The broad, long-term objective of this research is to prevent and intervene in alcohol abuse and dependence through the recognition of alcohol abusing and dependent patients in primary health care settings. Three common screens for alcohol abuse and dependence will be evaluated in this study: 1) the CAGE, 2) the Self-Administered Alcohol Screening Test (SAAST), and 3) the Alcohol Use Disorders Identification Test (AUDIT).
The specific aims of this study are 1) to determine the prevalence of alcohol abuse and dependence among White, Black, and Mexican-American, male and female primary care patients, and 2) to test for bias in the CAGE, SAAST, and AUDIT as screens for alcohol abuse and dependence by examining the operating characteristics (sensitivity, specificity, and predictive value) across patient sex and race/ethnicity. A sample of 1,350 patients (150 males and 300 females from each ethnic/racial subgroup) will be drawn using a systematic probability sampling procedure from patients presenting at the Family Medicine Clinic, University of Texas Medical Branch, over a 19-month period. After undergoing an initial demographic profile, eligible patients will complete the SAAST and AUDIT questionnaires. An interview will follow, including the CAGE questions and a diagnostic interview using the Alcohol Use Disorders and Associated Disabilities Interview Schedule (AUDADIS) to make a diagnosis of alcohol abuse or dependence consistent with DSM-III-R diagnostic criteria (and ICD-10, and proposed DSM-IV). Data from the AUDADIS will serve as the gold standard against which to determine the sensitivity, specificity and predictive value of the CAGE, SAAST, and AUDIT for each patient subgroup. Receiver Operating Characteristic (ROC) Curve analysis will be performed to compare the screens. Likelihood ratios and posterior probabilities will be computed for various cut-off scores and prior probabilities. As a secondary analysis, discriminant function analysis will be used to evaluate the contribution of age, depression, and anxiety (and acculturation for Mexican-American patients) to overall classification accuracy of each screen. It is expected that the findings from this study will provide primary care providers with valid and accurate data on the likelihood of alcohol problems in their patients, accounting for patient sex and ethnicity (White, Black, or Mexican-American), and lead to the development of more optimal decision making algorithms for the recognition and management of alcohol problems in primary health care settings.
Showing the most recent 10 out of 11 publications
