Screening for alcohol problems: a comparison of instrument performance among black emergency department and primary care patients

Abstract

While alcohol problems are over-represented in primary care settings and in emergency departments (EDs), screening for alcohol use disorders has not been a routine part of care in the United States, and little is known of the performance of screening instruments for problem drinking, particularly among women and ethnic minorities. The sensitivity and specificity of the CAGE, BMAST, AUDIT, TWEAK, and RAPS are compared against DSM-IV and ICD-10 criteria for alcohol dependence in probability samples of black patients interviewed in the ED (n= 1091) and primary care clinics (n = 711) in Jackson, Mississippi. Instruments appeared to perform better in the ED than in primary care. The CAGE and RAPS appeared to perform best in the primary care sample, and the AUDIT and RAPS in the ED sample. While the prevalence of alcohol dependence in the primary care clinics was lower than in the ED, findings suggest that both sites are important for screening and identification of patients with alcohol use disorders. Further research is needed for determining those screening instruments which perform optimally in identifying problem drinking patients across clinical sites.