Presenting Concerns and Psychotherapy Utilization in University Counseling Centers: What Do We Know about Asian American and International Asian Students?

ABSTRACT

Individuals from Asian American (AA) and international Asian (IA) backgrounds are often lumped together in mental health research, which may obscure important differences in mental health presentation and need. While previous research has examined differences amongst AA and IA students seeking therapy at a single university counseling center, our study sought to better generalize these findings by utilizing a national data set to examine differences in treatment utilization, presenting concerns, overall distress, and improvement in treatment. Our results showed that IA students utilize therapy at lower rates than AA students. However, both groups attended appointments at roughly the same rate. Distress at intake was significantly higher for IA students across most domains measured. Finally, both groups improved in treatment at similar rates, with some differences observed between IA undergraduate and graduate students. We discuss implications for college counseling professionals and urge further research in this domain.

KEYWORDS:

• Asian
• Asian American
• international Asian
• university counseling center
• treatment utilization

Numerous studies have shown that colleges and universities are experiencing an increasing demand for mental health services (Center for Collegiate Mental Health, 2019; Oswalt et al., 2020; Xiao et al., 2017). Common concerns that contribute to college students’ psychological distress include generalized anxiety, social anxiety, depression, family distress, and academic pressures (Oswalt et al., 2020). These concerns can be severe, with about eight percent of those who seek help in college counseling centers requiring extensive psychiatric help (Oswalt et al., 2020). Unattended mental health concerns can negatively impact school performance, leading to increases in anxiety, depression, mental health stigma, and suicidal thoughts and actions (Hirsch, Rabon, Reynolds, Barton, & Chang, 2019). Accordingly, college counseling centers are implementing different models of treatment to meet these demands and serve their students (Brunner, Wallace, Reymann, Sellers, & McCabe, 2014).

Mental health concerns can be particularly problematic for students with racial or ethnic minority (REM) status. Research suggests that REM students’ rates of mental health diagnoses are impacted by cultural factors such as stigma surrounding mental health and help-seeking (Kim, Oh, & Mumbauer, 2019; Liu, Stevens, Wong, Yasui, & Chen, 2019; Sanchez, Ybarra, Chapa, & Martinez, 2016). Because of the stigma associated with help-seeking behaviors, REM students often have difficulty recognizing mental health problems, which may contribute to lower treatment utilization (Huang & Zane, 2016). College students today are more ethnically diverse than ever before, with about one-third having REM status (Brunner et al., 2014). As diversity on college campuses continues to increase, it is vital that college counseling center professionals further develop their understanding of the worldviews and experiences of REM students seeking mental health services.

Among REM students, those with Asian heritage tend to underutilize psychotherapy at significantly higher rates (Miller, Yang, Hui, Choi, & Lim, 2011; Ruzek, Nguyen, & Herzog, 2011). Many Asian-Americans (AA) and people from Asian and South Asian countries report that stigma associated with mental health concerns is a significant barrier to seeking treatment (Kim, Atkinson, & Umemoto, 2001; Kudva et al., 2020). Traditional Asian cultural values of modesty, discretion, humility, and conservatism of interpersonal expression may lead students of Asian descent to conceal mental health concerns in order to avoid standing out (Kim et al., 2001). Consequently, individuals from Asian and AA backgrounds tend to seek out therapy with a greater level of initial distress and drop out of therapy prematurely (Kim, Park, La, Chang, & Zane, 2015; Stokes et al., 2021).

Even among those with a similar ethnic heritage, considerable cultural differences exist that are dependent on the environment in which one is raised. For example, international Asian (IA) students often face more challenges adapting to an American university system than AA students, who may be more familiar with U.S. culture (Heggins & Jackson, 2003). Acculturation is the process of assimilating to a new culture, typically the dominant one. Researchers have found that, while both AA and IA students report that acculturative stress is a significant concern, AA students have higher levels of acculturation than IA students (Hwang & Ting, 2008; Zhou, Zhang, & Stodolska, 2018). Cultural differences, lack of social support, and language proficiency represent distinct stressors that are often faced by IA students, and these stressors can lead to international students seeking help at lower rates than native students (Kim et al., 2019).

