|Home | About | Journals | Submit | Contact Us | Français|
Research suggests that Asian Americans underutilize mental health services but an understanding of the multiple factors involved in utilization has not been examined in a nationally representative sample. The current study analyzed data from the National Latino and Asian American Study (NLAAS) and examined 368 individuals with disorders to understand utilization and what factors were related to the utilization of specialty mental health services. Significant underutilization was found for Asian Americans; moreover, underutilization was especially acute among Asian American immigrants. For U.S.-born Asian Americans, use of primary care services was significantly associated with use of mental health services, but for foreign-born Asian Americans, use of primary care services was unrelated to mental health services use. For both U.S.-born and foreign-born Asian Americans, use of alternative services appeared to significantly affect whether Asian Americans with disorders utilize mental health services, but the nature of the influence varied depending on the individual’s level of English-language proficiency. These findings revealed that a major mental health disparity, the underutilization of mental health services by Asian Americans, was nuanced by use of other health-related services and immigration-related factors.
It is well documented that ethnic minorities who suffer from a mental health problem tend not to seek mental health services (U.S. Department of Health and Human Services, 2001). Abe-Kim et al. (2007) found that for Asian Americans with a diagnosis consistent with criteria in the Diagnostic and Statistical Manual (4th ed. [DSM–IV]; American Psychiatric Association, 1994) during a 12-month period, 34.1% sought any services compared with 41.1% of all individuals in the National Comorbidity Survey–Replication (NCS-R) sample (Wang et al., 2005). However, these studies only examined service use within a 12-month period, which may yield conservative estimates of utilization given previous research documenting the tendency for Asian Americans to delay seeking services (e.g., Leong, 1986; Zane & Sue, 1991). We examined lifetime rates of service use in the mental health, primary care, and alternative/indigenous sectors.
Few investigations have examined how other types of care may constrain or facilitate mental health service use among Asian Americans. This study tested (a) the inhibition hypothesis, that other types of health care (e.g., primary or alternative care) inhibit individuals from seeking specialty mental health services, versus (b) the facilitation hypothesis, that other types of care are positively associated with the use of specialty mental health services. Asian Americans may underutilize mental health services because they use other support services that can compete with mental health services. It is also possible that the utilization of primary or alternative care in lieu of specialty mental health services (e.g., as delivered by psychologists or psychiatrists) may be influenced by culturally based beliefs about the etiology of mental health problems (Cheung & Snowden, 1990; Leong, Wagner, & Tata, 1995; U.S. Department of Health and Human Services, 2001), holistic beliefs about illness (Sue & Sue, 1999; Sue & Morishima, 1982), and the shame and face loss associated with mental illness (Zane & Yeh, 2002).
A limitation of previous utilization research is that many studies do not distinguish between Asian Americans who are foreign-born versus those who are U.S. born. Research has suggested that patterns of help seeking and quality of care may be quite different for immigrant and U.S.-born Asian Americans (Leong & Lau, 2001; U.S. Department of Health and Human Services, 2001). Moreover, previous research has tended not to focus on individuals with a diagnosable psychiatric disorder. Studies that measure only distress or that aggregate clinical with nonclinical cases can be misleading in terms of utilization rates. The notion that Asian Americans underutilize mental health services may be premature if most of the underutilization is accounted for by those with less severe problems. Previous studies of Asian American mental health utilization have been derived mostly from unrepresentative samples (e.g., treated populations in the public sector, convenience samples of college students). These studies have focused on a limited number of psychiatric disorders and, hence, could overlook the effects of other mental health conditions on service use.
This study addresses these prior limitations by using data from the most rigorously conducted nationally representative sample of Asian Americans—the National Latino and Asian American Study (NLAAS)—which included a wide range of psychiatric disorders. In the current investigation, we first conducted a precise study of underutilization by examining only individuals with a psychiatric disorder and by distinguishing between U.S.-born and foreign-born Asian Americans. Second, we tested competing hypotheses regarding the role of primary and alternative care on specialty mental health services use. Asian Americans may seek services in these sectors of care rather than specialty mental health services, so it is important to understand what role, if any, these services play in an individual’s decision to use services.
