[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address 54.163.129.96. Please contact the publisher to request reinstatement.
[Skip to Content Landing]
Download PDF
Figure.
Risk of Postoperative Complication for Selected Variables
Risk of Postoperative Complication for Selected Variables

Risk of postoperative complications by surgeon volume (P < .001), hospital volume (P < .001), hospital region (P < .001), urban/rural location (P = .41), and hospital teaching status (P = .37).

Table 1.  
Descriptive Statistics of the Study Population
Descriptive Statistics of the Study Population
Table 2.  
Adjusted Weighted Odds Ratio for the Risk of Being Managed by a Low-Volume Surgeon and at a Low-Volume Hospital Based on Race/Ethnicity and Socioeconomic Characteristics, Stratified by Hospital Region
Adjusted Weighted Odds Ratio for the Risk of Being Managed by a Low-Volume Surgeon and at a Low-Volume Hospital Based on Race/Ethnicity and Socioeconomic Characteristics, Stratified by Hospital Region
Table 3.  
Adjusted Weighted Odds Ratio for the Risk of Being Managed at a Rural Hospital and at a Nonteaching Hospital Based on Race/Ethnicity and Socioeconomic Characteristics, Stratified by Hospital Region
Adjusted Weighted Odds Ratio for the Risk of Being Managed at a Rural Hospital and at a Nonteaching Hospital Based on Race/Ethnicity and Socioeconomic Characteristics, Stratified by Hospital Region
1.
Healthcare Cost and Utilization Project (HCUP). Hospital-based ambulatory surgery, 2007: statistical brief 86. Rockville, MD: Agency for Healthcare Research and Quality; 2010. http://www.hcup-us.ahrq.gov/reports/statbriefs/sb86.jsp. Accessed February 4, 2012.
2.
Chandrasekhar  SS, Randolph  GW, Seidman  MD,  et al; American Academy of Otolaryngology-Head and Neck Surgery.  Clinical practice guideline: improving voice outcomes after thyroid surgery. Otolaryngol Head Neck Surg. 2013;148(6)(suppl):S1-S37.
PubMedArticle
3.
Davies  L, Welch  HG.  Increasing incidence of thyroid cancer in the United States, 1973-2002. JAMA. 2006;295(18):2164-2167.
PubMedArticle
4.
Birkmeyer  JD, Siewers  AE, Finlayson  EV,  et al.  Hospital volume and surgical mortality in the United States. N Engl J Med. 2002;346(15):1128-1137.
PubMedArticle
5.
Birkmeyer  JD, Sun  Y, Wong  SL, Stukel  TA.  Hospital volume and late survival after cancer surgery. Ann Surg. 2007;245(5):777-783.
PubMedArticle
6.
Finlayson  EV, Goodney  PP, Birkmeyer  JD.  Hospital volume and operative mortality in cancer surgery: a national study. Arch Surg. 2003;138(7):721-726.
PubMedArticle
7.
Cheung  MC, Hamilton  K, Sherman  R,  et al.  Impact of teaching facility status and high-volume centers on outcomes for lung cancer resection: an examination of 13,469 surgical patients. Ann Surg Oncol. 2009;16(1):3-13.
PubMedArticle
8.
Chen  H, Zeiger  MA, Gordon  TA, Udelsman  R.  Parathyroidectomy in Maryland: effects of an endocrine center. Surgery. 1996;120(6):948-953.
PubMedArticle
9.
Zelen  J, Bilfinger  TV, Anagnostopoulos  CE.  Coronary artery bypass grafting. The relationship of surgical volume, hospital location, and outcome. N Y State J Med. 1991;91(7):290-292.
PubMed
10.
Birkmeyer  JD, Stukel  TA, Siewers  AE, Goodney  PP, Wennberg  DE, Lucas  FL.  Surgeon volume and operative mortality in the United States. N Engl J Med. 2003;349(22):2117-2127.
PubMedArticle
11.
Dimick  JB, Cowan  JA  Jr, Stanley  JC, Henke  PK, Pronovost  PJ, Upchurch  GR  Jr.  Surgeon specialty and provider volumes are related to outcome of intact abdominal aortic aneurysm repair in the United States. J Vasc Surg. 2003;38(4):739-744.
PubMedArticle
12.
Courcoulas  A, Schuchert  M, Gatti  G, Luketich  J.  The relationship of surgeon and hospital volume to outcome after gastric bypass surgery in Pennsylvania: a 3-year summary. Surgery. 2003;134(4):613-623.
PubMedArticle
13.
Nguyen  NT, Paya  M, Stevens  CM, Mavandadi  S, Zainabadi  K, Wilson  SE.  The relationship between hospital volume and outcome in bariatric surgery at academic medical centers. Ann Surg. 2004;240(4):586-594.
PubMed
14.
Ho  V, Heslin  MJ.  Effect of hospital volume and experience on in-hospital mortality for pancreaticoduodenectomy. Ann Surg. 2003;237(4):509-514.
PubMed
15.
Sosa  JA, Bowman  HM, Gordon  TA,  et al.  Importance of hospital volume in the overall management of pancreatic cancer. Ann Surg. 1998;228(3):429-438.
PubMedArticle
16.
Casson  AG, van Lanschot  JJ.  Improving outcomes after esophagectomy: the impact of operative volume. J Surg Oncol. 2005;92(3):262-266.
PubMedArticle
17.
Sosa  JA, Bowman  HM, Tielsch  JM, Powe  NR, Gordon  TA, Udelsman  R.  The importance of surgeon experience for clinical and economic outcomes from thyroidectomy. Ann Surg. 1998;228(3):320-330.
PubMedArticle
18.
Loyo  M, Tufano  RP, Gourin  CG.  National trends in thyroid surgery and the effect of volume on short-term outcomes. Laryngoscope. 2013;123(8):2056-2063.
PubMedArticle
19.
Gourin  CG, Tufano  RP, Forastiere  AA, Koch  WM, Pawlik  TM, Bristow  RE.  Volume-based trends in thyroid surgery. Arch Otolaryngol Head Neck Surg. 2010;136(12):1191-1198.
PubMedArticle
20.
Cherenfant  J, Gage  M, Mangold  K,  et al.  Trends in thyroid surgery in Illinois. Surgery. 2013;154(5):1016-1023.
PubMedArticle
21.
Kandil  E, Noureldine  SI, Abbas  A, Tufano  RP.  The impact of surgical volume on patient outcomes following thyroid surgery. Surgery. 2013;154(6):1346-1353.
