Global Burden of Thyroid Cancer From 1990 to 2017

Key Points Question What were the epidemiologic patterns and variation in the trends of thyroid cancer worldwide from 1990 to 2017? Findings In this cross-sectional study covering data on incidence, deaths, and disability-adjusted life-years and their temporal trends from 195 countries and 21 regions, increasing trends of thyroid cancer burden were observed, with significant differences by sex, region, country, age, and sociodemographic index. Almost half of the thyroid cancer burden was noted in Southern and Eastern Asia, and a third of patients with thyroid cancer resided in countries with a high sociodemographic index. Meaning This study suggests an increasing global burden of thyroid cancer; the geographic disparities may provide support for cancer health care planning and resource allocation.


Introduction
Thyroid cancer is the most pervasive endocrine cancer worldwide. 1 During the past decades, published studies reported that the incidence of thyroid cancer continues to increase 2 in countries and regions such as Canada, 3 the US, 4 Australia, 5 Asia, [6][7][8] South America, 9 and Europe. [10][11][12][13] Although some regional studies have provided data on the incidence and mortality associated with thyroid cancer, 14,15 studies on thyroid cancer examining the association between the disease and country, sex, age, sociodemographic index (SDI), and other factors are lacking. Comprehensive, in-depth analysis of thyroid cancer in all regions of the world based on a variety of factors may be beneficial for health care planning and resource allocation.
The Global Health Data Exchange is a public website available for querying the burden of 354 human diseases and injuries in 195 countries and territories worldwide, providing an opportunity to investigate the distribution and changes in the patterns of thyroid cancer. 16 Analyses based on age-standardized rates may help policy makers assess the burden of thyroid cancer, measure the progress of specific treatments, allocate resources, and formulate relevant policies. This study aimed to explore the current pattern and alteration of thyroid cancer incidence, deaths, and disabilityadjusted life-years (DALYs).

Study Population and Data Collection
The data were obtained using the Global Health Data Exchange, covering annual incidence, deaths, DALYs, and age-standardized rate of thyroid cancer in 21 regions and 195 countries, from January 1,1990, to December 31, 2017. 17 Data on both sexes of 4 age groups (5-14, 15-49, 50-69, and Ն70 years) were collected. Detailed descriptions of the methods are presented in the eAppendix in the 3.5.2 (R Project for Statistical Computing). A 2-tailed P value <.05 was considered statistically significant.

Thyroid Cancer Incidence
At a global level, from 1990 to 2017, thyroid cancer ASIR in most countries presented an upward trend ( Figure 1A). Incident cases were greater among females than among males (female to male ratio, 1.92 in 1990 and 2.36 in 2017), as was the ASIR of thyroid cancer (female to male ratio, 2.65 in 1990 and 2.23 in 2017); however, the EAPC was larger in males (2.18      Age-standardized incidence rate (ASIR) (A), age-standardized death rate (ASDR) (B), and age-standardized disability-adjusted life-year rate (AS DALY) (C).       addition, the EAPC was correlated with ASIR (ρ = −0.35, P < .01) ( Figure 3A) but was not correlated with SDI (ρ = −0.08, P = .29) (eFigure 20 in the Supplement).

Discussion
In this study, from 1990 through 2017, the incidence, deaths, and DALYs of thyroid cancer and ASIR increased by 60% to 200%, whereas the ASDR and age-standardized DALY rates decreased. The increasing incidence of thyroid cancer in all SDI quintiles raises notable points. The ASIR in areas within a high SDI quintile continued to increase until 2010 and then began decreasing. Compared with the decreased age-standardized rates among females in all SDI quintiles, those rates among males continued to increase until 2010. In 2009, the American Thyroid Association's guidelines on thyroid cancer diagnosis and treatment were revised substantially and other countries developed guidelines, which might be associated with the change noted from 2010. 20 A previous study reported that males in communities with low socioeconomic status had poorer thyroid cancerspecific survival, but these findings did not appear to apply to women. 21

Limitations
This study has limitations. As with other estimates of disease burden, the most important limitation of GBD is the lack of data at many sites. The key principle of GBD is to make full use of the data sources of all relevant resources. Data are available from a wide range of sources (>90 000 data sources). Although the diagnosis of thyroid cancer with inadequate, insufficiently specific, or unreliable registration has been corrected by a redistribution algorithm, the accuracy of diagnosis still has some unreliability. In addition, information bias regarding the epidemiologic evaluation of thyroid cancer was inevitable, as data are scarce in a few parts of the world. Given the restrictions of data type, to our knowledge, further investigation on thyroid cancer stratified by histologic characteristics, grade, and risk factors has not been conducted.