Mental health outcomes in HIV and childhood maltreatment: a systematic review

Background High rates of childhood maltreatment have been documented in HIV-positive men and women. In addition, mental disorders are highly prevalent in both HIV-infected individuals and victims of childhood maltreatment. However, there is a paucity of research investigating the mental health outcomes associated with childhood maltreatment in the context of HIV infection. The present systematic review assessed mental health outcomes in HIV-positive individuals who were victims of childhood maltreatment. Methods A systematic search of all retrospective, prospective, or clinical trial studies assessing mental health outcomes associated with HIV and childhood maltreatment. The following online databases were searched on 25–31 August 2010: PubMed, Social Science Citation Index, and the Cochrane Library (the Cochrane Central Register of Controlled Trials and the Cochrane Developmental, Psychosocial and Learning Problems, HIV/AIDS, and Depression, Anxiety and Neurosis registers). Results We identified 34 studies suitable for inclusion. A total of 14,935 participants were included in these studies. A variety of mixed mental health outcomes were reported. The most commonly reported psychiatric disorders among HIV-positive individuals with a history of childhood maltreatment included: substance abuse, major depressive disorder, and posttraumatic stress disorder. An association between childhood maltreatment and poor adherence to antiretroviral regimens was also reported in some studies. Conclusion A broad range of adult psychopathology has been reported in studies of HIV-infected individuals with a history of childhood maltreatment. However, a direct causal link cannot be well established. Longer term assessment will better delineate the nature, severity, and temporal relationship of childhood maltreatment to mental health outcomes.


Background
Abuse is a common phenomenon in countries where the prevalence rate of HIV is also high and can include physical, sexual and emotional violence and deprivation or neglect [1]. Studies conducted in developing countries such as South Africa and other African countries have reported high rates of abuse in both adults and children.
This includes intimate partner violence (IPV), rape, and childhood abuse or maltreatment [1][2][3]. Childhood maltreatment has been defined in many different ways. However, for the present review, childhood maltreatment included emotional, physical, and sexual abuse and emotional and physical neglect. According to Bernstein et al. [4] sexual abuse is defined as 'sexual contact or conduct between a child younger than 18 years of age and an adult or older person.' Physical abuse is defined as 'bodily assaults on a child by an adult or older person that posed a risk of or resulted in injury.' Emotional abuse is defined as 'verbal assaults on a child's sense of worth or well-being or any humiliating or demeaning behaviour directed toward a child by an adult or older person.' Physical neglect is defined as 'the failure of caretakers to provide for a child's basic physical needs, including food, shelter, clothing, safety, and health care.' Emotional neglect is defined as 'the failure of caretakers to meet children's basic emotional and psychological needs, including love, belonging, nurturance, and support' [4]. Although women are more vulnerable and regarded as particularly at risk for abuse, men are also victims of rape and childhood maltreatment.
Many studies have investigated the link between adverse childhood experiences such as physical and/or sexual abuse and HIV risk. The experience of childhood maltreatment may increase HIV infection risk indirectly by increasing high-risk behaviors or by interfering with HIV prevention choices [5]. For example, many of the outcomes associated with childhood maltreatment place individuals at increased risk of contracting HIV through behaviors such as transactional sex, unprotected sex, inability to negotiate condom use, alcohol and/or drug abuse, early onset of sexual activities, and multiple sex partners [6][7][8][9][10]. In addition, childhood maltreatment may directly increase the risk of HIV infection through sexual abuse. Injury and the tearing of tissue resulting from sexual violence may increase the likelihood of HIV infection [11]. Studies have also found that childhood maltreatment is strongly associated with adult revictimization which can further increase the risk for HIV among women [5].
The mental health outcomes of HIV-infected individuals have been well documented to date. Research suggests a significant burden of mental illness in individuals living with HIV/AIDS, both globally and in the developing world. Mental illnesses documented in HIVinfected individuals include predominantly substance use, anxiety, and mood disorders [12][13][14][15][16][17][18][19]. Moreover, it has been suggested that HIV disease progression may be hastened by mental disorders such as depression and anxiety [20].
Similarly, research suggests the long-term mental health outcomes of childhood maltreatment include predominantly substance, anxiety, and mood disorders [21,22]. Interestingly, Kaplow and Widom [23] followed 496 individuals with neglect, physical and sexual abuse prior to the age of 12 into adulthood. Their research suggests that an earlier onset of maltreatment predicted more symptoms of anxiety and depression in adulthood, while controlling for gender, race, current age and reports of other abuse. Later onset of maltreatment was predictive of more behavioral problems in adulthood [23]. In a review of child sexual abuse, Johnson [11] outlines a number of child and adult psychological and behavioral consequences of child sexual abuse. These include substance use disorders, and anxiety and mood disorders, amongst others [11].
Although many studies have focused on mental health outcomes in childhood maltreatment and HIV separately, there is a paucity of research investigating childhood maltreatment and HIV in combination, and the associated mental health outcomes in dually affected men and women. HIV-infected women may face more current and past negative life events than men in developing parts of the world [13] and this may lead to significant adult psychopathology and poor adherence to antiretroviral medications [24,25]. In light of this, it is evident that HIV-positive individuals, women in particular, are vulnerable to risk factors associated with abuse, and abuse-related changes in behavioral functioning, which may complicate HIV infection. A systematic assessment and summary of the available evidence is therefore warranted in order to add to the available evidence for both clinical and research decision making.

