A Longitudinal Study of the Aftermath of Rape among Rural South African Women

Abstract

Objective:

Sexual assaults against women are a global health crisis, with rates alarmingly high in South Africa. However, we know very little about the circumstances and the aftermath of these experiences. Further, there is limited information about how factors specific to the rape (e.g., fighting back) versus those that are specific to the individual – and potentially modifiable - influence mental health outcomes. This study examined how situational characteristics of rape as well as individual and situational factors confer risk for symptoms of depression, PTSD and dysfunctional sexual beahvior at 12-month follow-up.

Method:

Two-hundred and nine (N = 209) South African women were recruited from rural rape clinics in the Limpopo (LP) and North West Provinces (NWP) of South Africa. Interviews were conducted at baseline (within 6 months of the rape incident), at 6 and 12 months by trained staff at the clinics in English or their native language. Women were interviewed after services were provided in a private room.

Results:

One-hundred and thirty-two (n = 132) women were lost to follow-up at 12 months, resulting in 77 women with interview data for all time-points. Undermining by the survivor’s social support system and an increased belief in myths about rape were associated with increased dysfunctional sexual practices and depression significantly change from baseline to follow-up.

Conclusions:

These findings demonstrate the need for interventions that address the most pervasive effects of rape erode over time. These behaviors can increase risks for revictimization and reduce psychological well-being in the aftermath of rape.

Introduction

The World Health Organization (WHO) reports that sexual violence against women is a global health crisis (World Health Organization, 2013). While between 2008 and 2015, recorded cases dropped by 7.4% from 46.647 to 43.195 per year, the rates of rape in South Africa continues to be one of the highest in the world (Studies, 2015; Watch, 2010).

It is important to understand the circumstances of the events and the effects on women’s psychological and emotional health such as symptoms of depression and post-traumatic stress. Research in the US has found that 55% of sexual violence victims reported being diagnosed with depression or an anxiety disorder compared to 17% of a non-victimized sample (Bossarte, Choudhary, & Smith 2012), and between 33% and 45% of rape survivors report symptoms of PTSD (Campbell, 2008). In a recent study of South African rape survivors, major depression and PTSD rates at 2 months were 10.5% and 10.1%, respectively (Nöthling, Lammers, Martin, & Seedat, 2015). However, the authors note that the low rates may have been attributed to the counseling provided to severely traumatized patients at the recruitment site.

Personal and Interpersonal Sequelae.

More research is needed to identify the characteristics that heighten South African rape survivors’ risks for long-term psychological distress and dysfunction. One such characteristic discussed in women’s rights groups and not well understood is whether women fought their alleged perpetrators, especially those residing in rural areas of South Africa (Bugwandeen, Collings, Wiles, 2008; Kalichman 2005). While studies have reported that different circumstances of rape may relate in unique ways to a victims' help-seeking decisions due to stigma issues, self-blame, and other factors (Starzynski, 2005; Ullman & Filipas, 2001), it is less clear how these characteristics contribute to the persistence of mental health problems that may diminish the quality of their lives.

There are many personal and interpersonal characteristics that may heighten the negative aftermath of rape. Several studies have reported greater psychological distress among those who reported greater severity of child sexual abuse histories (Wyatt & Newcomb, 1990), and higher baseline levels of distress. Many women who are victims of prior abuse report more frequent and longer durations of depression and anxiety compared to those without histories of trauma (Zlotnick, 1997). Among survivors of childhood sexual abuse, the most common coping strategies are avoidance and denial in response to the abuse, as well as from the experiences of rape. The impact of rape is a devastating personal trauma where the survivors’ lives are often shattered and their psychological and physical privacy invaded. The long-term effects can be significant. Rape survivors are reportedly the largest group with Post Traumatic Stress Disorder (PTSD). Depending on health care and support received, the length of time for recovery can range from months to years (Dunmore, Clark, Ehlers, 2001).

The psychological consequences of rape are likely to be influenced by several factors, including the characteristics of the event (e.g., severity of the assault, force, alcohol use), immediate psychological reaction to the assault (e.g., self-blame, loss of self-esteem, shame, fear), reactions from their social network (e.g., criticism, blame, ostracism), and the use of denial and avoidance coping (Littleton and Brietkopf, 2006; Littleton 2006) The more severe the rape the greater the likelihood of denial, emotional distancing or the greater the self-blame (Meyer & Taylor, 1986), and as a consequence, the greater the likelihood of PTSD, as well as sexual, eating and mood disorders (Faravelli, 2004).