Despite the wide variety of cultural practices, norms, and backgrounds for AA and IA students, studies examining the experiences of students in college campus settings often combine the two, using labels such as “Asian” or “Asian American” (e.g., Iwamoto & Liu, 2010). Such simplistic combination of ethnocultural groups has been termed ethnic gloss (Trimble & Bhadra, 2013) in psychological research. The process of glossing AA and IA students lumps two distinct cultural groups together in a manner that frequently conceals or discounts the differences that naturally exist across these groups (Stokes et al., 2021). For example, Asia consists of many countries, and within each Asian country there are numerous groups of people with distinct ethnic/cultural backgrounds (Chu & Sue, 2011; Yan & Berliner, 2013). The idea that those with lived experiences in Asian countries outside of the U.S. will have the same experiences as those raised in the U.S. is problematic. Clearly, different cultural upbringing and lived experiences impact the way people view mental health, treatment of mental health concerns, and views toward traditional psychotherapy (Yoon & Jepsen, 2008). For example, Wong, Wang, and Maffini (2014) found that IA students who strongly endorsed humility and emotional self-control, values that tend to be traditional in many Asian cultures, were more likely to have negative attitudes toward seeking professional psychological help (Wong et al., 2014). Researchers have found that stronger identification with ethnic culture contributes to a lower likelihood of accessing mental health services, with these students instead turning to family and friends for help (Kahn & Williams, 2003; Soorkia, Snelgar, & Swami, 2011). IA students tend to seek out counseling services when academic distress is high (e.g., Zhang & Dixon, 2003), and they are also more likely to view counseling for personal concerns as a last resort, suitable only for severe mental illness (Li, Wong, & Toth, 2013).

Stokes et al. (2021) examined the similarities and differences of IA and AA students’ experiences in seeking treatment in a college counseling center. Specifically, they examined differences between the two groups in terms of treatment utilization, presenting concerns, severity of concerns at intake, and treatment outcomes in one college counseling center. They found that IA students were slightly less likely to attend therapy than AA and European American (EA) students, and that there were no significant differences between the groups in terms of their rates of treatment discontinuance. They also found significant differences in the presenting concerns of the three groups. Finally, the authors found that AA and IA students had less improvement in treatment than EA students.

Due to limitations in their study, Stokes et al. (2021) noted the need for further exploration to better understand similarities and differences among AA and IA populations. Notably, their study explored the experiences of students at one, private, religiously-oriented university. Consequently, the generalizability of their findings is limited, and further research is necessary to understand the broader experiences of AA and IA students using a more diverse sample.

No published studies to date have taken the next steps suggested by Stokes et al. (2021) to examine the similarities and differences of AA and IA university students using a broader sample. The purpose of the present study was to follow up and expand upon Stokes et al. (2021) using a national archival dataset from The Center for Collegiate Mental Health (CCMH). In this current study, we compared AA and IA students who attended counseling in university counseling centers in terms of the number of sessions used, presenting concerns, level of distress at intake, and treatment outcomes. Based on the results of Stokes et al. (2021), we hypothesized that IA students would show lower numbers of attended sessions in university counseling centers across the nation than AA students. We also hypothesized that IA students would show a higher severity of initial presenting concerns and overall distress at intake than their AA counterparts. Finally, we hypothesized that IA students would demonstrate similar or slower treatment improvement rates compared to AA students.

Method

Procedure

Data for the study were obtained from the Center for Collegiate Mental Health (CCMH), a multidisciplinary practice-research network of over 400 college counseling centers and other organizations dedicated to providing current information about the mental health of college students. The data were collected between the years 2015–2017 from 163 unique college counseling centers across the United States. Data collected more recently than 2017 were not obtained due to CCMH researcher access restriction policies.

Previous to this study, clients of CCMH-affiliated counseling centers signed consent forms and completed demographic and psychological distress measures (described later). The information from these tools was de-identified to protect the confidentiality of student participants. Furthermore, the institutions involved in providing data were de-identified as well. Both Institutional Review Board (IRB) approval and informed consent were previously obtained from each institution contributing to the data. IRB approval for this study was deemed exempt due to the archival nature of the data used.

Data set

The initial data set (N = 364,025) contained demographic and outcome variables for 301,345 unique students. At the time of their first course of therapy, 41.4% of students identified as White, 6.0% as Hispanic, 5.7% as African American, 5.4% as Asian American/Asian, 3.2% as Multi-racial, 1.1% as self-identify, 0.3% as American Indian or Alaskan Native (AIAN), and 0.1% as Native Hawaiian or Pacific Islander (NHPI), with 36.7% missing data. Participants identified as 55.9% non-International Student and 3.6% International Student, with 40.6% missing data. In terms of gender identity, 40.3% identified as woman, 21.5% as man, 0.7% as self-identify, and 0.3% transgender, with 37.1% missing data. The average age was 22.2 (SD = 4.7) with 3.6% missing data.

