Evidence-Based Treatment & Patient Diversity: What We Know & Why It Matters

Transcription

1 Evidence-Based Treatment & Patient Diversity: What We Know & Why It Matters Stan Huey, Ph.D. Department of Psychology University of Southern California 13 th Integrating Care Conference Universal City, CA 10/20/16

2 Without data, you re just another person with an opinion -Andreas Schleicher-

3 Cultural Responsiveness Hypothesis Potential Problem with Conventional Therapies: Developed for White, Western, English-speaking Majority of clinicians are White Not consider language, beliefs, worldview of culturally different When culture is ignored: Value conflicts & miscommunication Client discomfort & poor engagement Dropout & treatment failure So treatments must be culturally responsive & clinicians must be culturally competent

4 Questions We Asked in Grad School Are EVTs effective with ethnic minorities? Sometimes maybe. But often not. Do White youth benefit more than minorities from the same treatments? Of course Whites benefit more Do cultural adaptations enhance outcomes for ethnic minorities? Yes, definitely

5 6 Questions To Address Are treatments effective with ethnic minorities? Are treatment outcomes worse for minorities compared to Euro-Americans? Does cultural tailoring enhance treatment outcomes for minorities? What about other domains of diversity? E.g., class, religious beliefs, sexual orientation Does culture matter? How to best address diversity in the treatment context?

6 Are Therapies Effective with Ethnic Minorities?

7 Meta-Analysis Primer What is Meta-Analysis? Quantitative Review of Therapy Effects Active Treatment vs. Control Group in Randomized Trials (RCTs) Effect Size d=.20 is small effect d=.50 is medium effect d=.80 is large effect Effects adjusted for sample size

8 Treatment Outcome Meta-Analyses with Ethnic Minorities 1 Effect Size (d) Asian Ams Latino Youth EBTs for Minority Youth Minorities/ Conduct Probs Minorities/ Drug Probs Gillespie & Huey, 2015; Huey & Polo, 2008; Huey et al., 2014; Huey & Tilley, 2016

9 Mental Health Treatment Effects for Ethnic Minorities Across 140 Randomized Trials Effect Size (d) Anxiety Depression Externalizing Probs Schizophrenia Substance Use Smoking Trauma-Related Misc/Other Huey et al., 2014

10 EBPs for Minorities More than 50 EBTs for ethnic minorities with diverse mental health problems Family systems therapies (e.g., FFT, BSFT, MDFT, MST) Interpersonal psychotherapy (IPT) Diverse cognitive-behavioral treatments (CBTs) Infant-parent relationship therapy, motivational interviewing, play therapy, and other therapies Modality doesn t seem to matter much Family vs. group vs. individual (e.g., Bernal; Nayamathi; Szapocznik)

11 Common Elements of Minority EBPs Theoretical coherence, with underlying theory of change Structured or semi-structured protocol, or treatment manual Standard number of sessions or clear termination criteria

12 Are Treatment Outcomes Worse for Ethnic Minorities vs. Euro-Americans?

13 Differential Effectiveness? Reviews by Huey & Polo (2008) & Miller et al. (2007) Most relevant studies show no ethnic differences in treatment effects 15%-23% show effects favoring minorities Results from 29 Meta-Analyses (Huey & Smith, 2014) 62% show no ethnicity effects 14% show effects that favor whites 17% show effects that favor minorities Summary: No consistent ethnicity effects

14 Does Cultural Tailoring Enhance Outcomes for Ethnic Minorities?

15 Why Cultural Sensitivity Matters

16 General Reasons to Consider Culture Implicit bias against outgroups is the norm, & such biases affect judgment & behavior Interracial interactions can be cognitively taxing Majority & minority groups often have different perceptions of opportunity & discrimination Banaji & Greenwald, 2013; Norton & Sommers, 2011; Richeson & Shelton, 2007

17 Clinical Reasons to Consider Culture Stigma Help-Seeking Underutilization Patient preferences Clinician/system biases Greater psychopathology or severity Symptom profile/presentation Unique MH correlates for minorities & immigrants Attrition/dropout Treatment barriers

