INTRODUCTION Evaluation of Treatment

T he Eva difficulties involved in attempting to evaluate the effects of treatment for alcohol‐related problems are, in complexity and subtlety, second only to the process of treatment itself. Much of the research done in the past utilizes a simple input—output model of evaluation. People presumed to be “alcoholic” by some criteria are studied upon entry into a specific in‐or outpatient setting. This is defined as “day 1” of treatment, regardless of how little or how much care the subjects have had in the past. Treatment is categorized rather grossly by quantity or type. The people leaving treatment are followed and remeasured.