HIV, Violence and Women: Unmet mental health care needs

Abstract

Background

HIV-infected (HIV+) women have high rates of Gender Based Violence (GBV). Studies of GBV find that approximately 50-90% of survivors develop mood and anxiety disorders. Given that women in sub-Saharan African constitute the largest population of HIV+ individuals in the world and the region's high GBV prevalence, mental health research with HIV+ women affected by GBV (HIV+GBV+) in this region is urgently needed.

Methods

Qualitative methods were used to evaluate the mental health care needs of HIV+GBV+ female patients at an HIV clinic in the Kisumu County, Kenya. Thirty in-depth interviews and four focus groups were conducted with patients, healthcare providers and community leaders. Interviews were transcribed, translated and analyzed using qualitative data software.

Results

Respondents stated that physical, sexual and emotional violence against HIV+ women was widely prevalent and perpetrated primarily by untested husbands accusing a wife of marital infidelity following her positive HIV test result. Mental health problems among HIV+GBV+ women included depressive, anxiety, traumatic stress symptoms and suicidal thoughts. Participants opined that emotional distress from GBV not only caused HIV treatment default, but also led to poor HIV health even if adherent. Respondents agreed that mental health treatment was needed for HIV+GBV+ women; most agreed that the best treatment modality was individual counseling delivered weekly at the HIV clinic.

Limitations

Emotional distress may be higher and/or more varied among HIV+GBV+ women who are not engaged in HIV care.

Conclusions

Mental health care is needed and desired by HIV+GBV+ women in Kisumu County, Kenya.

Introduction

HIV infection in women is consistently associated with violent victimization (Silverman 2010). HIV-infected (HIV+) women report extraordinary levels of Gender Based Violence (GBV), particularly intimate partner violence (IPV) (Cohen et al. 2000; Jewkes et al. 2010; Maman et al. 2000). Studies of women affected by GBV find that 60-90% develop anxiety disorders, including Posttraumatic Stress Disorder (PTSD), and approximately 50% develop mood disorders, such as depression (Rees et al. 2011; Golding 1999). In the setting of HIV, depression and PTSD not only cause suffering and debility, but also correlate with decreased adherence to antiretroviral therapy (ART), a key factor in treatment failure (Blashill, Perry, and Safren 2011; Boarts et al. 2006; Gonzalez et al. 2011; Starace et al. 2002). Interventions that target depression have shown sustained improvements in ART adherence (Sin and Dimatteo 2013).

Despite awareness of GBV and associated mental illness among HIV+ women, little interventional research for diagnosed mental illness on HIV outcomes has been published. The current lack of mental health treatment for HIV+ populations in low and middle income countries (LMICs) represents a critical research gap (Collins et al. 2006). Given that women in sub-Saharan African now constitute the largest proportion of HIV+ individuals in the world and have high GBV prevalence, mental health research with HIV+ women affected by GBV (HIV+GBV+) in sub-Saharan Africa is urgently needed.

The objective of this study was to conduct a mental health care needs assessment of HIV+GBV+ women served by the Kenya Medical Research Institute (KEMRI)-University of California at San Francisco (UCSF) Family AIDS Care Education & Services (FACES) collaborative in Kisumu County, Kenya. Established in 2004, FACES is a President's Emergency Plan for AIDS Relief (PEPFAR)-funded care and research collaboration serving more than 140,000 HIV+ individuals in western Kenya. Kisumu County has the highest prevalence of HIV (19.3%) and physical violence against women (57% of women aged 15-49) in Kenya (Kenya National AIDS and STI Control Program 2007; Kenya Demographic and Health Survey 2008-09” 2010). The information from this study will be used to adapt a scalable, capacity building mental health treatment for HIV+GBV+ women at FACES for our upcoming Randomized Controlled Trial (RCT) at the same site.

Methods

Between April 2013 and June 2013, we conducted in-depth interviews and focus group discussions with 61 study participants. Inclusion criteria included age of 18 or older, ability to give verbal informed consent, attend the duration of the interview, and absence of severe cognitive dysfunction, such as advanced dementia, severe intellectual disability, current intoxication and psychosis. An onsite referral system was in place for prospective participants found to be in health crisis or in need of legal aid related to GBV.