18 What is Culturally-Responsive Tx? No uniform view Many opinions, many frameworks, many labels: Culturally-competent, minority-specific, ethnicallysensitive, culturally-tailored, culturally compatible, etc. CRT = Efforts to make treatments more appropriate for ethnic minorities

19 What is Culturally Responsive Tx? How do clinicians define? Survey by Zayas et al. (1996) 150 White members of APA & NASW Awareness of existence of differences (33%) Knowledge of client s culture (12%) Distinguishing between culture and psychopathology in assessment (21%) Taking culture into account in therapy (34%)

20 What is Culturally Responsive Tx? APA Guidelines for Multicultural Education, Training, Research, Practice Recognize that attitudes & beliefs can neg. influence interactions with culturally different Recognize importance of multicultural sensitivity/responsiveness to understanding culturally different Apply culturally appropriate skills in clinical and other applied psychological practices

21 Cultural Competence Models Rogler (1) Increase access, (2) Select traditional treatments that fit, (3) Modify traditional treatments Bernal Metaphors, language, etc. Sue et al. Tripartite Multicultural Competencies Lopez Shifting Cultural Lenses Model Fuertes & Gretchen 8 Theories of Multicultural Counseling

22 What is Culturally Responsive Tx? Some Pan-Minority Recommendations: Short-term, time-limited, pragmatic, directive, goaloriented, problem-focused treatment Attentive to effects of minority status or discrimination Assess whether behavior matches values & norms of host culture (i.e., is it adaptive in client s culture) Assess & validate client experiences w/racism Attend to nonverbal/indirect forms of communication Role induction

23 What is Culturally Responsive Tx? Recommendations for African Americans: Incorporate spirituality & faith-based coping Selected use of AAVE Recommendations for Asians/Asian-Americans: Accept & tolerate low levels of expressivity Avoid comments construed as critical or disapproving Recommendations for Latinos: Involve family in treatment Use polite form of you (usted) with adults

24 Content Analysis of CRTs Huey, Wood, & Arizago (2010) Based on 35 randomized trials that include cultural adaptations Must be clear link to race, ethnicity, culture 12 cultural tailoring categories

25 Content Analysis of CRTs Provide Education/Training E.g., Teach providers about values or beliefs of cultural group Cultural Content E.g., Use pictures, images, video, or objects that depict or target ethnic minorities Client-Provider Cultural Match E.g., Use providers who are knowledgeable of or sensitive to cultural background/needs of clients Cultural Themes/Values E.g., Adopt treatment structure/modality that matches the norms, values, or expectancies of ethnic group Huey, Wood, & Arizago, 2014

26 Content Analysis of CRTs Linguistic Matching E.g., Conduct treatment in the client s preferred or needed language Culturally-Responsive Interaction Style E.g., Attend to the hierarchical relationship between provider and client Use Family, Peer, or Community Agents E.g., Include community peers as intervention agents Cultural Labeling of Program or Concepts E.g., Use culturally-relevant sayings, proverbs, idioms, or honorifics Huey, Wood, & Arizago, 2014

27 Content Analysis of CRTs Design/Validation by Cultural Agents or Experts E.g., Use cultural agents/experts to review/endorse/rate the appropriateness of intervention or intervention components Individualizing Treatment E.g., Allow client to direct the course of treatment Prior Support with Ethnic Group E.g., Use interventions or strategies that are empirically-supported with the cultural group Miscellaneous E.g., Address how ethnic/cultural factors affect treatment engagement, process, or outcomes Huey, Wood, & Arizago, 2014

28 Evidence Most minority-focused treatments are culturallytailored E.g., Huey & Polo (2008), Gillespie & Huey (2015) 10 meta-analyses summarized by Huey et al. (2014) All show that culturally tailored treatment better than no treatment, placebo, & services-as-usual controls BUT, do culturally tailored treatments work better than generic treatments? Huey (2013) meta-analysis Rigorous, direct comparison of tailored vs. generic treatments Overall effect size of d=.01, no effect

29 Effect Sizes for 10 Randomized Trials of Culturally Tailored vs. Generic Treatments Effect Size (d) Studies showing positive tailoring effects Average Effect Size Across All Studies Studies showing negative tailoring effects Huey, 2013 Note: A positive effect size means that outcomes favor the culturally tailored condition; a negative effect size means that results favor the generic condition.

