Abstract
Background:
Racially minoritized patients with thyroid cancer are less likely to receive high-quality and guideline-concordant care. Inaccessibility of high-volume centers may contribute to inequalities in thyroid cancer outcomes. This study sought to understand the extent to which access to higher volume thyroid cancer centers is associated with patient outcomes.
Methods:
We queried linked California Cancer Registry and California Office of Statewide Health Planning and Development databases for thyroid cancer patients who received thyroid surgery between 1999 and 2017. Hospitals were stratified by their median annual volume of thyroid cancer operations: ultra-low volume (0–5 cases/year), low-volume (6–25 cases/year), mid-volume (26–50 cases/year), and high-volume (>50 cases/year). We analyzed the rates of complications, rates of reoperation for cancer recurrence, use of radioactive iodine (131I), and mortality by median hospital volume of thyroid surgery. A multivariable regression controlled for high-risk tumor features. Differences in access by center volume were assessed based on patient demographics.
Results:
We studied 52,599 thyroid cancer patients who underwent thyroidectomy. Patients who underwent thyroidectomy at ultra-low volume centers were more likely to undergo reoperations for recurrent/persistent disease compared with patients at low- (odds ratio [OR] 1.17 [CI 1.02–1.35]), mid- (OR 1.25 [CI 1.06–1.46]), and high-volume centers (OR 1.26 [CI 1.03–1.56]). Patients who received thyroid operations at ultra-low volume centers were also less likely to receive guideline-concordant 131I ablation compared with patients at higher volume centers (OR 0.77 [CI 0.72–0.82]). A pair-wise comparison between all volume categories for all outcomes revealed no statistically significant differences in outcomes between low-, mid-, or high-volume centers. Only ultra-low volume centers had significantly higher rates of adverse outcomes. Ultra-low volume centers were disproportionately accessed by women (p < 0.05), Hispanic, Asian/Pacific Islander, and American Indian people (p < 0.01), those from the lowest three quintiles of socio-economic status (p < 0.01), and the uninsured and those on Medicaid or Medicare (p < 0.01) when compared with higher volume centers.
Conclusions:
Patients receiving thyroid cancer surgery at centers performing ≤5 such operations per year were more likely to require reoperation for recurrent/persistent disease and less likely to receive appropriate 131I ablation. Ultra-low volume centers served higher proportions of socially and economically marginalized communities.
Introduction
Racially minoritized patients with thyroid cancer are less likely to receive high-quality and guideline-concordant care. 1 The mechanisms underlying these inequities are poorly understood. The surgeon volume-outcome association for thyroid surgery is well described. 2 For patients with well-differentiated thyroid cancer (WDTC), disease-free survival is higher when the operative surgeon performs more than 40 thyroidectomies a year. 3
Racial disparities in access to high-volume surgeons appear to have downstream effects on patient outcomes. For example, Hispanic and Black patients are more likely to undergo surgery by low-volume surgeons, 4,5 which is associated with higher in-hospital mortality, longer length of stay, and higher rates of complications. 5 –7
While prior work has demonstrated disparities in thyroid surgery outcomes based on surgeon volume, data characterizing the impact of hospital volume on complications and cancer outcomes, as well as inequalities in patient access to high-volume centers, remain limited. Although there is general consensus that surgeon volume for thyroid surgery matters, there is no such consensus on hospital volume, which extends beyond the surgeon's expertise and considers the systems, resources, and support available at differentially volumed centers.
The current study investigates whether the thyroid surgery volume-outcome relationship may contribute to the observed racial and socio-economic inequalities in thyroid cancer treatment. We examined the rate of complications, reoperations, radioactive iodine (131I) use, and mortality in thyroid cancer patients undergoing thyroid surgery, with the hypothesis that thyroid cancer patients receiving care at low-volume centers experience poorer outcomes than those receiving care at high-volume centers. We further hypothesized that unequal access to higher volume centers for thyroid cancer treatment may contribute to the observed disparities in thyroid cancer outcomes.
Methods
Patient selection
Institutional Review Board (IRB) approvals from the California Committee for the Protection of Human Subjects (IRB No. 14-10-1768) and the UCLA Office of the Human Research Protection Program (IRB No. 14-001553-CR-00005) were obtained. We obtained the linked California Cancer Registry (CCR) and California Office of Statewide Health Planning and Development Patient Discharge Data (OSHPD-PDD) and California Office of Statewide Health Planning and Development Ambulatory Surgery Center (OSHPD-ASC) databases for thyroid cancer patients from 1999–2017.
