OBJECTIVES:

The current study aimed to investigate the relationship between advanced pubertal development and adolescent dating abuse (ADA) and to test if this relationship is moderated by friendship group characteristics in a nationally representative sample of US girls.

METHODS:

Data were drawn from wave 1 and 2 (1995–1996) of the National Longitudinal Study of Adolescent to Adult Health. The sample included 3870 girls aged 13 to 17 years, all of whom were in romantic and/or nonromantic sexual relationships. Relative pubertal development was measured as perceived physical development as compared with peers of the same age and race and age at menarche at wave 1. Participants reported at wave 2 whether they had experienced any verbal or physical abuse in their relationships. Friendship group characteristics included the percentage of boy friends, older friends, and friends’ risk behavior level.

RESULTS:

Negative binomial regression analyses revealed an interaction between advanced pubertal development and percentage of boy friends on ADA victimization, adjusted for age, race, parents’ marital status, household income, number of relationships, self-esteem, self-control, and antisocial behavior history. Advanced pubertal development was associated with more ADA victimization when girls’ friendship groups comprised a higher percentage of boys.

CONCLUSIONS:

Findings highlight the importance of pubertal timing and friendship group characteristics to ADA victimization. Early pubertal development is a risk marker for ADA victimization, particularly when a higher percentage of girls’ friends are boys. Pediatricians and adolescent health specialists should be sensitive to the elevated risk for ADA victimization in early-maturing girls.

What’s Known on This Subject:

One previous study found that girls’ pubertal timing predicted risk for adolescent dating abuse (ADA). Under what contexts early-maturing girls may be at increased risk for ADA victimization is unknown.

What This Study Adds:

We report that when girls make an early transition to puberty, they are at heightened risk for ADA victimization if their friendship group includes a greater percentage of boys.

According to the 2013 Youth Risk Behavior Survey, 20% and 10% of high school-attending girls and boys, respectively, in the United States reported experiencing sexual or physical abuse by a dating partner in the past year.1 Other studies using nationally representative samples have also found that 31% to 40% of girls and 27% to 30% of boys are victimized by psychological aggression, physical aggression, or both by a dating partner during adolescence.2,3 Not only is adolescent dating abuse (ADA) relatively common, but youth who are victimized are subsequently at elevated risk for depression, anxiety, eating disorders, substance use,2,4 and suicidal behavior.2,5,6 Thus, ADA is a significant public health problem. One previous study has shown that girls who make an early transition to puberty are at elevated risk for ADA,7 but more research is needed to understand the conditions under which early-maturing girls are at heightened risk.

The current study is designed to inform prevention efforts by assessing the relationship between pubertal timing and ADA in a nationally representative sample of US girls. Our hypothesis was that early puberty would elevate the risk for subsequent ADA, especially when girls’ friendship groups included a high percentage of boys, older friends, and when friends participated in high levels of risk behavior.

Off-time pubertal development is defined as making the transition to puberty relatively earlier or later than one’s peers.8,10 The stage-termination hypothesis proposes that early-maturing individuals have less time to acquire, integrate, and consolidate the coping and adaptive skills needed to successfully master the developmental challenges associated with the transition to adolescence.10,12 Consistent with this hypothesis, early puberty has been associated with emotional and behavioral problems, including externalizing and internalizing problems, substance abuse, eating disorders, and poor academic achievement.13,14 

Girls who make an early transition to puberty may be at heightened risk for ADA. Girls who are physically mature tend to attract the attention of boys13,15 and girls who make an early transition to puberty are often sexually active at a younger age than later-maturers.15,16 However, there are relatively few data on the prevalence of ADA among early-maturing girls. Studying a closely related topic, Craig et al17 found that early-maturing adolescents were at higher risk for same-sex and opposite-sex sexual harassment and opposite-sex relational aggression victimization. Similarly, a study revealed that early maturation was associated with physical violence victimization for both boys and girls, although not necessarily in the context of dating relationships.18 In the only previous study focused specifically on ADA and pubertal timing, Foster et al7 found that adjusting for the impact of various covariates, early-maturating girls were at greater risk for ADA than their on-time and late-maturing peers. However, they did not test how the effect of pubertal timing varied as a function of adolescents’ social environment.

Although early-maturing girls may be at elevated risk for ADA, it is also true that not all early-maturing girls will be victimized.19 The contextual amplification hypothesis postulates that certain contextual conditions will accentuate the negative effect of early maturation.20 In the current study, we focused on peer context (ie, friendship) as a potential modifier of the association between early puberty and ADA, because peer context is highly salient in adolescence and relevant to dating, sexual, and romantic encounters. Adolescents spend increasingly more time with their friends than with their parents, and friends and romantic partners are influential and extremely important sources of social support.21 At least 3 aspects of the friendship group could increase the likelihood that girls will be exposed to ADA. First, girls with a higher percentage of boys in their friendship group may be more likely to be dating, which could potentially expose them to ADA. Second, older friends are also more likely to be involved in ADA,22 as victims or perpetrators, and having friends who are involved in ADA is a risk factor for one’s own ADA.23,24 Thus, as the proportion of older friends in the friendship groups increase, so may girls’ risk for ADA. Third, friends who engage in high levels of risk behaviors (eg, underage alcohol consumption, fighting) are likely to be involved in aggressive behavior and delinquency both inside and outside of dating relationships.6,25 

Compared with on-time and later-maturing girls, early-maturing girls may be more vulnerable to the effect of these friendship group characteristics26,27 because they will have had less time to develop coping skills10,12 and may have experienced a variety of hardships that limit their capacity to remove themselves from or resolve situations in sexual or romantic relationships that could become abusive. Thus, early-maturing girls may be at greater risk for ADA relative to later-maturing girls if their network includes a high percentage of boy friends, older friends, or friends who engage in risk behavior. The current study tests these hypotheses.