Search strategy and selection criteria
We searched the electronic databases PubMed, Social Science Citation Index, the Cochrane Library (The Cochrane Central Register of Controlled Trials: CENTRAL) and the Cochrane Developmental, Psychosocial and Learning Problems, HIV/AIDS, and Depression, Anxiety and Neurosis registers on 25-30 August 2010. No limit on the time period was applied to the search in order to avoid omission of relevant studies. Reference lists of articles identified through database searches and bibliographies of systematic and non-systematic review articles were examined to identify further relevant studies. We included all English language, original research (retrospective and prospective studies) and clinical trials reporting mental health outcomes of childhood trauma in HIV-positive individuals. The population included adult men and women already infected with HIV/AIDS who experienced childhood maltreatment prior to 18 years of age. We excluded systematic and non-systematic review articles and studies of no direct relevance to the comprehensive search. The PubMed search included the following terms: childhood abuse AND HIV. . No filters were included to ensure that all relevant papers were retrieved. The PubMed search selected those studies that addressed childhood abuse and HIV in all fields. An initial search of titles was undertaken by the reviewer (GS). Studies were included irrespective of sample size and period of follow-up. Titles and abstracts of studies that appeared relevant were then assessed to determine whether they met the inclusion criteria. Abstracts that did not meet the inclusion criteria were rejected. The reviewer assessed full texts of articles that appeared to meet the inclusion criteria of the present study. Information was extracted regarding population characteristics and sample size, study design, outcomes measured and results. No exploration of publication bias was undertaken and it was not possible to conduct a sensitivity analysis for the current review article due to the fact that no meta-analyses were conducted (see Figure 1).

