Ending intimate partner violence among women living with HIV: How attachment and HIV stigma inform understanding and intervention

ABSTRACT

This mixed-methods, exploratory study examined why women living with HIV (WLHIV) stay in intimate partner violence (IPV) relationships and what helps end IPV in their lives. WLHIV (n = 108) who experienced IPV and were patients at two HIV primary care clinics in San Francisco completed quantitative surveys; 15 participants also completed a qualitative interview. Qualitative data showed HIV stigma was the most cited reason for staying in an IPV relationship, followed by substance use, and then by themes of attachment insecurity. Quantitative data indicated that most participants accessed HIV services and rated them as more helpful than other community resources to end IPV in their lives. Enduring attachment relationships with HIV medical and social service providers and their attachment-enhancing actions and attributes were critical to participants addressing IPV and coping with HIV stigma. This study highlights the important role that HIV providers and clinics can play in addressing IPV among WLHIV.

KEYWORDS:

• attachment
• stigma
• HIV/AIDS
• trauma
• IPV
• substance use

Introduction

Women living with HIV (WLHIV) experience disproportionately high rates of trauma compared to the general population of women in the United States (Machtinger et al., 2012). The majority of WLHIV in the United States have experienced childhood abuse (58.2%) and/or intimate partner violence (IPV) in their lives (55.3%), and these percentages are substantially greater than those of women in the general population (Brezing et al., 2015; Machtinger et al., 2012).

WLHIV who experience IPV – including sexual violence, physical violence, stalking, and/or psychological aggression by a current or former intimate partner – have difficulty engaging in regular health care (Lichtenstein, 2006), report less compliance with their HIV treatment (Brady et al., 2002), and report low levels of health-related quality of life (McDonnell et al., 2005). While multiple studies have found similar rates of IPV among WLHIV and socio-demographically similar (urban, poor, women of color) HIV-negative women (Burke et al., 2005; Cohen et al., 2000), WLHIV experience more frequent and severe abuse (Wyatt et al., 2002). Not surprisingly, WLHIV’s mental health and quality of life are more negatively affected by IPV compared to HIV-negative women (McDonnell et al., 2005).

In both HIV-positive and HIV-negative women, higher rates of IPV are associated with higher rates of alcohol and substance use (Burke et al., 2005). Although WLHIV have higher overall rates of substance use than HIV-negative women, the relationship between IPV and use of illicit drugs is stronger among HIV-negative women (Burke et al., 2005). Research has shown comparable help-seeking behaviors of socio-demographically similar HIV-positive and HIV-negative women who experience IPV, with the majority (70%) of women not getting help for violent episodes (O’Campo et al., 2002). Those who receive help to leave violent relationships receive help from family (65%), friends (52%), and the police (49%); only (11.5%) receive help from social services to end IPV (O’Campo et al., 2002).

HIV stigma

Structural traumas such as HIV stigma may interact with and influence WLHIV’s experiences of IPV. HIV/AIDS-related stigma, defined as “prejudice, discrediting, and discrimination directed at people perceived to have AIDS or HIV” (Herek, 1999, p. 1107), can lead to internalization of the stigmatized identity and impaired psychological health and functioning (Goffman, 1963). Experiencing HIV stigma is associated with poor adherence to antiretroviral therapy (ART) (Katz et al., 2013), greater HIV symptoms (Vanable et al., 2006), and worse depression, anxiety, and hopelessness (Riggs et al., 2007). WLHIV are more likely than men living with HIV (MLHIV) to experience the multiple health consequences of HIV stigma, including depression, social isolation, and rejection (Lichtenstein et al., 2002); and WLHIV also report higher levels of perceived stress than MLHIV (Riggs et al., 2007). When African-American WLHIV disclose their HIV status, they receive less social support and are more likely to be rejected by their families and communities than white WLHIV and MLHIV (Clark et al., 2003; Lichtenstein et al., 2002).

Attachment theory

WLHIV who experience IPV may also experience attachment insecurity. While stigma and internalized stigma reflect one’s relationship with society and taking in society’s negative view of self, attachment theory focuses on the internalization of intimate relationships throughout one’s lifetime – initially emphasizing the child and caretaker bond and later including romantic and other significant relationships (Bowlby, 1988; Sable, 2008). Consistent, sensitive, and empathic relationships create attachment security that facilitates affect regulation and sense of internal security. Threatening, abusive, and erratic relationships instill attachment insecurity, fear, anxiety, and/or avoidance (Davis, 2011; Sable, 2008).

