Factors associated with mental health symptoms in women living with HIV in Southern India : An exploratory study

Limited research has focused on the mental health of HIV-infected women in resource poor settings such as rural India. This study attempts to fill this gap in the literature, through conducting standardized interviews with 20 HIV-infected women in rural, Southern India. Variables of interest included trauma exposure, mental health symptoms, shame, guilt, social support, negative social reactions, coping, and HIV knowledge. Results indicate most women experienced HIV-related stigma in the form of negative social reactions, and limited social support. Many reported a history of interpersonal violence, and moderate to severe symptoms of post-traumatic stress and depression. Feelings of shame, guilt, and self-blame were significantly correlated with mental health symptoms. Seeking comfort in religion, such as engaging in meditation or prayer, was a preferred strategy for coping with HIV-infection. Despite the small sample size, this exploratory study provides important information about the challenges facing women living with HIV in settings where HIV is highly stigmatized. Future research should examine predictors of mental health outcomes in a larger sample, and evaluate the efficacy of interventions designed to reduce negative social reactions, shame, and self-blame, and increase social support.


INTRODUCTION
More than two and half million people in India are estimated to be living with HIV and AIDS; of these, 39% are women (World Bank, 2012).The southern states have some of the highest rates of infection, including the Indian state of Tamil Nadu (World Bank, 2012).Factors such as a large trucking industry, cultural prohibitions against talking about sexual practices even among married couples (Pallikadavath et al., 2005), and the disenfranchised status of women (Claeson and Alexander, 2008) have been identified as contributing to the spread of HIV among women in India."Inaddition to the physical health challenges HIV-infected women may face, HIV status has been associated with poor mental health outcomes, especially in resource poor communities (Das and Leibowitz, 2011;Wu et al., 2007).Various factors such as HIV-related stigma, including negative social reactions, lack of social support, risk of interpersonal violence, shame and guilt, may increase the risk of mental health difficulties, especially for HIVinfected women.
HIV-related stigma and negative social reactions are common among women living with HIV (Brown et al., 2003;Hollen, 2010;Vlassoff et al., 2012).HIV-related stigma is defined as the perception and experience of discrimination from others based on HIV-related attitudes and beliefs (Mahajan et al., 2008).HIV-related stigma has been associated with mental health symptoms (Brown et al., 2003), and diminished quality of life (Steward et al., 2009;Thomas et al., 2007;Nyamathi et al., 2013).Stigma has also been associated with decreased likelihood of attending medical appointments, increased alcohol consumption, high rates of social isolation (Nyamathi et al., 2013;Moskowitz et al., 2009), higher CD4 counts, and a lack of medication adherence (Rintamaki et al., 2006).Specifically for HIV-infected women of Southern India, stigma and negative self-image have been linked with risk for depression (Charles et al., 2011).Such women are also likely to be victims of overt discrimination that may include social banishment (Van Hollen, 2010;Vlassoff et al., 2012).
In addition to stigma, is not uncommon for HIV-infected persons to report low levels of social support, and resulting social isolation (Sivaram et al., 2009).Inadequate social support for HIV-infected populations has been associated with poor mental and physical health outcomes (Moskowitz et al., 2009), and lack of medication adherence (Gonzalez et al., 2004;Simoni et al., 2006;Weaver et al., 2005;Vyavaharkar et al., 2007).Adequate social support has been associated with a reduction in negative affect (Simoni et al., 2006), decreased depression (Gonzalez et al., 2004), reductions in stress, enhanced coping (Weaver et al., 2005), and other positive mental health outcomes for persons living with HIV and AIDS (Gielen et al., 2001;Serovich et al., 2001).One study of HIV-infected women in India indicated that women experienced a decline in social support after contracting the virus, and experienced resulting mental health symptoms, including a lack of hope for the future (Majumdar, 2004).