PubMedArticle
22.
Ahuja  N, Chang  D, Gearhart  SL.  Disparities in colon cancer presentation and in-hospital mortality in Maryland: a ten-year review. Ann Surg Oncol. 2007;14(2):411-416.
PubMedArticle
23.
Doubeni  CA, Field  TS, Buist  DS,  et al.  Racial differences in tumor stage and survival for colorectal cancer in an insured population. Cancer. 2007;109(3):612-620.
PubMedArticle
24.
Singh  JA, Lu  X, Rosenthal  GE, Ibrahim  S, Cram  P.  Racial disparities in knee and hip total joint arthroplasty: an 18-year analysis of national Medicare data [published online September 18, 2013]. Ann Rheum Dis. doi:10.1136/annrheumdis-2013-203494.
PubMed
25.
Hravnak  M, Ibrahim  S, Kaufer  A, Sonel  A, Conigliaro  J.  Racial disparities in outcomes following coronary artery bypass grafting. J Cardiovasc Nurs. 2006;21(5):367-378.
PubMedArticle
26.
Yeung  M, Kerrigan  J, Sodhi  S,  et al.  Racial differences in rates of aortic valve replacement in patients with severe aortic stenosis. Am J Cardiol. 2013;112(7):991-995.
PubMedArticle
27.
Martinez  SR, Chen  SL, Bilchik  AJ.  Treatment disparities in Hispanic rectal cancer patients: a SEER database study. Am Surg. 2006;72(10):906-908.
PubMed
28.
Tammemagi  CM.  Racial/ethnic disparities in breast and gynecologic cancer treatment and outcomes. Curr Opin Obstet Gynecol. 2007;19(1):31-36.
PubMedArticle
29.
Lim  II, Hochman  T, Blumberg  SN, Patel  KN, Heller  KS, Ogilvie  JB.  Disparities in the initial presentation of differentiated thyroid cancer in a large public hospital and adjoining university teaching hospital. Thyroid. 2012;22(3):269-274.
PubMedArticle
30.
Radowsky  JS, Helou  LB, Howard  RS, Solomon  NP, Stojadinovic  A.  Racial disparities in voice outcomes after thyroid and parathyroid surgery. Surgery. 2013;153(1):103-110.
PubMedArticle
31.
Zevallos  JP, Xu  L, Yiu  Y.  The impact of socioeconomic status on the use of adjuvant radioactive iodine for papillary thyroid cancer. Thyroid. 2014;24(4):758-763.Article
32.
Harari  A, Li  N, Yeh  MW.  Racial and socioeconomic disparities in presentation and outcomes of well-differentiated thyroid cancer. J Clin Endocrinol Metab. 2014;99(1):133-141.
PubMedArticle
33.
Sosa  JA, Mehta  PJ, Wang  TS, Yeo  HL, Roman  SA.  Racial disparities in clinical and economic outcomes from thyroidectomy. Ann Surg. 2007;246(6):1083-1091.
PubMedArticle
34.
Noureldine  SI, Abbas  A, Tufano  RP,  et al.  The impact of surgical volume on racial disparity in thyroid and parathyroid surgery. Ann Surg Oncol. 2014;21(8):2733-2739.
PubMedArticle
35.
Keppel  KG, Pearcy  JN, Wagener  DK. Trends in Racial and Ethnic-Specific Rates for the Health Status Indicators: United States, 1990-98: Healthy People Statistical Notes, No. 23. Hyattsville, MD: National Center for Health Statistics; January 2002.
36.
Healthcare Cost and Utilization Project. Overview of the National (Nationwide) Inpatient Sample (NIS). Rockville, MD: Agency for Healthcare Research and Quality; 2013. http://www.hcup-us.ahrq.gov/nisoverview.jsp. Accessed February 2, 2014.
37.
Quan  H, Sundararajan  V, Halfon  P,  et al.  Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005;43(11):1130-1139.
PubMedArticle
38.
Lantz  PM, House  JS, Lepkowski  JM, Williams  DR, Mero  RP, Chen  J.  Socioeconomic factors, health behaviors, and mortality: results from a nationally representative prospective study of US adults. JAMA. 1998;279(21):1703-1708.
PubMedArticle
39.
Sorlie  PD, Backlund  E, Keller  JB.  US mortality by economic, demographic, and social characteristics: the National Longitudinal Mortality Study. Am J Public Health. 1995;85(7):949-956.
PubMedArticle
40.
Andrulis  DP.  Access to care is the centerpiece in the elimination of socioeconomic disparities in health. Ann Intern Med. 1998;129(5):412-416.
PubMedArticle
41.
Fiscella  K, Franks  P, Gold  MR, Clancy  CM.  Inequality in quality: addressing socioeconomic, racial, and ethnic disparities in health care. JAMA. 2000;283(19):2579-2584.
PubMedArticle
42.
Liu  JH, Zingmond  DS, McGory  ML,  et al.  Disparities in the utilization of high-volume hospitals for complex surgery. JAMA. 2006;296(16):1973-1980.
PubMedArticle
43.
Rothenberg  BM, Pearson  T, Zwanziger  J, Mukamel  D.  Explaining disparities in access to high-quality cardiac surgeons. Ann Thorac Surg. 2004;78(1):18-25.
PubMedArticle
44.
Epstein  AJ, Gray  BH, Schlesinger  M.  Racial and ethnic differences in the use of high-volume hospitals and surgeons. Arch Surg. 2010;145(2):179-186.
PubMedArticle
45.
Bristow  RE, Zahurak  ML, Ibeanu  OA.  Racial disparities in ovarian cancer surgical care: a population-based analysis. Gynecol Oncol. 2011;121(2):364-368.
PubMedArticle
46.
Chang  CM, Yin  WY, Wei  CK,  et al.  The association of socioeconomic status and access to low-volume service providers in breast cancer. PLoS One. 2013;8(12):e81801.
PubMedArticle
47.
Quan  H, Li  B, Saunders  LD,  et al; IMECCHI Investigators.  Assessing validity of ICD-9-CM and ICD-10 administrative data in recording clinical conditions in a unique dually coded database. Health Serv Res. 2008;43(4):1424-1441.
PubMedArticle
48.
Bell  N, Schuurman  N, Hayes  MV.  Using GIS-based methods of multicriteria analysis to construct socio-economic deprivation indices. Int J Health Geogr. 2007;6:17.
PubMedArticle
Original Investigation
December 2014