The CCR is a comprehensive central cancer registry that mandates cancer reporting by law. The OSHPD-PD and OSHPD-AS databases include admissions and procedure details from all non-federal licensed hospitals and ASCs in California. We limited our sample to patients who had thyroid surgery during the study period.
Thyroid cancer operations were extracted from OSHPD to include any procedure code indicating a thyroid lobectomy, thyroidectomy, or lymph node dissection. The subsequent patient cohort was examined for clinically inconsistent outlier data, and these patients were excluded after individual review.
Patient characteristics, including age, sex, race, ethnicity, Charlson Comorbidity Index, and insurance status, were analyzed. For a measure of patient socioeconomic status, Yang quintiles of socioeconomic status, which are area-based socioeconomic measures from American Community Survey data, were used.
We also evaluated the census tract characteristics of patients, which are provided by the CCR as socioeconomic indicators, including median household income, average years of schooling, percent of residents with blue collar jobs, percent of people living below the 200% Federal Poverty Line, and percent unemployment.
Outcome measures
The main adverse outcomes of interest, specified below, were defined and evaluated in the database. This part of the study required variable operationalization, as discrete data for complications and surgical outcomes are not directly reported in either the CCR or OSPHD databases (Supplementary Appendix S1).
The adverse outcomes of interest were: complications, stratified by subtype (endocrine, nerve, airway, wound, and bleeding); use of guideline-concordant adjuvant 131I ablation; need for reoperation for cancer recurrence; and both 30-day and in-hospital mortality.
To determine the occurrence of these outcomes, the OSHPD-PD and ASC databases were evaluated for complications in the following way: nerve complications included procedure codes consistent with treatment or diagnosis of vocal cord paralysis; bleeding complications included diagnosis codes consistent with hemorrhage or hemorrhage-related procedure, such as blood transfusion or evacuation of hematoma; wound complications included diagnosis codes for seroma, wound dehiscence, infection, or drainage and debridement procedures; endocrine complications included diagnosis codes for hypoparathyroidism and hypocalcemia; and airway complications included procedure codes for tracheostomy (Supplementary Appendix S2).
Reoperation was defined and operationalized as any subsequent thyroid or lymph node operation beyond the index operation, excluding two lobectomies ≤90 days apart, or a lobectomy followed by either a subtotal or total thyroidectomy within 90 days.
Guideline-concordant use of 131I was extracted from the radiation summary and stratified by patients satisfying “hard indication” or “no indication” criteria. All patients who underwent either a total thyroidectomy or two lobectomies ≤90 days apart were assessed for whether they were administered 131I.
Patients who had either distant metastases, nodal disease, or extrathryoidal extension were considered to have a “hard indication,” while patients with tumors less than 1 cm in size and no nodal disease were considered to have “no indication” for 131I ablation. In-hospital and 30-day mortality was operationalized as discharge status “expired” either at the initial hospitalization or at any hospitalization within 30 days of the index operation for thyroid cancer. These two variables for mortality were aggregated in analysis given the rarity of such events.
Independent variables/covariates
Based on prior literature about the surgical volume-outcome relationship in thyroid cancer, the hospital and ambulatory centers were stratified by their median annual volume of thyroid cancer operations across the study period. 2,3,7 –9 Thus, four bands of facility volume were selected: ultra-low (0–5 thyroid cancer cases/year), low (6–25 cases/year), mid (26–50 cases/year), and high (>50 cases/year). The relationship between outcomes and hospital volume was additionally examined as a continuous variable to assess concordance with analyses performed using volume categories.
Statistical analysis
Descriptive statistics of all adverse outcomes by subtype (mortality, reoperation, complication, use of 131I) by hospital volume status were reviewed. Chi-square test was used to compare categorical variables, and Wilcoxon rank-sum test or t-test was used to compare continuous variables. Subsequently, a mixed-effects logistic regression controlled for thyroid cancer tumor features that make the occurrence of adverse outcomes more likely inherent to the disease process: tumor size >4 cm, high-risk tumor histology, extrathyroidal extension, and lymphovascular invasion, with a random intercept for hospital.
