Mental disorders are common amongst special populations in whom they both precede and follow HIV infection. In this chapter, the term mental disorders is used to refer to both addictive and non-addictive disorders. Whilst people with intellectual disabilities are also a very important population, there is no data that strongly links this disability to the HIV epidemic. Below we discuss each special population with particular focus on how mental illness and vulnerability to HIV infection travel together.

MSM is a term that describes men of various identities and social contexts who engage in sexual behaviour with other men. Throughout the world, MSM have elevated rates of HIV infection when compared to the general population, and this is particularly pronounced for black MSM. In Africa and across the African diaspora (including the Americas), pooled estimates show that black MSM are 15 times more likely to be HIV-positive compared with the general population and 8.5 times more likely compared with black non-MSM populations [8].

In many parts of the world, HIV surveillance, prevention and treatment are impeded by the stigma, secrecy, discrimination, violence and legal penalties that surround same-sex behaviour. Disparities in the prevalence of HIV infection are greater in countries that criminalize sexual behaviour between men than those that do not [8].

Stigma, discrimination, interpersonal violence and legal sanctions probably also contribute to the elevated rates of psychiatric disorders that have been documented amongst MSM in countries where such studies have been conducted [9]. See the commentary by Magidson and O’Cleirigh for an in-depth discussion of the syndemic conditions that have a synergistic effect on HIV risk amongst MSM.

King et al. [10] conducted a systematic review of mental disorder, suicide and deliberate self-harm amongst lesbian, gay and bi-sexual people. Twenty-five studies, mostly from high-income countries, were considered to be of sufficient quality to be included in a meta-analysis. Using data drawn from this group of studies, the authors concluded that MSM had elevated rates of depression, anxiety disorders, alcohol and other substance dependence and suicide attempts when compared to their heterosexual counterparts. The extent to which prevalent HIV infection explained some of this mental illness comorbidity is not clear [11].

Using data from a comprehensive health survey conducted amongst 571 gay men in Geneva, Switzerland, Wang et al. [12] found the following 12-month rates of the five disorders studied: major depression, 19%; specific and social phobia, 22% and alcohol and drug dependence, 17%. These rates contrasted with the European Study of the Epidemiology of Mental Disorders, which provided population prevalence for 10 psychiatric disorders amongst Western European men and found the following 12-month prevalences: any mood disorder, 2.8%; any anxiety disorder, 3.8% and any alcohol disorder, 1.7% [13].

Some psychiatric disorders amongst MSM may, in turn, reduce HIV-related health care utilization [14].

PWID are becoming more important in the global landscape of HIV/AIDS because injection drug use (IDU) is accounting for a larger proportion of new cases of HIV infection [2].

IDU has been identified in 148 countries, with the largest numbers of injectors identified in China, the United States and Russia [15]. Data about the extent of injection drug use is absent for many countries in Africa, the Middle East and Latin America. Africa, the continent with the largest HIV epidemic, has had an increasing role in global drug-trafficking routes, leading to increasing IDU transmission of HIV [16]. This is especially true in Nigeria, Kenya, Tanzania, South Africa and Mauritius. Kenya alone had 7% of its new HIV infections attributable to IDU in 2008, and 42% of Kenya’s PWID population is HIV infected [16].

It is estimated that outside of Sub-Saharan Africa, one in three new HIV cases is attributed to injection drug use [2]. In parts of the world where the epidemic is growing rapidly, such as Eastern Europe and Central Asia, some estimates show more than 80% of HIV transmission occurring amongst PWID [2]. Estimates vary widely on the prevalence of HIV amongst PWID; however, a 2008 systematic review suggests that the average global prevalence rate of HIV amongst PWID is 18% [15], compared to 0.8% of the general population [17].

Because PWID play a central role in the acquisition and transmission of HIV infection through risky sexual and drug use practices, it is very important for them to have access to safe injection equipment and opioid substitution treatment. Those with HIV infection need to be able to access and adhere to antiretroviral treatment.

