnancy and over five times more likely to have given birth as a teenager.

As described by many of the studies presented, sexual abuse can lead to an overall tendency or proclivity towards early sexual behavior. Abuse-specific reasons for this ten- dency may include exposure to early unwanted sexual behav- ior and subsequent acceptance of sexual behavior as a way to connect with others (Finkelhor, 1988) or a means to recreate/ attempt to master the earlier traumatic event (American Psychiatric Association, 2013). All of these explanations imply that teenage pregnancy is not necessarily the desired outcome or conscious choice per se, but is the end result of a trauma-related trajectory.

Teenage pregnancy and, indeed, teenage childbearing, may occur more often in CSA females for non-abuse reasons.

There is a large body of research which finds that having a lower SES and being of particular racial/ethnic groups may increase the likelihood of teenage pregnancy – creating a culture which not only condones but supports early child- bearing as a means of obtaining status within that commu- nity (Furstenberg, 1992; Miller, 2002; Quinlivan, Tan, Steele & Black, 2004). Therefore, to the extent that CSA occurs within lower socioeconomic groups and cultural groups supporting teenage reproductive behavior, this may create an environment in which teenage pregnancy and childbearing is more normative than in higher SES and cultures not endors- ing earlier reproduction.

Third, increased teenage pregnancy in CSA females may be somewhat intentional. Some research has examined not only the presence of positive attitudes about having a baby as a teenager (Rocca, Doherty, Padian, Hubbard & Min- nis, 2010) but also the absence of negative attitudes toward having babies as predictive of sexual behavior in adolescents

Post Trauma Moderators continued from previous page

continued on next page

A pril M

atew s

Volume 29 Number 1 January 2014 | International Journal of Childbirth Education | 17

(Stevens-Simon, Kelly, Singer, & Cox, 1996). It may be that female teens with a history of sexual abuse are more apt to be positive or ambivalent about becoming pregnant than their comparison counterparts.

Family Environment The current findings are also consistent with previous

research regarding perceptions of family environment. Previ- ous research indicates that perceptions of family cohesion or strength are associated with adolescent sexual outcome, which specify that decreases in cohesion and strength relate to increased sexual behavior and teenage pregnancy (Mc- Clure et al., 2008; Miller, 2002; Quinlivan et al., 2004). Salient in the current analysis is the fact that family environ- ment interacted with abuse status in the prediction of both teenage pregnancy and teenage childbearing, suggesting the potential scope of family influence to be considerable in terms of subsequent adolescent female sexual and reproduc- tive behavior. Maternal perception of decreased family orga- nization was associated with increased teenage pregnancy in abuse females. On the other hand, maternal perception of decreased family control was associated with teenage child- bearing in abused females. These findings collectively suggest increased organization and control might be beneficial in abuse family environments with regard to teenage reproduc- tive behavior.

Overall, CSA females’ maternal perceptions of family environment appear quite influential regarding the likeli- hood of teenage pregnancy and teenage childbearing, both of which can have sizeable impact not only on the individu- als involved but on the community more generally (WHO, 2013). Given the tremendous stress that sexual abuse imposes upon the family as a whole, it is not surprising that perceptions of that environment influence abused females’ behavior. More informative, perhaps, is the fact that percep- tions of the family environment significantly influence abuse females’ adolescent sexual and reproductive behavior. To the extent that maternal perceptions accurately or sufficiently reflect the family environment and the family environment lacks control and organization – the abused female may see herself as disorganized and lacking control. Teenage preg- nancy, for example, may be perceived as a way to change her environment such that the teenager presumably has control (as parent) and believes she can create an environment that

is more organized than the one in which she was raised. Alternately, if an abused female grows up with less con-

trol and organization in her surroundings, she may simply not have the resources to control her own choices and be- havior in the form of using birth control. In either case, this finding warrants continued attention and offers a possible application for treatment. If CSA females enter treatment, it might be important to thoroughly evaluate maternal percep- tions of the family environment with the understanding that these perceptions may be influential upon subsequent adolescent sexual and reproductive behavior and not merely descriptive.

The current analysis is one of very few known studies to assess maternal perceptions of family environment in relation to teenage pregnancy and teenage childbearing. Compli- mentary to the results of Mannarino and Cohen (1996) who found that mothers of CSA females’ reported their families as low on cohesion and adaptability, the current analysis finds that CSA mothers identify organization and control as salient family issues.

