Directive and non-directive therapist styles: Brief intervention for subsyndromal depression for Asian and European Americans
Table of Contents Dedication ii Acknowledgements iii List of Figures vii Abstract viii Specific Aims and Hypotheses 1 Introduction 4 Method 16 Results 30 Table Table 1: Multiple Comparison Tests on Adjusted Means and Standard 32 Error for Depression, Coping, and Working Alliance at 1-month (T2) and 6-month follow-up (T3) Discussion 41 References 54 Appendices Appendix A: ANOVA Analyses Examining Pre-Treatment Scores 68 of Completer vs. Lost to Follow-Up (T2 & T3) Appendix B: Multiple Comparison Tests on Raw Means and 70 Standard Deviation for Depression and Coping at Pre-treatment (Time 1) Appendix C: ANCOVA Analyses Examining Outcomes for 71 Author as Therapist Versus Other Therapists Appendix D: ANCOVA Analyses Examining Therapist-Participant 72 Ethnic Match as Moderator of Treatment Outcomes Appendix E: Number of Participants Experiencing Clinically 75 Significant Change by Treatment Condition and Time Point Appendix F: Raw Means and Standard Deviations for Depression, 76 Coping, and Working Alliance at 1-month (T2) and 6-month follow-up (T3) by Ethnicity
vi Appendix G: ANCOVA Analyses Examining Participant Ethnicity 78 as Moderator of Treatment Outcomes Appendix H: Raw Means and Standard Deviations for Depression, 81 Coping, and Working Alliance at 1-month (T2) and 6-month follow-up (T3) by Acculturation Level – Asian American Sample Only Appendix I: ANCOVA Analyses Examining Acculturation Level 85 (EAVS-R) as Moderator of Treatment Outcomes Appendix J: ANCOVA Analyses Examining Acculturation Level 88 (AAVS-M) as Moderator of Treatment Outcomes Appendix K: ANCOVA Analyses Examining Culture Related 91 Variables as Moderators of Working Alliance Appendix L: Beck Depression Inventory-II 92 Appendix M: DSM-IV Based Depression Scale 95 Appendix N: Brief COPE 97 Appendix O: Asian American Values Scale-Multidimensional 99 Appendix P: European American Values Scale for Asian Americans- 102 Revised Appendix Q: Working Alliance Inventory 103 Appendix R: Therapist Styles Checklist 104
vii List of Figures
Figure 1: Flowchart of Study's Recruitment, Allocation, Follow-up, and 17 Analysis Figure 2: Moderation of T3 BDI Depression by Participant Ethnicity Using 37 Adjusted Means Figure 3: Moderation of T3 DDS Depression by Participant Ethnicity Using 38 Adjusted Means Figure 4: Moderation of T3 maladaptive coping by Participant Ethnicity Using 39 Adjusted Means
Abstract Subsyndromal depression (SSD) is a highly prevalent and disabling mood disorder. Research indicates no differences between SSD and major depressive disorder with regard to impairment and strain. This study evaluated the efficacy of a directive and non-directive intervention for SSD compared to a placebo control group using an Asian American and European American sample. Studies with Asians and Asian Americans suggest that using a directive approach to therapy promotes positive therapeutic process, which may lead to improved treatment outcomes. Participants in the directive and non- directive condition met with a therapist for a twenty minute session that focused on addressing symptoms of depression, providing psychoeducation, and offering feedback on coping strategies. Analyses indicate that the directive approach was generally more effective at improving depressive symptoms and coping strategies and strengthening working alliance compared to non-directive and placebo interventions. Additionally, significant treatment outcome and ethnicity moderation effects were found for the placebo control condition. Implications for research and clinical practice are discussed.