Internal working models of attachment refer to the mental templates of what one comes to expect of relationships, the world, and the self that are shaped by early intimate relationships (Bowlby, 1969; Sable, 2008). Attachment styles categorize the way people behave, feel, and think in interaction with intimate others and are rooted in one’s internal working models of attachment (Collins et al., 2004). Secure adult attachment style includes the ability to form dependable relationships that provide physical and psychological protection and affect regulation (Sable, 2008). Adults with insecure attachment styles have difficulty with anxiety about rejection, abandonment, or being unloved, and/or avoidance of intimacy or interdependency with others (Collins et al., 2004; Hazan & Shaver, 1987). To create or influence positive internal working models and secure attachment style, attachment figures must provide two important functions. First, they need to act as a safe haven by responding sensitively to the distress of their child or loved one through soothing and comforting contact to facilitate a sense of trust and confidence that others will be helpful when needed (Bowlby, 1969; Feeney & Collins, 2004). Second, attachment figures must also provide a secure base of available support and noninterfering encouragement from which a person can confidently explore the world independently (Bowlby, 1969; Feeney & Collins, 2004). The perpetration of IPV stands out in its contrast to providing safe haven and secure base functioning as the attachment figure is the source of danger and distress, is overly controlling, and aims to prevent autonomy.

While research on attachment style among people living with HIV (PLHIV) is limited, high rates of attachment insecurity have been found in low-income, urban populations of PLHIV (Riggs et al., 2007). Insecure attachment styles among PLHIV have been associated with poorer coping, overall psychological distress, and HIV stigma (Riggs et al., 2007; Turner-Cobb et al., 2002). PLHIV with secure attachment styles report significantly less stress and depression compared to those with insecure attachment styles (Riggs et al., 2007). WLHIV who experience IPV may have high rates of attachment insecurity, because attachment insecurity is significantly associated with and influenced by key challenges facing WLHIV – childhood trauma (Alexander, 2009), HIV stigma (Riggs et al., 2007), and chronic illness (Feeney, 2000).

The complex interplay of IPV, HIV stigma, insecure attachment, and the variety of poor health and behavioral outcomes experienced by WLHIV requires further research and clarification for providers to most effectively care for WLHIV. To address this gap in the research literature and provide practical guidance to providers, this exploratory study aims to: 1) identify the factors associated with WLHIV staying in or leaving IPV relationships (or otherwise ending the violence in such relationships); 2) understand the specific roles that HIV stigma and attachment play in WLHIV’s IPV relationships and ending IPV in their lives; and 3) learn how medical and social service providers can support WLHIV to safely end IPV in their lives.

Methods

A mixed methods study design was used to understand the breadth and depth of IPV experiences among WLHIV. Participants were cis-and transgender WLHIV who reported ever experiencing IPV in adulthood and who obtained health care at one of two women’s HIV primary care clinics in San Francisco. The study was approved by the UCSF Committee on Human Research and the Smith College School for Social Work Human Subjects Review.

Procedures

Medical and social service providers at the two primary care clinics referred patients to the study, or patients self-referred in response to flyers posted in the clinics. A researcher was stationed at each program’s weekly clinic over a 3-month period to recruit participants for the study. The researcher met with each potential participant in a private space in the clinic to screen for eligibility, review the consent form, and explain the instructions for the different scales on the survey. The researcher remained available to answer any questions while the participants completed the approximately 30-minute paper and pencil survey on their own. Surveys were later administered by the researchers once it became clear that literacy and comprehension were significant barriers for some of the participants. Participants received a $15 grocery gift card upon completing the survey. A total of 108 WLHIV participated in this quantitative survey.

Qualitative interviews were conducted with 15 of the survey participants who were randomly selected from a pool of participants who had indicated on their surveys that they had experienced IPV since their HIV diagnosis, but were no longer in an IPV relationship. These criteria for participating in the qualitative interview were chosen to gain in-depth understanding of how IPV relationships ended and the impact of HIV and HIV stigma on those relationships. Qualitative interviews were completed in a private clinic office over a 4-month period (post-survey data collection). Participants received a $25 grocery gift card for the qualitative interview, which lasted 60 to 90 minutes. Qualitative interviews were audio-recorded and transcribed.

