Racial and Ethnic Disparities in the Diagnosis and Treatment of Thyroid Disease

Abstract

Context

There are differences in diagnosis, treatment, and outcomes for thyroid between racial and ethnic groups that contribute to disparities. Identifying these differences and their causes are the key to understanding and reducing disparities in presentation and outcomes in endocrine disorders.

Evidence Acquisition

The present study reviews original studies identifying and exploring differences between benign and malignant thyroid diseases. A PubMed, Web of Science, and Scopus search was conducted for English-language studies using the terms “thyroid,” “thyroid disease,” “thyroid cancer,” “race,” “ethnicity,” and “disparities” from inception to December 31, 2022.

Evidence Synthesis

Many racial and ethnic disparities in the diagnosis, presentation, treatment, and outcomes of thyroid disease were found. Non-White patients are more likely to have a later time to referral, to present with more advanced disease, to have more aggressive forms of thyroid cancer, and are less likely to receive the appropriate treatment than White patients. Overall and disease-specific survival rates are lower in Black and Hispanic populations when compared to White patients.

Conclusions

Extensive disparities exist in thyroid disease diagnosis, treatment, and outcomes that may have been overlooked. Further work is needed to identify the causes of these disparities to begin to work toward equity in the care of thyroid disease.

Recent events over the past decade including the COVID-19 pandemic and civil rights unrest have drawn attention to the pervasive causes and effects of health care disparities in surgical and endocrine diseases (1-4). These disparities in health care are likely multifactorial, relating to limited access to care, environmental exposures, and resource-poor living conditions (5-8). In addition to disparities in overall health and health care, racial and ethnic minorities experience differences in outcomes in the treatment of specific disease processes, which include differences in the rates of detection and diagnosis of disease, postoperative complications, and cancer recurrence and survival (3, 9, 10).

Thyroid disease affects approximately 20 million Americans, with about 12% likely to develop thyroid disease at some time in their life (11). This includes benign and malignant disease, and hyperfunctioning, hypofunctioning, and normal function (euthyroid) thyroid disease. The majority of thyroid dysfunction results from hypothyroidism and predominately affects women. Hyperthyroidism affects approximately 1.3% of Americans and is generally treated with medication, but often requires more definitive treatment with thyroid surgery (thyroidectomy) or radioactive iodine (RAI) ablation (12). Lastly, euthyroid disease, which includes thyroid nodules and thyroid cancers, is relatively common, as thyroid nodules can be found in population autopsy studies in 30% to 60% of individuals (13). Depending on patient-specific factors, the risk of thyroid malignancy is 7% to 15% for any given thyroid nodule (13). Thyroid cancer is steadily increasing in incidence and is currently the sixth most common cancer in women (14). New thyroid cancer cases are estimated to be 15.7 per 100 000 people per year with a mortality rate of 0.5 per 100 000 individuals per year (14). This increase in incidence has been attributed to improvement in diagnosis, increase in incidental findings, as well as true increase in thyroid cancer rates (15, 16).

There is an exploding field of research on endocrine disease health care disparities both in benign and malignant thyroid diseases. There exists a spectrum of clinical equipoise when discussing and selecting treatment options with patients in regard to thyroid disease. Whenever subjectivity is prevalent in disease management, the potential for bias and discrimination arises (17, 18). Treatment selection among minorities can vary by access to subspecialist, speed of definitive treatment, and appropriateness or “guideline-concordant” care. Examined outcomes may include both patient-reported (satisfaction with care) and so-called “textbook” outcomes such as disease recurrence, surgical morbidity and mortality, as well as adherence to follow-up (19).

Much work has been done over the past few decades to explore these issues through the lens of health care disparities, and there exists a need to systematically catalog the overall findings in this area. Toward this end, a review was conducted to map the research being actively conducted in the area of health care disparities among patients with thyroid disease. The overall goal was to summarize differences between racial and ethnic groups that contribute to disparities in the (1) diagnosis, (2) treatment/access to care, and (3) outcomes of thyroid disease.

