Screening Leads to Overestimated Associations of Thyroid Dysfunction and Thyroiditis with Thyroid Cancer Risk

SIMPLE SUMMARY: The association of thyroid dysfunction and thyroiditis with the risk of thyroid cancer is an important issue in clinical practice, and is controversial due to various confounders, including screening effects. In this large-sample, population-based, nationwide study, we demonstrated that the screening effect likely significantly contributed to the positive associations of thyroid dysfunction and thyroiditis with thyroid cancer. After adjustment for this confounder, thyroid cancer risk was positively associated with hypothyroidism and thyroiditis but negatively associated with hyperthyroidism and Graves’ disease. Multiple subgroup analyses showed consistent results. Given these overestimated associations, we suggest that unnecessary and excessive concerns or screening tests for thyroid cancer could be reduced in patients with thyroid dysfunction or thyroiditis. ABSTRACT: We aimed to assess the relationships of functional thyroid disease and thyroiditis with subsequent thyroid cancer, which is controversial due to various confounders, and the effect of thyroid disease workup on this association. We used the cohort data from 2002 to 2015 (Study I, n = 28,330) and the entire data from 2002 to 2019 (Study II, n = 883,074) of the Korean National Health Insurance Service database, and performed logistic regression and subgroup analyses with various covariates. In Study I, hypothyroidism, thyroiditis, autoimmune thyroiditis, hyperthyroidism, and Graves’ disease showed positive associations with thyroid cancer. In Study II, after adjustment for covariates including the number of thyroid function tests, the ORs for thyroid cancer were significantly reduced in all thyroid diseases. Hypothyroidism, thyroiditis, and autoimmune thyroiditis were positively associated (adjusted odds ratio, OR (95% confidence interval, CI) 1.28 (1.25–1.32), 1.36 (1.31–1.42), and 1.17 (1.11–1.24), respectively), whereas hyperthyroidism and Graves’ disease were negatively associated with thyroid cancer (adjusted OR (95% CI) 0.80 (0.77–0.83) and 0.69 (0.65–0.74), respectively). Multiple subgroup analyses in both studies showed consistent results. In this large population-based, nationwide study, we confirmed that thyroid disease workup leads to overestimation of associations of thyroid dysfunction and thyroiditis with thyroid cancer risk.