Complementary and Alternative Medical Service Use for Mental Health Problems among Chinese Americans: The Roles of Acculturation-related Factors

Abstract

The author used data from the Collaborative Psychiatric Epidemiology Surveys to examine the 12-month prevalence and predictors of the use of complementary and alternative medicine (CAM) relative to conventional Western medical services among Chinese Americans. The author examined the differences in service utilization patterns between Chinese Americans and non-Hispanic whites and the effects of acculturation factors such as generational status and English proficiency within the population of Chinese Americans. Multinomial logistic regression analyses revealed significant heterogeneity of exclusive CAM use in response to mental health problems by race/ethnicity and generational status, as well as English proficiency, gender, age, marital status, education, employment status, having insurance, and having any probably psychiatric disorder. Specifically, first-generation Chinese immigrants lagged behind second-, third-, or higher generation Chinese Americans and non-Hispanic whites in the likelihood of using exclusive CAM services for mental health problems. In addition, this study revealed that exclusive CAM service use was more popular than the use of only conventional Western medicine or a combination of both among all Chinese Americans except for the second generations. The findings provide a more nuanced understanding of the pattern of mental health service use among Chinese Americans. Implications for policy and research are discussed.

Keywords

Chinese Americans mental health service use CAM conventional Western medicine acculturation generational difference

Introduction

Complementary and alternative medicine (CAM) has been defined as “medical interventions not taught widely at U.S. medical schools or generally available at U.S. hospitals” (Eisenberg et al. 1993:246). They are, in a sociological perspective, “practices that are not in conformity with the standards of the medical community” (Eisenberg et al. 1993:246). Some typical CAM therapies include herbal medicine, acupuncture, and chiropractic (Ong et al. 2005). CAM use among U.S. adults substantially increased in the 1990s and has remained stable for the past decade (Barnes, Bloom, and Nahin 2008; Eisenberg et al. 1998; Tindle et al. 2005). According to data from the 2012 National Health Interview Survey, approximately one third of U.S. adults have used some type of CAM in the past 12 months (Clarke etal. 2015). Researchers have proposed several speculative explanations for the rising popularity of CAM (Astin 1998; Barnes et al. 2004; Coulter and Willis 2004), including

dissatisfaction with conventional Western medical service (as delivered by psychologists, psychiatrists, etc.) due to a growing emphasis on chronic illness in an aging population;

financial and language barriers to conventional services in a multicultural society;

a societal change toward postmodernism, individualism, and a holistic philosophy of health, which entails a decline in the acceptance of traditional authority and a desire for greater autonomy and control over medical decision making; and

marketing forces of local CAM providers and insurance companies, the rise of the Internet, and the transmission of “foreign” CAM therapies through globalization and migration.

However, researchers have found it hard to operationalize broad concepts such as “postmodern thesis,” “holistic philosophy of health,” and “trend of globalization” in empirical studies, which, in addition to other conceptual and methodological difficulties, have limited understanding of the prevalence, predictors, and patterns of CAM use.

One way to overcome some of the challenges is to examine one concept or one set of factors within one specific population. Therefore, the present study was focused on the effects of generational status and English proficiency on CAM use among Chinese Americans. Acculturation is generally defined “as the process by which individuals adopt the attitudes, values, customs, beliefs, and behaviors of another culture” (Abraído-Lanza et al. 2006:1342). Specifically, in the context of CAM use, the concept of acculturation encompasses various barriers to conventional treatment, culturally based beliefs about health and medicine, and immigration-related factors (Clarke et al. 2015; Eisenberg et al. 1993). Therefore, it provides a good angle to examine the issue at hand.

Research over the past three decades suggests that Asian immigrants underuse health care services in general (see Mayberry, Mili, and Ofili 2000) and mental health–related services specifically (Harris, Edlund, and Larson 2005), even when psychiatric disorder prevalence was controlled for (Wang et al. 2005). Among those who met the diagnostic criteria for mental disorders, the rate of using any service among Asian Americans was lower than the national average (U.S. Department of Health and Human Services 2001). This well-documented underuse of mental health services among Asian Americans “breeds the myth that Asian Americans are the so-called model minority and glosses over their mental health needs” (Kung 2003:110).

Studies conducted in China provide meaningful baseline data, and the findings shed lights on the patterns of help seeking behaviors of Chinese immigrants in the United States. These studies have found high levels of CAM use in general (Bodeker and Kronenberg 2002; Xu and Yang 2009), particularly for illnesses such as cancer (Chen et al. 2008) and hypertension (Gu et al. 2002). In China, CAM is practiced in a dual system with conventional Western medicine, and practices such as acupuncture and cupping are fully insured (Park et al. 2012). Although there are very limited empirical findings on how CAM is being used in relation to conventional Western medicine in Chinese populations, scholars have suggested that economic factors and cultural believes play important roles in CAM use (Bodeker and Kronenberg 2002). One study conducted in Hong Kong found that only 3 percent of patients in a primary health care setting used only CAM, and 16 percent used both CAM and conventional Western medicine. Research in China provided some background for understanding the behaviors of Chinese immigrants.

With 4 million people, Chinese Americans are the largest Asian group in the United States, according to the 2010 U.S. census (Hoeffel et al. 2012). However, the literature on mental health–related help-seeking behaviors among Chinese Americans is limited in several aspects (Bauer, Chen, and Alegría 2010; Kim and Keefe 2010; Sue et al. 2012), including the frequent use of regional community samples or college student samples, examining Asian Americans as one panethnic group, and inconsistent findings on the effects of several factors such as age and having health insurance. In addition, previous research has yet to provide a clear demonstration of acculturation effects, especially generational differences, in the patterns of mental health service use among Chinese Americans. Although complex and nuanced effects of acculturation-related factors such as English proficiency and generational status on the use of CAM and other mental health services have been suggested, stronger empirical evidence is much needed. Furthermore, because the acculturation phenomenon results from the conjunction of two or more cultures, research on acculturation must be comparative (Berry 1990). So far, few studies have directly compared Chinese Americans with non-Hispanic whites. Considering that non-Hispanic whites constitute the majority of the U.S. population, such a comparison is critical for understanding the sociocultural aspects of behavioral health and informing mental health service planning in a multicultural and multiethnic society.

The process of migrating to a new country and adjusting to a new environment has complicated effects on the levels of distress among migrants and their access to and compliance with mental health treatment (Lee et al. 2017; Tabora and Flaskerud 1994; Zhu 2017, 2018). At the time of the 2010 census, more than 76 percent of Chinese Americans were foreign born (Hoeffel et al. 2012), which means that Chinese cultural values and beliefs are very likely to be prominent among the majority of Chinese Americans. It is therefore important that we understand how immigration and acculturation influences the help-seeking behaviors of this population.

In this study I explore two research questions. First, how do Chinese Americans differ from non-Hispanic whites in their use of CAM relative to the conventional Western medical services? Second, what acculturation-related, sociodemographic, and health-related factors are associated with CAM use in response to mental health problems among Chinese Americans? I use a nationally representative sample of Chinese Americans, make comparisons between Chinese Americans and non-Hispanic whites, and examine the effects of generational status, English proficiency, sociodemographic characteristics, and other mental health–related factors on CAM use in response to mental health problems.

Literature Review

Mental Health Service Use among Chinese Americans

Empirical studies of mental health service use among Chinese immigrants confirm the general trend of underuse in this population. For example, using archival data from the Mental Health Division of Hawaii’s Department of Health, Leong (1994) found a 0.02 percent use rate of inpatient mental health facilities and a 0.11 percent use rate of outpatient mental health facilities among Chinese American respondents, much lower than the rates among other racial/ethnic groups. In a Canadian study, Tiwari and Wang (2008) found that only 2.5 percent of Chinese immigrants had talked to health professionals about their emotional or mental health conditions, again, much lower than that among white Canadians (9.8 percent). Note that the prevalence rates vary significantly across studies because of the varying scope of service sectors examined.

Research also found significant heterogeneity of mental health service use within the populations of Asian or Chinese Americans. Several significant predictors include being female (Bishop and Lewith 2010; Tata and Leong 1994), having a psychiatric disorder (Spencer and Chen 2004), and perceived discrimination (Spencer et al. 2010). Studies also found elevated likelihood of using any mental health service to be associated with being a U.S. citizen (Jang, Lee, and Woo 1998), being U.S.-born, higher generation (Abe-Kim etal. 2007; Le Meyer et al. 2009), better English proficiency (Kim et al. 2011), and high acculturation scores (Kung 2004).

However, several limitations exist in these studies. First, from a methodological point of view, the use of regional community samples or convenience samples of college students undermines the generalizability of the results. Second, many studies examined Asian Americans as one panethnic group and controlled for Asian ethnicity in their analyses (e.g., Abe-Kim et al. 2007; Kim et al. 2011). This is not enough to fully demonstrate the potentially unique pattern of service use for each Asian ethnic group. Given the unique immigration experience and cultural background of Chinese Americans, it is important to examine a nationally representative sample of this ethnic group, which was not done in previous studies.