Materials and Methods

A PubMed, Web of Science, and Scopus search was conducted for English-language studies using the terms “thyroid,” “thyroid disease,” “thyroid cancer,” “race,” “ethnicity,” and “disparities” from inception to December 31, 2022. The search was conducted using Boolean logic, excluding studies that address only disparities in socioeconomic status. To be included in this review, studies must have measured or focused on the specific presentation or outcomes of North American, adult, racial and ethnic minorities, and thyroid diseases. Studies on children and in the state of pregnancy were excluded. Only peer-reviewed original data, excluding reviews, editorials, and case reports, were included for review. Additional references were identified through review of the citations of the retrieved articles. All studies were reviewed by 2 independent authors (S.D. and A.G.), who came to a consensus for inclusion.

Results

Benign Thyroid Disease

Diagnosis

The initial evaluation and management of thyroid disease begins with assessment of thyroid function via clinical exam and biochemical testing. Additionally, a search for the presence of thyroid nodules that may be concerning for thyroid cancer is performed with imaging. The detection of disparities in the assessment of thyroid function is somewhat limited because it is nearly impossible to calculate the true number of patients who should be evaluated for thyroid disease in the general population. Due to the vagueness of symptoms of thyroid dysfunction, it also is difficult to know which patients should be screened and which are missed. In a survey of 16 081 patients, a Canadian study found that fewer minorities and immigrants were likely to report being on thyroid medication or that their last set of bloodwork performed by their physician was to rule out a new thyroid disorder (Table 1) (20). While this study was unable to evaluate true screening rates, it did suggest that immigrants and minorities, especially those with limited access to a physician, were potentially being screened and treated for thyroid disease a lower rate than White Canadians (20).

Table 1.

Selected references related to benign thyroid disease

Reference	Demographics compared			Findings compared to White controls
Benign disease
Stoll (20)	n = 16 081	Minorities/Canadian immigrants	White	Less likely to undergo ultrasound or laboratory tests for thyroid disease
Kuo et al (21)	n = 1199	Black	White	Hypothyroidism: older age
				Higher referral for goiter, compressive symptoms, dysphagia Time to referral: > 1 y vs <1 y Gland mass: larger in Black patients
Iwata et al (22)	n = 1369	Non-White	White	White: 1.6× incidental nodule discovery Non-White malignancy 4.9% vs 12.9%

Treatment/Outcomes

In a retrospective review of an institutional database of patients with benign thyroid disease, Kuo et al (21) evaluated patient characteristics, disease course, thyroid pathology, and postoperative outcomes evaluating 1189 patients (1023 White and 166 Black). Black patients with hypothyroidism were referred to consultation at an older age than White patients. Black patients were more commonly referred for goiter, while White patients were more commonly referred for nodular disease. Additionally, Black patients were more likely to be referred with more severe symptoms, with 45% of Black patients having compressive symptoms and 19% having dysphagia, compared to 21.2% and 10.2% of White patients, respectively (P < .01) (21). Black patients also had a higher median time to referral (1 year vs <1 year) and larger median gland mass (71 g vs 24.3 g) when compared to White patients (21).

Patients with concerns for thyroid nodules or thyroid cancer should receive a thyroid ultrasound during their work-up. Any concerning findings should prompt fine-needle aspiration biopsy (FNAB) for further diagnosis and to guide management, but these diagnostic tests are not equally applied in all races. Iwata et al (22) performed a retrospective review of 1369 patients undergoing FNAB for thyroid nodules. The authors compared the method by which thyroid nodules were discovered prior to biopsy and assessed the effect a patient's race may have had on their workup. White patients were 1.6 times more likely to have an FNAB for an incidentally discovered thyroid nodule compared to non-White individuals (P < .0001). Additionally, incidentally biopsied thyroid nodules in non-White patients were found to have malignancy in 4.9% of cases, compared to 12.9% of White patients (22). These findings suggest that White patients are more likely to receive FNAB for thyroid nodules. It additionally raises concerns that non-White patients may be underdiagnosed as many thyroid nodules are not receiving an indicated FNAB.

Thyroid Cancer

Once a thyroid nodule has been biopsied and the diagnosis of thyroid cancer established, patients are generally treated with thyroidectomy (total thyroidectomy or thyroid lobectomy). In the case of differentiated thyroid cancers (DTC) (papillary, follicular, and oncocytic thyroid cancer), the addition of suppressive-dose thyroid hormone replacement and/or RAI ablation is made depending on the pathologic characteristics of the cancer (23). As noted earlier by differences in time to referral and FNAB rates, racial and ethnic disparities exist before patients are diagnosed with malignancy, which may continue to manifest in cancer presentation and survival.