Measures

Demographics. The demographic portion of the survey included gender identity, age, race/ethnicity, income amount and source, employment status, highest level of education, years since HIV diagnosis, and length of time being a patient at their HIV primary care clinic.

Childhood Abuse. Three questions from the Childhood Trauma Questionnaire – Short Form (CTQ-SF) were used to screen for physical, sexual, and emotional abuse (Thombs et al., 2007). The questions for physical and sexual abuse are an established two-question screen taken from the CTQ-SF and are 84.8% sensitive and 88.1% specific for accurate screening (Thombs et al., 2007). A third question was chosen from the CTQ-SF to screen for emotional abuse – “When I was growing up, people in my family said hurtful and insulting things to me.”

IPV. The Revised Conflict Tactics Scale Short Form (CTS2S) was used to measure prevalence and chronicity of IPV (Strauss & Douglas, 2004). The CTS2S asks participants to quantify their experiences of different types of abuse and behavior, including: sexual coercion, injury, psychological aggression, physical assault, and/or negotiation (as a strategy to manage conflict) during a particular time – in the case of this study, during the last relationship in which the participant experienced IPV. Concurrent validity between the five constructs in CST2S and the longer, widely used CTS2 ranged from .77 to .89 for participant reported behaviors, and .65 to .88 for participant report of partner behaviors (Strauss & Douglas, 2004). Participants were also asked researcher-developed (K.D.) multiple-choice questions about IPV (e.g., are you currently in an IPV relationship; if no, how did your most recent IPV relationship end).

HIV Stigma. HIV stigma was measured using the HIV Stigma Scale (Berger et al., 2001), a 40-item instrument with four subscales – Personalized Stigma, Disclosure Concerns, Negative Self-Image, and Concerns with Public Attitudes about People with HIV/AIDS. Along with calculating an overall perceived HIV stigma score, the subscale score for Negative Self-Image was recorded as it reflects the notion of internalized HIV stigma. The HIV Stigma Scale is internally reliable. Cronbach’s alpha for the overall scale and all subscales was .90 or greater, and test-retest-reliability was .87 or greater (Berger et al., 2001).

Attachment. Adult attachment was measured using the 36-item Experiences in Close Relationships (ECR) scale (Brennan et al., 1998). The measure can be scored to classify participants’ romantic attachment styles dichotomously (secure versus insecure) (Collins et al., 2004). The instrument demonstrates high internal consistency (Brennan et al., 1998).

Substance use. Participants completed the 4-item CAGE-AID as a screening for a history of alcohol and/or other drug use problem. The CAGE-AID exhibited .79 sensitivity and .77 specificity for one or more yes answers and .70 sensitivity and .85 specificity to two or more yes answers (Brown & Rounds, 1995). Participants were also asked two separate yes/no questions about whether they or their partners were using substances in their most recent IPV relationship.

Services utilized and rated. This researcher-developed (K.D.) measure asked participants to identify what services, agencies, and people attempted to help them end IPV in their lives. Personal and professional resources identified in previous research (O’Campo et al., 2002) were included in the scale (e.g., family, friends, police, IPV shelter, etc.) plus HIV-specific resources. Participants then rated (on a 5-point scale) how helpful or unhelpful each used resource was.

Qualitative data. The qualitative interview guide consisted of both open-ended and directed questions whose scope covered IPV, experiences of living with HIV and disclosure of HIV status, childhood abuse, and resource questions about who and what was helpful to participants in coping with IPV and HIV stigma. The interview questions specifically addressed stigma, attachment, and supports and served to fill in the details of the broad quantitative questionnaire data and gain more thorough understanding of what women needed to end IPV in their lives.

Data analysis

Data from the quantitative surveys were analyzed using IBM SPSS Statistics for Windows, Version 22.0. Descriptive statistics were calculated for all variables. Bivariate analyses were used to test associations between primary variables (overall perceived HIV stigma, internalized HIV stigma, IPV, substance use, child abuse, attachment). Pearson’s Correlation Coefficients were used to investigate the relationships between HIV stigma and IPV, substance use, and child abuse, as well as the relationships between internalized stigma and these same variables. Cross-tabulations, Chi Square Tests, and Mann-Whitney Tests were used to examine relationships between types of attachment and the secondary variables. T-tests investigated the relationship between insecure attachment style and overall perceived HIV stigma and internalized stigma.

