Clinical recommendations made in dermatology publications are frequently not supported by adequate evidence

Objectives: Absolute risk and cost-benefit analyses are necessary to guide clinical decision making. The purpose of this study was to evaluate whether clinical recommendations in dermatology publications were supported by adequate evidence. We also assessed whether dermatology residents report sufficient evidence when they author publications. Methods: We analyzed the clinical cohort, case-control, and case series studies published in JAMA Dermatology and the Journal of American Academy of Dermatology from January 2018 to December 2018 for statistical significance, absolute associations, diagnostic and therapeutic clinical recommendations, and risk-benefit analysis. We also identified articles with a U.S. dermatology resident as first or second author. Results: We found that the majority of articles reported statistical significance, but only 3% included absolute risk analyses and none had risk-benefit analysis. Furthermore, 42% of studies with a dermatology resident as a primary author reported statistical significance, but none provided absolute risk or risk-benefit analyses. Conclusions: Reviewers need to be more aware of the evidentiary needs required for clinical recommendations, and dermatology residents may benefit from additional statistics training.

Automated summary

The majority of dermatology publications had statistical significance, but very few included absolute risk and risk-benefit analyses. Additionally, 42% of studies authored by dermatology residents reported statistical significance but lacked absolute risk or risk-benefit analyses. Reviewers need to pay more attention to evidentiary requirements for clinical recommendations, and dermatology residents may benefit from additional statistics training.