Rape in South African Context.

In the context of South Africa’s Apartheid era and the oppression of its Black population, the severity and frequency of rape is likely associated with a high population prevalence of depression and PTSD among Black women. In the meantime, a woman may develop health and psychological problems like PTSD or depression, but may only complain of symptoms that are misunderstood by family and friends, people in the workplace and in the community. As a result of rape occurring without adequate care, maladaptive coping strategies, financial and relationship problems and more unintentional injuries can lead to poor health, revictimization and isolation (Clark, Dunmore, & Ehlers, 2001). A longitudinal study of rape survivors who are in their childbearing years will increase our understanding of the effects of these experiences. Interventions can be developed to help women to improve health and mental health, develop positive coping strategies and use existing or new resources.

It is possible that women’s beliefs about rape, such as their dress or behavior, being factors that increased their chances of being victimized, may heighten their psychological distress over time. In a recent study, 17% of the South African women agreed that rape usually results from what a woman says or does (Kalichman et al., 2005). These myths may vary by women’s educational background, ethnicity or culture. Contributing factors such as these may require specific attention both as a prevention strategy and in interventions developed for survivors to address the circumstances of rape and in so doing, enhance anticipatory coping, as well as minimize distress and improve women’s abilities to cope with these experiences when they occur.

The literature on interpersonal characteristics such as social undermining suggests that victims’ reactions depend largely on their perceptions of the offense, and have damaging outcomes such as reciprocated social undermining, and as well as depression, decreased self-esteem, and psychosomatic symptoms (Duffy et al., 2006). Individuals who experience social undermining by those they initially perceive as a source of social support may experience feelings of betrayal and more severe depression and engage in self-blame following the rape.

Beliefs about rape and experiences of social undermining by one’s social resources for support may also increase the likelihood of engaging in self-destructive or risky behaviors (e.g. heavy drinking) following the rape (Gidycz et al 2006; Chin et al, 2010).

In sum, there is a dearth of information about the modifiable personal and interpersonal characteristics that are known to influence psychological well-being such as symptoms of depression and trauma within the South African context. However, while it is critical to identify factors that will minimize the aftermath of rape and increase the number of women who report their experiences, it is possible that women may minimize their symptoms immediately post-rape but evidence an increase in symptoms over time. The need for culture-informed interventions to address the long-term effects of rape and other trauma in resource-limited settings are clearly needed.

The purpose of this paper is to examine how personal and interpersonal factors confer risk for rape-related trauma symptoms and depression over time. The goal is to better understand some of the factors that either result in symptom exacerbation or reduction and that might contribute to the framework for the development and testing of future interventions.

Methods

Participants

The present study recruited and interviewed 209 treatment-seeking Black South African women. Permission to conduct the study was granted by the Internal Review Board at UCLA, the North West University, and relevant local health authorities and informed consent was obtained from all who volunteered to participate. Participants were recruited from two separate public trauma clinics: the LP clinic (N=87) and the NWP clinic (N=122). One hundred thirty two women (n=132) were lost to follow up at 12 months, resulting in 77 women with interview data for all time points. Women were eligible to participate if they were between 18 and 50 years of age, reported a rape incident within the last six months, and spoke one of the local languages. The demographic characteristics of the participants at each of the clinics are provided in Table 1.

Table 1:

