CHAPTER 6

Mental Health Care for Hispanic Americans

The Need for Mental Health Care

Historical and Sociocultural Factors That Relate to Mental Health

Historical and sociocultural factors suggest that, as a group, Latinos are in great need of mental health services. Latinos, on average, have relatively low educational and economic status. In addition, historical and social subgroup differences create differential needs within Latino groups. Central Americans may be in particular need of mental health services given the trauma experienced in their home countries. Puerto Rican and Mexican American children and adults may be at a higher risk than Cuban Americans for mental health problems, given their lower educational and economic resources. Recent immigrants of all backgrounds, who are adapting to the United States, are likely to experience a different set of stressors than long-term Hispanic residents.

Key Issues for Understanding the Research

Much of our current understanding of the mental health status of Latinos, particularly among adult populations, is derived from epidemiological studies of prevalence rates of mental disorders, diagnostic entities established by the Diagnostic and Statistical Manual of Mental Disorders (DSM; American Psychiatric Association, 1994). The advantage of focusing on rates of disorders is that such findings can be compared with and contrasted to findings from studies in other domains (e.g., clinical studies) using the same diagnostic criteria. In addition, diagnostic entities are now often associated with specific pharmacological and psychosocial treatments.

Although there are several advantages to examining DSM-based clinical entities, there are at least three disadvantages. One limitation is that individuals may experience considerable distress—a level of distress that disrupts their daily functioning—but the symptoms associated with the distress fall short of a given diagnostic threshold. Thus, if only disorder criteria are used, some individuals’ need for mental health care may not be recognized. A second disadvantage is that the current definitions of the diagnostic entities have little flexibility to take into account culturally patterned forms of distress and disorder. As a result, disorders in need of treatment may not be recognized or may be mislabeled. A third limitation is that most of the epidemiological studies using the disorder-based definitions are conducted in community household surveys. They fail to include nonhousehold members, such as persons without homes or those who reside in institutions. Because of these limitations, it is important to broaden the review of research on mental health needs to include not only studies that report on disorders, but also studies that report on symptoms, symptom clusters, culturally patterned expressions of distress and disorder, and high-need populations not usually included in household-based surveys.

Mental Disorders

Adults

As noted in previous chapters, researchers have conducted two large-scale studies to identify the rates of psychiatric disorders among adults in the United States. The first, the Epidemiologic Catchment Area Study (ECA) (Robins & Regier, 1991), examined rates of psychiatric disorders in five communities (N = 19,182): New Haven, Baltimore, St Louis, Durham, and Los Angeles. Investigators at the Los Angeles site conducted inter-views in English and Spanish and oversampled Mexican Americans (N = 1,243), so that rates of psychiatric disorders in this subpopulation could be estimated (Karno et al., 1987). The second study, the National Comorbidity Study (NCS) (Kessler et al., 1994), examined psychiatric disorders in a representative sample of individuals living throughout the United States (N = 8,098), excluding Alaska and Hawaii. This survey included Hispanics (N = 719), but was conducted only in English; thus, Spanish-speaking Hispanics were not represented (Ortega et al., 2000).

The ECA study found that Mexican Americans and white Americans had very similar rates of psychiatric disorders (Robins & Regier, 1991). However, when the Mexican American group was separated into two sub-groups, those born in Mexico and those born in the United States, it was found that those born in the United States had higher rates of depression and phobias than those born in Mexico (Burnam et al., 1987). The NCS found that relative to whites, Mexican Americans had fewer lifetime disorders overall and fewer anxiety and substance use disorders. Like the Los Angeles ECA findings, Mexican Americans born outside the United States were found to have lower prevalence rates of any lifetime disorders than Mexican Americans born in the United States. Relative to whites, the lifetime prevalence rates did not differ for Puerto Ricans, nor for “Other Hispanics.” However, the sample sizes of the latter two subgroups were quite small, thus limiting the statistical power to detect group differences (Ortega et al., 2000).