Women living with HIV are also at increased risk of interpersonal violence, including intimate partner abuse (IPA) (Campbell et al., 2008;Fonk et al., 2005;Gielen et al., 2001).Across various studies, including India, women with a history of intimate partner abuse appear to have higher rates of HIV (Stockman et al., 2013;UNAIDS, 2013).Although actual rates are likely influenced by underreporting, research indicates one-third of married woman in India may experience IPA in their lifetime (Silverman et al., 2008).IPA appears to be a risk factor for contracting HIV among Indian women (Go et al., 2003).Intimate partner abuse has also been linked to low quality of life among women living with HIV (McDonnell et al., 2005), and has been associated with negative mental health outcomes, including post-traumatic stress, depressive symptoms (Campbell et al., 2008), and substance abuse (McDonnell et al., 2005).Domestic violence may also be a barrier to receiving HIV services (Decker et al., 2009).
Considering this, it is perhaps not surprising that women living with HIV are at increased risk for poor mental health outcomes.Depression is one of the most common psychological disorders experienced by those living with HIV (Chandra et al., 2005).In one study of HIV-infected women in urban Tamil Nadu, India, 40% reported depressive symptoms (Charles et al., 2011).Much like IPA, depression is considered to be both a risk factor for contracting HIV, and a common outcome of living with a life-threatening illness (Collins et al., 2006;Prince et al., 2007).Persons living with HIV are at increased risk of -suicide (Cooperman and Simoni, 2005), anxiety disorders and post-traumatic stress (PTSD) (Breuer et al., 2011;Jin et al., 2006;Nyamathi et al., 2013).Mental health symptoms can have a direct impact on disease state; those with clinical levels of depression and PTSD are more likely to have a lower CD4 count and higher viral loads (Boarts et al., 2006;Breuer et al., 2011).Due to the relationship between mental health symptoms and HIV status, there is a need to further understand factors which may be associated with poor mental health among this population.
Given the myriad challenges HIV-infected women may face, including the possibility of societal rejection, it may not be surprising that women living with HIV experience feelings of shame and guilt.Shame has been linked to physiological responses that negatively impact HIV disease outcomes (Dickerson et al., 2004).Shame has also been associated with a decreased likelihood of being tested for HIV (Kalichman and Simbayi, 2003), and an increase in risky sexual behaviors (Sikkema et al., 2009).Specifically, in Southern India, shame and guilt were found to be associated with HIV-infection and subsequent reductions in social and familial support (Tarakeshwar et al., 2007).
In light of these concerns, it is important to consider coping strategies that may be effective in mitigating mental health consequences associated with HIV.Some research indicated that HIV-infected women in rural India may use religion and spirituality as preferred coping strategies (Majumdar, 2004;Nyamathi et al., 2013), and may focus an increasing amount of energy on securing the well-being of their children (Majumdar, 2004).Although such approaches have typically been considered healthy or adaptive forms of coping, navigating daily life with HIV and AIDS can also result in harmful or seemingly maladaptive forms of coping.For example, some HIV-infected individuals may use avoidant strategies such as hiding medications, avoiding healthcare appointments, and not disclosing one's HIV status (Nyamathi et al., 2013).Avoidant coping strategies have been associated with negative health outcomes, reduced quality of life, and psychological distress (Nyamathi et al., 2013).

Current study
There is limited research examining various factors associated with mental health outcomes among HIVinfected women in countries such as India.This study is designed to explore challenges facing HIV-infected women in rural Southern India in hopes of informing future research and clinical interventions.Specifically, we investigated -HIV-related stigma in the form of negative social reactions and social support; interpersonal violence; mental health symptoms including depression and PTSD; shame, guilt, and HIV knowledge.

Participants
Study participants were selected from HIV-infected women receiving nutrition and health services at a local non-governmental organization (NGO).Data collection took place over several weeks during the summer of 2013.An institutional review board (IRB) at a university in the U.S. and an IRB-equivalent in New Delhi approved study procedures prior to data collection.

