2005 HSR&D National Meeting Abstract
3068 — Can Health Behaviors Identify Patients at Increased Risk for Potentially Avoidable Hospitalizations?
The Agency for Healthcare Research and Quality (AHRQ) has proposed 13 discharge diagnoses to estimate the quality of ambulatory care for health systems, but little is known about how differences in patient health behaviors might impact these measures. We evaluated the association between two types of potentially avoidable hospitalizations and screening results for alcohol use and smoking.
This cohort study included VA primary care outpatients who completed alcohol and smoking screening questions for the VA Ambulatory Care Quality Improvement Project (1997-2000). Patients were divided into four groups based on AUDIT-C alcohol screening scores: 0 (nondrinking), 1-3 (low-risk drinking), 4-7 (medium-risk driking), 8-12 (high-risk drinking). Smoking was categorized “current” or “never/past” smoking. VA and Medicare data were used to identify two types of primary discharge ICD-9 diagnoses for potentially avoidable hospitalizations as the main outcomes: 1) those proposed by AHRQ as Adult Ambulatory Care Sensitive Conditions and 2) those indicating trauma. Main analyses used Cox proportional hazards regression to model the association between alcohol and smoking measures and time to the first avoidable hospitalization, adjusting for demographics. Patients with low-risk drinking and never/past smoking were used as referent groups.
Among 35,055 patients followed for 2 years, the risk of admission for any potentially avoidable AHRQ condition was not significantly increased for patients with medium or high-risk drinking. The risk of admission for trauma, however, was increased for patients with high-risk drinking (HR=1.82; 1.35-2.45). Current smoking was associated with increased risk of admission for any potentially avoidable AHRQ admission (HR=1.52;1.42-1.64), and trauma (HR=1.42;1.20-1.70).
High-risk drinking was not associated with increased avoidable hospitalizations proposed by AHRQ to evaluate quality of care, but was associated with increased risk of admission for trauma. Current smoking was associated with increased risk of both types of potentially avoidable hospitalizations
VA patients are screened for alcohol consumption and smoking history. These may be readily available markers for potentially avoidable hospitalizations.