Demographics

Measures	Baseline			12 months
	LP (N=51)	NWP(N=26)	p	LP (N=50)	NWP(N=26)	P
Age	28.4(12)	25.8(7.6)	0.2483	29.3(12.2)	27.6(7.7)	0.474
ethnicity
Venda	46(92%)	1(4%)	<.0001
Tswana	0(0%)	23(92%)
Other	4(8%)	1(4%)
Marital Status
Married	9(17.65%)	0(0%)	0.0246	10(19.61%)	1(3.85%)	0.087
Not married	42(82.35%)	26(100%)		41(80.39%)	25(96.15%)
Relationship status
No relationship	16(32%)	6(24%)	0.0676	16(31.37%)	6(24%)	0.0633
Dating	11(22%)	12(48%)		8(15.69%)	10(40%)
Living with partner	23(46%)	7(28%)		27(52.94%)	9(36%)
education
Less than matric	36(72%)	16(61.54%)	0.3519	28(54.9%)	15(57.69%)	0.8156
Matric and higher	14(28%)	10(28.46%)		23(45.1%)	11(42.31%)
employment
Not working	38(80.85%)	20(76.92%)	0.6908	40(83.33%)	20(76.92%)	0.5436
Part/full time	9(19.15%)	6(23.08%)		8(16.67%)	6(23.08%)
monthly income
Less than $55	37(74%)	15(68.18%)	0.6116	33(64.71%)	21(84%)	0.0814
$61 and higher	13(26%)	7(31.82%)		18(35.29%)	4(16%)
HIV Status
Positive	5(12.82%)	8(30.77%)	0.0763
Negative	34 (87.18%)	18(69.23%)
taking antiretroviral medication
Yes	8(20.51%)	6(23.08%)	0.8054
No	31(79.49%)	20(76.92%)
Did the victim fight back during the rape
Yes	14(27.45%)	15(57.69%)	0.0096
No	37(72.55%)	11(42.31%)
How long has it been since the rape (in days)	42.2(35.6)	80(32.9)	0.0002
Was there more than one person involved?
Yes	21(42%)	10(38.46%)	0.7659
No	29(58%)	16(61.54%)
Who was (were) the perpetrator(s)
Parents	1(2%)	0 (0%)	1
Relative	1(2%)	0(0%)	1
Stranger	28(56%)	14(53.85%)	0.8578
Husband/partner	1(2%)	4(15.38%)	0.0256
someone known	20(40%)	9(34.62%)	0.6466

Procedure

The present analyses were conducted as part of a larger longitudinal study called Project Fulufhelo (Hope) in conjunction with other services provided to sexual assault survivors at each clinic. Both clinics were connected to local hospitals and specialized in providing sexual assault services. Clients were provided psychological support and/or psychiatric referral as needed. Unfortunately, there are no other trauma services supporting rape survivors in either town. Participation in the study did not interfere with any aspect of the survivor’s standard care (e.g., counseling, medical exams or legal referrals, etc.). The manner in which rape survivors were admitted to both clinics was similar. Following standard care procedures, rape survivors were treated upon arrival to the clinic in a private room by designated clinic staff. They received a standard medical examination using a rape crime kit to collect medical evidence for forensic analysis. They also provided a formal statement disclosing the rape to a police officer. At the next clinic or home visit, women were informed about the current study and asked to participate if they met eligibility criteria. If they agreed, a trained counselor read the consent form and women chose the time, private place and language in which the interview would be administered (English or a local language). English-language surveys were translated into local languages by a local expert and a study coordinator and back translated to confirm accuracy.

The survey protocol included both qualitative and quantitative behavioral, psychological, and health measures that took approximately 2 hours to complete. If the participants asked for additional explanations about questions that were unclear, the interview was stopped and the participant was debriefed. Thus, women had the right to discontinue the interview at any time without any consequences to their care or compensation for participation. Following the interview, the women were paid $10.00 for their time, transport and food vouchers.

Measures

Demographics.

Participants provided background demographic information regarding their age, clinic site (LP vs. NWP), marital or relationship status, number of children, education (less than matric vs. matric or higher), work status, and income (see Table 1).

Sexual Abuse History.

Participants also answered questions from the Wyatt Sexual History Questionnaire which consisted of 54 questions about sexual abuse incidents that occurred prior to the age of 16 ranging from frottage to rape involving a relative, acquaintance or stranger (Wyatt, 1984). There were few reports of childhood abuse before age 16 in this sample, with only 8 (4.4%) women reporting being sexually abused and 23 (11%) reporting being physically abused since age 16. Thus, this variable was dichotomized as reported sexual and physical abuse or no reported abuse.

Circumstances of Rape Incident.

Women were also asked to provide some details of the rape incident, including the number of perpetrators, type of perpetrator (e.g., parent, stranger, etc.), the time between the assault and the interview, the location of the assault, etc. In particular, we assessed whether any physical force was used by the perpetrator(s) during the assault (e.g. hitting, slapping, choking, beating, threatening to use or used a weapon, and being held down). Participants reported how much they were hurt during the assault (no injury, minor injury, or major injury), and whether they fought back (yes/no).