Procedure
Participant recruitment began with an information session about the research for all clients receiving services from the local NGO.The information session allowed potential participants an opportunity to meet the research team and ask questions about the project.After the information session, the research team reached out by phone to determine interest.Participants were contacted at random.A final sample of 20 participants was recruited in this manner for this exploratory research.
Consistent with local consent procedures, the consent form was verbally reviewed with each participant prior to engaging in a structured interview.The voluntary nature of the research was emphasized in the consent form, and an 'opt-in' procedure was used to determine participation, making it relatively easy for women to choose to not participate.Finally, consent was obtained by signature or thumbprint, as some women were not literate.
Structured interviews were used to collect information about mental health symptoms, HIV-related stigma in the form of negative social reactions, social support, interpersonal violence and exposure to other adverse events, shame, guilt, coping skills, HIV transmission knowledge, and demographics (e.g.religion, age, marital status, number of children).The structured interview was based on a standardized questionnaire detailed in the following section.
The questionnaire was translated into Tamil by a local translator and back translated into English by U.S. and Indian researchers.
Each item was reviewed to ensure clarity and to maintain construct equivalence between Tamil and English versions.
Data were collected at participants' residences in order to reduce the burden on participants associated with asking them to travel to the NGO office.The research team traveled to participant homes at a pre-arranged day/time.Each interview took approximately one hour.Upon completion participants received a bag of rice or packet of oil as compensation for their time.

Measures
Participant demographics: Demographics including age, marital status, number of children, resources compared to others in the community, education, employment status, and religion were measured.
HIV-related stigma, negative social reactions: Social reactions were measured using three adapted items based on the Social Reactions Questionnaire (SRQ; Ullman, 2000).Participants were asked -Did anyone do any of the following when they found out about your health situation: indicate you are to blame for the situation; stop spending time with you; treat you differently?Cronbach's alpha for these items was 0.97.
Social Support: Social support was measured using three items from the Interpersonal Support Evaluation List (Cohen and Hoberman, 1983).Although this scale has not been validated for use with this specific population, it has been administered in various settings including Taiwan (Chen et al., 2000) and Venezuela (Bastardo and Kimberlin, 2000).It has also been administered in populations with HIV (Bastardo and Kimberlin, 2000;Yi et al., 2006).The following questions were included with a 4-item response scale from definitely true to definitely false -there are several people I trust to help solve my problems; when I feel lonely there are several people I can talk to; if I were sick I could easily find someone to help me.Because the Cronbach's alpha was low for the combined items (0.25), items were examined individually.
Trauma exposure, including interpersonal violence: Trauma exposure was measured using an adapted 5-item version of the Life Events Checklist (Blake et al., 1995).We did not ask about "life threatening illness," because HIV-infection is a criterion for receipt of services at the local NGO where participants were recruited.We asked about childhood and adulthood exposure to the following events: natural disasters, motor vehicle accidents, physical assault by a family member (e.g. partner, husband, mother-in-law), physical assault by someone outside of the family, and sexual assault.

Depression:
The Hopkins Symptom Checklist (HSCL-25) is a selfreport scale intended to measure symptoms of anxiety and depression.The HSCL has been validated for use with a variety of cultural groups (Klein et al., 2001).For this study, the 15-item depression subscale was used, with response options from 1 = not at all to 5 = extremely.One researcher-created somatic symptom item was added to the scale based on suggestions from local NGO staff: Headaches or other pain in my body when thinking about the problems in my life.An additional item about sexual interest or pleasure was dropped based on the difficulties local staff had administering the item.Cronbach's alpha indicated good internal consistency for all items (0.89).

PTSD:
The Harvard Trauma Questionnaire (HTQ) is a self-report scale developed for research and assessment of trauma symptoms in Indochinese refugees (Mollica et al., 1992).Various adaptations of the HTQ have been validated for use with Cambodian, Laotian, Vietnamese, Arabic, Farsi, Serbo-Croation, and Russian-speaking populations (Klein et al., 2001).The HTQ has been used in Tamil Nadu for research and screening of PTSD after the 2004 tsunami (Kumar et al., 2007).For this study, 13 HTQ items were selected for use based on recommendations for cultural appropriateness.
Participants were asked about problems people may experience after a stressful event and how much these problems affected [them] in the past week.For all 13 symptom statements, response options ranged from 1 = not at all to 4 = extremely.Cronbach's alpha for the 13 items was good (0.83).