Participants were drawn from the in-home interview sample of the National Longitudinal Study of Adolescent to Adult Health (Add Health). Add Health is a longitudinal study of a nationally representative sample of adolescents from the United States who were in grades 7 through 12 in 1994 to 1995 when the study began. Our final analytic sample comprised 3870 girl participants aged 13 to 17 years self-identified as white (61%), African American (22.5%), and Hispanic (16.5%). Detailed information regarding how the final sample was derived can be found in in the Supplemental Information.

ADA

ADA was measured at the wave 2 in-home interview. Each participant identified up to 3 romantic relationships and 3 nonromantic relationships that involved sexual activity in the past 18 months and reported ADA victimization information (but not perpetration) in each relationship. Of the 3870 participants, all of whom had been in romantic and/or nonromantic sexual relationships, 9.15% (n = 354) were involved in both romantic and nonromantic sexual relationships, 89.17% (n = 3451) were involved only in romantic relationships, and 1.68% (n = 65) were involved only in nonromantic sexual relationships. Sixty-seven percent of the participants had only 1 relationship. For each relationship, participants reported whether their partners insulted them in public, swore at them, threatened them with violence, pushed/shoved them in public, or threw something at them (0 = no, 1 = yes). Responses were coded as missing if participants did not have a partner in the past 18 months or if they refused or did not know how to answer the question. The same 5 items have been used to measure ADA in previous studies.2,7 A participant was considered to have experienced a particular form of abuse if she reported experiencing that form of abuse ≥1 time in any relationship. Thus, across all relationships, participants could have experienced up to 5 forms of ADA. Thirty percent of participants reported that they experienced at least 1 form of ADA.

Relative Pubertal Development

Both a subjective and an objective measure of relative pubertal development were used. The subjective measure was a single item reported at wave 1 indicating how advanced participants thought their physical development was compared with other girls their age (1 = “I look younger than most” to 5 = "I look older than most”). One-third of the participants reported that they looked average compared with their peers. The objective measure of relative pubertal development was based on participants’ reports (from wave 1 or 2) of how old they were when they had their first menstrual period. Menarcheal age was standardized within participants of the same age and race. We multiplied this standardized score by the value of −1 so that higher scores indicated more advanced pubertal development compared with peers. We use the phrase “advanced pubertal development” to refer to girls who either had a relatively earlier age at menarche compared with peers of the same age and race or described themselves as being more physically developed compared with peers of the same age and race.

Friendship Group Characteristics

The Add Health study collected network data at wave 1 from all students who attended each participating school, enabling a comprehensive assessment of a participant’s friendship group. Only schools in which >50% of the student body completed the questionnaire (90% of all participating schools) were used to calculate the network measures. Each participant was asked to nominate up to 5 girl and 5 boy friends from the roster of all students enrolled in the participant’s school and/or in the sister school. Nominations were used in the current study if both the participant (ie, nominator) and the individuals named as the participant’s friends (ie, nominees) were uniquely identifiable students who completed an in-school questionnaire. In the current study, we used the ego-send network, which included all individuals the participant nominated as friends, even if these nominated friends did not nominate the participant as a friend in return. The use of peer network data allows us to determine features of the friendship group based on the nominated friends’ self-reports of behavior and demographic characteristics rather than on the participant’s perceptions of these characteristics.

We identified 3 characteristics of participants’ friendship groups. The first was the percentage of older friends, defined as the percentage of nominated friends who were at least 1 year older than the participant. The second was the percentage of boy friends, which was defined as the percentage of boy friends in a participant’s friendship group. The third was risk behaviors among friends, which was based on the nominated friends’ reports of how often (0 = never; 1 = once or twice; 6 = nearly every day) they engaged in 5 risk behaviors in the past 12 months including (1) smoking cigarettes, (2) drinking beer, wine, or liquor, (3) getting drunk, (4) lying to parents or guardians, (5) skipping school without an excuse, and how often (0 = never; 4 = >7 times) they were in physical fights. To adjust for the density of the friendship groups, the item score was divided by the number of friends nominated. Risk behavior items were z-transformed and then summed to represent the level of risk behavior among friends. The measure of risk behavior among friends had good internal consistency (Cronbach’s α = 0.79).

Covariates

We included a range of covariates in the analysis (Table 1). All covariates were drawn from the wave 1 in-home interview. Demographic covariates at wave 1 were age, participants’ race, whether parents were married, and the household income. We also included the number of relationships, self-esteem, self-control, and antisocial behavior at wave 1 because of their association with ADA in the literature.28,30 All but parents’ marital status was correlated with ADA, and all covariates were significantly correlated with at least 1 of the main variables of interest (ie, friendship group characteristics and relative pubertal development). Detailed information about the covariates are reported in the Supplemental Information.