Association of Socioeconomic Status, Race, and Ethnicity With Outcomes of Patients Undergoing Thyroid Surgery

Author Affiliations
  • 1Department of Surgery, Tulane University School of Medicine, New Orleans, Louisiana
  • 2Department of Otolaryngology, Tulane University School of Medicine, New Orleans, Louisiana
JAMA Otolaryngol Head Neck Surg. 2014;140(12):1173-1183. doi:10.1001/jamaoto.2014.1745
Abstract

Importance  For the management of thyroid diseases, there have been few studies aimed at examining the association between disparities and outcomes.

Objective  To measure the effects of race, ethnicity, and socioeconomic status on outcomes following thyroid surgery.

Design, Setting, and Participants  Cross-sectional analysis of 62 722 thyroid procedures identified in the Nationwide Inpatient Sample (NIS) from 2003 through 2009.

Interventions  Thyroidectomy.

Main Outcomes and Measures  The first set of outcomes included postoperative complication, length of stay (LOS), and overall cost in relation to selected hospital and surgeon characteristics. The second set encompassed accessibility to different surgeon and hospital volumes, hospital locations, and hospital teaching status based on race/ethnicity, income, and health service payer.

Results  The majority of cases were total thyroidectomies (57.9%) for benign conditions (60.8%). Low-volume surgeons performed most operations (90.8%). Low surgeon volume was associated with higher risk of postoperative complications compared with higher surgeon volume (17.2% vs 12.1%; P < .001). Low-volume compared with high-volume hospitals had higher rates of postoperative complications (17.7% vs 15.1%; P < .001). High surgeon volume was associated with a decreased LOS (mean [SD], 1.74 [0.02] vs 1.20 [0.07] days; P < .001). In addition, LOS was longer at low-volume hospitals (1.85 [0.02] vs 1.57 [0.03] days; P = .001). Hispanics were more likely to be operated on by low-volume surgeons (odds ratio [OR], 2.04; 95% CI, 1.19-3.48), and in certain regions throughout the United States, blacks were more likely to be operated on by low-volume surgeons. Patients with Medicare (OR, 1.30; 95% CI, 1.13-1.53) and lower income (OR, 1.73; 95% CI, 1.19-2.53) were more likely to be treated at low-volume centers. Minorities, including Hispanics, blacks, and other race/ethnicity, were more likely to have their operation in an urban setting (P < .005 for all). Blacks were less likely to have operations performed at nonteaching institutions (OR, 0.48; 95% CI, 0.38-0.60), as were people without private insurance (P < .05 for Medicare, Medicaid, and self-pay).