It is noteworthy that maternal perceptions of family ex- pression and conflict affect teenage pregnancy and childbear- ing, at p<.10. Mothers of abused females indicated T1 family environments to be less expressive but more conflicted than do mothers of the comparison females. This suggests that abused families are particularly vulnerable to decreased organization, control, and to a lesser extent, expression along with increased conflict. This is consistent with literature that finds both households with a history of CSA and households with teenage pregnancies to be especially chaotic and disor- ganized (Miller, Benson, & Galbraith, 2001; Swanston et al., 2003). The results of the maternal data also suggest potential clinical application, as it may be important for clinicians to recognize first, that mothers’ perspectives of family environ- ment are important predictors in their daughters’ likelihood of teenage pregnancy. Second, as these moderating variables were assessed at T1, shortly after the abuse occurred, these findings suggest the need for swift intervention for CSA families. It may be that early therapeutic support for CSA families might help mitigate these effects on adolescent reproductive behavior.

Limitations This research contributes to previous research on child-

hood sexual abuse and the relationships CSA has with both teenage pregnancy and teenage childbearing and introduces

Post Trauma Moderators continued from previous page

continued on next page

18 | International Journal of Childbirth Education | Volume 29 Number 1 January 2014

new findings regarding these direct relationships in the inter- action of family environment with CSA. There are limita- tions to this work. First, the predictor variables of perceived family environment are measured solely through maternal self-report. In addition, some of the alphas reported on these subscales, while acceptable by some standards (Mitchell & Jolley, 2004), would not be considered acceptable by others; therefore, it would be beneficial to measure fam- ily environment using a different measure to validate these findings. Second, the sample size of the females included in this analysis for which the T1 family environment was both appropriate and obtained is relatively small. Thus, generaliz- ability is cautioned. Lastly, abuse participants in this analysis likely provide a conservative estimate of the effects of abuse as these are females for whom the abuse was both identified and substantiated and who were willing and able to partici- pate in a long-term research study.

Future Research The current analysis supports the role of childhood sex-

ual abuse as directly contributing to teenage pregnancy and to teenage childbearing. As the current analysis measured teenage pregnancy in terms of any pregnancy occurring in teenage years regardless of outcome (i.e. birth, miscar- riage, abortion), future research should attempt to replicate this finding using this definition of teenage pregnancy. The current analysis finds that family environment significantly interacts with abuse history in predicting both teenage pregnancy and teenage childbearing, particularly the impact of family control and organization. Thus, future work may want to expand upon this by increasing the number of family members for whom a measure of family environment can be taken (e.g., sisters and brothers), use certain legal docu- ments such as police records or social service documents to more objectively examine family environment, and perhaps comparing those measures to CSA females self-report of the family environment.

Finally, while work has begun in this area, future work may want to examine the role of extended family influences, including boyfriends or partners, to measure their impact upon both family environment and teenage pregnancy.

Conclusion The current analysis is one of very few prospective stud-

ies to examine the direct relationship between childhood sexual abuse and teenage reproductive behavior. This analysis uses a stringent definition of sexual abuse and compares a sample of CSA females to a well-matched comparison group, which many studies are unable to do. Results from this research caution the interpretation of current statistics reporting a decrease in national rates of teenage pregnancy and suggests that for particular subpopulations, teenage pregnancy continues to be a significant concern. This analysis underscores the importance of family environment in affect- ing teenage pregnancy and childbearing, specifically mother perceptions of family control and organization. Considerable research has documented the deleterious effects of CSA on females – particularly on their subsequent sexual behavior. The current analysis extends this research and suggests that the family environment as construed by the CSA female mother may have particular influence on the occurrence of teenage reproductive behavior in females with a history of childhood sexual abuse.

Implications of CSA, Teenage Pregnancy, and Teenage Childbirth • Declining rates of teenage pregnancy may not ac-

curately reflect population differences in teenage pregnancy and childbirth.

• Each increase in abuse status corresponds to a 166% increase in teenage pregnancy.

• Each increase in abuse status corresponds to a greater than 400% increase in teenage childbirth.

• Pregnant and childbearing teenagers with a his- tory of CSA may be at risk of having complicated, dysfunctional, or unhealthy family environments.

• Childbirth educators have an opportunity to identify CSA and to offer resources and referrals for healing as well as parenting classes for expecting and new teenage parents.

• Information on healthy communication, boundar- ies, rule-setting, discipline, and emotional expres- sion may be useful topics to include in parenting seminars for teenage parents, particularly if there is a suspected or verified history of CSA.