Differentiated thyroid cancer

Diagnosis

Several studies have noted differences in DTC size at presentation, rates of metastasis, and differences in high-risk features at presentation when examining patient race and ethnicity. In a retrospective review of 53 990 patients with thyroid cancer from the Surveillance, Epidemiology, and End Results (SEER) database from 1973 to 2009, Morris et al (24) found Black patients more likely to present with larger tumors (relative risk [RR] tumor >1 cm: 1.12; 95% CI, 1.05-1.20; P < .0005; RR of tumor >4 cm: 1.13; 95% CI, 1.11-1.15; P < .001) and more likely to have follicular carcinomas than White patients (22.3% vs 13.3%; P < .001) (Table 2). This ethnic size disparity has been reinforced by several other studies (25-27, 30). Despite this size difference, Black patients have been found less likely to have nodal metastases (RR 0.89; 95% CI, 0.87-0.91; P < .0001) (24, 28, 33), extracapsular invasion (RR 0.96; 95% CI, 0.94-0.98; P = .0008) (24, 33), or regional metastases (RR 0.96; 95% CI, 0.93-0.99; P < .0001) (24, 28).

Table 2.

Selected references related to differentiated thyroid cancer

Reference	Demographics compared			Findings compared to White controls
Malignant disease
DTC diagnosis
Morris et al; Harari et al; Tang et al;  Hollenbeak et al (24-27)	n = 18 047-53 900	Black	White	Presentation: larger tumors >1 cm and >4 cm
Hollenbeak et al (27)				Follicular tumors: more likely
Morris et al (24)				Nodal mets: less likely ETE: less likely Regional mets: less likely
Moo-Young et al (28)	n = 18 047	Black, Hispanic	White	Hispanic: younger age, more lymph node mets Black: least likely for mets
Harari et al (25)	n = 25 945	Minorities	White	Metastasis: higher ratesa
Nash et al (29)	n = 210	Alaskan Native	White	Age: younger Tumor size: larger Advanced stage: more likely
Shah et al (30)	n = 282 043	Black, Hispanic, Asian	White	Black patients: larger tumors
Moten et al;  Bonner et al (31, 32)	n = 7894-322 923	Black, Hispanic, Asian	White	Hispanic/Asian: greater odds high-risk disease Black: lower odds of high risk disease
Tang et al;  Mehta et al (26, 33)	n = 70,346-258 973	Black	White	Classic PTC: less likely, less distant mets Follicular variant PTC: more likely, more distant mets Follicular thyroid cancer: more likely, less distant mets Distant mets: no difference
Mehta et al (33)	n = 258 973	Black, Hispanic, Asian	White	Black: lower odds ETE, lymph node mets Asian, Hispanic: higher odds ETE, lymph nodes mets
DTC treatment
Shah et al (30)	n = 282 043	Black	White	Appropriateness of operation: lower odds Under treatment with RAI: higher odds
DTC outcomes
Morris et al;  Tang et al (24, 26)	n = 25 945-70 346	Black	White	Worse overall survival No difference in disease-specific survival rates
Harari et al;  Tang et al;  Moten et al;  Bonner et al (25, 26, 31, 32)	n = 25 945-322 923	Non-Hispanic Black, Hispanic, API	Non-Hispanic White	API/Hispanic: higher overall survival Black: lower overall survivala
Shah et al (30)	n = 282 043	Black	White	Higher postoperative mortality 30, 90 d
Nguyen et al (34)	n = 19 940 952	Filipino, Non-Filipino Asian	White	Filipino: more likely to die of thyroid cancer Foreign-born Filipino: higher mortality rate than American born
Asban et al (35)	n = 1584	Black	White	Black males: higher mortality than Black females, White males, or White femalesa
Keegen et al (36)	n = 16 827	African American/Hispanic	White	Worse disease-specific survival
Megwalu and Saini (37)	n = 17 688	Black, other race	White	Overall survival Black < White < othera

Abbreviations: API, Asian/Pacific Islander; DTC, differentiated thyroid cancer; ETE, extrathyroidal extension; mets, metastases; PTC, papillary thyroid cancer; RAI, radioactive iodine.

^a Relationship consistent on multivariable analysis.

