1. Alcohol, Drug and Mental Health Comorbidity on the Texas-Mexico Border Lynn Wallisch and Richard Spence University of Texas at Austin, School of Social Work, Addiction Research Institute Introduction In a representative sample of adults living in residential households on the Texas-Mexico border, 12.9% had DSM-IV defined problems of alcohol abuse or dependence. Some 5.6% had alcohol problems only, another 2.8% had alcohol problems and also used illicit drugs, and 3.5% had alcohol problems and problems with gambling and/or mental health. This study compares characteristics and treatment attitudes and behaviors among these three groups. Comorbidity is expected to be associated with acculturation; with substance problem severity; and with increased likelihood of seeking services. Comorbidity Measures and Hypotheses Alcohol Problems Alcohol problems were defined as meeting the DSM-IV criteria for abuse or dependence. Gambling Problems Positive response to three or more of the following questions, drawn from the South Oaks Gambling Screen: 1. When you participated in these gambling activities during the last 12 months, how often did you go back another day to win back money you lost? (most or every time) 2. During the past year, did you ever spend either more time or more money gambling than you intended? 3. During that year, did you ever feel guilty about the way you gambled or about what happened when you gambled? 4. During that year, did you ever feel that you would like to stop gambling, but didn’t think that you could? 5. During that year, did you ever borrow from someone and not pay them back as a result of your gambling? 6. In the past 12 months, do you feel that you have had a problem with betting money or gambling, such as feeling guilty over what happens when you bet or wishing you could cut down on your gambling? Conclusions Results Presented at College on Problems of Drug Dependence 67 th Annual Scientific Meeting, Orlando, FL, June, 2005. This study was supported by NIDA grant R01DA14794. Demographic Characteristics of Sample Mental Health Questions Reporting poor mental health or scoring in the top 20 th percentile of the sample on the depression questions (short version of CES-D): 1. Would you say your emotional or mental health is excellent, good, fair, poor, DK (poor). 2. I am going to read a list of ways you may have felt. Please tell me how often you have felt this way during the past week (none of the time, rarely, some of the time, most of the time). a. I didn’t not feel like eating; my appetite was poor. b. I had trouble keeping my mind on what I was doing. c. I felt depressed. d. Everything I did was an effort. e. My sleep was restless. f. I felt sad without a real reason. g. I lost interest in my usual everyday activities. Hypotheses Comorbidity is expected to be associated with -being born in the US vs Mexico and greater acculturation to Anglo culture (Cherpitel et al., 2004; Grant et al., 2004; Ortega et al., 2000); - having a greater number of drug or alcohol problems (Brooner et al., 1997; Merikangas et al., 1998; Grant, 1997); and -greater motivation for and prior use of substance abuse treatment (Vega et al.,1999; Grant, 1997; SAMHSA, 2002). Sample and Methods 1200 adults living in households on the Texas-Mexico border were interviewed in person in 2002-2003 about their substance use and related factors. The sample was randomly selected under a multistage cluster sampling design, and respondents were drawn equally from El Paso city, the urban areas of the lower Rio Grande Valley, and colonias in the Valley. In statistical analysis, data were weighted to account for probabilities of selection and to conform the sample to the census age-sex-ethnic-site distribution. This study of comorbidity included only those respondents who had past-year problems of alcohol abuse or dependence (N=200). Three groups of these respondents were compared: 1. Alcohol problems only (N=86) 2. Alcohol problems and past-year illicit drug use (N=53) 3. Alcohol problems and mental health or gambling problems (N=61) (about half of this group had also used an illicit drug) The very small cluster sizes in the analysis subgroups (average cluster = 1.5) suggest that any potential within-group clustering would result in a negligible impact on the standard errors for the analysis (Maas & Hox, 2002); accordingly, statistical analysis was done in SAS, with weighted sample sizes adjusted to equal the raw sample. The groups did not differ significantly by gender or education. Those with alcohol problems who were also drug users were most likely to be young and to have the highest household income; however, these characteristics were not strongly associated with the outcome variables. The hypotheses were mostly supported. Those with comorbid problems were more likely than those with alcohol problems only to have been born in the US (as compared to Mexico) (p=.01), and were more likely to be culturally Anglo-oriented (as compared to Mexican-oriented) (p=.01). Those with comorbid gambling or mental health problems had higher alcohol severity (as measured by number of symptoms reported) than those with alcohol problems only (p=.02), and higher drug severity than those with alcohol problems who used drugs but didn’t have mental health or gambling problems (p=.03). Those with comorbid problems were more likely than alcohol only respondents to have ever talked to someone about their drug or alcohol problem (p=.03). The percentage who had ever received substance abuse treatment was two to three times higher (but not significantly different) for those with comorbid problems, and the percentage who had wanted but failed to get treatment was significantly higher for those with comorbid gambling or mental health problems than for the other two groups (p=.00). The percentage motivated for treatment at this time, if it were affordable and convenient, was also significantly higher for those with comorbid gambling or mental health problems than for the other two groups (p=.01). These findings of greater substance severity level and treatment propensity are especially interesting because the measures used to represent comorbidity do not necessarily indicate serious pathology; past-year drug use is not the same as abuse or dependence (although three-quarters of these drug users did have problems of abuse or dependence); and the gambling and mental health measures are merely suggestive of some kind of a problem in those domains, not necessarily of severe disorder. It is also interesting to note that the rates of comorbidity found among this primarily Mexican American Texas border population are almost identical to those found in a major study of over 3000 Mexican Americans in Fresno County, California (Vega et al., 2003): alcohol abuse/dependence alone = 5.6% in TX vs 6.3% in CA; alcohol and drugs = 2.8% in TX vs 2.2% in CA; and alcohol and non-substance problems=3.5% in TX vs 3.4% in CA. The greater comorbidity among those born in the US was also a finding of both studies. * p<.05