«The Mediating Effect of Acculturation on the Effectiveness of Culturally Adapted Cognitive Behavioral Therapy with Mexican Americans Suffering From ...»
SETTING The setting for the study was the city of El Paso, Texas, one of the major border cities located along the 2000-mile U.S.-Mexico border. El Paso is located in the outmost
Chihuahua, Mexico to the South. El Paso’s population was 755,085 in 2006 and Ciudad Juarez had a population of 1,563,973. Together, they equal more than two million people (American Community Survey, 2005). El Paso and Juarez have often been referred to as the “twin cities” because they are socially, economically, and politically interdependent.
In 2004, there was an average of 1,295,865 privately owned vehicles that crossed into El Paso from Juarez, Mexico and an average of 706,927 pedestrians that entered through the same ports of entry (U.S. Customs Service, 2004). The El Paso, Juarez, Southern New Mexico region is considered one of the largest manufacturing centers in North America, having 267,500 manufacturing workers. El Paso is home for more than 70 Fortune 500 companies, including Hoover, Eureka, and Boeing (Regional Development Economic Corporation, 2008).
The population breakdown in El Paso is 80% Hispanic or Latino of any race (75% are of Mexican descent), 15% White, 3.5% African American, 1.5% Asian, 1% American Indian and Alaska Native, and 0.1 Native Hawaiian and other Pacific Islander (American Community Survey, 2006). In addition, it is estimated that 26.9% of the El Paso population are foreign-born. The median household income in El Paso for 2006 was $33,103 and 23.6% of the population lived below the poverty level (American Community Survey, 2006). Approximately 20% of the population in El Paso attained a Bachelor’s degree or higher. The languages spoken are predominantly English and Spanish, with 29% of households speaking English only and 69% of the households speaking Spanish (U.S. Census Bureau, 2000).
Prior to initiating data collection, approval from The University of Texas at Austin’s Institutional Review Board (IRB) for the protection of human subjects was obtained (Appendix A). In addition, approval was obtained from the Executive Director of Family Service of El Paso to conduct the study in the FSEP facility (Appendix B).
Study participants for the dissertation were recruited in conjunction with the larger Hogg Foundation study. Individuals who called FSEP for mental health services were screened for eligibility into the larger Hogg Foundation study by the executive director, a Licensed Professional Counselor, based on information regarding income, ethnicity, and symptoms on the initial intake form (Appendix C). Of the individuals who consented to participate in the Hogg foundation study, a subsample who self-identified as Mexican or Mexican-American were invited to participate in the dissertation study. These consenting participants comprised the experimental treatment group and were routed to receive culturally adapted cognitive behavioral therapy by specially trained therapists who used a structured CBT protocol (Appendix D). Therapists for the treatment group were Masters level social workers or professional counselors. If the individual did not meet eligibility to participate in the Hogg Foundation study because they had insurance or Medicaid or could not be matched to a research therapist due to scheduling issues, the individual was asked to participate in the comparison group for this dissertation study.
The participants of the comparison group received treatment as usual provided by a variety of individuals ranging from Licensed Professional Counselors to social work and counseling interns. Treatment as usual included supportive therapy, behavioral therapy, existential therapy, and narrative therapy, depending on the theoretical orientation of the individual providing therapy for the comparison group. Participants in the treatment
comparison group were adult Mexican Americans who were diagnosed with depression who had insurance, Medicare, or Medicaid.
PARTICIPANTSThe participants in this study were individuals who self-identified as Mexican American or Mexican, male or female over the age of 18, and were diagnosed with depression. Both English and Spanish-speaking clients were recruited. Clients who were taking medication for depression or other physical ailments were included. Persons with comorbid mental diagnoses with psychotic features were excluded. Psychotic features were identified through clinical assessment conducted by the treating therapists at the time of the initial assessment and during the course of treatment using DSM-IV-TR criteria. Psychotic features are defined in the DSM-IV-TR as: 1) delusions 2) hallucinations 3) disorganized speech 4) grossly disorganized or catatonic behavior. The treating therapists were instructed to notify the principal investigator if a study participant presented the above-mentioned symptoms and the study participant was excluded from the study at that time. The principal investigator made the final determination as to whether an individual was included or excluded from the study.
Based on a power analysis, a minimum of 68 participants, with each group comprised of 34 were required to yield an effect size of.95. Using GPower (Faul et al.,
2007) to conduct a priori t test power analysis indicated that the sample size required for an α=0.05, power (1-β)=0.95, with 2 groups of an effect size of 0.9 is N=68. This analysis
detecting a clinically meaningful difference in depression if there really is one.
Participants were assigned to either a treatment group or a comparison group. The treatment group (n=48) received culturally adapted cognitive behavioral therapy. The comparison group (n=33) received treatment as usual.
The theoretical population examined in this study was adult Mexican Americans with depression. The participants were between the ages of 18 and 84. All of the participants were either English or Spanish-speaking individuals recruited from FSEP.
The target population included 236 potential participants who reported symptoms of depression during the intake process. Of the pool of 236, 71 individuals were not invited to participate due to the absence of trained agency staff at the time of the intake and 20 were excluded because they did not meet criteria due to being minors, experiencing bereavement rather than depression, not scoring as depressed, or requiring marital or family counseling. Four individuals refused to participate in the study. A total of 141 cases (of 145) were included in the study, which resulted in a response rate of 97%. Of the 141 cases, 78 cases met criteria to be included in the treatment group and 63 cases met criteria to be included in the comparison group. Only respondents who were administered three Patient Health Questionnaires were included in the study, for a total sample of 81 cases. Of these, 48 met criteria to be placed in the treatment group and 33 met criteria to be placed in the comparison group.