Conclusions and Relevance  There are significant socioeconomic and racial disparities in thyroid surgery outcomes. Low-volume centers and surgeons had a significantly longer LOS and higher risk of complications, and inequalities were prevalent concerning access to these high-volume hospitals and surgeons.

Introduction

The rates of thyroid surgery have tripled over the past 3 decades. In 2007, the US Agency for Healthcare Research and Quality (AHRQ) statistics indicated that 37.4 thyroidectomies were performed per 100 000 population.1 Through increased detection and the growing US population, some investigators estimate that the current rates of thyroid surgery are between 118 000 and 166 000 patients in the United States per year.2 In addition, the incidence of thyroid cancer in the United States has risen from 3.6 per 100 000 in 1973 to 8.7 per 100 000 in 2002—a 2.4-fold increase.3

Surgeons perform thyroidectomies for a range of benign and malignant conditions. Like many studies that have demonstrated improved patient outcomes following complex surgical procedures in high-volume centers47 and those that have shown that increased surgeon experience results in better patient outcomes,816 similar volume-outcome relationships have been observed in thyroid surgery.1721 Thyroid surgery is associated with excellent outcomes, including very low morbidity and mortality and short length of hospital stay (LOS), especially when performed at well-equipped and qualified centers by experienced surgeons.

With respect to disparities in health care, there has been an ample amount of literature published on the topic as it pertains to colon cancer, orthopedic surgery, cardiac surgery, and breast and gynecologic surgery.2228 Whether it is owing to clinical differences among racial or ethnic groups, cultural differences and their effect on behavior or decision making, or socioeconomic factors, such studies have consistently demonstrated that disparities exist and are an integral factor in obtaining access to preventive care and surgery. For the management of thyroid diseases, there has been much less effort to examine the association between disparities and outcomes.2934

Health disparities have been documented for decades, and although improvements in health care have led to advances in the management of a variety of diseases, gaps between white individuals and racial minorities have shown little change.35 As such, it seems imperative to continue to examine the influence of race/ethnicity and socioeconomic status on outcomes for a variety of surgical procedures. To date, this is the largest study of its kind to look at the effects of socioeconomic status and race/ethnicity on access to care and clinical and economic outcomes following thyroid surgery.

Methods

This study is a cross-sectional analysis using the Nationwide Inpatient Sample (NIS) database for the years 2003 through 2009. The NIS database is part of the Healthcare Cost and Utilization Project (HCUP) and is sponsored by the Agency for Healthcare Research and Quality. It is the largest all-payer inpatient database that is publicly available in the United States, containing de-identified data from approximately 8 million hospital stays from about 1000 hospitals sampled to approximate a 20%-stratified sample of US community hospitals.36 The International Classification of Disease, Ninth Revision (ICD-9) was used to define the diagnoses and procedures of interest.

The study population consisted of adult (age ≥18 years) inpatients who underwent total/complete thyroidectomy (TT) (ICD-9 code 06.4) or unilateral thyroidectomy (UT) (ICD-9 code 06.2) as the primary procedure. The primary diagnoses were classified into malignant thyroid diseases (ICD-9 codes 193 and 237.4), benign thyroid diseases (ICD-9 codes 245.0-245.4, 245.8, 245.9, 240.0, 240.9, 241.0, 241.1, 241.9, 246.0-246.3, 246.8, 246.9, 242.10, 242.11, 242.20, 242.21, 242.30, 242.31, 242.80, 242.81, 242.90, 242.91, and 226), or Graves disease (ICD-9 codes 242.0, 242.00, and 242.01).

The study had 2 main sets of outcomes: the first set included postoperative complication, LOS, and overall cost in relation to selected hospital and surgeon characteristics, namely surgeon volume (low surgeon volume [<100 thyroidectomies per year] vs high surgeon volume [≥100 thyroidectomies per year]), hospital volume (low hospital volume [<median] vs high hospital volume [≥median] [median = 42 thyroidectomies per year]), hospital location (urban vs rural), and hospital teaching status (teaching hospital vs nonteaching). The HCUP-NIS database contains data on total charges representing the amount billed for services for each hospital. To calculate cost, hospital-specific cost to charge ratios available through HCUP were used, which allow for the conversion of charges to costs. Information used to calculate these ratios was obtained from hospital accounting reports collected by the Centers for Medicare & Medicaid Services (CMS).36

The second set of the outcomes encompassed the accessibility to and availability of different surgeon volumes, hospital volumes, hospital locations, and hospital teaching status to patients based on patients’ race/ethnicity (white, black, Hispanic, other [ including, but not limited to, Asians, Pacific Islanders, and Native Americans]), income (quartile classification), and health service payer (Medicare, Medicaid, private insurance, self-pay).