Moo-Young (28) et al similarly reviewed the SEER database (1996–2006, n = 18 047) but combined results with a local minority-majority hospital database (Cook County Hospital 1978-2006; n = 291). In this study, the clinical, pathologic, and demographic differences in patients with papillary thyroid cancer (PTC), follicular thyroid cancer (FTC), and medullary thyroid cancer (MTC) were compared by race/ethnicity. Analysis from both databases demonstrated that Hispanic patients were more likely to be younger than other race groups (P < .001) and more likely to present with lymph node metastasis than other races (P < .001) (28). Uniquely, Harari et al (25) in a review of the California Cancer registry found minorities (Black, Hispanic, and Asian/Pacific Islanders [API]) to present with higher rates of metastatic disease when compared to White patients (P < .001), which persisted after adjusting for age, sex, socioeconomic status, and insurance type (Black, odds ratio [OR] = 1.36 [CI, 1.01-1.84]; Hispanic, OR = 1.89 [CI, 1.62-2.21]; API, OR = 1.82 [CI, 1.54-2.15]).

In a 45-year review of thyroid cancers diagnosed in Alaskan Native individuals, Nash et al (29) identified 210 patients diagnosed with thyroid cancer. When comparing the characteristics of these patients to White patients, they found that Alaskan Native patients were more likely to present at a younger age (44.2 years vs 48.4 years; P = .0002) and were more likely to have larger tumors (P = .030). The authors also noted that Alaskan Native patients were more likely to present with more advanced stage disease than White patients, with lower likelihood of presenting with localized disease (58.5% vs 68%; P = .027) (29).

Only examining patients with high risk features of thyroid cancer using the SEER database, Moten et al (31) found that Hispanic patients (OR, 1.67; 95% CI, 1.56-1.79) and Asian patients (OR, 1.49; 95% CI, 1.37-1.62) had greater odds of presenting with high-risk disease when compared to White patients, while Black patients were at lower odds of presenting with high-risk disease (OR, 0.46; 95% CI, 0.40-0.53). Mehta et al (33), using the National Cancer Database (NCDB), similarly found that Asian and Hispanic patients were more likely to have extrathyroidal extension and lymph node metastasis than White patients.

When considering subtypes of DTC, Tang et al examined 70 346 patients using the SEER database, comparing White and Black patients. They, with others, noted White patients were more likely to have classic PTC, (62%), while Black patients were more likely to have follicular variant PTC (FV-PTC) (41.1%) or FTC (10.9%) (26, 27, 33). Rates of distant metastases were not different between groups, except in cases of Black patients with FV-PTC (OR = 1.715; P = .026) (26). They notably found that lymph node metastasis were more common in White than in Black patients except in FTC (OR = 1.715; P = .026) (26).

These studies found that the presentation and severity of thyroid cancer vary among different minority populations. This is possibly related to a difference in the type of DTC (Black patients are more likely to be diagnosed with FTC and FV-PTC), a delay in diagnosis, or limited access to care. Lower rates of lymph node metastasis may be explained by either less aggressive assessment of lymph node status in Black patients, or to higher rates of FTC/FV-PTC, which tend to metastasize through hematogenous spread not lymphatic channels (26, 30, 33).

Treatment

When considering treatment choices between racial/ethnic groups in those with DTC, there exist differences in appropriateness of care. Using the NCBD, Shah et al (30) recorded the date of surgery for each patient and the most current American Thyroid Association and Nation Comprehensive Cancer Network guidelines for that date. With this information, the authors determined if the extent of surgery and postoperative treatments were appropriate for each patient’s risk category (30). They found that Black patients were at lower odds than White patients of undergoing the appropriate operation for their thyroid cancer (OR 0.78; 95% CI, 0.71-0.87; P < .001) (30). The authors also determined that minority groups were more likely to be undertreated with RAI when compared to White patients (Hispanic OR = 1.27; 95% CI, 1.18-1.36; Black OR = 1.26; 95% CI, 1.17-1.37; Asian OR = 1.25; 95% CI, 1.14-1.37; P < .001); while White patients were more likely to be overtreated with RAI than other groups (30).

