To the Editor We read with interest the recent article by Kravitz and colleagues1 describing a randomized clinical trial comparing n-of-1 trials with standard care for treatment of chronic musculoskeletal pain.
The goal of the study was to establish the “benefits of participating in an n-of-1 trial, not to assess the superiority or inferiority of any particular treatment.”1(1369) However, there appears to be a disconnect between the study goal and the choice of outcomes, which were focused on pain interference scores across different treatment regimens. Therefore, the null results should be interpreted with respect to treatment efficacy, not design. The n-of-1 participants who demonstrated a better response to 1 of 2 treatments were likely to experience improved pain outcomes as a result of continuing to receive the superior treatment. However, there was a high proportion (>75%) of n-of-1 participants who had no treatment superiority, and this may explain the trial’s findings.