Dichotomized complications were defined as the presence of 1 or more of the general or specific complications based on the secondary diagnoses made during the hospital stay. We included specific complications related to TT: (1) vocal fold paralysis or hoarseness, (2) hypoparathyroidism, hypocalcemia, or tetany, (3) tracheomalacia, (4) neck seroma or hematoma, and (5) wound complications, as well as those that were indirectly related but that merited routine discussion during the process of informed consent, such as cardiac, pulmonary, and renal complications.21

Other secondary independent factors that were assessed as potential confounders were (1) patient demographics (age [<64, 64-79, ≥80 years], and sex); (2) clinical factors (primary diagnosis [malignant, benign, or Graves disease]), obesity (body mass index ≥30 [calculated as weight in kilograms divided by height in meters squared]), admission type (elective vs nonelective), inpatient death, whether neck dissection was performed, and a modification of the Charlson Comorbidity Index Score, which was used to measure patient comorbidity (scores 0-1 were categorized as low; 2-3, medium low; 4-5, moderate; and ≥6, high)37; and (3) hospital characteristics (hospital bed-size [small, medium, large] and hospital geographic region [Northeast, Midwest, South, West]).

Statistical analyses used weighted data reflecting the national estimate. The recorded weights are available in the NIS data and are calculated based on the stratification variables that were used in sampling methodology. These variables include hospital geographic region, urban/rural location, teaching status, bed size, and ownership.

Cross-tabulation and χ2 test were used to examine the association between each of the independent factors and the outcomes of interest. In examining for potential confounders, the following variables with significant association were considered to have a possible confounding effect and were thus included in multivariable logistic regression models: age, sex, diagnosis, neck dissection status, inpatient death, admission type, hospital bed size, Charlson Comorbidity Index Score, obesity, and hospital region, besides the main factors of interest (race, income, and service payer).

Multivariable logistic regression models were used to calculate odds ratios (ORs) and 95% CIs. Stratification by different hospital regions was performed as a means to examine the variations across the United States. The t test was used to test for the differences in cost and LOS. Significance level was set as α = .05, and all analyses were conducted using SAS 9.2 for Windows (SAS Institute Inc).

Results

A total of 62 722 discharge records were identified based on the inclusion criteria between 2003 and 2009 (Table 1). The mean (SD) age of the study population was 51.8 (0.12) years. The majority of the study population was white (71.7%), was female (80.6%), and had private insurance (65.4%). Most patients had only 1 or no comorbidities (97.5%). Thirty-one subjects died during the hospital stay, which accounted for approximately 0% in the weighted analysis. The low surgeon volume group performed the majority of thyroidectomies (90.8%), while the remaining operations (9.2%) were performed by the high surgeon volume group. Slightly more than the half of the procedures were TT (57.9%). In regard to diagnosis, benign thyroid disease was the most common indication for operation (60.8%), and Graves disease was the least common (3.3%). Malignant conditions made up the remainder (35.9%). Postoperative complications were reported in 10 257 patients (16.4%). The mean (SD) LOS was 1.71 (0.02) days, while the mean (SD) cost was US $6449.82 ($108.62).

Surgeon Volume, Hospital Volume, Hospital Location, and Teaching Status

The first part of the study focused on examining the association of surgeon volume, hospital volume, hospital location, and hospital teaching status with postoperative complications, LOS, and cost. Postoperative complications were higher in the low surgeon volume group compared with the high surgeon volume group (17.2% vs 12.08%; P < .001) (Figure). Also, complications were higher in the low hospital volume group compared with the high hospital volume group (17.7% vs 15.1%; P < .001). Furthermore, the risk of complications were associated with hospital region as well (P < .001), with the highest risk recorded in the South (19.9%) and the lowest risk in the Northeast (13.0%). On the other hand, postoperative complications were not associated with hospital location or hospital teaching status.

With respect to LOS and cost, patients had longer mean (SD) LOSs when managed by low-volume surgeons compared with high-volume surgeons (1.74 [0.02] days vs 1.20 [0.07] days; P < .001). In addition, LOS was longer at low-volume hospitals compared with high-volume hospitals (1.85 [0.02] days vs 1.57 [0.03] days; P = .001), while LOS was not significantly different with respect to hospital location (rural, 1.71 [0.02] days vs urban, 1.68 [0.03] days; P = .52) and teaching status (nonteaching, 1.74 [0.03] days vs teaching, 1.69 [0.04] days; P = .26). Cost (mean [SD]) was not significantly associated with any of the selected characteristics: hospital volume (low, $6557.55 [$80.14] vs high, $6342.37 [$199.94]; P = .32), hospital location (rural, $6457.31 [$183.84] vs urban, $6443.08 [$115.99]; P = .95), surgeon volume (low, $6162.41 [$138.28] vs high, $5505.69 [$424.95]; P = .10), or hospital teaching status (nonteaching, $6254.2 [$132.70] vs teaching, $6584.77 [$162.14]; P = .11).

