Tacrolimus for myasthenia gravis patients with a previous thymectomy: effect-modification, confounding, and caution

Powell, T.N. (2008) Tacrolimus for myasthenia gravis patients with a previous thymectomy: effect-modification, confounding, and caution. Epidemiologic Focus, 1. e5. ISSN 1944-3307

Full text available as:

PDF - Requires Adobe Acrobat Reader or other PDF viewer. 101 Kb

Official URL: http://www.epifocus.org/index.php/epifocus/article/view/5/46

Abstract

Mitsui et al. recently reported that patients with thymoma and previous thymectomy had improved outcomes compared to non-thymoma patients with a previous thymectomy when tacrolimus is used as the immunosuppressive agent in the treatment regimen. However, the study design may not have been conducive to evaluating potential effect modification by thymoma status. Comparisons of stratified populations that do not incorporate within-group controls may be misleading. Furthermore, statistical control for confounding by prognostic factors is required for a non-randomized trial because confounding is not addressed by randomization. Clinically significant differences in the baseline distribution of several prognostic factors between the groups could have contributed to confounding in the trial by Mitsui et al. Clinical significance is stressed in this scenario because the over-emphasis of statistical significance when evaluating prognostic factor distribution between groups often undermines the validity of a clinical trial, particularly a trial with a small sample size and thus underpowered analyses.

Archive Staff Only: edit this record