The third limitation concerns the demonstration of an acculturation effect. There seems to be a gap between the two bodies of literature, one comparing Chinese Americans with non-Hispanic whites and the other examining variation among Chinese Americans. Studies in the latter body of literature found significant generational difference in mental health service use, while studies of the first type did not distinguish generational status of Chinese Americans at all. The acculturation literature suggests that “the behaviors and values of the later generations are much more similar to those of white-Americans than are those of their own parents or grandparents” (Leong 1986:198). Two comprehensive reviews of literature on immigration and assimilation, 25 years apart, have both supported this theory (Massey 1981; Waters and Jiménez 2005). Applying this perspective to mental health service use, one would expect greater similarity between higher generation immigrants and non-Hispanic whites than between more recent immigrants and the comparison group. Therefore, comparing different generations of Chinese Americans directly with non-Hispanic whites would generate a clear demonstration of acculturation effects. The results would also help us identify the high-risk group, the subpopulation in which “underutilization” is most acute (Abe-Kim et al. 2007).

CAM Use and Conventional Western Medical Service for Mental Health Problems

While trying to differentiate individuals who use mental health service from those who do not, researchers have also come to question “whether ‘use’ is a homogeneous category” (Pescosolido et al. 1998:284). There are a wide range of sources from which people may seek help for their mental health problems, from conventional Western health care services such as psychiatrists and psychologists, to complementary and alternative medical practitioners such as acupuncturists and chiropractors, each type with distinctive underpinnings of treatment and clienteles. With the increasing popularity of CAM use in the past two decades, a growing body of literature has accumulated on the prevalence and predictors of CAM use in the United States.

Several sociodemographic factors were consistently found to be associated with higher CAM use, including being female (Wolsko et al. 2002), a higher level of educational attainment (Barnes et al. 2004; Bishop and Lewith 2010), higher income (Tindle et al. 2005), and poor health status (Rafferty et al. 2002). Less consistent were the findings on the effects of age and having health insurance, as well as the racial/ethnic disparities. Elevated level of CAM use was found to be associated with younger age (Wang, Kennedy, and Wu 2015), middle age (Clarke et al. 2015), and older age (Oldendick et al. 2000). Mackenzie et al. (2003) found that the uninsured were more likely to use CAM than those with insurance, while Clarke et al. (2015) reported the highest level of CAM use among those with private health insurance, followed by the uninsured and then those with public insurance. Such inconsistency is likely due to differences in sample frame (national vs. limited geographic areas) and measurements of CAM use (i.e., types of CAM therapies included). Therefore, it is necessary to account for the effects of sociodemographic factors on CAM use.

One salient issue regarding the study of CAM use is whether it is a replacement for or a complement to conventional Western medical services (Druss and Rosenheck 1999). Because alternative treatment such as herbal medicines could cause drug interactions with conventional Western medical treatment with potentially serious clinical consequences (Izzo and Ernst 2012), it is imperative that we understand how various types of remedies are used in relation to one another (Silverstein and Spiegel 2001). So far, the findings of the role of CAM in affecting the use of other sectors of mental health services among the general U.S. population are conflicting (Berthold et al. 2007; Druss and Rosenheck 1999; U.S. Department of Health and Human Services 2001). Most of the studies focused on specialty mental health service (e.g., psychiatrists and psychologists) or conventional Western medicine in general (e.g., specialty mental health service, primary care). Le Meyer et al. (2009) summarized the two competing hypotheses as (1) the facilitation hypothesis, that the use of other types of services is positively associated with the use of specialty mental health services, and (b) the inhibition hypothesis, that the use of other types of services inhibits the use of specialty mental health services. Several studies found support for the facilitation hypothesis among the general U.S. population (Druss and Rosenheck 1999; Kessler et al. 2001; Upchurch et al. 2007), and Asian Americans (Berthold et al. 2007; Le Meyer et al. 2009), as well as Chinese Americans (Ma 1999). In contrast, other researchers have found that Asian Americans prefer nonprofessional services such as self-help group or traditional Chinese medicine to conventional Western mental health professionals (Chu, Hsieh, and Tokars 2011; Yang et al. 2009). Some researchers argued that people use CAM therapies in lieu of conventional Western medical services because the former are more empowering, offer more personal autonomy to patients, and acknowledged better “the role of nonphysical (mind/spirit) factors in creating health and illness” (Astin 1998:1548), while the latter was viewed as “impersonal, costly, inconvenient, unavailable, or inaccessible” (Murray and Rubel 1992:62). This perspective suggests a competing relationship between CAM therapies and conventional Western medicine, and is supported by several studies (U.S. Department of Health and Human Services 2001; Vincent and Furnham 1996).

To summarize, how CAM is used relative to conventional Western medicine remains unclear. If the majority of patients tend to seek help from alternative therapies in addition to conventional Western medical mental health treatment, then Western medical professionals need to be at least aware of, and preferably to proactively probe, CAM use in order to evaluate its potentially “dangerous” impact on Western medical treatment (Fang and Schinke 2007). If CAM use overlaps very little with the use of conventional Western medical care, then we need to identify the factors that distinguish the users of one from the users of another, in order to further understand how each sector is catering to the different needs of people.

Acculturation and CAM Use among Chinese Americans

The body of literature on assimilation and acculturation has grown significantly in the past century. One of the most influential early models is human ecology. For example, Park’s (1928) famous work on migration assumes a linear and directional model of acculturation, that the loss of the original culture occurs through greater acculturation to the host culture. Since then, the conceptual frameworks of acculturation have become more elaborate. Clark and Hofsess (1998) argued that immigrants selectively adopt the dominant cultural beliefs and practices for various reasons, and maintain certain values of their original cultures. Berry (2003) proposed a model of orthogonal relations between the original and the host culture; acculturation was considered a dynamic process of negotiating between maintaining original cultural characteristics and being involved in the larger society. Furthermore, Berry pointed out that acculturation is influenced by group-level factors such as the national/ethnical history of and reason for immigration, and ideology, attitudes in the host society, and individual-level factors such as personality and sociodemographic characteristics.

Meanwhile, empirical studies of acculturation and health seem to have lagged behind in applying this multidimensional, nonlinear conceptualization of acculturation. Although researchers have constructed acculturation scales with the intention of reflecting the diverse conceptualization of acculturation (e.g., language use, identity, and cultural practice), using such scales makes it almost impossible to capture the nuanced or potentially completely opposite effects of different factors on health believes and behaviors among immigrants (Abraído-Lanza et al. 2006). In fact, acculturation should be used as a “latent variable” with various indicators to reflect the cultural belief systems, values, linguistic and other behavioral preferences, and it should be measured “by considering factors relevant to the particular health issue at hand, rather than by a monolithic ‘acculturation’ concept” (Abraído-Lanza et al. 2006:1343). This “theory-driven model of acculturation” would help us understand how specific components of acculturation affect certain health behaviors.

Studies of service use for health problems in general suggest that acculturation and proxy measures are associated with CAM use among Chinese Americans. One study conducted in California found that Chinese Americans were more likely than non-Hispanic whites to use CAM for health problems in general and that English proficiency and length of U.S. residence were found to be associated with CAM use among Chinese Americans (Hsiao et al. 2006; Lee et al. 2010). Studies of Asian immigrants’ help seeing behaviors also examined the roles of citizenship as a proxy measure of acculturation and its relationship with access to health care (Atkinson, Lowe, and Matthews 1995; Jang et al. 1998; Nguyen 2011).

Specifically regarding help-seeking behaviors in response to mental health problems, a study of Asian American adults with psychiatric disorders found that the association between CAM use and conventional Western mental health service use was moderated by acculturation. Specifically, among individuals with lower levels of English proficiency, CAM use in response to mental health problems was negatively associated with specialty mental health service use. However, for those with good or excellent English proficiency, CAM use was positively associated with the use of specialty mental health services (Le Meyer et al. 2009). This nuanced role of English proficiency, as the authors argue, may help explain the conflicting findings on the use of CAM and conventional Western medical services among different populations, and suggests the necessity to examine the effects of acculturation-related factors in mental health-related service use among immigrants.

Two empirical studies also shed light on how acculturation affects CAM use in response to mental health problems among Chinese Americans. Using survey data of Chinese American patients in three community health centers in New York City, Fang and Schinke (2007) found that although CAM users were less acculturated than nonusers, the acculturation effect on CAM use in response to mental health problems was not significant when sociodemographic factors were controlled. However, they did not examine the use of conventional medical service at all. Another study only analyzed the use of formal mental health services (psychiatrists, psychologists, etc.). Interestingly, they found a gradation of increase in the willingness to seek help for their mental health problems as the level of Western influence increases, from Chinese students in mainland China and Hong Kong, to Chinese American (U.S.-born and foreign-born) students, and then European American students (Chen and Mak 2008).

Pieced together, these findings seem to suggest that from lower to higher generations of Asian or Chinese Americans and then to non-Hispanic whites, the likelihood of using conventional Western medicine would increase, while the likelihood of using CAM would decrease. Additionally, how CAM is used relative to conventional Western medicine might differ by English proficiency. These nuanced effects have yet to be tested in empirical studies.