The qualitative interview data were analyzed using an interative content analysis approach (Patton, 2002) in both a deductive and an inductive manner as relevant concepts were searched for within the data and other patterns and themes were allowed to emerge from the data. Atlas.ti 7.0 software was utilized as an annotation and note-keeping aid. The primary researcher (K.D.) open-coded the interviews and developed 47 different codes that were organized into 10 families. Major themes were developed within the families and text was organized under the themes. Reliability was attained by another reviewer checking families and themes for internal homogeneity and external heterogeneity and by using the quantitative data to triangulate the data. Pseudonyms are used to represent qualitative participants in the Results section below.

Results

Table 1. Participant demographics, trauma, and substance use (N = 108)

	Mean (SD), range, or Number (%)
Age	49.6 (SD 8.6), range 28–66
Gender Identity
Cisgender Female	96 (89.0%)
Transgender Female	12 (11.0%)
Race/Ethnicity
African-American/ Black	63 (58.3%)
Caucasian/White	19 (17.6%)
Multi-ethnic	10 (9.3%)
Latina	5 (4.6%)
Asian/ Pacific Islander	3 (2.8%)
Native American	3 (2.8%)
Additional Race/Ethnicity	5 (4.6%)
Monthly Income	$918.00 (SD $481.90), range $0-$3500
Source of Income
Disability	95 (88.0%)
Employment only	5 (4.6%)
No income	3 (2.8%)
Other source of income	5 (4.6%)
Education
Less than high school	24 (22.2%)
High school/equivalent	63 (58.3%)
Associate’s Degree	16 (14.8%)
Bachelor’s Degree	4 (3.7%)
Graduate Degree	1 (1.0%)
Years Since HIV Diagnosis	15.7 (SD 7.8), range 1–31
Years in HIV Care at Study Clinics (N = 106)	9.0 (SD 6.4), range 0–25.7
Childhood Abuse Experienced
Childhood Emotional Abuse	94 (87.0%)
Childhood Physical Abuse	74 (68.5%)
Childhood Sexual Abuse	77 (71.3%)
Any Type of Childhood Abuse	102 (94.4%)
Intimate Partner Violence Experienced (N = 104)
Physical Assault	101 (97.1%)
Psychological Aggression	104 (100.0%)
Injury	97 (93.3%)
Sexual Coercion	82 (78.8%)
HIV Stigma
Overall Perceived HIV Stigma (possible range 40–160)	97.0 (SD 20.3), range 46–151
Negative Self-Image Subscale (possible range 13–52)	30.0 (SD 7.9), range 13–52
Attachment Style (N = 107)
Insecure Attachment	93 (87.0%)
Secure Attachment	14 (13.1%)
Substance Use
Participant history of substance use problem (CAGE-AID)	92 (85.2%)
Participant substance use during most recent IPV relationship	66 (61.1%)
Partner substance use during most recent IPV relationship	87 (81.3%)

The 108 participants were predominantly very low income women of color and the majority had a high school education or less (Table 1). A small subset of participants identified as transgender; very few participants were employed. Participants had been living with HIV and receiving care at their HIV primary care clinic for many years. The overall study population suffered a high degree of trauma, insecure attachment, and substance use, and a mid-range level of overall perceived and internalized HIV stigma. Most participants screened positive for childhood abuse, insecure attachment, and substance abuse history, and had experienced psychological aggression, physical assault, injury, and/or sexual coercion from intimate partners in adulthood. There were no statistically significant differences between survey and qualitative interview participants in terms of demographics, measures of IPV, substance use, childhood abuse, HIV stigma levels, and attachment style.

Why WLHIV stay in IPV relationships

Survey and interview data revealed key factors that kept participants in IPV relationships and those that helped them end IPV. Qualitative data revealed that the most common reason for staying in an IPV relationship was HIV stigma, followed closely by drugs and alcohol, and then attachment insecurity-related themes of sense of commitment and negative self-image.