A third study examined rates of psychiatric disorders in a large sample of Mexican Americans residing in Fresno County, California (Vega et al., 1998). This study found that the lifetime rates of mental disorders among Mexican American immigrants born in Mexico were remarkably lower than the rates of mental disorders among Mexican Americans born in the United States. Overall, approximately 25 percent of the Mexican immigrants had some disorder (including both mental disorders and substance abuse), whereas 48 percent of the U.S.-born Mexican Americans had a disorder (Vega et al., 1998). Furthermore, the length of time that these Latinos had spent in the United States appeared to be an important factor in the development of mental disorders. Immigrants who had lived in the United States for at least 13 years had higher prevalence rates of disorders than those who had lived in the United States fewer than 13 years (Vega et al., 1998).

Figure 6-2 compares data for the lifetime prevalence of CIDI disorders among immigrant and U.S. born Mexican Americans in Fresno, California versus the general population rates from the National Comorbidity Survey.

It is interesting to note that the mental disorder prevalence rates of U.S.-born Mexican Americans closely resembled the rates among the general U.S. population. In contrast, the Mexican-born Fresno residents’ lower prevalence rates were similar to those found in a Mexico City study (e.g., for any affective disorder: Fresno, 8 %, Mexico City, 9 %) (Caraveo-Anduaga et al., 1999). Together, the results from the ECA, the NCS, and the Fresno studies suggest that Mexican-born Latinos have better mental health than do U.S.-born Mexican Americans and the national sample overall.

A similar pattern has been found in other sets of studies. One study examined the mental health of Mexicans and Mexican Americans who were seen in family practice settings in two towns equidistant from the Mexican border (Hoppe et al., 1991). This investigation found that 8 percent of the Mexican American participants had experienced a lifetime episode of depression, whereas only 4 percent of Mexican participants had. A group of earlier studies conducted in the mid-1980s also examined rates of depression in English- and Spanish-speaking Latinos, including Cuban Americans (N = 857) in Miami (Narrow et al., 1990); Mexican Americans (N = 3,118) in the Southwest (Moscicki et al.,1987); Puerto Ricans (N = 1,140) in New York City (Moscicki et al., 1987); and Puerto Ricans (N = 1,513) on the island (Canino et al., 1987). One of the most salient findings is that Puerto Ricans from the island had lower rates of lifetime depression (4.6 %) than those from New York City (9 %) (Canino et al., 1987; Moscicki et al., 1987).

The most striking finding from the set of adult epidemiological studies using diagnostic measures is that Mexican immigrants, Mexican immigrants who lived fewer than 13 years in the United States, or Puerto Ricans who resided on the island of Puerto Rico had lower prevalence rates of depression and other disorders than did Mexican Americans who were born in the United States, Mexican immigrants who lived in the United States 13 years or more, or Puerto Ricans who lived on the mainland. This consistent pattern of findings across independent investigators, different sites, and two Latino subgroups (Mexican Americans and Puerto Ricans) suggests that factors associated with living in the United States are related to an increased risk of mental disorders.

Some authors have interpreted these findings as suggesting that acculturation may lead to an increased risk of mental disorders (e.g., Vega et al., 1998; Escobar et al., 2000; Ortega et al., 2000). The limitation of this explanation is that none of the noted epidemiological studies directly tested whether acculturation and prevalence rates are indeed related. At best, place of birth and number of years living in the United States are proxy measures of acculturation. Moreover, acculturation is a complex process (LaFromboise et al., 1993); it is not clear what aspect or aspects of acculturation could be related to higher rates of disorders. Is it the changing cultural values and practices, the stressors associated with such changes, or negative encounters with American institutions (e.g., schools or employers) that underlie some of the different prevalence rates (Betancourt & Lopez, 1993)? Before acculturation can be accepted as an explanation for this observed pattern of findings, it is important that direct tests of specific acculturation processes be carried out and that alternative explanations for these findings be ruled out. Longitudinal research would be especially helpful in identifying the key predictors of Latinos’ mental health and mental illness.

