Correlates of African American Sexual Assault Survivors’ Medical Care Seeking

Abstract

African American women are at higher risk for sexual assault than other racial/ethnic groups and have an overall high prevalence of lifetime sexual assault. Despite elevated risk and prevalence, African American survivors are often reluctant to use services in the aftermath of sexual assault. Yet, little research has focused exclusively on African American women’s sexual assault experiences including their experiences of medical care seeking. A mail survey study was conducted in Chicago (2010-2011) to understand better African American women’s sexual assault experiences in relationship to post-assault medical care seeking in a large community sample (N = 836). Multivariable regression analyses examined whether demographics, assault characteristics, trauma history, and post-assault psychosocial factors were related to medical care seeking. Results revealed unique correlates of immediate and long-term help-seeking from a variety of medical/health sources. Being of older age and lower income, perceived life threat, and delayed disclosure were related to less medical care seeking. Survivors who were assaulted by strangers, experienced interpersonal and contextual traumas, and who received tangible aid and mixed social reactions were related to medical care seeking. Implications for research and clinical practice with this population are provided.

Rape prevalence is high in African American women with recent data from the National Sexual Violence Survey (NISVS) showing 21.2% have had such experiences during their lifetimes, similar to 20.5% of White women (Breiding et al., 2014). Though other studies have reported higher assault rates, a recent (non-random) community interview study revealed over half (53.7%) of participants reported rape, and 44.8% reported sexual coercion during their lifetimes (Basile et al., 2016). Overall, African Americans are at greater risk of victimization by violent crime than are individuals of other racial and ethnic groups. A national telephone survey of community and college women showed African American women in both samples had experienced more forcible rape in their lifetimes than non-Hispanic White, Asian, or Hispanic women (Kilpatrick et al., 2007). Sexual assault risk is also elevated for African American women with intersecting marginalized identities, with low-income, HIV positive, bisexual, or incarcerated women at increased risk for sexual victimization (West & Johnson, 2013). Together, these studies highlight the high risk for and nature of sexual victimization of African American women.

While African American women face a high risk for sexual assault victimization, they also face unique health risks associated with being assaulted. Various health impacts associated with sexual assault have been documented, including poorer perceived health, greater odds of physical injuries from the assault, sexual and reproductive problems in some women, and acute and chronic medical conditions (Basile et al., 2016). African American women have reported a myriad of physical and mental health consequences from sexual assault, including genital injuries, sexually transmitted infections (STIs), bruises, post-traumatic stress disorder (PTSD), suicide ideation, and low self-esteem (West & Johnson, 2013). Mental health issues stemming from abuse may trigger maladaptive coping mechanisms, such as self-harm or substance abuse, that can exacerbate physical health consequences. African American women aged 18-24 years reporting sexual abuse in the past year are 4.5 times as likely to test positive for an incident of high-risk human papilloma virus (HPV) infection (Wingood et al., 2009). Despite these health problems, African American survivors often do not receive the medical care they need, exacerbating these issues, and little research has examined their medical care seeking.

Medical care seeking

Survivors of sexual assault seek help from medical sources at relatively low rates (Lanthier, Du Mont, & Mason, 2018). Less than two-thirds (60%) of women with a lifetime history of rape from a representative community sample reported seeking some form of formal assistance (Amstadter et al., 2008), of which 38% sought medical care. Zinzow and colleagues (2012) reported that only 21% of rape survivors from a national sample received medical care. Research has also documented factors related to medical care seeking.

Studies have shown that demographics (age, race, sexual orientation, and education) may relate to post-assault help-seeking (Amstadter et al., 2008; 2010; Zinzow et al., 2012), although findings vary depending on the nature of the sample (representative or not), population (college or community) and analyses (bivariate/multivariate). For example, Zinzow and colleagues’ (2012) final multivariate model of correlates of greater seeking of medical care included black race, rape-related injury, concerns about STIs, pregnancy concerns, and reporting the incident to police. Amstadter and colleagues’ (2008) study of national data showed only depression was related to medical care seeking when controlling for demographics and rape characteristics. Study results from other non-representative community samples have shown that specific assault characteristics may be important with more violent assaults, stranger assaults, and those in which survivors perceived life threat (i.e., that they might be killed during the incident) related to more formal help-seeking (Starzynski et al., 2007). Histories of interpersonal and contextual traumas related to increased health consequences of such experiences (Meshberg-Cohen, Presseau, Thacker, Hefner, & Svikis, 2016) may lead to greater medical care seeking.

