The present study evaluated a commitment-based intervention to improve adherence with a 10-day antibiotic regimen. Subjects were 48 undergraduate students receiving treatment from the Student Health Services at Virginia Polytechnic Institute and State University. Adherence was assessed by pill counts obtained in unannounced home visits 7 to 10 days after the regimen was prescribed and by subjects’ self-reports. Pre- and posttest measures of self- efficacy and outcome expectancy were completed by subjects. Significantly more subjects in the intervention group were adherent (85%) than in the control group (64%) when adherence was defined as at least 80% of medication taken and nonadherence defined as less than 80% or more than 110% of medication taken. Self-efficacy and outcome expectancy scores, when multiplied together for a predictive index, were significantly correlated with self-reported adherence at pre- and posttest, but were not correlated with pill count adherence. Self-efficacy at pre- and posttest and outcome expectancy at pretest were significantly correlated with self-reported adherence. Self-efficacy at posttest was significantly correlated with pill count adherence. An intervention designed to increase commitment to medical regimen resulted in greater adherence with a short-term regimen. Self-efficacy and outcome expectancy predicted self-reports of adherence behavior, but were poor predictors of objective measures of adherence. At best, self-efficacy appeared to reflect recent behavior. / Master of Science