Children and Youth

Most epidemiological studies of Latino children and adolescents have been conducted with symptom indices and problem behavior checklists, not diagnostic instruments. Efforts to study diagnostic entities among Latino children in community samples have been limited. In one study carried out in Puerto Rico, psychiatrists administered a standard diagnostic instrument, the Diagnostic Interview Schedule for Children (DISC), and found high rates of mental disorders (49 %) among Puerto Rican children who had previously been identified as having significant behavioral problems. However, the rate dropped to 18 percent when a diagnosis with some associated impairment was required (Bird et al., 1988). The importance of including impairment as a criterion for disorders in children was established in another recent study. Children living in Georgia, Connecticut, New York, and Puerto Rico were assessed to establish rates of mental disorders; the Puerto Rican children had rates comparable to the multiethnic sample from the U.S mainland (Shaffer et al., 1996). For all groups, rates of disorders dropped dramatically when impairment was required as part of the diagnosis.

An examination of studies of mental health problems reveals a generally consistent pattern: Latino youth experience a significant number of mental health problems, and in most cases, more problems than whites. Studies of child mental health problems typically used versions or portions of a popular screening instrument, the Childhood Behavior Checklist (CBCL, Achenbach & Edelbrock, 1983). Glover and colleagues (1999) found that Hispanic children in middle schools, specifically Mexican-origin youth from Texas, reported more anxiety-related problem behaviors than white students. In addition, Hispanic sixth- and seventh-graders from a Southwestern city reported more delinquency-type problem behaviors than white students (Vazsonyi & Flannery, 1997). Youth in Puerto Rico were also found to have a significantly higher total problem score (35% versus 20%) and prevalence rate of “cases” (36% versus 9 %) than a three-State sample comprised primarily of whites (Achenbach et al., 1990). A study of Hispanic 10- to 16-year-old boys in Dade County, Florida, was the only exception. This investigation did not reveal any differences in total problem behaviors when comparing Hispanic, non-Hispanic white, and African American boys (Vega et al., 1995).

Studies of depressive symptoms and disorders also revealed more distress among Hispanic children and adolescents, particularly among Mexican-origin youth. This was evident in a community study in Las Cruces, New Mexico (Roberts & Chen, 1995), as well as in a national study within the 48 coterminous States (Roberts & Sobhan, 1992). In both these investigations, Mexican American adolescents reported more depressive symptoms than did white adolescents. In a recent study that used a self-report measure of major depression among middle school (grades 6–8) students in Houston, Texas, Mexican American youth were found to have a significantly higher rate of depression than white youth (12 % versus 6 %) (Roberts et al., 1997). These findings held even when level of impairment and sociodemographic factors were taken into account.

A large-scale survey of primarily Mexican American adolescents in schools on both sides of the Texas-Mexico border revealed high rates of depressive symptoms, drug use, and suicide (Swanson et al., 1992). Like the adult epidemiological studies, this investigation found that living in the United States is related to elevated risk for mental health problems. More Texas youth (48 %) reported high rates of depressive symptoms than did Mexican youth (39 %). Also, youth residing in Texas reported more illicit drug use in the last 30 days (21 %) and more suicidal ideation (23 %) than youth residing in Mexico.

Together the data indicate that Latino children and adolescents are at significant risk for mental health problems, and in many cases at greater risk than white children. At this time, it is not clear why a differential rate of mental health problems exists for Latino and white children. Special attention should be directed to the study of Latino youth, as they may be both the most vulnerable and the most amenable to prevention and intervention.

Older Adults

Few studies have examined the mental health status of older Hispanic American adults. A study of 703 Los Angeles area Hispanics age 60 or above found over 26 percent had major depression or dysphoria. Depression was related to physical health; only 5.5 percent of those without physical health complications reported depression (Kemp et al., 1987). Similar findings associated chronic health conditions and disability with depressive symptoms in a sample of 2,823 older community-dwelling Mexican Americans (Black et al., 1998). The findings from in-home interviews of 2,723 Mexican Americans age 65 or older in Southwestern communities revealed a relationship between low blood pressure and higher levels of depressive symptomatology (Stroup-Benham et al., 2000). These data are somewhat difficult to interpret. Given the fact that somatic symptoms (e.g., difficulty sleeping and loss of appetite) are related to poor health, these studies could simply document that these somatic symptoms are elevated among older Hispanics who are ill. (See Box 6–1, an illustration of the importance of considering the physical problems of older Latinos. This is one of many cases that Celia Falicov, 1998, uses to illustrate how the social and cultural world of Latino families expresses itself in clinical domains.) On the other hand, presence of physical illness is also related to depression. Taken together, these findings indicate that older Hispanics who have health problems may be at risk for depression. Furthermore, a recent study suggests that the risk for Alzheimer’s disease may be higher among Hispanic Americans than among white Americans (Tang et al., 1998).