30 Why Might Tailoring Diminish Effects? Reactivity Some cultural content may evoke negative emotional rxns Chang; Webb Less Activation of Change Mechanisms Some tailoring may distract from core strategies or create inefficiencies that interfere with active ingredients E.g., Speech anxious Latinos undergoing exposure feel less anxious in front of a Latino audience (Perez, 2008) Paradoxically, cultural concordance could make clients too comfortable Castro; Lau; Kumpfer et al.; Kliewer et al.

31 Kliewer, Lepore Et Al. (2011) Sample & Design Black youth (91%) in high-violence, urban neighborhoods Randomly assigned to Standard vs. Enhanced Expressive Writing Standard Expressive Writing Write about their deepest thoughts and feelings related to violence Culturally Enhanced Expressive Writing Given option to write stories, skits, songs, or poetry about violence, and to share their work with others in the classroom Rationale? Strong oral tradition within African American culture Popularity of Spoken Word & role of rap in popular culture Reflects cultural experience of African Americans Results Enhanced less effective at reducing teacher-rated aggression! Why? Maybe less emotional processing in enhanced condition

32 When Might Tailoring Diminish Effects? Survey N=175 subject pool undergrads asked to rate importance of 62 practices used when working with ethnic minorities Students tend to favor implicit cultural adaptations & disfavor explicit adaptations Huey et al., 2016

33 Most & Least Important When Treating Ethnic Minorities 5 Most Important Individualize treatment to match needs of the client (38%) Interact with client in warm, informal, or personal manner (31%) Learn about values, norms, experiences of diverse cultural groups (28%) Refer client to therapists who speak same language (27%) Be attentive to issues of respect with client (26%) 5 Least Important (& Most Harmful) Give treatment approach a culturally-relevant title/label (41%) (H:14%) Encourage use of culturally-defined healing practices, such as prayer or use of herbal medicines (41%) (H:8%) Discuss ethnic/cultural differences between therapist & client (34%) (H:22%) Adopt a directive/authoritative interaction style w/client (33%) (H:25%) Focus on treating clients of same ethnic group as therapist (27%) (H:9%) Huey et al., 2016

34 Summary What we know so far Therapies are generally effective for ethnic minorities In lab and real-world settings Many EBTs for Blacks & Latinos And increasing for Asians Americans, indigenous populations, & ethnic minorities in other countries Mostly CBTs, but not exclusively Minorities & Euro-Ams mostly benefit equally No persuasive evidence that cultural tailoring necessarily enhances treatment effects

35 Other Aspects of Diversity?

36 Low-Income Clients Do EBTs work for low-income clients? Dozens of RCTs include predominantly low-income youth & families E.g., Most trials for Coping Power & MST focus on low-income youth EBTs are generally effective with this population Not much discussion of tailoring efforts

37 LGBT Clients Do EBTs work for LGBT clients? No RCTs focused specifically on LGBT youth with mental health problems Lots of adult-focused RCTs, but mostly focused on HIV prevention or treating sequelae of HIV (e.g., depression) EBPs generally effective with LGBT adults with anxiety-related problems, depression, & substanceuse problems But efficacy with LGBT youth mostly unknown, & importance of LGBT-specific tailoring unclear

38 Other Diversity Categories Religion Gender Immigrant status Age Region of country Disability Family structure Etc.

39

40 Evidence that Culture Matters

41 Race/Culture Matters Race, affirmation, & schooling Self-affirmation intervention with minority youth (Cohen & Sherman, 2014) Cultural adaptation with Asian Americans OST with Asian Americans (Pan et al., 2011) Directiveness with Asian Americans (Pan & Huey, 2016) Ethnicity and treatment process Ethnic differences in treatment resistance (Sayegh et al., 2016)