We conducted 30 in-depth interviews and four focus groups. Interviews and focus groups were completed by the study research team, all of whom had been involved with prior qualitative research at the clinic and were fluent in the local languages and dialects of Dholuo, Kiswahili and English. Supplemental training in the conduct of qualitative interviews and ethical research principles was provided by authors SM and SD through a week long course prior to the start of the study. There were a total of four interviewers, including two men and two women. In-depth interviews were conducted with one participant and one interviewer. Focus groups were conducted with two interviewers in order to keep track of responses and follow up probes. None of the focus group participants were also key informant interviewees. A separate team of four research assistants transcribed and translated the audio-recordings into interview transcripts.

Interviewers followed a structured interview guide consisting of open-ended questions and standardized probes. Interview domains included: violence against HIV+ women, mental and physical health problems among HIV+ GBV+ women, current coping methods, and barriers and preferences for mental health care. Respondents were asked to report on the experience of HIV+GBV+ women in their community, as opposed to their own personal experiences. While aware that participants’ responses would be informed by their own experiences, this approach was used in order to avoid potential emotional distress caused by request for information regarding the respondent's personal traumatic experiences and also to gather information reflective of shared experiences among HIV+GBV+ women.

Interviews were audio-recorded with no patient identifiers. Transcribers and translators all had prior experience with translation from Dholuo and Kiswahili into English. Translators/transcribers were given accuracy tests during the training phase in which they translated and transcribed audio recordings from Dholuo/Kiswahili into English. Ethical approval was obtained from the UCSF Committee on Human Research and the KEMRI Ethical Review Committee.

Data analysis

Grounded theory guided the process of data analysis (Strauss 1990). Grounded theory sorts data in three ways: through concepts, categories and propositions. Concepts are identified by seeking conceptual similarities among focus groups/interviews. These concepts are grouped to develop categories of concepts. Categories are then compared and grouped to develop propositions or theoretical constructs about the phenomena of interest.

Central to qualitative data analysis is an iterative, inductive process (Starks and Trinidad 2007) in which interview transcripts are reviewed line by line and as a concept becomes apparent, a primary code is assigned (Bradley et al. 2007). As more data are reviewed, the specifications of codes (secondary codes) are developed. To ascertain whether a code is appropriately assigned, the analyst compares text segments to segments that have been previously assigned the same code and decides whether they reflect the same concept. Using this “constant comparison” method (Glaser 1965) the researchers refine dimensions of existing codes and identify new codes.

Dedoose software version 4.5 was used to code excerpts of interviews and group concepts according to categories. The primary and secondary codes were established through a process of open coding on specified domains (see above) of all (100%) interviews (SMM) and by comparing results with a separate open coding of a random selection of 15% of interviews (SLD). Discrepancies were discussed and resolved with slight modification of primary and secondary codes. The resulting codebook was applied by another author (BZ) to 50% of the interviews. Application of codes was discussed with minor modifications to secondary domains. The remaining 50% interviews were coded independently by BZ. The overall concordance rate for the application of primary and secondary codes was calculated by to be 98.7% across the 34 interviews based on random reliability checks of 50% of interviews. Specifically, the number of codes applied correctly was divided by the number of codes applied correctly + the number of omissions + the number of commissions (incorrect applications). 2750 codes were applied correctly in the 50% random check, with 35 omissions and no commissions = 2750/(2750+35) = .9874.

Results

With some individuals belonging to more than one group, twenty-five of the in-depth participants were female FACES patients (83%); five were community health workers (17%), three were healthcare providers (10%) and six were members of the clinic's community advisory board (20%). The four focus groups ranged in size from 4 to 10 participants and consisted of: (1) community health care workers; (2) FACES patients (all female, Dholuo speaking); (3) FACES patients (all female, Kiswahili speaking); (4) community advisory board members (Table 1).

Table 1.