To address the issues discussed above, I performed two different analyses. In the first analysis I examined the differences of CAM use outcomes among different generations of Chinese Americans and non-Hispanic whites while controlling for sociodemographic and health-related factors. The first hypothesis tested is as follows:

Hypothesis 1: First-generation Chinese Americans are the most likely to use only CAM therapies, followed by second-generation, third- or higher generation, and then non-Hispanic whites; the rates of using any conventional Western medicine will follow an ascending trend from first-, second-, and third- or higher generation Chinese Americans to non-Hispanic whites.

In the second analysis, I turned to the effects of three acculturation-related factors (generational status, U.S. citizenship, and English proficiency) on CAM use outcomes among Chinese Americans alone, while controlling for sociodemographic and health-related factors, testing the following hypotheses, controlling for other acculturation-related factors:

Hypothesis 2: Chinese Americans who speak good or excellent English will be less likely to use only CAM therapies, and more likely to use any conventional Western medical services, than those with fair or poor English proficiency.

Hypothesis 3: Among Chinese Americans, U.S. citizens are less likely to use only CAM therapies, and more likely to use any conventional Western medical services, than non-U.S. citizens.

Because the length of U.S. residence is a meaningful measure only for foreign-born immigrants, I examined its association with mental health service use among only first-generation Chinese Americans:

Hypothesis 4: Among first-generation Chinese Americans, those with longer U.S. residency will be less likely to use only CAM therapies, and more likely to use any conventional Western medical services, than non-U.S. citizens.

The following section discusses the methods used in this study, including the data and sample, measurement, and a variety of issues regarding statistical modeling and interpretation.

Methods

Data Source

I use publicly available data from the Collaborative Psychiatric Epidemiology Surveys (CPES)m funded by the National Institute of Mental Health (Alegria et al. 2016). This survey joins together three nationally representative surveys: the National Comorbidity Survey Replication (NCS-R), the National Survey of American Life (NSAL), and the National Latino and Asian American Study (NLAAS). The CPES population includes adults aged 18 and older living in households in the 48 coterminous United States (NCS-R, NSAL), and the population for the Latino and Asian ancestry groups extends to Hawaii (NLAAS). Respondents in all three surveys were selected from a four-stage clustered area probability sample of households, with selection using area data from the 2000 U.S. census. Although each survey has unique features, they share a common core of measurements of primary mental health diagnostic symptoms, symptom severity, and mental health–related service use. More details on CPES sample design features are provided by Heeringa et al. (2004). This study focuses on Chinese American respondents, using non-Hispanic whites as a comparison group. The CPES data consist of 600 self-identified Chinese Americans from NLAAS and 7,587 non-Hispanic whites from NCS-R and NSAL. The non-Hispanic white subsample from NSAL (n = 891) was designed for the focus of the survey, black/white contrasts, and is therefore considered not optimal for comparative analyses between whites and nonblack minority groups (Jackson et al. 2004) and therefore was dropped. Also excluded are 2,516 non-Hispanic white cases from NCS-R Part I, who did not complete the more in-depth Part II interview. As a result, the sample analyzed in this study consists of 4,180 non-Hispanic white respondents from NCS-R and 600 Chinese Americans from NLAAS.

Measures

Dependent Variable

Mental health service use was assessed with the question “In the past 12 months, did you go to see [provider on list] for problems with your emotions, nerves, or your use of alcohol or drugs?” Four sectors of service were originally constructed: (1) specialty mental health service (psychiatrist, psychologist, social worker, counselor, any other mental health professionals, such as a psychotherapist or mental health nurse), (2) general health care service (general practitioner, family doctor, nurse, or other health professional, etc.), (3) religious or spiritual advisers, and (4) alternative medicine therapies (acupuncture, biofeedback, relaxation or meditation techniques, etc.) (Wang et al. 2005). After examining each type of service separately, I combined the first two categories into “conventional Western medical service” and the last two into “CAM therapy” because of the very limited number of respondents who used only general health care service or only religious or spiritual advisers. Although it would be interesting to examine those who used both CAM and conventional Western medical service as one separate category, the limited number of the cases that belong to this category (14 Chinese Americans and 353 non-Hispanic whites) would prove difficult for statistical analysis. Therefore, the outcome variable is constructed as three categories:

“any conventional Western medical service” included respondents who had used any conventional Western medical services in the past 12 months,¹

“CAM only” included respondents who only used the other two sectors of services, and

“none” included those who had used none of the services for problems with their emotions, nerves, or alcohol or drug use.

This last category is used as the reference group of the outcome variable in the multinomial logistic regression models.

Acculturation-related Factors

Three factors related to acculturation were included in the analysis. Generational status of the respondents was measured in three categories: first generation (foreign-born immigrant), second generation (born in the United States with at least one parent being an immigrant), and third or higher generation (born in the United States and both parents born in the United States). Length of U.S. residence was coded in three categories: “U.S. born,” “fewer than 10 years,” and “10 or more years.” The 10-year cutoff point was used because it was associated with socioeconomic status, insurance access, health behaviors, and health outcomes among immigrants (Carrasquillo, Carrasquillo, and Shea 2000; Li and Hummer 2015; Stephen et al. 1994). Citizenship was measured by whether respondent was a U.S. citizen or not. English language proficiency was assessed using the question “How well do you speak English?” The response was dichotomized as “fair/poor” or “excellent/good.”

Control Variables

Diagnostic criteria from the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV), were used to assess the mental health status of respondents. A probable DSM-IV endorsed diagnosis of any psychiatric disorder in the past 12 months was measured on the basis of whether respondents’ reported symptoms met the diagnostic criteria of any one out of the four categories: (1) depressive disorders (major depressive disorder or dysthymia), (2) anxiety disorders (panic disorder, agoraphobia without panic disorder, social phobia, generalized anxiety disorder, or posttraumatic stress disorder), (3) substance use disorders (alcohol abuse or dependence, drug abuse or dependence), or (4) impulse-control disorder (intermittent explosive disorder). Having health insurance was dichotomously coded. Other variables considered include gender, age, marital status, having a college degree, and work status. Refer to Table 1 for more details on how they are coded.

Table 1.

Weighted Descriptive Statistics of Chinese Americans and Non-Hispanic Whites.

	Non- Hispanic White	First- generation CA	Second- generation CA	Third- or Higher Generation CA	p Value
n	3,983	470	70	40
Mental health service use					<.001
Any conventional	15.03 (.01)	3.42 (.01)	13.73 (.05)	15.90 (.07)
CAM only	20.72 (.01)	6.41 (.01)	12.91 (.05)	16.40 (.07)
None	64.25 (.01)	90.17 (.02)	73.36 (.07)	67.70 (.09)
Gender					>.05
Female	51.91 (.01)	54.02 (.02)	57.28 (.07)	42.09 (.05)
Male	48.09 (.01)	45.98 (.02)	42.72 (.07)	57.91 (.05)
Age (mean)	46.91 (.52)	43.28 (1.19)	38.78 (2.81)	40.67 (1.63)	<.001
Marital status					<.001
Never married	19.24 (.01)	16.13 (.02)	40.96 (.09)	29.28 (.06)
Married/cohabiting	60.04 (.01)	73.77 (.02)	44.54 (.09)	51.81 (.11)
Divorced/separated/widowed	20.36 (.01)	10.10 (.02)	14.50 (.05)	18.91 (.11)
Education (years)					<.001
≤11	12.95 (.01)	21.08 (.03)	.05 (.03)	0
12s	31.75 (.02)	17.65 (.02)	11.09 (.05)	16.94 (.05)
13–15	29.01 (.01)	18.43 (.03)	28.10 (.05)	40.27 (.09)
College or more	26.29 (.01)	42.84 (.04)	55.43 (.06)	42.79 (.06)
Work status					>.05
Employed	65.50 (.01)	66.62 (.03)	58.53 (.05)	67.62 (.11)
Unemployed	5.01 (.01)	5.45 (.01)	4.52 (.02)	10.05 (.08)
Not in labor force	29.49 (.01)	27.94 (.03)	36.95 (.06)	22.34 (.07)
Probable DSM-IV diagnosis					<.001
Yes	21.07 (.01)	8.93 (.02)	18.24 (.05)	29.50 (.09)
No	78.93 (.01)	91.07 (.02)	81.76 (.05)	70.50 (.09)
Has health insurance	90.11 (.01)	84.88 (.02)	93.39 (.03)	97.75 (.03)	<.05
Citizenship	—				<.001
U.S. citizen	—	40.53 (.03)	0	0
Non-U.S. citizen	—	59/47 (.03)	100	100
English proficiency	—				<.001
Poor/fair English	—	53.44 (.04)	10.78 (.05)	.04 (.004)
Good/excellent English	—	46.56 (.04)	89.22 (.05)	96.55 (.004)
Length of U.S. residence
U.S. born	100	0	100	100
<10 years	0	17.75 (.03)	0	0
≥10 years	0	82.25 (.03)	0	0

Note: Data are expressed as percentage (standard error). CA = Chinese American; CAM = complementary and alternative medicine; DSM-IV = Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition.

