CRD summary

This review concluded that the CAGE questionnaire is of limited diagnostic value for screening purposes at the recommended cut-point of two or more. The review was generally well-conducted and the authors' conclusions are likely to be reliable.

Authors' objectives

To determine the accuracy of the CAGE questionnaire in screening for alcohol abuse or dependence in a general clinical population.

Searching

MEDLINE was searched from 1974 to December 2001 using the reported search terms. In addition, a number of general practice journals (in Dutch) and reference lists were screened and experts were contacted. The studies could be reported in any language apart from Japanese.

Study selection

Assessment of study quality

Inclusion was based on the fulfilment of several quality-related factors: appropriate study population, appropriate reference standard, avoidance of verification bias. The included studies were assessed for the possibility of selection bias. Two reviewers independently assessed validity and resolved any disagreements through recourse to a third reviewer, if required.

Data extraction

One reviewer extracted the data and a second reviewer checked them. The sensitivity, specificity, positive and negative likelihood ratios, and positive and negative predictive values were calculated from data reported in the original articles or provided by the authors of the original studies. All measures of test performance were calculated for each of the CAGE thresholds, together with the 95% confidence intervals (CIs).

Methods of synthesis

Results of the review

Ten studies (n=4,562) were included. Of these, five were in primary care patients (n=2,708), three in ambulatory medical patients (n= 1,202) and two in in-patients (n=652).

CAGE threshold >= 1: the sensitivity ranged from 61 to 100% and the specificity from 37 to 88%.

CAGE threshold >= 2: the sensitivity ranged from 46 to 92% and the specificity from 62 to 95%.

There was significant heterogeneity between studies for all CAGE thresholds and for all measures of test performance. Sensitivity was higher and specificity lower in in-patients compared with ambulatory or primary care patients.

The pooled area under the ROC curve was 0.87 (95% CI: 85, 89).

Authors' conclusions

The CAGE questionnaire is of limited diagnostic value for screening purposes at the recommended cut-point of two or more.

CRD commentary

The review addressed a focused objective supported by clearly defined inclusion criteria that limited the review to high-quality studies. Several relevant sources were searched and attempts were made to minimise language bias. It was not clear whether the authors made specific attempts to locate unpublished studies. Appropriate steps were taken to minimise bias in the review process and the methods used to pool the data were appropriate. The authors' conclusions were supported by the results presented and are likely to be reliable

Implications of the review for practice and research

The authors did not state any implications for practice or further research.

Bibliographic details

Aertgeerts B, Buntinx F, Kester A. The value of the CAGE in screening for alcohol abuse and alcohol dependence in general clinical populations: a diagnostic meta-analysis. Journal of Clinical Epidemiology 2004; 57(1): 30-39. [PubMed: 15019008]

Other publications of related interest

This additional published commentary may also be of interest. Schorling JB. Review: sensitivity of the CAGE questionnaire for the DSM diagnosis of alcohol abuse and dependence in general clinical populations was 71% at cut points >=2. Evid Based Med 2005;10:26.

Indexing Status

Subject indexing assigned by NLM

MeSH

Alcoholism /diagnosis; Evidence-Based Medicine; Humans; Likelihood Functions; Mass Screening /methods /statistics & numerical data; Questionnaires; ROC Curve; Sensitivity and Specificity

AccessionNumber

12004000479

Database entry date

31/08/2007

Record Status

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.