Aim: The aim of the present study was to identify influencing factors for sexual violence in sexually experienced teenage girls.
Methods: Eighty sexually experienced teenage girls aged between 14 and 19 attending obstetrics and gynecology clinics in suburban Tokyo were selected as subjects. After completing a simple questionnaire on sexual behavior and lifestyle the subjects were interviewed using a semistructured interview technique. Logistic regression analysis was used to identify factors associated with sexual violence.
Results: Police and government statistics for rape, sexual and domestic violence in Japan are extremely low. However, the present study showed that 52.5% of subjects had been victims of sexual violence. Furthermore, of these, 69.0% had been forced to have sex with a partner and 64.3% had been raped by someone other than a partner. Factors influencing the risk of sexual violence were sexual behavior and attitudes towards sexuality and lifestyle. Having a sexually transmitted infection was not an influencing factor.
Conclusions: It is important to identify teenage girls attending obstetrics and gynecology clinics at risk of sexual violence and to teach them about the risks of sexual behavior associated with imbalances of power. Furthermore, it is essential to support victims of sexual violence to protect them from the recurrence of such abuse. (Reprod Med Biol 2008; 7: 181–187)
ACCORDING TO JAPANESE Police statistics,1 there were 2076 reported rapes and 8751 reported cases of indecent assault in 2005, with 55.5 and 59.5% of the victims under 20 years of age, respectively. Furthermore, according to a study by the Japanese Cabinet Office on violence towards women,2 incidences of spousal violence and marital rape were reported to be 3.8 and 2.9% for ‘multiple assaults’ and 17.1 and 7.0% for ‘once or twice’, respectively.
In Japan, crimes such as child molestation, rape and domestic violence (DV) are likely to remain unreported. This is because victims and their families want to ‘cover them up’ or feel too ‘ashamed’ to talk about them. However, unwanted sexual experiences, such as sexual abuse, rape and molestation during childhood, can affect victims both mentally and physically in later life. Compared with women who have no history of sexual abuse, those who do tend to have a higher rate of: early age of sexual debut, sexual experiences, multiple pregnancies, unprotected sex, multiple partners, sexual anxiety, antisocial behavior, drug and/or alcohol abuse, as well as running away from home or getting involved in prostitution.3–8
Sexual abuse/violence almost always results in unprotected sex and is likely to lead to further sexual violence, unwanted pregnancies and an increased risk of sexually transmitted infection (STI). To prevent sexual violence, including child molestation, it is essential to understand the factors that result in teenage girls becoming victims of sexual violence, and to implement an appropriate support system for these victims. Therefore, the aim of the present study was to identify influencing factors for sexual violence in sexually experienced teenage girls.
MATERIALS AND METHODS
AFTER OBTAINING WRITTEN informed consent, 85 sexually experienced teenage girls attending obstetrics and gynecology clinics in suburban metropolitan areas of Japan from 30 July 2004 to 9 July 2005 were recruited as subjects. The majority of these girls had come to the clinics for the treatment of STI (33.8%), diagnosis of pregnancy/antenatal care (28.8%) or termination of pregnancy (20%). Regarding obstetric history, although 46 (57.5%) of the participants had been pregnant before, only six (13.0%) had actually given birth.
The mean age of the participants was 17.8 years at the time of the study (ranging from 14 to 19 years). The most common ‘occupation’ was student (junior high school, high school, vocational school, university) (42.5%), followed by unemployed or part-time worker (36.3%), full-time worker (11.3%) and full-time homemaker (10.0%). All full-time homemakers (n = 8) had married because of pregnancies. At the time of pregnancy these participants were either high school students (n = 3), unemployed after graduation from junior high school or high school drop-outs (n = 4) or vocational school students (n = 1).
After administering a simple questionnaire, examiners interviewed the subjects using a semistructured interview technique. After reassuring the participants that neither their parents nor their school would be informed of the answers, the examiners went over each question again with the participants. Often additional information was obtained or answers were changed so the examiners added to the original answers or modified them when necessary. Items investigated in the survey included questions on sexual behavior, attitudes towards sexuality, lifestyle, reproductive history and STI history.
Definitions of the terms used in this paper are explained as follows:
Sexual violence: forcing a person verbally, physically or by other means to have sex against their will. This includes DV and rape.
DV: forced sex by a former or current partner.
Rape: forced sex by a person (stranger, friend, acquaintance or family member) other than a partner. This includes attempted rape where penetration did not take place.
