Objective.— To describe the performance of alcohol screening questionnaires in female
patients.
Data Sources.— We searched MEDLINE from 1966 to July 1997 for alcoholism or alcohol-drinking and for CAGE, AUDIT, BMAST, TWEAK, T-ACE, MAST, SMAST, or SAAST; Citations Indexes for newer screening questionnaires and those
without acronyms; and MEDLINE from 1996 to July 1997 for alcoholism or alcohol-drinking and screening.
Study Selection and Data Extraction.— Reviewed studies presented data for women comparing brief alcohol screening
questionnaires with valid criterion standards for heavy drinking (≥2 drinks
per day) or alcohol abuse or dependence in US general clinical populations.
Sensitivities, specificities, and areas under receiver operating characteristic
curves (AUROCs) were extracted.
Data Synthesis.— Thirteen articles (9 studies) were reviewed. The CAGE questionnaire
had AUROCs of 0.84 to 0.92 for alcohol abuse and dependence in predominantly
black populations of women, but using the traditional cut point of 2 or more
resulted in low sensitivities (38%-50%) in predominantly white female populations.
The TWEAK and Alcohol Use Disorders Identification Test (AUDIT) questionnaires
had high AUROCs (0.87-0.93) for past-year alcohol abuse or dependence in black
or white women, but had sensitivities less than 80% at traditional cut points.
For detecting heavy drinking, the AUDIT questionnaire had AUROCs of at least
0.87 in female primary care patients. The TWEAK and T-ACE questionnaires had
higher AUROCs (0.84-0.87) than the CAGE questionnaire (0.76-0.78) for detecting
heavy drinking before pregnancy was recognized in black obstetric patients.
Conclusions.— The CAGE questionnaire was relatively insensitive in predominantly white
female populations. The TWEAK and AUDIT questionnaires have performed adequately
in black or white women, using lower cut points than usual.