HIV stigma. The study population had mid-level overall perceived HIV stigma scores and internalized HIV stigma scores (Table 1). While also in the mid-level range for both stigma measures, African-American participants reported significantly higher levels of overall perceived HIV stigma (M = 101.11, SD = 19.358, t(80) = −2.647, p = .010) and internalized HIV stigma (M = 31.33, SD = 7.746, t(80) = −5.386, p = .009) compared to white participants. Despite the survey data showing a mid-level range of stigma, almost all qualitative respondents stated that HIV stigma played a significant role in keeping them in an IPV relationship. At the time of the interviews, only three qualitative participants were in an intimate relationship, and all but one stated that being HIV-positive made them more hesitant to engage in intimate/sexual relationships. They described how their partners’ willingness to be in an intimate relationship with them despite their HIV status kept them in the abusive relationship. Regina shared,

I found out in 2007 that I was positive. To find someone who accepts you for that and never said nothing bad about me because of that, that meant a lot to me. I felt like I couldn’t have nobody else who would be okay with me having this disease … So that’s why I stayed, that’s why I allowed to be talked to crazy, because who’s going to want me?

They also described how their partners used their HIV status and HIV stigma to control them and keep them in the relationship. Keisha stated, “The guy that I used to be with, he had threatened if I got with someone else that he was going to tell them. He would always say ‘Nobody going to want you when they know you’re HIV-positive.’” HIV stigma led to fear that they would not find another partner who would accept them, and their partners used and amplified this fear to keep them in relationships. This fear likely developed from first-hand experience of being rejected for being HIV-positive as all but one participant reported experiencing rejection from others upon disclosing their positive HIV-status.

Drugs and alcohol. Most quantitative participants reported that they and their partners used substances in their most recent IPV relationship (Table 1). Three-quarters of the qualitative interviewees reported that substance use played an important role in staying in abusive relationships. Substance use led to isolation from support systems that could have helped them leave the violent relationships. Kim discussed ignoring her friends’ attempts to help her leave an abusive relationship because she and her partner used drugs together and he supplied her with drugs. “I pushed most of my friends away because I was doing drugs. They kept telling me, ‘You gotta leave him.’ And I kept telling them, ‘You don’t know.’” Addiction led participants to prioritize substances above all else including personal safety and prevented participants from acknowledging and addressing the violence in their relationships. Beverly shared,

My relationship with drugs and alcohol … the drugs covered a lot of what was there. It just like blinded me, like something over my eyes. I think if drugs and alcohol weren’t involved, I wouldn’t have been in this relationship. The only way we even knew each other was through drugs and alcohol.

Attachment insecurity. The vast majority of the quantitative sample had insecure attachment style (Table 1). Insecurely attached participants had experienced significantly more childhood abuse of any type (U = 395, p = .017), childhood sexual abuse (U = 401.5, p = .018), and childhood emotional abuse (U = 445, p = .040), as well as significantly more sexual coercion in their most recent IPV relationship (U = 425.5, p = .046) than securely attached participants. Insecure attachment was more highly associated with high levels of overall perceived HIV stigma (t(105) = 2.774, p = .007) and internalized stigma (t(105) = 3.098, p = .002) than secure attachment.

More than two-thirds of qualitative participants discussed themes of attachment insecurity as reasons for being and staying in an IPV relationship. They described commitment to their partners despite the abuse and that abusive relationships were what they knew from experiencing and/or witnessing abuse in childhood. Briana stated, “I didn’t see the women in our family not stick and stay [in their relationships] … I’ve seen the women go through a lot and I’ve seen them stick in it. It’s just for me, the way I came up … ”

Participants exhibited negative internal working models of attachment in their expectation of relationships and their views of themselves and the world. Specifically, they described childhood abuse influencing their enduring negative sense of self and that negative self-image keeping them in IPV relationships. Bernadette shared, “I didn’t look for men to be honest. I didn’t look for a man to love me. I looked for a man to use me because I had never been loved.” Kim stated, “I was always told that I would never amount to nothing; that I would end up just like my mom. There for a while I did. I was on drugs and alcohol, and I was fixin’ to lose my son.” Participants described internalizing negative messages about relationships and themselves; this contributed to their belief that abusive relationships were to be expected. Abusive dynamics with their partners mimicked relationships throughout their lives, and reflected and reinforced what they felt inside.