Box 6-1: Emotional or physical problems?

Mrs. Corrales (age 70)
Mrs. Corrales, a 70-year-old Puerto Rican, was referred to a mental health clinic by her local priest. Mrs. Corrales had no friends within the urban barrio. She had migrated from Puerto Rico eight years earlier to live with her two sons and her 45-year-old single and mildly developmentally impaired daughter. Two years before she came to the clinic, her sons had moved to a nearby city in search of better jobs. Mrs. Corrales remained behind with her daughter, who spoke no English and did not work. Among other questions, the Latin American therapist asked her if she was losing weight because she had lost her appetite, to which she quipped: "No, I've lost my teeth, not my appetite! That's what irks me!" Indeed, Mrs. Corrales had almost no teeth left in her mouth. Apparently, her conversations with the priest (an American who had learned to speak Spanish during a Latin American mission and was sensitive to the losses of migration) had centered on the emotional losses she had suffered with her sons' departure. The priest thought this was the cause of her "anxious depression." Though well meaning, he had failed to consider practical issues. Mrs. Corrales had no dental insurance, did not know any dentists, and had no financial resources.

Source: Falicov (1998), p. 255

Mental Health Problems

Symptoms

The early epidemiological studies of Latinos examined the number of symptoms, not the number of mental disorders, reported by groups of Hispanic Americans, and in some cases compared them to the number of symptoms reported by white Americans. Much of this research found that Latinos had higher rates of depression or distress than whites (Frerichs et al., 1981; Roberts, 1981; Vernon & Roberts, 1982; Vega et al., 1984). In a large-scale study of Hispanics, Cuban Americans (Narrow et al., 1990) and Mexican Americans (Moscicki et al., 1989) were found to have lower rates of depressive symptoms than Puerto Ricans from the New York City metropolitan area (Moscicki et al., 1987; Potter et al., 1995). In another line of inquiry, Latina mothers who have children with mental retardation were found to report high levels of depressive symptomatology (Blacher et al., 1997a, 1997b).

It is important to note that measures of symptoms may reflect actual disorders that may not be measured in a given study, as well as general distress associated with social stressors but not necessarily associated with disorders. Two studies provide evidence that depressive symptom indices used with Latinos tend to measure distress more than disorder. In one study, rates of depressive symptoms were found to be similar among poor Puerto Ricans living in New York City and in Puerto Rico (Vera et al., 1991), even though earlier analyses indicated different rates of major depression for the two samples (Canino et al., 1987; Moscicki et al., 1987). In the second study, symptoms of depression were less related to diagnosis of depression for those Hispanics who were economically disadvantaged than for those Hispanics more socially advantaged (Cho et al., 1993). If an index of depressive symptoms were an indicator of both general distress and disorder, then that index would have been related to a diagnosis of depression for both economically advantaged and disadvantaged samples. An under-standing of the interrelation of psychological distress, specific mental disorders, and social conditions would help shed light on how distress and disorder are moderated by social factors. (See Box 6–2 as an example of how the social world relates to family mental health problems.)

Box 6-2: Rebellious teenager and father's mal trato

Javier (age 16)
Javier Reyes Balan, a 16-year-old boy, was referred by his school for persistent truancy. Nine years ago, his mother, father, and four younger siblings moved from Michoacan, Mexico, to San Diego, California, to better their economic situation. Javier was bilingual and served as the family interpreter in their dealings with outside institutions. He preferred to speak English and was clearly more savvy about American values and ways than his parents.