Social support following the assault may also contribute to medical care seeking decisions, with informal support sources (e.g., family, friends, partners), either increasing or decreasing rape survivors’ help seeking (Paul & Sasson, 2013). Negative social reactions to disclosure of rape may affect African American survivors’ psychological symptoms to a greater extent (Hakimi, Bryant-Davis, Ullman, & Gobyn, 2018). Thus, if African American women are consequently less likely to disclose to those sources or delay doing so, this may also decrease their assault-related medical care seeking or lead them not to disclose in the medical setting, particularly in settings dominated by White providers and/or those without appropriate cultural competence.

Race and Help-seeking

Women of color have experienced years of oppression, creating barriers to getting medical care including stereotypes and cultural beliefs specific to African American women’s sexuality and victimhood. Culturally-specific stereotypes about African American women’s sexuality, such as the Jezebel stereotype that they are sexually promiscuous and immoral and subsequently “unrapeable”, remain today, and serve as a barrier to disclosure and help-seeking (Tillman, Bryant-Davis, Smith, & Marks, 2010). White gender role expectations of women being passive lead African American women who fight back during an assault to be less likely to be perceived as victims, thus creating a barrier to seeking assault-related services. African American women have often been socialized to embrace values such as perseverance and strength (i.e., the “strong black woman”; Wyatt, 2008). While this may be a positive self-image that empowers African American woman to help themselves and others, internalizing this persona may make it difficult for women to admit that they need help coping with the trauma of sexual assault (West & Johnson, 2013) and has been associated with silencing (Tillman et al., 2010). Previous studies have revealed “White dominated helping professions” as a barrier to help seeking (Alvidrez et al., 2011; NSVRC, 2016). Survivors who are racially/ethnically matched with intake clinicians are more likely to engage in treatment (Alvidrez et al., 2011); so, culturally relevant organizations staffed with diverse employees who work to establish rapport with the community and have high quality training about sexual assault and other forms of violence in the population are needed. African American women may receive racial and gender-based socialization that helps them survive, but that also makes them less able to express their needs to themselves or others thus delaying help seeking (Anyikwa, 2015). Only a few studies have examined help-seeking in African American only survivors (Bryant-Davis et al., 2015; Long & Ullman, 2013; Tillman et al., 2010; West & Johnson, 2013), whereas most research has compared experiences of women from different racial/ethnic groups (Amstadter et al., 2008; Weist et al., 2014; Zinzow et al., 2010) or statistically controlled for race in multivariate analyses (Amstadter et al., 2010; Zinzow et al., 2012).

Nondisclosure and delayed disclosure are frequent occurrences among sexual assault survivors and can negatively affect their access to and receipt of physical and psychological services by medical professionals (Tillman et al., 2010). This may occur because not telling informal support sources (e.g., family, friends, partners) leads to less formal support seeking, as those sources often encourage survivors to seek formal help (Paul and Sasson, 2013). If African American survivors receive tangible aid from informal sources they tell, that may lead them to seek medical care. Nondisclosure or delayed disclosure may lead survivors to ignore possible somatic or other health issues that may be assault-related sequelae and could contribute to long-term health problems. Survivors tend to not seek medical/ health care services immediately after the assault (Weist et al., 2014), either due to not labeling the experience as an assault, shame, embarrassment, or lack of additional physical injuries in addition to the sexual victimization, which often occurs in acquaintance/known offender assaults (Littleton, Axsom, & Grills-Taquechel, 2009). Few studies have examined factors associated with disclosure/nondisclosure in African American survivors in general or to specific formal support sources like medical/health professionals.