Hispanic Patients

In the next part of the study, we examined the accessibility to health care in terms of a patient’s racial/ethnic and economic backgrounds. At the national level, Hispanics were more likely to be managed by low-volume surgeons than whites (OR, 2.04; 95% CI, 1.19-3.48 [P = .009]) (Table 2). This finding remained significant when considering each region separately except for in the Midwest. In addition, in the West, Hispanics were more likely to be managed at low-volume hospitals compared with whites (OR, 1.64; 95% CI, 1.02-2.65 [P = .04]) (Table 2). Also, Hispanics were less likely to be managed in rural hospitals compared with whites at the national level (OR, 0.34; 95% CI, 0.21-0.56 [P < .001]) (Table 3). This association remained significant only in the South on regional analysis. Finally, in the Midwest, Hispanics had less probability of being managed in a nonteaching hospital compared with whites (OR, 0.24; 95% CI, 0.07-0.77 [P = .02]) (Table 3).

Black Patients

Blacks in the Northeast were more likely to be managed by low-volume surgeons compared with whites (OR, 1.87; 95% CI, 1.01-3.48 [P = .047]) (Table 2), and in the Midwest and West, blacks were at greater risk of being managed in low-volume hospitals compared with whites (OR, 2.25; 95% CI, 1.09-4.63 [P = .03]; and OR, 1.69; 95% CI, 1.09-2.60 [P = .02], respectively) (Table 2). Blacks were less likely to go to rural hospitals compared with whites on a national level (OR, 0.34; 95% CI, 0.20-0.60 [P < .001]) (Table 3), and this relationship maintained significance in all regions except in the South. In addition, on both a national and regional level, blacks were less likely to be managed in nonteaching hospitals compared with whites (all regions, OR, 0.48; 95% CI, 0.38-0.60 [P < .001]) (Table 3).

Household Income

We next examined the effect that household income had on the accessibility to health care. Nationally, the association that was most prominent and consistent was the risk of being managed at rural hospitals. Risk of going to a rural hospital was linearly and inversely related to the quartile of household income. The risk of the lowest income (<$39 000) compared with the highest income (≥$62 999) was OR, 19.67; 95% CI, 8.06-48.03 (P < .001), while that of the second highest income ($48 000-$62 999) compared with the highest income was OR, 3.73; 95% CI, 1.91-7.29 (P < .001) (Table 3). Furthermore, the likelihood of being managed by the low surgeon volume group or at a low-volume hospital (Table 2) was also significantly associated with income; however, this association was less generalizable across the entire nation and was most prevalent in certain regions. In the South, households with the lowest quartile income were at greater risk of being managed by low-volume surgeons compared with those with the highest quartile income (OR, 4.4; 95% CI, 1.80-10.72 [P = .001]) (Table 2). The risk appears to decrease as income increases, and a similar association was identified in the Midwest. Also in the South, those with lower income were more likely to be managed at a low-volume hospital (OR, 2.48; 95% CI, 1.30-4.70 [P = .006]) (Table 2).

Health Insurance Type

Nationally, Medicare patients were more likely to be managed at low-volume hospitals (OR, 1.31; 95% CI, 1.13-1.53 [P < .001]) (Table 2) . In the South, Medicare patients were more likely to be managed by low-volume surgeons compared with those with private/health maintenance organization (HMO) coverage (OR, 1.74; 95% CI, 1.17-2.59 [P = .007]) (Table 2). The same applied to patients receiving Medicaid. In the Northeast and South, receiving Medicare was associated with higher risk of being managed in low-volume hospitals compared with patients with private/HMO coverage (OR, 1.43; 95% CI, 1.10-1.85 [P = .008]; and OR, 1.29; 95% CI, 1.03-1.61 [P = .03], respectively) (Table 2). Alternatively, receiving Medicaid had variable association with being managed at rural hospitals compared with patients with private/HMO coverage. In the South, for instance, there was a greater risk of being managed in rural hospitals (OR, 1.59; 95% CI, 1.05-2.39 [P = .03]), while in the West; the opposite was true (OR, 0.26; 95% CI, 0.10-0.68 [P = .006]) (Table 3).

Collectively, at the national level, patients receiving Medicare, Medicaid, or self-pay were significantly less likely to be managed at nonteaching hospitals compared with those with private/HMO coverage (OR, 0.84; 95% CI, 0.72-0.99 [P = .03]; OR, 0.67; 95% CI, 0.52-0.86 [P = .002]; and OR, 0.48; 95% CI, 0.31-0.75 [P = .001], respectively) (Table 3). When stratifying by regions, the Midwest was the only region in which association between type of coverage and access to nonteaching hospital was lost.

Discussion

Through a variety of factors such as substandard housing and nutrition, lower educational and economic opportunity, and greater environmental risks, both lower socioeconomic position and minority race/ethnicity contribute to poor health and shortened survival.38,39 These disparities in health-related outcomes are fueled most notably by the inequalities that exist in the access to health care,40 and some would argue that being a member of a minority race or ethnicity predisposes a patient to lower-quality care.41

Many studies have demonstrated volume-outcome relationships between improved patient outcomes following complex operations at high-volume centers47 and by more experienced surgeons.816 Recent studies have shown this to be true for thyroid surgery as well.1721 Given the arguments regarding accessibility to quality health care, one would suspect that the disparities that exist for minorities and those of lower socioeconomic status may potentially be explained by a lack of access to high-volume surgeons and high-volume centers. Liu et al42 showed in their study of Californians between 2000 and 2004 undergoing a variety of cardiovascular, oncologic, and orthopedic surgical procedures that nonwhites, Medicaid patients, and uninsured were more likely to receive care at low-volume hospitals. They emphasized the need for policy development that would include specific efforts to identify the patients and system factors required to reduce the inequalities in access to care at these higher-volume centers.