Mr. Reyes began the session by complaining bitterly about Javier's unruly behavior, lack of cooperation with his mother, and lack of respect toward his parents. Mrs. Reyes appeared to agree with her husband's view of Javier, although she protested that she didn't need much help around the house.

An inquiry about Mr. Reyes's occupation revealed that he had hoped to start his own small business as a car mechanic after moving from Mexico. He had not succeeded and was supporting the family precariously with occasional small jobs. He was proud of his competence and honesty as an automobile mechanic. But now he refused to work in a company under an Anglo-American foreman who would subject him to mal trato. In his view, "they [Americans] don't respect us Mexicans, and when you turn around they exploit you." The father's position in the family appeared to be debilitated by his unemployment.

Source: Falicov (1998), pp. 128-129.

Somatization

The expression of distress through somatic symptoms has been observed in many groups, including Latinos (Escobar et al., 1987). Early research, influenced by psychodynamic theory, suggested that the expression of psychic distress via bodily complaints reflected limited psychological development. Current perspectives, however, accept somatic and psychological forms of expressing distress as equally valid. The two modes of expression are thought to mirror the sociocultural context; they do not necessarily reflect a lack of insight or psychological sophistication. The critical questions today concern how social and cultural processes shape the expression of distress that emphasizes the soma, the psyche, or both (Kirmayer & Young, 1998).

Some research has examined the extent to which Latinos express physical symptoms, particularly in comparison to whites. Many of these studies have used symptom indices derived from the diagnostic interview used in the ECA studies. According to these studies, Mexican American women, particularly those over age 40, are more likely to report somatic symptoms; however, no differences were found between Mexican American and white men (Escobar et al., 1987). In an additional study, Puerto Rican men and women had higher rates of somatic symptoms than Mexican American and non-Hispanic men and women (Escobar et al., 1989).

A group of primary care patients that included Central American immigrants, Mexican immigrants, U.S.-born Mexican Americans, and whites were assessed for psychiatric disorders and somatization. After controlling for education and income differences, the immigrants reported fewer psychiatric disorders but higher rates of somatic symptoms when compared with the U.S.-born sample (Escobar et al., 2000). However, a more recent study questions the validity of those findings (Villasenor & Waitzkin, 1999), arguing that differences in use of health care services, different cultural under-standings of the questions, and differences in socioeconomic status lead to spurious reports of somatic symptoms. For example, symptoms could have been considered “medically unexplained” because Latinos failed to receive adequate medical care and did not receive a diagnosis from a physician. Because high levels of somatic symptoms are related to disability (Escobar et al., 1987), research in this area is most important. Of particular significance are service factors (accessibility to care) and cultural factors (the meaning of physical and mental health) as they relate to somatization and distress.

Culture-Bound Syndromes

DSM-IV recognizes the existence of culturally related syndromes, referred to in the appendix of DSM as culture-bound syndromes. Relevant examples of these syndromes for Latinos are susto (fright), nervios (nerves), and mal de ojo (evil eye). One expression of distress that is most commonly associated with Caribbean Latinos but has been recognized in other Latinos as well is ataques de nervios (Guarnaccia et al., 1989). Symptoms of an ataque de nervios include screaming uncontrollably, crying, trembling, and verbal or physical aggression. Dissociative experiences, seizure-like or fainting episodes, and suicidal gestures are also prominent in some ataques. In one study carried out in Puerto Rico, researchers found that 14 percent of the population reported having had ataques (Guarnaccia et al., 1993). Furthermore, in detailed interviews of 121 individuals living in Puerto Rico (78 of whom had had an ataque), experiencing these symptoms was related to major life problems and subsequent psychological suffering (Guarnaccia et al., 1996). Clinical and ethnographic studies of individuals living in Boston and New York City also report observations of ataques, which in some instances required treatment (Guarnaccia et al., 1989; Liebowitz et al., 1994).