42 OST for Phobic Asian Americans One Session Treatment (OST) for Phobias (Ӧst, 1996) Participants: 30 Asian Americans, English speaking, screened for at least one phobia Fears of spiders, crickets, worms, & dead fish Design: Randomized into three conditions: OST-S, OST-CA, & self-help manual 7 Cultural Adaptations: E.g., Normalize problem; Emphasize/facilitate emotional control; Exploit vertical nature of therapy Pan, Huey, & Hernandez, 2011

43 OST Phobic Stimuli Common House Spider Cellar Spider Pan, Huey, & Hernandez, 2011

44 Behavioral Approach Holds animal for 20 seconds Holds animal for less than 20 seconds Touches animal with one finger Puts hand in cage Removes lid from cage Touches cage Stops close to table with cage Approaches within 1 meters of animal Approaches within 2 meters of animal Approaches within 3 meters of animal Approaches within 4 meters of animal Stops 5 meters from animal Refuses to enter test room Pan, Huey, & Hernandez, 2011

45 T2 Catastrophic Thinking Acculturation Status as a Moderator of Treatment Effects Standard OST Adapted OST Pan, Huey, & Hernandez, 2011

46 Brief Directive Intervention for Depression 120 Asian & White undergrads with subsyndromal depression Randomized to brief psychoed + feedback interventions: Directive tx: Probing for information, using direct questions, & eliciting specific responses that control the conversation in a session Non-directive tx: Probing for affect, reflection of feelings, restatement Placebo: Discuss highly rated cultural values Results Overall, directive more effective than non-directive & placebo Ethnicity moderated, but in somewhat unexpected way For Asians, DI > Placebo For Whites, Placebo > NI Pan & Huey, 2016

47 Ethnicity as Treatment Moderator Pan & Huey, 2016

48 Treatment Resistance in MST Ethnic diffs in Struggle & Working Through Hypothesis? 41 youth and families in MST clinical trial Juvenile drug offenders (M=15.4 years) 59% African American, 41% White Resistance Coding Chamberlain et al. (1984) Therapy Process Code Client behavior that appeared to block, divert, or impede the direction set by the therapist or if the client criticize[d] present family members Coded 3 audio-taped sessions from each client Cannabis use & recidivism at post-tx & follow-up Results Different resistance trajectories for Black vs. White families Negative quadratic Whites who desisted from crime; but positive quadratic for Black desisters Sayegh et al., 2016

49 Resistance Proportion Treatment Resistance, by Race 0.16 Resistance M (SD) Sample Category n Phase I Phase II Phase III Total (.08).06 (.07).07 (.07) Adolescent Ethnicity European African American (.08).04 (.04).07 (.09) 0.08 American European American (.08).10 (.10).07 (.05) African 0.04 American 0 I II III Treatment Phase Sayegh et al., 2016

50 Resistance Proportion Results 0.16 Resistance M (SD) Sample Category n Phase I Phase African II Phase III Follow-Up Americans Rearrested (.08).07 Who (.08) Desisted.06 (.06) African 0.08 at Follow-up American (.04).03 (.02).05 (.07) European American (.10).10 All (.10) Other.06 (.05) Desisted (.09).06 Youth (.07).09 (.09) African American (.10).04 (.06).09 (.10) European 0 American 6.02 (.02).09 (.09).07 (.06) I II III Treatment Phase

51 Addressing Diversity in Evidence- Based Practice

52 Strategies for Addressing Diversity Strategies that allow one to consider a variety of diversity concerns while minimizing stereotyping Many consistent with manualized approaches & common sense clinical practice Mostly derived from review of hundreds of EBPs But many don t have gold standard evidence

53 Strategies for Addressing Diversity Inclusivity Reduce access barriers Role induction Start & stick with client goals Fit analysis Strength-focus Cultural knowledge but avoid assumptions Humility

54 Inclusivity Diversity Cues Claude Steele & Whistling Vivaldi Diverse staff, pictures, brochures, etc. U Wisconsin brochure Wood website Use sensitive and inclusive language Use partner vs. boyfriend Donald Trump Bernie Sanders

55 Reducing Access Barriers Linguistic access Geographic access Home- & school-based approaches Mobile treatments (e.g., Mobile Triple P) Transportation, insurance, sliding scales, work schedule, childcare, etc. Engagement sessions to reduce barriers E.g., McKay et al., 1998