Social Undermining.

Women completed a three-item questionnaire that asked them to identify whether an important person in their life had engaged in criticism or caused their life to become more difficult following the rape. The questions were rated on a 3-point Likert-type scale from “Not at all or a little” to “Quite a lot”. The composite includes the means of the three items. Using the Cronbach Alpha, resulting in a reliable composite social undermining score (r=0.90) (Sarason et al, 1981).

Coping:

The 28-item Brief COPE Inventory (Carver, 1997) was used to assess the coping strategies used in dealing with the rape. The inventory includes functional as well as dysfunctional responses. It also included 2 pairs of polar-opposite tendencies because each scale is unipolar (the absence of this response does not imply the presence of its opposite). Use of emotional support (2 items), positive reframing (5 items), humor (10 items), and active approach coping (65 items) were assessed as subscales and responses were recorded on a 4-point Likert type scale from 1=“Not at all” to 4=“Doing this a lot”. This measure has been used in several languages (Carver, 1997), and two reliable sum scores for emotional support (Alpha = 0.79) and approach coping (Alpha = 0.83) were calculated.

Rape Myth.

The Rape Myth Acceptance Scale is a six-item measure that was originally developed by Burt (1980) and adapted for use in South Africa (Burt, 1980; Kalichman et al, 2005). The rape myth measure included items such as “Many rapes happen because women lead men on” were reported on a 4-point Likert-type scale from 1 =“Disagree” to 4=“Agree” and a reliable sum score was calculated (r = 0.81).

Mental Health Outcomes.

Sexual Trauma Symptom (STS) subscale is a component of the Trauma Symptom Inventory (TSI, Brier, 1995) and evaluated acute and chronic posttraumatic symptomatology, including the effects of rape, spousal abuse, physical assault, combat experiences, major accidents, and natural disasters, as well as lasting sequelae of childhood abuse and other early traumatic events. STS items are rated according to frequency of occurrence over the prior 6 months using a 4-point scale ranging from 0 (“Never”) to 3 (“Often”). A 16-item sum score that assessed dysfunctional sexual behavior (DSB) was calculated as 0.73 reliability score for this sample. Examples of questions included how frequently women engaged in unhealthy sexual thoughts or behaviors (e.g., How often did you think about or had sex with someone you hardly knew?). The STS has been standardized for men and women ranging from 18 and older and has good reliability (Alpha = 0.86) for this sample (Briere and P.A. Resources, 1995).

Post-traumatic stress disorder was measured using the seventeen-item Posttraumatic Stress Diagnostic Scale (PDS), a measure that has shown reasonable internal and external validity in populations of women who have experienced trauma (Foa et al, 1997; Foa, 1995). The PDS was developed based on the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) criteria and includes questions regarding avoidance/numbness, physiological arousal, and re-experiencing/intrusive thoughts (e.g. “Did you relive the traumatic event, acting or feeling as if it were happening again?”). Participants were asked to rate how often each symptom had bothered them in the past month on a 4-point Likert-type scale from “Not at all” to “Almost always.” Scores on each item were summed to create a composite. The traditional cutoffs for symptom severity ratings are 0 =”No rating”, 1-10 =“Mild”, 11-20 =“Moderate”, 21-35 =“Moderate to Severe”, and over 36 =“Severe”. This scale demonstrated excellent internal consistency with an Alpha of 0.92.

Depressive symptoms were measured using the twenty-one-item Beck’s Depression Inventory-II (BDI-II), a widely used instrument designed to assess the severity of depression in normal and clinical populations (Beck, Brown, & Steer 1996). This measure contains items that list four statements arranged in increasing severity about a specific symptom of depression. The scale measures a range of depressive symptoms in line with criteria from the DSM-IV, including a loss of pleasure, sadness, agitation, and changes in appetite or sleeping patterns. Cutoff values for depressive symptoms for this scale are: 0-13 = minimal, 14-19 = mild, 20-28 = moderate, 29-69 = severe. Reliability in the sample was high, with an Alpha of 0.91.

To examine how changes between social undermining, endorsement of rape myths, coping efforts and emotional support were associated with changes in dysfunctional sexual behaviors, PTSD and depression across twelve months, a series of repeated measures regression analysis were conducted (See Table 3 for the depression results and Table 4 for Dysfunctional Sexual Behavior. There were no significant findings for PTSD).