These tumor features were extracted from the CCR, and the discrete values for each category were operationalized into dichotomous variables for the existence or absence of each subset of high-risk tumor feature. We examined the relationship between outcomes and hospital volume as a continuous variable to confirm that the set tranches of hospital volume were consistent with our findings.
A pair-wise comparison was performed between all four levels of hospital volume for all outcomes to identify any statistically significant differences occurring across differentially volumed centers. A sensitivity analysis was performed to assess complication subtypes (wound, bleeding, airway, endocrine, and nerve) by hospital volume status. Patient demographics were extracted from CCR and OSHPD databases to evaluate the risk ratio of receiving care at ultra-low volume centers by both patient demographics (age, sex, and census tract characteristics).
All analyses were conducted using R (version 4.2.0; R Foundation for Statistical Computing, Vienna, Austria).
Results
The cohort comprised 52,599 thyroid cancer patients treated surgically at 505 hospital-based and ASCs during the study period (Fig. 1). High- and low-volume centers performed the majority of thyroid cancer operations (34% and 39%, respectively), while mid- and ultra-low volume centers performed fewer such operations (15% and 12%, respectively) (Fig. 2). The annual number of thyroid operations at each center remained stable and similar to the calculated median volume throughout the study period, suggesting that each center was appropriately categorized (Fig. 3).
Study population. CCR, California Cancer Registry.
Comparison of the number of centers as categorized by median annual thyroid cancer surgical volume with the total number of thyroid cancer operations performed by the respective center volume categories over the study period, 1999–2017.
Box-and-whisker plots demonstrating stability over time of annual center thyroid cancer surgical volume when median surgical volume over the study period is used to define the center's volume category, 1999–2017 (n = 505 centers).
A pair-wise comparison between all volume categories for all outcomes revealed no statistically significant differences in outcomes between low-, mid-, or high-volume centers. Only ultra-low volume centers had significantly higher rates of adverse outcomes.
Ultra-low volume centers were disproportionately accessed by women (p = 0.03), Hispanic, Asian/Pacific Islander, and American Indian people (p < 0.01), those from the lowest three quintiles of socio-economic status (p < 0.01), and the uninsured and those on Medicaid or Medicare, (p < 0.01) when compared with higher volume centers (Fig. 4). Men, White people, those from the highest two quintiles of socio-economic status, and the privately insured had significantly lower odds of receiving care at an ultra-low volume center (p < 0.01) (Table 1).
RR of receiving surgery at ultra-low versus higher volume centers by selected patient demographics; patients with an RR >1 are more likely to receive surgery at ultra-low volume centers. RR, risk ratio; SES, socioeconomic status.
Demographic and Treatment Characteristics of Patients by Center Volume
p-Value comparing demographic and treatment characteristics between Ultra-low and Aggregated Higher Volume Centers.
Anaplastic, High-risk Papillary or Follicular thyroid cancer, or Medullary Thyroid Cancer.
Receipt of 131I ablation for patients who had either a total thyroidectomy or 2 thyroid lobectomies within 90 days.
I, radioactive iodine; ETE, extrathyroidal extension; HMO, health maintenance organization; LVI, lymphovascular invasion; NOS, not otherwise specified; PPO, preferred provider organization; SD, standard deviation; SES, socioeconomic status.
Patients at ultra-low volume centers were more likely to come from census tracts with lower median household incomes, fewer years of schooling, more blue-collar jobs, more unemployment, and more people living below the 200% Federal Poverty Line (p < 0.01) (Table 1).
Patients at ultra-low volume centers received more limited index operations, with higher rates of thyroid lobectomies, and the lowest rate of total thyroidectomy with lymph node dissection of all hospital volume categories (Table 1).
Unadjusted rates of complications, reoperations, in-hospital and 30-day mortality, and use of 131I were reviewed (Table 2). After adjusting for high-risk tumor features, patients who underwent thyroid operations at ultra-low volume centers were more likely to undergo reoperations for recurrent disease compared with patients at low- (odds ratio [OR] 1.17 95% confidence interval [CI 1.02–1.35]), mid- (OR 1.25 [CI 1.06–1.46]), and high-volume centers (OR 1.26 [CI 1.03–1.56]) (Table 2) (Supplementary Appendix S3).
Outcomes After Surgery for Thyroid Cancer Occurring by Hospital Volume
Significant values p < 0.05 are bolded.
