In their recent commentary, Largent et al1 propose guidance to help clinicians distinguish between well-supported and unsupported off-label prescribing. In 2008, we published a related analysis using nationally representative data from a survey of office-based physicians to identify the frequency of unsupported off-label use of specific therapies.2 We identified medications for which research on off-label use is most urgently needed based on recency of market entry, known adverse effects, high medication cost, high levels of promotional expenditures, and high volume of uses with insufficient evidence. Largent et al1 discussed these same features with the exception of promotional efforts, a factor we consider crucial, given the potential impact of marketing on prescribing. Among the top 25 prioritized medications, antidepressants (escitalopram, trazodone, sertraline, bupropion, amitriptyline, and venlafaxine), antipsychotics (quetiapine and risperidone), and anxiolytics/sedatives (zolpidem, lorazepam, and clonazepam) were prominent. Given the complexities that clinicians face in assessing the evidence base supporting different therapies,3 particular vigilance may be appropriate when using these treatments.