Data for the present study were derived from the NLAAS, conducted between May 2002 and December 2003 (see Alegria et al., 2004, for detailed sampling and data collection procedures). Data and informed consent were collected by trained bilingual interviewers. This work received institutional review board approval. Of the original sample, the present study examined the 368 Asian Americans who met criteria for a disorder. Among these individuals, 55.7% met the criteria for an affective disorder, 54.9% for an anxiety disorder, and 23.1% for a substance abuse disorder. The ethnic breakdown was as follows: Chinese: 30.4%; Filipino: 25.8%; Other Asian: 23.9%; and Vietnamese: 19.8%.1 Other sample characteristics are displayed in Table 1.
The primary diagnostic instrument used to establish a mental disorder was the World Mental Health Survey Initiative Version of the World Health Organization Composite International Diagnostic Interview (WMH–CIDI; Kessler & Üstün, 2004). Individuals met criteria for a disorder if they had been diagnosed with an affective, an anxiety, and/or a substance abuse disorder according to DSM–IV (American Psychiatric Association, 1994) criteria.2 Service use was assessed with the question, “Which of the following types of professionals did you ever see about problems with your emotions or nerves or your use of alcohol or drugs?” We chose lifetime rates because research has shown that most individuals with psychiatric disorders do eventually make treatment contact for these disorders (Kessler, Olfson, & Berglund, 1998).3 The dependent variable in this study was the use of specialty mental health services (e.g., as delivered by a psychiatrist, a psychologist, a counselor, or another mental health professional). The service use independent variables in this study were primary care services (e.g., as provided by a general practitioner, a nurse, an occupational therapist, or any other medical doctor or health professional) and alternative/indigenous services (e.g., as provided by a religious or spiritual advisor, a healer, a doctor of Oriental medicine, a chiropractor, or a spiritualist). A similar measure of service use has been applied in other studies (Berthold et al., 2007; Wang et al., 2005). English-language proficiency was assessed with the following item: “How well do you speak English?” Responses were separated into two categories: “poor/fair” and “good/excellent.” The NLAAS obtained respondents’ age, gender, ethnicity, level of educational attainment, household income, household size, marital status, health insurance coverage, and nativity status. Age at immigration was also assessed (12 years or younger vs. 13 years or older).4
To conduct our analyses, we used SAS 9.1 (SAS Institute, 2003). All analyses are based on weighted data. Separate logistic regression analyses were conducted for U.S.-born and immigrant participants to identify factors associated with mental health service use. The association between selected variables and utilization was estimated in the form of logistic regression weights and corresponding standard errors, which may be converted into odds ratios (ORs) and associated 95% confidence intervals (CIs).
Table 1 displays the weighted mean or percentage distribution of variables for all individuals by nativity status. We assessed for group differences on the predictor variables (e.g., psychiatric morbidity) as well as the covariates (e.g., gender). A logistic regression and a t test were used for dichotomous and continuous variables, respectively. U.S.-born Asian Americans were more likely to be diagnosed with a substance abuse disorder, Wald χ2 = 18.43, p < .001. These participants were also more likely to use specialty mental health services, Wald χ2 = 6.79, p < .01, and alternative services, Wald χ2 = 4.82, p < .05.
All of the individuals had a diagnosable mental health disorder, but among the entire sample, only 28% used specialty mental health services, 16% used primary care services, and 11% used alternative services. In comparison, drawing on data from the NCS-R, we calculated that 54% of those with the same disorders as in this study used mental health services. U.S.-born individuals utilized specialty mental health services at a rate almost double that of immigrants (40% vs. 23%).
Table 2 presents the regression model for variables associated with mental health utilization for U.S.-born and foreign-born Asian Americans separately. For the U.S.-born participants, the first model (estimated without including interaction effects) indicated that individuals using primary care services were almost 15 times more likely to use specialty mental health services (OR = 14.85; 95% CI = 2.90, 76.06). English-language proficiency and the use of alternative services were related to mental health services use, but these effects were qualified by an interaction. Results from the final model indicated that for individuals with poor or fair English-language proficiency, those who used alternative services were less likely to use specialty mental health services (OR < .001).5 For individuals with good or excellent English-language proficiency, the odds of using mental health services were almost seven times greater for individuals also using alternative services (OR = 6.91; 95% CI = 2.39, 20.00). Primary care use was again positively related to mental health service use (OR = 13.27; 95% CI = 2.69, 65.41).