Similar results have been demonstrated looking at discharge data from the state of New York. Rothenberg and colleagues43 examined data from 1996 and 1997 for patients undergoing coronary artery bypass graft surgery, and later Epstein et al44 looked at data from 2001 to 2004 for New York City. Not only was there a disadvantage for minorities to receive care at high-volume hospitals, but also from high-volume surgeons for procedures with a well-established volume-mortality association. Other studies examining outcomes for ovarian and breast cancers have also demonstrated racial and socioeconomic influences on achieving access to high-volume providers (both surgeons and hospitals).45,46 In the study by Chang et al,46 despite the nationwide universal health care system in Taiwan, patients with breast cancer from a lower socioeconomic status were still more likely to obtain care from low-volume service providers.46

In 2007, Sosa et al33 were the first group to explore these issues as they relate to thyroidectomy. In their study they found that there are significant racial and ethnic disparities in outcomes, that these disparities are pervasive, and that they may in fact be worsening as the gap in access to high-volume thyroid surgeons increases among different races.

To date, our study is the largest of its kind using the most recently available data from the NIS database to examine the correlation between volume and outcomes and the socioeconomic and racial disparities present for thyroid operations. Unlike Sosa et al,33 we categorized surgeon volume into 2 groups for simplicity and used a 2-step algorithmic multivariable analysis, which allowed us to control for possible race-related intrinsic factors, such as genetics. Instead of linking race/ethnicity to the outcome directly and controlling for variables that do not necessarily give weight to intrinsic factors, we linked outcome to both hospital and surgeon characteristics, and consequently we examined what race/ethnicity and socioeconomic backgrounds had access to those characteristics on both a national and regional level. This allows a more confident inference regarding the inequality and disparity in health management with respect to patients’ demographic attributes.

Similar to previously published findings,1721 in our study postoperative complications were significantly more likely to occur when low-volume surgeons performed the procedures (17.2% vs 12.1%) and when patients were managed at low-volume centers (17.7% vs 15.1%) (Figure). Furthermore, we found that complications seemed to be associated with hospital region as well, with the highest risk recorded in the southern part of the United States (19.9%) and the lowest risk in the northeast (13.0%).

Compared with whites, other races and ethnicities, such as blacks and Hispanics, were more likely to be managed by low-volume surgeons and at low-volume institutions throughout different regions of the United States (Table 2). Furthermore, LOS was longer when patients were managed by low-volume surgeons (1.74 vs 1.20 days) and at low-volume hospitals (1.85 vs 1.57 days). Using income and payer status as surrogates for socioeconomic status, we demonstrated the likelihood of being managed by a low-volume surgeon or at a low-volume hospital to be significantly associated with income in certain regions of the United States (Table 2). In the South, households with the lowest quartile income were at greater risk of being managed by low-volume surgeons and at low-volume hospitals compared with those with the highest quartile income. This risk appears to decrease as income increases, and a similar association was identified in other regions, such as the Midwest.

Finally, on a national level, Medicare patients were more likely to be managed at low-volume hospitals, and in the South, Medicare patients were also more likely to be managed by low-volume surgeons compared with those with private/HMO coverage. The same applied to patients on Medicaid. In the Northeast and South, receiving Medicare was also associated with higher risk of being managed in low-volume hospitals. All of these findings together lend credence to access limitations being one of the principal explanations for the observed discrepancy in outcomes by race.

Interestingly, despite showing that there are disparities regarding access to teaching and urban hospitals (Table 3), teaching status and whether a hospital was in an urban or rural location did not appear to have an effect on outcomes (Figure). Other studies have suggested that patients are more likely to present to public hospitals with more advanced-stage of thyroid cancers compared with teaching hospitals29 and that race and low socioeconomic status often lead to delayed presentation and more advanced disease.32 Furthermore, Harari et al32 showed that black patients consistently present with later disease and have worse survival compared with any other racial group and that there appears to be a persistent protective survival advantage of Asian-Pacific Islander and Hispanic patients despite presenting with later-stage disease. Our study encompassed operations for Graves disease (3.3%) and benign (60.8%) and malignant (35.9%) conditions and did not specifically focus on outcomes for thyroid malignancy. Because teaching status and whether the hospital was in an urban or rural location did not represent any significant associations with respect to complications, LOS, or cost within this analysis, it is possible that this is a limitation of the comprehensiveness of the NIS database or that further regional classifications may have been necessary but were outside the scope of this analysis.