Outcomes

Prompt diagnosis and treatment are key factors in optimizing care for thyroid cancer. Additionally, treatment by a high-volume surgeon, limited in those with poor health access, has the potential to drive outcomes from thyroid malignancy. Morris et al (24), in 2008, published on disparities in survival after treatment and did not note a difference in mortality rates between Black and White patients. Subsequent examinations since that time, however, have found otherwise. Using the California Cancer registry, Harari et al (25) examined 25 945 non-Hispanic White, non- Hispanic Black, Hispanic, and API patients with DTC. The authors found the unadjusted overall survival rates were higher among API and Hispanic patients, but lower among Black patients when compared with White patients; P < 0.001). Adjusting for stage, age, sex, comorbidities, and treatment, analysis demonstrated overall survival advantage for Asian patients (hazard ratio [HR] 0.83; CI, 0.71-0.97) and survival disadvantage for Black patients (HR 1.4; CI, 1.10-1.73). Shah et al (30) similarly found Black patients not only had lower overall survival than White patients at 5 years (94% vs 93%; P < .001) and 10 years (87%-84%; P < .001), but also had the highest rate of postoperative mortality at 30 (0.2%) and 90 (0.4%) days (P < .001) when compared to other races. Others have similarly found Black patients not only had lower overall survival than White patients at 5 years (94% vs 93%; P < .01) and 10 years (87% vs 84%; P < .001), but also had the highest rate of postoperative mortality at 30 (0.2%) and 90 (0.4%) days (P < .001) when compared to other races (26, 30).

Bonner et al (32) found that among an NCDB cohort (2004-2018), Hispanic (OR = 1.49; P < .001) and Asian (OR = 1.49; P < .001) patients were more likely to develop advanced-stage well-differentiated thyroid cancer when compared to the White population, but did not exhibit worse survival. Conversely, Black patients were less likely to present with advanced disease, but had worse overall survival (HR = 1.24; P < .001) than the White population.

In regard to disparities within the Asian population, Nguyen et al (34) sought to determine potential disparities in people of Filipino heritage when compared to non-Filipino Asian and non-Hispanic White patients. Using US death records from 2003 to 2012, they found that Filipino patients were 1.8 and 3.9 times more likely to die of thyroid cancer than non-Filipino Asian and non-Hispanic White patients (34). They also noted that mortality rates for thyroid cancer were higher in foreign-born Filipino people than American-born Filipino people (34). The authors propose potential environmental exposure (carcinogenic volcanic lava in the Philippines), difference in approach to accessing health care, and high-iodine diet as potential causes for these differences in mortality (34).

Asban et al (35) explored the effect of intersectionality, that is, being both a racial and sex minority (Black male) among thyroid cancer populations. Among 1584 local patients with thyroid cancer, the authors found that Black males had the highest mortality risk compared to non-White females (HR = 1.58; 95% CI, 1.03-2.43; P = .0376), White males (HR = 3.5; 1.88-6.54; P < .0001), and White females (HR = 4.78; 3.31-9.90; P < .0001), after controlling for age, cancer type, stage, surgical procedure, and RAI.

When evaluating if these health outcome disparities in survival were present in adolescents and young adults between the ages of 15 and 39 years with DTC, Keegan et al (36) found no difference in overall survival across race/ethnicity; however, African American and Hispanic individuals had worse disease-specific survival, with thyroid cancer mortality being more than 6 times (HR = 6.7; CI, 2.46-18.23) and 3 times (HR = 3.6; CI, 1.87-6.92) more likely than White patients, respectively.

Papillary thyroid microcarcinoma, or PTCs smaller than 1 cm, may be managed differently from larger DTC including with close surveillance, forgoing surgery unless concerning changes are seen in close ultrasound follow-up. Megwalu and Saini (37) sought to evaluate the effect of race on papillary thyroid microcarcinoma. They included 17 688 patients from the SEER program database and found that Black patients (5-year survival 94.6%, 10-year survival 89.4%) patients had lower overall survival when compared to White patients: White (5-year survival 98%, 10-year survival 95.4%) and “other race” patients (5-year survival 98.8%, 10-year survival 95.7%) (P < .001). When adjusting for sex, marital status, age, year of diagnosis, multifocal disease, and type of surgery, Black patients persistently had worse overall survival (HR = 2.59; CI, 1.92-3.51; P < .001).