For the foreign-born participants, only age at immigration was related to specialty mental health services use in the initial model; the older an individual was when he or she arrived in the United States, the lower the likelihood of using mental health services (OR = 0.23; 95% CI = 0.08, 0.64). When we included interactions in the model, age at immigration was no longer associated with service use. Moreover, the same interaction found in the U.S.-born sample emerged between use of alternative services and English proficiency. For individuals with poor or fair English language proficiency, those who used alternative services were less likely to use mental health services (OR < .001). However, those with good or excellent English-language proficiency who used alternative services had greater odds of using mental health services (OR = 25.58) compared with those who did not use alternative services.6
Our findings regarding utilization are broadly consistent with previous research (e.g., Matsuoka, Breaux, & Ryujin, 1997) in that the majority of Asian Americans (even though all met criteria for a psychiatric disorder) did not use specialty mental health services. However, utilization rates of specialty mental health services did vary by nativity status. The rate of mental health service use by U.S.-born individuals was almost twice that of immigrant Asian Americans. This is consistent with Abe-Kim et al. (2007), who also found that U.S.-born individuals used mental health services at higher rates than immigrants. Moreover, this substantial difference in utilization rate by nativity status may help reconcile previous findings demonstrating overutilization (Hu, Snowden, Jerrell, & Nguyen, 1991) and underutilization of services (Matsuoka et al., 1997). It is possible that in the research finding relative overutilization, the sample of Asian Americans may have been predominantly U.S. born, whereas in the research demonstrating underutilization, the samples either may have included equal proportions of immigrant and U.S.-born Asian Americans or may have consisted predominantly of immigrant Asian Americans.7
Studies examining the influence of primary and alternative care on specialty mental health services have demonstrated conflicting results (Berthold et al., 2007; Fang & Schinke, 2007). We found that primary care service use was positively associated with mental health services for U.S.-born Asian Americans but unrelated to mental health services for immigrant Asian Americans. The reasons for this variance in pattern of use remain unclear and merit the focus of future investigations. The findings provide partial support for the facilitation hypothesis for at least a subgroup of Asian Americans. There seems to be little support for the notion that primary care use inhibits the use of mental health services, even for less acculturated individuals.
Berthold et al. (2007) found that the use of alternative services was related to greater specialty mental health services use, which also has been observed in the general U.S. population (Druss & Rosenheck, 1999; Kessler et al., 2001). Our results provide partial support for this contention. For Asian Americans with a lower level of English-language proficiency, use of alternative services was negatively associated with the use of mental health services. However, for individuals with good or excellent English-language proficiency, use of alternative services was associated with a greater likelihood of using mental health services. There appears to be empirical support for the facilitation hypothesis for English-proficient Asian Americans, whereas the findings support the inhibition hypothesis for non–English-proficient Asian Americans. Findings from this study may provide some basis for understanding the conflicting findings in past research regarding the role of alternative care (Druss & Rosenheck, 1999; Kessler et al., 2001; U.S. Department of Health and Human Services, 2001).
There may be a number of reasons for the varying effects of alternative service use with level of English proficiency. First, non–English-proficient Asian Americans may be using alternative services in lieu of mental health care because almost all providers in the former speak Asian native languages, whereas such language capability may not be available among the latter group of service providers. For English-proficient Asian Americans, utilization of services, regardless of the type of service, for mental health problems may reinforce patients’ belief that they can benefit from treatment, which may make subsequent involvement in other types of care more likely (Berthold et al., 2007). Second, Asian Americans who are more proficient in English may be better able to articulate their emotional problems to their alternative care provider, increasing the likelihood of referrals to mental health services. Finally, English-language proficiency is a proxy variable, reflecting acculturation level, and those who are less acculturated may perceive the root of their problems as spiritual or transcendent and, as such, choose to seek help from alternative/indigenous services as opposed to specialty mental health services (Lin & Cheung, 1999; Sue & Sue, 1999). Those who are more acculturated (i.e., have higher levels of English-speaking proficiency) may be influenced by both Western and Eastern conceptions of illness and therefore seek services in both alternative care and mental health specialty sectors.