TABLE 1

Bivariate Correlation Between Main Variables and Descriptive Statistics

12345678910111213
1. ADA             
2. Relative pubertal development (objective) 0.06**            
3. Relative pubertal development (subjective) 0.07** 0.25**           
4. Risk behavior among friends 0.11** 0.08** 0.07**          
5. Percentage of boy friends 0.03 0.05** 0.03 0.14**         
6. Percentage of older friends 0.06** 0.04* 0.05** 0.10** 0.18**        
7. No. relationships 0.24** 0.02 0.07** 0.03 0.04 0.01       
8. Age, W1 0.05** −0.001 −0.09** 0.19** −0.01 −0.12** −0.03      
9. Parents married, W1 −0.03 −0.04* 0.01 −0.04* 0.01 −0.02 0.02 0.01     
10. Household income, W1 ($1000) −0.04* −0.03 0.06** −0.04 0.01 −0.04 0.06** −0.03 0.42**    
11. Self-control, W1 −0.11** 0.01 −0.02 −0.09** 0.02 −0.01 −0.08** 0.04* −0.03 −0.03   
12. Self-esteem, W1 −0.11** −0.03 −0.03 −0.13** 0.01 −0.002 −0.05** −0.04** −0.01 −0.01 0.38**  
13. Antisocial behavior, W1 0.18** −0.03* 0.10** 0.19** −0.02 0.02 0.13** −0.04* −0.05** −0.03 −0.30** −0.26** 
Mean 0.57 12.05a 3.39 0.03 0.37 0.33 1.45 15.81 69.25% 0.02 15.79 46.18 2.36 
SD 1.06 1.31 1.07 4.13 0.24 0.29 0.71 1.28 — 3.94 2.69 49.47 2.19 
12345678910111213
1. ADA             
2. Relative pubertal development (objective) 0.06**            
3. Relative pubertal development (subjective) 0.07** 0.25**           
4. Risk behavior among friends 0.11** 0.08** 0.07**          
5. Percentage of boy friends 0.03 0.05** 0.03 0.14**         
6. Percentage of older friends 0.06** 0.04* 0.05** 0.10** 0.18**        
7. No. relationships 0.24** 0.02 0.07** 0.03 0.04 0.01       
8. Age, W1 0.05** −0.001 −0.09** 0.19** −0.01 −0.12** −0.03      
9. Parents married, W1 −0.03 −0.04* 0.01 −0.04* 0.01 −0.02 0.02 0.01     
10. Household income, W1 ($1000) −0.04* −0.03 0.06** −0.04 0.01 −0.04 0.06** −0.03 0.42**    
11. Self-control, W1 −0.11** 0.01 −0.02 −0.09** 0.02 −0.01 −0.08** 0.04* −0.03 −0.03   
12. Self-esteem, W1 −0.11** −0.03 −0.03 −0.13** 0.01 −0.002 −0.05** −0.04** −0.01 −0.01 0.38**  
13. Antisocial behavior, W1 0.18** −0.03* 0.10** 0.19** −0.02 0.02 0.13** −0.04* −0.05** −0.03 −0.30** −0.26** 
Mean 0.57 12.05a 3.39 0.03 0.37 0.33 1.45 15.81 69.25% 0.02 15.79 46.18 2.36 
SD 1.06 1.31 1.07 4.13 0.24 0.29 0.71 1.28 — 3.94 2.69 49.47 2.19 

W1, wave 1.

a

The mean age at menarche is reported, the relative pubertal development (objective) has a mean of 0 because of the standardization reported in the Methods section.

*

P < .05.

**

P < .01.

We used negative binomial regression to test the hypothesis that the association between advanced pubertal development and ADA at wave 2 would be strongest when the friendship groups included a greater percentage of older friends and boy friends and when friends engaged in more risk behaviors at wave 1. Friendship group characteristics and relative pubertal development were mean-centered to facilitate interpretation. First, we tested the main effects of advanced pubertal development and the 3 friendship group characteristics on ADA, adjusting for the effects of covariates. Next, we included the three 2-way interaction terms between relative pubertal development and each of the 3 friendship group characteristics. We ran the analyses separately for the objective and the subjective measure of relative pubertal development. Analyses were conducted in Mplus 631 by using maximum likelihood estimation with robust SE, and missing data were handled with the full information maximum likelihood method. Analyses were adjusted for the stratified and clustered sampling design of Add Health and incorporated sampling weights that adjusted for differential response and attrition rates over time.

Girls with more advanced pubertal development were more likely to be dating (see Supplemental Information) and reported more ADA (Table 1). The correlations between the 2 measures of relative pubertal timing and ADA were statistically significant but small in effect sizes (r = 0.06 and 0.07 for objective and subjective pubertal timing, respectively). Having more relationships, being older, living in a family with a lower household income, having lower self-esteem, engaging in more antisocial behaviors at wave 1, and being white versus Hispanic were associated with higher levels of ADA across all 4 models (Table 2). As seen in Table 2 (model 1), the main effects of the objective measure of relative pubertal development and risk behaviors among friends were statistically significant. One unit increase in relative pubertal development was associated with a 14% increase in ADA. In addition, 1 unit increase in risk behaviors among friends was associated with a 3% increase in ADA. A similar pattern of results emerged for the subjective measure of relative pubertal development (Table 2, model 3).