Sample Characteristics

In-Depth Interviews	Category	No. (%)
Gender	Female	25 (83)
	Male	5 (16)
Stakeholder Group (some belong to more than one group)	Clinic patient	25 (83)
	Health care provider	3 (10)
	Community health worker	5 (16)
	Community advisory board	6 (20)
Stakeholder Focus Groups
Health care workers (any type, 9 participants) community healthcare worker, data clerk, pharmacist, clinicians(2), nurse, record keeper, lab tech, nutritionist	Female	6 (67)
	Male	3 (33)
Clinic Patients, Dholuo speaking (10 participants)	Female	10 (100)
	Male	0
Clinic Patients, Kiswhaili speaking (8 participants)	Female	8 (100)
	Male	0
Community Advisory Board (4 participants) area chief, church elder, business people (2)	Female	1 (25)
	Male	3 (75)

Seventeen of the 29 key informant interviews with women were conducted by women and 12 were conducted by men. We analyzed the number and type of codes for interviews that were matched by gender versus those that were not. Both the average number and range of codes was similar between interviews that were gender matched and those that were not.

We also examined differences in the themes generated by key informant and focus group interviews based on participant type. The two largest groups of key informant interviews were healthcare workers and patients. Interviews with both groups produced codes in every domain with no systematic differences in code-type. Focus group interviews with patients, health care workers and community advisory board members also covered all domains. The community advisory board had relatively fewer numbers of codes than the other focus groups (196), compared with health care workers (498) and patient focus groups (525 and 324).

The majority of the results presented here are concepts that were present in 50% or more of the interviews. Exceptions were made for commonly reported secondary codes on the identified domains, the least frequent of which was reported in 20% of interviews. An overview of key themes is presented in Table 2 and discussed by domain, below. Participants’ names were removed and pseudonyms were used for the quotes.

Table 2.

Overview of Key Themes and Codes

Violence against HIV+ women*
• Percent and number of interviews citing violence against HIV+ women by type:
physical: 100% (n=34)	sexual: 76% (n=26)
emotional: 82% (n=28)	financial/food abuse: 62% (n=21)
• Begins after disclosure of HIV status to husband: 79% (n=27)
• Perpetrated by primarily by husband: 97% (n=33)
• Based on assumption of wives’ infidelity, despite conflicting community knowledge
Mental health problems among HIV+GBV+ women*
• Sadness: 56% (n=19)	• Sleep disturbances: 29% (n=10)
• Suicidality: 50% (n=17)	• Anxiety: 47% (n=16)
• Worthlessness/guilt: 50% (n=17)	• Alienation from others: 75% (n=25)
• Memory/Concentration difficulties: 26% (n=9)	• Irritability: 35% (n=12)
Effects of gender based violence on HIV health among HIV+GBV+ women*
• ART default–decreased appetite interferes with adherence to ARTs taken with food: 59% (n=20)
• ART default–hopelessness and passive suicidality among HIV+GBV+ women: 53% (n=18)
• ART treatment is ineffective for raising CD4 in setting of GBV-associated stress: 56% (n=19)
Methods of coping with emotional problems*
• Social support from other women, including support groups: 82% (n=28)
• Economic self-sufficiency: 71% (n=24)
• Religion: 62% (n=21)
Preferences and barriers for mental health care*
• Strong preference for receiving mental health care: 100% (n=34)
• Integration of mental health care within HIV clinic: 32% (n=11)
• Individual, weekly counseling preferred by those who commented

^*Percent and number of interviews that cite the issue as a problem/method/preference for HIV+GBV+ women served by the clinic.

Violence against HIV+ Women

Physical Violence

All participants perceived that violence against HIV+ women was highly prevalent. The majority of participants (79%, n=27) reported that violence against HIV+ women began after women learned of and disclosed their HIV+ status to their male spouse. Women often learned of their status through the national mandatory prenatal HIV testing program. Ninety-seven percent (33 interviews) reported that the husbands of HIV+ women inflicted the violence. While accusation of marital infidelity was typically cited as the basis for victimization of women, many husbands initiated violence without knowledge of their own HIV status. Fifty-three percent (18 interviews) reported violence from community members toward HIV+ women and 50% (17 interviews) stated that husband's family members inflicted violence. For example, participants reported that:

“when a woman is pregnant she must be tested for HIV and when she is found to be positive she is told to bring the husband to be tested and that is where the problem will start because after the woman has tested positive the husband will not take it lightly. Violence will start right away because he will be pointing fingers at you” (Linnett, female community health worker, in-depth interview)

“... you might have a husband who has not gone for a HIV test. They can never go. It is almost impossible to convince them to go. You come back home to tell your husband that you are HIV positive. This man will hurt you. You will be beaten. You might even be sent away from your home. You will leave the kids behind or be forced to go with the kids. Your husband might also die and your in-laws will want to inherit you [forced marriage to a brother of the late husband]. This happens a lot in Dholuo land. You will have to tell them that you cannot be inherited because of one, two or three. They will start saying, ‘You have killed our son.’ They will beat you up and send you away from that home. They believe that it is you who brought them the disease” (Jane, female community health worker, in-depth interview).