Statistical Analysis

All statistical analyses were conducted using Stata 14. Because the outcome, mental health service use, consists of three nominal categories, multinomial logistic regression to study the correlates of service use was used. The CPES supplied weighting was applied in all analytical procedures to adjust for the study’s complex sampling methods using the SVY command. Consequently, multivariate significance testing uses Wald χ² tests based on coefficient variance-covariance matrices that adjusted for design effects using the Taylor’s series linearization method. To examine any generational differences of sociodemographic and mental health–related characteristics in Table 1, I conducted the Rao-Scott χ² test (Rao and Scott 1984) a design-adjusted version of the Pearson χ² test for categorical variables and weighted linear regression and the Wald test for age, a continuous variable.

For each covariate in the multinomial logistic regression models, the average of the marginal effects, an increasingly preferred option used by researchers, is computed taking the average of the marginal effect of a covariate for each observation in the estimation sample at its observed values; the computation is done with the MCHANGE command (Long and Freese 2014). Specifically, for all variables, the author computed the discrete change (DC), the difference in predicted probability for each category of mental health service use between the covariate categories (e.g., men vs. women, employed vs. unemployed). For age, which is entered as a continuous variable in single and quadratic terms, DC was computed for the change in predicted probability for each category of the outcome variable by one standard deviation change. To compare the effect size across variables, I computed a measure of the effect size for each covariate, whether a continuous or dummy variable, by taking the average of the absolute values of the DC for all three categories of mental health services.

For the two multinomial logistic regression models presented later, Wald test results² confirmed that all three categories of the dependent variable are distinguishable from one another. Also, even though the Hausman-McFadden test is not available after the svy command, I computed the odds for each pair of mental health service use outcomes for the multinomial logistic regression models presented later. I compared the odds of the full model with those with the third outcome category excluded for any given pair of outcomes, the results³ of which indicated no violation of the independence of irrelevant alternatives assumption in either model. It is worth noting thatto my best knowledge, there is currently no goodness-of-fit test for multinomial logistic regression analysis using complex survey data, hence the inability to evaluate the power of the models in predicting the patterns of service use.

Results

Table 1 presents the descriptive statistics of sociodemographics and health-related factors for non-Hispanic white and three generational groups of Chinese Americans, along with significance test results (p values) that indicate any differences among the four groups. First-generation (3.42 percent) and second-generation (13.73 percent) Chinese Americans had significantly (p < .001) lower rates of using at least one type of conventional Western medical services for mental health problems than did third- or higher generation Chinese Americans and non-Hispanic whites (15.90 percent and 15.03 percent, respectively). Also, first-generation (6.41 percent) and second-generation (12.91 percent) Chinese Americans had significantly (p < .001) lower rates of using exclusively CAM for mental health problems than did third- orhigher generation Chinese Americans and non-Hispanic whites (16.40 percent and 20.72 percent, respectively). Compared with non-Hispanic whites, Chinese Americans were in general younger and more likely to have a college or higher degree (p < .001). First-generation (91.07 percent) and second-generation (81.76 percent) Chinese Americans were also significantly less likely to have a probable DSM-IV diagnosis than the other two groups (p < .001). There were no significant differences in gender and work status composition among the four groups. Among the three groups of Chinese Americans, there were significant differences in U.S. citizenship status and English proficiency (p < .001), with the first-generation Chinese Americans having a lower rate of being citizenship and good or excellent English proficiency.

Table 2 presents a closer look at Chinese Americans’ use of mental health service by acculturation-related factors. Weighted crosstabulation results show that mental health service use varied significantly by generational status and English proficiency but not U.S. citizenship. Specifically, first-generation immigrants lag behind second, third, and higher generations, on the proportion having used any conventional Western medical health services and the proportion using CAM exclusively. Those with poor or fair English proficiency also have a significantly lower level of mental health service use of any conventional Western medicine, as well as CAM only, than those reporting speaking good or excellent English. Although these bivariate results are suggestive of the potential importance of these acculturation-related factors, it is important to determine if such associations hold up while controlling for other predictors.

Table 2.

Weighted Percentages of Mental Health Service Use among Chinese Americans by Acculturation-related Factors (n = 580).

	Any conventional	CAM Only	None	p Value
Generational status				<.001
First generation	3.70 (.01)	7.14 (.02)	89.16 (.02)
Second generation	13.41 (.05)	12.11 (.04)	74.48 (.06)
Third or higher generation	14.50 (.06)	17.16 (.06)	68.34 (.08)
Citizenship				>.05
U.S. citizen	5.72 (.01)	7.54 (.01)	86.73 (.01)
Non-U.S. citizen	5.40 (.02)	10.25 (.03)	84.35 (.04)
English proficiency				<.05
Poor/fair English	3.20 (.01)	5.10 (.02)	91.70 (.02)
Good/excellent English	7.59 (.02)	11.14 (.02)	81.27 (.03)
Length of U.S. residence (years)^a				<.05
<5	6.95 (.03)	14.97 (.06)	78.09 (.07)
≥5	3.03 (.01)	5.51 (.01)	91.46 (.02)

Note: CAM = complementary and alternative medicine.

Length of U.S. residence was analyzed only among 470 first-generation Chinese Americans.

Using multinomial logistic regression, I examined how first, second, and third or higher generations of Chinese Americans compare with non-Hispanic white respondents on the CAM use outcomes, controlling for sociodemographic characteristics and health-related factors. Regression coefficients and DCs for the covariates are presented in Table 3.

Table 3.

Weighted Multinomial Logistic Regression Results and AMEs of Correlates of 12-month Use of Mental Health Service among Chinese Americans and Non-Hispanic Whites (n = 4,535).

	Any Conventional vs. None, b (SE)	Any Conventional, AME (SE)	CAM Only vs. None, b (SE)	CAM Only, AME (SE)
Non-Hispanic white (reference)
First-generation CA	−1.82 (.33)^‡	−.10 (.01)	−1.68 (.22)^‡	−.15 (.01)
Second-generation CA	−.42 (.54)	−.02 (.05)	−.97 (.42)^†	−.11 (.04)
Third- or higher generation CA	−.51 (.37)	−.03 (.04)	−.56 (.46)	−.06 (.06)
Female (reference: male)	.63 (.10)**	.04 (.01)	.83 (.09)**	.10 (.01)
Age	.10 (.02)**	−.001 (.0003)	.03 (.02)	−.001 (.0005)
Age²	−.001 (.0002)**	-	−.0004 (.0002)	-
Marital status (reference: never married)
Married/cohabitating	−.36 (.13)**	−.03 (.01)	−.09 (.14)	−.001 (.02)
Divorced/separated/widowed	.30 (.17)	.03 (.02)	.03 (.21)	−.006 (.03)
Education (reference: ≤11 years)
12 years	−.03 (.17)	−.02 (.02)	.56 (.15)**	.06 (.01)
13–15 years	.02 (.15)	−.01 (.02)	.68 (.18)**	.08 (.02)
College or higher	.38 (.17)*	−.001 (.02)	1.29 (.22)**	.17 (.03)
Work status (reference: employed)
Unemployed	.30 (.16)	.03 (.02)	.10 (.23)	−.005 (.03)
Not in labor force	.67 (.15)**	.08 (.02)	−.01 (.13)	−.02 (.02)
Probable DSM-IV diagnosis	2.02 (.10)**	.24 (.02)	.98 (.07)**	.07 (.01)
Health insurance	.43 (.17)*	.05 (.02)	−.21 (.19)	−.05 (.03)
Intercept	−4.60 (.49)		−2.59 (.42)
Average predicted probability		.14		.19

Note: Discrete changes are presented for all variables except age, for which marginal change is presented. AME = average marginal effect; CA = Chinese American; CAM = complementary and alternative medicine; DSM-IV = Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition.

p ≤ .05, two-tailed test. ^†p ≤ .05, one-tailed test; **p ≤ .01, two-tailed test; ^‡p ≤ .01, one-tailed test.

All covariates are significantly associated with mental health service use. First and foremost, regression coefficients from the multinomial logistic regression confirmed the underuse of mental health service among first- and second-generation Chinese Americans, but not for the third- or higher generation Chinese Americans. The predicted probability for first-generation Chinese Americans to use any conventional Western medical services is, on average, 10 percentage points (DC = –.10) lower than that among non-Hispanic whites. Interestingly, the disparity was even greater (15 percentage points) on the likelihood of using exclusively CAM therapies. These differences are large in magnitude, considering that the sample average predicted probability of using any conventional services was .14, and .19 for only CAM use. As for second-generation Chinese Americans, even though they did not differ significantly from non-Hispanic whites on the likelihood of using any conventional services rather than none, they were less likely to use only CAM therapies than non-Hispanic whites (DC = –.11). This difference is smaller than that between first-generation Chinese Americans and non-Hispanic whites but still substantial in size. Third- or higher generation Chinese Americans did not differ significantly from non-Hispanic whites on the use of mental health services.

Unsurprisingly, having a probable DSM-IV diagnosis was associated with service use. Those with a probable DSM-IV diagnosis were also more likely to use any conventional services (DC= .24) and more likely to only use CAM (DC = .07), compared with those without. Gender, age, education, marital status, employment status, and age were also significantly factors in the model.