Of the 85 teenage girls who completed the questionnaire and interview, 80 subjects who provided complete data were selected for analysis. The participants were divided into two groups: those who had experienced sexual violence and those who had not. Statistical analysis was carried out using SPSS software for Windows (Ver.12.0; SPSS Japan, Tokyo, Japan). A Mann–Whitney U-test was used to compare mean values for each group with regard to sexual behavior, attitudes towards sexuality, lifestyle, reproductive history and STI history, and a Fisher's exact test was used to analyze the ratio. In addition, a multiple logistic regression analysis was conducted using SPSS software for Windows (Ver.11.5) and the odds ratio was obtained for the independent effects of each factor on sexual violence. Factors with a high correlation among the components of the regression analysis were excluded, and the absence of multicollinearity was confirmed. P-values < 0.05 in a two-sided test were considered to be statistically significant.
The content and purpose of the study was explained to each subject and written informed consent was obtained from all girls enrolled in the study. For girls under 14 years of age, consent from a parent was also obtained. Participants were guaranteed that refusal to participate in the study would not result in any clinical disadvantage and that they could withdraw from participation at any time during the study. This study was approved by the Tsukuba University ‘Ethical Review Board’ (Notification no. 354, 2004).
History of sexual violence
INCLUDING THOSE PARTICIPANTS who had experienced molestation during childhood or forced sexual intercourse with verbal and physical violence, 52.5% of the girls had experienced sexual violence (Table 1). Of these subects, 69.0% had been victims of DV and 64.3% had been raped. The most common rapists were strangers (70.4%), followed by friends, acquaintances and family members/relatives in that order.
Table 1. Experience of sexual violence in the 80 study participants
|Details of sexual violence (n = 42)|
| Domestic violence||29||69.0|
|Rapist (n = 27)|
| Friends|| 5||18.5|
| Acquaintances|| 4||14.8|
| Family members/relatives|| 1|| 3.7|
The most common age at first intercourse was 15 years (33.8%), followed by 14 years (22.5%) and 16 years (6.3%). Half of the subjects had experienced sex within 2 months of meeting their first sexual partner: The period from first meeting to sex for the first time was highest for ‘1 week to 1 month’ (28.8%), with ‘within 1 week’ and ‘1 to 2 months’, 8.8% and 12.5%, respectively. The most common reason for having sex was ‘love’ (72.5%), followed by ‘curiosity’ (20.0%), while six girls (7.5%) responded ‘my partner forced me’, and one each (1.3%) responded ‘I wanted money’ or ‘I was raped’.
The most common number of sexual partners during the 3 months leading up to the interview was 1 (73.8%), followed by 0 (11.3%) and 2 (8.8%), which accounted for more than 90% of the participants. Three girls responded that they had had more than five partners. In more than half of the cases (51.3%), the age of the current partner was ‘over 20 years’, while ‘under 19 years’ was 32.5%. In several cases the age of the partner was ‘unknown’ (16.3%) because of prostitution or one-night stands. The most common dating period with the current main sexual partner was ‘within 3 to 6 months’ (18.8%). The responses ‘within 1 year’ and ‘more than 1 year’ were 48.8% and 31.3%, respectively, while 8.8% responded ‘casual sex only’.
Regarding the total number of sexual partners, less than five accounted for just over half of the responses to this question, with 20.0% having only one and 8.8% having four. The most common overall age of the sexual partners was ‘between 20 and 29 years’ (55%), while ‘over 30 years’ accounted for 11%. Only 33.8% of participants had had a sexual partner of a similar age, that is, ‘under 19 years’. The percentage of girls who had a steady sexual partner, but also had sex with another man was 37.5%.
Attitudes towards sexuality
General ideas about sex included ‘it is acceptable (or permissible) to have sex for money or gifts’ (22.5%), ‘it is acceptable (or permissible) to have sex by offering money or gifts’ (17.5%), ‘it is acceptable (or permissible) to have sex with more than one partner for a certain period’ (22.5%), and ‘it is acceptable (or permissible) to have sex with a man whom you have met for the first time’ (31.3%).
The percentage of participants who had three meals a day was 45.7, while 7.5% had only one meal. Breakfast was eaten everyday by 38.8% of participants, while 18.8% never ate breakfast. The number of subjects who never ate out of the home in a week was 14 (17%). The percentage of participants who did not eat at home more than 5 days a week was 13.8, while 17% never ate out.
With regard to smoking, 71.3% responded that they had smoked. Of these smokers, 63.2% still smoked. For 26.3%, the average number of cigarettes per day was between 10 and 19, while 22.8% smoked more than 20 cigarettes a day. Of the smokers, 3.5% responded that they had started smoking before entering primary school, while 29.8% started at the age of 14.