Shame and Guilt:
The 15-item State Shame and Guilt scale (SSGS) was originally developed to capture state shame and guilt in psychology experiments (Marschall et al., 1994).Although this measure has not been previously validated in this cultural context, it has been used in a variety of naturalistic settings (Gruenewald et al., 2004;Tangney and Dearing, 2006).There are five items for each of three subscales.An example of a shame item is I want to sink into the floor and disappear.An example of a guilt item is I felt bad about something I did.Participants are asked to rate how they are feeling currently on a 5-point scale from 1 = not feeling this way at all, to 5 = feeling this way very strongly.The Cronbach's alpha for the shame subscale was 0.64.For the guilt subscale the Cronbach's alpha was low (0.36); however, we included these scale items given the exploratory nature of this study.
Coping: The Brief COPE is a 28-item scale that measures utilization of 14 types of coping derived from both theory and research (Carver et al., 1989;Carver, 1997).The instrument was piloted in a racially diverse sample in the USA and has since been translated into numerous languages (Carver, 1997;Kapsou et al., 2010).Previous studies using the Brief COPE have focused on immigrants, caregivers, students, and refugees (Chase et al., 2013;Sarfo-Mensah, 2009).The dispositional version of the Brief COPE was administered (Carver, 1997).Participants were asked: In order to cope with stressful life events, such as your current health status, how often do you do the following?Cronbach's α for the full coping scale was poor (0.59), suggesting the scale did not measure a unified construct, so items were examined individually.
HIV knowledge: An investigator created 11-item inventory was used to capture knowledge about HIV transmission: Yes or no, "Is HIV transmitted by…?"sex with an infected person; contaminated needles; blood transfusion; God's will; hugging; shaking hands; sharing the same bathroom; insects; donating blood; sharing eating utensils.This measure was used based on feedback from local staff and considering what was relevant for the local cultural context.

RESULTS
All data were analyzed using Statistical Package for the Social Sciences (IBM, SPSS Statistics, 20.0 and 21.0).Prior to analysis, variables of interest were examined for violations of statistical assumptions (e.g., skew, kurtosis, extreme outliers).No outliers were in need of modification.Descriptive statistics were run on variables of interest to examine prevalence rates (e.g. for exposure to interpersonal violence, depression symptoms).Pearsoncorrelations were used to explore associations between mental health symptoms and other variables described previously in the literature review and measures sections.
Participant demographics: Average age of participants (N = 20) was 43 years (SD =7.10, range= 28-55 years).Eighty-five percent indicated Hinduism was their primary religion (n=17), ten percent (n=2) reported Christianity, and five percent (n=1) did not indicate any religion.All participants reported being married, but only 15 percent (n=3) indicated that they currently lived with her husband.Twenty-five percent (n=5) reported being separated from their husbands and 60 percent (n=12) were widows at the time of the survey.Ninety-five percent (n=19) reported having children (M =1.90; SD=1.21).
Seventy-five percent of women (n=15) reported having no more than fifth standard (or approximately 6th grade in the US school system) education (years of education M= 4.80; SD=3.72).Seventy-five percent (n=15) were employed in some capacity at the time of the survey, with the majority engaged with day labor or other sporadic work.Ninetypercent (n=18) reported having less than (money, other resources) or much less than others in the community.