One study comparing service seeking between White and African American survivors found no differences in seeking medical care post-assault based on race; however, White survivors with low education were less likely than African American survivors with low education to seek medical care (Weist et al., 2014). A national sample of adult women showed higher rates of post-assault medical care among African American women than White women, such that African American women were more than three times as likely to seek medical attention (30% of African American sample sought post-assault medical care; Zinzow et al., 2012). Injury, reporting to the police, sexually transmitted disease (STD) concerns, and pregnancy concerns were positively related to receipt of post-assault medical care. The authors discussed whether greater medical care seeking among African American women was reflective of increased likelihood of experiencing a stereotypic sexual assault versus patterns of medical help-seeking associated with specific assault characteristics (e.g., no condom use, violence) (Zinzow et al., 2012). However, Flicker and colleagues (2011) found that only severity of physical abuse was associated with increased likelihood of seeking medical help. Overall, research exploring African American survivors’ medical care seeking is lacking and is often only considered in comparison to White survivors’ medical care seeking.

The limited research to date examining African American women’s experiences of mental and physical heath help-seeking has revealed little use of formal services and significant barriers to reaching them. Greeson and Campbell (2011) found survivors were often revictimized by medical systems with over half (52% of the sample) of African American survivors reporting several revictimizing experiences (e.g., disbelief, blame, lack of empathy, being treated in a cold manner) when seeking post-assault medical care and indicating that they would be unlikely to seek further help as a result of their negative experiences. Additional research is needed using an all African American sample of survivors to further our understanding of survivors’ pathways to medical help-seeking and their experiences in receiving services.

Current Study

Based on prior research on help-seeking correlates in general among women and sexual assault in women, we conducted an exploratory analysis of demographics, victimization and trauma history, sexual assault characteristics and post-assault experiences in relationship to seeking medical help both immediately post-assault and “ever” (e.g., whether they ever told a medical professional about assault) specifically for African American sexual assault survivors. The following hypotheses guided the present study: 1) History of more traumatic events, including child abuse, would be related to more immediate and ever seeking of medical care for assault; 2) More violent sexual assaults, greater perceived life threat, assaults perpetrated by a stranger, no pre-assault drinking, and earlier disclosure post-assault would be related to more medical care seeking immediately and ever in life; and 3) Greater PTSD, depressive symptoms, positive coping, telling informal sources, negative and mixed social reactions, and tangible support would be related to more medical care seeking immediately and ever in life.

Method

Sample

The participants in this study were 836 African American women who were part of a larger study from the Chicago metropolitan area of women who had an unwanted sexual experience at the age of 14 years or older.

Procedure

Data were collected over one year from a sample of women volunteers recruited from the community through advertisements (both online and print) and/or referrals from other participants. Fliers advertising the study were posted at community agencies serving women and survivors of violence, at local colleges and universities, and businesses catering to women (e.g., women’s bookstores, nail and hair salons) across the metropolitan area; and also specifically targeted agencies located in areas serving women of color. To participate, the women were screened for eligibility during an initial phone call to the study office using the following criteria: a) they had an unwanted sexual experience at the age of 14 years or older, b) were 18 years or older at the time of study participation and c) had told someone previously about their unwanted sexual experience. Even if ineligible to participate, women were sent a list of community resources. Qualified and interested women were sent packets containing the survey, a written informed consent document, a list of community resources, and a stamped return envelope for the completed mail survey. Women were sent $25 money orders after returning completed surveys (85% response rate). All documentation and study procedures were approved by the University of Illinois at Chicago’s Institutional Review Board.

Measures

Sexual assault.

Sexual victimization in childhood (prior to the age of 14 years) and adulthood (at age 14 years or older) was assessed with a modified 11-item version of the Sexual Experiences Survey - Revised (SES-R; Testa et al., 2004) that assesses various forms of sexual assault, including: unwanted sexual contact, verbally coerced intercourse, attempted rape, and rape due to force or incapacitation (e.g., from alcohol or drugs). The SES-R has good reliability (α = 0.73; Testa et al., 2004)); similar reliability was found in this sample (α = 0.77). Assault severity was computed by coding the highest severity sexual assault type on the SES-R. If women had multiple assaults, the assault severity and details about the assault were assessed for the most serious incident. Most women had experienced a completed rape (80%), or a history of child sexual abuse (CSA; 78.7%).

Assault characteristics.

