José R Maldonado1, Yelizaveta Sher2, Smita Das2, Kelsey Hills-Evans2, Anna Frenklach2, Sermsak Lolak3, Rachel Talley4, Eric Neri2. 1. Stanford University School of Medicine, Stanford, CA, USA jrm@stanford.edu. 2. Stanford University School of Medicine, Stanford, CA, USA. 3. The George Washington University School of Medicine and Health Sciences, Washington, DC, USA. 4. New York Presbyterian/Columbia University Medical Center, NewYork, NY, USA.
Abstract
AIMS: The prevalence of alcohol use disorders (AUDs) among hospitalized medically ill patients exceeds 40%. Most AUD patients experience uncomplicated alcohol withdrawal syndrome (AWS), requiring only supportive medical intervention, while complicated AWS occurs in up to 20% of cases (i.e. seizures, delirium tremens). We aimed to prospectively test and validate the Prediction of Alcohol Withdrawal Severity Scale (PAWSS), a new tool to identify patients at risk for developing complicated AWS, in medically ill hospitalized patients. METHODS: We prospectively considered all subjects hospitalized to selected general medicine and surgery units over a 12-month period. Participants were assessed independently and blindly on a daily basis with PAWSS, Clinical Institute Withdrawal Assessment-Alcohol, Revised (CIWA-Ar) and clinical monitoring throughout their admission to determine the presence and severity of AWS. RESULTS: Four hundred and three patients were enrolled in the study. Patients were grouped by PAWSS score: Group A (PAWSS < 4; considered at low risk for complicated AWS); Group B (PAWSS ≥ 4; considered at high risk for complicated AWS). The results of this study suggest that, using a PAWSS cutoff of 4, the tool's sensitivity for identifying complicated AWS is 93.1% (95%CI[77.2, 99.2%]), specificity is 99.5% (95%CI[98.1, 99.9%]), positive predictive value is 93.1% and negative predictive value is 99.5%; and has excellent inter-rater reliability with Lin's concordance coefficient of 0.963 (95% CI [0.936, 0.979]). CONCLUSION: PAWSS has excellent psychometric characteristics and predictive value among medically ill hospitalized patients, helping clinicians identify those at risk for complicated AWS and allowing for prevention and timely treatment of complicated AWS.
AIMS: The prevalence of alcohol use disorders (AUDs) among hospitalized medically ill patients exceeds 40%. Most AUD patients experience uncomplicated alcohol withdrawal syndrome (AWS), requiring only supportive medical intervention, while complicated AWS occurs in up to 20% of cases (i.e. seizures, delirium tremens). We aimed to prospectively test and validate the Prediction of Alcohol Withdrawal Severity Scale (PAWSS), a new tool to identify patients at risk for developing complicated AWS, in medically ill hospitalized patients. METHODS: We prospectively considered all subjects hospitalized to selected general medicine and surgery units over a 12-month period. Participants were assessed independently and blindly on a daily basis with PAWSS, Clinical Institute Withdrawal Assessment-Alcohol, Revised (CIWA-Ar) and clinical monitoring throughout their admission to determine the presence and severity of AWS. RESULTS: Four hundred and three patients were enrolled in the study. Patients were grouped by PAWSS score: Group A (PAWSS < 4; considered at low risk for complicated AWS); Group B (PAWSS ≥ 4; considered at high risk for complicated AWS). The results of this study suggest that, using a PAWSS cutoff of 4, the tool's sensitivity for identifying complicated AWS is 93.1% (95%CI[77.2, 99.2%]), specificity is 99.5% (95%CI[98.1, 99.9%]), positive predictive value is 93.1% and negative predictive value is 99.5%; and has excellent inter-rater reliability with Lin's concordance coefficient of 0.963 (95% CI [0.936, 0.979]). CONCLUSION: PAWSS has excellent psychometric characteristics and predictive value among medically ill hospitalized patients, helping clinicians identify those at risk for complicated AWS and allowing for prevention and timely treatment of complicated AWS.
Authors: Pratik P Pandharipande; E Wesley Ely; Rakesh C Arora; Michele C Balas; Malaz A Boustani; Gabriel Heras La Calle; Colm Cunningham; John W Devlin; Julius Elefante; Jin H Han; Alasdair M MacLullich; José R Maldonado; Alessandro Morandi; Dale M Needham; Valerie J Page; Louise Rose; Jorge I F Salluh; Tarek Sharshar; Yahya Shehabi; Yoanna Skrobik; Arjen J C Slooter; Heidi A B Smith Journal: Intensive Care Med Date: 2017-06-13 Impact factor: 17.440
Authors: S Jesse; G Bråthen; M Ferrara; M Keindl; E Ben-Menachem; R Tanasescu; E Brodtkorb; M Hillbom; M A Leone; A C Ludolph Journal: Acta Neurol Scand Date: 2016-09-01 Impact factor: 3.209
Authors: Tessa L Steel; Shewit P Giovanni; Sarah C Katsandres; Shawn M Cohen; Kevin B Stephenson; Ben Murray; Hillary Sobeck; Catherine L Hough; Katharine A Bradley; Emily C Williams Journal: Addict Sci Clin Pract Date: 2021-03-24