56 Treatment Engagement Rates, by Engagement Condition (McKay et al., 1998)* At least 1 session (%) 100% 80% 60% 40% 20% 0% 44% Usual Intake 86% 89% Phone Only Phone + In-Person *Predominantly poor, Black & Latino families

57 Role Induction Key elements of role induction Review expected frequency of attendance & services available Elicit treatment expectations & correct misperceptions Clarify therapist & client responsibilities Elicit reasons for entering treatment & discuss how treatment relates to identified problems Elicit barriers to attendance & problem-solve Improves engagement & reduces dropout for ethnic minority clients (Katz et al., 2004)

58 Focus on Client Goals

59 Focus on Client Goals Goals often reflect client priorities and values Elicit treatment goals from relevant family members Let client s goals direct course of treatment Resist impulse to change vs. collaboration on tx goals

60 Fit Analysis In CBT terms, a functional analysis Evaluate ABCs antecedents & reinforcing consequences of behavior In the MST world, involves fit circles Modify & individualize treatment based on evolving fit Maybe the implicit norm among clinicians??

61 Strength-Focused Emphasize what client/family is doing right Build on preexisting values & competencies Utilize preexisting resources With kids, engage parents & other feasible natural agents As explainers, interpreters, idea generators, supervisors, coaches, parent surrogates, etc. Who takes over when treatment ends? Caveat: Equivocal effects of child-only vs. parent + child treatment (e.g., Pina et al.; Szapocznik et al.)

62 Cultural Knowledge Some familiarity with norms, experiences, & challenges of population you re working with E.g., coming out & higher suicidality for LGBT youth E.g., higher substance abuse among LBT women E.g., discrimination & higher schizophrenia in AfrAms You should NOT be colorblind E.g., Apfelbaum, Norton research Avoid assumptions about the importance/relevance of race, class, sexual orientation, etc. for client

63 Humility

64 Humility We often don t know what we don t know We have self-serving & self-enhancing biases Therapists give inflated ratings of competence/adherence Generally, experienced therapists no better than novices Solicit client s perspective & experiences, but Don t presume you ll truly understand that experience, esp. if cultural differences Southpark

65 Contact: Stan Huey, Ph.D. Phone: University of Southern California Department of Psychology, SGM S. McClintock Ave. Los Angeles, CA References: Huey, S. J., Jr., & Polo, A. J. (2008). Evidence-based psychosocial treatments for ethnic minority youth. Journal of Clinical Child and Adolescent Psychology, 37, Huey, S. J., Jr., Tilley, J. L., Jones, E. O., & Smith, C. (2014). The contribution of cultural competence to evidence-based care for ethnically diverse populations. Annual Review of Clinical Psychology, 10, Huey, S. J., Jr., & Wood, L. (Winter, 2014). A marriage of research and practice: Adapting evidence based practices for diverse populations. The California Psychologist, 8-9. Sayegh, C.A., Hall-Clark, B.N., McDaniel, D.D., et al. (2016). A preliminary investigation of ethnic differences in resistance in Multisystemic Therapy. Journal of Clinical Child and Adolescent Psychology.

66 Hyperlinks

67

68 Bias as Zero-Sum Game

69 Struggle-and-Working-Through Patterson & Chamberlain, 1994

70 What NOT to do

71 What NOT to do

72

73 % of Clinicians Who Report Consideration of Culture in Working with Ethnic Minority (or Diverse) Clients % Culturally Competent Allison et al., 1996 Hansen et al., 2006 Harper & Iwamasa, 2000 Lopez & Hernandez, 1996 Maxie et al., 2006 Note: For Allison et al. (1996), reflects % reporting some level of competence in serving diverse clients; For Hansen et al. (2006), reflects % who consider themselves somewhat or very multiculturally competent; For Harper & Iwamasa (2000), reflects % who address ethnicity-related issues frequently or always when relevant; For Lopez & Hernandez (1996), reflects % who take culture into account for most or all culturally different patients ; For Maxie et al. (2006), reflects % who discussed ethnic/racial differences with a client at least once in the past 2 years.

74 Title