Medullary thyroid cancer

Diagnosis, treatment, and outcomes

The diagnosis and treatment of MTC is inherently different from DTC as it derives from thyroid C-cells and not follicular cells and may be linked to genetic syndromes (Table 3) (38). The tumors may be aggressive with poor responsiveness to RAI and with lower survival rates than DTCs (42). In a retrospective review of 1647 patients with MTC, Roche et al (39) examined presentation and rates of types of treatment for non-Hispanic White, Hispanic, Black, and other races with MTC. Hispanic (30.2%) and Black (24.8%) patients were more likely to present with tumors larger than 4 cm than non-Hispanic White patients (20.7%). The authors also found that Black patients were less likely to undergo lymph node examination than non-Hispanic White patients (adjusted OR, 0.61; 95% CI, 0.39-0.93) (39). Additionally, Black (74.8%) and Hispanic (79.1%) patients had lower overall survival that non-Hispanic White (83.4%) patients (P = .06) (39). Black (79.6%) patients also had lower disease-specific survival than White (88%) and Hispanic (86.1%) patients (P = .046) (39). Guideline-concordant care has also been examined for MTC by Chang et al (40). They identified factors such as African American race, localized disease at diagnosis, and median household income as predictors of receiving guideline-discordant care, while guideline-concordant care was predictive of overall survival.

Table 3.

Selected references related to medullary and anaplastic thyroid cancer

Reference	Demographics compared			Findings compared to White controls
Malignant disease
MTC
Roche et al (39)	n = 1647	Non-Hispanic Black, Hispanic, other	Non-Hispanic White	Non-White: larger tumors Black: less likely lymph node examination Black/Hispanic: lower overall survival
Chang et al (40)	n = 3693	African American, Hispanic, other	White	Hispanic: more likely regional disease African American: less likely to receive guideline concordant care
ATC
Hollenbeak et al (27)	n = 26 902	Black	White	Diagnosis: 2.3× more likely
Roche et al (41)	n = 719	Non-White	Non-Hispanic White	Tumor size, lymph node involvement, and distant mets no difference Treatment: more likely to receive surgery and radiation Overall survival: lower at 20 mo converged at 60 mo

Abbreviations: API, Asian/Pacific Islander; ATC, anaplastic thyroid cancer; mets, metastases; MTC, medullary thyroid cancer.

Anaplastic thyroid cancer

Diagnosis, treatment, and outcomes

Anaplastic thyroid cancer (ATC) is a rapidly progressing thyroid cancer that often arises from untreated advanced DTC (43). Patients with ATC often succumb to their disease within months of diagnosis. Advances in care are promising, but extremely limited to patients that are able to access academic centers that can care for this rare disease (44). Black patients are known be 2.3 times more likely to be diagnosed with ATC than White patients (27). Roche et al (41) examined 719 patients diagnosed with ATC using the SEER database. Tumor size, lymph node involvement, and distant metastasis did not vary between non-Hispanic patients, White patients, and non-White patients. The authors did find that non-Hispanic White (82.6%) patients were more likely to receive any treatment than non-White patients (74.4%) (P = .012) (41). Adjusted logistic regression analysis demonstrated Non-White race/ethnicity to be predictive of receiving no treatment (OR = 0.29; 95% CI, 0.16-0.54). When comparing overall survival, non-White patients (15.5%) had lower survival rates at 20 months when compared to non-Hispanic White patients (20.8%) but converged by 60 months. They also noted that 5-year survival was significantly improved by receipt of any treatment or radiation treatment (41).

Overall Surgical Outcomes

Surgical outcomes after thyroidectomy have been known to be associated with the volume of thyroid operations that a surgeon performs. This has been demonstrated time and time again to the point that the American Thyroid Association recommends thyroidectomy ideally be performed by a high-volume thyroid surgeon (23). Several studies have identified access to high-volume surgeons and socioeconomic status as associated with postoperative outcomes but few have focused on race/ethnicity (45, 46).

