The efficacy of attention rehabilitation after an acquired brain injury was examined meta-analytically. Thirty studies with a total of 359 participants met the authors' selection criteria. Studies were categorized according to whether training efficacy was evaluated by comparing pre- and posttraining scores only or included a control condition as well. Performance improved significantly (using the d+ statistic) after training in pre-post only studies but not in pre-post with control studies. Further analyses showed that specific-skills training significantly improved performance of tasks requiring attention but that the cognitive-retraining methods included in the meta-analysis did not significantly affect outcomes. These findings demonstrate that acquired deficits of attention are treatable using specific-skills training. Implications of these results for rehabilitation theory and future research are discussed.