Post Trauma Moderators continued from previous page

continued on next page

Volume 29 Number 1 January 2014 | International Journal of Childbirth Education | 19

Childbirth educators are in a unique position to identify suspected childhood sexual abuse in pregnant teenagers. Furthermore, childbirth educators have the opportunity to intervene directly by identifying potentially unsafe home environments for teenage mothers and their infants and indirectly through sensitive and appropriate parenting educa- tion for expecting or new parents. These parenting classes can focus on establishing healthy family dynamics, including family control and organization. Finally, childbirth educators can provide resources and referrals for families in crisis and serve as an important first step in the process of healing for young victims of childhood sexual abuse.

References American Psychiatric Association. (2013). Diagnostic and Statistical Manual of Mental Disorders (5th ed). Washington, DC: Author.

Abramovich, E. (2005). Childhood sexual abuse as a risk factor for subse- quent involvement in sex work: A review of empirical findings. Journal of Psychology & Human Sexuality, 17(1/2), 131-146. doi:10.1300/J056v17n01_08

Bailey, H. N., Moran, G., & Pederson, D. R. (2007). Childhood maltreat- ment, complex trauma symptoms, and unresolved attachment in an at-risk sample of adolescent mothers. Attachment & Human Development, 9(2), 139-161. doi:10.1080/14616730701349721

Centers for Disease Control and Prevention. (2012). About teen pregnancy. Retrieved February 11, 2013, from http://www.cdc.gov/teenpregnancy/ aboutteenpreg.html

Chen, X. K., Wen, S. W., Fleming, N., Demissie, K., Rhoads, G. G., & Walker, M. (2007). Teenage pregnancy and adverse birth outcomes: A large population based retrospective cohort study. International Journal of Epidemi- ology, 36(2), 368-373.doi:10.1093/ije/dyl284

Cornelius, M. D., Goldschmidt, L., Willford, J. A., Leech, S. L., Larkby, C., & Day, N. L. (2009). Body size and intelligence in 6-year-olds: Are offspring of teenage mothers at risk? Maternal & Child Health Journal, 13(6), 847-856. doi:10.1007/s10995-008-0399-0

De Genna, N., Larkby, C., & Cornelius, M. (2011). Pubertal timing and early sexual intercourse in the offspring of teenage mothers. Journal of Youth & Adolescence, 40(10), 1315-1328. doi:10.1007/s10995-011-0750-8

Dittus, P. J., & Jaccard, J. (2000). Adolescents’ perceptions of maternal disapproval of sex: Relationship to sexual outcomes. Journal of Adolescent Health Care, 26(4), 268-278. doi: 10.1016/S1054-139X(99)00096-8

Finkelhor, D. (1988). The trauma of child sexual abuse: Two models. In G.E. Wyatt and G.J. Powell (Eds.), Lasting effects of child sexual abuse (pp. 61-82). Newbury Park, CA: Sage.

Francesconi, M. (2008). Adult outcomes for children of teenage mothers. Scandinavian Journal of Economics, 110(1), 93-117. doi:10.1111/j.1467- 9442.2008.00526.x

Francisco, M. A., Hicks, K., Powell, J., Styles, K., Tabor, J. L., & Hulton, L. J. (2008). The effect of childhood sexual abuse on adolescent pregnancy: An integrative research review. Journal for Specialists in Pediatric Nursing, 13(4), 237-248.

Furstenberg, F. F. (1992). Teenage childbearing and cultural rationality: A thesis in search of evidence. Family Relations, 41, 239-243.

Gottfried, A.W. (1985). Measures of socioeconomic status in child develop- ment research: Data and recommendations. Merrill-Palmer Quarterly, 31(1), 85-92.

Hamilton, B. E., Martin, J. A., & Ventura, S. J. (2013). Births: Preliminary data for 2012. Hyattsville, MD: National Center for Health Statistics. Re- trieved from http://www.cdc.gov/nchs/data/nvsr/nvsr62/nvsr62_03.pdf

Hillis, S. D., Anda, R. F., Dube, S. R., Felitti, V. J., Marchbanks, P. A., & Marks, J. S. (2004). The association between adverse childhood experiences and adolescent pregnancy, long-term psychosocial consequences, and fetal death. Pediatrics, 113(2), 320-327. doi:10.1542/peds.113.2.320

Hollingshead, A.B. (1975). Four factor index of social status. Unpublished Manuscript, Yale University, New Haven, CT.

