Methodological aspects of randomized controlled trials on cognitive interventions.

The aims of this article were to assess the methodological challenges of randomized controlled trials (RCTs) on cognitive interventions for people with multiple sclerosis, and to discuss specific problems related to rehabilitation trials. A bibliographic search was carried out. Selected articles were examined using five methodological quality criteria related to description of the target population, method of allocation, masking of patient and examiner, the outcome measures and analysis. A total of 12 RCTs, published between 1996 and 2010, were found. Five RCTs were on drug therapy (two on disease-modifying therapies, one on donepezil and two on drugs for fatigue). Seven RCTs were on rehabilitation (six on PC-based retraining techniques). The quality of the studies was overall low, inadequate selection of the target population and limited study size being the most frequent limitations. No clear improvement was found in the most recent studies. Additional methodological problems specific to rehabilitation trials are difficulties in defining and reproducing the study intervention.