Introduction

HIV/AIDS remains an important public health concern in the United States and, while major gains in treatment have been made, many Americans continue to cope with the heavy burden of this pandemic. Though risky sexual behavior and injection drug use are the major risk factors for transmission, socioeconomic and sociocultural factors are also strongly and consistently associated with risk. Those with the fewest socioeconomic resources––the poor, the uninsured, and the drug addicted––carry the greatest burden of HIV/AIDS in this country. Sociocultural and ethnic minority groups––men who have sex with men (MSM), African Americans, and LatinosFootnote 1––are also disproportionately affected. In this review, we examine reasons for the disproportionate impact of HIV/AIDS on Latinos in comparison to whites and address the ways in which psychosocial research could contribute to better addressing the needs of this underserved, rapidly growing U.S. minority group.

In 2000, Latinos made up 12.5% of the population, surpassing African Americans at 12.3% (United States Bureau of the Census 2002). By 2005, the Latino population had grown to 14.4% of the national population, with over 41 million Latinos living in the United States (United States Bureau of the Census 2006). In this same year, Latinos accounted for 18% of HIV/AIDS diagnoses (Centers for Disease Control 2007). While we recognize that African Americans face the most disproportionate burden of HIV/AIDS among all US ethnic groups, little research has focused on Latinos and reviews of black-white disparities in HIV/AIDS are readily available (e.g., Blankenship et al. 2005; Millett et al. 2006; Smith et al. 2000). Therefore, we focus on secondary and tertiary prevention opportunities with HIV-positive Latinos.

In 2002, HIV/AIDS was the third leading cause of death among Latino males aged 35–44, accounting for 10.4% of all deaths, and the fourth leading cause of death among Latinas in the same age group (6.5% of deaths). These rates are more than double the HIV/AIDS death rates of white males (4.8%) and females (1.9%) in the same age group (Anderson and Smith 2002). Since the advent of highly active antiretroviral therapy (HAART), annual AIDS incidence has fallen significantly. However, the proportion of AIDS cases for U.S. Latinos in 2005 was 3-times higher than the proportion of AIDS cases among whites. Between 2001 and 2005, the estimated number of deaths of persons with AIDS decreased among whites and blacks but remained stable among Latinos (CDC 2007).

In the general U.S. population, while AIDS incidence has decreased over time, AIDS prevalence has increased. This increase can be attributed to the longer life expectancies of HIV-positive patients taking new antiretroviral therapies, including those who have already progressed to AIDS. However, estimated AIDS prevalence among Latinos increased by 130% from 1993 to 2001, compared to a 68% increase among whites (CDC 2001). In 2004, the annual rate of HIV/AIDS diagnosis among Latino males was 60.2 per 100,000 and among Latino females was 16.3 per 100,000. The overall rates for non-Latino white males and females in 2005 were 18.7 and 3.2, respectively (CDC 2006). For men between the ages of 25 and 44, the HIV/AIDS-related death rates per 100,000 in 2003 were 10.3 for Latino men, and 6.2 for white men. For women in the same age range, the rates were 3.8 for Latinas and 1.3 for white women (National Center for Health Statistics 2005). These data indicate that HIV-positive Latinos have significantly worse outcomes than HIV-positive whites.

One challenge in calculating HIV/AIDS estimates by ethnic group is the survey terminology that attempts to group together the heterogeneous peoples of Latin America, often resulting in inaccuracies or distortions. The term ‘Hispanic’ is an ethnic designation first used in U.S. surveillance reports beginning in the 1970s to identify all persons of Latin American heritage, regardless of race or country of origin. However, many Latinos—persons of Latin American heritage—especially those born outside of the U.S., may not self-identify, or be accurately identified as ‘Hispanic.’ Studies reviewed below employed different systems for ethnic designations, making it difficult to synthesize the information. Unless otherwise noted, the data presented here refer primarily to Latin Americans living in the U.S.; however, it is likely that there are inconsistencies in the make-up of the samples.

In the following pages, we review the literature to attempt to capture the reasons underlying the HIV/AIDS disparities between Latinos and whites. First, we focus on demographic, socioeconomic, and socio-cultural factors relevant to the imbalance in HIV/AIDS outcomes compared to whites. We then review the impact of drug abuse and mental health problems among Latinos. Finally, we discuss opportunities for behavioral science intervention research to enhance HIV/AIDS outcomes in HIV-positive Latinos. Our review will show that there is a great need for behavioral science research to address the needs of HIV-positive Latinos. The initial evidence suggests that culturally sensitive interventions focusing on improving mental health, substance abuse, treatment adherence, health literacy, and access to care could be especially effective in improving outcomes for this group.

Factors Relevant to HIV/AIDS Disparities among U.S. Latinos

Demographic, socio-economic, and socio-cultural factors influence the degree to which individuals access and utilize health care services, and these factors may play a role in explaining racial and ethnic differences in health care (Institute of Medicine 2002). Differences in HIV/AIDS outcomes and treatment between Latinos and whites in the U.S. must be considered in the context of these factors. Important demographic factors include nation of origin, area of residence, and mode of HIV infection. Socio-economic factors include income, education, language skills, and access to appropriate care. Socio-cultural factors include knowledge, beliefs, and health-care provider communication. Such elements may be associated with aspects of culture and acculturation, including the preference for Spanish language versus English, norms for sexual behavior, stigma, and social oppression.