Relationship to the perpetrator was coded as stranger, acquaintance/date, romantic partner/husband, other relative, and multiple perpetrator types. Perceived life threat was assessed with a no/yes question asking women whether they feared their life was in danger during the assault. Physical injury was assessed with several items asking about which types of injuries women experienced as a result of the assault, including minor cuts and scratches, bruises, cuts, knife and/or gunshot wounds. Finally, pre-assault drinking of alcohol was assessed by asking women if they were drinking prior to the assault.

Coping strategies.

Participants completed the Brief COPE, a 28-item self-report scale of coping strategies (Carver, 1997) used in the past 12 months to cope with the assault using a scale ranging from 0 (I didn’t do this at all) to 3 (I did this a lot). Maladaptive coping was computed based on a factor analysis as the average of responses to eight items composing the behavioral disengagement, denial, self-blame, and substance use subscales (α = 0.81; total possible range of scores = 0-24). In the present analyses, positive coping was analyzed including both adaptive individual coping, which includes twelve items assessing adaptive, active forms of individual coping (α = 0.83) such as: “I thought hard about what steps to take” and adaptive social coping which includes four items assessing active adaptive interpersonal forms of coping (α = 0.87; total possible range of scores = 0-36) such as: “I tried to get advice or help from other people about what to do.”

Social reactions to assault disclosures.

Women completed the Social Reactions Questionnaire (SRQ; Ullman, 2000), reporting how often they received 48 different social reactions from any support provider they told since the assault on a scale ranging from 0 (never) to 4 (always). The scale contains seven subscales (blame, stigmatization, control, distract, egocentric, emotional support, and tangible support. Responses were averaged to create subscales (all ranges: 0-4) assessing the frequency with which participants received positive reactions to assault disclosure (e.g., emotionally or informationally supportive reactions such as “Held you or told you that you are loved,” “Helped you get information of any kind about coping with the experience,” or tangible aid “Took you to the police”) and negative reactions to assault disclosure (e.g., blaming or stigmatizing the survivor reactions such as “Told you that you could have done more to prevent this experience from occurring” or “Said he/she feels you’re tainted by this experience”). The SRQ has good test-retest reliability (rs = 0.68 to 0.77) and evidence of several forms of validity as reported by Ullman (2000). The subscales were also reliable in this sample with Cronbach’s α = 0.93 for negative reactions to assault disclosure and 0.92 for positive reactions to assault disclosure. In addition, we coded mixed reactions that were not clearly positive or negative which were low in average frequency of occurrence. Specific negative reactions that involved turning against the survivor (e.g., blame, stigma) were also assessed and low in frequency of occurrence.

Posttraumatic stress symptoms.

PTSD symptoms were assessed with the Posttraumatic Stress Diagnostic Scale (PDS; Foa, 1995), a standardized 17-item instrument based on DSM-IV criteria on which women rated how often each symptom (i.e., reexperiencing/intrusion, avoidance/numbing, hyperarousal) bothered them in relation to the assault in the past 12 months. The 17 items were summed to assess extent of posttraumatic symptomatology (α = 0.93; total possible range of scores: 0-51).

Depressive symptoms were assessed with the Center of Epidemiologic Studies Depression Scale (CESD-7) modified by Mirowsky and Ross (1990). In our study, participants were asked to rate their symptoms over the past 12 months using a five-point Likert scale from 0 (never) to 5 (always). In our sample, reliability (α = 0.86) was satisfactory. Items were averaged with higher scores indicating more depressive symptoms (total possible range of scores = 0-21).

Traumatic life events.

Traumatic event history was measured with the Stressful Life Events Screening Questionnaire (SLESQ-Revised; Green et al., 2006) to assess interpersonal trauma (e.g., stalking, child abuse, abuse by a romantic partner, abuse by someone else, emotional abuse), and contextual trauma (e.g., military war zones, dangerous neighborhood, being threatened with a weapon, been in other frightening situations, force used against you in robbery, close friends died from homicide, witnessed death).” Interpersonal trauma scores ranged from 0 to 6 and contextual trauma scores from 0 to 5.