Husbands of HIV+ women were also reported to discard their wives’ anti-retroviral medications or otherwise interfere with their ART adherence, for example:

“For those who live together, it may be that I stay with my husband yet he does not allow me to go and take drugs. It will force me to hide these drugs where he does not know yet this is a hard task. Personally I have this problem, there was a time when he took them and threw them into a pit latrine” (Amina, participant in focus group of female clinic patients).

Emotional Violence

Emotional violence against HIV+ women was cited as a problem in the majority of interviews (82%, n=28). Participants reported that HIV+ women experienced emotional abuse from male partners, in-laws and community members. Common themes were disparaging remarks regarding women's infection as well as explicit accusations that HIV+ women were sexually immoral and had contracted HIV by having sex outside of the marriage. These beliefs persisted despite simultaneous acknowledgement that local norms favor male promiscuity over female and that (untested) husbands are often responsible for their wives’ HIV infection. For example, study participants discussed that:

“If I go out there and tell people that I am HIV positive, at the back of their mind, they will think this lady was a prostitute at one particular time. They will not think of the other ways or modes of transmission. Some people will feel sorry for you while others will not. The in-laws will say that it is you who brought it into the family” (Esther, female social worker, in-depth interview).

“Emotional problem[s] can come as a result of disclosure and it might cause her to be emotional and also the blame game may be after realizing that she is positive when she goes home she discloses to the husband and then the husband will say that she is the one who has brought the HIV infection something like that. So it is being blamed on her and maybe she is the one who has not brought it and she will then be affected emotionally” (Akili, participant in focus group of healthcare providers).

Sexual violence

Over three-quarters of the interviews referenced sexual violence against HIV+ women (76%, n= 26). Husbands were described as resistant to using condoms because they wanted to have many children. Wives were described as wishing to use condoms in order to adhere to HIV care providers’ instructions to prevent transmission/co-infection with multiple strains of HIV. Participants reported that the discord led to sexual violence against HIV+ women, for example:

“I have another point like they really suffer sexual violence in the point that may be here at the clinic we encourage and give condoms, that when they go back home they should use them so that it prevent any infection in case it occurs. Then and when they reach home and may be all of them are positive and there he say that we are all positive and why should we use the condoms or something like this. So the man will end up having sex without using the protection and may as the woman will not be willing to give in it will be like a rape case in such a situation” (Akili, participant in focus group of healthcare providers)

Another major theme related to lack of sexual desire among HIV+ women secondary to HIV-associated fatigue—husbands were described as forcing sex even when women felt too ill. For example, participants described the following:

“The lady was too sick to make love yet the man insisted on making love. The lady could complain that the guy hurt her because she was weak. ‘You are just pretending. Open up your thighs.’ Isn't that being raped? You know a sick person cannot make love” (Sarah, female healthcare provider in-depth interview).

Financial and Food Abuse, Emotional Neglect

Financial and food abuse, involving withholding of money and food, as well as emotional neglect, were reported in over half of the interviews (62%, n=21). Participants reported themes such as:

“Before you told him that you were HIV positive, maybe he [your husband] used to give you 500 shillings per day. Now that you are HIV positive, he starts frustrating you. He will give you 200 shillings and you are supposed to use that for the whole day” (Esther, female social worker, in-depth interview)

“He sees you as just dead, you are just like an image . . . He has nothing to lose, he sees you as a dead person. You are just there, there is no serious love.” (Mildred, female clinic patient, church elder and community board advisor, in-depth interview).

Mental Health Problems for HIV+GBV+ Women

All interviewees and focus groups noted that HIV+GBV+ women experienced emotional problems.