Demographic Factors Relevant to HIV/AIDS Disparities

While HIV/AIDS disproportionately affects the entire U.S. Latino population, there are significant within-group variations. For example, Latinos born in the U.S. and Puerto Rico have higher rates of HIV/AIDS compared to foreign-born Latinos (National Minority AIDS Council 1999). Likewise, there are significant variations by U.S. area of residence. Latinos are concentrated by geography and AIDS prevalence follows this concentration, with five states/territories—New York, Puerto Rico, California, Texas, and Florida—accounting for three-quarters of AIDS cases among Latino adults and adolescents between the years 2001 and 2002 (Henry J. Kaiser Family Foundation 2003). Therefore, it is important to note that significant within-group variability exists in the U.S. Latino population.

Similar to their white American counterparts, most HIV-positive Latino men are infected through either unprotected sex with men or shared needles from injection drug use (IDU). Recent data demonstrate this pattern, with 59% of HIV-positive Latino males becoming infected through male-to-male sexual contact and 19% contracting the virus through IDU. An additional 4% reported a combination of male-to-male sexual contact and IDU (CDC 2007). However, there appear to be differences in risk depending on nationality. For example, MSM make up the largest risk group among Latinos born in Mexico (48%), the U.S. (37%), Central or South America (36%), and Cuba (34%). However, IDU is the largest risk category for Puerto Ricans (46%; National Minority AIDS Council 1999).

For Latinas, the majority of HIV-infections occur through heterosexual contact (73%), with an additional 23% infected through IDU (CDC 2007). Many Latinas infected through heterosexual contact are in long-term, perceived monogamous relationships, and are at risk due only to their partners’ risky behavior. For example, Rhodes et al. (2007) found that Latino MSM were more than three times as likely as white MSM to have had sex with a woman in the past 2 years, and 2.3 times as likely as white MSM to have used condoms inconsistently during anal sex over the past 3 months. Furthermore, in a sample of 4928 male respondents from the 2002 national Survey of Family Growth, Adimora et al. (2007) found that Latino men were more than twice as likely as white men to have concurrent sexual partnerships. Latina partners of non-monogamous or IDU men often remain unaware of their HIV risk until they or their partners become ill, or until they are screened for HIV during pregnancy care (Rios-Ellis 2003).

Socioeconomic Factors Relevant to HIV/AIDS Disparities

Socio-economic vulnerabilities of U.S. Latinos in terms of income, education, language skills, and access to health care may be associated with aspects of HIV prevention and care. For instance, data from the U.S. Census Bureau, which may be biased due to under-reporting by illegal Latino immigrants, shows that the 2005 median income for Latino households was $35,967, while the median income for non-Latino white households was $50,784 (DeNavas-Walt et al. 2005). The 2002 National Survey of Latinos, conducted by the Pew Hispanic Center and the Kaiser Family Foundation, found that half of all Latinos report having an annual household income under $30,000. In terms of education, over half (55%) of foreign-born Latinos have less than a high school education compared to fewer than a quarter (23%) of native-born Latinos. Native-born Latinos are more likely to have completed high school than foreign-born Latinos (35% vs. 29%), have some college education (29% vs. 9%), or have graduated from college or received a degree after college (13% vs. 7%). With regards to language, native-born Latinos are more likely than foreign-born Latinos to speak English as their primary language (61% vs. 4%) or to be bilingual (35% vs. 24%). Finally, regarding health care, this survey also found a higher rate of Latinos who reported being without health insurance (35%) than whites (14%). A significant number of Latinos also reported difficulty paying for medical bills (22%), delaying seeking care because of costs (20%), or challenges in getting needed health care services (15%). A substantial number of Latinos also report problems communicating with health care providers due to language barriers (29%), or difficulty obtaining care due to their ethnicity (18%; Pew Hispanic Center and Kaiser Family Foundation 2002). Thus, fewer educational and economic resources may represent vulnerabilities to worse HIV/AIDS outcomes for U.S. Latinos.

The socioeconomic barriers that HIV-positive Latinos face mirror the barriers faced by the general Latino population. While Latinos appear to get tested for HIV as often and as soon as whites, they generally have much poorer health insurance coverage. In the HIV Cost and Services Utilization Study (HCSUS), which included a nationally representative sample of people with HIV/AIDS in 1996, 24% of Latinos reported being uninsured, compared to 17% of whites (Bing et al. 2001). Furthermore, Latinos were more likely than whites to have had less than two outpatient visits in the prior six months (Shapiro et al. 1999).