Table3.

Repeated measure analyses for BDI

Measures	Estimate	Standard Error	DF	t Value	p Value
Social undermining	6.6277	1.6217	68	4.09	0.0001*
Emotional support	1.2502	0.8463	68	1.48
Approach Coping	−0.7903	1.1274	68	−0.7
Rape myth	4.8474	1.5699	68	3.09	0.0029*

Table4.

Repeated measure analyses for Dysfunctional sexual behavior

Measures	Estimate	Standard Error	DF	t Value	p Value
Social undermining	4.3053	1.0997	69	3.91	0.0002*
Emotional support	0.153	0.5741	69	0.27
Approach Coping	2.0089	0.765	69	2.63	0.0106*
Rape myth	3.5073	1.0652	69	3.29	0.0016*

Results

Overall Sample at Baseline.

A total of 209 women were interview at baseline, 100 at six months and 77 remained at twelve months, and interviewers reported that most women left the area after the rape in order to avoid the perpetrator(s) and to start a new life. There were 100 women who were interviewed at the 6-month follow up, and there were 77 women who were re-interviewed at twelve months after the rape incident. This paper describes the aftermath and adjustments of women who were victims of rape over 12 months and the mental health consequences of those changes.

In Table 1 we report the demographic characteristics of the sample. Using t-tests for continuous variables and chi-square tests for categorical variables, we found few differences between women from the two clinics with the exception of ethnicity (LP was mainly Venda; VWP was mainly Tswana; p<.001), marital status (most not married; p=.02), 73% of the women at LP did not fight back; p= .0096), and length of time since rape before seeking services (NWP women waited 80 days compared to 42.2 days for the LP women; p=.0002). The samples were combined for subsequent analyses. Overall, women in the sample were 27 years of age on average, not married (88%), less than high school educated (68%), and unemployed (75%), with $55 or less per month as income (68%). They were seen in the clinics approximately 12 months after the rape incident. About 30% of the sample was dating and 39% were living with partners prior to the assault incident. Over half (55%) of the alleged perpetrators were strangers, with the remainder either husbands or partners.

We compared the 77 women who agreed to be interviewed at baseline and at twelve months versus the 132 who dropped out of the study because they could not be located at the 12-month follow up. While there were no significant differences between groups on age, employment, income, marital status and whether they were living with their partners or not, 79 percent of women from NWP clinic versus 46 percent of women from the LP clinic dropped out of this study (p<.001). Further, 72 percent of women with a matric level of education (high school) dropped out versus 58 percent of women with less than matric level of education (p=.0479).

The Dysfunctional Sexual Behavior (DSB) scores were in the moderate range and increased significantly from baseline (mean=22, SD=6) to 12 months (mean=24, SD=7) p=0.004. Depression scores were low and the change from baseline to 12-months was not significant (baseline mean=9.5, SD=9.5, 12-month mean=9.3, SD=10.2; p=0.9209). PTSD scores were elevated, but the change in symptoms from baseline to 12 months was not significant (baseline mean=31.2, SD=12.9, 12-month mean=30.9, SD=14.3; p=0.8997). Attitudes about rape myths were low, but the change from baseline to 12-months was significant (baseline mean=1.5, SD=0.5, 12-month mean=1.6, SD=0.5; p=0.0175).

Women also described the characteristics of the perpetrator(s) and the assault, as illustrated by this 42 year-old woman’s description:

“I was on my way from a field when I came across an unknown guy who grabbed me by the throat and pulled me down. I collapsed and when I awoke, I found that I was raped, stabbed in the shoulder and my neck was strangled. I was unable to talk. I forced myself to walk to the house where someone could see me and help.”

Fewer than half (45%) of women reported being sexually assaulted by someone known to them, including husbands or partners. Over half (58%) reported that there was only one perpetrator involved, 40% reported more than one perpetrator involved, and 2% did not report this information.

Women reported a range of force used in their rape: 7.79% were forced to have sex by verbal threats, 22.08% were forced by the perpetrator’s strength, 25.97% were hit, slapped, choked or beaten, and 32.47% were threatened with the use of a weapon.

Pre-Post Changes.