Kramer, K. L., & Lancaster, J. B. (2010). Teen motherhood in cross- cultural perspective. Annals of Human Biology, 37(5), 613-628. doi:10.3109/03014460903563434

Levine, J. A., Emery, C. R., & Pollack, H. (2007). The well-being of chil- dren born to teen mothers. Journal of Marriage & Family, 69(1), 105-122. doi:10.1111/j.1741-3737.2006.00348.x

Mannarino, A .P., & Cohen, J. A. (1996). Family-related variables and psychological symptom formation in sexually abused girls. Journal of Child Sexual Abuse, 5(1), 105-120.

McClure, F. H., Chavez, D. V., Agars, M. D., Peacock, M. J., & Matosian, A. (2008). Resilience in sexually abused women: Risk and protective factors. Journal of Family Violence, 23, 81-88. doi:10.1007/s10896-007-9129-4

Miller, B. C. (2002). Family influences on adolescent sexual and contracep- tive behavior. The Journal of Sex Research, 39(1), 22-26.

Miller, B. C., Benson, B., & Galbraith, K. A. (2001). Family relationships and adolescent pregnancy risk: A research synthesis. Developmental Review, 21(1), 1-38. doi:10.1006/drev.2000.0513

Mitchell, M. L., & Jolley, J. M. (2004). Research Design Explained (5th ed.). Belmont, CA: Wadsworth/Thomson Learning.

Mollborn, S. (2010). Exploring variation in teenage mothers’ and fathers’ educational attainment. Perspectives on Sexual & Reproductive Health, 42(3), 152-159. doi:10.1363/4215210

Moos, R. H. & Moos, B. S. (1981). Family Environment Scale Manual. Palo Alto, CA: Consulting Psychological Press, Inc.

Noll, J. G. & Shenk, C. E. (2013). Teen birth rates in sexually abused and neglected females. Pediatrics, 131, e1181-e1187. doi: 10.1542/peds.2012-3072

Noll, J. G., Shenk, C. E., & Putnam, K. T. (2009). Childhood sexual abuse and adolescent pregnancy: A meta-analytic update. Journal of Pediatric Psychology, 34(4), 366-378. doi:10.1093/jpepsy/jsno98

Noll, J. G., Trickett, P. K., & Putnam, F. W. (2003). A prospective inves- tigation of the impact of childhood sexual abuse on the development of sexuality. Journal of Consulting and Clinical Psychology, 71(3), 575-586. doi: 10.1037/0022-006X.71.3.575

Office of Adolescent Health. (2013). Trends in teen pregnancy and child- bearing. Retrieved from http://www.hhs.gov/ash/oah/adolescent-health- topics/reproductive-health/teen-pregnancy/trends.html#_ftn5

Post Trauma Moderators continued from previous page

continued on next page

20 | International Journal of Childbirth Education | Volume 29 Number 1 January 2014

Pallitto, C. C., & Murillo, V. (2008). Childhood abuse as a risk factor for adolescent pregnancy in El Salvador. Journal of Adolescent Health, 42(6), 580- 586. doi: 10.1016/j.jadohealth.2007.11.148

Petersen, A. C., Schulenberg, J. E., Abramowitz, R. H., Offer, D., & Jarcho, H. D. (1984). A self-image questionnaire for young adolescents (SIQYA): Reliability and validity studies. Journal of Youth & Adolescence, 13(2), 93-111.

Quinlivan, J. A., Tan, L. H., Steel, A., & Black, K. (2004). Impact of demo- graphic factors, early family relationships and depressive symptomatology in teenage pregnancy. Australian and New Zealand Journal of Psychiatry, 38, 197-203.

Raj, A., Silverman, J. G., & Amaro, H. (2000). The relationship between sexual abuse and sexual risk among high school students: Findings from the 1997 Massachusetts Youth Risk Behavior survey. Maternal & Child Health Journal, 4(2), 125-134. doi:10.1023/A: 1009526422148

Roberts, R., O’Connor, T., Dunn, J., & Golding, J. (2004). The effects of child sexual abuse in later family life; mental health, parenting & adjust- ment of offspring. Child Abuse & Neglect, 28, 525-545.

