Screening for Hazardous Drinking in Nursing Home Residents: Evaluating the Validity of the Current Cutoffs of the
Monika Dreher-Weber1, Anton-Rupert Laireiter2, Anton Kühberger2
1Department of Psychosomatic Medicine and Psychotherapy, Medical Park Chiemseeblick, Bernau-Felden, Germany.
Background:
Because of physiological changes, elderly people are much more exposed to the adverse effects of alcohol. Therefore, hazardous drinking is defined at lower levels as compared to younger adults. This work aimed to evaluate the validity of the current cutoff levels of the Alcohol Use Disorder Identification Test-Consumption (AUDIT-C) questions to detect hazardous drinking in the elderly by using ethyl glucuronide in hair (HEtG).
Methods:
In a border region between Austria and Germany, 344 nursing home residents were included from 33 of the 107 nursing homes. Residents were asked to answer the AUDIT-C questions, hair samples were obtained, and nursing staff members were asked for their assessments of the residents' alcohol consumption. Hair samples were analyzed for HEtG using gas chromatography-mass spectrometry. Receiver-operating characteristic (ROC) curve analysis was performed to determine the validity of cutoff values for the AUDIT-C to detect an alcohol consumption of ≥10 g of alcohol/d.
Results:
A total of 11.3% of the nursing home residents (n = 344) drank ≥10 g of alcohol/d (4.9% >60 g of alcohol/d, 6.4% 10 to 60 g of alcohol/d, 88.7% <10 g of alcohol/d)). For the drinking limit of ≥10 g of alcohol/d, ROC curve analysis showed a balanced sensitivity and specificity, with an AUDIT-C cutoff of ≥4 for men (sensitivity: 70%, specificity: 83.6%; AUC = 0.823, CI = 0.718 to 0.928, p < 0.001) and ≥2 for women (sensitivity: 73.7%, specificity: 81.9%; AUC = 0.783, CI = 0.653 to 0.914, p < 0.001). Nursing staff (n = 274) underestimated alcohol consumption and evaluated 40% of the chronic-excessive alcohol consumers (>60 g of alcohol/d) as being abstinent.
Conclusions:
Our data suggest that an AUDIT-C cutoff of ≥4 for men and ≥2 for women can be recommended to detect the consumption of ≥10 g of alcohol/d in the elderly. Because the nursing staff to a large extent underestimates the alcohol consumption among nursing home residents, further teaching of the staff, improvement of screening instruments for the elderly, and the use of objective biomarkers might be helpful for recognizing hazardous drinking and can thus help improve the quality of life of the elderly.