In 2006, Sosa et al (47) used the Health Care Utilization Project National Inpatient Sample (HCUP-NIS) national database (1999-2004) to retrospectively review 16 878 patients who underwent thyroid procedures. They found that White patients (43%) were more likely to have a total thyroidectomy, while Hispanic and Black patients were more likely (57% for both) to have subtotal thyroidectomy, a now well-known inferior procedure (Table 4). In this cohort, Black patients carried the highest rate of undergoing thyroid surgery for benign disease (Black 80%, White 63%, Hispanic 55%, “other race” 59%), while Hispanic patients were the most likely to receive thyroid surgery for malignancy (Black 20%, White 37%, Hispanic 45%, “other” 42%) (47). Black patients had the lowest rate of elective admission for thyroidectomy, were more likely to have surgery by low-volume surgeons (1-9 cases/year), and had the longest average lengths of stay (2.5 days) compared with White patients (1.8 days; P < .001) (47). Additionally, White patients (7%) were more likely to have surgery performed by highest-volume surgeons (>100 cases/year) when compared to Black (2%) and Hispanic (1%) patients (47). Black patients (4.9%) had higher complication rates than Hispanic (3.6%) and White (3.8%) patients (P = .056); they also had the highest rate of in-hospital mortality (0.4%) (47). Cost of care was lower for White patients ($5447/patient) compared to Black patients ($6587), Hispanic patients ($6294), and patients in the other race group ($5479) (47). Differences in length of stay and hospital cost remained after adjusting for age, sex, hospital region, procedure, diagnosis, comorbidity, surgeon volume, household income, primary payer, and admission type (47). These findings have been supported in other more recent studies in the Healthcare Cost and Utilization Project Databases (HCUP) and NSQIP (30, 48-50). Similar findings by Kuo et al (21) found Black patients were also noted to have higher rates of reintubation postoperatively (2.2% vs 0.2%), but other postoperative complication rates were not different from White patients. The authors also found that Asian patients were more likely to suffer from recurrent laryngeal nerve injury than White patients (OR = 1.83; 95% CI, 1.12-2.97).

Table 4.

Selected references related to postsurgical outcomes

Reference	Demographics compared			Findings compared to White controls
Surgical outcomes
Sosa et al (47)	n = 16 878	Black, Hispanic, other race	White	Black: more subtotal thyroidectomies (inferior) Black: more thyroid surgery for benign disease Black: longest average length of staya Black: highest in hospital mortality Hispanic/other: more thyroid surgery for malignant disease
Noureldine et al (48)	n = 106 314	African American, Hispanic, Asian	White	Black: lowest rate of elective admission Black: longest average length of stay
Noureldine et al;  Sosa et al (47, 48)	n = 16 878-106 314	African American, Hispanic, Asian	White	Black: higher risk of low-volume surgeon Black/Hispanic: lowest risk of high-volume surgeon Cost of care: Black > Hispanic > White > othera
Sosa et al; Noureldine et al;  Al-Qurayshi et al;  Shah et al;  Kuo et al;  Maduka et al;  Radowsky et al (21, 30, 47-51)	n = 16 878-106 314	Black, Hispanic	White	Highest complication rates: Black > White > Hispanic (complications: neck hematoma, hoarseness, reintubation, surgical)
Al-Qurayshi et al (49)	14 220 inpatient 7215 outpatient	African American	White	More likely operated on by low-volume surgeons
Maduka et al (50)	n = 6187	Asian	White	More recurrent laryngeal nerve injury
Radowsky et al (51)	n = 87	Black	White	Higher odds of negative voice outcomes

^a Relationship consistent on multivariable analysis.

When examining patient-reported outcomes, Radowsky et al (51) performed a prospective observational trial among patients undergoing thyroidectomy and parathyroidectomy and found that Black patients were more likely to report subjective negative voice outcomes (OR = 2.6; P = .034) when compared to White patients.

Discussion

Data on racial and ethnic disparities in the diagnosis, treatment, and outcomes of thyroid disease, including both benign and malignant disease processes, are being generated at a rapid rate. The data that are published tend to be retrospective and database driven. There is much heterogeneity in study design, studied populations, and outcome measurement. However, a few common threads emerge. In the case of benign disease, non-White patients are more likely to have later time to referral and to be referred for treatment when they experience more severe symptoms than White patients. In the case of thyroid cancer, Black patients are more likely to present with larger tumors and more likely to have high-risk features, such as extrathyroidal extension, lymph node metastasis, and distant metastasis than White patients. While Black patients are found to have lower rates of lymph node metastasis, they are more likely to present with more aggressive forms of thyroid cancer (FV-PTC, FTC, and ATC) and less likely to have lymph node evaluation. Additionally, non-White patients are less likely to receive the guideline-concordant operation and adjuvant treatment for thyroid malignancy than White patients.