TABLE 2

Negative Binominal Regression Results for the Effects of Relative Pubertal Development and Friendship Group Characteristics on ADA (N = 3870)

Model 1Model 2Model 3Model 4
B (95% CI)B (95% CI)B (95% CI)B (95% CI)
Intercept −0.87*** (−0.19 to −0.65) −0.87*** (−1.09 to −0.65) −0.84*** (−1.06 to −0.62) −0.85*** (−1.07 to −0.63) 
Dispersion 1.56** (1.27 to 1.85) 1.54*** (1.25 to 1.83) 1.58** (1.29 to 1.87) 1.55*** (0.1.26 to 1.84) 
Covariates     
 No. relationships 0.45*** (0.37 to 0.53) 0.45*** (0.37 to 0.53) 0.45*** (0.37 to 0.53) 0.45*** (0.37 to 0.53) 
 Age, W1 0.07* (0.01 to 0.13) 0.07* (0.01 to 0.13) 0.08* (0.02 to 0.14) 0.08** (0.02 to 0.14) 
 Race (reference: white)     
  African American −0.12 (−0.32 to 0.08) −0.12 (−0.32 to 0.08) −0.11 (−0.31 to 0.09) −0.10 (−0.30 to 0.10) 
  Hispanic −0.22* (−0.42 to −0.02) −0.22* (−0.42 to −0.02) −0.23* (−0.43 to −0.03) −0.22* (−0.42 to −0.02) 
 Parents married, W1 0.07 (−0.17 to 0.31) 0.07 (−0.17 to 0.31) 0.05 (−0.19 to 0.29) 0.05 (−0.19 to 0.29) 
 Self-control, W1 −0.01 (−0.03 to 0.01) −0.01 (−0.03 to 0.01) −0.01 (−0.03 to 0.01) −0.01 (−0.03 to 0.01) 
 Self-esteem, W1 −0.53** (−0.88 to −0.18) −0.52** (−0.87 to −0.17) −0.54** (−0.89 to −0.19) −0.53** (−0.86 to −0.20) 
 Household income, W1 ($1000) −0.15** (−0.25 to −0.05) −0.15** (−0.25 to −0.05) −0.16** (−0.26 to −0.06) −0.15** (−0.25 to −0.05) 
 Antisocial behavior, W1 0.10*** (0.06 to 0.14) 0.07*** (0.03 to 0.11) 0.07*** (0.03 to 0.11) 0.07*** (0.03 to 0.11) 
Relative pubertal development     
 Objective measurea 0.13** (0.05 to 0.21) 0.12** (0.04 to 0.20) — — 
 Subjective measureb — — 0.09* (0.01 to 0.17) 0.09* (0.01 to 0.17) 
Friendship group characteristics     
 Risk behavior among friends 0.03** (0.01 to 0.05) 0.03** (0.01 to 0.05) 0.03* (0.01 to 0.05) 0.03** (0.01 to 0.05) 
 Proportion of boy friends 0.18 (−0.17 to 0.53) 0.17 (−0.18 to 0.52) 0.20 (−0.17 to 0.57) 0.20 (−0.17 to 0.57) 
 Proportion of older friends 0.27 (−0.02 to 0.56) 0.27 (−0.02 to 0.56) 0.23 (−0.06 to 0.52) 0.22 (−0.07 to 0.51) 
Interaction     
 Risk behavior among friends by relative pubertal development — −0.007 (−0.03 to 0.01) — −0.001 (−0.02 to 0.02) 
 Proportion of boy friends by relative pubertal development — 0.39* (0.08 to 0.70) — 0.40* (0.09 to 0.71) 
 Proportion of older friends by relative pubertal development — −0.22 (−0.49 to 0.05) — −0.04 (−0.31 to 0.23) 
Model 1Model 2Model 3Model 4
B (95% CI)B (95% CI)B (95% CI)B (95% CI)
Intercept −0.87*** (−0.19 to −0.65) −0.87*** (−1.09 to −0.65) −0.84*** (−1.06 to −0.62) −0.85*** (−1.07 to −0.63) 
Dispersion 1.56** (1.27 to 1.85) 1.54*** (1.25 to 1.83) 1.58** (1.29 to 1.87) 1.55*** (0.1.26 to 1.84) 
Covariates     
 No. relationships 0.45*** (0.37 to 0.53) 0.45*** (0.37 to 0.53) 0.45*** (0.37 to 0.53) 0.45*** (0.37 to 0.53) 
 Age, W1 0.07* (0.01 to 0.13) 0.07* (0.01 to 0.13) 0.08* (0.02 to 0.14) 0.08** (0.02 to 0.14) 
 Race (reference: white)     
  African American −0.12 (−0.32 to 0.08) −0.12 (−0.32 to 0.08) −0.11 (−0.31 to 0.09) −0.10 (−0.30 to 0.10) 
  Hispanic −0.22* (−0.42 to −0.02) −0.22* (−0.42 to −0.02) −0.23* (−0.43 to −0.03) −0.22* (−0.42 to −0.02) 
 Parents married, W1 0.07 (−0.17 to 0.31) 0.07 (−0.17 to 0.31) 0.05 (−0.19 to 0.29) 0.05 (−0.19 to 0.29) 
 Self-control, W1 −0.01 (−0.03 to 0.01) −0.01 (−0.03 to 0.01) −0.01 (−0.03 to 0.01) −0.01 (−0.03 to 0.01) 
 Self-esteem, W1 −0.53** (−0.88 to −0.18) −0.52** (−0.87 to −0.17) −0.54** (−0.89 to −0.19) −0.53** (−0.86 to −0.20) 
 Household income, W1 ($1000) −0.15** (−0.25 to −0.05) −0.15** (−0.25 to −0.05) −0.16** (−0.26 to −0.06) −0.15** (−0.25 to −0.05) 
 Antisocial behavior, W1 0.10*** (0.06 to 0.14) 0.07*** (0.03 to 0.11) 0.07*** (0.03 to 0.11) 0.07*** (0.03 to 0.11) 
Relative pubertal development     
 Objective measurea 0.13** (0.05 to 0.21) 0.12** (0.04 to 0.20) — — 
 Subjective measureb — — 0.09* (0.01 to 0.17) 0.09* (0.01 to 0.17) 
Friendship group characteristics     
 Risk behavior among friends 0.03** (0.01 to 0.05) 0.03** (0.01 to 0.05) 0.03* (0.01 to 0.05) 0.03** (0.01 to 0.05) 
 Proportion of boy friends 0.18 (−0.17 to 0.53) 0.17 (−0.18 to 0.52) 0.20 (−0.17 to 0.57) 0.20 (−0.17 to 0.57) 
 Proportion of older friends 0.27 (−0.02 to 0.56) 0.27 (−0.02 to 0.56) 0.23 (−0.06 to 0.52) 0.22 (−0.07 to 0.51) 
Interaction     
 Risk behavior among friends by relative pubertal development — −0.007 (−0.03 to 0.01) — −0.001 (−0.02 to 0.02) 
 Proportion of boy friends by relative pubertal development — 0.39* (0.08 to 0.70) — 0.40* (0.09 to 0.71) 
 Proportion of older friends by relative pubertal development — −0.22 (−0.49 to 0.05) — −0.04 (−0.31 to 0.23) 