Adjusted for covariates: high-risk tumor histology, tumor size >4 cm, extrathyroidal extension, and lymphovascular invasion.
Aggregated in-hospital and 30-day mortality.
Denotes patients who had either a total thyroidectomy or a completion thyroidectomy AND had evidence of regional nodal disease based on pathology, evidence of distant metastatic disease, or extrathyroidal extension on pathology.
Denotes patients who had either a total thyroidectomy or a completion thyroidectomy AND had with tumors <1 cm in size and no nodal disease.
Patients who underwent total thyroidectomies with pathological evidence of nodal disease or extrathyroidal extension were less likely to receive guideline-concordant adjuvant 131I ablation compared with patients at higher volume centers. Of patients with these hard indications for 131I ablation, 7824 people (64.9%) at high volume, 3291 (60.7%) at mid volume, 8494 (63.7%) at low volume, and 2316 (57.4%) at ultra-low volume centers actually received it.
When comparing ultra-low volume centers with aggregated higher volume centers, this represented an odds ratio of 0.77 [CI 0.72–0.82] of receiving guideline-concordant 131I ablation. Interestingly, for patients who had no indication for 131I ablation but received it regardless, there were no differences by hospital volume (Table 2).
Finally, patients at ultra-low volume centers were more likely to experience wound (OR 1.44 [CI 1.06–1.94], p = 0.02) and airway complications (OR 1.21 [CI 1.02–1.43], p = 0.03) compared with patients at mid-volume centers (Supplementary Appendix S1). There were no statistically significant differences in 30-day or in-hospital mortality or aggregated complications.
Discussion
Our study found that patients receiving thyroid cancer surgery at centers performing ≤5 such operations per year are more likely to require reoperation for structural disease recurrence. We also found that patients receiving thyroid cancer surgery at centers performing ≤5 such operations per year are less likely to receive guideline-concordant 131I ablation, a possible mechanistic explanation for the observed higher rates of cancer recurrence among this patient cohort. There were no differences in aggregated rates of common surgical complications or in-hospital and 30-day mortality by hospital volume.
While prior literature for thyroid surgery has focused mostly on surgeon volume, center volume is a well-described variable in predicting better surgical outcomes. 10 –12 Our study found that for centers treating more than five thyroid cancer patients annually, there was no difference in outcomes. This is a much lower threshold than previous thyroid surgeon-volume literature has described. 2,3
While surgical technique for thyroid cancer is important and may improve with volume, cancer outcomes are dependent on the multidisciplinary networks of care available to the patient. We suspect that a center's volume is a marker of experience with the continuum of thyroid cancer care, including detection, diagnosis, timely receipt of surgery, appropriate 131I ablation, and ongoing surveillance.
The inferior patient outcomes at ultra-low volume centers expand on our current understanding of the relationship between surgical volume and patient outcomes, specifically in thyroid cancer. A National Cancer Database retrospective study demonstrated that high-volume centers for thyroidectomies (>12 cases/year) achieve higher rates of negative margins for WDTC, but did not evaluate surgical complications, reoperation, or cancer recurrence. 8
An additional retrospective cohort study set in the UK's National Health Service demonstrated that higher hospital volume is associated with lower 30-day readmission and lower hypoparathyroidism rates one year post-operatively, but also did not evaluate cancer recurrence. 13
The goal of thyroid cancer index operations is to remove all visible thyroid cancer. Incomplete resections or undetected nodal disease increase the disease recurrence risk. The influence of surgeon volume on thyroid surgery outcomes has been extensively documented, 2,3 and prior reports have described that low-volume centers achieve lower rates of margin negativity, 8 which may partially explain our findings on reoperation rates. The CCR does not collect data on margin status, and thus we cannot assess whether ultra-low volume centers had higher rates of incomplete resection in our cohort.