HIV-related Stigma, Negative Social Reactions:
The majority of respondents reported HIV-related stigma, indicating others blamed them for their HIV status (74%), and stopped spending time when them due to their status (68%).
Social Support: Perceived social support was low: nearly all respondents indicated they did not have many people to talk with (95%), or to help them problem-solve (84%).More than half (68%) indicated that they don't have anyone to help them when they are sick.
Trauma exposure, interpersonal violence: Most women reported multiple types of trauma exposure.On average each participant reported 1.5 events (median= 2; range=0-4).Ten percent reported a history of exposure to natural disasters.Fifteen percent reported having been in a motor vehicle accident.Thirty-five percent of women reported physical assault by a stranger.Twenty sixpercent reported a history of an unwanted sexual experience.Notably the most frequent form of trauma exposure (65%) was physical assault by a close family member (e.g.partner, husband, mother-in-law).Furthermore, 80 percent reported interpersonal abuse by either a family member or stranger.Eighty-percent of all reported incidents occurred during adulthood, with 20 percent occurring during childhood.Only fifteen percent did not endorse at least one form of exposure to adverse events.

Depression:
Traditionally, an average score of >1.75 on the HSC is comparable to a diagnosis of clinical depression.When calculating each participant's average score, the denominator was adjusted to reflect excluded items.Ninety-four percent of respondents (n=18, 2 missing cases) had an average score >1.75 (M=3.32;SD=0.71).However, because the HSC has not been normed or validated for use in this specific population, diagnostic cut-offs may be misleading.
As a result, we also examined the range of scores in this sample (range = 19-54; M = 43; SD = 9.27), and split the sample into thirds representing low, medium and high
The guilt subscale score was M=17.11;SD=4.12 (range=7-20, out of a possible 5-25).Feeling very small and as if I am drowning in this problem were frequently endorsed items.
Coping: Each item was scored on a scale of 1 to 4, where 1= I haven't been doing this at all and four= I have (1.28)I will say this is not true 1.9 (1.04)I will compromise 1.9 (1.14)I would make a joke of it 1.8 (1.20)I would take it as a joke 1.5 (.89) *this item significantly associated with symptoms of depression, r (20) = .74;p <.000, and PTSD, r (20) =.74; p <.000.
been doing this a lot.Seeking comfort in religion, including engaging in prayers or meditation, was the most common means of coping with HIV status and other stressful life events.This was followed by seeking advice, counseling, and support from others.See Table 3 for the mean scores of each item.
HIV knowledge: Despite the fact that participants were receiving some form of HIV support services from the local NGO (e.g.medication, counseling), 33 percent held at least one inaccurate belief about how HIV is transmitted.All participants agreed that HIV cannot becontracted by shaking hands or sharing the same bathroom.Ninety-five percent agreed HIV is transmitted by sex with an infected person and contaminated needles.However, some believed that HIV can be transmitted by hugging, sharing eating utensils, or through insects.See Table 4 for participant HIV transmission knowledge by item.
Correlations: variables associated with mental health symptoms: We ran a basic correlation matrix for variables of interest (Table 5).Feelings of shame and guilt were significantly associated with depression.Guilt was also associated with symptoms of PTSD.Negative social reactions and social support items were not significantly associated with depression or PTSD symptoms.Self-blame (I will judge myself again and again) was associated with depression, r (20) = .74;p <.000, and PTSD, r (20) =.74; p <.000.women in rural India is experiencing HIV-related stigma, resulting in negative social reactions and limited social support.They have also reported interpersonal violence, including abuse by partners and family members.Rates of depression and post-traumatic stress symptoms are high, and feelings of shame and guilt are common.