CI, confidence interval; W1, wave 1; B, unstandardized coefficient; —, variable not included in the model.

a

Age at menarche was standardized within participants of the same age and race and then multiplied by the value of −1 so that higher scores indicated more advanced pubertal development compared with peers.

b

A 1-item measure (“How advanced is your physical development compared to other girls your age?”).

*

P < .05.

**

P < .01.

***

P < .001.

The effects of the objective and subjective measures of relative pubertal development were moderated by the percentage of boys in girls’ friendship groups (Table 2, model 2 and 4). Figure 1A reveals the region of significance for the slope of the objective measure of relative pubertal development on ADA. When the proportion of boys in a friendship group was <29%, the association between relative menarcheal age and ADA was not statistically significant. However, when the percentage of boys in a girls’ friendship group was ≥29%, having a relatively earlier menarcheal age was associated with ADA, and the magnitude of this association increased as the percentage of boy friends increased. Figure 2A illustrates the simple slope effects. Results involving the subjective measure of relative pubertal development were similar to those involving the objective measure (Table 2, model 4; Figs 1B and 2B).

FIGURE 1

Region of significance for the slope of relative pubertal development on ADA. A, Objective measure of relative pubertal development. B, Subjective measure of pubertal development.

FIGURE 1

Region of significance for the slope of relative pubertal development on ADA. A, Objective measure of relative pubertal development. B, Subjective measure of pubertal development.

FIGURE 2

The effects of proportion of boy friends and relative pubertal development on ADA. A, Objective measure of relative pubertal development. B, Subjective measure of pubertal development. * P < .05.

FIGURE 2

The effects of proportion of boy friends and relative pubertal development on ADA. A, Objective measure of relative pubertal development. B, Subjective measure of pubertal development. * P < .05.

To test the robustness of our findings, we repeated analyses using an alternative ADA score that summed across adolescents’ reports of all forms of ADA across all relationships. Negative binomial regression revealed similar results to those reported in Table 2 when using this alternative ADA score. The observed relationship between pubertal timing and ADA did not differ significantly between younger and older girls (see Supplemental Information).

Using a nationally representative sample of US girls, we found that girls who made an early transition to puberty relative to their peers of the same age and race were at elevated risk for ADA. Early-maturing girls whose friendship groups included a greater percentage of boys were at exacerbated risk, even accounting for the fact that girls who were relatively more physically advanced had a higher number of romantic or sexual relationships than girls who were less physically advanced. Findings were replicated with both objective and subjective ratings of relative pubertal development, the first reflecting girls’ age at menarche, and the second reflecting their perceptions of their physical development relative to girls their age and race.

One reason that having a greater percentage of boys in a friendship group may specifically increase risk for ADA for early-maturing girls is because early-maturing girls are more likely to be in romantic, dating, or sexual relationships with their friends than later-maturing girls are.15 Boys may find early-maturing girls attractive because they are more physically developed than their on-time/later-maturing peers15,32,33 and early-maturing girls may be more likely than their later-maturing peers to seek out boy friends to develop romantic or sexual relationships because of their relative sexual maturity. Thus, early-maturing girls are at heightened risk for ADA than their later-maturing peers because of the higher exposure. Second, compared with on-time or late-maturing girls, those who make an early transition to adolescence engage in higher levels of externalizing behavior.34 The spectrum of risky behaviors that includes externalizing problems is hypothesized to result from the maturational gap between the brain’s reward system (which is activated by pubertal hormones35) and the cognitive control system (which is not fully mature until individuals are in their mid-to-late 20s).36,37 Thus, early maturing girls may be at increased risk for ADA when their friendship group comprises a higher percentage of boys not only because they have more opportunities to be in romantic or sexual relationships, but also because they may perpetrate abuse that subsequently elevates the chances that they will be victimized.38 

Risk behaviors among friends did not moderate the association between relative pubertal development and ADA, but it had a statistically significant main effect. Individuals who engage in risk behaviors are at elevated risk for perpetrating ADA.6,39 Adolescents with many peers who engage in delinquent and aggressive behavior might perceive such behavior, including ADA, as normative. If girls are dating these risk-taking friends, they would be at elevated risk for ADA. Thus, risk behaviors among friends may increase girls’ risk of ADA in general, and on-time/late-maturing girls were just as susceptible to this impact as their early-maturing counterparts.