However, margin status could be unrelated to lymph node recurrence. Failure to detect and clear positive lymph nodes at the time of the index operation may contribute to higher rates of recurrence. Ultra-low volume centers are performing fewer total thyroidectomies with lymph node dissections than higher volume centers, which could suggest an opportunity for improvement. Even so, reoperations for lymph node recurrence are independently associated with mortality, highlighting the importance of achieving negative margin status at the index operation. 14
Our study found that patients receiving thyroid cancer surgery at centers performing ≤5 such operations per year are less likely to receive guideline-concordant 131I ablation. The American Thyroid Association recommends considering the use of 131I ablation for low- to intermediate-risk thyroid cancers, and recommends its use for high-risk cancers after total thyroidectomy. 15
In these patients, appropriate use of adjuvant 131I may improve disease-specific survival and progression-free survival. 15 –17 Therefore, failure to receive these guideline-concordant adjuvant therapies at ultra-low volume centers may increase the risk of recurrence. Our findings concerning 131I ablation must be interpreted cautiously, as overall rates of 131I ablation, particularly for low-risk tumors, have been declining in recent years as part of ongoing efforts to avoid overtreatment. 18
Several reasons may underlie the lower rate of appropriate 131I use at ultra-low volume centers. Administration of 131I necessitates the availability of nuclear medicine, a highly specialized field involving specialty equipment, infrastructure, and personnel to deliver care.
These may not be available in lower resourced settings. Many ultra-low volume centers, particularly in rural areas, do not have endocrinologists, the practitioners who typically prescribe and oversee 131I ablation. Encouraging more fellowship-trained endocrinologists to provide at least partial coverage in rural areas, particularly by telemedicine, may close the aperture on some of these disparities.
Patients receiving 131I also require physical isolation for a period of time, during which most are unable to work. 19 These restrictions may prove impossible for some patients, particularly those of limited financial means. Indeed, this may be especially true at ultra-low volume centers, where we found that patients are more likely to be from financially marginalized backgrounds.
Patients receiving care at ultra-low volume centers were more likely to be from racially, ethnically, and financially marginalized populations. Although 5-year survival for locoregional WDTC approaches 100%, 20 the need for a potentially avoidable reoperation for cancer recurrence should be interrogated closely.
Reoperations impose significant costs upon patients, and they may increase the financial toxicity and psychosocial strain of ongoing treatment for thyroid cancer. 21,22 Higher reoperation rates at ultra-low volume centers may carry profound consequences for financially marginalized patients who are least able to cope with additional health care costs.
We acknowledge several limitations to this study. Many ultra-low volume centers provide an essential safety net to patients who cannot otherwise access care at higher volume centers, whether due to geographic, personal, or insurance payor restrictions. Low-volume and rural hospitals are capable of providing the highest quality, guideline-concordant cancer care, especially when they participate in quality standards programs. 23,24
Patients prefer to undergo surgery close to home, even when travel to a higher-volume center would result in lower operative mortality risk. 25 It is essential to recognize the excellent work that is being done by lower-volume centers, which in our study provide parity of care above a threshold of 5 cases per year. Patient preference in seeking care close to home should be respected: partial coverage at ultra-low volume centers by fellowship-trained endocrine surgeons should be evaluated as a potential bridge to improve outcomes locally.
In conclusion, patients receiving thyroid cancer surgery at centers performing ≤5 such operations per year were more likely to require reoperation for recurrent/persistent disease and less likely to receive appropriate 131I ablation. Ultra-low volume centers served higher proportions of socially and economically marginalized communities.
Quality improvement initiatives at ultra-low volume centers may focus on these areas to improve delivery of high-quality thyroid cancer treatment. Partial coverage by fellowship-trained endocrinologists and endocrine surgeons at these centers may provide an opportunity to close the aperture on these disparities. Equally, higher volume centers should interrogate their financial barriers to patients receiving thyroid cancer care.
Footnotes
Acknowledgments
The authors would like to thank Dr. Iuliana Bobanga, the only board-certified endocrine surgeon in Alaska, for her valuable insight on delivering the highest quality care at lower-volume centers.
Authors' Contributions
H.H.-P.: Conceptualization, methodology, and writing–original draft. Y.M.: Writing–original draft, review and editing. C.-H.T. and J.K.: Methodology, formal analysis. M.W.Y.: Conceptualization, methodology, and writing–review and editing. J.X.W.: Supervision, methodology, and writing–review and editing.
Author Disclosure Statement
No competing financial interests exist.
Funding Information
H.H.-P. was supported by the VA Office of Academic Affiliations through the National Clinician Scholars Program. The contents do not represent the views of the U.S. Department of Veterans Affairs or the United States Government.
Supplementary Material
Supplementary Appendix S1
Supplementary Appendix S2
Supplementary Appendix S3