1 Specific Aims and Hypotheses This study assessed the efficacy of a brief intervention for subsyndromal depression (SSD) with Asian and European Americans. Geisner, Neighbors, and Larimer (2006) demonstrated that a brief, mailed intervention was effective in alleviating depressed mood and associated features, such as maladaptive coping. Research also indicates that a directive therapist style positively affects outcomes for Asian Americans compared to a non-directive therapist style (e.g., Atkinson & Matsushita, 1991; Exum & Lau, 1988; Hill et al., 2007). The primary goal of this study was to treat Asian and European American participants with SSD using directive and non-directive adaptations of Geisner et al.’s (2006) brief intervention. Participants were assigned to one of three treatment conditions: 1. Directive Intervention (DI): DI is a brief intervention that focuses on (1) psychoeducation, (2) personalized feedback on depressive symptomatology, and (3) personalized feedback on coping strategies. DI therapists used a directive style in sessions. 2. Non-directive Intervention (NI): This condition incorporated the same content as DI. However, therapists used a non-directive style in sessions. 3. Placebo control: In this condition, participants met with a therapist, but discussed culture-related topics rather than depression-related topics. This study adopted a 3 (treatment condition: DI, NI, placebo) X 3 (time: pre- treatment, 1 month follow-up, 6-month follow-up) X 2 (ethnicity: Asian American,
2 European American) design including 120 individuals (60 Asian American, 60 European American) with SSD. The specific aims and hypotheses were as follows: Aim 1. To evaluate the efficacy of directive and non-directive brief interventions for SSD. Hypothesis 1. Both DI and NI will be more effective at reducing depressive symptoms and maladaptive coping than placebo control. Additionally, both conditions will be more effective at improving adaptive coping than placebo. Aim 2. To evaluate culture-related moderators of intervention effects. Hypothesis 2a. Participant ethnicity (Asian American vs. European American) will moderate the effects of treatment condition on all outcomes. Specifically, DI will be more effective than NI for Asian Americans, whereas the active treatments (i.e., DI and NI) will be equally effective for European Americans. Placebo control will be the least effective intervention for both Asian and European Americans. Hypothesis 2b. Among Asian Americans, acculturation level will moderate the effects of treatment condition on all outcomes. Specifically, DI will be more effective than NI for less acculturated Asian Americans, whereas the active treatments will be equally effective for more acculturated Asian Americans. Placebo control will be the least effective intervention. Aim 3. To evaluate how working alliance is affected by treatment condition and whether culture-related variables moderate these effects.
3 Hypothesis 3a. Both DI and NI will lead to stronger working alliance between therapist and client than placebo control. Hypothesis 3b. Participant ethnicity will moderate the effects of treatment condition on working alliance. Specifically, DI will be associated with higher ratings of working alliance than NI for Asian Americans. No differences in ratings of working alliance will be found between European Americans in the DI and NI conditions. Placebo control will result in the lowest ratings of working alliance for both ethnic groups. Hypothesis 3c. Among Asian Americans, acculturation level will moderate the effects of treatment condition on working alliance. Specifically, DI will lead to higher ratings of working alliance than NI for less acculturated Asian Americans. Among more acculturated Asian Americans, there will be no differences in level of working alliance between DI and NI. Finally, placebo control will have the lowest ratings of working alliance regardless of acculturation level.
4 Introduction Major Depression, Subsyndromal Depression, and Correlates Major depressive disorder (MDD) ranks as the number one cause of worldwide disability among mental health disorders (Murray & Lopez, 1996). In the general population, prevalence rates for MDD are estimated at 6.7% for the past year (Kessler, Chiu, Demler, & Walters, 2005) and 17.1% for lifetime occurrence (Blazer, Kessler, McGonagle, & Swartz, 1994). Additionally, MDD has a negative impact on productivity and is a burden on health care resources (Greenberg, Stiglin, Finkelstein, & Berndt, 1993). Overall, MDD has been a major focus for psychological researchers, as evidenced by numerous studies addressing MDD onset, impact, and treatment (see Cuipers, van Straten, Andersson, & van Oppen, 2008; Moras, 2006). While the problems associated with MDD are well-documented, there is also a large set of individuals who report subsyndromal depression (SSD). Individuals with SSD often endure clinically significant levels of depression, but either do not meet DSM- IV criteria for MDD or dysthymia (Brent, Birmaher, Kolko, Baugher, & Bridge, 2001; Forsell, 2007; Judd, Akiskal, & Paulhus, 1997), or their scores on depression measures do not meet threshold for a major depressive episode (Barkham, Shapiro, Hardy, & Rees, 1999; Weinberg, Tronick, Beeghly, Olson, Kernan, & Riley, 2001). Approximately 11% of the U.S. population currently suffers from SSD compared to 4% of the population experiencing MDD (Judd, Paulus, Wells, and Rapaport, 1996).
