N-of-1 trials are a tapestry of heterogeneity.

Objectives: To summarize the methods of design, analysis, and meta-analysis used in N-of-1 trials.
Study Design and Setting: Electronic search for English language articles published from 1950 to 2013. N-of-1 trials were selected if they followed an ABAB design and if they assessed a health intervention for a medical condition. Elements of design, analysis, and meta-analysis were extracted.
Results: We included 100 reports representing 1,995 participants. N-of-1 trials have been conducted in over 50 health conditions. Most reports incorporated the use of elements that maintain methodological rigor, including randomization, blinding, and formal outcome assessment; however, many failed to address trial registration, funding source, and adverse events. Most reports statistically analyzed individual N-of-1 trials; however, only a small proportion of included series meta-analyzed their results.
Conclusions: N-of-1 trials have the ability to assess treatment response in individual participants and can be used for a variety of health interventions for a wide range of medical conditions in both clinical and research settings. Considerable heterogeneity exists in the methods used in N-of-1 trials.
(Copyright © 2016 Elsevier Inc. All rights reserved.)