Why and how WLHIV end IPV relationships

Most survey participants (75%) reported that they were not in an active IPV relationship at the time the survey was conducted. When asked how their most recent IPV relationship ended, 81.5% of these participants selected at least one proactive response (i.e. “I ended the relationship” or “I moved away”). All but one of the qualitative participants described an event or situation that triggered them to end their last IPV relationship. These included escalation in abuse, partner’s abuse of their child(ren) or involvement of Child Protective Services, growing tired of abuse/hitting a limit, and feeling betrayed by their partners. They then either ended the relationship immediately or started the process of leaving or otherwise ending the abuse.

Table 2 Participants accessed personal and professional resources to end their most recent IPV relationship. The majority of participants received help from friends, family, primary medical care, case management, police, individual therapy, and substance use programs. Notably, 70% to 90% of the participants who utilized services affiliated with their HIV primary care clinic rated them as helpful or very helpful (Figure 1). Family, substance use programs, and the emergency medical department were other entities from which the majority of participants received help that were rated very positively. Group therapy ranked highest in helpfulness ratings as 90% of participants who utilized this service ranked it as helpful or very helpful. Figure 1

Table 2. Use of supports to end IPV

	Number (%)
Friends (N = 107)	84 (78.5%)
Family (N = 103)	66 (64.1%)
Primary Medical Care (N = 106)	63 (59.4%)
Case Management (N = 106)	59 (55.7%)
Police Department (N = 105)	58 (55.2%)
Individual Therapy/ Counseling (N = 107)	50 (46.7%)
Substance Use Program (N = 106)	48 (45.3%)
Emergency Department Medical Care (N = 106)	41 (38.7%)
Medical Social Work (N = 105)	39 (37.1%)
Mental Health Clinic (N = 105)	39 (37.1%)
Church/Spiritual Community (N = 106)	39 (36.8%)
Psychiatry (N = 106)	34 (32.1%)
Domestic Violence Shelter (N = 104)	29 (27.9%)
Group Therapy (N = 105)	29 (27.6%)
Domestic Violence Specific Agency (N = 105)	27 (25.7%)
District Attorney’s Office (N = 104)	26 (25.0%)
Couples Therapy/Counseling (N = 106)	21 (19.8%)
Prenatal Medical Care (N = 106)	17 (16.0%)

Supports available at or affiliated with the study clinics are bolded.

Figure 1. Helpfulness ratings of IPV supports

Attachment to HIV clinics and providers. Participants utilized the services affiliated with their HIV primary care clinics more commonly and rated them more positively than most other supports (including IPV specific resources). They also rated HIV services as more helpful than friends and the police (two of the most highly utilized resources in this study and previous research). Participants described their positive attachment to their HIV primary care clinic and providers as key to being able to address IPV and other difficult experiences. All of the qualitative participants reported having strong, long-term, even familial relationships with their HIV medical and social service providers. Briana shared,

My doctor always is my girl … She’s always my right hand … Without her, I don’t know what I would do. Me and her, that’s my sister. To really take her in like a sister, like nobody really know me in and out outside my family like she do.

Lisa stated,

I love everybody [at the clinic]. I’ve been knowing them for a long time. It took me a long time to open up because I was very shy and very ‘you don’t tell nobody nothing.’ It was something with my doctor that I liked about him. I always felt a sense of trust … It was like okay, these people care …

Ongoing, long-term contact helped participants build a sense of trust and closeness with their health care team, allowing them to share what was going on in their lives.

Qualitative participants identified specific actions taken by their HIV providers that helped in addressing IPV: 1) inquiring about relationships, 2) listening and supporting (without pushing) participants to make their own decisions, 3) reminding participants that they deserve better than an abusive relationship, and 4) providing a sense of hope. Michelle describes the care she received at her clinic after an incident of being choked by her partner.

The people here [in clinic], they gave me a lot of support. They talked to me. They made sure I wasn’t currently being physically abused and that I was in a safe place … I was so hysterical when I came in about it, but they calmed me down. I know that if something happened that I could talk to them about anything … I was so scared to tell anybody about my relationship, but I felt like I couldn’t take it alone. I decided to let them know what was going on.

The clinic’s helpful response to her distress created a strong sense of trust and a sense that she could lean on her providers during difficult times. Lisa discussed the patience and consistency of her clinic care team as she struggled with whether or not to leave her abusive partner.