Each participant completed a client information form, which all FSEP clients complete and is kept in the agency client folders. In addition, each participant completed a consent form, an Acculturation Rating Scale for Mexican Americans-II (ARSMA-II) and a Patient Health Questionnaire (PHQ-9). Each case was then assigned an identification number and written in a master log. All questionnaires were reviewed for legibility, completeness, and adherence to the eligibility criteria listed in the study protocols. The ARSMA-II scales and the PHQ-9 questionnaires were reviewed for missing data. The identification number was written on all pages of the consent forms, as well as, on each page of the corresponding ARSMA-II scales and PHQ-9 questionnaires.
The completed ARSMA-II scales and the PHQ-9’s were then separated from the consent forms. All documents were kept in a locked cabinet at FSEP and only the principal investigator had access to these documents.
A codebook was developed to assign numerical coding to selected demographic, cultural and clinical characteristics obtained from the client information form, the initial assessment form, the ARSMA-II and the PHQ-9 questionnaire. Dichotomous responses were coded numerically with a “0” and “1” and each non-continuous variable was given numerical codes for each of the possible indicators of the variable. For multivariate analyses, the ordinal level measures of age, education, and marital status were re-coded into dummy variables. Once the data was appropriately coded, it was entered directly into
Depression Scale Measures of depression were collected using the Patient Health Questionnaire (PHQ-9) (Appendix E) in either English or Spanish. The instruments were administered by trained bilingual agency staff. Depression scores for the participants were measured during the initial therapy visit and every subsequent visit during the data collection period.
The PHQ-9 is a self-administered 9-item instrument that measures both the presence of depression symptoms and severity of depression as a continuous measure (Martin, Rief, Klaiberg, & Braehler, 2006). Each item on the instrument pertains to one of the nine symptoms of depression as outlined in the DSM-IV-TR. The instrument is available in English and in Spanish and takes approximately ten minutes to administer.
Although there are a number of scales that measure depression, the Patient Health Questionnaire (PHQ-9) was used for this study for several important reasons. After considering the Beck Depression Inventory-II (BDI-II) and the Center for Epidemiological Studies Depression Screening Index (CES-D), the PHQ-9 was selected because of the ease of its administration and its availability in English and Spanish.
Another criteria used for selecting the PHQ-9 was its availability in the public domain free of charge. Feasibility and cost issues were considered since FSEP is a non-profit organization.
Besides being brief and written in simple language, the PHQ-9 has good psychometric properties in both English and Spanish. The PHQ-9 in English has a sensitivity of 88% and a specificity of 88% for major depression. The internal reliability of the PHQ-9 is high with a Chronbach’s alpha of.89 (Kroenke, Spitzer, & Williams,
86% in a sample of Spanish-speaking general hospital inpatients (Diez-Quevedo, Rangil, Sanchez-Planell, Kroenke, & Spitzer, 2001).
Acculturation Measure Acculturation was measured using the Acculturation Rating Scale for Mexican Americans-II (ARSMA-II) (Appendix F). Study participants completed Scale 1 of the Acculturation Rating Scale for Mexican Americans-II (ARSMA-II) (Cuellar et al., 1995).
The ARSMA-II includes a section for identifying generation status. The section includes the criteria for the five generation statuses. Individuals who were born in Mexico are considered First Generation, all other generation statuses include individuals who were born in the United Status.
The ARSMA-II is comprised of two scales. Scale 1 measures the integration and assimilation modes of acculturation through a self-administered 30-item Likert-type scale that poses questions regarding language use and preference, ethnic identity and classification, cultural heritage and ethnic behaviors and ethnic interaction. Participants are asked to select from (1) Not at all, (2) Very little or not very often, (3) Moderately, (4) Much or very often, and (5) Extremely often or almost always. Scale 1 is scored by calculating the mean for the Anglo-Orientation subscale (by adding items 2,4,7, 9, 10, 13, 15, 16, 19, 23, 25, 27, and 30 and dividing the result by 13) and then calculating the mean for the Mexican-Orientation subscale (adding items 1, 3, 5, 6, 8, 11, 12, 14, 17, 18, 20, 21, 22, 24, 26, 28, and 29 and dividing the result by 17). The mean for the MexicanOrientation subscale is then subtracted from the mean for the Anglo-Orientation subscale to provide a raw acculturation score. This score is then compared to the acculturation level cutoff scores to identify an acculturation level.
acculturation in research. Since its development, the original ARSMA has been used in numerous studies measuring acculturation and has been used in the development of other acculturation scales (Cuellar et al., 1995) The ARSMA-II was selected for this study because the simplicity of the questions can be understood by people of low literacy and low English proficiency, which matches a large proportion of the population under study.
In addition, it was selected because it is easy to administer and takes approximately fifteen minutes to complete. This scale fits well for a population that is depressed and has low energy.
The ARSMA-II is available in both English and Spanish. The test-retest reliability of Scale 1 at one-week interval was found to be.96 and the concurrent validity with the original ARSMA is.89. Chronbach’s Alpha for the items on the scale is.86 for the items that comprise the Anglo Orientation Score and.88 for the items that comprise the Mexican Orientation Score (Cuellar et al., 1995).
Demographic Data Measures of education level, gender, age, marital status and language preference were collected from the agency’s Client Information Form (Appendix G). Measures of history of mental health treatment, medical conditions, medications used and comorbid psychiatric diagnoses were collected from the initial assessment form (Appendix H) that was completed by agency therapists.