The HCSUS also provides information about other socioeconomic barriers to care faced by HIV-positive Latinos. For example, 19% of Latinos with HIV reported postponing care due to lack of transportation, compared to 11% of whites, and were more likely to postpone care because they were too sick or had competing needs. They also delayed care after diagnosis more often compared to whites (23% vs. 15%; Turner et al. 2000). Similarly, among HIV-positive patients receiving care in or near Boston and Providence, only 50% of Latinos received HAART compared to 75% of whites (Stone et al. 1998). Thus, Latinos have less access to appropriate HIV-care than whites and face more barriers to obtaining care, possibly due to socioeconomic disparities.

Sociocultural Factors Related to HIV/AIDS Disparities

National survey data provide interesting information on Latinos’ views of HIV/AIDS. For example, 83% of Latinos say that access to HIV care is a problem, 73% report that access is affected by income, and 64% report that access is affected by race/ethnicity (Kaiser Family Foundation 2001). This survey also found that respondents who preferred to be interviewed in Spanish expressed greater concern about HIV/AIDS, were significantly more likely to have misperceptions about risk behaviors associated with HIV infection, and expressed a greater need for information about HIV/AIDS. Misperceptions and need for HIV information were also greater among Latinos with less education.

Other empirical investigations have suggested a lack of accurate knowledge concerning HIV/AIDS among Latinos (Scott et al. 1998; DiClemente et al. 1988; Aruffo et al. 1991). Latinos seem more likely to believe myths about HIV transmission by casual contact and less acculturated Latinos are more likely to have these beliefs (Marín and Marín 1990). Acculturation, or the level of familiarity with and incorporation of the American culture (e.g., one aspect of acculturation is English language preference), has been positively related to AIDS knowledge (Organista et al. 1998; London and Driscoll 1999). However, in a multivariate analysis of 121 Latinos from New Jersey that included markers of acculturation and educational attainment, only educational attainment was found to significantly predict HIV/AIDS knowledge (Miller et al. 2002). This suggests that educational disparities may account for much of the difference between Latinos and whites in HIV knowledge.

In addition to having poorer access to care and less accurate knowledge of HIV, Latinos report more sexual risk behavior than whites. Several studies have found increased sexual risk behavior in heterosexually active Latinos (Marin et al. 1993; Grinstead et al. 1997). Latino men report high rates of heterosexual anal intercourse (Billy et al. 1993; Sabogal et al. 1993) and are more likely than Latinas to report having multiple sexual partners (Marin et al. 1993). Researchers have examined possible sociocultural explanations for these higher levels of risk behavior. For example, Díaz and colleagues, using a combination of qualitative and quantitative methods, postulated a theory of health and well-being that is specific to Latino MSM (e.g., Díaz 1998; Díaz et al. 2001; Díaz and Ayala 1999; Díaz et al. 2004). Research demonstrates that various forms of social oppression (i.e., homonegativity, racism, poverty) experienced by Latino MSM are associated with sexual risk behavior and indices of mental health. Likewise, Landrine et al. have advocated for the increased measurement of perceived ethnic discrimination (2006) and acculturation (2004) as stressors that impact health behavior and decisions. This work identifies social oppression experienced by Latinos as a risk factor for HIV.

It is important to recognize the role of normative cultural values in understanding behavior and communication. Latino cultural norms can result in mismatches between patient expectations and health-care interactions that are based on the values of the white majority. Comprehensive reviews are available on the importance of these factors for cognitive-behavioral therapy (Organista 2006), substance abuse counseling (Gloria and Perogy 1996), and the patient-physician relationship (Flores 2000). We briefly describe some examples of the relevance of these values in the context of HIV/AIDS.

Latino culture emphasizes personalismo, which values warmth and the personal dimension in both business and personal relationships. Such relationships entail a ‘formal friendliness’ that may involve the use of physical contact, including handshakes or placing a hand on the other’s shoulder (Flores 2000). Relatedly, simpatia, or “kindness,” emphasizes politeness and pleasantness in the face of stress (Triandis et al. 1984). When these expectations are met with the relatively neutral attitude of many doctors, there is a potential for dissatisfaction with care (Flores 2000). Respeto, or respect, is another important value involving an expectation of appropriate deferential behavior on the basis of a position of authority, age, gender, social position, and economic status, which could result in a patient not asking questions or expressing dissatisfaction with care (Flores 2000). Personalismo, simpatia, and respeto are each tied to the importance of developing confianza, or trust, in all important relationships (Organista 2006). Failing to meet cultural expectations for communication style, level of warmth, and respect in the patient-provider interaction may result in distrust of the provider. It is important to note that quality of the patient-provider relationship has been associated with treatment adherence in HIV-positive Latinos (van Servellen and Lombardi 2005; Murphy et al. 2003).

Latinos also tend to emphasize a collective or familial orientation, which has been termed familialismo. The importance of familial obligations, relying on family for support, and using family members as referents for decisions are concepts related to familialismo (Sabogal et al. 1987). Social support from family as well as ‘needing someone to live for’ have each been related to HIV treatment adherence in Latinos (van Servellen and Lombardi 2005; Murphy et al. 2003). Finally, fatalismo, or the belief that one can do little to alter one’s fate, is common among Latinos; religious faith bears heavily on this norm. Fatalismo can result in use of avoidance and delays in seeking care (Flores 2000). These understudied norms may have important impacts on health behavior, treatment seeking, and patient-provider communication for HIV-positive Latinos.

