Treatment of alcoholism consumes a large proportion of health care resources. In the private sector, research has demonstrated that inpatient alcohol treatment reduces health care utilization. However, little is known about its effects on alcoholics with minimal financial resources, elderly, ethnic minorities, and chronic users of health care. Analysis of health service use among alcoholics who receive treatment on the Veterans Administration (VA) Alcohol Dependency Treatment Programs (ADTPs) will lead to understanding the """"""""offset effect"""""""" on alcoholism treatment in such patients. The major objective of this study is to examine the effect of inpatient alcohol treatment on health care utilization by a large sample (approximately 15,000) alcoholic veterans, half of whom were treated for alcoholism and half were not. The study is also designed to examine: 1) the effect of inpatient alcohol treatment on the health care utilization of special groups of alcoholics, specifically those who are elderly and ethnic minorities; 2) the effect of system-specific characteristics such as ADTP admission policies, other measures of access to care, and geographic location on the health care utilization patterns of alcoholics. A nonequivalent control group design will be used to compare pretreatment and posttreatment health care utilization among veterans treated on ADTPs. Using the Patient Treatment File (PTF) and Outpatient (OPC) Survey, the VA's secondary databases relative to yearly inpatient and outpatient health care utilization by all VA patients, the utilization patterns of a random sample of approximately 7,400 alcoholics will be examined three years before and three years after ADTP treatment, compared to a nontreatment group of a similar number of alcoholics not treated on an ADTP. The treatment and nontreatment groups will be stratified by age, ethnicity and severity of alcoholism. Measures of health care utilization include frequency of hospitalizations, total length of hospital stays, major medical diagnosis categories, and number of outpatient clinic visits. Statistical methods used to analyze the data will include analysis of covariance with weighted least squares regression models for categorical data (e.g., GSK method), basic analysis of covariance with a single covariate and logistic regression analyses.