Overall and thyroid cancer survival rates are lower in Black and Hispanic populations. When examining disparities within other subgroups, such as Filipino or Native Alaskan patients, it becomes apparent that further disparities are revealed that are often overlooked in those categorized as “other” race. In regard to surgical outcomes, non-White patients are less likely to have their operation performed by a high-volume thyroid surgeon, which has been shown to improve outcomes and decrease rates of complications. As with other disparities in health, more aggressive disease patterns in presentation are likely multifactorial, relating to environmental exposure, lack of resources to seek care, delay in diagnosis, access to treatment options, and possible underlying provider bias (2, 4, 9, 52-55).

Moving forward, multidisciplinary teams need to create innovative solutions to better understand and reduce these disparities at multiple stages of the care of patients with thyroid disease. For example, this review could find no biologically driven studies exploring potential sources of racial disparities. Most existent studies that explore the influence of certain molecular drivers of disease (BRAF, TERT) have been completed in mostly homogeneous populations (or ethnic makeup is not mentioned), potentially missing important genes or mutations (56-58). The field of epigenetics as a driver of disparate cancer outcomes is exploding in many other disease states but, to date, no studies have explored this among patients with thyroid cancer (59-61).

Next, at the diagnosis and treatment phase, we feel that increasing access to well-trained, high-volume endocrinologists and surgeons will improve care and outcomes for minorities and the socioeconomically vulnerable. There are endocrine surgery “deserts” in the United States but the best approach to resolve these remains unknown (62). Other work in primary care medicine has shown that providing incentives such as student loan repayment and access to visas helps to increase access to underserved regions (63). The recent COVID-19 pandemic has also shown us how to leverage technology and mobile clinics to improve access (64). Those who care for patients with endocrine diseases should continue to learn from these examples and implement measures to increase access to care, particularly in remote regions.

However, even in regions of the United States with ready access to high-volume centers and countries such as Canada with universal medical care, racial/ethnic disparities are seen in referral patterns and patient outcomes, even when controlling for socioeconomic factors (20, 37, 65). This may be explained by the need for improved referral patterns and provider education as well as attention to the biologic and environmental influence on disease. Additionally, clinically, we believe more regimented, less clinically nuanced decision-making regarding treatment (eg, development of protocoled RAI indications and dosing) may reduce disparities. Leveraging protocolized treatment has been shown to reduce disparities in the field of colorectal surgery with enhanced recovery pathways, for example (17). By taking implicit bias out of the picture by incorporating evidence-based, guideline-driven protocols, we may mitigate disparities in the diagnosis and treatment of thyroid diseases.

Some limitations in the presented studies include the database-driven sources of data. Although national databases present large amounts of data on a wide variety of patients, the missing granularity may cause contradictory results and limits firm conclusions. For example, how well can we compare rates of complications (ie, nerve injury) when different databases define outcomes so differently? Additionally, large databases are unable to comment on individual-level social determinants of health and other risk factors with accuracy. High-quality studies that can comment on the granularity of risk factors such as social determinants of health, adequacy of evaluation, treatment, and follow-up must be conducted on the single- or multi-institutional level to improve our understanding. Finally, studies conducted retrospectively limit comments on causality and the effect of missing variables that could influence bias and outcomes.

In conclusion, more high-quality, original studies on racial/ethnic disparities in thyroid disorders must be conducted. In addition, innovative solutions to identifying and reducing the causes of these disparities must be explored to begin to work toward equity in the care of thyroid disease.

Abbreviations

API

Asian/Pacific Islanders

ATC

anaplastic thyroid cancer

DTC

differentiated thyroid cancer

FNAB

fine-needle aspiration biopsy

FTC

follicular thyroid cancer

FV

follicular variant

MTC

medullary thyroid cancer

NCDB

National Cancer Database

OR

odds ratio

PTC

papillary thyroid cancer

RAI

radioactive iodine

RR

relative risk

SEER

Surveillance, Epidemiology, and End Results

Funding

A.G. is supported by the National Cancer Institute (NCI grant No. 3U54CA118948-17S1).

Disclosures

A.G., H.C., T.S.W., and S.D. have nothing to declare. The authors report no proprietary or commercial interest in any product mentioned or concept discussed in this article. No authors report any financial or intellectual conflicts of interest.

Data Availability

Original data generated and analyzed during this study are included in this published article or in the data repositories listed in “References.”