There is value in identifying specific culture-bound syndromes such as ataques de nervios because it is critical to recognize the existence of conceptions of distress and illness outside traditional psychiatric classification systems. These are often referred to as popular, lay, or common sense conceptions of illness or illness behavior (Koss-Chioino & Canive, 1993). Some of these popular conceptions may have what appear to be definable boundaries, while others are more fluid and cut across a wide range of symptom clusters. For example, many people of Mexican origin apply the more general concept of nervios to distress that is not associated with DSM disorders, as well as to distress that is associated with anxiety disorders, depressive disorders (Salgado de Snyder et al., 2000), and schizophrenia (Jenkins, 1988). Though it is valuable for researchers and clinicians alike to learn about specific culture-bound syndromes, it is more important that they assess variable local representations of illness and distress. The latter approach casts a wider net around understanding the role of culture in illness and distress.

In the following quote, Koss-Chioino (1992) points out that a given presenting problem can have multiple levels of interpretation: the mental health view, the folk healing view (in this case, spiritist), and the patient’s

The same woman, during one episode of illness, may experience “depression” in terms of hallucinations, poor or excessive appetite, memory problems, and feelings of sadness or depression, if she presents to a mental health clinic; or, alternatively, in terms of “backaches,” “leg aches,” and “fear,” if she attends a Spiritist session. However, she will probably experience headaches, sleep disturbances, and nervousness regardless of the resource she uses. If we encounter her at the mental health clinic, she may explain her distress as due to disordered or out-of-control mind, behavior, or lifestyle. In the Spiritist session she will probably have her distress explained as an “obsession.” And if weencounter her before she seeks help from either of these treatment resources, she may describe her problems as due to difficulties with her husband or children (or to their having abandoned her). (p. 198)

In the treatment setting, integrating consumers’ popular or common sense notions of health and illness with biomedical notions has the potential to enhance treatment alliances and, in turn, treatment outcomes (Leventhal et al., 1997; Lopez, 1997).

Suicide

According to national statistics, Latinos had a suicide rate of approximately 6 percent in 1997 compared to a rate of 13 percent for the white population (DHHS, 1990). Overall, this lower rate suggests that Hispanic Americans are not demonstrating excess psychopathology through high rates of suicide. However, a national survey of 16,262 high school students in grades 9 through 12 found that Hispanics, both young women and young men, reported more suicidal ideation and specific suicidal attempts proportionally than whites and blacks. Over 10 percent of the Hispanics had attempted suicide, and 23 percent had considered the possibility of suicide (Centers for Disease Control and Prevention, 1998). Although this survey provided no data on actual suicides, these data suggest significant distress among Hispanic youth and are consistent with the several studies that found greater distress among Latinos than among largely white American youth.

High-Need Populations

Given that poverty is associated with homelessness and that many Hispanic American subgroups experience high rates of poverty, high rates of homelessness might be anticipated. However, the fact is that Hispanics are underrepresented among those without shelter (National Survey of Homeless Assistance Providers and Clients, 1996). Likewise, the need to place children in foster care is related to socioeconomic factors. Again, few Hispanic children are in the foster care system (DHHS, 1999). The fact that Hispanics are more likely to live with extended family members and with unrelated individuals suggests that family or friends may be taking care of those in need. Although Hispanics are relatively underrepresented among persons who are homeless or in foster care, they are present in high numbers within other vulnerable, high-need populations, such as incarcerated individuals, war veterans, survivors of trauma, and persons who abuse drugs or alcohol.

Individuals Who are Incarcerated

Low family socioeconomic status is associated with rates of chronic delinquency and crime (Wadsworth, 1979; Farrington, 1987; Tracy et al., 1990; Werner & Smith, 1992). The socioeconomic status of a neighborhood also predicts delinquency; that is, neighborhoods with high rates of adult unemployment, overcrowding, poor housing, low-achieving students, and high rates of mobility are all associated with high rates of delinquency (Rutter, 1979; Byrne & Sampson, 1986; McGahey, 1986; Schuerman & Kobrin, 1986). Given that many Latinos are poor and live within impoverished inner cities, relatively high rates of criminal involvement might be expected.