Drug Abuse and Mental Health Problems in U.S. Latinos

Drug abuse and mental health problems are closely related with each other and are associated with increased HIV risk (e.g., Latkin et al. 2007). Substance abuse and/or mental illness may complicate the treatment needs of HIV-positive patients and may have a negative impact on outcomes (Walkup et al. 2008). Although rates of mental illness appear to be roughly similar between whites and Latinos, there are significant disparities in access to treatment. Compounding the problem is the gross under-representation of Latinos in drug abuse and mental health treatment research.

Latino Substance Use

Sharing needles when using injection drugs is a source of HIV risk for all ethnic groups in the U.S. However, reflecting another disparity, Latinos who inject drugs are at least one and a half times more likely than white IDUs to be infected with HIV (CDC 1992). Data collected between 1996 and 1998 suggest that Latinos who inject drugs may be as much as four times as likely as whites to progress to AIDS, with the number of Latinos living with drug-related AIDS more than doubling between 1995 and 2000 (Day 2003).

The National Survey on Drug Use and Health (NSDUH) provides a cross-sectional snapshot of drug use in the U.S. and shows comparable prevalence in white and Latino populations (NSDUH 2006). Data from this survey show that combined rates of substance dependence and abuse were equivalent among Latinos (9.8% reporting abuse or dependence) and whites (9.6% reporting abuse or dependence). However, among individuals with HIV, according to the HIV Cost and Services Utilization Study (HCSUS), 13% reported substance dependence, with an additional 12% reporting marijuana use and one quarter (26%) reporting other illicit drug use. Only 50% of HIV-positive individuals surveyed reported abstinence from illicit drug use. Thus, drug abuse and dependence are equivalent among Latinos and whites and are quite high among HIV-positive individuals, regardless of ethnicity.

Latino Mental Health

Mental Health Problems in U.S. Latinos

A minority supplement to the Surgeon General’s Report on Mental Health (1999) notes that the prevalence rates of mental disorders in U.S. Latinos are similar to those for whites (US Department of Health and Human Services [DHHS] 2001). The National Epidemiologic Survey on Alcoholism and Related Conditions (Hasin et al. 2005) found a 4.27% 12-month prevalence of major depressive disorder among Latinos, compared with 5.53% for whites. The National Survey on Drug Use and Health (2006) found that rates of serious psychological distress were equivalent between adult Latinos (10.8%) and whites (10.3%). Past year prevalence of major depressive episode was also similar for whites (8.5%) and Latinos (8%).

Latinos infected with HIV, according the HCSUS, do not have higher rates of psychiatric disorders than HIV-positive whites. However, rates of psychiatric distress among HIV-positive individuals are quite high. Nearly half (48%) of the sample screened positive for one or more psychiatric disorders. More than one-third screened positive for major depression, and more than one-quarter experienced symptoms of dysthymia during the previous 12 months. These HCSUS data were compared to data from the National Household Survey on Drug Abuse (NHSDA), which interviewed a nationally representative sample of 22,181 people in 1994 using the same screeners as the HCSUS. The proportion of HCSUS participants screening positive for major depression was nearly 5-times greater than in the NHSDA, and the proportion for Generalized Anxiety Disorder was nearly 8-times higher. Rates of positive screens for panic attacks were 4-times greater. These data suggest a strikingly high prevalence of mental health problems among HIV-positive individuals.

More recent data from a prospective cohort study with a large multiethnic sample of HIV-positive women show that 73% of the women met criteria for mild depressive symptoms and 47% met criteria for moderate-to-severe depressive symptoms (Cook et al. 2002). A significantly higher percentage of Latinas were represented in each of these cutoff categories than white women. Latino MSM living in U.S. urban centers also exhibit a relatively high frequency of symptoms of psychological distress and disorder and these psychological symptoms are predicted by experiences of social discrimination (i.e., homophobia, racism, and poverty; Diaz et al. 2001). The potential direct and indirect effects of perceived discrimination on the mental and physical health of individuals in the African-American community have been described by Clark et al. (1999) and may generalize to other ethnic minority groups. The chronic social discrimination experienced by U.S. Latinos may not only be associated with the increased risk of contracting HIV, but it may also adversely affect the health trajectory of persons living with the virus. Thus, understanding the forms of discrimination perceived by Latinos in the U.S. and their possible direct and indirect effects on health may be essential to the development of psychosocial HIV prevention and post-infection intervention programs.