5 In light of the recent addition of SSD to the spectrum of depressive disorders, more research has emerged highlighting problems secondary to SSD. Compared to non- depressed individuals, those with SSD report lower quality of life, energy level, emotional well-being, social functioning, health, physical activity, and work performance (Chachamovich, Fleck, Laidlaw, & Power, 2008; da Silva Lima & de Almeida Fleck, 2007; Goldney, Fisher, Dal Grande, & Taylor, 2004; Judd et al., 1996; Martin, Blum, Beach, & Roman, 1996; Rapaport & Judd, 1998). SSD is also related to greater utilization of mental health services, higher rates of attempted suicide and work days lost, and higher levels of social irritability and household and financial strain (Horwath, Johnson, Klerman, & Weissman, 1994; Johnson, Weissman, & Klerman, 1992; Judd et al., 1996; Wells et al., 1989). Moreover, Judd et al. (1996) report no differences between individuals with MDD versus SSD with regard to functional impairment and strain. Judd et al.’s (1996) findings are important to note because of the dynamic nature of depressive disorders. It is not uncommon to see individuals move in and out of depressive subtypes (Forsell, 2007). Initially, an individual might only meet criteria for SSD, but can later develop more severe depressive symptoms to meet an MDD diagnosis. Moreover, individuals with SSD are at high risk for developing MDD at some point in their lives (Judd et al., 1997; Judd et al., 1998). Overall, the problems related to SSD are clinically relevant and provide a strong rationale for continued research. Correlates of depressive symptomatology that are of particular interest for this study are the use of coping strategies. Correlational research indicates that using a
6 variety of coping strategies is associated with better depression outcomes (Clarke & Goosen, 2009; Wright, Banerjee, Hoek, Rieffe, & Novin, 2010). This study will examine whether the active interventions are associated with changes in the use of both adaptive and maladaptive coping strategies among individuals with SSD. Depression among Asian Americans Asian Americans are among the fastest growing minority groups in the United States (U.S. Census Bureau, 2008). As such, treatment outcome research with Asian Americans is becoming increasingly important. Though some view Asian Americans as a “model minority” who experience few mental health problems (Sue, Sue, Sue, & Takeuchi, 1995), research indicates significant mental health issues in this population (U.S. Department of Health and Human Services, 2001). However, it is unclear whether prevalence rates of MDD for Asian Americans are equal to rates in the general population. Some research reports lower rates (Takeuchi et al., 1998; Ying, 1988) while others report higher rates (Hsu, Wan, Adler, Rand, Choi, & Tsang, 2005; Kuo, 1984; Yeung et al., 2004; Young, Fang, Zisook, 2010). Still others report no difference in MDD prevalence between Asian and European Americans (Chang, 1996; Edman et al., 1998; Hinton et al., 1998; Jackson-Triche, Sullivan, Wells, Rogers, Camp & Mazel, 2000; Rosenthal & Schreiner, 2000; Siegel, Aneshensel, Taub, Cantwell, & Driscoll, 1998; Tabora & Flaskerud, 1994). Despite equivocal findings regarding cross-cultural differences, there is research showing that depressive symptomatology is common among Asian Americans. For
7 example, Bemak and Greenberg (1994) reported that 63% of Vietnamese Americans self- reported experiencing mild to severe depression and that 31% believed they felt more depressed than their friends. Similar research appears to confirm the prevalence of depressive symptoms among Asian Americans (Aldwin & Greenberger, 1984; Chung et al., 2003; Ying, 1988; Young et al., 2010). This research indicates that depression- related disorders are common among Asian Americans and continued research with this population is needed. Mental Health Treatment and Asian Americans There is little