My therapist … She helped me because I talked to her for about a year about that relationship … She and my doctor and [the medical social worker], they helped me through that time … My therapist was dealing with my emotional part and helped me to know that I wasn’t going to be in this space forever and that I could do things to make me feel good … And [the medical social worker] just, she had an ear … She just listened …

The team at the HIV clinic provided the support she needed to come to her own decisions.

Providers’ attachment-enhancing qualities. Participants also described key provider characteristics that helped them cope with IPV and also reduced HIV stigma, including: 1) accepting and nonjudgmental attitudes, 2) attentiveness and engagement, 3) availability and accessibility, and 4) setting limits and respecting boundaries. These characteristics reflect the qualities of attachment figures that carry out safe haven and secure base functions that build and enhance attachment security across the life span. Gloria emphasized feeling completely comfortable talking to her doctor about anything and knowing she would be accepted.

When you have that communication where you can talk to your doctor about any part of your body, you feel comfortable with them … From the time I was diagnosed in ’96 all the way up until now, so far, I had a doctor where we can get in this four corner room and it can just be me and them and we can talk about anything that’s going on with me.

Kim discussed the availability and responsiveness of her provider.

My doctor now, I have her email address. I didn’t have to go through the front desk and then

this person and this person and then the doctor. I had her email address. If I emailed her in

the morning, by the end of the day, I’d have a response. I had her personal phone number out

here to the hospital because I kept getting kidney infections. I’d call her and I’m on my way

to the hospital. One night, I was in the emergency room around 7-8 o’clock, she pops up ‘I

just had to come check on you before I go home.’ Awesome.

Her doctor’s accessibility and clear investment in her health had a very positive impact on her.

Discussion

This study highlights the crucial role that HIV medical and social service providers can play in addressing IPV for WLHIV by acting in ways that promote healthy attachment and reduce HIV stigma. HIV stigma, substance use, and attachment insecurity are primary factors described by participants for keeping them in abusive relationships.

Despite the overall survey population having only a mid-range level of perceived HIV stigma and internalized stigma, almost all qualitative study participants cited HIV stigma as a key factor keeping them in an IPV relationship. Perhaps, because HIV transmission is strongly associated with sex with an intimate partner, the weight of stigma specific to intimate relationships and fear of not being able to find another partner who would accept their HIV status remained strong even if participants had found ways to mitigate HIV stigma to a mid-range level in their day-to-day lives. African-American participants had significantly higher levels of overall perceived and internalized HIV stigma as well as higher levels of disclosure concerns compared to White participants. This racial difference matches previous research in which African-American women identify substantial perceived HIV stigma from their communities and society and greater concerns about HIV disclosure (Clark et al., 2003). It is important for HIV medical and social service providers to understand this poignant layer of vulnerability for African-American WLHIV; they may be more isolated and alone due to their experiences of HIV stigma and require more sensitive outreach and support to engage in care and to address IPV.

Substance use was prevalent among participants across their lives, in their most recent IPV relationships, and among their partners in their most recent IPV relationship, and influenced them to stay in their abusive relationships. Three-quarters of participants who engaged in substance use treatment rated it as helpful/very helpful in ending IPV in their lives, pointing to substance use treatment as a potentially powerful IPV intervention.

Attachment insecurity was prevalent among all participants, and qualitative participants discussed themes of attachment insecurity keeping them in IPV relationships. Specifically, participants described an internalized negative sense of self that was instilled in them by abusive attachment figures and environments during their childhoods; they repeated in adulthood the abusive relationships they experienced in childhood. Notably, attachment insecurity was significantly correlated with overall perceived and internalized HIV stigma, sexual coercion in most recent IPV relationship, and childhood abuse. Attachment insecurity highlights the enduring impact of these multiple traumas in creating and reinforcing participants’ negative sense of self, relationships, and the world and points to the need for intervention that addresses attachment insecurity and trauma.

Most survey participants utilized HIV medical and social services to a high degree and rated them as being helpful or very helpful in ending IPV suggesting participants were much more connected to medical and social services than previous studies have found and that participants find these services particularly helpful in addressing IPV. These data speak to the value of this multidisciplinary model of HIV care in which medical providers, social workers, case managers, therapists, and pharmacists collaborate to provide comprehensive and holistic assessment and treatment of medical, psychological, and social needs.