Multiple studies have shown that even in settings where antiretroviral treatment (ART) is widely available, PWID are less likely to start ART than other HIV-positive populations [18]. Studies have found that PWID commonly present for HIV treatment in the later stage of the disease when AIDS symptoms are present. Later HIV stage initiation of ART predicts worse prognosis and survival outcomes [18] and, as already noted, poses a greater likelihood of HIV transmission to others. PWID have also been found to be less adherent to ART than other HIV-positive populations [18, 19] and are more likely to discontinue ART outright after it has been started [19].

The most commonly injected drugs are opioids and stimulants [20]. Whilst injection of these substances could be an intermittent behaviour that is not occurring in the context of addiction, studies of current and former drug injectors have found that the vast majority of them suffer from one or more addictive disorders that include both injected and non-injected substances.

Brooner et al. [21] studied 716 opioid abusers seeking methadone maintenance treatment in Baltimore, Maryland. Using the Structured Clinical Interview for DSM (SCID) for DSM-III-R to establish diagnoses, the authors found that 100% of subjects met the criteria for current opioid dependence. In addition, lifetime rates of drug dependence with other substances were as follows: cocaine, 65%; cannabis, 51%; alcohol, 50%; sedatives, 45%; stimulants, 19% and hallucinogens, 18%. In another US study of 158 people who had a lifetime history of injection drug use, 99.4% met the criteria for a lifetime substance abuse or dependence diagnosis when assessed with the Diagnostic Interview Schedule (DIS) [22]. In a study of 55 young adult incarcerated heroin users in Taiwan, of whom 46 were injectors, 87% had a current diagnosis of substance dependence as assessed by the Mini International Neuropsychiatric Interview (MINI) [23].

Type of drug addiction is also linked to risky drug injection. For example, to maintain a high, individuals injecting cocaine, which is rapidly metabolized, must inject much more frequently than those using longer acting drugs. As a result, obtaining new needles in a safe manner for each injection may not always be possible [24].

Comorbidity between substance use disorders and other mental illnesses has been another focus of research in this population [25]. Mental disorders often precede the onset of injection drug use. Again, the most obvious link is to current addictive disorders as already described above. In addition, studies show that early onset of alcohol and polysubstance use is an important risk factor for IDU in adulthood.

Between 1980 and 1985, a series of US-based studies found that up to 80% of opioid users met the criteria for at least one non-substance use psychiatric disorder, with rates of mood disorder and antisocial personality disorder far exceeding general population estimates [21]. In an attempt to distinguish between psychiatric syndromes induced by drug intoxication/withdrawal from those that are independent of drug use, Brooner et al.’s study [21] assessed 716 opioid abusers after they had been stabilized on methadone maintenance and found that 24% of subjects met lifetime criteria for an Axis-I non-substance psychiatric disorder, most commonly major depression. On Axis-II, 35% had a personality disorder, most commonly antisocial personality disorder.

A team in Chicago [26] used various outreach techniques to gather mental health information on 570 young and mostly white PWID who were not currently in treatment and who had injected less than 1 month ago. Lifetime prevalences of major depression, post-traumatic stress disorder (PTSD) and other anxiety disorders and antisocial and borderline personality disorders were much higher than in the general population.

Depression has been found to increase unsafe injection drug use practices. In a study of 343 opioid-dependent adults recruited from 12 sites across the United States and enrolled in multisite studies of the National Drug Abuse Treatment Clinical Trials Network (CTN001-002), depressive symptoms were associated with an increased level of injection risk behaviours [27]. In a large study in Sweden examining almost 7000 criminal justice clients with suspected substance-related problems, a history of drug injection was one of the factors associated with past suicide attempts [28]. There is also evidence that psychological distress can shorten the time from HIV to AIDS, particularly amongst IDU women [29].

Unfortunately, studies of rates of mental disorders amongst PWID in most countries are not available.

Commercial sex work, whether legal or illegal, is an economic exchange in which specific sexual activities are purchased or traded for other goods. Many social and economic factors are associated with sex work, including extreme poverty, illiteracy and unaddressed (or even sanctioned) violence against women and MSM.