Almost all of the girls (98.8%) had a drinking habit. Of these, 5.1% reported drinking ‘almost every day’ and 73.8%‘mainly weekends/sometimes’, while 18.8% responded ‘once or twice a month for social drinking only’.
Relationship between history of sexual violence and other factors
In the participants with a STI, there was no significant difference between those who had a history of sexual violence (59.1%) and those who did not (44.4%) (P = 0.261).
With regard to the relationship with sexual behavior, the mean age at first intercourse of those subjects who had experienced sexual violence was 14.9 years, which was significantly lower than the participants who had no such experience (15.5 years) (U = 598.0, P = 0.047) (Table 2). The mean number of previous sexual partners was 10.5 in those participants who had a history of sexual violence, while the mean number was 5.5 in those who had not, indicating that subjects who had had a violent sexual experience had significantly more previous sexual partners (U = 574.0, P = 0.030). The mean ages of the oldest previous sexual partners of those subjects who had a history of sexual violence compared with those who had not were 24.2 and 21.8 years old, respectively. This is significant because the latter are still students and do not have the social power or financial means that the former, salaried men, do; including the means to pay for sex. Participants who had experienced sexual violence had significantly older sexual partners in the past (U = 559.0, P = 0.031).
Table 2. Experience of sexual violence and sexual behavior
|Age at first intercourse (n = 80)||14.9|| 1.5||14.0||15.4|| 1.1||16.0||598.0||0.047|
|Period of time (weeks) from when they met to when they had sex for the first time (n = 75)†|| 4.8 || 6.9|| 2.0|| 7.3||11.7|| 2.0||630.0||0.435|
|Total number of sexual partners (n = 80)||10.5||20.8|| 6.0|| 5.5||11.1|| 3.0||574.0||0.030|
|Age of the oldest partner (n = 79)||24.1|| 7.5||22.0||21.7|| 6.5||21.0||559.0||0.031|
|Number of sexual partners whom they had met within 3 months (n = 80)|| 2.3 || 5.3|| 1.0|| 1.0|| 0.9 || 1.0||646.5||0.059|
|Oldest age of the current sexual partner (n = 67)‡||21.3|| 3.8 ||20.0||20.4|| 3.3 ||21.0||477.5||0.350|
|Friendship period (months) with current partners (n = 65)†||12.9||11.4|| 7.0||13.9||13.0|| 9.0||505.0||0.863|
With regard to the subjects’ attitudes towards sex, when asked whether they thought it was acceptable to have sex with a person whom they had met for the first time, 42% of those subjects with a history of sexual violence responded ‘yes’, while only 18.4% of the subjects with no history of sexual violence gave the same reply. Thus, the response ‘yes’ was given significantly more often in those subjects who had experienced sexual violence (P = 0.029).
As for lifestyle and the frequency of eating breakfast per week, 40.5% of those subjects who had a history of sexual violence responded ‘more than 5 days’, while 59.5% responded ‘less than 4 days’. In contrast, 63.2% of the subjects who had no such experience responded ‘more than 5 days’ and 36.8% responded ‘less than 4 days’. Thus, the frequency of breakfast per week was significantly lower in those subjects who had a violent sexual experience (P = 0.048) (Table 3). No relationship with smoking and drinking was noted.
Table 3. Experience of sexual violence and lifestyle (N = 80)
|Smoking||Yes||34 (81.0)||23 (60.5)||0.052|
|No|| 8 (19.0)||15 (39.5)|| |
|Current smoking||Yes||22 (52.4)||14 (36.8)||0.184|
|No||20 (47.6)||24 (63.2)|| |
|Frequency of drinking||Every day|| 1 (2.4)|| 3 (8.1)||0.501|
|Sometimes||33 (78.6)||26 (70.3)|| |
|Social drinking|| 8 (19.0)|| 8 (21.6)|| |
|Frequency of meals per day||Three times||19 (45.2)||19 (50.0)||0.823|
|Once or twice or more than four times||23 (54.8)||19 (50.0)|| |
|Frequency of breakfast per week||More than five times||17 (40.5)||24 (63.2)||0.048|
|Less than four times||25 (59.5)||14 (36.8)|| |
|Freqency of eating out per week||Less than 2 days||24 (57.1)||26 (68.4)||0.358|
|More than 3 days||18 (42.9)||12 (31.6)|| |
Statistically significant differences in five factors, namely: ‘age of sexual debut’, ‘total number of sexual partners’, ‘age of the oldest partner’, ‘attitudes to sexual intercourse with a person they met for the first time’ and ‘the frequency of breakfast per week’ were found with the Mann–Whitney U-tests and the Fisher's exact tests. The ‘age of subjects at the interview’ was not statistically significant. A multiple logistic regression analysis of these six factors was carried out using the presence or absence of a violent sexual experience as the criterion variable (Table 4). There were no significant correlations among these factors. There appeared to be some relationship between sexual violence and ‘it is acceptable to have sex with a person whom you have met for the first time’ (P = 0.097). However, there was no significant relationship with the other factors, namely: ‘age of sexual debut’ (P = 0.195), ‘total number of sexual partners’ (P = 0.994), ‘age of the oldest partner’ (P = 0.928), ‘the frequency of breakfast per week’ (P = 0.145) and ‘age of subjects at the interview’ (P = 0.162). Furthermore, the odds ratio for ‘it is acceptable to have sex with a person whom you have met for the first time’, ‘the frequency of breakfast per week’, ‘age of subjects at the interview’ and ‘age at first intercourse’ was 2.6, 2.1, 1.4 and 0.8, respectively.