The current study highlighted the importance of early puberty and friendship group characteristics in ADA victimization. A recent study showed that only 55% of parents talked to their children about dating abuse, and that parents were significantly less likely to discuss dating abuse than other sensitive topics, such as drugs or family finance.40 Our findings suggest that parents need to talk to their children about ADA before puberty and that they should be aware of who their children’s friends are. In addition, schools can provide developmentally appropriate healthy dating curricula for elementary and middle school, so that girls who make an early transition to puberty are better prepared to differentiate abusive dating behaviors from normative dating behaviors. Finally, pediatricians can also play an active role in ADA prevention/intervention by being aware of the risks that early-maturing girls face and being alert for signs of ADA.

Limitations of this study include the fact that pubertal timing was assessed by using self-report. The measure of pubertal timing could be improved by having a trained third party (eg, nurses) assess pubertal development according to the Tanner stages.41 Nevertheless, the current findings held across an objective (ie, age at menarche) and a subjective measure of relative pubertal development, bolstering confidence in the results. Second, ADA was assessed in up to 3 romantic relationships and up to 3 nonromantic sexual relationships, which means that ADA could have been underdetected if some girls had abusive experiences in a seventh or eighth relationship that was not assessed. This seems unlikely, however, because none of the participants reported having had 6 relationships, only 1 reported having had 5 relationships, and 67% reported that they had only 1 romantic or sexual relationship. Third, the data analyzed in this study were from 20 years ago. The reason is that Add Health is the only study that collected peer network data as part of a nationally representative sample; there are not more recent data we could use.

ADA is prevalent and consequential. In recent years, more prevention programs have been tested for middle school youth,42,43 because it has become clear that ADA is already prevalent by the time youth reach high school.44,45 However, the effects of these prevention programs are typically small and difficult to sustain. Biological indicators that a girl may be at increased risk for ADA are important for the field because they give pediatricians, parents, and others who work with youth a marker that should prompt extra conversation about the importance of healthy relationships. Our study provides 1 marker for increased risk for ADA among US girls that should be included in lists of risk factors and educational materials. Adolescents and their parents should be reassured that only some girls who make an early transition to puberty (32% as compared with 28% of the on-time and 27% of the late-maturing girls based on the subjective rating of pubertal timing; see Supplemental Information) experience dating abuse, but that it is nevertheless worthwhile to be vigilant with this group, particularly when these girls have a greater percentage of boys in their friendship groups. Continued study of the mechanisms underlying the linkage of early puberty, boy friends, and ADA may additionally enhance prevention efforts.

     
  • ADA

    adolescent dating abuse

  •  
  • Add Health

    National Longitudinal Study of Adolescent to Adult Health

Dr Chen conceptualized the study, carried out data analysis, and drafted the initial manuscript; Dr Rothman interpreted the results, edited the manuscript, and critically reviewed the manuscript; Dr Jaffee conceptualized the study, contributed to data analysis, edited the initial manuscript, and critically reviewed the manuscript; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

FUNDING: No external funding.

This research uses data from Add Health, a program project directed by Kathleen Mullan Harris and designed by J. Richard Udry, Peter S. Bearman, and Kathleen Mullan Harris at the University of North Carolina at Chapel Hill, and funded by grant P01-HD31921 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, with cooperative funding from 23 other federal agencies and foundations. We thank Ronald R. Rindfuss and Barbara Entwisle for assistance in the original design. Information on how to obtain the Add Health data files is available on the Add Health Web site (www.cpc.unc.edu/addhealth). No direct support was received from grant P01-HD31921 for this analysis.