Data analyses

All variables were examined with descriptive statistics to ensure appropriate distributional assumptions were met in terms of normality and frequencies, and correlations were conducted to evaluate possible multi-collinearity of predictors. Analyses were conducted in two steps to examine whether demographics, assault characteristics and trauma history, and post-assault psychosocial factors were related to medical care seeking. First, separate initial three blockwise logistic regressions were conducted with three variable sets: a) demographics, b) assault and trauma history and c) post-assault factors for each dependent variable. From each of these models we considered any factors significantly related to the dependent variable at p<0.10 as eligible for inclusion in the next set of models. Second, those significant factors from each of the initial models using the separate variable sets were then entered into final combined blockwise logistic regression models, entering the three variable sets in sequential blocks (e.g., demographics, assault/trauma history, post-assault) for each dependent variable and producing odds ratios (ORs) and 95% confidence intervals (CIs). Dependent variables included immediate seeking of medical attention after assault, and whether the participant ever told a medical professional about the assault. Each model was examined for the significance of factors (i.e., odds ratios, p-values) in relationship to each dependent variable. Overall model fit was not the focus of these exploratory analyses, only the identification of potentially relevant factors related to medical care seeking; however, overall fit statistics [−2 x Log Likelihood Ratio (LLR), degrees of freedom, p value] are provided for final models in the text.

Results

Descriptive characteristics of assault and post-assault experiences

Women ranged in age from 18 to 71 years at the time of the survey (Mean = 31.1, SD = 12.2 years). The sample was well-educated (44.5% some college; 21.7 % college graduate or beyond, 19.9% high school or GED, 13.6% less than 12^th grade). Almost one-quarter of women were currently in school (23.4%). Over one-third of women were employed at the time of the study (33.6%). Fewer than 5% were of Hispanic ethnicity (4.8%). Income was relatively low in the sample with over half making less than $10,000 per year (52.7%); 17.5% made $10,000-$20,000; 11.6% made $20,000 to $30,000; 7.5% made $30,000-$40,000; and just over 10% made over $40,000. Sexual orientation was identified by women as only heterosexual for most of the sample (68%), followed by mostly heterosexual (13.7%), then bisexual (9.4%), and 3.5% other, and 0.7% mostly lesbian. Over two-thirds of women had children (70.8%) and over half were single at the time of the study (56%).

Most assaults were by known perpetrators (77%), including acquaintance/friend/dates (35.3%), romantic partner/husbands (17.6%), other relatives (13.5%), and multiple perpetrator types (1.9%), whereas 22.7% were by strangers (Table 1). In terms of the highest severity of sexual assault as assessed by the SES-R, responses showed that 0.5% reported none of the items (but had some other type of unwanted sexual experience), 3.3% had unwanted sexual contact, 8.0% had sexual coercion, 7.8% had attempted rape, and 80.4% had completed rape experiences. Most women had a childhood sexual assault (CSA) (78.7%) history in addition to adult sexual assault (ASA), the focus of this study. A majority (70.3%) reported perceiving their life was in danger at the time of the assault. Women reported being physically injured with more minor injuries (soreness 66.7%, bruises 46.5%, cuts 12.1%) and fewer severe injuries (e.g., broken bones 2.3, gunshot wounds 3.3%) during the assault. Under one-fifth of survivors were drinking prior to assault (19.1%). Assaults occurred an average of 16.69 years ago, (SD = 12.13, range 0-52 years).

Table 1.

Adult sexual assault characteristics and outcomes (N = 836)

	Total %	Mean	SD
Depressive symptom score	-	2.02	0.74
PTSD symptom score	-	21.81	13.03
Assault type
None	0.5	-	-
Sexual contact	3.3	-	-
Sexual coercion	8.0	-	-
Attempted rape	7.8	-	-
Completed rape	80.4	-	-
Child sexual abuse history	78.7	-	-
Perpetrator
Stranger	22.7	-	-
Acquaintance/date	35.3	-	-
Romantic partner/husband	17.6	-	-
ther relative	13.5	-	-
Multiple perpetrator types	10.9	-	-

Note. PTSD = Posttraumatic stress disorder

Other traumas and post-assault disclosure

In terms of other lifetime traumatic events, women reported an average of over two other contextual traumatic events and almost three other interpersonal trauma types other than the sexual assault. Contextual trauma scores had a Mean = 2.36, SD = 1.55, and interpersonal trauma scores had a Mean = 2.98, SD = 1.50.

