The design, logic, and results of a two-year health education study directed at improving rates of patient adherence to antituberculosis medical regimens are presented. An incentive scheme to reward positive health behaviors plus targeted educational counseling sessions was implemented in a randomized clinical controlled trial. The 205 subjects who participated in the study are categorized according to patients with active tuberculosis (n = 88) or preventive patients with no evidence of active disease (n = 117). Patients in each of these groups were randomly assigned to a special intervention (SI) group or a usual care (UC) control group and were followed monthly throughout their treatment program. While SI patients with active tuberculosis demonstrated higher levels of appointment-keeping behavior and mean percent of medication taken compared to UC patients, no statistically significant differences between the two groups were found. Preventive therapy patients assigned to the SI group, however, were significantly more likely than UC patients to remain in care during their 12-month regimen (64% vs 47%; p = .003). Furthermore, SI patients had significantly higher levels of adherence to their medical regimen compared to UC patients (68% vs 38%; p less than .001). These results demonstrate the positive effects of a structured health education program on the improvement of continuity of care and adherence behavior among patients with tuberculosis.