1
Vagi
KJ
,
O’Malley Olsen
E
,
Basile
KC
,
Vivolo-Kantor
AM
.
Teen dating violence (physical and sexual) among us high school students: findings from the 2013 national youth risk behavior survey.
JAMA Pediatr
.
2015
;
169
(
5
):
474
482
[PubMed]
2
Exner-Cortens
D
,
Eckenrode
J
,
Rothman
E.
Longitudinal associations between teen dating violence victimization and adverse health outcomes.
Pediatrics
.
2013
;
131
(
1
):
71
78
3
Haynie
DL
,
Farhat
T
,
Brooks-Russell
A
,
Wang
J
,
Barbieri
B
,
Iannotti
RJ
.
Dating violence perpetration and victimization among U.S. adolescents: prevalence, patterns, and associations with health complaints and substance use.
J Adolesc Health
.
2013
;
53
(
2
):
194
201
[PubMed]
4
Foshee
VA
,
Reyes
HLM
,
Gottfredson
NC
,
Chang
L-Y
,
Ennett
ST
.
A longitudinal examination of psychological, behavioral, academic, and relationship consequences of dating abuse victimization among a primarily rural sample of adolescents.
J Adolesc Health
.
2013
;
53
(
6
):
723
729
[PubMed]
5
Ackard
DM
,
Eisenberg
ME
,
Neumark-Sztainer
D
.
Long-term impact of adolescent dating violence on the behavioral and psychological health of male and female youth.
J Pediatr
.
2007
;
151
(
5
):
476
481
[PubMed]
6
Bossarte
RM
,
Simon
TR
,
Swahn
MH
.
Clustering of adolescent dating violence, peer violence, and suicidal behavior.
J Interpers Violence
.
2008
;
23
(
6
):
815
833
[PubMed]
7
Foster
H
,
Hagan
J
,
Brooks-Gunn
J
.
Age, puberty, and exposure to intimate partner violence in adolescence.
Ann N Y Acad Sci
.
2004
;
1036
(
1
):
151
166
[PubMed]
8
Dorn
LD
,
Nottelmann
ED
,
Susman
EJ
,
Inoff-Germain
G
,
Cutler
GB
 Jr
,
Chrousos
GP
.
Variability in hormone concentrations and self-reported menstrual histories in young adolescents: menarche as an integral part of a developmental process.
J Youth Adolesc
.
1999
;
28
(
3
):
283
304
9
Graber
JA
,
Seeley
JR
,
Brooks-Gunn
J
,
Lewinsohn
PM
.
Is pubertal timing associated with psychopathology in young adulthood.
J Am Acad Child Adolesc Psychiatry
.
2004
;
43
(
6
):
718
726
[PubMed]
10
Peskin
H
.
Influence of the developmental schedule of puberty on learning and ego functioning.
J Youth Adolesc
.
1973
;
2
(
4
):
273
290
[PubMed]
11
Adelson
J
.
Handbook of Adolescent Psychology
.
New York, NY
:
Wiley
;
1980
12
Brooks-Gunn
J
,
Petersen
AC
,
Eichorn
D
.
The study of maturational timing effects in adolescence.
J Youth Adolesc
.
1985
;
14
(
3
):
149
161
[PubMed]
13
Mendle
J
,
Turkheimer
E
,
Emery
RE
.
Detrimental psychological outcomes associated with early pubertal timing in adolescent girls.
Dev Rev
.
2007
;
27
(
2
):
151
171
[PubMed]
14
Mrug
S
,
Elliott
MN
,
Davies
S
,
Tortolero
SR
,
Cuccaro
P
,
Schuster
MA
.
Early puberty, negative peer influence, and problem behaviors in adolescent girls.
Pediatrics
.
2014
;
133
(
1
):
7
14
[PubMed]
15
Cavanagh
SE
.
The sexual debut of girls in early adolescence: the intersection of race, pubertal timing, and friendship group characteristics.
J Res Adolesc
.
2004
;
14
(
3
):
285
312
16
Negriff
S
,
Susman
EJ
,
Trickett
PK
.
The developmental pathway from pubertal timing to delinquency and sexual activity from early to late adolescence.
J Youth Adolesc
.
2011
;
40
(
10
):
1343
1356
[PubMed]
17
Craig
WM
,
Pepler
D
,
Connolly
J
,
Henderson
K
. Development context of peer harassment in early adolescent: the role of puberty and the peer group. In:
Juvonen
J
,
Graham
S
, eds.
Peer Harassment in Schools: The Plight of the Vulnerable and Victimized
.
New York, NY
:
Guilford Press
;
2001
:
242
262
18
Haynie
DL
,
Piquero
AR
.
Pubertal development and physical victimization in adolescence.
J Res Crime Delinq
.
2006
;
43
(
1
):
3
35
19
Skoog
T
,
Stattin
H
.
Why and under what contextual conditions do early-maturing girls develop problem behaviors?
Child Dev Perspect
.
2014
;
8
(
3
):
158
162
20
Ge
X
,
Brody
GH
,
Conger
RD
,
Simons
RL
,
Murry
VM
.
Contextual amplification of pubertal transition effects on deviant peer affiliation and externalizing behavior among African American children.
Dev Psychol
.
2002
;
38
(
1
):
42
54
[PubMed]
21
Furman
W
,
McDunn
C
,
Young
B
. The role of peer and romantic relationships in adolescent affective development. In:
Allen
NB
,
Sheeber
L
, eds.
Adolescent Emotional Development and the Emergence of Depressive Disorders
.
New York, NY
:
Guilford Press
;
2008
:
299
317
22
Rennison
CM
,
Welchans
S
. Intimate partner violence. Bureau of Justice Statistics Special Report. NCJ 178247. May 2000. Available at: www.popcenter.org/problems/domestic_violence/PDFs/Rennison%26Welchans_2000.pdf. Accessed June 22, 2016
