CRD summary

The authors concluded that the Alcohol Use Disorders Identification Test (AUDIT) was a useful screen for detecting harmful and hazardous drinking in the elderly and that the CAGE test was valuable when screening for dependence. The authors' conclusions reflected the evidence presented, however, due to poor reporting of review processes the reliability of the conclusions is unclear.

Authors' objectives

To evaluate paper-and-pen screening tools for alcohol misuse in elderly primary care patients.

Searching

The electronic databases MEDLINE and PsycINFO were searched to identify relevant articles written in English. Search terms were reported. The reference lists from relevant publications were searched.

Study selection

Eligible for inclusion in the review were studies of alcohol misuse pen-and-paper screens tested in a population aged mostly 60 years and over who self-presented to primary care. Studies had to use a gold standard reference test to detect alcohol dependence or abuse. The gold standard reference test for detecting alcohol dependence was a clinical interview with structured or semi-structured interviewing tool such as the Diagnostic Interview Schedule (DIS), using either the Diagnostic and Statistical Manual of Mental Disorder (DSM-IVR) or the International Classification of Diseases (ICD-10) criteria. The gold standard reference test for detecting hazardous drinking was careful questioning about consumption. Studies that gave an average age but no cut off, allowed the test to influence whether to perform the reference test or that did not have data that allowed the calculation of sensitivity and specificity were excluded from the review.

Pen-and-paper screens included in the review were: Alcohol Use Disorders Identification Test (AUDIT), AUDIT-Consumption Questions (AUDIT-C), CAGE (acronym from four key words in the questions of the test; Cut down, Annoyed, Guilty and Eye opener), Michigan Alcohol Screening Test (MAST), the geriatric version of MAST (MAST-G), short-MAST (SMAST), Alcohol Related Problems Survey (ARPS) and shortened ARPS (shARPS).

Most studies included both genders. A variety of cut-off points were used.

Two reviewers independently selected the studies for inclusion in the review; any disagreements were resolved by consensus.

Assessment of study quality

Study quality was assessed using the following criteria: demographics provided; comorbidities described; eligibility criteria and participation rate provided; criterion standard evaluation of all screen subjects; blinding; and analysis of pertinent subgroups. The authors did not state how the validity assessment was performed.

Data extraction

Data were extracted in order to calculate sensitivity and specificity. The authors did not state how many reviewers performed the data extraction.

Methods of synthesis

The authors stated that studies were compared using sensitivity and specificity at the same cut offs and, where possible, Receiver Operator Curves (ROCs) were used to determine the cut off where the test was most efficient. No further details were provided.

Results of the review

Nine studies were included in the review (6,353 patients). Sample size ranged from 103 to 5,065.

Detecting alcohol abuse and dependence: CAGE (three studies) at a cut-off of 1 or more was shown by Receiver Operator Curves from two of the studies to be the most efficient. Sensitivity ranged from 79.1% to 88% and specificity ranged from 55.8% to 88%.

Variations of MAST (four studies): The Receiver Operator Curves from two studies showed a cut-off of 4 or more was most efficient. However, only one study used used this cut-off, which gave a sensitivity of 91.4% and a specificity of 83.9%. A Receiver Operator Curve for the MAST-G suggested a cut-off of 5 or more. Two studies that used the MAST-G found that sensitivity ranged from 69.8% to 91% and specificity ranged from 80.5% to 84%.

Detecting hazardous or excessive drinking:

AUDIT (one study): a cut-off of 8 or more gave a sensitivity of 33.3% and a specificity of 90.7%.

CAGE (three studies): two studies looked at a cut off of 1 or more and found that sensitivity ranged from 31% to 60% and specificity ranged from 92% to 100%; two studies that looked at a cut off of 2 or more found that sensitivity ranged from 14% to 38.9% and specificity ranged from 97% to 97.1%.

SMAST (one study): SMAST performed poorly at a cut off of 2 or more, with a sensitivity of 48% and a specificity of 100%.

Variations of AUDIT (two studies): AUDIT at a cut-off of 8 or more showed a sensitivity of 66.7% and a specificity of 95.3%; AUDIT C at a cut-off of 3 or more showed a sensitivity of 100% and a specificity of 80.7%.

ARPS and shARPS in comparison to AUDIT and SMAST-G (one study): the study did not report the cut-off for ARPS and shARPS. Sensitivity was 93% (ARPS) and 92% (shARPS) compared to 28% for the AUDIT and 52% for the SMAST-G. Specificity was not so good at 63% and 51% for the ARPS and shARPS compared with 100% and 96% for the AUDIT and SMAST G.

Authors' conclusions

The Alcohol Use Disorders Identification Test (AUDIT) was a useful screen for detecting harmful and hazardous drinking in the elderly and the CAGE test was valuable when screening for dependence.

CRD commentary

This review addressed a clear research question. Inclusion criteria were adequate with the exception of study design. The search strategy was adequate however, it was limited to studies in English and there was no attempt to locate unpublished studies. This means that relevant studies might have been missed. The validity assessment tool was appropriate. However, the authors reported neither how many reviewers performed the validity assessment nor the data extraction, which meant that these processes may have been subject to reviewer error or bias. There was insufficient information on individual studies and on the methods of synthesis used in the review, which meant that it was not possible to determine the reliability of the results. The authors' conclusions reflected the evidence presented, however, due to the lack of reporting on review processes (in particular of methods of synthesis) the reliability of the conclusions is unclear.

Implications of the review for practice and research

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

Funding

Not stated.

Bibliographic details

Berks J, McCormick R. Screening for alcohol misuse in elderly primary care patients: a systematic literature review. International Psychogeriatrics 2008; 20(6): 1090-1103. [PubMed: 18538045]

Original Paper URL

http://journals.cambridge.org/action/displayAbstract?aid=2452960

Indexing Status

Subject indexing assigned by NLM

MeSH

Age Factors; Aged; Alcohol Drinking /epidemiology /psychology; Alcohol-Related Disorders /diagnosis /epidemiology; Alcoholism /diagnosis /epidemiology; Diagnostic and Statistical Manual of Mental Disorders; Geriatric Assessment /methods; Health Surveys; Humans; International Classification of Diseases; Mass Screening /methods; Middle Aged; Primary Health Care /statistics & numerical data; Psychiatric Status Rating Scales; Questionnaires

AccessionNumber

12009101309

Database entry date

02/09/2009

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.