Participants

The entire CCMH data set included 301,345 unique students. Of these students, 11,905 (5.4% of the total sample) identified as Asian American/Asian. For the present study, we only included participants if they identified as Asian American/Asian and provided international student status through the Standardized Data Set (SDS). Although “self-identify” was an option for reporting racial and ethnic identity, no free-response descriptions were provided in the data received from CCMH; thus, we did not include participants who marked “self-identify” in our analyses. We also only included participants if they identified as either an undergraduate or graduate student, and if they had completed mental health outcome data (both first and last session) through the Counseling Center Assessment of Psychological Symptoms-34 (CCAPS-34). In terms of gender identity, 63.0% of individuals in our data set identified as woman, 35.4% as man, 0.4% as self-identify, and 0.7% transgender, with 0.49% missing data. The average age was 22.53 (SD = 3.92) with 0.02% missing data (See Table 1 for more detailed demographic information). Additionally, as students may have been included in the initial data set more than once (owing to multiple courses of therapy being reported), only instances of the first course of therapy, and only those who engaged exclusively in individual therapy, were included.

Table 1. Asian/Asian American Demographics (N = 11,905).

Demographic Category	Undergraduate		Graduate
	Domestic (n = 6,397)	Int’l (n = 2,355)	Domestic (n = 1,055)	Int’l (n = 2,098)
	n (%)	n (%)	n (%)	n (%)
Gender Identity
Woman	4,134 (64.6)	1,349 (57.3)	742 (70.3)	1,278 (60.9)
Man	2,133 (33.3)	976 (41.4)	299 (28.3)	809 (38.6)
Transgender	36 (0.6)	3 (0.1)	2 (0.2)	3 (0.1)
Other/Missing	94 (1.5)	27 (1.1)	12 (1.2)	8 (0.3)
Sexual Orientation
Heterosexual/Straight	5,200 (81.3)	1,909 (81.1)	884 (83.8)	1,704 (81.2)
Lesbian	87 (1.4)	20 (0.8)	13 (1.2)	16 (0.8)
Gay	114 (1.8)	49 (2.1)	25 (2.4)	54 (2.6)
Bisexual	410 (6.4)	137 (5.8)	48 (4.5)	70 (3.3)
Questioning	200 (3.1)	79 (3.4)	24 (2.3)	62 (3.0)
Other/Missing	386 (6.1)	161 (6.8)	61 (5.8)	192 (9.2)
Country of Origin
Bangladesh	72 (1.1)	26 (1.1)	16 (1.5)	56 (2.7)
China	457 (7.1)	1,117 (47.4)	62 (5.9)	825 (39.3)
Hong Kong	21 (0.3)	29 (1.2)	5 (0.5)	11 (0.5)
India	454 (7.1)	308 (13.1)	97 (9.2)	659 (31.4)
Indonesia	20 (0.3)	44 (1.9)	2 (0.2)	14 (0.7)
Japan	38 (0.6)	38 (1.6)	8 (0.8)	16 (0.8)
South Korea	232 (3.6)	226 (9.6)	46 (4.4)	125 (6)
Malaysia	6 (0.1)	71 (3.0)	0 (0.0)	18 (0.9)
Nepal	34 (0.5)	58 (2.5)	4 (0.4)	43 (2.0)
Pakistan	146 (2.3)	42 (1.8)	32 (3.0)	37 (1.8)
Philippines	237 (3.7)	16 (0.7)	22 (2.1)	13 (0.6)
Taiwan	43 (0.7)	43 (1.8)	11 (1.0)	65 (3.1)
United States	3,587 (56.1)	13 (0.6)	590 (55.9)	2 (0.1)
Vietnam	148 (2.3)	66 (2.8)	17 (1.6)	25 (1.2)
Other/Missing	902 (9.4)	258 (11.0)	143 (13.7)	189 (9.0)
Relationship Status
Single	4,274 (66.8)	1,620 (68.8)	548 (51.9)	1,144 (54.5)
Serious Dating/Committed	1,799 (28.1)	560 (23.8)	363 (34.4)	507 (24.2)
Married	71 (1.1)	26 (1.1)	98 (9.3)	274 (13.1)
Divorced/Separated	22 (0.4)	16 (0.7)	11 (1.1)	54 (2.6)
Other/Missing	231 (3.7)	133 (5.7)	35 (3.4)	120 (5.6)

Measures

Standardized Data Set (SDS)

The SDS is a comprehensive set of exclusively demographic questions developed by CCMH and used by participating counseling centers, typically upon intake (Center for Collegiate Mental Health, 2019). The SDS includes demographic questions about gender, age, ethnicity, religion, sexual identity, etc., as well as presenting concerns, previous counseling experience, and mental health history.