A number of clinical implications can be drawn. First, the findings affirm that underutilization continues to be a major mental health disparity in Asian American communities, suggesting the need for renewed efforts in outreach and community education. Second, outreach efforts should focus on developing more effective networks and referral bridges with alternative, indigenous healers and care providers in view of the possible inhibition effects of such services on mental health service use (Hall, 2001). Because facilitation effects were found for primary care services, clinicians should continue to capitalize on integration programs such as The South Cove Bridge Project in engaging Asian Americans in mental health treatment services (Yeung et al., 2004). This seems to be a fruitful and productive avenue in reducing underutilization.
The current study has several limitations. Results were based on self-report data, which may be subject to recall bias as well as social desirability concerns. Another limitation involved sampling predominately from three Asian American ethnic groups. Clearly, research needs to examine the utilization patterns of other Asian American groups. Finally, the NLAAS was a cross-sectional survey, which prevented the examination of temporal patterns of mental health, primary care, and alternative services use. Future longitudinal studies can better determine the natural sequencing of the help-seeking process.
Prior to the current investigation, no study to our knowledge has examined the relationship between nonspecialty mental health and specialty mental health services in such a representative sample of Asian Americans with disorders. Including other types of health care in the current study was informative and affirmed the ecological reality that related services can either compete with and inhibit or enhance and facilitate service use in another sector. These effects appear to vary depending on certain personal characteristics of Asian Americans pertaining to immigration-related factors.
This study was supported by the Asian American Center on Disparities Research through Research Grant P50MH073511 from the National Institute of Mental Health (NIMH) of the National Institutes of Health (NIH) and by NIH Research Grants MH06220, MH62207, and MH62209, funded by the NIMH, and RWJ DA18715, with generous support from the Substance Abuse and Mental Health Services Administration and the NIH Office of Behavioral and Social Sciences Research.
We thank Stanley Sue, Manveen Dhindsa, Carmel Gabriel, Lauren Berger, and Janice Cheng for their thoughtful advice and editorial comments.
1The “other Asian” category included individuals who identified themselves as Japanese, Korean, Asian Indian, or other.
2Affective disorder includes major depressive disorder and dysthymia; anxiety disorder includes panic disorder, agoraphobia without panic, social phobia, generalized anxiety disorder, and posttraumatic stress disorder; substance disorder includes alcohol abuse, alcohol dependence, drug abuse, and drug dependence.
3There is typically a delay between onset of symptoms and treatment seeking that averages between 6 and 14 years. Approximately 80% of individuals with anxiety and depression eventually make treatment contact (Kessler et al., 1998). Studies that examine only 12-month rates of service may be biased against individuals who do not seek treatment right away. Our study was able to directly address this issue of delay by examining lifetime rates of use.
4Research has shown that after the age of 12 years, children’s language acquisition skills markedly decline. Because immigrants who arrive in their host country prior to age 12 are better able to acquire the language of that particular culture, their cultural adjustment may be much less problematic than that of immigrants who arrive after age 12. Immigrants who arrive prior to adolescence may also be less likely to have adopted the cultural identity of their native country and, consequently, may be more likely to adopt the cultural identity of their host culture (Tsai, Ying, & Lee, 2000).
5Because of the limited sample size, a confidence interval could not be established. That is, because the number of individuals whose English-speaking proficiency was poor or fair and who were born in United States was only 10, there were very limited possibilities for the combination of using mental health and alternative services.
6Again, no confidence interval could be computed on account of the sample size.
7Nativity status was not reported in either of these studies.
Oanh Le Meyer, University of California, Davis.
Nolan Zane, University of California, Davis.
Young Il Cho, University of California, Davis.
David T. Takeuchi, University of Washington.