Compliance among heavy alcohol users in clinical drug trials

A high level of patient compliance is essential in determining the efficacy and safety of treatment with new drugs. Alcohol abusers are generally considered to exhibit a low level of compliance with respect to the taking of medication. This study describes a selection strategy and compares methods of determining compliance in two clinical trials assessing the effects of zimelidine and citalopram on ethanol intake in nondepressed male heavy drinkers (greater than 4 drinks/day). Objective methods included the use of a tracer substance (riboflavin), measurement of neurochemical effects (serotonin uptake inhibition), and measurement of drug and metabolite concentrations in body fluids (blood and urine). Other methods included self-report and pill count. Mean compliance estimates for the taking of medication ranged from 78 to 97% depending on the method used. Methods to determine ethanol use included daily self-monitoring and daily urine alcohol measurement. Alcohol consumption correlated with urine alcohol measurement (r = 0.62, p less than .001). These comprehensive assessments of compliance allowed accurate evaluations of drug efficacy and toxicity. The study demonstrates that with stringent selection and reinforcement procedures, exceptionally high levels of compliance can be attained in this group of subjects. Similar procedures should be included in trials evaluating new drugs.