This study compared the Glasgow Modified Alcohol Withdrawal Scale (GMAWS) and a newly devised 3-item “Anxiety Sweats Tremor” Scale (AST) to the Revised Clinical Institute Withdrawal Assessment Scale (CIWA-Ar)—the standard of care for symptom-triggered management of alcohol withdrawal syndrome.
Our study took place over 2 separate 1-week observational periods, and included 332 serial evaluations from 85 unique patients. All study participants were treated per hospital protocol based on CIWA-Ar, with supplemental scoring initially by GMAWS and later by AST in tandem. Internal consistency, interitem correlation, and operational characteristics were explored.
Median CIWA-Ar score across both phases was 6 (range 0–13), with a median GMAWS score of 2 (range 0–5) and an AST score of 3 (range 0–7). The internal consistency of CIWA-Ar and GMAWS were both poor, with Cronbach alpha scores of 0.46 (n = 156) and 0.41 (n = 156), respectively. The internal consistency of the AST scale was significantly better, with a Cronbach alpha of 0.68 (n = 176). AST identified individuals with CIWA-Ar ≥8 with an area under the receiver-operating characteristic curve of 0.83 (95% confidence interval 0.77–0.89), compared with 0.81 (95% confidence interval 0.74–0.88) for GMAWS. An AST score of ≥3 (out of a possible 9) predicted CIWA-Ar ≥8, with a sensitivity of 93% and a specificity of 63%, whereas the GMAWS had a sensitivity and specificity of 100% and 12%, respectively, based on previously defined cut-offs.
A simple 3-item scale demonstrated good internal consistency and reliably identified individuals experiencing significant alcohol withdrawal. This scale needs to be tested in other settings and among patients with a broader spectrum of withdrawal severity.