How big are the differences between first-generation Chinese Americans and non-Hispanic whites in the outcome categories? As previously noted, we can compare the average absolute DC across all three categories of mental health service use for this purpose. The average difference between first-generation Chinese Americans and non-Hispanic whites is .17, which is only slightly lower than that the average difference between having a probable diagnosis of DSM-IV psychiatric disorder versus not (.21).⁴ This indicates that the degree of difference between first-generation Chinese Americans and non-Hispanic whites is comparable in magnitude to the degree of difference between those who have a probable psychiatric disorder diagnosis and those who do not. This alarming yet not entirely unexpected finding confirms the underuse of mental health service use among recent immigrants and at the same time begs the question of what factors other than the ones already included in the present model can contribute to such ethnic disparity. In their review of the health care use among Asian immigrants, Clough, Lee, and Chae (2013) pointed out that the cultural incompetency of and discrimination in the health care system, in addition to linguistic and economic issues related to access and quality of service, are the major barriers Asian immigrants face to participating fully in the health care system in the United States. Future research should examine the effects of cultural competency and discrimination on help-seeking behaviors among Asian immigrants to help health providers develop efficient interventions.

Figure 1 provides a visual presentation of the predicted probabilities of mental health service use among Chinese Americans of different generations and non-Hispanic whites, when other variables were held at the mean values for each group (the local mean).⁵ As Figure 1 shows, first-generation Chinese Americans lag behind the other three groups in the use of any conventional services and exclusively CAM therapies in the past 12 months. The stacked bars formed a gradation, from first-, second-, and third- or higher generation Chinese Americans to non-Hispanic whites. The predicted probabilities of not using any mental health services (the blank bars) are .91, .71, .69, and .362, respectively, with first generation significantly different from all three other groups (p < .05, results not shown). The predicted probabilities of exclusive CAM use (the black bars) show a similar gradation, increasing from .06, .13. 17, to .21 from left to right across the four groups, with first-generation Chinese Americans significantly lower than non-Hispanic whites (p < .05, results not shown). As for the use of any conventional Western services (the gray bars), the predicted probability for the first generation (.03) is noticeably different from the other three groups, with the differences reaching statistical significance for the comparison with the third- or higher generation Chinese Americans (.14), and with non-Hispanic whites (.17). In summary, the first part of hypothesis 1 regarding CAM use is rejected, while the second part regarding the use of conventional Western medicine is confirmed. It is worth noting that second-generation Chinese Americans presented a distinctive pattern of service use, being the only group with a higher predicted probability of using any conventional services (.16) than that of using only CAM therapies (.13). In other words, first and third or higher generations of Chinese Americans, as well as non-Hispanic whites, when seeking help for their mental health problems all tend to use exclusively CAM therapies somewhat more than any conventional Western medical services, whereas second-generation Chinese Americans rely slightly more on at least some conventional Western medical treatment.

Figure 1.

Predicted probability of mental health service use among Chinese Americans and non-Hispanic whites (n = 4,563).

An important issue to consider is whether these generational differences among Chinese Americans can be explained by other indicators of acculturation. To assess this, I restricted the sample to Chinese Americans and added the other available acculturation indicators (citizenship and English proficiency) to the analysis of mental health service usage. Table 4 presents the results of this multinomial logistic regression analysis. The last row of the table shows that on average, Chinese Americans have a predicted probability of .06 to use any conventional Western services and .08 to use CAM exclusively. Generational status and English proficiency were associated with mental health service use among Chinese Americans. Compared with individuals with poor or fair English proficiency, those with good or excellent English proficiency were more likely to use any conventional services. This confirms the impact of language barriers to use of conventional Western medical services. English proficiency was not significantly associated with CAM use. Figure 2 shows the predicted probabilities of mental health service use by English proficiency among Chinese Americans when other variables were set at the local means. Specifically, Chinese Americans with good or excellent English proficiency are more likely to use any conventional Western medicine (Pr = .05) than those with poor or fair English proficiency (.02). The former also have higher predicted probability of using only CAM (.07) than the latter (.02). Therefore, the part regarding exclusive CAM use in hypothesis 2 is rejected, and the part regarding conventional Western medical service use is confirmed. It is worth noting that the pattern of service use differs by English proficiency: those with good or excellent English proficiency have a higher probability of using only CAM than the probability of using any conventional Western medical service, while those with poor or fair English proficiency have equal probabilities of the two outcomes. This result challenges the previous finding of a positive correlation between CAM use and specialty mental health service by Le Meyer et al. (2009). In addition, U.S. citizenship and length of U.S. residency were not significant predictors, rejecting hypotheses 3 and 4.

Table 4.

Weighted Multinomial Logistic Regression Results and AMEs of Correlates of 12-Month Use of Mental Health Service among Chinese Americans (n = 580).

	Any Conventional vs. None, b (SE)	Any Conventional, AME (SE)	CAM Only vs. None, b (SE)	CAM Only, AME (SE)
First generation (reference)
Second generation	1.64 (.75)^†	.09 (.06)	.28 (.71)	.003 (.04)
Third or higher generation	1.03 (.59)^†	.04 (.04)	1.16 (.50)^†	.08 (.05)
≥10 years U.S. residence (reference: <5 years)^a	.19 (.64)^†	.01 (.02)	.12 (.43)	.01 (.02)
Good/excellent English	.89 (.54)^†	.04 (.02)	.07 (.40)	−.001 (.03)
Female (reference: male)	.09 (.52)	−.005 (.02)	1.03 (.41)*	.01 (.02)
Age	.15 (.09)	.001 (.001)	.02 (.07)	−.001 (.001)
Age²	−.002 (.001)	—	−.0003 (.001)	—
Marital status (reference: never married)
Married/cohabitating	−1.70 (.78)*	−.07 (.05)	−1.12 (.57)	−.07 (.05)
Divorced/separated/widowed	.19 (.98)	.03 (.08)	−1.06 (1.01)	−.08 (.06)
Education (reference: ≤11 years)
12 years	−1.27 (.91)	−.08 (.06)	.92 (1.25)	.02 (.02)
13–15 years	−1.26 (.71)	−.08 (.05)	1.78 (1.32)	.05 (.03)
College or higher	−1.39 (.64)*	−.10 (.05)	2.86 (1.19)*	.13 (.02)
Work status (reference: employed)
Unemployed	−.40 (1.10)	−.02 (.03)	.96 (.43)*	.08 (.05)
Not in labor force	.55 (.53)	.03 (.03)	.04 (.38)	−.002 (.02)
Probable DSM-IV diagnosis	2.51 (.45)**	.17 (.05)	1.25 (.40)**	.08 (.04)
Health insurance	.45 (1.32)	.03 (.04)	−1.00 (.53)	−.09 (.05)
Intercept	−5.65 (2.85)		−4.13 (1.78)
Average predicted probability		.06		.08

Note: Discrete changes are presented for all variables except age, for which marginal change is presented. AME = average marginal effect; CAM = complementary and alternative medicine; DSM-IV = Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition.

Length of U.S. residence was analyzed only among 470 first-generation Chinese Americans, with other covariates, except for generational status, controlled for.

p ≤ .05, two-tailed test. ^†p ≤ .05, one-tailed test. **p ≤ .01, two-tailed test.

Figure 2.

Predicted probability of mental health service use by English proficiency among Chinese Americans (n = 580).

The association between mental health service use and another acculturation-related factor, generational status, was more complicated. Specifically, second-generation Chinese Americans were more likely to use any conventional services than first-generation Chinese Americans (DC = .09), but they did not differ significantly on using only CAM therapies. As for third- or higher generation Chinese Americans, they were more likely to use any conventional services (DC = .04) or to use only CAM therapies (DC = .09) than the first generation.

Figure 3 presents the predicted probabilities of 12-month service use among Chinese Americans by generational status, when other factors were set at local means. First-generation Chinese immigrants lagged behind their second- and third- or higher generation counterparts on the overall use of mental health services, which is consistent with previous findings (e.g., Kung 2004). Similar to the previous model, the predicted probability for exclusive CAM use also climbed up from .04 to .08, then to .15, from left to right across the three groups. And once again, second-generation Chinese Americans showed a strong preference of using any conventional Western services (.18) to only CAM therapies (.08). Note that the predicted probabilities for the three Chinese American groups in Figure 3 are different from those in Figure 1, which is due primarily to the two extra variables, English proficiency and citizenship, that were introduced in the Chinese American–alone model, the results presented in Table 4.

Figure 3.

Predicted Probability of Mental Health Service Use by Generational Status among Chinese Americans (n = 580).

Once again, having a probable DSM-IV diagnosis were found to be associated with service use among Chinese Americans. Those with a probable DSM-IV diagnosis were much more likely than those without to use any conventional services (DC = .17), as well as use only CAM (DC = .08). Gender, marital status, education, and work status were also significant predictors in this model.