well-controlled research examining the effectiveness of mental health interventions with Asian Americans. Some evidence suggests that matching therapist-client ethnicity and language or providing services in Asian-specific clinics leads to reductions in premature termination (Flaskerud & Hu, 1992; Sue, Fujino, Hu, Takeuchi, & Zane, 1991; Yeh, Eastman, & Cheung, 1994), increases in client satisfaction (Gamst et al., 2003), and improvements in overall client functioning (Gamst, Dana, Der- Karabetian, & Kramer, 2001). However, caution is needed in interpreting these findings as many of these studies were conducted in community-based mental health centers where standardization of treatment was unknown and difficult to replicate. Additionally, the culture-specific services in these studies were often not clearly defined, nor were they randomized clinical trials. More recently, a few well-controlled treatment outcome studies with defined cultural adaptations to services have been conducted with Asian and Asian American
8 populations (Hinton, Pham, Tran, Safren, Otto, & Pollack, 2004; Yu & Seligman, 2002). For example, Otto et al. (2003) made modifications to a cognitive-behavioral treatment for post-traumatic stress disorder (PTSD) for Southeast Asian immigrants that targeted culture-specific fears and provided language-appropriate services. The modified treatment led to greater reduction of PTSD symptoms compared to treatment as usual. Additionally, the author and colleagues (Huey & Pan, 2006; Pan, Huey, & Hernandez, 2011) conducted a randomized pilot trial that made culture-responsive modifications to an intervention for specific phobia. Results demonstrated that the culture-responsive intervention was related to greater benefits compared to the unmodified intervention and a self-help control group. The current study builds upon principles and findings from the Pan et al. (2011) study. This study focuses on one of the seven cultural adaptations described by Pan et al. (2011) – directiveness. Prior theory and research suggests that directive therapist styles may lead to positive outcomes with Asian Americans compared to non-directive styles (Kim, Atkinson, & Umemoto, 2001; Kim, Liang, & Li, 2003; Li & Kim, 2004; Sue & Zane, 1987). Though directiveness was found to be a robust outcome predictor, we were unable to isolate its unique effects on treatment outcomes (Pan et al., 2011). Therefore, the primary aim in this study addresses whether a directive versus non-directive therapist style influences treatment outcomes.
9 Directive and Non-directive Therapist Styles Some scholars suggest that directiveness is an important therapeutic approach when working with Asian Americans (Leong, Chang, & Lee, 2007; Sue & Zane, 1987) because East Asians are generally more comfortable with hierarchical, structured relationships (Sue & Zane, 1987). This approach may be viewed positively by Asian Americans, presumably because the therapist will be perceived as an expert and solution- focused (Leong et al., 2007; Sue & Zane, 1987). Indeed, research with Asian Americans indicates that directiveness is related to positive outcomes. For Asian Americans, directiveness resulted in higher ratings
of counselor effectiveness (Atkinson, Maruyama, & Matsui, 1978), session value (Hill et al., 2007), and counselor credibility than non-directiveness (Atkinson & Matsushita, 1991; Exum & Lau, 1988; Merta, Ponterotto, & Brown, 1992). Additionally, directive counselors were rated higher than non-directive counselors by Asian Americans on counselor cross-cultural competence, empathic understanding, working alliance, and session depth (Kim, Li, & Liang, 2002; Li & Kim, 2004). However, as this research has come largely from non-symptomatic college student samples, generalizability to clinical populations is unknown. Pope-Davis, Liu, Toporek, and Brittan-Powell (2001) recommend that clinical research rely less on analogue designs and focus on more clinical contexts. Therefore, this study focused on evaluating interventions for individuals with mild to moderate depressive symptoms.