Table 2 reports the changes from baseline to the 12-month follow up interview. A series of independent T-tests for matched pairs were run to test for significance of pre-post change. The results indicated that while women reported an increased ability to discuss and think about the sexual assault incident, there were significant pre-post changes on all of predictors. Incidents of social undermining (p<.0111), approach coping (p<0.014), and agreement with rape myths (p<0.0175) increased significantly, while emotional support from family and friends decreased significantly over time (p=0.0232).

Table2.

Measures with significant changes from baseline to 12 month follow-up (N=77)

Measures	Baseline	12 months	T value	P value
Social undermining	1.2 (0.43)	1.4(0.5)	−2.61	0.0111*
Emotional support	2.8(1.09)	2.4(1.1)	2.32	0.0232*
Approach coping	2.1(0.82)	2.4(0.96)	−2.52	0.014*
Rape Myth	1.5(0.5)	1.6(0.5)	−2.43	0.0175*
outcome measures
Dysfunctional sexual behavior	21.7(6.6)	24.4(7.7)	−2.97	0.004*
PTSD	31.2 (12.9)	30.9 (14.3)	0.13
BDI	9.5 (9.5)	9.3 (10.2)	0.1

Women also reported moderate to severe levels of dysfunctional sexual behaviors at both time points (m = 21.7 & 24.4) and the pre-post change was significant (p< 0.004). The PTSD scores were also in the moderately severe range (m = 31.2 and 30.9), but the pre-post change was not significant. Finally, depression scores were in the modest range (m = 9.5 and 9.3), but they did not change significantly.

Two simple multivariate regressions were conducted to test the relative contribution of the four predictor variables in accounting for variance in pre-post changes in severity of depression and dysfunctional sexual behaviours. Variables were entered in invariant order as follows: social undermining, emotional support, approach coping and rape myths. As shown in Tables 3 and 4, significant increases in scores on social undermining (p< 0.0001; p<0.0002) and rape myth (p< 0.0029; p<0.0016) scales were associated with higher BDI and dysfunctional sexual behavior scores, respectively. Changes in approach coping scores were also significant, but only for dysfunctional sexual practices, with the greater the increase in the use of approach coping the greater the report of dysfunctional sexual behavior (p< 0.01).

Self-Defense.

Finally, there were 29 women who reported fighting their perpetrators in an attempt to get away or minimize the effects of the physical and sexual assault. T-tests for matched pairs were examined for pre-post changes. Among those women, scores on dysfunctional sexual behaviours that might heighten risks for HIV transmissions significantly increased from baseline (mean=22, SD=6) to 12 months (mean=26, SD=7) (p=.0013). In this sub-sample there were no significant changes over time for symptoms of depression from baseline (mean=8.7 SD=7.7) to 12 months (mean=13.3 SD=11.9) (p=0.0862), or on attitudes about rape myths from baseline (mean=1.5 SD=0.5) to 12 months (mean=1.7 SD=0.5) (p=0.0626) or on PTSD from baseline (mean=35.5 SD=9.4) to 12 months (mean=36.3 SD=15.4) (p=0.7573).

Discussion

This is one of a few studies that described the rape experience and reported long-term health and mental health (depression and traumatic sexual symptoms) outcomes of women survivors at two rural clinics in South Africa. This study extends the literature on rape in South Africa (SA) in two ways: (1) this report focuses on the experience and psychological consequences of rape rather than its prevalence; and (2) it identifies factors that contribute to the negative aftermath of rape among rural women in SA. The findings are important because they inform the development of interventions to address this significant public health problem. While symptoms of depression and dysfunctional sexual behaviors were reported by all women immediately after the rape, changes in symptoms over time were associated with changes in perceived social undermining and adherence to beliefs related to myths about rape. In other words, for women who reported an increase in the amount of negative behaviors and criticism by others as well as an increase in their beliefs related to myths about rape, there was an increase in depression and dysfunctional or risky sexual behaviors over time. However, there was a non-significant statistical trend towards changes in coping behaviors relating to changes in depression and dysfunctional sexual behaviors over time.