Most women delayed disclosing sexual assault for months or years after it occurred (53.7%) whereas 46.3% disclosed immediately up to weeks after the incident (Table 2). Assaults perpetrated by someone known to the victim were associated with delayed disclosure, r (692) = 0.16, p<0.001. Virtually all women (96.2%) told an informal support source (family, friend, romantic partner/spouse). Women reported a moderate level of positive/adaptive coping (active: Mean = 29.19, SD = 7.81, interpersonal: Mean = 9.07, SD = 3.72) and greater maladaptive coping (Mean = 16.35, SD = 5.78). On average, women reported “rarely” receiving negative reactions to assault disclosure, such as turning against the survivor (e.g., victim blame) (Mean = 0.78, SD = 0.95), low levels of mixed social reactions (Mean = 1.24, SD = .89), and “sometimes” receiving positive reactions to assault disclosure (Mean = 2.22, SD = 0.95), with a low level of tangible support overall.

Table 2.

Traumatic life events and post-assault disclosure (N = 836)

	Total (%)	Range	Mean	SD
Received tangible support score	-	0 – 4	1.40	1.23
Negative turning against social reactions	-		0.78	0.95
Mixed social reactions	-		1.24	0.89
Positive coping score	-	16 – 64	39.67	10.34
Maladaptive coping score	-	0-24	16.35	5.78
Number of contextual traumas score	-	0 – 6	2.36	1.56
Number of interpersonal traumas score	-	0 – 5	2.98	1.50
Victim pre-assault drinking	19.1	0 – 1	-	-
Life threat	70.3	0 – 1	-	-
Disclosure timing			-	-
Immediately to two weeks after	46.3	0 – 1		-
Months or years later	53.7	0 – 1
Told informal source	96.2	0 – 1	-	-
Receipt of mental health care
Post-assault	29.1	0 – 1	-	-
Anytime	36.7	0 – 1	-	-
Receipt of medical care
Post-assault	22.9	0 – 1	-	-
Anytime	23.6	0 – 1	-	-

Women reported a low level of depressive symptoms (Mean = 2.01, SD = 0.75) and a moderate level of PTSD (Mean = 20.75, SD = 12.76) symptoms related to the assault. In response to a question asking if they received medical attention, few women indicated they had done so post-assault (22.9%). In response to a question asking if they ever told “a medical doctor, other medical person, or emergency room staff,” only 23.6% reported ever having sought medical care related to the assault.

Multivariate regression analyses.

Analyses were conducted to examine whether demographics, assault characteristics and trauma history, and post-assault psychosocial factors are related to medical care seeking. Separate initial logistic regressions were conducted with three variable sets: a) demographics; b) assault and trauma history; and c) post-assault factors. Significant factors from each of the three variable sets were then entered in final combined blockwise logistic regression models for the dependent variable. Dependent variables included immediate seeking of medical attention after assault, and whether they ever told a medical professional about the assault.

Results from the final models showed that in older age and lower income were associated with less medical care seeking both immediately post-assault and ever (Table 3). Survivors of assaults by strangers, not only known perpetrators, were more likely to seek medical care both immediately post-assault and ever. Life threat was related to less medical care seeking both immediately post-assault and ever. Experiencing additional interpersonal traumas was related to greater medical care seeking ever, whereas experiencing contextual traumas was associated with greater immediate medical care seeking. Delayed sexual assault disclosure was related to less medical care seeking both immediately and ever. Greater tangible support/aid reactions from others one told about sexual assault related to greater seeking of medical care immediately and ever. Mixed social reactions (i.e., acknowledgement without support – distraction, egocentric, some control) were related to more immediate medical seeking. Model fits were: for immediate medical care seeking χ² (9) = 155.47, p < 0.001; −2 x LLR = 466.55, Nagelkerke R² = 0.36. and for ever medical care seeking χ² (7) = 136.44, p < 0.001; −2 x LLR = 496.25, Nagelkerke R² = 0.32, indicating that the significant factors accounted for over one-third of the variance in immediate and ever seeking of medical care.