Counseling Center Assessment of Psychological Symptoms-34 (CCAPS-34)

The CCAPS-34 is a multidimensional assessment of psychological symptoms intended for routine use in college populations (Locke et al., 2012). It is an abbreviated version of the Counseling Center Assessment of Psychological Symptoms-62 (CCAPS-62; Locke et al., 2011), containing a subset of the same items but revised for more frequent administration. Items are rated on a 5-point Likert-scale, where zero indicates Not at all like me and four indicates Extremely like me. The national development of the CCAPS-62 – assessed with a sample of 22,060 participants – indicated acceptable reliability, with internal consistency alphas ranging from .78-.91 across indices (depression = .91, generalized anxiety = .85, and social anxiety = .82), and test-retest reliability ranging from .78 – .92 and .76 – .91 between 1-week and 2-week test-retest correlations respectively. It also demonstrates good convergent validity, with strong and significant Pearson product-moment correlations shown between every index and referent measures.

The CCAPS-34 likewise indicates strong convergent validity, with strong and significant Pearson product-moment correlations between indices with the same referent measures (r = .52 – .78). It similarly has acceptable reliability, with internal consistency alphas ranging from .76 – .89 across indices (depression = .89, generalized anxiety = .82, and social anxiety = .8), and test-retest reliability ranging from .79 – .87 and .74 – .86 between 1-week and 2-week test-retest correlations respectively. Bartholomew (2021) provides evidence that the CCAPS-34 is valid for use with international student clients.

The CCAPS-34 is made of up seven factors (i.e., depression, generalized anxiety, social anxiety, academic distress, eating concerns, hostility, alcohol use) that encompass a general distress index. This factor structure differs from the CCAPS-62 in that it omits family distress and changes the name of substance use to alcohol use, as the subset of questions for that factor only contained questions pertaining to alcohol.

Although we received data for both the CCAPS-62 and the CCAPS-34, most participants were only administered the CCAPS-62 upon intake and were assessed with the CCAPS-34 for subsequent visits. Because the current study was conducted over multiple sessions, we decided to use only the CCAPS-34 items which were available for analysis after intake. For the CCAPS-62 data available at intake, we only used items appearing in the CCAPS-34, and, therefore, essentially treated each measure as the CCAPS-34.

Results

Hypothesis 1

We hypothesized that IA students would show a lower number of attended sessions of mental health services than AA students. In examining both the mean number of sessions scheduled and the mean number of sessions attended, AA students had significantly higher rates of utilization. Thus, our hypothesis that IA students attend fewer sessions of mental health services than AA students was confirmed, but with negligible effect sizes. Additionally, there was no significant difference in the proportion of sessions attended between the two groups (see Table 2).

Table 2. International Asian (IA)/Asian American (AA) Utilization Rates (Independent Samples t-test).

Category	Undergraduate				Graduate
	AAMean (SD)	IA Mean (SD)	$\bar{X}$ diff	d	AAMean (SD)	IA Mean (SD)	$\bar{X}$ diff	d
Age	20.85 (2.66)	21.43 (2.33)	−0.58	-	26.32 (4.25)	26.95 (3.72)	−0.63	-
# Sessions Scheduled	5.11 (5.15)	4.51 (5.07)	0.60^a	0.11	6.10 (6.38)	5.33 (6.17)	0.77^a	0.12
# Sessions Attended	3.93 (4.04)	3.38 (3.85)	0.55^a	0.14	4.78 (5.05)	4.12 (4.89)	0.66^a	0.13
Attended/Scheduled	0.81 (0.21)	0.80 (0.21)	0.01	-	0.83 (0.19)	0.82 (0.20)	0.01	-

^ap < .05.

Hypothesis 2

We also hypothesized that IA students would show a higher severity of initial presenting concerns and overall distress at intake than their AA counterparts. For undergraduates, IA students showed significantly higher initial distress on Depression, Hostility, and Eating Concerns indices on the CCAPS-34, while AA students showed significantly higher initial distress on the Anxiety index. Our hypothesis for undergraduate IA versus AA was confirmed, except for Anxiety, where AA students were higher. For graduate students, IA students showed significantly higher levels of distress on the Depression, Social Anxiety, Academics, and Hostility indices, as well as on the overall Distress Index. Graduate AA students showed significantly higher distress only on the Alcohol Use index. With graduate students our hypothesis was confirmed, except that graduate AA students exhibited higher distress with Alcohol Use (see Table 3).