Results
All databases searched yielded abstracts, and there were duplicates between the databases. All the studies had published results in peer-reviewed journals. Two hundred and five abstracts were identified and reviewed. Of the 205 abstracts identified, 171 articles were excluded. Studies were excluded if they were: of no relevance to the present review, systematic or non-systematic review articles, or not conducted in the population of interest. Titles and abstracts of 52 studies that appeared relevant were then assessed to determine whether they met the inclusion criteria. Of the relevant studies reviewed, 34 articles met inclusion criteria. Three of the 34 articles were sourced from reference lists of other manuscripts. Full text articles for all 34 studies were accessed and reviewed. The reviewed articles are summarized in Table 1. There was heterogeneity in sample characteristics, study methodologies and outcome measures among all studies reviewed in this article. Moreover, some studies included an HIV comparison group whereas others did not, further limiting comparability. For these reasons, it was decided that a meta-analysis of these data was not feasible. Variability in measurement of mental health impairment was noted. Psychiatric symptoms and disorders were assessed according to standard diagnostic criteria, using a structured clinician administered interview and/or through self-report (see Table 1). Although some studies differentiated symptoms and diagnoses, others reported more global levels of psychological distress. For example, two articles sourced reported on global psychological distress and mental health in general, without delineating whether symptoms were depressive in nature or anxiety related, for example [26,27]. Furthermore, some studies simply stated the percentage of HIV-positive maltreatment victims reporting symptoms of anxiety. Although these studies reported global anxiety levels, they failed to differentiate by diagnosis [7,[28][29][30].
A history of childhood maltreatment was also assessed in different ways, but all studies relied on self-reported history of childhood maltreatment, and most assessments were retrospective in nature. In some studies, childhood maltreatment included various forms/types such as physical abuse and neglect, emotional abuse and neglect, and sexual abuse [43]. Other studies only examined childhood sexual abuse (CSA) [6,26,40] or combined sexual and physical abuse into one category of child abuse [25,39,41]. Some studies utilized validated self-report measures sensitive in tapping into various forms of childhood abuse and neglect [38,43]    (1) Sexual revictimization.
(2) PTSD.                HIV-positive women were more likely to report a history of CSA.
HIV-positive women with a history of CSA were more likely to report posttraumatic stress, substance abuse, chronic stress, and psychiatric history than HIV-negative counterparts.
HIV-positive women with a history of CSA were more likely to report risky health behaviors than HIVnegative counterparts.   Quantitative cross-sectional survey design.
Self-administered surveys were used to assess demographics, sexual abuse history, substance use and sexual risk behaviors.
(3) Having undergone treatment for substance abuse.
93 (15%) of the men reported being forced to have sex when they were 16 years or younger by a man at least 5 years older. Of these 93 men, the average age of first abuse was 9.3 years. Sexually abused men were more likely to report childhood physical abuse relative to nonabused men (41% vs. 12%). Men who were sexually abuse were more likely to have tested HIV-positive (40%) relative to non-abused men (19%). 77% of the men were HIV-negative and 22% were (9%).
HIV-positive. Abused men endorsed more symptoms of borderline personality disorder. Contrary to expectations, abused men did not differ in dissociation symptoms or trauma-related anxiety when compared to non-abused counterparts. Abused men were more likely to report alcohol and drug abuse in the past 6 months and having undergone treatment (28%) compared to non-abused men. ( infected victims also reported significantly greater distress with physical symptoms and higher rates of AIDS-defining conditions than did non-victims. Trauma Questionnaire [4]. However, many studies established a history of childhood abuse by simply asking a single question such as 'have you ever experienced a sexual assault or rape as a child or teenager, that is, when you were 18 years of age or younger?' and using a dichotomous response option (Yes/No) [26,28,42].

Childhood maltreatment
Childhood maltreatment, such as physical and sexual abuse is a common phenomenon in the general population (uninfected individuals). CSA is reported by as many as 32% of women and 14% of men in the general population, whereas physical abuse is experienced by 22% of males and 19.5% of females in the general population [55]. However, rates of childhood maltreatment in HIV-positive individuals are significantly higher, suggesting that the experience of childhood maltreatment in the context of HIV is worthy of greater attention. Rates of CSA among HIV-positive individuals range from 32% to 76%, respectively [28,56,57].
The percentage of participants in individual studies who received a diagnosis of mood disorders included (study number in Table 1): 46.4% (4) 68% (10), 15% (20), 30% (22), 34.7% (23) and 39% (28). Other studies reported depression scores on self-report/interviewer administered instruments. The mean depression score on the Center for Epidemiologic Studies Depression Scale (CES-D) was 23 in one study (6) and in another study, 81% of women and 76% of men had a depression score higher than 16 on the CES-D (17). The mean score for depressive symptoms on the Beck Depression Inventory was 29.8 in women and 28.2 in men in another study (24). Mood disorders was not an inclusion criterion in the research study but rather an unselected observation for most studies (20,23,24,28). However, in other studies, participants were only included if there was evidence of psychological distress or criteria for mood or anxiety disorders were met (4,22).
In comparison to the general population (i.e. uninfected and non-abused counterparts), evidence suggests that ongoing risk behaviors and rates of psychopathology are higher in HIV-infected individuals with histories of abuse [7,8,10,31,37,53]. HIV-positive individuals were more likely to report posttraumatic stress, risky health behaviors, substance abuse, chronic stress, and psychiatric history compared with HIV-negative counterparts [53]. In addition, abused individuals reported higher rates of mental illness, compared to non-abused counterparts, suggesting that a history of abuse in childhood increases the likelihood of psychopathology [58]. These findings lend credence to the argument that childhood maltreatment in the context of HIV is worthy of greater attention.