A larger proportion of Hispanic Americans (9 %) compared to white Americans (3 %) is incarcerated (Bureau of Justice Statistics, 1999). Among men, Hispanics are nearly four times as likely as whites to be in prison at some point during their lifetimes. Among women, less than 2 percent of Hispanics will enter prison compared to less than 1 percent of white women (Bureau of Justice Statistics, 1999). In addition, Hispanic youth make up 18 percent of juvenile offenders in residential placement (Bureau of Justice Statistics, 1999). Current epidemiological studies of incarcerated men and women include Hispanics and, in general, find that the rates of mental disorders among incarcerated individuals are higher than among community residents (Teplin, 1994; Teplin et al., 1996). Few ethnic differences among Hispanic Americans, white Americans, and African Americans were found. For those that were found, the small subsample of Latinos raises questions about the reliability of the findings.

Vietnam War Veterans

High rates of post-traumatic stress disorder (PTSD) exist among Vietnam War veterans. In a national study of Vietnam veterans (Kulka et al.,1990), Hispanics were found to be at higher risk for war-related PTSD than their white counterparts. In a further examination of Kulka’s work, Ruef and her colleagues (2000) found the risk for Hispanics also higher than that for black veterans, suggesting that the risk is not just related to minority status. In another recent reexamination of the Kulka study, Puerto Rican veterans in particular were found to have a higher probability of experiencing PTSD than were others with similar levels of war zone stress exposure (Ortega & Rosenheck, 2000). Because these differences in prevalence were not explained by exposure to stressors or acculturation and were not accompanied by substantial reductions in functioning, the authors suggest that differences in symptom reporting may reflect features of expressive style rather than different levels of illness. Another plausible factor in explaining the higher likelihood of experiencing PTSD is greater exposure to violence and trauma prior to entering the military (Bremmer et al., 1993).

Refugees

Many Hispanics, particularly Central Americans, have come to the United States as refugees, and only a small number of them were granted refugee status as defined by the U.S. Government. During the period of civil wars in Nicaragua, El Salvador, and Guatemala, an estimated 2 million Central Americans migrated to Mexico, the United States, and Canada. From 1990 to 1997, from 4 to 8 percent of the refugees who entered the United States legally were from Central America. Many others are believed to have entered the country through unauthorized channels. Although self-help groups and assistance centers were set up by religious organizations, these refugees did not have official U.S. Government sanction and thus received no U.S. Government resettlement benefits (Carillo, 1990).

Because Central American refugees often experienced the systematic violation of human rights in their own countries (Farias, 1994), they are at high risk for mental disorders such as PTSD and depression. Adults attending three schools in Los Angeles were examined for symptoms of PTSD and depression (Cervantes et al., 1989). Half of the Central American participants reported symptoms that were consistent with a diagnosis of PTSD. In comparison with recent Mexican immigrants, a greater proportion of Central American refugees reported symptom clusters of PTSD (50% versus 25%) (Cervantes et al., 1989). In another study, 60 percent of adult Central American refugee patients were diagnosed with PTSD (Michultka et al., 1998). Central American immigrant children seeking care at refugee service centers also had high rates of PTSD (33 %) (Arroyo & Eth, 1984). Thus, Central American refugees who have been exposed to trauma have a high need for mental health care.

Individuals with Alcohol and Drug Problems

Studies have consistently shown that rates of substance abuse are linked with rates of mental disorders (Kessler et al., 1996: Ross et al., 1988; Rounsaville et al., 1991). Most studies of alcohol use among Hispanics indicate that rates of use are either similar to or slightly below those of whites (Kessler et al., 1994). However, two factors influence these rates. First, gender differences in rates of Latinos’ use are often greater than the gender differences observed between whites. Latinas are particularly unlikely to use alcohol or drugs (Gilbert, 1987). In some cases, Latino men are more likely to use sub-stances than white men. For example, in the Los Angeles ECA study, Mexican American men (31 %) had significantly higher rates of alcohol abuse and dependence than non-Hispanic white men (21 %). In addition, more alcohol-related problems have been found among Mexican American men than among white men (Cunradi et al., 1999).

