Commentary: The history and place of n-of-1 trials: a commentary on Hogben and Sim

Randomized trials estimate the average difference between two treatments, but nobody is average. Clinicians have, therefore, long been wary in how they apply the results of parallel group trials to indi-vidual patients. Their concern is the variation in re-sponse to treatment, seen daily in clinics and hospital wards. A treatment may be better on average, but we cannot be sure how that applies to an individual until it has been tested on him or her. In their 1953 article1 on ‘self-controlled trials’, Hogben and Sim echo this long-standing concern: ‘such variability of response to treatment may be of paramount interest in practice.’ Rather than simply complain about this problem of generalizability, however, they offered an alternative: the ‘self-controlled and self-recorded ’ clinical trial, o