To further examine the effect of a probable DSM-IV diagnosis, I computed the predicted probability of the outcome categories separately for individuals with and without a probable DSM-IV diagnosis. Figure 4 shows that a probable DSM-IV diagnosis not only affect the probability, but also the pattern of using mental health services. Individuals with a probable DSM-IV diagnosis were not only more likely to use any services for mental health problems (.47), they also prefer to use at least some conventional Western medicine (.33) than only CAM (.14). Those without a probable DSM-IV diagnosis, however, prefer to use only CAM (.04) than to use any conventional Western medicine (.02), even though the likelihood using any mental health services was low (.06). This finding suggests that those with subthreshold or subclinical mental disorder symptoms prefer using only CAM.

Figure 4.

Predicted probability of mental health service use by psychological disorder status among Chinese Americans (n = 580).

Discussion

This study has three major findings. First, the results confirm the overall underuse of mental health services among Chinese Americans compared with non-Hispanic whites (Su and Li 2011). This calls for greater efforts in mental health service outreach to this ethnic minority community, including providing linguistic support given the effects for English proficiency, and more ethnic responsive services.

Second, this study reveals that exclusive use of CAM therapies was more popular than using any conventional Western medical services among first- and third- or higher generation Chinese Americans, though not for the second generation. This finding contradicts, partially at least, the findings of Druss and Rosenheck (1999) that CAM was used as an add-on rather than an alternative to formal medical services. In addition to confirming the rising popularity of CAM therapies found in previous studies (e.g., Wolsko et al. 2002), this finding underscores the need for research to better understand the efficacy of CAM services and the satisfaction of patients/customers with such treatments. It also highlights the need for more discussions on insurance policies regarding alternative and complementary mental health services: should health insurance be more inclusive of this sector of services, given that CAM use was more prevalent than conventional services for the majority of the Chinese American population?

Third, a higher level of acculturation is associated with greater use of mental health service, overall and by sector, among Chinese Americans. First-generation immigrants were the least likely to seek help for problems with their emotions, nerves, or substance use. Second-generation Chinese Americans present a unique pattern of service use; though slightly lagging behind third or higher generations on the overall use of mental health services, the former has a higher probability of using any conventional services than the latter. What factor prompted second-generation Chinese Americans to prefer using conventional services for their mental health problems? This question merits future research. Third and higher generations were not significantly different from their white counterparts in mental health service use pattern. To summarize, the probability of Chinese Americans using mental health services, overall and by sector, increases from lower (more recent) to higher (earlier) generations, catching up with the non-Hispanic white Americans at the third generation. English proficiency proved to be a barrier for Chinese Americans to conventional Western medical services, but not CAM therapies, which was consistent with previous study results (Ying and Miller 1992).

Cultural factors play important roles in how individuals seek help for health problems (Kleinman, Eisenberg, and Good 1978; Rogler and Cortes 1993). To understand the underuse of mental health services among Asian populations andthe acculturation effects, scholars have offeredthree general explanations: (1) differences between Asian and Western beliefs in the etiology of mental illness, (2) stigma and shame in Asian cultures, and (3) conflicts of Asian values and Western psychotherapy treatment (Chen and Mak 2008; Lau and Takeuchi 2001). Chinese culture, like other collectivist Asian cultures, regards mental health problems as personal failure andattributes them to internal, personal causes (Tabora and Flaskerud 1994). It opposes “the values associated with Western psychotherapy . . . [including] open verbal communication, exploration of intrapsychic conflicts, and a focus on the individual” (Leong and Lau 2001:204). Therefore, people with strong roots in Chinese culture may choose to seek help from alternative health services, in lieu of conventional Western mental health care, because the former is more compatible with their cultural beliefs.

A practical implication of the findings is that mental health professionals must pay attention to the sociocultural background of those seeking help, in both identifying causes and providing treatment suggestions. In addition, public health professionals, policy makers, and society need to be aware of the various barriers Chinese Americans and other ethnic minorities experience. The provision of services in ethnic languages, education to the vulnerable populations, greater outreach to ethnic minorities, and making services more affordable are all essential efforts to be taken.

Furthermore, an increasing body of literature has shown a pattern of mental health service use among Asian Americans, that although the overall use rate is low, those who do use mental health services tend to be highly disturbed in terms of psychiatric conditions (Chu and Sue 2011). For health professionals and researchers, it is of great importance to better understand this pattern, as well as its implications. What are the factors and pathways that hold Asian Americans from seeking helps for their mental health problems until the conditions are severe? Does this low-prevalence, high-severity pattern of service use apply to different sectors of services? These questions warrant future studies.

Several limitations should be noted. First, the measures this study relied on are primarily self-reported, which is subject to recall and response biases. Future study should consider taking into account social desirability response effects by the respondents. Second, the CPES data are cross-sectional, hence the lack of power to establish causation between the predictors (e.g., English proficiency) and the outcome variable, the patterns of mental health service use. This topic merits future longitudinal research. Third, the acculturation-related factors (i.e., generational status and English proficiency) are only proxies rather thandirect measurement of acculturation. As previously discussed in this article, the operationalization of acculturation is still under debate. To advance our knowledge of the effects of such a process on the health and help-seeking behaviors of ethnic minorities and immigrants, further refinement of measurements is necessary. Furthermore, generational status measurements (i.e., nativity status of participants and parents) were not included in the NCS-R, hence the lack of differentiation on generational status within non-Hispanic whites; how generational status affected the help seeking behaviors of this subsample remains to be reexamined.

In addition, the sample size of Chinese Americans made it difficult to examine those who used both CAM and conventional Western medicine as a separate outcome group, which would further advance our understanding on the use patterns of CAM. Last but not least, although this study has shed light on some of the factors underlying the patterns of mental health service use among Chinese Americans, further research is still needed to understand the help-seeking trajectories and the underlying factors. For example, more in-depth analyses are needed to understand the length of use and satisfaction with various types of mental health services among Chinese Americans. Also, the type of health insurance and income may also influence CAM use and deserve attention in future research. Furthermore, the underlying roles of cultural factors on the perception of needs and their influence on help-seeking behaviors are hypothesized but have yet to be examined in empirical findings (Nguyen and Bornheimer 2014).

Footnotes

ORCID iD

Lin Zhu

Notes

References

Abe-Kim

Jennifer

Takeuchi

David T.

Hong

Seunghye

Zane

Nolan

Sue

Stanley

Spencer

Michael S.

Appel

Hoa

Nicdao

Ethel

Alegría

Margarita

. 2007. “Use of Mental Health-Related Services among Immigrant and U.S.-born Asian Americans: Results from the National Latino and Asian American Study.” American Journal of Public Health 97(1):91–98.

Abraído-Lanza

Ana F.

Armbrister

Adria N.

Flórez

Karen R.

Aguirre

Alejandra N.

2006. “Toward a Theory-driven Model of Acculturation in Public Health Research.” American Journal of Public Health 96(8):1342–46.

Alegria

Margarita

Jackson

James S.

Kessler

Ronald C.

Takeuchi

David

. 2016. “Collaborative Psychiatric Epidemiology Surveys (CPES), 2001–2003.” Retrieved December 25, 2017 (http://www.icpsr.umich.edu/icpsrweb/ICPSR/studies/20240).

Astin

John A.

1998. “Why Patients Use Alternative Medicine: Results of a National Study.” JAMA 279(19):1548–53.

Atkinson

Donald R.

Lowe

Susana

Matthews

Linda

. 1995. “Asian-American Acculturation, Gender, and Willingness to Seek Counseling.” Journal of Multicultural Counseling and Development 23(3):130–38.

Barnes

Patricia M.

Bloom

Barbara

Nahin

Richard L.

2008. “Complementary and Alternative Medicine Use among Adults and Children: United States, 2007.” Hyattsville, MD: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics. Retrieved August 5, 2015 (http://purl.access.gpo.gov/GPO/LPS108504).

Barnes

Patricia M.

Powell-Griner

Eve

McFann

Kim

Nahin

Richard L.

2004. “Complementary and Alternative Medicine Use among Adults: United States, 2002.” Seminars in Integrative Medicine 2(2):54–71.

Bauer

Amy M.

Chen

Chih-Nan

Alegría

Margarita

. 2010. “English Language Proficiency and Mental Health Service Use among Latino and Asian Americans with Mental Disorders.” Medical Care 48(12):1097–1104.

Berry

John W.

1990. “Acculturation and Adaptation: A General Framework.” Pp. 90–102 in Mental Health of Immigrants and Refugees, edited by Holtzman

W. H.

Bornemann

T. H.

Austin, TX: Hogg Foundation for Mental Health.

10.

Berry

John W.

2003. “Conceptual Approaches to Acculturation.” Pp. 17–37 in Acculturation: Advances in Theory, Measurement, and Applied Research, editedby Chun

K. M.

Balls Organista

Marín

Washington, DC: American Psychological Association.

11.

Berthold

S. Megan

Wong

E. C.

Schell

T. L.

Marshall

G. N.

Elliott

M. N.

Takeuchi

Hambarsoomians

2007. “U.S. Cambodian Refugees’ Use of Complementary and Alternative Medicine for Mental Health Problems.” Psychiatric Services 58(9):1212–18.

12.

Bishop

Felicity L.

Lewith

G. T.

2010. “Who Uses CAM? A Narrative Review of Demographic Characteristics and Health Factors Associated with CAM Use.” Evidence-Based Complementary and Alternative Medicine 7(1):11–28.