Table 3. Undergraduate International Asian (IA)/Asian American (AA) Severity of Initial Concerns (Independent Samples t-test).

CCAPS Index	Undergraduate				Graduate
	AAMean (SD)	IA Mean (SD)	d		AAMean (SD)	IA Mean (SD)	d
Depression	1.99 (1.02)	2.05 (1.01)	−0.06^a		1.73 (0.98)	1.99 (0.97)	−0.27^a
Anxiety	1.97 (0.99)	1.83 (0.94)	0.20^a		1.86 (0.99)	1.83 (0.93)	-
Social Anxiety	2.12 (0.99)	2.15 (0.93)	-		1.96 (1.00)	2.08 (0.93)	−0.12^a
Academics	2.23 (1.08)	2.26 (1.12)	-		1.95 (1.04)	2.16 (1.08)	−0.20^a
Hostility	1.01 (0.86)	1.20 (0.92)	−0.21^a		0.85 (0.77)	1.11 (0.87)	−0.32^a
Eating	1.06 (1.14)	1.15 (1.2)	−0.08^a		1.00 (1.16)	1.01 (1.15)	-
Alcohol	0.52 (0.81)	0.49 (0.78)	-		0.49 (0.80)	0.38 (0.69)	0.15^a
Distress Index	1.95 (0.81)	1.95 (0.83)	-		1.75 (0.79)	1.89 (0.78)	−0.18^a

^ap < .05.

Hypothesis 3

We hypothesized IA students would demonstrate similar or slower treatment improvement rates compared to AA students. To evaluate this hypothesis, we conducted a Repeated Measures Analysis of Variance, controlling for number of sessions attended. While number of sessions attended was a significant covariate across indices in the undergraduate sample, for the graduate sample it was significant only for the Depression, Hostility, and overall Distress indices. For undergraduate students, there was no significant difference in treatment improvement rates with the exception of IA students showing significantly higher rates of improvement in terms of Hostility. For graduate students, the rate of change was significantly higher for IA students on the Depression, Social Anxiety, Academics, and Hostility indices, and on the overall Distress Index (see Table 4).

Table 4. Rate of Change: Main Effects of Time and Int’l Status, Covariate of Number of Attended Sessions, and Interaction Terms.

Variable (CCAPS Index)	Undergraduate				Graduate
	F^a	p	$\bar{X}$ diff^b		F	p	$\bar{X}$ diff
Time (Depression)	40.43	0.00*	0.24*		29.12	0.00*	0.25*
Int’l Status	6.17	0.01*	−0.06*		42.43	0.00*	−0.23*
Sessions Attended	12.53	0.00*			4.10	0.43*
Time ^x Attended	867.80	0.00*			290.57	0.00*
Time ^x Int’l	1.59	0.21			10.27	0.00*
Time (Anxiety)	16.24	0.00*	0.15*		12.98	0.00*	0.18*
Int’l Status	34.80	0.00*	0.13*		2.67	0.10	0.06
Sessions Attended	6.63	0.01*			2.04	0.15
Time ^x Attended	486.77	0.00*			240.07	0.00*
Time ^x Int’l	0.41	0.52			5.94	0.02*
Time (Social Anxiety)	8.90	0.00*	0.11*		3.96	0.05*	0.12*
Int’l Status	2.85	0.09	−0.04		9.74	0.00*	−0.11*
Sessions Attended	6.05	0.01*			0.73	0.39
Time ^x Attended	433.29	0.00*			174.76	0.00*
Time ^x Int’l	0.62	0.43			4.73	0.03*
Time (Academic)	32.49	0.00*	0.14*		15.49	0.00*	0.18*
Int’l Status	0.87	0.35	−0.03		22.43	0.00*	−0.18*
Sessions Attended	44.47	0.00*			3.11	0.08
Time ^x Attended	221.91	0.00*			137.15	0.00*
Time ^x Int’l	0.49	0.48			7.80	0.01*
Time (Hostility)	21.52	0.00*	0.14*		27.04	0.00*	0.13*
Int’l Status	78.30	0.00*	−0.18*		63.02	0.00*	−0.24*
Sessions Attended	7.61	0.01*			9.44	0.00*
Time ^x Attended	623.55	0.00*			95.23	0.00*
Time ^x Int’l	8.70	0.00*			8.50	0.00*
Time (Eating)	4.43	0.04*	0.06*		0.29	0.59	0.06*
Int’l Status	10.05	0.00*	−0.09*		0.04	0.85	0.01
Sessions Attended	9.35	0.00*			1.50	0.22
Time ^x Attended	75.68	0.00*			35.13	0.00*
Time ^x Int’l	0.16	0.69			2.86	0.09
Time (Alcohol)	5.15	0.02*	0.04*		7.59	0.01*	0.05*
Int’l Status	4.02	0.05*	0.03		18.41	0.00*	0.11*
Sessions Attended	28.22	0.00*			1.61	0.20
Time ^x Attended	80.24	0.00*			14.65	0.00*
Time ^x Int’l	0.13	0.72			0.16	0.69
Time (Distress Index)	44.92	0.00*	0.19*		29.84	0.00*	0.20*
Int’l Status	0.01	0.92	−0.00		16.61	0.00*	−0.12*
Sessions Attended	12.74	0.00*			4.61	0.03*
Time ^x Attended	918.08	0.00*			328.68	0.00*
Time ^x Int’l	0.71	0.40			10.96	0.00*