Stress

The most commonly cited emotional problem for HIV+GBV+ women was described as stress. Eighty-two percent (28 interviews) used the term “stress” to describe the suffering of HIV+GBV+ women. The primary sources of stress were violent victimization from husbands and social stigma associated with HIV infection. Many reported themes similar to the following:

“The problems that women in Nyanza [Kisumu County] face [when] you are HIV+, whenever you pass people tend to talk about you and this will cause you stigma and stress . . . Yes, I had a lot of stress and this made me to be very thin since there was no appetite. I was worried and felt defensive of his [abusive husband's] presence and wondered what to do in case he was around” (Anna, female clinic patient, in-depth interview).

While stress was used as a general term with varied definitions, the emotional sequelae of stress among HIV+GBV+ women were more specific and included depression and anxiety symptoms, as well as their impact on HIV health—described below.

Sadness

Sadness was cited in more than half of interviews (56%, n=19) as an emotional problem for HIV+GBV+ women, including its association with impairment of daily function.

“They [HIV+GBV+ women] cannot associate with others...they are always sad...she [an HIV+GBV+ woman] cannot do any job. These are the suffering[s]” (Mary, female clinic patient, in-depth interview).

Suicidality

Thoughts of suicide were described as a problem for HIV+GBV+ women in half of interviews (50%, n=17) interviews, for example:

“The state of being so hopeless comes in and this comes because all their [HIV+GBV+ women's] things are taken away and even some rights taken away as well. She [an HIV+GBV+ woman] will think of committing suicide at some point because of the emotional problems” (Kamili, participant in focus group of clinic community advisory board).

Worthlessness and Guilt

Feeling worthless and guilty was cited as a common emotional problem (50%, n=17).

“They [HIV+GBV+ women] feel as if they just don't add value. Others feel like even dying because they are so valueless in the eyes of everyone” (Martha, female clinic patient and nurse, in-depth interview).

Memory and Concentration Difficulties

Problems with memory and concentration were cited in approximately one-quarter of interviews (26%, n=9).

“There is this problem of memory lapse. You will only realize that you needed to do something when it is already late. There is a lot of memory lapse once there is violence.” (Faiza, participant in focus group of female clinic patients).

Sleep Disturbances

Both increased and decreased sleep were reported as symptoms of emotional distress among HIV+GBV+ women (29%, n=10).

“Now you are not at peace you are emotionally stressed, you develop poor sleeping habits and at times in the night you ask yourself what you have done wrong” (Lisa, female healthcare provider, in-depth interview).

Anxiety

Forty-seven percent of interviewees (n=16) reported anxiety/worry as a problem for HIV+GBV+ women. The major themes revolved around concern about finances, food security and providing for children.

“They like sitting alone figuring out how their life will be better and fulfil where she would to reach. Sometimes you will find the person worrying too much even if you talk to her it's like the mind is very far, and thinking of what she will eat for supper or the ugali flour is over or the child slept hungry and they still going to lack food” (Christa, female clinic patient, in-depth interview).

Feeling alienated from others

A commonly cited emotional problem for HIV+ women affected by GBV was feeling alienated from others (75%, n=25). Participants described this feeling in response to being HIV+, suffering GBV and the two combined.

“The reason that may result to a hurtful feeling is when a man alienates you and does not want the [your] child” (Elizabeth, female clinic patient, in-depth interview).

“Women who are productive before their HIV status is known, tend to shy away when their status is known and there is that violence in the family. They tend to shy away from the community activities that they were participating before” (Radhi, participant in focus group of healthcare providers).

Irritability and aggression

Less frequently reported symptoms of emotional distress among HIV+GBV+ women included irritability/aggression (35% of interviews, n=12).

“They feel neglected, they blame everyone and they are angry with everyone. This affects even their future relationship[s]” (Mosi, participant in focus group of clinic's community advisory board).

Emotional distress and HIV health

Almost all participants described emotional distress among HIV+GBV+ women as having a negative impact on their HIV health (91%, n=31).

Adherence to ARTs

Decreased appetite and weight loss

More than half of interviews (59%, n=20) reported that HIV+GBV+ women experienced decreased appetite and weight loss. This was described as both a symptom of emotional distress and a danger to HIV-related health because low appetite interfered with adherence to ART medication. Some ART medications must be taken with food – if women's appetites were low, they stopped eating and did not take medication.