10 This study examined the efficacy of brief interventions for individuals with SSD. In the active interventions, the therapist employed either a directive or non-directive style. Prior research has characterized directiveness as conveying information, probing for information, directing behavior, and making interpretations (Li & Kim, 2004). Also, directive therapists use direct questions, elicit specific responses, and make statements that control the verbal activity in a session (Atkinson & Matsushita, 1991; Hagebak & Parker, 1969). The DI condition integrated these diverse therapist style features. By contrast, prior studies have operationalized non-directiveness as the use of restatement, wherein the therapist literally repeats a client’s responses, and summarization, wherein the therapist uses paraphrasing to condense a client’s statements (Atkinson & Matsushita, 1991). Other non-directive behaviors include reflection of feeling and probing for affect (Li & Kim, 2004). Finally, non-directive therapists often give clients responsibility for choosing topics of conversation and emphasize clarification of affect (Hagebak & Parker, 1969). These features of a non-directive therapeutic approach will be used in the NI condition. It is important to note that while the two therapist styles in this study have different characteristics, elements of both styles may be effective and appropriate for treatment (Friedli, King, Lloyd, & Horder, 1997; Li & Kim, 2004; Ward et al., 2000; Waxer, 1989). While theory and research indicate that directiveness is related to positive outcomes for Asian Americans, comparative studies that include European American samples are rare. However, a study by Waxer (1989) did compare therapist style
11 preferences for Chinese and European Canadians. Results indicated that Chinese Canadians preferred a directive therapist style, whereas European Canadians preferred a non-directive style. Overall, therapist styles research has not led to specific hypotheses concerning whether European Americans as a group prefer directive or non-directive therapist styles. Much of the existing research that links therapeutic approach to treatment outcomes has been conducted using samples where participant ethnicity is either not reported or not the target of interest. Given general trends in the literature (e.g., Barbe, Bridge, Birmaher, Kolko, & Brent, 2004; Bond, Wingrove, Curran, & Lader, 2002; Ward et al., 2000), it can be reasonably assumed that the majority of participants in these studies are European American unless reported otherwise. In general, this research has produced mixed results. Some studies suggest that a non-directive therapist style is preferable as it can lead to less client resistance (Miller, Benefield, & Tonigan, 1993) and non-directive counselors are viewed as more capable of addressing social, personal, and emotional issues (Lee & Mixson, 1995). Other studies find no differences in outcomes between directive versus non-directive approaches (Bond et al., 2002; Markowitz, Spielman, Sullivan, & Fishman, 2000; Ward et al., 2000), or that non-directive approaches result in worse outcomes (Barbe et al., 2004). Thus, there are no clear trends that consistently indicate whether therapist style affects outcomes for European Americans or non-minority populations. For this reason, it is hypothesized that the
12 directive and non-directive interventions will be equally effective for the European American sample. Acculturation Effects Research suggests that among Asians, depressive symptoms and acculturation level are associated. Less acculturated Asians tend to experience higher levels of depressive symptoms compared to more acculturated Asians (Chung et al., 2003; Hwang & Ting, 2008). Less acculturated Asians also appear to cope with depression differently; they engage in less adaptive (e.g., problem solving, use of social support) and more maladaptive coping strategies (e.g., avoidance, self-blame) compared to more acculturated Asians (Noh & Kaspar, 2003; Wong, Kim, & Tran, 2010). For Asian Americans, acculturation level may also influence how therapist intervention styles affect outcomes. Some theorize that less acculturated Asian Americans prefer a directive therapist style due to the greater respect given to authority figures in Confucian-oriented cultures (Kim et al., 2001). Less acculturated Asian Americans may view therapists as mental health experts who will provide structure in their sessions and work towards immediate problem resolution (Kim et al., 2001). Though research indicates that directiveness is related to positive outcomes for Asian Americans (e.g., Exum & Lau, 1988; Atkinson & Matsushita, 1991), only a small amount of research exists on the relationship between acculturation and therapeutic approach. Li and Kim (2004) found no effects of varying therapist styles and acculturation on counseling process variables. In contrast, a more directive approach resulted in better