Table 4. Logistic regression analysis of the effect factors on sexual violence
|Age at first intercourse (years)|| ||80||0.8||0.5–1.2||0.195|
|Total number of sexual partners|| ||80||1.0||1.0–1.0||0.994|
|Age of the oldest partner (years)|| ||79||1.0||0.9-1.1||0.928|
|It is acceptable to have sex with a man whom you have met for the first time||Yes||25||2.6||0.8–8.1||0.097|
|No||55|| || || |
|Frequency of breakfast per week||<4||39||2.1||0.8–5.8||0.145|
|>5||41|| || || |
|Age of the subjects at interview (years)|| ||80||1.4||0.9–2.1||0.162|
IN THE PRESENT study, 52.5% of the participants had experienced sexual violence. A total of 69.0% were victims of DV and 64.3% had been raped. Most rapists were ‘strangers’ (70.4%), but in 29.6% of participants the rapist was known to the victim. According to a 2006 survey by the Cabinet Office,2 10% of married women had been victims of spousal sexual violence in the past year. However, in the present study, the number of participants who had experienced DV was 29 (36.3%), which is considerably higher than the results of the Cabinet Office survey. One reason for this discrepancy is that the Cabinet Office Survey had no age limitations and obtained information on relationships with spouses only. Another reason might be that participants in the present study were teenage girls, the majority of whom were not married to their partners, and seemed to be unaware that they were victims of DV until we asked the question ‘how does your sex partner usually act when you refuse sex’.
Furthermore, the number of rape victims in our study was 27 (33.8%). According to 2005 criminal police statistics,1 the number of reported rapes and sexual assaults was 2076 and 8751, respectively, which accounted for a mere 0.01% of the Japanese population. Very few subjects in our study reported their rape to the police. Cases that had been reported included a case where a victim had been held captive for several days and raped or where a passerby had rescued a child victim. Victims of sex crimes in Japan tend not to take action and the majority of victims in our study did not even tell their family. Therefore, Japanese police statistics grossly underrepresent the actual number of sex crimes. Furthermore, during individual interviews, there were many cases were the victims had not only kept silent about their attacks, but did not even realize that they had been victims of rape because they were too young to know what sex was. These girls would have been excluded from general questionnaires or police reports.
In the present study, with regard to the number of sexual partners, 20% of the subjects responded ‘1’ and 52% responded ‘less than 4’, while 20% responded ‘more than 10’. Kihara9 found that more than 70% of Japanese teenage girls had had multiple sexual partners, while Nonoyama10 found that sexually experienced Japanese women aged 15 to 24 years had had between 1 and 35 previous sexual partners. In terms of the period from the time when they met to when they had sex for the first time, 8.8% of subjects in the present study responded ‘within 1 week’, while 37.5% responded ‘within 1 month’. Furthermore, 37.5% had also experienced sex with someone other than their current sexual partner. A questionnaire conducted by Ito11 among junior high and high school students indicated that 20% of sexually experienced junior high school students and 15% of sexually experienced high school students had sex on the same day they met someone for the first time. Likewise, subjects in the present study also responded that they had sex shortly after meeting for the first time, including the same day. According to our study question about the age of the oldest sexual partner, 34% responded that their sexual partners were ‘under the age of 19’, which was almost the same age as the subjects themselves; however, 55% responded ‘20 to 29 years old, and 11% responded ‘more than 30 years old’, which is more than a 10-year difference. This suggested that subjects tended to have sex with older persons, which might be a factor for the high rate of DV and rape recorded in the present study.
