Department of Dermatology, Seoul St. Mary's Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea.
Department of Dermatology, Yonsei University Wonju College of Medicine, Wonju, Korea.
J Invest Dermatol. 2023 Sep;143(9):1717-1723. doi: 10.1016/j.jid.2023.02.031. Epub 2023 Mar 21.
Although the study design for identifying specific disease associations using a health insurance database has been well-established, few studies explore unknown comorbidities. We conducted a series of automated case-control studies for all International Classification of Disease, Tenth Revision, Clinical Modification diagnostic codes (A01-Z99) using the Korean National Health Insurance database from 2007 to 2017 to reveal undiscovered disease associations of vitiligo. A total of 90,297 patients with vitiligo and 90,297 age- and sex-matched controls without vitiligo were included, and disease associations for 1,265 relevant diagnostic codes were screened. A meta-analysis of the individual ORs for each International Classification of Disease, Tenth Revision code was performed to identify the possibility of selection bias. Finally, the association with vitiligo was significantly increased in 45 diseases and decreased in 6 diseases. We not only reaffirmed the positive correlation between vitiligo and other autoimmune diseases but also observed associations with obsessive-compulsive disorder and melanoma. In contrast, femur fracture showed a negative correlation. In this study, we attempted an automated mass screening and suggested a possible selection bias. In the era of large-scale databases, a systematic and comprehensive approach might be needed.
虽然使用健康保险数据库来确定特定疾病关联的研究设计已经成熟,但很少有研究探索未知的合并症。我们使用韩国国家健康保险数据库,对 2007 年至 2017 年的所有国际疾病分类第 10 次修订临床修正版诊断代码(A01-Z99)进行了一系列自动化病例对照研究,以揭示白癜风的未发现疾病关联。共纳入了 90297 例白癜风患者和 90297 名年龄和性别匹配的无白癜风对照组,并筛选了 1265 个相关诊断代码的疾病关联。对每个国际疾病分类第 10 次修订版代码的个体优势比进行了荟萃分析,以确定选择偏倚的可能性。最后,45 种疾病与白癜风的关联显著增加,6 种疾病的关联显著降低。我们不仅再次证实了白癜风与其他自身免疫性疾病之间的正相关关系,还观察到了与强迫症和黑色素瘤的关联。相比之下,股骨骨折表现出负相关。在这项研究中,我们尝试了自动大规模筛查,并提出了可能存在选择偏差。在大数据时代,可能需要一种系统和全面的方法。