13.

Bodeker

Gerard

Kronenberg

Fredi

. 2002. “A Public Health Agenda for Traditional, Complementary, and Alternative Medicine.” American Journal of Public Health 92(10):1582–91.

14.

Carrasquillo

A. I.

Shea

2000. “Health Insurance Coverage of Immigrants Living in the United States: Differences by Citizenship Status and Country of Origin.” American Journal of Public Health 90(6):917–23.

15.

Chen

Sylvia Xiaohua

Mak

Winnie W. S.

2008. “Seeking Professional Help: Etiology Beliefs about Mental Illness across Cultures.” Journal of Counseling Psychology 55(4):442–50.

16.

Chen

Zhi

Zheng

Shu

X. O.

2008. “The Use of Complementary and Alternative Medicine among Chinese Women with Breast Cancer.” Journal of Alternative and Complementary Medicine 14(8):1049–55.

17.

Chu

Joyce P.

Hsieh

Kun-Yueh

Tokars

Di Ann

. 2011. “Help-seeking Tendencies in Asian Americans with Suicidal Ideation and Attempts.” Asian American Journal of Psychology 2(1):25–38.

18.

Chu

Joyce

Sue

Stanley

. 2011. “Asian American Mental Health: What We Know and What We Don’t Know.” Online Readings in Psychology and Culture 3(1). Retrieved (http://scholarworks.gvsu.edu/orpc/vol3/iss1/4).

19.

Clark

Lauren

Hofsess

Lisa

. 1998. “Acculturation.” Pp. 37–59 in Handbook of Immigrant Health, edited by Loue

New York: Springer.

20.

Clarke

Tainya C.

Black

Lindsey I.

Stussman

Barbara J.

Barnes

Patricia M.

Nahin

Richard L.

2015. “Trends in the Use of Complementary Health Approaches among Adults: United States, 2002–2012.” National Health Statistics Reports (79):1–16.

21.

Clough

Juliana

Lee

Sunmin

Chae

David H.

2013. “Barriers to Health Care among Asian Immigrants in the United States: A Traditional Review.” Journal of Health Care for the Poor and Underserved 24(1):384–403.

22.

Coulter

Ian D.

Willis

Evan M.

2004. “The Rise and Rise of Complementary and Alternative Medicine: A Sociological Perspective.” Medical Journal of Australia 180(11):587–89.

23.

Druss

Benjamin G.

Rosenheck

Robert A.

1999. “Association between Use of Unconventional Therapies and Conventional Medical Services.” JAMA 282(7):651–56.

24.

Eisenberg

David M.

Davis

R. B.

Ettner

S. L.

Appel

Wilkey

Van Rompay

Kessler

R. C.

1998. “Trends in Alternative Medicine Use in the United States, 1990–1997: Results of a Follow-up National Survey.” JAMA 280(18):1569–75.

25.

Eisenberg

David M.

Kessler

R. C.

Foster

Norlock

F. E.

Calkins

D. R.

Delbanco

T. L.

1993. “Unconventional Medicine in the United States—Prevalence, Costs, and Patterns of Use.” New England Journal of Medicine 328(4):246–52.

26.

Fang

Lin

Schinke

Steven P.

2007. “Complementary Alternative Medicine Use among Chinese Americans: Findings from a Community Mental Health Service Population.” Psychiatric Services 58(3):402–4.

27.

Dongfeng

Shenshen

Sang

Feng

Nicole

Liang

Lam

Xian Li Christopher

Yan

Lijing L.

Zhou

Yujie

Yangfeng

. 2002. “Prevalence, Awareness, Treatment, and Control of Hypertension in China.” Hypertension 40(6):920–27.

28.

Harris

Katherine M.

Edlund

Mark J.

Larson

Sharon

. 2005. “Racial and Ethnic Differences in the Mental Health Problems and Use of Mental Health Care.” Medical Care 43(8):775–84.

29.

Heeringa

Steven G.

Wagner

Torres

Duan

Adams

Berglund

2004. “Sample Designs and Sampling Methods for the Collaborative Psychiatric Epidemiology Studies (CPES).” International Journal of Methods in Psychiatric Research 13(4):221–40.

30.

Hoeffel

Elizabeth M.

Rastogi

Sonya

Ouk Kim

Myoung

Shahid

Hasan

. 2012. “The Asian Population: 2010.” 2010 Census Briefs. Retrieved October 13, 2014 (http://www.census.gov/prod/cen2010/briefs/c2010br-11.pdf).

31.

Hsiao

An-Fu, M.

Wong

Goldstein

M. S.

Becerra

L. S.

Cheng

E. M.

Wenger

N. S.

2006. “Complementary and Alternative Medicine Use among Asian-American Subgroups: Prevalence, Predictors, and Lack of Relationship to Acculturation and Access to Conventional Health Care.” Journal of Alternative and Complementary Medicine 12(10):1003–10.

32.

Izzo

Angelo A.

Ernst

Edzard

. 2012. “Interactions between Herbal Medicines and Prescribed Drugs.” Drugs 69(13):1777–98.

33.

Jackson

James S.

Torres

Caldwell

C. H.

Neighbors

H. W.

Nesse

R. M.

Taylor

R. J.

Trierweiler

S. J.

Williams

D. R.

2004. “The National Survey of American Life: A Study of Racial, Ethnic and Cultural Influences on Mental Disorders and Mental Health.” International Journal of Methods in Psychiatric Research 13(4):196–207.

34.

Jang

Michael

Lee

Evelyn

Woo

Kent

. 1998. “Income, Language, and Citizenship Status: Factors Affecting the Health Care Access and Utilization of Chinese Americans.” Health & Social Work 23(2):136–45.

35.

Kessler

Ronald C.

Soukup

Davis

R. B.

Foster

D. F.

Wilkey

S. A.

Van Rompay

M. I.

Eisenberg

D. M.

2001. “The Use of Complementary and Alternative Therapies to Treat Anxiety and Depression in the United States.” American Journal of Psychiatry 158(2):289–94.

36.

Kim

Giyeon

Aguado Loi

C. X.

Chiriboga

D. A.

Jang

Parmelee

Allen

R. S.

2011. “Limited English Proficiency as a Barrier to Mental Health Service Use: A Study of Latino and Asian Immigrants with Psychiatric Disorders.” Journal of Psychiatric Research 45(1):104–10.

37.

Kim

Wooksoo

Keefe

Robert H.

2010. “Barriers to Healthcare among Asian Americans.” Social Work in Public Health 25(3–4):286–95.

38.

Kleinman

Arthur M.

Eisenberg

Leon

Good

Byron

. 1978. “Culture, Illness, and Care: Clinical Lessons from Anthropologic and Cross-cultural Research.” Annals of Internal Medicine 88(2):251–58.

39.

Kung

Winnie W.

2003. “Chinese Americans’ Help Seeking for Emotional Distress.” Social Service Review 77(1):110–34.

40.

Kung

Winnie W.

2004. “Cultural and Practical Barriers to Seeking Mental Health Treatment for Chinese Americans.” Journal of Community Psychology 32(1):27–43.

41.

Lau

Anna

Takeuchi

David

. 2001. “Cultural Factors in Help-seeking for Child Behavior Problems: Value Orientation, Affective Responding, and Severity Appraisals among Chinese American Parents.” Journal of Community Psychology 29(6):675–92.

42.

Lee

Jennifer H.

Goldstein

Michael S.

Brown

E. Richard

Ballard-Barbash

Rachel

. 2010. “How Does Acculturation Affect the Use of Complementary and Alternative Medicine Providers among Mexican- and Asian-Americans?” Journal of Immigrant and Minority Health 12(3):302–9.

43.

Lee

Minsun

Takeuchi

David

Gellis

Zvi

Kendall

Philip

Zhu

Lin

Zhao

Shanyang

Grace X.

2017. “The Impact of Perceived Need and Relational Factors on Mental Health Service Use among Generations of Asian Americans.” Journal of Community Health 42(4):688–700.

44.

Le Meyer

Oanh

Zane

Nolan

Cho

Young Il

Takeuchi

David T.

2009. “Use of Specialty Mental Health Services by Asian Americans with Psychiatric Disorders.” Journal of Consulting and Clinical Psychology 77(5):1000–1005.

45.

Leong

Frederick T. L.

1986. “Counseling and Psychotherapy with Asian-Americans: Review of the Literature.” Journal of Counseling Psychology 33(2):196–206.

46.

Leong

Frederick T. L.

1994. “Asian Americans’ Differential Patterns of Utilization of Inpatient and Outpatient Public Mental Health Services in Hawaii.” Journal of Community Psychology 22(2):82–96.

47.

Leong

Frederick T. L.

Lau

Anna S. L.

2001. “Barriers to Providing Effective Mental Health Services to Asian Americans.” Mental Health Services Research 3(4):201–14.

48.

Jing

Hummer

Robert A.

2015. “The Relationship between Duration of U.S. Residence, Educational Attainment, and Adult Health among Asian Immigrants.” Population Research and Policy Review 34(1):49–76.

49.