Intervention studies
The review revealed six intervention studies that have been conducted with this population [8,[31][32][33][34][35]. Three of these interventions were carried out in bisexual men and men who have sex with men (MSM) [8,31,35] and three were carried out in mixed samples of males and females [32][33][34]. A total of 4295 MSM were enrolled into a behavioral intervention trial over 48 months. Behavioral assessments were conducted every 6 months. However, the results revealed that among men reporting a history of CSA, the intervention had no effect in reducing HIV infection rates. Moreover, men reporting a history of CSA were more likely to display depressive symptomatology and use nonprescription drugs [8]. Similarly, 49 gay and bisexual HIV-infected men with histories of CSA were enrolled into an intervention study, consisting of 15 coping group sessions. When compared to an alternative support group intervention and a control condition, the coping group intervention proved to be efficacious in treating HIV-positive adults with histories of CSA. This was attributable to the inclusion of a coping skills training component in the aforementioned treatment condition [31]. Support for the efficacy of the aforementioned coping group intervention was reported in a separate study assessing 28 men and women with HIV and histories of CSA [33]. Similarly findings were reported in another study utilizing the same coping group intervention in 198 HIV-infected men and women with histories of CSA [34]. Reductions in intrusive traumatic stress symptoms were exhibited among participants in the coping group intervention compared to the waitlist condition and in avoidant traumatic stress symptoms compared to the support group condition [34]. Moreover, the efficacy of the aforementioned coping group intervention in reducing sexual transmission risk behavior was assessed [32]. The sexual behavior of 247 HIVpositive men and women with histories of CSA was assessed at baseline, postintervention, and at 4, 8, and 12 month follow-up periods. The frequency of unprotected sexual intercourse for all partners decreased more among participants in the coping group intervention than participants in the support intervention condition [32]. Lastly, a randomized clinical trial comparing the effects of two six-session interventions was carried out in a sample of 137 bisexual men and MSM. Results from both interventions revealed reductions in sexual risk behaviors and number of sexual partners from baseline to posttest, and from 3 to 6 month follow-ups. No significant differences in depression were evident between the two conditions; however, at 6 months the total sample reported a significant decrease in depressive symptoms [35].

Adherence to antiretroviral medication
In examining mental health outcomes, two articles also reported an association between childhood maltreatment and poor treatment adherence to antiretroviral regimens [25,43]. In one study, a lower percentage of women with a history of physical or sexual abuse reported using Highly Active Antiretroviral Therapies (HAART). Experiencing any physical or sexual abuse increased the likelihood of no HAART use. Women with a history of any physical or sexual abuse were more than 1.5 times more likely to lack HAART, even when clinically eligible [25]. Moreover, the use of HIV medications has been found to be negatively associated with CSA experiences [43].

High risk behaviors
Many studies found participants commonly reporting engagement in high-risk behaviors such as transactional sex or compulsive sexual behaviors [3,6,39,40,42,53]. Individuals who experienced abuse regularly were more likely to be HIV-positive, exchanged sex for payment, and be a current user of sex-related drugs [40]. It has also been reported that women experiencing CSA were more likely than women without such histories to have used drugs, to have had more than ten sexual partners, to have traded sex for money, drugs, or shelter; and to have been forced to have sex with a person known to be HIV-positive [6]. Moreover, women who had been sexually assaulted were significantly more likely to have shared injection drug equipment [3].

