Background Clinical practice guidelines (CPG’s) are predominantly based on randomised controlled trials (RCT’s). However, the number of published observational, non-randomised trials is high and yet, the information they contain is often excluded while compiling CPG’s.

Objectives We compared randomised and observational clinical treatment trials, published within one year, using multiple sclerosis (MS) as a model. The aim was to find out whether valuable information for CPG’s can be discovered by using observational studies.

Methods We collected all publications of MS treatment in Medline during year 2012, using Scottish Intercollegiate Guidelines Network (SIGN) search filtres for randomised and observational trials for making a systematic literature review. The clinical drug trials on adults with MS, published in English were included. Case reports and symptomatic treatment trials were excluded.

Results A total of 24 RCT’s and 45 observational trials were found. The median number of participants in RCT’s was 430 (range 66 to 2244) and in observational trials 118 (range 5 to 22255). Twenty RCT’s were efficacy trials, 2 evaluated health economics and one each safety and pharmacology. Twenty-eight observational trials had efficacy as a primary end-point, 10 addressed safety, 5 treatment adherence and one each pharmacology and health economics.

Discussion Most trials addressing safety or compliance issues were observational, therefore important safety and adherence data could be lost by omitting them.

Implications for Guideline Developers/Users Observational trials should be considered while compiling CPG’s.