13 treatment outcomes among less acculturated Asian Americans in a study on phobic anxiety (Pan et al., 2011). Additionally, Hill et al. (2007) found that Asian Americans with higher adherence to Asian values rated directive sessions more positively, whereas those less adherent to Asian values rated non-directive sessions higher. As a directive approach has demonstrated some positive effects with less acculturated Asian Americans, it is hypothesized that their depressive symptoms will be more improved following the DI condition compared to NI. Working Alliance Establishing a strong working alliance between therapist and client is a vital component to overall success and satisfaction with therapy (Fischer, Jome, & Atkinson, 1998; Horvath & Greenberg, 1994; Horvath & Symonds, 1991; Iacoviello, McCarthy, Barrett, Rynn, Gallop, & Barber, 2007; Krupnick et al., 1996; Martin, Garske, & Davis, 2000). One aim of this study is to examine how a directive or non-directive approach affects the working alliance. To date, the research on the relationship between working alliance and therapist style is inconsistent. For example, more directive interventions, such as cognitive behavior therapy and solution-focused brief therapy, have led to stronger working alliance compared to non-directive interventions (Loeb et al., 2005; Wettersten, Lichtenberg, & Mallinckrodt, 2005). Conversely, non-directive therapist attributes such as warmth, empathy, acceptance, and openness and therapist techniques such as reflection, supportiveness, and facilitating the expression of affect result in strong working alliance (Ackerman & Hilsenroth, 2003; Watson & Geller, 2005). Finally, other
14 research indicates that strong working alliance can be developed regardless of therapy modality (Krupnick et al., 1996; Watson & Geller, 2005). It is hypothesized that both active therapy styles, directive and non-directive, will lead to higher ratings of working alliance compared to placebo control Examining how participant ethnicity and acculturation level interact with treatment condition to affect working alliance is another aim of this study. It has been shown that among Asian Americans a stronger working alliance is achieved with a directive (versus non-directive) approach, but no European American comparison group was included (Li & Kim, 2004). Shonfeld-Ringel (2001) suggests that to develop a strong working alliance with Asian clients, treatment sessions should be therapist- directed. Based on prior research indicating that a directive therapeutic approach generally lead to positive outcomes among Asian Americans (Kim et al., 2002; Li & Kim, 2004), it is hypothesized that the directive intervention will result in the highest levels of working alliance for this ethnic group. For European Americans, working alliance is hypothesized to not differ between active treatment groups. Additionally among Asian Americans, it is hypothesized that less acculturated individuals will report higher levels of working alliance following the directive intervention compared to the non-directive intervention. This hypothesis stems from research indicating that a directive approach is rated more positively among individuals with high adherence to Asian values (Hill et al., 2007). No differences are expected between treatment conditions among more acculturated Asian Americans.
15 Brief Interventions for SSD This study used a brief intervention for SSD, derived from a mailed-focused intervention by Geisner et al. (2006) consisting of psychoeducation and personalized feedback. The intervention led to significant reductions in depressive symptoms and increases in willingness to use adaptive coping strategies. Participants in the Geisner et al. (2006) study were primarily European American (60%) and Asian American (34%), and participant ethnicity did not influence the effectiveness of the intervention. Many brief, face-to-face interventions have been effective for treating SSD as well. For example, highly structured, time-limited therapies have led to increased self- esteem, social skills, social support, and client satisfaction, and decreased depressive symptomatology (Barkham, Shapiro, Hardy, & Rees, 1999; Detert, Llewelyn, Hardy, Barkham, & Stiles, 2006; Heywood, Stancombe, Street, Mittler, Dunn, & Kroll, 2003). Additionally, brief interventions with minimal therapist contact such as telephone- assisted therapy (Bilich, Dean, Phipps, Barisic, & Gould, 2008; Mohr, Hary, & Marmar, 2006), and computer- or internet-based therapy (Andersson, Bergstrom, Hollandare, Ekselius, & Carlbring, 2004; Cavanagh, Shapiro, Van Den Berg, Swain, Barkham, & Proudfoot, 2006; Spek, Cuijpers, Nyklicek, Riper, Keyzer, & Pop, 2007) have been effective. By expanding on Geisner et al.’s (2006) design to include a brief, face-to-face session with a therapist, this study integrated aspects of successful interventions for SSD that emphasize brevity and minimal therapist contact.
