ABSTRACT
A clinical trial conducted on a single individual patient, in the absence of evidence on which to substantiate treatment decisions for that same patient. Ethical considerations and patient compliance are important, as with any medical investigation. N-of-1 trials are, for obvious reasons, more limited than larger clinical trials, but can still be useful if conducted scientifically. This involves some form of controlling, randomization and concealment of treatment allocation (with regard to the order in which the treatments are to be administered), and blind assessment of responses, in addition to repetition of observation. GUYATT et al. (2014) and STR AUS et al. (2010) recommend the use of period pairs, in which the treatment in question and a placebo or alternative treatment are both administered in random order. These two-period trials are repeated until evidence of benefit or lack thereof emerges. N-of-1 trials are in many ways similar to crossover trials, not only because patients act as their own controls, but also because they raise the same practical issues and have similar limitations. See SENN (2002; 2008), STRAUS et al. (2010) and GUYATT et al. (2014) for further details and discussion. See also hierarchy of evidence.