A second factor associated with Latinos’ rates of substance abuse is place of birth. In the Fresno study (Vega et al., 1998), rates of substance abuse were much higher among U.S.-born Mexican Americans compared to Mexican immigrants. Specifically, substance abuse rates were seven times higher among U.S.-born women compared to immigrant women. For men, the ratio was 2 to 1. U.S.-born Mexican American youth also had higher rates of substance abuse than Mexican-born youth (Swanson et al., 1992).

Strengths

The study of mental disorders and substance abuse among Latinos suggests two specific types of strengths that Latinos may have. First, as noted, Latino adults who are immigrants have lower prevalence rates of mental disorders than those born in the United States. Among the competing explanations of these findings is that Latino immigrants may be particularly resilient in the face of the hardships they encounter in settling in a new country. If this is the case, then the identification of what these immigrants do to reduce the likelihood of mental disorders could be of value for all Americans. One of many possible factors that might contribute to their resilience is what Suarez-Orozco and Suarez-Orozco (1995) refer to as a “dual frame of reference.” Investigators found that Latino immigrants in middle-school frequently used their families back home as reference points in assessing their lives in the United States. Given that the social and economic conditions are often much worse in their homelands than in the United States, they may experience less distress in handling the stressors of their daily lives than those who lack such a basis of comparison. U.S.-born Latinos are more likely to compare themselves with their peers in the United States. Suarez-Orozco and Suarez-Orozco argue that these Latino children are more aware of what they do not have and thus may experience more distress.

A second factor noted by the Suarez-Orozcos that might be related to the resilience of Latino immigrants is their high aspiration to succeed. Particularly noteworthy is that many Latinos want to succeed in order to help their families, rather than for their own personal benefit. Because the Suarez-Orozcos did not include measures of mental health, it is not certain whether their observations about school achievement apply to mental health. Nevertheless, a dual frame of reference and collective achievement goals are part of a complex set of psycho-logical, cultural, and social factors that may explain why some Latino immigrants function better than Latinos of later generations.

A second type of strength noted in the literature is how Latino families cope with mental illness. Guarnaccia and colleagues (1992) found that some families draw on their spirituality to cope with a relative’s serious mental illness. Strong beliefs in God give some family members a sense of hope. For example, in reference to her brother’s mental illness, one of the inform-ants commented:

We all have an invisible doctor that we do not see, no? This doctor is God. Always when we go in search of a medicine, we go to a doctor, but we must keep in mind that this doctor is inspired by God and that he will give us something that will help us. We must also keep in mind that who really does the curing is God, and that God can cure us of anything that we have, material or spiritual. (p. 206)

Jenkins (1988) found that many Mexican Americans attributed their relatives’ schizophrenia to nervios, a combination of both physical and emotional ailments. An important point here is that nervios implies that the patient is not blameworthy, and thus family members are less likely to be critical. Previous studies from largely non-Hispanic samples have found that both family criticism (for a review see Bebbington & Kuipers, 1994) and family blame and criticism together (Lopez et al., 1999) are associated with relapse in patients with schizophrenia. Mexican American families living with a relative who has schizophrenia are not only less likely to be critical, but also those who are Spanish-speaking immigrants have been found to be high in warmth. This is important because those patients who returned from a hospital stay to a family high in warmth were less likely to relapse than those who returned to families low in warmth (Lopez et al., 1998). Thus, Mexican American families’ warmth may help protect the relative with schizophrenia from relapse. The spirituality of Latino families, their conceptions of mental illness, and their warmth all con-tribute to the support they give in coping with serious mental illness.

Although limited, the attention given to Latinos’ possible strengths is an important contribution to the study of Latino mental health. Strengths are protective factors against distress and disorder and can be used to develop interventions to prevent mental disorders and to promote well-being. Such interventions could be used to inform interventions for all Americans, not just Latinos. In addition, redirecting attention to strengths helps point out the overemphasis researchers and practitioners give to pathology, clinical entities, and treatment, rather than to health, well-being, and prevention.