Background: There is conflicting evidence regarding the benefits of intensive speech and language therapy (SLT), particularly because intensity is often confounded with total SLT provided. Aims: A two-centre, randomized, rater-blinded, parallel study was conducted to compare the efficacy of 100 h of SLT in a regular (RT) versus intensive (IT) treatment in sub-acute post-stroke aphasia. Methods & Procedures: Consecutive patients with aphasia, within 3 months of a left hemisphere ischemic stroke, were randomized to IT (2 h per day x 5 days per week, 10 weeks) or RT (2 h per week x 50 weeks). Evaluations took place at 10, 50 and 62 weeks. Primary outcome was the frequency of responders, defined by 15% increase of Aphasia Quotient (AQ) from the baseline to 50 weeks. Secondary outcomes were changes from the baseline in AQ and functional communication profile (FCP) at 50 and 62 weeks and improvement stability between 50 and 62 weeks. Outcomes & Results: Thirty patients were randomized and 18 completed the study. No significant differences were found between groups in primary or secondary outcomes, although IT patients ("N" = 9) obtained higher scores in language measures between 10 and 62 weeks in per protocol analysis. The number of non-completions was identical between groups. Conclusions & Implications: This study suggests that, in the sub-acute period following stroke and controlling for the number of hours of SLT provided, there is a trend for a greater improvement in language and functional communication measures with IT compared with RT. The lack of statistical significance in results was probably due to the small sample size. (Contains 6 tables and 3 figures.)