Table 3.

Standardized beta coefficients (b’s) from logistic regression models for seeking medical care post-assault and ever telling medical professionals

Factor	Blocka	Final	Block	Final
	Immediate medical care seeking		Ever seeking medical care
Age per year	−0.26*	-	−0.02*	0.01
Education	-	0.00	-	-
Income	−0.15+	−0.09	−0.18*	−0.05
Life threat	−0.13***	−0.98**	−0.64*	−0.32
Victim-offender relationship	0.91***	−0.79**	0.77**	0.64*
Interpersonal trauma	−	−	0.23**	0.21**
Contextual trauma	0.16*	0.10	-	-
Delayed disclosure	-	0.61**	-	−0.64*
Told informal source	-	−0.99+	-	-
Tangible support	-	0.73***	-	0.82**
Mixed social reactions (UA)b	-	0.23*	-	-

Note.

⁺p < 0.10;

^*p < 0.05;

^**p < 0.01;

^***p < 0.001.

^aBlock: Significant factors from each of three variable sets.

^bUA: unsupportive acknowledgement.

Discussion

The present study’s findings showed that in a large diverse sample of African American sexual assault survivors some demographic and pre-assault characteristics, other traumas and assault characteristics, disclosure timing, and social reactions of tangible aid were associated with immediate and ever seeking of medical care. That older age and income related to less medical care seeking, perhaps related to poverty that existed at high rates for older African American women who were less likely to be insured and/or have access to and use medical care (Snipes, Wilson, Esparza, & Jones, 2009; Wyn, Ojeda, Ranji, & Salganicoff, 2004). In our sample, most survivors had incomes of under $20,000 per year, and just under half had health insurance at the time of assault as reported on the initial mail survey (49%); however, that variable did not reach statistical significance in preliminary analyses and may not reflect insurance status at the time of post-assault help-seeking, which is often delayed. Future research needs to examine that variable as well as examine correlates of help-seeking for women with and without insurance at the time of assault, especially as past similar research on survivors generally has shown that lack of insurance can also affect the quality of care, not just access to care, leading to revictimizing experiences (Starzynski, Ullman, & Vasquez, 2017).

The finding that stranger assaults were associated with greater medical care seeking both immediately and ever is not surprising, given the greater legitimation of stranger rape in society, increased likelihood of women labeling it as rape, and potentially greater health concerns (e.g., somatic symptoms, STIs, pregnancy) following such assaults (Conoscenti & McNally, 2006; Littleton, Breitkopf, & Berenson, 2008). Perceived life threat was related to less medical care seeking, perhaps due to greater PTSD that may have led survivors to withdraw and isolate and be less willing to seek formal help (Roberts et al., 2011). Importantly, survivors with more other interpersonal and contextual traumas had greater medical care seeking, which may suggest that the context of sexual violence is unique at least for the African American women in our sample who faced high rates of other forms of violence, including child abuse, domestic violence, and community violence, all of which also have negative health effects (Sabri & Gielen, 2017). The cumulative health impact of multiple forms of violence likely exacerbates the impact of sexual assault on African American women’s health, as found in research on cumulative trauma exposure in this population (Long & Ullman, 2013; Sabri & Gielen, 2017).

Delayed disclosure of assault was related to less medical care both immediately and ever, which may reflect lower severity and/or known perpetrator assaults, but also may reflect other internal and external barriers for African American survivors. Greater tangible support/aid reactions from those told about assault was related to seeking greater medical care, which may reflect receipt of help from such sources which often occurs and/or that social reactions were assessed from all possible informal and formal sources. Mixed social reactions were related to more immediate medical care seeking, which may reflect distinct sets of support sources women told post-assault yielding combinations of social reactions that gave them needed support yet still needing medical care for assault impacts.