Rocca, C. H., Doherty, I., Padian, N. S., Hubbard, A. E., & Minnis, A. M. (2010). Pregnancy intentions and teenage pregnancy among Latinas: A mediation analysis. Perspectives on Sexual and Reproductive Health, 42(3), 186- 196. http://dx.doi.org /10.1363/4218610

Rubin, D. B. (1987). Multiple imputation for nonresponse in surveys. New York, NY: Wiley.

Saewyc, E. M., Magee, L. L., & Pettingell, S. E. (2004). Teenage pregnancy and associated risk behaviors among sexually abused adolescents. Perspec- tives on Sexual and Reproductive Health, 36(3), 98-105. Retrieved from http:// www.guttmacher.org/archive/PSRH.jsp

Schafer, J. L., & Graham, J. W. (2002). Missing data: Our view of the state of the art. Psychological Methods, 7(2), 147-177. doi:10.1037/1082- 989X.7.2.147

Stevens-Simon, C., Kelly, L., Singer, D., & Cox, A. (1996). Why pregnant adolescents say they did not use contraceptives prior to conception. Journal of Adolescent Health, 19(1), 48-53.

Swanston, H. Y., Plunkett, A. M., O’Toole, B. I., Shrimpton, S., Parkinson, P. N., & Oates, R. K. (2003). Nine years after child sexual abuse. Child Abuse & Neglect, 27, 967-984. doi:10.1016/S0145-2134(03)00143-1

Trickett, P. K., Noll, J. G., & Putnam, F. W. (2011). The impact of sexual abuse on female development: Lessons from a multigenerational, longitudi- nal research study. Development and Psychopathology, 23, 101-124.

Tyano, S., Keren, M., Herrman, H., & Cox, J. (2010). Parenthood and mental health: A bridge between infant and adult psychiatry. West Sussex, England: Wiley-Blackwell. doi: 10.1002/9780470660683

World Health Organization. (2013). Adolescent pregnancy. Retrieved from http://www.who.int/maternal_child_adolescent/topics/maternal/adoles- cent_pregnancy/en/

Amy Sickel received her PhD in Applied Social Psychology from George Washington University and is currently faculty in the School of Psychology at Walden University. Her research interests include the intersection of social cognitive factors and health and her research experience includes: psychosocial functioning and mental health; the study of reproductive/health outcomes in

females and child abuse survivors; socioeconomic status/health; the doctor/patient relationship and health outcome; the psycho- social functioning/health care needs of aging and the impact of stigma on mental/physical health. Dana Dillard is working toward a PhD in Health Psychology through Walden University and serves as Associate Faculty with Ashford University. Dana is primarily interested in the intersec- tions of health and experiential stress, including abuse and trauma, and holistic health approaches to stress management. Penelope Trickett received her PhD in Psychology from the New School for Social Research and is presently the David Lawrence Stein/Violet Goldberg Sachs Professor of Mental Health, School of Social Work, and Professor of Psychology, Dornsife College of Letters, Arts, and Sciences, University of Southern California. She has earned multiple awards including: the Independent Scientist Award from the National Institute of Mental Health; the Sterling C. Franklin Faculty Award for Research and Scholarship; and the Alfred J. Marrow Award for outstanding dissertation in the field of psychology. Penelope’s research has focused on the developmental consequences of child abuse and/or neglect on children and ado- lescents and on the characteristics of families in which such abuse occurs and she has published extensively within her field. Frank W. Putnam received his MD from the Indiana University School of Medicine and is currently a Professor of Psychiatry at UNC School of Medicine. Frank did a residency in Psychiatry at Yale University and fellowships in Child & Adolescent Psychia- try and Clinical Research respectively, at George Washington University and the National Institute of Mental Health. He has conducted ground breaking research on the long-term sequelae of childhood sexual abuse with particular focus on the role of dissociation as a product of the abuse. He has published exten- sively including two books and over 200 articles and chapters and received Lifetime Achievement Awards from multiple profes- sional societies. Jennie Noll received her PhD in Developmental Psychology/Sta- tistical Methodology from the University of Southern California and is a Professor in the department of Human Development and Family Studies and Director, Network on Child Protec- tion and Wellbeing at Pennsylvania State University. Jennie has been the Principal Investigator on six NIH funded grants since 2002. Her primary research interests include: 1) bio-psycho-social consequences of childhood sexual abuse, 2) pathways to teen pregnancy/high-risk sexual behaviors for abused and neglected youth, 3) long-term adverse health outcomes for victims of sexual abuse and 4) high-risk internet and social media behaviors in abused/neglected teens.

Post Trauma Moderators continued from previous page

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.