In the USA, the incidence of thyroid cancer increased rapidly for several decades, although some studies have suggested that it has now plateaued or even begun to decrease. We aimed to establish whether incidence in the USA has truly decreased or merely plateaued, and to understand some of the underlying factors driving these changes. We conducted a retrospective, population-based study using the National Cancer Institute (NCI)'s Surveillance, Epidemiology, and End Results database and the National Center for Health Statistics database. We used incidence data from these registries obtained between 1975 and 2019, including patients with a diagnosis of thyroid cancer according to the third edition ICD for Oncology (site code C73.9) and malignant histology. We used the NCI's Joinpoint Regression Program to estimate trends in the incidence over time and age-period-cohort modelling to identify the factors influencing these trends. Our sample included 91 968 patients with thyroid cancer, of whom 23 467 (25·5%) were men and 68 501 (74·5%) were women. The annual incidence of thyroid cancer increased from 5·0 cases per 100 000 people in 1975 to 14·6 cases per 100 000 people in 2009, before plateauing until 2019. The age-period-cohort analysis suggests that the changes in incidence were primarily due to time period effects. Furthermore, the increase in incidence was most prominent among women aged 40-69 years and men aged 50-79 years. Throughout all time periods, incidence increased with each successive birth cohort among both women and men. The rise and subsequent plateau in the incidence of thyroid cancer in the USA have been primarily driven by time period effects, likely due to changing patterns in diagnostic pressure. Variations in the incidence of thyroid cancer by age, which increased during the time frame of this study, seem to be driven predominantly by overdiagnosis. Although the incidence of thyroid cancer has plateaued, it remains at peak levels, suggesting that overdiagnosis remains a crucial unresolved public health issue. Further work is needed to help limit the current drivers of overdiagnosis and to implement novel solutions aimed at both physicians, patients, and policy makers. None.
Read full abstract