Childhood sexual and physical abuse histories are common amongst male and female sex workers [30–35]. Moreover, throughout the world, sex workers report being raped and physically assaulted during the course of their work. These childhood and adult traumas are associated with significant suicidal ideation and risk, and high rates of mental disorders, especially alcohol/substance use disorders, depression and PTSD [11, 30–38].

Using a variety of outreach techniques, Farley et al. [31] interviewed 475 people engaged in prostitution in five countries: South Africa, Thailand, Turkey, the United States and Zambia [31]. The sample included women, men and transgendered people who were primarily working on the street. Histories of abuse and violence were pervasive: 54% of interviewees reported physical abuse in childhood, and 58% reported childhood sexual abuse. In the context of prostitution, 73% had been physically assaulted, 68% had been threatened with a weapon and 62% had been raped. The Unites States had some of the highest rates of these violent events and offered some of the poorest support services. On using the PTSD checklist (PCL), a 17-item scale that assesses Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV) symptoms for this disorder, 67% of the respondents met the criteria for PTSD. On average, 92% stated that they wanted to leave prostitution. The authors argue that prostitution, whether legal or not, primarily exploits the poorest people with the fewest choices and should be seen as a human rights violation.

Farley and Kelly’s review of the literature on prostitution published in 2000 cited many other studies that found similar rates of abuse and violence: sexual abuse in childhood, 50–90%; physical abuse in childhood, 60–90%; rape during prostitution, 40–85% and physical assault during prostitution, 60–87% [32]. The authors conclude that women in prostitution are socially invisible battered women, and that focusing on them as simply vectors of HIV transmission overlooks their many vulnerabilities and needs.

Roxburgh et al.’s study of 72 female street-based sex workers in Australia found that 75% of these sex workers experienced sexual abuse in childhood with the mean age of the first episode being 7 years of age [34]. Amongst these women, 47% met a lifetime DSM-IV diagnosis of PTSD.

A study of suicidal behaviour amongst 326 female sex workers in Goa, India, found that in the previous 3 months, 35% of the women had suicidal ideation and 19% had attempted suicide. This compared with a 0.8% annual incidence of attempted suicide in the general population of women in Goa [35].

A study of 310 female sex workers in China found that 30% had elevated depressive symptoms, 18% had suicidal ideation and 9% had made a suicide attempt in the past 6 months [33]. Greater perceived stigma on the part of sex workers was associated with poorer mental health.

Interviews were conducted with 193 female sex workers in Zurich, Switzerland, using the Composite International Diagnostic Interview (CIDI) to determine rates of mental disorders. One-year prevalence data showed that 50% of the women had a mental disorder: 24% had major depression, 12% had dysthymia, 13% had PTSD and 34% had other anxiety disorders [37].

Two studies found that both male and female sex workers had elevated rates of psychopathology when compared to matched controls [32, 34]. It was unclear if psychopathology preceded or followed sex trading.

Some studies suggest that mental illness is associated with higher HIV prevalence and lower rates of condom use amongst sex workers. In one study conducted in Puerto Rico [39], sex workers with high levels of depressive symptoms had a 70% HIV infection rate, whereas those with low depressive symptoms had a 30% infection rate. This did not appear to be a consequence of HIV infection, since depressive symptoms were independent of HIV status.

Vanwesenbeeck [40] conducted an exhaustive review of the research literature on commercial sex work from 1990 through 2000 and analysed it from a ‘pro-sex work feminist frame of reference’, meaning that sex work is on principal considered legitimate work. Vanwesenbeeck emphasized that the tendency of researchers to focus on street sex workers and those recruited in jails or through social service agencies paints a much bleaker picture than would studies focused on indoor sex workers. She further concludes that in the Western world, IDU and non-commercial sexual activity are the most important risk factors for HIV infection in female sex workers. Vanwesenbeeck agrees that in Africa, the current epicentre of the HIV epidemic, commercial sex work also has a role in the spread of HIV. Documented rates of HIV amongst sex workers in Africa include 58% amongst female sex workers in Burkina Faso [41] and 61% and 43%, respectively, amongst female sex workers recruited from truck stops who did or did not have anal sex with their clients [42]. Extensive social networking of commercial sex workers has been documented in Nigeria and Uganda [43, 44].