^aFor all models, Time, Time ^x Int’l, and Time ^x Attended = Within-Subjects F; Attended and Int’l = Between-Subjects F. ^b Mean Difference (I – J) based on estimated marginal means. For Time, I = Initial score and J = Final score; for Int’l, I = American and J = International Student. *p < .05.

Discussion

Increased understanding of the relationship between AA and IA students may help mental health providers in university counseling centers support their students more efficiently. In this study, we used a national archival dataset from the CCMH to compare AA and IA students in terms of the number of sessions used, presenting concerns, level of distress at intake, and treatment outcomes.

Hypothesis 1

Our first hypothesis was partially confirmed with small to negligible effect sizes. When examining the number of scheduled and attended sessions, IA students from both undergraduate and graduate levels showed slightly lower rates of utilization. Though these differences were statistically significant, their practical significance may be negligible. AA and AI students attended at the same rate or proportion as AA students whether at the undergraduate or graduate level.

It is important to note that previous literature suggests that Asian individuals (combining AA and IA into one group) tend to attend therapy at relatively lower rates than their European American (EA) counterparts (Kim et al., 2015; Leong, Lee, & Kalibatseva, 2016; Smith & Khawaja, 2011). Stokes et al. (2021) used data from a university counseling center and took the next step to examine the differences between AA, IA, and EA students. Our results from a national dataset seem to align with the findings of Stokes et al. (2021) (that IA students were less likely to utilize psychotherapy services than AA students) statistically, but may diverge in the practical significance of the finding (where we showed effect sizes less than 0.2).

Due to the nature of our data and the lack of any exit-interview information, it is unclear why IA students may utilize psychotherapy less than AA students. Given previous research findings, it is possible that IA students experience higher levels of distress associated with acculturation in the U.S. than their AA counterparts, particularly in terms of language barriers (Zhang & Brunton, 2007), psychological and sociocultural adjustment (Zhang & Goodson, 2011), perceived discrimination (Smith & Khawaja, 2011), and financial concerns (Poyrazli & Grahame, 2007). These factors associate with inhibiting formal help-seeking. Given that psychotherapy is more common in the U.S. than it is in Asian countries, seeking psychological services might be a foreign or even taboo concept for many IA students (Rogers-Sirin, 2013). In addition, Asian cultural values, such as emotional self-control, face-saving, shame avoidance, and family recognition through achievement (Speller, 2005) may cause IA students to be reluctant to engage in psychotherapy because of the degree of self-disclosure and perceived emotional vulnerability required in treatment. As such, IA students may experience greater distress for longer periods before seeking out treatment. Our findings suggest the potential need for more outreach activities to IA students aiming to increase mental health awareness and destigmatize mental health services, but may also suggest that they are closer to attending therapy at the same rate as their peers than previous literature has indicated.