23
Vagi
KJ
,
Rothman
EF
,
Latzman
NE
,
Tharp
AT
,
Hall
DM
,
Breiding
MJ
.
Beyond correlates: a review of risk and protective factors for adolescent dating violence perpetration.
J Youth Adolesc
.
2013
;
42
(
4
):
633
649
[PubMed]
24
Foshee
VA
,
Reyes
HLM
,
Ennett
ST
.
Examination of sex and race differences in longitudinal predictors of the initiation of adolescent dating violence perpetration.
J Aggress Maltreat Trauma
.
2010
;
19
(
5
):
492
516
[PubMed]
25
Felson
RB
,
Lane
KJ
.
Does violence involving women and intimate partners have a special etiology?*.
Criminology
.
2010
;
48
(
1
):
321
338
26
Burt
SA
,
McGue
M
,
DeMarte
JA
,
Krueger
RF
,
Iacono
WG
.
Timing of menarche and the origins of conduct disorder.
Arch Gen Psychiatry
.
2006
;
63
(
8
):
890
896
[PubMed]
27
Harden
KP
,
Mendle
J
.
Gene-environment interplay in the association between pubertal timing and delinquency in adolescent girls.
J Abnorm Psychol
.
2012
;
121
(
1
):
73
87
[PubMed]
28
Ehrensaft
MK
,
Cohen
P
,
Brown
J
,
Smailes
E
,
Chen
H
,
Johnson
JG
.
Intergenerational transmission of partner violence: a 20-year prospective study.
J Consult Clin Psychol
.
2003
;
71
(
4
):
741
753
[PubMed]
29
Foshee
VA
,
Benefield
TS
,
Ennett
ST
,
Bauman
KE
,
Suchindran
C
.
Longitudinal predictors of serious physical and sexual dating violence victimization during adolescence.
Prev Med
.
2004
;
39
(
5
):
1007
1016
[PubMed]
30
Vézina
J
,
Hébert
M
.
Risk factors for victimization in romantic relationships of young women: a review of empirical studies and implications for prevention.
Trauma Violence Abuse
.
2007
;
8
(
1
):
33
66
[PubMed]
31
Muthén
LK
,
Muthén
BO
.
Mplus User’s Guide.
6th ed.
Los Angeles, CA
:
Muthén & Muthén
;
1998
32
Flannery
DJ
,
Rowe
DC
,
Gulley
BL
.
Impact of pubertal status, timing, and age on adolescent sexual experience and delinquency.
J Adolesc Res
.
1993
;
8
(
1
):
21
40
33
Skoog
T
,
Özdemir
SB
.
Explaining why early-maturing girls are more exposed to sexual harassment in early adolescence.
J Early Adolesc
.
2016
;
36
(
4
)
490
509
34
Lynne
SD
,
Graber
JA
,
Nichols
TR
,
Brooks-Gunn
J
,
Botvin
GJ
.
Links between pubertal timing, peer influences, and externalizing behaviors among urban students followed through middle school.
J Adolesc Health
.
2007
;
40
(
2
):
181.e7
181.e13
[PubMed]
35
Sisk
CL
,
Zehr
JL
.
Pubertal hormones organize the adolescent brain and behavior.
Front Neuroendocrinol
.
2005
;
26
(
3–4
):
163
174
[PubMed]
36
Steinberg
L
.
A social neuroscience perspective on adolescent risk-taking.
Dev Rev
.
2008
;
28
(
1
):
78
106
[PubMed]
37
Paus
T
.
Mapping brain maturation and cognitive development during adolescence.
Trends Cogn Sci
.
2005
;
9
(
2
):
60
68
[PubMed]
38
Jennings
WG
,
Higgins
GE
,
Tewksbury
R
,
Gover
AR
,
Piquero
AR
.
A longitudinal assessment of the victim-offender overlap.
J Interpers Violence
.
2010
;
25
(
12
):
2147
2174
[PubMed]
39
Johnson
WL
,
Giordano
PC
,
Manning
WD
,
Longmore
MA
.
The age-IPV curve: changes in the perpetration of intimate partner violence during adolescence and young adulthood.
J Youth Adolesc
.
2015
;
44
(
3
):
708
726
[PubMed]
40
Rothman
EF
,
Miller
E
,
Terpeluk
A
,
Glauber
A
,
Randel
J
.
The proportion of U.S. parents who talk with their adolescent children about dating abuse.
J Adolesc Health
.
2011
;
49
(
2
):
216
218
[PubMed]
41
Morris
NM
,
Udry
JR
.
Validation of a self-administered instrument to assess stage of adolescent development.
J Youth Adolesc
.
1980
;
9
(
3
):
271
280
[PubMed]
42
Foshee
VA
,
Bauman
KE
,
Ennett
ST
,
Linder
GF
,
Benefield
T
,
Suchindran
C
.
Assessing the long-term effects of the Safe Dates program and a booster in preventing and reducing adolescent dating violence victimization and perpetration.
Am J Public Health
.
2004
;
94
(
4
):
619
624
[PubMed]
43
Taylor
BG
,
Stein
ND
,
Mumford
EA
,
Woods
D
.
Shifting Boundaries: an experimental evaluation of a dating violence prevention program in middle schools.
Prev Sci
.
2013
;
14
(
1
):
64
76
[PubMed]
44
Avery-Leaf
S
,
Cascardi
M
,
O’Leary
KD
,
Cano
A
.
Efficacy of a dating violence prevention program on attitudes justifying aggression.
J Adolesc Health
.
1997
;
21
(
1
):
11
17
[PubMed]
45
Lavoie
F
,
Vézina
L
,
Piché
C
,
Boivin
M
.
Evaluation of a prevention program for violence in teen dating relationships.
J Interpers Violence
.
1995
;
10
(
4
):
516
524

Competing Interests

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.

Supplementary data