Data Sparsity in Study on Human Leukocyte Antigen Class I Genes Associated With Stevens-Johnson Syndrome and Severe Ocular Complications

To the Editor We have read the article by Wakamatsu and colleagues¹ that was published in JAMA Ophthalmology. The authors evaluated the association between human leukocyte antigen class I genes and cold medicine–associated Stevens-Johnson syndrome/toxic epidermal necrolysis with severe ocular complications. One of their main findings was that human leukocyte antigen-A*66:01 was associated with cold medicine–associated Stevens-Johnson syndrome/toxic epidermal necrolysis with severe ocular complications among Europeans (odds ratio [OR], 21.2; 95% CI, 0.97-465.0; P = .04), which is questionable. The scientists consider an association to be statistically significant when its P value is less than .05 or its confidence interval (CI) has not crossed 1.² In the aforementioned association reported by Wakamatsu et al,¹ the CI had crossed 1, whereas its corresponding P value was reported to be less than .05, which is controversial and should be corrected.