Access to Mental Health Services

As noted previously, Latinos are the least likely of all the ethnic groups living in the U.S. to have health insurance and 37% of Latinos are estimated to have neither public nor private health insurance, which is roughly double the rate of uninsured whites (DHHS 2001). Recent data from the first 1500 outpatients with major depression who enrolled in the Sequenced Treatment Alternatives to Relieve Depression trial illustrate the consequences of insurance status on mental health. Patients with public or no insurance who were more likely to be from racial or ethnic minority groups, were also more likely to have comorbid medical and psychiatric conditions, have a greater severity of depression, and to suffer greater functional impairment (Lesser et al. 2005). Latinos are less likely than white Americans to receive necessary mental health care and when they do receive care, they are more likely to receive it from primary health providers than from mental health specialists (DHHS 2001). Furthermore, a study of insured U.S. federal employees found that white employees were 1.7 times more likely to visit an outpatient mental health provider and make 2.6 more mental health visits per year compared to both African Americans and Latinos (Padgett et al. 1994). Thus, even insured Latinos are less likely to receive mental health care than whites.

More recent data from a follow-up to the nationally representative Healthcare for Communities survey also show consistent ethnic differences in mental health care (Wells et al. 2001). Latinos reported less access to care, poorer quality of care, and greater unmet needs for alcoholism, drug abuse, and mental health treatment in comparison to whites. Latinos reported more delays in receiving care, lower satisfaction with care, and lower rates of active treatment among those in need. The percentage of those in need who were receiving active treatment was nearly 50% less for Latinos than for whites. Data from the 1980s Medical Outcomes Study suggest that providers were less likely to detect diagnosable mental disorders in Latino patients than in whites (Borowsky et al. 2000). However, more recent data suggest that although providers may recognize depression and recommend treatment at equal levels with white patients, Latinos are still less likely to obtain appropriate care (Miranda and Cooper 2004).

Representation in Mental Health Research

This disparity in access to treatment is, unfortunately, mirrored in treatment research. The Surgeon General’s Report notes a striking absence of ethnic/racial minorities in psychiatric clinical trials conducted between 1986 and 1994. Often, ethnic group comparisons were not made when examining outcomes of these trials and most trials lacked an adequate number of minorities to conduct such comparisons. Only 7% of those participating in depression trials conducted during this period were identified as ethnic minorities (DHHS 2001). While this state of affairs has been improving, there is still a paucity of data regarding the treatment of mental health problems in ethnic/racial minority patients (Schraufnagel et al. 2006).

Opportunities for Psychosocial Intervention Research with HIV-Positive Latinos

Effective interventions are urgently needed to improve HIV outcomes for HIV-positive Latinos, who have been under-represented in HIV intervention research to date (Amaro et al. 2001). A recent review located 20 trials of primary HIV risk reduction in U.S. Latinos and found evidence for an overall effect for increasing condom use and decreasing the number of sexual partners, the odds of acquiring sexually transmitted infections, injection drug use, and sharing of drug paraphernalia. However, few studies included HIV-infected participants (Herbst et al. 2007).

Below, we focus on research suggesting that culturally sensitive interventions addressing (1) HIV treatment adherence, (2) psychological distress, particularly depression, (3) substance abuse, (4) healthy literacy, and (5) access to care are needed to improve outcomes for HIV-positive Latinos. We conducted a systematic search of PsycINFO and MEDLINE for psychosocial intervention trials conducted with HIV-positive Latino samples published by the end of 2007. The search combining HIV/AIDS with Latino/Hispanic and various terms to identify psychosocial interventions returned 247 abstracts. Because only one trial conducted with an HIV-positive Latino sample was found (van Servellen et al. 2005), we also review an unpublished randomized trial conducted by a member of our group (R. Dúran) and include examples of interventions conducted with multi-ethnic HIV-positive samples that reported at least a representative proportion of Latino participants. We then provide examples of intervention work conducted with non-HIV-positive Latino populations, as these studies could serve as important models for further intervention work with HIV-positive Latinos. These intervention studies are presented in Table 1. We do not include substance abuse interventions reviewed by Herbst et al. (2007) due to space constraints.

Table 1 Intervention studies targeting depression, medication adherence, health literacy, or access to healthcare in Latinos
Full size table

HIV Treatment Adherence and Latinos

Adherence to HIV medications is critical for treatment success and numerous studies have established that poor adherence is associated with poor outcomes, as measured by viral load or CD4 cell count (e.g., Bangsberg et al. 2000; Catz et al. 2000; Paterson et al. 2000). These data reveal the necessity of high levels of adherence to maximize treatment outcomes. A number of intervention studies have been conducted to improve medication adherence in HIV-positive individuals (Simoni et al. 2006). However, with few notable exceptions (see below), ethnic minorities have been under-represented in studies of psychosocial factors in HIV-treatment adherence.

Possibly due to under-representation of ethnic minorities in antiretroviral adherence research, ethnicity has not been consistently associated with nonadherence to HIV treatment. This lack of a consistent association, however, does not rule out the possibility that different factors may be salient for Latinos than for other groups in influencing adherence. One published study focused on HIV-positive Latinos using a mixture of qualitative and quantitative approaches to data analysis (Murphy et al. 2003). This study used a sample of 81 HIV-positive monolingual and bilingual Spanish-speaking patients to identify important barriers and successful strategies for HAART adherence. The most important reasons for nonadherence cited by participants were feeling depressed or overwhelmed (21%), simply forgetting (19%), and sleeping through a dose (17%). Qualitative analyses revealed a number of important barriers and facilitators to adherence. Patient characteristics that were found to play a role in adherence included: (1) patients’ level of social support, (2) beliefs about the efficacy of antiretroviral medications, (3) use of alcohol and/or drugs, (4) willingness to take medications in front of others, (5) negative mood states, and (6) daily schedules. The quality of the patient’s relationship with the health care professional was also found to affect adherence. A number of patients expressed feeling uncomfortable discussing adherence issues with their providers and worried about provider reactions. Participants also stated that attending a health care facility that did not have a physician who spoke Spanish was detrimental to their adherence. Additionally, side effects and negative characteristics of the dosing regimen were reported to have a negative influence on adherence.

