BACKGROUND: Recent research suggests the effectiveness of short-term highly intensive treatment approaches in the chronic stage of aphasia. However, the effective elements of such treatment need to be determined. AIMS: The present study's aim was to evaluate which factors attribute to the success of aphasia therapy. An intensive (3 hours/day, 10 consecutive days) model-orientated aphasia therapy (MOAT), which considers patients' individual symptoms, was evaluated and therapy effects were compared to those of a similarly intensive training focusing on active speaking elements (constraint-induced aphasia therapy, CIAT) in order to identify the effective elements. METHODS AND PROCEDURES: 12 patients with chronic aphasia received 30 hours of MOAT over 10 days. Language functions were assessed with a standardised language test (Aachen Aphasia Test) and a naming task prior to therapy, after therapy, and at a 6-month follow-up. In addition, the amount and quality of communication were assessed with questionnaires as an indication of transfer to everyday communication. Results of this treatment group were compared to those of 27 patients who were treated according to principles of CIAT. OUTCOMES AND RESULTS: Language functions improved significantly following treatment relative to the pre-treatment scores, and the improvements remained stable across the follow-up period. Effects were comparable to those of CIAT for most variables, except for written language and perception of everyday communication which improved more after MOAT than after CIAT. The naming task disclosed generalisation to untreated items for MOAT. CONCLUSIONS: Results confirm that an intense training focused on individual deficits leads to substantial and durable improvements in language functions in patients with chronic aphasia. The comparison across treatments suggests consideration of the functional deficit, written language, and everyday communication as effective elements in the rehabilitation of chronic aphasia.