Factors that were significantly different with regard to the presence or absence of a history of sexual violence included: ‘age of sexual debut’, ‘total number of sexual partners’, ‘age of oldest partner’, ‘sexual intercourse with person they met for the first time’ and ‘the frequency of breakfast per week’. When a multivariate analysis of the above-mentioned items, as well as ‘age of the subjects at the interview,’ was carried out, there was no significant relationship between violent sexual experience and the above-mentioned factors concerning sexual partners. However, there appeared to be a relationship with ‘it is acceptable to have sex with a person you have met for the first time’ because the odds ratio of ‘it is acceptable to have sex with a person whom you have met for the first time’ was 2.6, which indicated a 2.6-fold risk of sexual violence. There was also a relationship between sexual violence and the age of sexual debut and between sexual violence and the number of sexual partners in the present study. Furthermore, it is likely that older sexual partners might force girls to engage in undesirable or risky sex. Most of the subjects in the present study were DV victims, including girls who were physically and verbally abused by their sexual partners when they refused sex and girls who had unwanted sex because their sexual partners told them they would engage in affairs with other women or prostitution if they did not comply.
The present study also included subjects whose first sexual experience involved rape and who were raped in their childhood. Some of these subjects had experienced several incidents of sexual violence or had unprotected sex with strangers who had picked them up or paid them to have sex. Regardless of the fact that they had a STI or induced abortion, those subjects who recognized the risks of unprotected sex demonstrated no protective behavior. Boyer3 stated that victims of sex crimes were likely to experience violence by their sexual partners. In the present study, prior victimization from a sex crime resulted in new victimization from another sex crime because subjects could not refuse sex because of their fear from a previous rape. Thus, previous rape experience appears to have a large impact on future sexual behavior and attitudes towards sexuality.
No relationship was noted with drinking, use of condoms or prostitution experience. However, the influence of alcohol and drug abuse in the younger generation, as well as in the sex industry, varies from country to country and culture to culture. Further study of other factors that showed no significant relationship in the present study should be conducted on a larger scale to prevent characteristic sexual violence in Japan and to support victims of sex crimes. Furthermore, further research is necessary on the different attitudes towards sex and sexual behavior before and after a violent sexual experience, including the reasons why victims had sex with strangers, the relationship between sexual violence and irregular lifestyle, and irregular lifestyle caused by a violent sexual experience.
In the present study, significant differences in ‘age of sexual debut’, ‘total number of sexual partners’, ‘age of oldest partner’, ‘sexual intercourse with person they met for the first time’ and ‘the frequency of breakfast per week’ were noted with regard to the presence and absence of sexual violence. It is suggested that sexual behavior, such as early sexual experience and many sexual partners, puts teenage girls at risk of experiencing sexual violence in later life. As the odds ratio of ‘it is acceptable to have sex with a man whom you have met for the first time’ was 2.6, this indicates that a low threshold towards engaging in sex might to some extent increase the risk of sexual violence.
From the results of the present study we were not able to elucidate whether being a victim of violent sex was caused by risky sexual behavior or whether being a victim of violent sex changed sexual behavior. It did, however, indicate that a young age of sexual debut and having sex with many partners, particularly partners that were significantly older, puts teenage girls at risk of being victims of sexual violence.
To prevent sexual violence, it is necessary that young women recognize that having sex with much older sexual partners or paid sex partners induces an unequal power balance and they must learn behavioral risk avoidance. Furthermore, it is essential to support victims of sexual violence who were sexually abused in their childhood to avoid repetition of such experiences. Young girls must be educated about the aspects of sexual behavior and lifestyle that minimize or prevent the risk of sexual violence before they enter into their first sexual relationship and appropriate support to victims of sexual violence must be made available to prevent them participating in risky sexual behavior or lifestyles.
The present study has several limitations. The study population was small because it focused on 80 subjects who visited specific medical institutions. Therefore, it is not possible to extrapolate these results to young Japanese women in general. The number of facilities and subjects investigated should be increased. We carried out an interview survey on participants with regard to violent sexual experiences, including molestation during childhood. However, the majority of participants did not perceive what had happened to them as ‘violence’, but rather ‘an inevitable accident’. They might have still kept what happened to them a secret, even if they recognized it as sexual violence, so precise and accurate information might not have been obtained. Further research is necessary to compare sexual behavior before and after a violent sexual experience, and to examine attitudinal changes toward sexuality.
WE DEEPLY APPRECIATE the cooperation of the staff and patients at the medical institutions where the research took place. We would also like to thank Sharon Hanley, Health Sciences University of Hokkaido, for her help with the English preparation of the manuscript. This study was conducted with a Grant-in-Aid for Scientific Research, which was a part of the Japanese 2005 ‘Effective Intervention Program to Prevent Sexually Transmitted Diseases’.