This model of multidisciplinary HIV care demonstrates how the Circle of Security in Adulthood attachment model (Feeney & Collins, 2004) may be useful to promote secure attachment and reduce HIV stigma in WLHIV who experience IPV (Davis, 2011). The Circle of Security model describes dyadic attachment, in which there is a caregiver (e.g., HIV provider) and a care receiver (e.g., clinic patient). The caregiver acts as both a safe haven and a secure base, providing security and affect regulation for the care receiver. Qualitative data repeatedly revealed participants’ experience of their HIV providers providing safe haven responses such as comfort, reassurance, and hope in the face of their life stressors and threats (IPV episodes, HIV diagnoses, etc.). Clinics provide healthy attachment experiences that could potentially combat participants’ abusive internal working models held from extensive trauma histories. For patients who receive satisfactory safe haven caregiving, develop physical and emotional stability, and are no longer in IPV relationships, the clinics can then act as a secure base of encouragement and non-interfering support for their goals and plans for the future – providing referrals and linkage to volunteer or job training or other activities. The clinics remain available and ready to provide safe haven support if crises arise again.

This model of HIV care can be supported and enhanced by adopting a Trauma-Informed Health Care (TIHC) framework (Machtinger et al., 2015, 2019) that includes substance use services and a strong collaboration with IPV-specific community resources as part of its response to trauma and IPV. Because participants were more likely to access HIV services than IPV-specific community resources to end IPV in their lives, it is important to find ways to integrate IPV expertise and services into HIV care. Partnership with IPV-specific agencies that can provide extensive training for HIV medical and social service providers on addressing IPV as well as an advocate to counsel and support patients are beneficial additions to this effective model of interdisciplinary care (Kimberg, 2007).

This study had limitations. The survey population size of 108 was small and relatively homogeneous. Fewer variables or a larger sample would make for a more rigorous quantitative study. Additionally, due to the small number of transgender participants enrolled in the study, their potentially unique experiences of IPV, attachment, stigma, and substance use could not be adequately captured and compared to the predominantly cisgender study population. A future study focused specifically focused on transgender WLHIV who experience IPV is needed. Adaptations were made to the data collection procedure after the start of the data collection process. In addition, while attachment theory is very useful in understanding and determining positive ways to intervene in the lives of WLHIV who experience IPV, it does not address the structural forces that may contribute to attachment figures’ actions and the caregiving environment. Including HIV stigma in this study helps to address some of those external forces, and future research should investigate other structural forces such as racism and poverty.

Conclusion

This mixed method study provides valuable insights about how medical and social service providers can support WLHIV who experience IPV to end IPV in their lives. The findings show that HIV stigma, substance use, and insecure attachment are key factors keeping WLHIV in IPV relationships. HIV primary care services are helpful in ending IPV among WLHIV, and participants describe the attachment enhancing behavior and characteristics of their HIV providers as being particularly beneficial to ending IPV in their lives and in reducing HIV stigma. The Circle of Security in Adulthood model of attachment provides a framework for understanding how multidisciplinary HIV programs can support WLHIV who experience IPV. This study informs and justifies larger scale, longitudinal studies to better understand the multifaceted experience of IPV among WLHIV and of interventions that address key reasons why WLHIV stay in abusive relationships – HIV stigma, substance use, and insecure attachment.

Disclosures

The authors report no real or perceived vested interests related to this article that could be construed as a conflict of interest.

Acknowledgments

We are grateful to all of the women who participated in this study. We also thank the UCSF Women’s HIV Program and the Women’s Clinic at Zuckerberg San Francisco General Hospital Ward 86 for their support of this project. Thank you to Laura Barhoum for data collection support, and Dr. Marsha Pruett and Dr, Kathryn Basham of Smith College School for Social Work for their advisement and guidance. This research was supported by the Sarah Haley Memorial Trauma Fund Award and the Smith College School for Social Work Clinical Research Institute Grant.

Author Contributions

Katy Davis, PhD, LCSW conceptualized and designed the study, led data collection, analyzed the data, and wrote the original draft of this manuscript. Carol Dawson-Rose, PhD, RN helped conceptualize the study and provided guidance on study design and data analysis. Yvette Cuca, PhD, MPH contributed to the revision of the manuscript. Martha Shumway, PhD contributed to the revision of the manuscript. Edward Machtinger, MD provided overarching support of the project including staff resources to help with data collection, and he contributed to data analysis and revision of the manuscript.