Limitations

The study had several limitations. First, the use of a retrospective convenience sample of sexual assault survivors who volunteered to participate in a study of previously disclosed unwanted sexual experiences provided a sample that was unlikely to be representative of survivors overall and may differ from a non-volunteer sample in terms of assault experiences, post-assault functioning, service-seeking, and social desirability bias and thus reduce the generalizability of the findings. Second, all women in the study had to have told someone in the past about their assault, which omitted those who never told before, thus precluding us from comparing nondisclosers to disclosers in the present study. Third, we could not link help seeking to specific sexual assault disclosures, nor determine the time ordering of correlates and outcomes based on the measures and cross-sectional study design.

Fourth, measures of medical care seeking were single item questions regarding assault disclosure immediately post-assault or ever to medical professionals that may underestimate the specific potentially myriad sources of medical care sought. Data were not obtained on reasons for telling or not telling medical professionals, and we could not separate general from assault-related medical care seeking, which may differ. Many women may seek medical care for health problems that are in fact assault-related but not disclose their assault history, and professionals may not all screen women for victimization histories or recognize injuries or other health concerns to be potentially due to sexual assaults. Much more comprehensive quantitative and qualitative data are needed to understand women’s health care seeking decisions and experiences post assault, including data from formally assessed medical history screening data not available in the current study. For instance, we did not ask questions about specific health concerns sexual assault victims may have had, such as about pregnancy and STDs. Finally, women in our sample reported a low level of depressive symptoms, but such symptoms often manifest differently in African Americans (Mental Health America, 2013), and thus may have been underestimated, given the brief self-report screening measure utilized. Longitudinal data are needed to examine patterns of health care seeking over time in survivors and to examine whether access to health care and other trauma informed services, like Sexual Assault Nurse Examiners (SANEs), may affect women’s health following assault. This is especially relevant for African American survivors, many of whom live in underserved areas for medical care and specialized services of rape crisis centers or hospitals with SANE programs.

Implications for intervention and treatment

Interventions and treatment for sexual assault need to address histories of other traumas that may influence odds of women seeking medical care for sexual assault. Demographics, such as age and socioeconomic status, affect seeking of care in African American women, and may mean more outreach is needed for other sociodemographic groups less likely to seek care. Stranger and life-threatening assaults are still more likely to lead for formal treatment seeking, so acquaintance and partner assault survivors may be less likely to seek medical care, especially immediately post-assault. Tangible support/aid from those one tells post-assault (most often family/friends/partners) plays a key role in facilitating formal help seeking, and negative/mixed social reactions may also play a role, so these reactions should be assessed and targeted in interventions and treatment. National samples of sexual assault survivors are needed to examine background, assault characteristics, trauma histories, and post-assault individual and social network factors on various forms of informal and formal help-seeking in African American women. Findings highlight that some factors may be unique in predicting post-assault and lifetime medical service seeking in this population, which is crucial for targeting clinical treatment and prevention efforts to African American survivors.

Implications for research, advocacy, and medical treatment providers

Various insights regarding African American sexual assault survivors service seeking and strategies for recovery can be gleaned from this study. A full physical, psychological, and sexual trauma history screening should be part of standard intake procedures at every agency. Greater outreach to older, lower income African American survivors may be warranted, especially in settings where they are comfortable, such as churches. Survivors who perceive life threat may have PTSD and require referrals from crisis counseling or mental health treatment settings and/or medical settings for physical health concerns. Rapes by strangers that often involve greater injury, are more legitimated as crimes are more likely to lead to medical attention, but women may also have other histories of trauma, including other CSA and/or known offender assaults to be considered in formulating health care plans. This is likely the case in our sample as known perpetrator assaults were associated with delayed disclosure. Interpersonal and contextual traumas should also be assessed to understand if women have ongoing safety concerns and risk of repeated assaults with appropriate referrals for domestic and/or community violence support sources. Telling informal sources was marginally related to greater medical care seeking as expected but further work is needed on which support sources women get specific types of support from and how that may facilitate or thwart medical help seeking. Both positive tangible support and mixed social reactions related to medical care seeking as expected demonstrating that positive and even mixed responses from others may facilitate women obtaining medical attention post-assault and or longer-term. Research is needed on correlates of African American sexual assault survivors’ assault-related sequelae including their health care needs and pathways to necessary medical care both post-assault and in the longer-term to improve their physical health and well-being.