Discussion
We performed a comprehensive systematic review of the literature to assess mental health outcomes in HIVpositive individuals with histories of childhood maltreatment. To our knowledge, this is the first review of its kind; no published systematic reviews assessing this association have been conducted to date.
The reported mental health outcomes in dually affected individuals (HIV-positive individuals with histories of childhood maltreatment) are in keeping with studies that have investigated these variables separately [11][12][13][14][15][16][17][18][19]23], supporting at least common outcomes, although assessment of the additive effects of HIV and childhood trauma is difficult in this retrospective review.
The most commonly reported mental illnesses in dually affected individuals included mood, anxiety, and substance abuse disorders. Very few studies examined Axis II disorders. It has been suggested that an HIV diagnosis alone may constitute a significant stressor and thus increase the likelihood of mental illnesses among HIV-positive individuals [59]. Apart from depression or anxiety being a secondary diagnosis to HIV/AIDS, anxiety and depressive symptoms measured over time were also associated with faster progression of the disease after five years. This finding may suggest a reinforcing relationship between HIV and mental illnesses such as depression or anxiety [20]. However, the majority of studies reviewed were cross-sectional in nature, therefore limiting their ability to make causal conclusions around the onset of mental illness in HIV. This highlights the importance of longer term assessment in order to better delineate the nature, severity, and temporal nature of mental health outcomes. Importantly, mental disorders such as depression or anxiety can further impact immune system functioning in HIV and, in turn, influence quality of life and health status [60].
Substance abuse was the most predominant mental health outcome reported in reviewed articles. For the most part, drugs and/or alcohol are used to numb emotional distress and feelings of anger and betrayal resulting from the experience of childhood maltreatment [50]. Not only does substance abuse have direct implications for the progression of the disease in infected individuals [61], it also has direct and indirect implications for the transmission of HIV. Antiretroviral regimens are known to have strong positive effects on quality of life and in improving health status in infected individuals [62]. Few articles have reported an association between childhood maltreatment and poor treatment adherence to antiretroviral regimens or HIV medications [25,43].
Some studies that investigated both early life trauma and adult trauma found an association between childhood trauma and later life trauma [6,28,32,37]. For example, the study by Simoni and Ng found that childhood abuse was correlated with both adult and recent trauma. Moreover, each type of trauma was also correlated with depression scores [49]. Several studies have also found that adult revictimization was very common in survivors of childhood maltreatment [6,32,33,39,50,51]. Further investigation of this relationship and the implications for prevention and intervention is warranted.
HIV-infected men and women may face many current and past negative life events [13] and this may lead to significant adult psychopathology and poor adherence to antiretroviral medications [24,25,63]. In light of this, it is evident that HIV-positive individuals, women in particular, are vulnerable to risk factors associated with abuse, and abuse-related changes in behavioral functioning. These risk factors and behavioral changes may in turn complicate HIV infection.
There are several limitations that warrant mention. First, no search strategy can guarantee the identification of all relevant research, and omission of important studies remains a possibility and may contribute to bias in inferences drawn. Selection or reviewer bias may be a possibility given that studies were not screened or abstracted in duplicate. Second, the heterogeneity across studies presents a problem as it impedes statistical pooling of studies. Third, it is important to note the controversy that abounds in the classification of search terms. There is no standard definition for the experience of abuse as a child. This construct is one that is classified and assessed in a variety of ways. Some studies use broader terms such as childhood adversity or maltreatment, whereas others use more specific terms such as child abuse. For example, in some studies reviewed, childhood maltreatment included diverse types of trauma such as physical abuse and neglect, emotional abuse and neglect, and sexual abuse. Other studies have used more restricted definitions and have only examined CSA or resorted to combining sexual and physical abuse into one category of child abuse. Childhood maltreatment in the present review included emotional, physical, and sexual abuse and neglect. The lack of a standard definition is a further source of bias. Another limitation that warrants mention is that of the design of the review process. The selection criterion for the inclusion of studies in the present review was manuscripts reporting on mental health and childhood maltreatment. However, this should not be mistaken or presented as an argument that childhood maltreatment and impairments in mental health are associated. However, despite these limitations, this review adds substantially to available evidence for both clinical and research decision making.

Implications for future studies
From the present review, it is clear that very few prospective studies have been executed in this domain [31,44]. The majority of research has been crosssectional and has included retrospective assessment of childhood maltreatment in HIV-infected individuals. This may be partly due to reasons associated with feasibility and logistics. As cross-sectional study designs preclude follow-up observations and longer term assessment of outcomes, future research should be prospective in nature and should better delineate the nature, severity, and temporal relationship of childhood maltreatment to mental health outcomes and treatment utilization, as well as the mediators and moderators of these outcomes. These studies will allow both clinicians and researchers to better understand the etiology of common mental disorders in HIV-infected samples and reduce bias when making causal inferences. Thus, longitudinal investigation of mental health outcomes in HIV infected individuals with childhood maltreatment will be key to explaining these causal relationships.

Conclusion
A broad range of adult psychopathology has been reported in studies of HIV-infected individuals with a history of childhood maltreatment. However, a direct causal link cannot be well established. The longer term assessment will better delineate the nature, severity, and temporal relationship of childhood maltreatment to mental health outcomes. There is a need to screen for childhood maltreatment, psychopathology, and associated functioning in HIV-positive individuals and to address these issues in management. Increased focus on the identification and support for children and youth who have experienced childhood maltreatment is necessary. HIV prevention interventions such as education in high-risk behaviors are also a necessity.