Hypothesis 2

Our second hypothesis was partially confirmed as well. When examining severity of initial presenting concerns and overall distress at intake, IA students showed significantly higher levels of distress than their AA counterparts in many categories. In particular, IA undergraduate students reported significantly higher initial distress for Hostility (d = 0.21). IA graduate students showed significantly higher Depression, Academic Concerns, and Hostility than their AA peers. AA students showed significantly higher distress with a substantial effect size for only one scale. More specifically, AA undergraduate students reported higher Anxiety than their IA peers. Though the Alcohol Use index showed an effect size lower than 0.2, AA graduate students indicated significantly higher distress on the Alcohol Use index than IA graduate students. If this finding is replicated in future research, it may indicate an important difference between groups. It is important to note that the drinking patterns and motivations among these two groups could be complex. Findings from a national survey showed that Asian Americans who were born in the United States were more likely to use alcohol compared to non-U.S.-born Asian Americans (Breslau & Chang, 2006). It is possible that AA graduate students are more likely to use drinking to cope with stress and demands from graduate school. Given the likelihood that AA graduate students are likely to experience longer and higher acculturative stress, it seems possible that they could therefore be at a higher risk for increased alcohol use (Thai, Connell, & Tebes, 2010). Overall, this finding (though a negligible effect size) may support previous research that AA individuals more frequently consume alcohol than their counterparts (Greene & Maggs, 2020). We noted that IA undergraduate and graduate students appeared to endorse greater levels of distress regarding Depression and Hostility than their AA counterparts at intake. Again, these findings support results from previous literature that IA students may wait until distress is high before seeking help from mental health professionals (Li et al., 2013; Stokes et al., 2021).

Hypothesis 3

Our third hypothesis was also partially confirmed. We found that IA students utilize treatment at lower rates than AA students. Therefore, we controlled for the number of sessions attended when examining treatment improvement rates. Interestingly, number of sessions attended was a significant covariate across all CCAPS indices for undergraduate students, but not for graduate students. Specifically, we found no significant differences in the rate of change between AA and IA undergraduate students over seven of the eight CCAPS indices. However, we did find that IA undergraduate students showed significantly higher rates of improvement on the Hostility index. Again, because we did not have exit interview information, it is unclear why these undergraduate students demonstrated a faster rate of improvement on the Hostility index. Nevertheless, our results indicated that AA and IA undergraduate students were improving at a similar rate for the majority of the CCAPS indices even though IA students scheduled and attended fewer sessions than AA students. IA students appeared to improve from the therapy much the same as AA students.

We also took the number of attended sessions into account when examining treatment improvement rates in the graduate sample. Contrary to our hypothesis, we found that AA and IA graduate students showed similar rates of improvement on three CCAPS indices (Anxiety, Eating Concern, and Alcohol Use). IA graduate students demonstrated faster change in the other five CCAPS indices (Depression, Social Anxiety, Academics, Hostility, and the overall Distress Index) than their AA counterparts. Mirroring the findings from the undergraduate sample, these results suggest that IA graduate students were obtaining the same amount of change, if not faster, in fewer sessions than the AA graduate students. Various factors may account for this finding. For example, unattended mental health concerns is one of the factors that may affect one’s academic performance (Eisenberg, Gollust, Golberstein, & Hefner, 2007), which would in turn jeopardize students’ academic status at the university and their ability to graduate. It is possible that IA students, who are required to maintain a full course of study each semester/term in order to keep their lawful visa status (Department of Homeland Security, n.d.), are more motivated to make changes than AA students.

Results from the above comparisons also suggested that IA graduate students improved at faster rates across different problem areas than the IA undergraduate students. Though speculative, it seems likely that level of English proficiency might play a role in these findings. Previous findings indicate that English proficiency positively influences therapy outcomes amongst international students, particularly psychological adjustment, feelings of security, and level of stress (Redmond, 2000; Yeh & Iose, 2010). Furthermore, research has shown that important common factors contribute to therapeutic success, which include therapeutic alliance, empathy, collaboration, and goal consensus (Wampold, 2015). These factors in therapy outcome suggest that therapy is highly dependent on communication. It is possible that IA graduate students have higher levels of English proficiency and therefore can communicate their struggles, feelings, and thoughts more effectively than the IA undergraduate students. For example, Van Nelson, Nelson, and Malone (2004), found that for both South Asian and Asian International students, the graduate students outperformed undergraduate students on all language proficiency tests. While these results could be attributed to developmental maturity or age, studies on therapy outcomes and age have shown that age has little to do with such outcomes (Bennett et al., 2013). Given that age does not seem to impact therapy outcomes, but language proficiency does, IA graduate students, who likely have higher English language proficiency, may benefit from therapy more than the IA undergraduate students.

Stokes et al. (2021) found that IA students had lower improvement rates than AA and European American students. However, we found that when students attended their appointments, both AA and IA students improved at equivalent rates regardless of undergraduate or graduate status. Our results seem to corroborate previous research on treatment effectiveness in symptom reduction over time for most individuals (Lambert & Bergin, 1994).