Hopelessness

Intentionally giving up on ARTs in the context of emotional distress was cited as a problem for HIV+GBV+ women in more than half of interviews (53%, n=18).

“Some people might get angry and lose hope in life. They will even prefer being dead than alive. They can stop medication and wait for their death” (Annette, female clinic patient, civil servant and community advisory board member, in-depth interview).

Effects of Emotional Distress on HIV Health

More than half of respondents stated that the emotional distress experienced by HIV+GBV+ women caused HIV health to deteriorate even if ARTs were taken as prescribed (56%, n=19). Several respondents specified that, in the setting of emotional distress, CD4 counts decreased even if fully adherent to ARTs.

“It's a must for me to take ART's whether there is violence. The disadvantage part of it is that you will take ART's without peace in your mind; CD4 will go down even if you are taking the medicine because there is nothing ... supporting your body with when there is no peace” (Mildred, female clinic patient, church elder and community board advisor, in-depth interview).

Current Emotional Coping Strategies among HIV+GBV+ Women

Eighty-two percent (n=28) of interviews identified social support from other HIV+ women as a common coping mechanism for emotional distress among HIV+GBV+ women.

“Most of them [HIV+GBV+ women] have made it a routine on dealing with problems by talking to each other and sharing deep about their emotions” (Martha, female clinic patient and nurse, in-depth interview).

Seventy-one percent (n=24) of interviews noted that economic self-sufficiency was an important coping mechanism for emotional distress. HIV+ women's support groups were noted to be helpful both emotionally and for economic assistance.

“I usually encourage women to join these support groups. From there, you will be helped emotionally and at times you might even be helped financially. Whatever little they have in the bank, might be given to you to start a business” (Esther, female social worker in-depth interview).

Sixty-two percent of interviewees (n=21) stated that religion was used as a coping skill.

“You will feel hurt since whenever a man comes, he will be quarreling you and end up abusing you physically, at times he does not beat you but quarrels you a lot, at times you assume that you have done something correctly yet to him you are wrong. As time goes by you will feel like losing hope in life and wish to die. This is the time when one turns to God and pray always. Whenever there is a problem I kneel down and ask for God's help” (Elizabeth, female clinic patient, in-depth interview).

Preferences and Barriers for Mental Health Care

All participants, without exception, reported that mental health care is needed for HIV+ women who suffer GBV. While most of the barriers cited were related to the stigma of receiving HIV care, as opposed to mental health care for HIV+ individuals (41%, n=14), respondents did observe that mental health care could confer additional stigma for recipients. For this reason, participants stated that the preferred location for mental health services was in the same building with the HIV care services, because it would better protect the privacy of those seeking mental health care services and decrease risk of additional stigma (32%, n= 11):

“I think that if they are integrated under one roof it will [be] better ... [for] the clients, because if they are separated once you go inside they might think that every time you go inside there everybody might think that you have a psychological problem. So if it's incorporated with all other medical care, then it will make you to access all the services you need and anytime you go you will say that you need the services” (Etana, participant in focus group of healthcare providers).

Among those who commented specifically on group versus individual format for counseling (n= 21), there was a strong preference for one-on-one counseling. Reasons included concern about gossip and differences in the emotional problems experienced by each woman.

“It should be done to individual. The reason for this is that each and every person has a different problem. We are now seated discussing. My physical and mental problem is different from another client” (Mary, female clinic patient, in-depth interview).

Most respondents agreed that counseling should occur approximately weekly. The rationale for weekly frequency included the observation that stressors and associated emotional distress accumulated daily for HIV+GBV+ women and longer intervals would not provide sufficient care.

Discussion

In this study of mental health care needs among HIV+GBV+ women at a large HIV care center in Kisumu, Kenya, all participants cited the negative impact of violence against HIV+ women. Respondents reported that violence was perpetrated primarily by the husbands of HIV+ women and it was severe, ranging from attempted murder, rape and physical assault requiring hospitalization, persistent emotional abuse accusing HIV+ women of being worthless, sexually immoral and non-human, as well as taking their children away from them, obstructing their HIV care and withholding of food, money and shelter. Emotional distress among HIV+GBV+ women included symptoms that are associated with depression in U.S. diagnostic systems (American Psychiatric Association 2013), such as sadness, suicidality, worthlessness/guilt, concentration/memory difficulties, sleep disturbances, low appetite/weight loss and hopelessness. General anxiety and traumatic stress symptoms such as alienation from others and irritability were also cited. Current coping methods included social support from other women.

