Reader response: Evaluating depression and suicidality in tetrabenazine users with Huntington disease

Abstract

Schultz et al.1 concluded that tetrabenazine (TBZ) use is not associated with depression or suicidality. This analysis of cross-sectional observational data from the Enroll-HD dataset contradicts prospective experimental evidence.2 This alone should caution against the authors' conclusion,1 but we have other concerns. Principally, 2 sources of bias may have distorted the observed associations. First, only individuals who tolerated TBZ are captured as users—those who abandoned treatment (e.g., due to depression/suicidality) are not counted. This survival bias would serve to induce a spurious negative association between TBZ use and depression/suicidality (i.e., a bias in the direction of an apparent protective effect). Second, antidepressant use may result from TBZ use (if this causes depression) and from depression itself (from other causes). Adjusting for this, as in this study,1 may introduce collider bias, and—again—a spurious negative association. These problems are compounded by the cross-sectional nature of the analysis, which precludes definition of incident depression/suicidality, despite prospective data being available within Enroll-HD. The conclusion that TBZ may be safe for patients with Huntington disease with a history of depression/suicidality is flawed and potentially dangerous. We welcome use of the Enroll-HD dataset, and CHDI Foundation funds an independent statistical standing committee to provide support and expertise (see supplemental full-length version, [links.lww.com/WNL/A825][1]). [1]: http://links.lww.com/WNL/A825