The findings of this study can be compared with a similar study by Murphy et al. (2000) conducted with English-speaking patients. One obvious difference was the role of language barriers in nonadherence for Spanish-speaking patients. Another key difference is that, although both groups reported that social support had a positive influence on adherence, Latinos were likely to report extremely strong feelings of needing ‘someone to live for.’ Consistent with these results, a quantitative study of 85 Spanish-speaking HIV-positive Latinos showed that emotional or informational support from friends and/or family was associated with aspects of improved medication adherence (van Servellen and Lombardi 2005). This may be related to the importance of familialismo.

One research group recently developed and tested an intervention to increase HAART adherence specifically for low-income, Spanish-speaking, HIV-positive Latinos (van Servellen et al. 2005; Table 1). These investigators reasoned that for an adherence intervention to be successful, it would need to address the needs of a population that is less familiar with the health care system, less adept in navigating and accessing the services they need, and less likely to be literate (van Servellen et al. 2003). Although preliminary, results suggest that a culturally adapted behavioral treatment can be successful in improving health literacy and patient-provider relationships in HIV-positive Latinos.

Depression and HIV-positive Latinos

Studies have consistently found significant relationships between depressive symptoms and HIV disease progression, (e.g., Leserman et al. 2000; Patterson et al. 1996; Ickovics et al. 2001). For example, Cook et al. (2004) examined longitudinal data from an ethnically-diverse cohort of 1716 HIV-positive women (23% Latina) and found that, even after controlling for antiretroviral therapy use, medication adherence, substance abuse, clinical indicators, and demographic factors, women who were chronically depressed were more than twice as likely to die from AIDS than those who were not. Additionally, although results from studies examining the relationship between depression and sexual risk are mixed (Crepaz and Marks 2001), lowered self worth and increased levels of depression have each been associated with greater frequency of high-risk sexual behavior among Latino MSM and Latino HIV-positive MSM, respectively (Dolezal et al. 2000; Poppen et al. 2004).

Research further demonstrates that depression is associated with poor adherence to medical regimens (DiMatteo et al. 2000). A number of studies have supported the relationship between depressive symptoms and treatment nonadherence among heterogeneous samples of HIV-positive men (Catz et al. 2000; Gonzalez et al. 2004; Gordillo et al. 1999; Hirschhorn et al. 1998; Kleeberger et al. 2001; Paterson et al. 2000; Safren et al. 2001; Simoni et al. 2002). Symptoms of depression make the already difficult task of adhering to HIV treatment regimens much more difficult for patients. As Latinos experience depression at the same rates as whites and are less likely to obtain appropriate care, they may be at higher risk for depression-related nonadherence.

The consistent association between depression and worse HIV outcomes suggests that interventions that are successful in treating depression in this population could also have positive effects on health. There is a small body of evidence for the efficacy of depression treatment interventions in Latino populations. Several studies, as shown in Table 1, examined depression treatment outcomes among various Latino populations, including Puerto Rican mothers (Comas-Diaz 1981), Mexican women (Lara et al. 2003), Puerto Rican adolescents (Rosselló and Bernal 1999), and Spanish speaking IDUs (González et al. 1993) with favorable results. Likewise, several successful depression treatment studies with minority samples have had a significant number of Latino participants (Miranda et al. 2003, 2004; Table 1). For example, Miranda et al. (2003) found that medication or cognitive-behavioral therapy reduced depressive symptoms significantly more than a community mental health referral. Another multi-ethnic study found that collaborative care significantly improved rates of depression care, lowered depression severity, and lowered health-related functional impairment for patients, regardless of ethnicity (Areán et al. 2005; Table 1). A final multi-ethnic study of HIV-positive patients with depressive symptoms found that interpersonal psychotherapy or supportive psychotherapy with imipramine resulted in significantly greater improvement in depressive symptoms than other treatment conditions (Markowitz et al. 2000; Table 1).

