Nearly 20% of hospitalized older adults return to the hospital within 30 days of discharge, and a large proportion visit emergency departments (ED) in this same period. Some of these readmissions and ED visits could be prevented if patients received discharge planning or care coordination. However, health behaviors could also influence outcomes following discharge. In particular, unhealthy alcohol use places older adults at increased risk of adverse outcomes due to poor medication adherence, medication interactions, vulnerability to infections, and falls. Despite these heightened vulnerabilities, clinicians rarely assess or address unhealthy alcohol use among older adults. Yet, there are many effective treatments that could be provided to hospitalized patients with unhealthy alcohol use. As a result, The Joint Commission recently developed measures that incentivize hospitals to offer alcohol-related care.
The aims of the proposed research are to determine whether unhealthy alcohol use that is identified through routine alcohol screening prior to hospitalization is associated with 1) increased risk of readmission, 2) increased risk of ED use, and 3) increased total acute care costs, during the 30 days following discharge. METHODS: The research will use a novel dataset that includes routinely collected alcohol screening results from the Veterans Affairs (VA) Health Care System, and VA and Medicare health care utilization data. The sample will include an estimated 431,349 VA patients age 65 or older (418,409 men, 12,940 women) who were enrolled in Medicare, had a VA or Medicare hospitalization for a medical or surgical condition (2008-2011), and were screened for unhealthy alcohol use in the prior year. Main outcomes for Aims 1 and 2 are any hospital readmission and any ED visit, in the 30 days after hospital discharge. The adjusted probabilities of 30-day readmission and ED use for patients with unhealthy alcohol use will be estimated using hierarchical logistic regression models to account for the correlation of patient outcomes within VA medical facilities (n=139), adjusting for important covariates. Generalized linear models will be used to examine the association between unhealthy alcohol use and total acute care costs in the 30 days post-discharge. IMPACT: The results of this study will have important implications for clinical practice and patient safety. If the proposed research demonstrates increased risk for 30-day hospital readmissions or ED visits among older adults with unhealthy alcohol use, these findings would highlight an important and often overlooked opportunity to target quality improvement efforts. Hospitals could use alcohol screening results to deliver appropriate care to patients at heightened risk for adverse outcomes. Analyses of whether unhealthy alcohol use is associated with increased acute care costs in the 30 days after discharge will suggest the magnitude of potential cost savings if outcomes can be improved.

Public Health Relevance

Older adults who drink alcohol at unhealthy levels may be at risk for adverse outcomes after hospital discharge. The proposed study is the first large-scale study to evaluate whether hospitalized older adults with unhealthy alcohol use are at increased risk for hospital readmission or emergency department visits, and associated costs, in the 30 days following hospital discharge. If older patients with unhealthy alcohol use have poorer outcomes and increased costs after discharge, this would highlight a critical opportunity to use proven interventions to improve the quality and safety of care for this vulnerable population.

National Institute of Health (NIH)
Agency for Healthcare Research and Quality (AHRQ)
Dissertation Award (R36)
Project #
Application #
Study Section
HSR Health Care Research Training SS (HCRT)
Program Officer
Willis, Tamara
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of Washington
Schools of Public Health
United States
Zip Code
Lapham, Gwen T; Rubinsky, Anna D; Williams, Emily C et al. (2014) Decreasing sensitivity of clinical alcohol screening with the AUDIT-C after repeated negative screens in VA clinics. Drug Alcohol Depend 142:209-15