The findings suggest that the immediate effects on these survivors and the risks for future longer term psychological problems need to be followed over time. The participants who were recruited from two rural clinics were all young, single, economically vulnerable women who reported rape incidents to the clinic staff, but also reported distress that required counseling to minimize more long-term lasting effects in the future. Some differences were noted, however, between women who participated in twelve months and those who were unavailable and dropped out. More work is needed to understand why women in a certain area were not available for the 12-month follow-up interview. Given that women with at least a high school education were more likely to have been unavailable, it is possible that they moved in search of employment but confirmation of these findings awaits further study.

Other differences were noted between ethnic groups of women. However, more research is needed to understand specific beliefs about rape that may differ between the ethnic groups included in this study. Scores on the PTSD measure were moderately high and remained unchanged, suggesting that the current services these women received from the rural clinics from which they were recruited may need to be expanded to include culturally adapted, evidenced based trauma treatments. These symptoms may be less likely to resolve on their own. Similar to other studies, women who ascribed to myths or misinformation about rape and how it occurred were also more likely to report symptoms of depression. Indeed, rape myth items essentially blamed the women for their own rape. It is of concern that endorsement of rape myths increased over one year. This finding suggests that more factual information about rape, its effects, and how to avoid these assaults should be available to rape survivors and to the community overall. Fighting the perpetrator back has been supported by other research reporting few additional physical injuries (Tark and Kleck, 2014). While the sample was small, women who fought back still reported psychological sequelae one year later. Fighting may have some effect on physical intrusion during rape, but the psychological effects still appear to endure.

Together, these findings support other studies that suggest that women who ascribed to myths or misinformation about rape such as “When a woman is raped, she usually did something careless to put herself in that situation” were more likely to report symptoms of depression (Kalichman et al., 2005). Facts about rape should be integrated into tribal ceremonies and religious and community based activities to help avoid “victim blaming”. Also, providing more information about how rape survivors may be negatively affected by social undermining that questions or blames the survivors needs to be part of public health messages to engender more social and community support for rape survivors. They can be more consistently supported over time. By providing more factual information and promoting the need for support, negative assumptions made by individuals who are part of the everyday lives of rape survivors may be minimized. The pros and cons of fighting a perpetrator back should be discussed so that women can make decisions about the advantages of self-protective readiness that may still require psychological counseling.

Although few women in this study reported past histories of violence and trauma, including childhood sexual and physical abuse, additional attention should be given to carefully assessing for these early experiences. Several studies have suggested that women who report having such histories were more likely to report symptoms of depression, including studies of American women of African descent (Banyard et al, 2002; Glover et al, 2010; Sigurdardottir and Halldorsdottir, 2013). The negative physical and psychological effects of cumulative incidents of violence and trauma on women who may have had prior histories of abuse before the rape incident should be included in future interventions that address PTSD symptoms in South Africa (Wyatt et al, 2002). This inclusion of the cumulative effects may be distinct from interventions that may address the rape incident alone for which women are currently seeking help. Whether women are ready to disclose their abuse histories or not, understanding that the effects can be long lasting may help women to seek help for all types of violence that they experience and to disclose them to trusted professionals.

While these findings are important, there are also several limitations of this study that are worth noting. First, this is a convenience sample of women who agreed to discuss their reported rape incidents and was not representative of rape survivors who did not choose to participate in this study. Consequently, the study did not compare women who disclosed their rape experiences with those who did not. Women who were raped, but did not agree to participate, could have had different experiences that are not reported here. While that number is small (fewer than 5), it is important to better understand the experiences of those who declined to be interviewed, as well. According to the interviewers, women were lost to follow-up at the 12-month check-in because they moved away from the area, perhaps to get a new start in their life. It will be important to identify what distinguishes those rape survivors with more education and who made the decision to relocate rather than stay versus those who remained in the communities where the incidents occurred. There is also a need for longitudinal studies beyond one year that follow survivors over time as they adapt different coping strategies and have more opportunities to finish their education and to work.

In conclusion, this is hopefully the start on ongoing work to more specifically address the cultural and contextual experiences of rape among South African women over time if effective and appropriate interventions are to be developed. Strategies for coping with past and current physical and sexual assault experiences are still needed along with expanded resources to address the longer term aftermath of rape among survivors, particularly among those who sought services in resource limited areas of South Africa.

Studies that investigate the social context in which rape has proliferated are needed so that new interventions can be designed that target men, families and the communities at large rather than limiting the focus of interventions to women survivors at risk for rape victimization.