Finally, a preliminary body of research suggests that cognitive-behavioral stress management (CBSM; Antoni et al. 2007), which integrates cognitive-behavioral therapy skills with relaxation and stress management strategies, can be effective in HIV-positive samples that include Latinos. However, future studies must include sufficient numbers of minority patients to facilitate ethnic comparisons of treatment outcomes. Recent data from group-based CBSM trials in relatively diverse samples of HIV-positive men have demonstrated efficacy for mood and other health-related outcomes (Carrico et al. 2005; Carrico et al. 2006; Antoni et al. 2006; See Table 1). An adaptation of CBSM for HIV-positive multi-ethnic women has also shown success with mood outcomes but results for quality of life and adherence have been mixed (Jones et al. 2003; LaPerriere et al. 2005; Lechner et al. 2003; See Table 1). While these studies did include Latinos, all were English-speaking and no analyses were reported to determine whether ethnicity moderated intervention effects. However, promising preliminary data from a full cultural adaptation of the CBSM intervention for HIV-positive monolingual Latino MSM suggest that CBSM is effective in improving several psychosocial and health-related outcomes (Durán et al. 2006; Table 1).

Substance Abuse and HIV-positive Latinos

Drug use and abuse is common among HIV-positive patients and is associated with worse outcomes. HCSUS data show that approximately 50% of HIV-positive patients reported using drugs in the prior 12 months and studies demonstrate that substance use negatively impacts HIV medication adherence (Chesney et al. 2000; Gebo et al. 2003; Golin et al. 2002; Haubrich et al. 1999; Lucas et al. 2002; Power et al. 2003; Wilson et al. 2002; Lucas et al. 2001). Studies have also identified a similar relationship between injection drug use and nonadherence (Aloisi et al. 2002; Gordillo et al. 1999; Stein et al. 2000; Gifford et al. 2000; Holzemer et al. 1999). See Uldall et al. (2004) for a detailed review of this literature.

Drug use has also been associated with high-risk sexual behavior (Stall and Purcell 2000; Kalichman 2000). Studies of MSM have shown that use of alcohol or other drugs prior to engaging in sexual activity is associated with increased frequency of unprotected receptive and insertive anal intercourse (e.g., Kelly and Kalichman 1998). This relationship has also been documented in samples of Latino MSM (Diaz et al. 1996; Dolezal et al. 2000) as well as in HIV-positive Latino MSM (Poppen et al. 2004).

Recent reviews note a general lack of both behavioral and pharmacotherapy treatment research that specifically targets Latino drug and/or alcohol abusers and note that Latinos are currently underserved by substance abuse treatment services (Alegria et al. 2006; Amaro et al. 2006). These authors recommend the reanalysis of existing treatment datasets to examine potential differences in effects between Latino and other groups, consistent with guidelines from the National Institutes of Health relating to the inclusion of minorities in research (National Institutes of Health 2000).

Herbst et al. (2007) review six HIV prevention studies (Castro and Tafoya-Barraza 1997; Colón et al. 1993; Dushay et al. 2001; Nyamathi et al. 1994; Robles et al. 2004; Schilling et al. 1991) targeting drug risk behaviors in Latinos and note an overall significant reduction in injection drug use and sharing of injection paraphernalia, but not for needle sharing. Still, the ability of these interventions to improve drug use outcomes is promising. Only two of the six studies reported efforts to culturally adapt the intervention for Latinos. Four studies included some HIV-positive participants but the maximum was 46% (see Herbst et al. 2007 for study details). In general, there have been few behavioral intervention trials for HIV-positive individuals and the results of these studies have been mixed (Strathdee and Patterson 2006). Interventions that decrease drug abuse among HIV-positive Latinos are needed, as they may be successful in reducing risk, improving adherence, and improving health outcomes.

Health Literacy Among HIV-positive Latinos

Healthy People 2010 (DHHS 2002) defines health literacy as “The degree to which individuals have the capacity to obtain, process, and understand basic health information and services needed to make appropriate health decisions.” The American Medical Association Council on Scientific Affairs concluded that low health literacy is associated with poor health outcomes, raising serious concerns about the ability of Americans to function adequately in health care settings (American Medical Association 1999). Studies of Latinos have shown that many have inadequate health literacy, even in Spanish. For example, a four-city study of 3260 new Medicare enrollees aged 65 or older found that 53.9% of Spanish-speaking respondents had inadequate or marginal health literacy in comparison to 33.9% of English-speaking respondents (Gazmararian et al. 1999). Nation-wide, approximately 56% of Latinos are classified as functionally illiterate (Kirsch et al. 1993).

As linguistic minorities in the U.S., Latinos, especially those who are monolingual, face additional challenges to reaching optimal HAART adherence. These challenges may include difficulties in obtaining and understanding accurate information about HIV and its treatment in an appropriate level of Spanish. In addition, language may be a proxy (albeit insufficient) for acculturation. Under-acculturated individuals may hold certain cultural values (e.g., respeto, fatalismo, machismo), beliefs about the cause of their condition (e.g., internalized and experienced homonegativity and racism), and inaccurate information about medications (e.g., they are poisonous because they make me sick) that could work against obtaining the best possible medical care. Moreover, English language skills in the U.S. are directly proportional to income, and it is an unfortunate fact that in the U.S., the quality of medical care one receives is often a function of the quality of health insurance one can afford. Health education for cardiovascular disease has been successfully incorporated into English-as-a-second-language classes for low-literacy Latinos and may be a useful model for HIV research (Elder et al. 2000; Table 1).

