CRD summary

This review assessed performance of brief alcohol screening questionnaires in identifying risk drinking in pregnant women. The authors concluded that T-ACE, TWEAK and AUDIT-C showed promise for identifying risk drinking and AUDIT-C for identifying alcohol dependency or abuse; further evaluation was needed. These conclusions were appropriately cautious.

Authors' objectives

To assess the performance of brief alcohol screening questionnaires in identifying risk drinking in pregnant women.

Searching

MEDLINE, EMBASE, PsycINFO, CINAHL and British Nursing Index were searched from inception to June 2008. Types of search terms used (included terms for alcohol and drinking behaviour and statistical terms associated with tests, such as sensitivity and specificity) were described. Bibliographies of reviews and included studies were screened for additional articles. Forward citation searching was undertaken using Science Citation Index and PubMed related articles functions.

Study selection

Cohort or cross-sectional studies that compared one or more brief alcohol screening questionnaires with an appropriate reference standard (structured interview) for identification of risk drinking, alcohol abuse or dependency in pregnant women were eligible for inclusion.

Most included studies were of women at first antenatal appointments in urban settings who reported some lifetime alcohol consumption; one study explicitly excluded substance abuse/dependence. All studies were conducted in USA. Three studies included only ethnic minority women of low socio-economic status and two included a more diverse group representative of the general population. Mean age of participants (where reported) ranged from 24 to 30 years.

Questionnaires evaluated by included studies were TWEAK (Tolerance, Worried, Eye-opener, Amnesia, Kut down), T-ACE [Take (number of drinks), Annoyed, Cut down, Eye-opener], CAGE (Cut down, Annoyed, Guilt, Eye-opener], NET (Normal drinker, Eye-opener, Tolerance), AUDIT (Alcohol Use Disorder Identification Test), AUDIT-C (AUDIT-consumption) and SMAST (Short Michigan Alcohol Screening Test). Various cut-off scores were used to define risk drinking (details given in paper). Various reference standard methods were used to define risk drinking (details given in paper) based on pre-pregnancy/lifetime, periconception or during pregnancy consumption.

The authors did not state how many reviewers selected studies for inclusion.

Assessment of study quality

Methodological quality of included studies was assessed using the QUADAS tool to assess appropriateness of patient spectrum, aspects of reporting quality, verification biases, incorporation bias, disease progression bias, review biases (blinding) and handling of indeterminate results and withdrawals.

Two reviewers independently assessed study quality. Disagreements were resolved by consensus.

Data extraction

Data were extracted on sensitivity, specificity and positive predictive value (PPV) for each questionnaire and cut-off threshold assessed. For studies that evaluated multiple cut-off thresholds for the same questionnaire, the area under the receiver operating characteristic (ROC) curve was also extracted, where possible.

Data were extracted by two reviewers independently. Discrepancies were resolved by discussion.

Methods of synthesis

Studies were combined in a narrative synthesis.

Results of the review

Five studies (n=6,724) were included in the review. Study quality was considered to be generally good. Only two QUADAS items were not met by any studies (interpretation of screening test result without knowledge of reference standard result and the converse). One study was susceptible to verification bias. In four studies questionnaires were not administered as independent instruments, but were derived from other questionnaires.

Risk drinking prevalence ranged from 4.3% to 30%, which reflected the different cut-offs applied.

Risk drinking: Sensitivity was highest for T-ACE (69% to 88%, using cut-off score ≥2; three studies), TWEAK (71% to 91%, using cut-off score ≥2; two studies) and AUDIT-C (95% using cut-off score ≥3; one study). Specificity ranges were 71% to 89% (T-ACE), 73% to 83% (TWEAK) and 85% (AUDIT-C). CAGE (cut off score ≥2; two studies) and SMAST (one study) performed poorly with reported sensitivities less than 50%.

Past year alcohol dependence: One study reported data on AUDIT-C for past-year alcohol dependence and alcohol abuse disorder. Using a cut-off score of ≥3, sensitivity was 100% for past year alcohol dependence and 96% for past year alcohol use disorder; specificity was 71% in both cases.

Lifetime alcohol dependence: One study assessed the AUDIT score for lifetime alcohol dependency and found very poor sensitivity (maximum 23% at a cut-off score ≥8).

Authors' conclusions

T-ACE, TWEAK and AUDIT-C show promise for screening for risk drinking, and AUDIT-C may also be useful for identifying alcohol dependency or abuse. However, the performance of these questionnaires as standalone tools is uncertain, and further evaluation is warranted.

CRD commentary

This review clearly defined its inclusion criteria, searched a number of sources for relevant studies and applied methods to reduce error and bias during the data extraction process. The authors stated that they did not use study design filters in their searches as these filters can lead to relevant studies being missed; however, it was likely that use of methodological search terms related to tests (sensitivity and specificity), as described, would similarly reduce search sensitivity. It was unclear whether measures were taken to minimise error/bias during the study selection process. Use of a narrative synthesis was appropriate given the range of instruments assessed and reference standards applied. Overall, the authors' conclusions were appropriately cautious.

Implications of the review for practice and research

Practice: The authors made no recommendations for practice in respect of screening for risk drinking using brief questionnaires.

Research: The authors stated that these screening tools should be evaluated in different settings and populations in order to select the optimal instruments. It would also be useful to evaluate their clinical and cost-effectiveness when used in conjunction with brief interventions for risk drinking and alcohol use disorders in pregnant women

Funding

No external funding.

Bibliographic details

Burns E, Gray R, Smith LA. Brief screening questionnaires to identify problem drinking during pregnancy: a systematic review. Addiction 2010; 105(4): 601-614. [PubMed: 20403013]

Original Paper URL

http://onlinelibrary.wiley.com/doi/10.1111/j.1360-0443.2009.02842.x/abstract

Indexing Status

Subject indexing assigned by NLM

MeSH

Adolescent; Adult; Alcohol Drinking /adverse effects; Alcoholism /complications /diagnosis; Cross-Sectional Studies; Databases, Bibliographic; Female; Humans; Mass Screening /methods; Pregnancy; Prenatal Diagnosis /methods; Prenatal Exposure Delayed Effects /etiology; Questionnaires /standards; Reference Standards; Sensitivity and Specificity; Young Adult

AccessionNumber

12010003109

Database entry date

20/10/2010

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.