Long

J. Scott

Freese

Jeremy

. 2014. Regression Models for Categorical Dependent Variables Using Stata. 3rd ed. College Station, TX: Stata Press.

50.

Mackenzie

Elizabeth R.

Taylor

Lynne

Bloom

Bernard S.

Hufford

David J.

Johnson

Jerry C.

2003. “Ethnic Minority Use of Complementary and Alternative Medicine (CAM): A National Probability Survey of CAM Utilizers.” Alternative Therapies in Health and Medicine 9(4):50–56.

51.

Grace Xueqin

. 1999. “Between Two Worlds: The Use of Traditional and Western Health Services by Chinese Immigrants.” Journal of Community Health 24(6):421–37.

52.

Massey

Douglas S.

1981. “Dimensions of the New Immigration to the United States and the Prospects for Assimilation.” Annual Review of Sociology 7:57–85.

53.

Mayberry

Robert M.

Mili

Fatima

Ofili

Elizabeth

. 2000. “Racial and Ethnic Differences in Access to Medical Care.” Medical Care Research and Review 57(4 Suppl.):108–45.

54.

Murray

Raymond H.

Rubel

Arthur J.

1992. “Physicians and Healers—Unwitting Partners in Health Care.” New England Journal of Medicine 326(1):61–64.

55.

Nguyen

Duy

. 2011. “Acculturation and Perceived Mental Health Need among Older Asian Immigrants.” Journal of Behavioral Health Services & Research 38(4):526–33.

56.

Nguyen

Duy

Bornheimer

Lindsay A.

2014. “Mental Health Service Use Types among Asian Americans with a Psychiatric Disorder: Considerations of Culture and Need.” Journal of Behavioral Health Services & Research 41(4):520–28.

57.

Oldendick

Robert

Coker

A. L.

Wieland

Raymond

J. I.

Probst

J. C.

Schell

B. J.

Stoskopf

C. H.

2000. “Population-based Survey of Complementary and Alternative Medicine Usage, Patient Satisfaction, and Physician Involvement.” Southern Medical Journal 93(4):375–81.

58.

Ong

Chi-Keong

Bodeker

Gerard

Grundy

Chris

Burford

Gemma

Shein

Kin

. 2005. WHO Global Atlas of Traditional, Complementary and Alternative Medicine. Geneva, Switzerland: World Health Organization.

59.

Park

Hye-Lim

Lee

Hun-Soo

Shin

Byung-Cheul

Liu

Jian-Ping

Shang

Qinghua

Yamashita

Hitoshi

Lim.

Byungmook

2012. “Traditional Medicine in China, Korea, and Japan: A Brief Introduction and Comparison.” Evidence-Based Complementary and Alternative Medicine 2012:429103. Retrieved September 27, 2017 (https://www.hindawi.com/journals/ecam/2012/429103/abs/).

60.

Park

Robert E.

1928. “Human Migration and the Marginal Man.” American Journal of Sociology 33(6):881–93.

61.

Pescosolido

Bernice A.

Wright

Eric R.

Alegría

Margarita

Vera

Mildred

. 1998. “Social Networks and Patterns of Use among the Poor with Mental Health Problems in Puerto Rico.” Medical Care 36(7):1057–72.

62.

Rafferty

Ann P.

McGee

Harry B.

Miller

Corinne E.

Reyes

Michele

. 2002. “Prevalence of Complementary and Alternative Medicine Use: State-specific Estimates from the 2001 Behavioral Risk Factor Surveillance System.” American Journal of Public Health 92(10):1598–1600.

63.

Rao

J.N.K.

Scott

A. J.

1984. “On Chi-squared Tests for Multiway Contingency Tables with Cell Proportions Estimated from Survey Data.” Annals of Statistics 12(1):46–60.

64.

Rogler

Lloyd H.

Cortes

Dharmar E.

1993. “Help-seeking Pathways: A Unifying Concept in Mental Health Care.” American Journal of Psychiatry 150(4):554–61.

65.

Silverstein

Daniel D.

Spiegel

Allen D.

2001. “Are Physicians Aware of the Risks of Alternative Medicine?” Journal of Community Health 26(3):159–74.

66.

Spencer

Michael S.

Chen

Juan

. 2004. “Effect of Discrimination on Mental Health Service Utilization among Chinese Americans.” American Journal of Public Health 94(5):809–14.

67.

Spencer

Michael S.

Chen

Juan

Gee

Gilbert C.

Fabian

Cathryn G.

Takeuchi

David T.

2010. “Discrimination and Mental Health-related Service Use in a National Study of Asian Americans.” American Journal of Public Health 100(12):2410–17.

68.

Stephen

E. H.

Foote

Hendershot

G. E.

Schoenborn

C. A.

1994. “Health of the Foreign-born Population: United States, 1989–90.” Advance Data (241):1–12.

69.

Dejun

Lifeng

. 2011. “Trends in the Use of Complementary and Alternative Medicine in the United States: 2002–2007.” Journal of Health Care for the Poor and Underserved 22(1):296–310.

70.

Sue

Stanley

Cheng

Janice Ka Yan

Saad

Carmel S.

Chu

Joyce P.

2012. “Asian American Mental Health: A Call to Action.” American Psychologist 67(7):532–44.

71.

Tabora

Betty

Flaskerud

Jacquelyn H.

1994. “Depression among Chinese Americans: A Review of the Literature.” Issues in Mental Health Nursing 15(6):569–84.

72.

Tata

Shiraz Piroshaw

Leong

Frederick T. L.

1994. “Individualism-collectivism, Social-network Orientation, and Acculturation as Predictors of Attitudes toward Seeking Professional Psychological Help among Chinese Americans.” Journal of Counseling Psychology 41(3):280–87.

73.

Tindle

Hilary A.

Davis

Roger B.

Phillips

Russell S.

Eisenberg

David M.

2005. “Trends in the Complementary and Alternative Medicine by U.S. Adults: 1997–2002.” Alternative Therapies in Health and Medicine 11(1):42–49.

74.

Tiwari

Suresh K.

Wang

JianLi

. 2008. “Ethnic Differences in Mental Health Service Use among White, Chinese, South Asian and South East Asian Populations Living in Canada.” Social Psychiatry and Psychiatric Epidemiology 43(11):866–71.

75.

Upchurch

Dawn M.

Chyu

Greendale

G. A.

Utts

Bair

Y. A.

Zhang

Gold

E. B.

2007. “Complementary and Alternative Medicine Use among American Women: Findings from the National Health Interview Survey, 2002.” Journal of Women’s Health 16(1):102–13.

76.

U.S. Department of Health and Human Services. 2001. “Mental Health: Culture, Race, and Ethnicity: A Supplement to Mental Health: A Report of the Surgeon General.” Rockville, MD: U.S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration, Center for Mental Health Services. Retrieved April 17, 2016 (http://www.ncbi.nlm.nih.gov/books/NBK44243/).

77.

Vincent

Charles

Furnham

Adrian

. 1996. “Why Do Patients Turn to Complementary Medicine? An Empirical Study.” British Journal of Clinical Psychology 35(1):37–48.

78.

Wang

Chi-Chuan

Kennedy

Jae

Chung-Hsuen

. 2015. “Alternative Therapies as a Substitute for Costly Prescription Medications: Results from the 2011 National Health Interview Survey.” Clinical Therapeutics 37(5):1022–30.

79.

Wang

Philip S.

Lane

Olfson

Pincus

H. A.

Wells

K. B.

Kessler

R. C.

2005. “Twelve-month Use of Mental Health Services in the United States: Results from the National Comorbidity Survey Replication.” Archives of General Psychiatry 62(6):629–40.

80.

Waters

Mary C.

Jiménez

Tomás R.

2005. “Assessing Immigrant Assimilation: New Empirical and Theoretical Challenges.” Annual Review of Sociology 31:105–25.

81.

Wolsko

Peter M.

Eisenberg

David M.

Davis

Roger B.

Ettner

Susan L.

Phillips

Russell S.

2002. “Insurance Coverage, Medical Conditions, and Visits to Alternative Medicine Providers: Results of a National Survey.” Archives of Internal Medicine 162(3):281–87.

82.

Judy

Yang

Yue

. 2009. “Traditional Chinese Medicine in the Chinese Health Care System.” Health Policy 90(2):133–39.

83.

Yang

Lawrence H.

Corsini-Munt

Serena

Link

Bruce G.

Phelan

Jo C.

2009. “Beliefs in Traditional Chinese Medicine Efficacy among Chinese Americans: Implications for Mental Health Service Utilization.” Journal of Nervous and Mental Disease 197(3):207–10.

84.

Ying

Yu-Wen

Miller

Leonard S.

1992. “Help-Seeking Behavior and Attitude of Chinese Americans Regarding Psychological Problems.” American Journal of Community Psychology 20(4):549–56.

85.

Zhu

Lin

. 2017. “Depression Risks and Correlates among Different Generations of Chinese Americans: The Effects of Relationships with Friends and Relatives.” Social Sciences 6(2):56.

86.

Zhu

Lin

. 2018. “Depression Symptom Patterns and Social Correlates among Chinese Americans.” Brain Sciences, 8(1):16.