Health literacy is repeatedly associated with HIV-treatment adherence (Fogarty et al. 2002). In a study of HIV-positive men and women, a significant correlation was found between years of education and health literacy, and between literacy and adherence (Kalichman et al. 1999). An additional study found that after adjusting for years of education, lower health literacy was associated with poorer knowledge of one’s HIV-related health status, poorer AIDS-related disease and treatment knowledge, and more negative health care perceptions and experiences (Kalichman and Rompa 2000). One study of low-income HIV-positive Latino men and women (van Servellen et al. 2003) found that years of education was associated with understanding HIV terms, as well as accurately reading and understanding instructions on prescription bottles. As noted above, this group piloted an intervention that sought to improve HIV treatment adherence and health literacy among HIV-positive Spanish-speaking Latinos (see Table 1).

Access to Care Among HIV-positive Latinos

Access to medical care for HIV-positive Latinos varies based on a variety of factors, according to data from the HCSUS (Morales et al. 2004). The most vulnerable HIV-positive Latinos include the uninsured, those seeking HIV care in the Southern U.S., and men, though previous studies have found that women have worse access to care (Shapiro et al. 1999). These disparities necessitate change on many levels, from individual to institutional to policy and law. At the level of psychosocial intervention and access to care, however, effective interventions connecting Latinos to health care may take place inside or outside the medical model (Wasserman et al. 2007). One study showed success in using outreach staff to link out-of-treatment HIV-infected minorities (36% Latino) to HIV care (Molitor et al. 2005; See Table 1). Additional creative strategies have been used to connect other patient populations of Latinos to care and work has also been done in HIV-prevention outreach. One strategy, developed in the migrant health community, is the use of promotoras—community health counselors who are linked to both the Latino community and the mainstream medical community—and act as mediators to care. For example, this community-based model has been tested in church-based communities for breast cancer screening (Sauaia et al. 2007; Table 1), Mexican-American communities for diabetes control (Lujan et al. 2007; Ingram et al. 2007; Table 1), and in an urban Latino community to increase HIV knowledge (Martin et al. 2005; Table 1). Such community-based approaches may be successful in reaching Latinos who are otherwise unlikely to access conventional services. For example, one group has examined the feasibility of reaching Latino men at risk for HIV through Latino soccer leagues in North Carolina (Rhodes et al. 2006). Overall, these inventive strategies and the use of culturally sensitive case management, community health workers, and patient navigators (Vargas and Cunningham 2006), hold promise to improve HIV-positive Latinos’ access to care.

General Conclusion

The literature reviewed suggests that HIV-positive Latinos are a rapidly growing group, are more severely impacted by HIV than whites, confront unique sociocultural challenges in coping with this disease, and are currently under-supported by available science and clinical service. Epidemiological data demonstrate the disproportionate impact of HIV/AIDS on Latinos, as well as less access to appropriate care. Data on drug abuse patterns and mental health characteristics of HIV-positive Latinos suggest that both problems are prevalent and may negatively impact HIV health outcomes and health behaviors, including risky sex and HIV treatment adherence. There is currently an unmet need for culturally sensitive intervention research to address treatment adherence, drug abuse, and mental health factors among HIV-positive Latinos, and preliminary evidence suggests potential efficacy for such interventions. Research also suggests that health literacy and access to health care may be particularly important targets for intervention in this group.

This review, along with another comprehensive review of psychosocial interventions for ethnic minorities (Miranda et al. 2005), provides evidence for the applicability of empirically supported psychosocial interventions in Latino populations. We have identified health behavior, mental health, and substance abuse targets of such interventions that could be especially important in the context of HIV/AIDS. Greater efforts should be made to enroll Latinos in behavioral intervention trials and to examine outcomes by ethnicity. Additionally, variables that might explain differences in outcome (reviewed above and including acculturation, literacy, discrimination, cultural norms, etc.) should be routinely assessed. Furthermore, more work is needed to operationalize the process of culturally adapting mental health interventions for use with Latinos. At this stage, the available literature provides little guidance beyond suggesting that culturally adapted interventions can be effective in this population, but we do not know whether they are more effective than treatments based on the majority culture. The issues contributing to the disproportionate impact of HIV/AIDS on U.S. Latinos are complex. Sex, sexual orientation, mode of infection, language proficiency/literacy, level of acculturation, nation of origin, and experiences with racism, homophobia, and other forms of marginalization are each important factors to be considered in work with this population. Comprehensive, methodologically rigorous studies that examine these demographic, socioeconomic, and sociocultural variables and their relationships to the HIV/AIDS disparity are needed.

Although we have considered Latinos as a group that differs from the white majority population in important respects, we have also noted that there is heterogeneity within this group. Therefore, generalizations are likely to result in distortions. It is clear that interventions targeted at the individual can only be partially successful in addressing the disparities in HIV/AIDS between whites and Latinos. Systems-level interventions are needed to address the problems of poverty, lower education, and less access to appropriate medical and mental health care in the U.S. Latino population. As the Latino population is growing rapidly in the U.S. and, unfortunately, continues to bear a heavy burden of HIV/AIDS, such interventions could have a broad and important public health impact.