16 Method Participants and Screening Participants were recruited through email announcements, flyers, and the USC Psychology subject pool. When recruited through email or flyers, potential participants were directed to an online survey where they completed a screening questionnaire. Participants recruited through the subject pool completed the same screening survey and were contacted after their eligibility was determined. Potential participants were asked to take part in the study if they met the following inclusion criteria: (1) self-identified as Asian American or European American, (2) fluent in English, (3) 18 years of age or older, and (4) reported depressive symptomatology congruent with SSD, based on the Beck Depression Inventory – II (BDI; Beck, Steer, Ball, & Ranieri, 1996; Beck, Steer, & Brown, 1996). Participants with BDI scores between 14 and 28 and no suicidal tendencies were eligible for inclusion. Individuals were excluded if they: (1) had a total BDI score below 14 or above 28, (3) endorsed either a 2 (“I would like to kill myself”) or 3 (“I would kill myself if I had the chance”) on the BDI’s suicidal thoughts or wishes item, (4) were currently receiving mental health services, or (5) were taking mood stabilizing medications. Figure 1 shows a flowchart detailing enrollment, allocation, follow-up, and analysis. If participants met study criteria, they were directed to another online survey that included additional questionnaires and asked to schedule a twenty minute session with a study therapist. One hundred sixty individuals met study criteria and were contacted for this next step. Forty
18 individuals declined to participate further due to scheduling difficulties, loss of interest, or not understanding the requirements of the study at prescreen (i.e., did not understand time commitment for study).
One hundred and twenty individuals participated in the study. Sixty were Asian American and sixty were European American. The sample consisted primarily of female (78%) undergraduate or graduate students (97%), with an average age of 21.2 years. The three largest ethnic groups among the sixty Asian American participants were Chinese (65%), Korean (17%), and Taiwanese (7%). Among the Asian American sample, 78% reported being bilingual; among the European American sample, 28% reported being bilingual. Measures Unless otherwise noted, the following measures were given at pre-treatment (T1), one-month follow-up (T2), and six-month follow-up (T3). Depressive symptoms. Two measures of depressive symptomatology were included in the study. As indicated earlier, the BDI was used during screening to determine study eligibility and, at subsequent follow-ups, as an indicator of depressive symptoms. The BDI is a self-report, 21-item measure that has respondents select among four statements that range in severity. Respondents are asked to choose the statement that best describes how they have been feeling over the past two weeks (Beck et al., 1996). The BDI assesses domains such as pessimism, worthlessness, self-dislike, loss of pleasure, changes in appetite, and loss of energy. This measure has demonstrated
19 convergent, divergent, and construct validity (Segal et al., 2008). Although the BDI has good internal consistency as well (α = .90) (Segal, Coolidge, Cahill, & O’Riley, 2008; Steer, Rissmiller, & Beck, 2000), for this sample, coefficient alpha was poor at T1 (α = .38). However, internal consistency was excellent at T2 (α = .92) and T3 (α = .93). The low alpha at T1 may be attributable to range restriction on the BDI (i.e., restricting inclusion to those with scores between 14 and 28) (Sackett, Laczo, & Arvey, 2004; Weber, 2001). This point will be further expanded upon in the Discussion section. The second measure of depressive symptomatology is the DSM-IV Based Depression Scale (DDS), which includes items that reflect DSM-IV symptoms for MDD. Participants are asked how severely they experienced depressive symptoms over the past two weeks on a 5-point Likert scale (0 = did not occur, 4 = very severe). Sample items include: “Feeling depressed or down most of the day, nearly every day”, “Decrease in interest of things you once found interesting (most of the day, nearly every day)”, and “Feeling worthless about yourself (nearly every day)”. The DDS has good reliability (α = .90) and demonstrates convergent validity with the BDI (r = .76) (Geisner et al., 2006). Internal consistency with this sample is α = .86 at T1, α = .92 at T2, and α = .94 at T3. Coping strategies. The Brief COPE (COPE) was used to assess respondents’ coping strategies (Carver, 1997). The COPE consists of 28 items, which measure 14 conceptually different coping strategies. Of these coping strategies, ten are considered to be adaptive (active coping, self-distraction, use of emotional support, use of instrumental support, venting, positive reframing, planning, humor, acceptance, and religion) and four