Clinical trial design - towards better practices

Objectives To review methodological approaches used in recent publications in onychomycosis clinical research. To identify key methodological criteria to ensure conclusive and reliable clinical results. Methods A Medline search for recent articles on onychomycosis treatment revealed considerable variation in inclusion criteria, definitions and methodology, rendering comparisons difficult. Results Careful diagnosis at trial entry is critical and predisposing factors affecting overall results are rarely considered at enrolment. Clear definitions are required; notably for the terms mycological, clinical and total cure. A consensus was reached that only studies that are evidence-based, controlled and double-blind with less than 10-15% of patients lost to follow-up should be accepted. Results should be interpreted with caution if a given treatment is found to be less effective than previously reported, even if the results are published in a recognized journal. It was agreed that an explanatory/pilot study should be done initially to determine whether there is a reason to believe that a new therapy is effective. If promising results are obtained, a double-blind, randomized study comparing the new therapy with either an existing therapy (preferably) or a placebo may be initiated. Conclusions Sample size, and inclusion and exclusion criteria should be clearly determined. Efficacy criteria should include mycological, clinical and total cure rates. Finally, studies of toenail and fingernail orrychomycosis must last at least 18 and 9 months, respectively.