Quality concerns with routine alcohol screening in VA clinical settings

Area of Science:

• Public Health
• Clinical Medicine
• Health Services Research

Background:

• Alcohol screening questionnaires are typically validated for self- or researcher-administration.
• Limited data exists on the performance of alcohol screening questionnaires in routine clinical settings.

Purpose of the Study:

• To compare alcohol screening results from routine outpatient clinical care within the Veterans Affairs (VA) Health Care System to results from the same questionnaire completed via mailed survey.
• To identify factors associated with discordant alcohol screening results between clinical and survey administration.

Main Methods:

• A cross-sectional study analyzed data from 6,861 VA outpatients (fiscal years 2007-2008).
• Compared Alcohol Use Disorders Identification Test-Consumption (AUDIT-C) results from clinical documentation and mailed surveys completed within 90 days.
• Used multivariable logistic regression to assess discordance prevalence across patient subgroups, VA networks, and temporal factors.

Main Results:

• A higher percentage of patients screened positive for unhealthy alcohol use on mailed surveys (11.1%) compared to clinical screens (5.7%).
• Discordance was substantial: 61.2% of patients who screened positive on the survey screened negative in the clinic.
• Factors associated with increased discordance included being Black, a positive survey AUDIT-C screen, and receiving care in specific VA networks.

Conclusions:

• The use of a validated alcohol screening questionnaire alone does not guarantee screening quality.
• Clinical alcohol screening quality requires monitoring, irrespective of the validation status of the questionnaire employed.