Results of this study indicate this sample of HIV-infected
Coping strategies include various efforts to mitigate distress, such as spiritual and religious activities.Although they are all receiving services from a local organization specializing in support for HIV-infected women, HIV knowledge was lacking in some areas.A few factors were associated with mental health symptoms, including feelings of shame, guilt, and self-blame.Somewhat surprisingly, other factors such as social support were not significantly associated with mental health symptoms.This may have been a result of the small sample size and associated lack of statistical power.
Findings from this exploratory study suggest HIVinfected women are highly stigmatized and may be rejected by their community.Negative social reactions and a lack of adequate social support were common among the women surveyed.A companion qualitative study was completed at that same time as this research, with participants from the same population of HIVinfected women in India.Qualitative results also indicated many HIV-infected women in this specific community face rejection by family members, and their broader social networks, following disclosure of HIV status (Hunter et al., 2015).
Among this group of women high rates of trauma exposure were reported, including high rates of physical abuse by family members or close others.Although it is not clear from this cross-sectional study whether the majority of the abuse occurred before or after women became HIV-infected, it is possible this is another form of social punishment and rejection associated with HIV status.Future research should examine the potential relationship between domestic abuse and HIV status more closely, especially in communities where HIV is highly stigmatized.
Although mental health measures were not normed and validated for specific use with this population, and as a result, conclusions should be tentative, psychological distress appears to be high among this group of women.Depressive symptoms, PTSD, shame, guilt, and selfblame were reported, and warrant further investigation.Specifically, feelings of sadness, associated crying, lack of hope, anger, rumination, fear and sleep disturbances were common.Feelings of shame and guilt, including feeling insignificant and overwhelmed, were common.
Despite the challenges, women appeared to be using a variety of coping strategies to deal with their HIV status.Religious and spiritual coping was common.These findings are consistent with the companion qualitative study in which women indicated that turning to religion and praying were a helpful means of coping with stressors associated with being HIV-infected.Mental health interventions should capitalize on this preferred coping strategy, ensuring women have adequate access to opportunities to practice religious and spiritual activities.Social support seeking was also a preferred coping strategy, despite the seemingly common negative social reactions and a reported lack of available social support.Given this preference, and the difficulties with access to adequate social support, community-wide antistigma campaigns should be investigated to determine the potential benefits for this population.
Finally, misconceptions surrounding HIV transmission were reported, despite the fact that all women surveyed are receiving services from a local organization, including HIV education.Additional HIV trainings with an emphasis on increasing knowledge about modes of transmission for HIV-infected women and community members could be a focus for future interventions.Community education may also decrease HIV-related stigma and increase social support for HIV-infected women.
Many of the aforementioned factors were examined for potential associations with mental health outcomes.The small sample size resulted in limited statistical power to detect associations however, and as such, may have resulted in inaccurate null results.For example, there was no significant correlation between a history of interpersonal violence and PTSD or depression symptoms.Previous research however, suggests a typically robust relationship between interpersonal violence and mental health outcomes (Campbell et al., 2008;Kilpatrick et al., 2003;Pico-Alfonso et al., 2006).In addition, somewhat surprisingly, negative social reactions and social support were not significantly correlated with mental health outcomes.Negative social reactions however, may influence feelings of shame, guilt, and self-blameall factors that were significantly associated with mental health outcomes in this sample.Future research should examine the potential relationship between negative social reactions and shame, guilt, and self-blame to determine how these constructs may be related.
This exploratory study has some limitations.Most notably, the sample size is small.In addition, the survey instruments were not specifically normed and validated for use with this population.Although the research team focused on equivalence of constructs during the translation process, it is possible that some variables of interest were not fully understood as intended.Finally, participant answers may have been influenced by the presence of researchers affiliated with the local service provider, and in some cases, family or other community members were coming in and out of the room during interviews.Attempts were made by the research team to address this, but these may have been inadequate, potentially resulting in under or over-endorsing of some items.
Despite the limitations, there are several strengths of this exploratory study.The research highlights challenges facing women living with HIV in settings such as rural Southern India where having HIV is highly stigmatized, and can leave one socially ostracized.This study also adds to the limited research examining various factors potentially associated with mental health outcomes for HIV-infected women in lower-income settings such as rural India.Finally, this study has the potential to inform therapeutic interventions for this population, including programs targetingviolence prevention, HIV awareness and anti-stigma campaigns, and feelings of shame, guilt, and self-blame.Future research should explore these potential applications with a larger sample and using a longitudinal design.
had a score of >2.5 (M =2.91; SD= 0.58), indicating an unusually high incidence of PTSD symptoms in this population (n = 18).However, as with the depression scale, diagnostic cut-offs may be misleading because the HTQ has not been normed or validated for use in this specific population.As with the depression measure, we looked at the range of scores in this sample (range = 13-

Table 3 .
Mean scores for individual coping items.

Table 4 .
HIV knowledge items.