Key populations are not mutually exclusive. An individual may simultaneously or over time belong to more than one key group. In particular, rates of drug use, including injection drug use, are high amongst sex workers in many countries [2]. Addiction to drugs can lead to sex work in exchange for drugs or for money to purchase drugs.

Vaddiparti et al. [45] developed a model to evaluate the association between childhood victimization, perpetration of violence and later cocaine dependence and adult sex trading amongst drug using women. A cohort of 594 women (362 sex traders) was recruited using community outreach strategies for HIV prevention studies in St. Louis (United States). Rates of cocaine dependence were higher amongst traders (85% vs 56%). Path analysis confirmed that childhood victimization had a significant and direct association with both adult cocaine dependence and sex trading. However, the association between childhood victimization and adult sex trading was mediated by cocaine dependence.

A study of 1606 women and 3001 men entering substance use treatment in the United States found high rates of prostitution [30]. In the past year, 41% of women and 11% of men reported having engaged in prostitution; lifetime rates were 51% amongst women and 19% amongst men. In both men and women, prostitution was associated with more mental health symptoms, injection drug use and HIV infection.

A study in Zanzibar, Tanzania, looked at 509 MSM living in the community, of whom 66 also injected drugs [46]. MSM-IDU were twice as likely to have HIV infection than non-IDU MSM. They were also five times less likely to wear a condom with a paid female partner and 10 times less likely with a non-paid female partner. MSM-IDU were much more likely to have engaged in group sex with other men in the past month. They also reported poor needle habits, with a majority indicating that they used a needle after someone else and had passed around a needle after using it themselves. This study demonstrates that whilst MSM are at high risk of HIV acquisition and transmission, those who are also injecting drugs are at even higher risk, in part through unsafe drug injection practices and in part through links to commercial sex work and other risky sexual practices.

In Roxburgh et al.’s study of sex workers in Australia, more than 80% were heroin dependent and injecting drugs [34]. About half of these sex workers had begun injecting drugs prior to sex work and used sex work to pay for their drugs. Half of the women reported using drugs to facilitate their sex work largely through their numbing effects. What emerges is a complex intertwinement of childhood abuse and neglect, PTSD, symptoms of depression, injection drug use and engaging in sex work. In Farley’s studies of 475 people who engaged in prostitution in five countries, 38% self-reported the need for alcohol or drug addiction treatment [31].

In a study in St. Petersburg (Russia), 81% of surveyed sex workers said they injected drugs at least once a day (65% had used non-sterile injecting equipment), and 48% of sex workers were HIV-positive [2]. A study in Puerto Rico found that 47% of female sex workers injected drugs [39]. Similar rates of HIV infection were reported amongst female sex workers who inject drugs in Ho Chi Minh City (Vietnam) [2]. Amongst men who inject drugs in Vietnam, having contacts with female sex workers was associated with a greater likelihood of being HIV-positive [47]. Illegal drug use, particularly with injection drugs, was the single greatest risk factor for HIV infection amongst female sex workers in Kaiyuan City, China [12].

In one Canadian study, transition to injection drug use was associated with involvement in sex work amongst Aboriginal people [48]. In another Canadian study, this time of HIV-infected IDU who had achieved viral suppression on ART, sex-trade involvement was associated with viral rebound [49].

Sub-Saharan Africa has had an alarming epidemic of HIV infection amongst adolescent girls and women [2]. Areas that have received particular attention in this context are sexual and gender-based violence, including during war and violent conflict; the power differential between men and women and the lack of economic opportunities that would empower women. See Mathews’ commentary for an in-depth review and discussion on intimate partner and sexual violence.

Numerous studies have shown positive associations between a history of being the victim of sexual violence, engaging in HIV-related sexual risk behaviours and being infected with HIV. Most of these studies cannot answer the question of which came first, sexual violence or HIV infection. However, one recent cohort study of young HIV-negative women in South Africa demonstrated that the acquisition of HIV infection was linked to intimate partner violence [50]. Other studies have demonstrated high rates of anxiety and depression amongst HIV-infected women in Sub-Saharan Africa [51, 52].