Implications for practice and future research

Although previous literature and our own study indicate that IA populations are less likely to engage in psychotherapy, the evidence suggests they benefit when they do. It appears that IA populations generally improve at the same rate, or even faster (particularly among graduate students), than AA populations. We believe that focused outreach to IA students can help decrease barriers that keep IA students from seeking therapy. Outreach efforts may focus on seeking therapy early rather than waiting until distress becomes overwhelming, explaining the confidentiality of services, discussing ways that therapists will be sensitive to cultural issues, and describing the potential benefits of mental health services.

Acculturation may play a role in whether IA populations choose to reach out for mental health services and/or how effective mental health services are. Research findings generally indicate that international students with higher levels of acculturation are more likely to seek therapeutic treatment in comparison to international students with lower acculturation levels (Zhang & Dixon, 2003). Gaining an understanding of how acculturation impacts IA students’ decision making related to mental health services may further improve understanding regarding more effective outreach services targeted at IA populations. Future research, perhaps qualitative in nature, examining acculturation and its impacts could help improve targeted outreach to IA students.

Given that our research findings suggest that, once in therapy, IA populations recover at the same rate and in some instances even faster than AA populations, it would be beneficial to investigate factors that impact this finding. For example, language barriers and/or differences between therapists and clients may play a role in therapeutic effectiveness. Thus, hiring clinicians with therapeutic proficiencies in multiple languages may benefit counseling centers’ outreach efforts to IA students. Having culturally diverse staff may decrease barriers IA students face (e.g., stigma) when deciding whether to seek help for mental health difficulties.

While previous research frequently grouped AA and IA populations together, future research needs to further examine differences among IA populations from different countries. We recognize that this may be a challenge with lower sample sizes when, for example, trying to compare students from Vietnam with students from South Korea. However, vast cultural and individual differences exist between individuals from various countries, making it essential to continue disaggregating as clearly as possible cultural impacts on mental health services and help seeking behaviors.

Future research may also examine IA populations and the role intersecting identities play in relation to help seeking behaviors and therapy outcome. For example, what role does religiosity and/or spirituality play for IA students? What impact may sexual orientation or gender identity have on mental health treatment for IA students? While it is beyond the scope of our current study, it is clear that much more work is needed in the area of identifying best practices for helping IA students with mental health.

Limitations

Our findings were based on a national archival data set from universities contributing to the CCMH. While CCMH includes many colleges and universities, we acknowledge that this is not a representative sample of all universities and colleges. Thus, while this sample does provide more generalizability than previous studies (e.g., Stokes et al., 2021), more research is needed. Furthermore, given the archival nature of this data, we were not able to conduct exit interviews to help provide explanations of our results. We do not know fully why IA students participated in mental health services at lower rates but improved at similar rates as AA students.

Additionally, as already noted to above, we acknowledge that grouping AA and IA students together in collective Asian populations may obscure differences between Asian nationalities and lead to ethnic glossing. Asia consists of 50 countries with unique cultures, practices, and mental health related beliefs. While we were unable to examine country specific data, we feel this study is a step forward in the work to disaggregate AA and IA populations. For this purpose, we suggest that future research compare AA and IA students from various Asian countries, beginning with those that have large enough numbers of IA students for meaningful comparisons (e.g., China, India, and South Korea). It also may be useful to examine the differences between international students from distinct Asian countries (e.g., comparing the experiences of Indian students to those of Chinese students). In short, we encourage further disaggregation whenever possible.

Conclusion

We sought to determine if the results of Stokes et al. (2021), which used a sample from a single university, would hold true with a national dataset. Likes Stokes and colleagues, we found that at a national level (using a much larger dataset) grouping AA and IA individuals is problematic. We also believe this indicates a need for further research in this domain. While there may be similarities between AA and IA students, ethnic glossing obscures potentially important differences that impact the effective delivery of mental health services. We found that IA students schedule and attend therapy at lower rates than AA students. However, when they do attend, both AA and IA students improve at similar rates. In fact, IA students may improve faster.

Our use of a large sample of students attending therapy at diverse college counseling centers included in the CCMH builds upon the work of Stokes et al. (2021) and provides better generalization to AA and IA university student populations. As we better understand differences between AA and IA students, we can begin to tailor outreach efforts to reach those who may be reluctant to seek out or attend therapy. We can also ultimately provide better services for those who do attend.

The historic grouping of all Asian individuals under the broad umbrella of “Asian” or “Asian American” or “Asian American/Pacific Islander” remains problematic in that we are likely to dismiss important differences that exist. Thus, we encourage continued research in the area to better understand IA students’ needs and improve outreach and treatment efforts.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.