BACKGROUND: Self-management education is associated with improvement in quality of life and reduction of hospital admissions. Despite this, data are insufficient to formulate clear recommendations regarding the type and content of education programs for COPD patients and few data are available on knowledge of the disease itself. AIM: To test: i) the level of patients' knowledge of their disease and therapy at baseline and after an educational program (COPD-EP), ii) the feasibility of structured educational sessions, iii) influence of clinical status (degree of severity of disease, presence of comorbidities, oxygen use), demographics status (age, sex), previous knowledge level, previous lessons attendance and adherence of COPD-EP to the variation of knowledge after program. Methods Selected COPD inpatients and outpatients referred to Rehabilitative hospital departments were enrolled. The study was divided in two parts: i) Pre Study Phase (educational materials and health team preparation) and ii) Study Phase: all COPD patients received one educational brochure and were invited to attend seven 30-min group lessons to complete the educational program. Learning effect was evaluated by a 20-questions multiple choice Learning Questionnaire (LQ). RESULTS: 158 patients were enrolled. 69.7 % of patients had previous formal education lessons on COPD management and 44.9 % had previous rehabilitative hospitalizations. At baseline, LQ total score was 15.2±3.5 points which increased to 16.9±3.0 point post COPD-EP (p<0.001). Pre-to-post change of LQ scores significantly correlated with adherence (R=0.24, p=0.002) and Severity Index of Cumulative Illness Rating Scale score (R= -0.22, p=0.001). Patients with low baseline knowledge were more likely to have improved LQ scores than patients with greater levels of knowledge. Patients without prior educational COPD lessons improved more than patients which attended previous education. CONCLUSIONS: A formal COPD-EP is feasible and effective in improving patient knowledge and self-management. Specific learning instruments to follow in this population should be validated.