In addition to the direct interference with HIV care experienced by HIV+GBV+ women, they described detrimental effects of GBV-associated emotional distress on their HIV health. Intentional default on ART medications was reported to be motivated by hopelessness, suicidality and anger. HIV+GBV+ women also repeated that HIV health (specifically CD4 cell count) could not be maintained in the setting of emotional distress even if ART medications were taken as directed. While little research has examined the effect of mental disorders on HIV health independent of ART adherence, this qualitative work suggests that the potential pathway may be worth exploring. If it does exist, it would be consistent with findings on the relationship between depression, PTSD and physical health (Dedert et al. 2010; Köhler et al. 2013).

All participants agreed that mental health care was needed for HIV+GBV+ women. There was also a high level of consensus regarding the type of treatment that would be most beneficial—individual psychotherapy in a private space at the FACES-supported clinic in weekly or twice weekly sessions. The barriers to mental health treatment were primarily related to lack of services and/or knowledge of service availability. The majority opinion was that any additional stigma associated with mental disorders would be obviated by embedding mental health services within the HIV care clinic.

Certainly, prevention efforts are critical responses to the health problems created by GBV against HIV+ women and such work is progressing in this region of Kenya, despite deeply entrenched social norms that support GBV (Hatcher et al. 2013; Turan et al. 2013). In addition, the FACES clinic is expanding its couples testing efforts in order to avoid situations in which women are the first in the family to be diagnosed with HIV. However, as reported here, the mental health of the many HIV+ women who have been and continue to be affected by GBV is a pressing issue that must be addressed in tandem with prevention efforts in order to remediate the devastating disability of treatable mental disorders and the HIV related deaths they can foster.

Limitations

One limitation of this study is that it focuses on HIV+GBV+ women who are engaged in HIV care in a single clinic, as opposed to those who never attempted or defaulted on HIV care. Given the disability associated with mental disorders and its interference with engagement in medical care, it is possible that the severity of the emotional disorders observed in this study of women engaged in care is the “tip of the iceberg” – the mental disorders among HIV+GBV+ women who are not engaged in HIV care may be more severe and perhaps more diverse than what was observed in this study. Another limitation is that individuals were asked to comment not about themselves but about local HIV+GBV+ women in general (see methods for rationale). Some participants spontaneously elected to provide personal information. Thus, our data analysis reflects a mix of data focused on perceptions of the self and others and thus the “second-hand” information may be less reliable. However, in light of the dearth of studies on the mental health care needs of HIV+GBV+ women, this exploratory qualitative study is an important first step towards the development of culturally-sensitive interventions in a region of the world experiencing high rates of both HIV and violence against women.

Conclusions

Participants interviewed from the FACES-supported clinic in Kisumu described severe, widespread violence against HIV+ women with symptoms of depression, anxiety and traumatic stress that interfered with HIV health. Women reported that mental health care was needed for HIV+GBV+ women and there was nearly full consensus on the type of treatment desired: individual counseling integrated with HIV care in approximately weekly sessions. This type of treatment has been effectively used in other low resource settings where evidence-based treatments such as Interpersonal Psychotherapy (IPT) or Cognitive Behavioral Therapy (CBT) have been adapted for local application and delivery in a sustainable manner (Bass et al. 2013; Bolton et al. 2003; Jiang et al. 2014; Meffert et al. 2009). The data from this study will be used to inform the development and testing of a sustainable, scalable, locally delivered mental health treatment, extending prior global mental health interventions by integrating treatment within an HIV clinic setting and measuring the impact of treatment on both HIV and mental health.

Highlights.

• We evaluate mental health care needs among HIV+ women affected by GBV in Kenya

• Thirty in-depth interviews and four focus groups were conducted at a large HIV clinic

• Participants stated that violence against HIV+ women is extremely common

• Participants stated that GBV against HIV+ women leads to poor mental and HIV health

• Mental health care is desired—individual, weekly counseling at HIV clinic preferred