There are several limitations and shortcomings to our study. The data set does not allow us to fully adjust for the extent of thyroid disease or stage of thyroid cancer, the NIS database underestimates 30-day complication and mortality rates following all procedures, and coding inconsistencies exist.47 Complications may be diagnosed and followed only on an outpatient basis or coded as separate diagnoses when requiring readmission. Because we are only capturing patients admitted to the hospital, it is possible that these patients represent a sicker demographic than patients typically undergoing outpatient operations, and thus complications may be more robust among this patient population. Some complications are missed by this database because of events occurring following discharge, and it is likely that our results represent a conservative estimate of the risks that may occur following thyroid surgery. In addition, because the NIS database does not capture ambulatory surgery cases, the LOS will be skewed upward since we are only capturing patients who are admitted. Finally, the NIS database provides us with some socioeconomic variables, such as income, primary payer, and rural vs urban locations, but it lacks other variables such as patient education, wealth, and employment status, which may also have an impact on outcomes and access to care.48

Despite these limitations, the NIS database uses weighted data points in an attempt to scale its data to reflect representative information at the national level. All calculations conducted in this study used the provided weighted data to produce the most accurate information on a national level, which makes these findings closely tailored to the US health system.

Conclusions

The risk of complications following thyroid surgery tends to be higher for low-volume surgeons, at low-volume hospitals, and within certain regions of the United States, notably the South. Furthermore, there are prevalent disparities between socioeconomic classes and races with respect to access to these safer high-volume surgeons and institutions.

Back to top
Article Information

Submitted for Publication: April 6, 2014; final revision received June 15, 2014; accepted June 23, 2014.

Corresponding Author: Adam Hauch, MD, MBA, Department of Surgery, Tulane University School of Medicine, 1430 Tulane Ave, SL22, New Orleans, LA 70112 (ahauch@tulane.edu).

Published Online: September 4, 2014. doi:10.1001/jamaoto.2014.1745.

Author Contributions: Drs Hauch and Kandil had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Hauch, Al-Qurayshi, Kandil.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Hauch, Al-Qurayshi.

Critical revision of the manuscript for important intellectual content: Hauch, Al-Qurayshi, Friedlander, Kandil.

Statistical analysis: Hauch, Al-Qurayshi, Kandil.

Administrative, technical, or material support: Kandil.

Study supervision: Friedlander, Kandil.

Conflict of Interest Disclosures: None reported.

Previous Presentation: This study was presented at the Fifth World Congress of the International Federation of Head and Neck Oncologic Societies and the Annual Meeting of the American Head & Neck Society; July 27, 2014; New York, New York.

References
1.
Healthcare Cost and Utilization Project (HCUP). Hospital-based ambulatory surgery, 2007: statistical brief 86. Rockville, MD: Agency for Healthcare Research and Quality; 2010. http://www.hcup-us.ahrq.gov/reports/statbriefs/sb86.jsp. Accessed February 4, 2012.
2.
Chandrasekhar  SS, Randolph  GW, Seidman  MD,  et al; American Academy of Otolaryngology-Head and Neck Surgery.  Clinical practice guideline: improving voice outcomes after thyroid surgery. Otolaryngol Head Neck Surg. 2013;148(6)(suppl):S1-S37.
PubMedArticle
3.
Davies  L, Welch  HG.  Increasing incidence of thyroid cancer in the United States, 1973-2002. JAMA. 2006;295(18):2164-2167.
PubMedArticle
4.
Birkmeyer  JD, Siewers  AE, Finlayson  EV,  et al.  Hospital volume and surgical mortality in the United States. N Engl J Med. 2002;346(15):1128-1137.
PubMedArticle
5.
Birkmeyer  JD, Sun  Y, Wong  SL, Stukel  TA.  Hospital volume and late survival after cancer surgery. Ann Surg. 2007;245(5):777-783.
PubMedArticle
6.
Finlayson  EV, Goodney  PP, Birkmeyer  JD.  Hospital volume and operative mortality in cancer surgery: a national study. Arch Surg. 2003;138(7):721-726.
PubMedArticle
7.
Cheung  MC, Hamilton  K, Sherman  R,  et al.  Impact of teaching facility status and high-volume centers on outcomes for lung cancer resection: an examination of 13,469 surgical patients. Ann Surg Oncol. 2009;16(1):3-13.
PubMedArticle
8.
Chen  H, Zeiger  MA, Gordon  TA, Udelsman  R.  Parathyroidectomy in Maryland: effects of an endocrine center. Surgery. 1996;120(6):948-953.
PubMedArticle
9.
Zelen  J, Bilfinger  TV, Anagnostopoulos  CE.  Coronary artery bypass grafting. The relationship of surgical volume, hospital location, and outcome. N Y State J Med. 1991;91(7):290-292.
PubMed
10.
Birkmeyer  JD, Stukel  TA, Siewers  AE, Goodney  PP, Wennberg  DE, Lucas  FL.  Surgeon volume and operative mortality in the United States. N Engl J Med. 2003;349(22):2117-2127.
PubMedArticle
11.
Dimick  JB, Cowan  JA  Jr, Stanley  JC, Henke  PK, Pronovost  PJ, Upchurch  GR  Jr.  Surgeon specialty and provider volumes are related to outcome of intact abdominal aortic aneurysm repair in the United States. J Vasc Surg. 2003;38(4):739-744.
PubMedArticle
12.
Courcoulas  A, Schuchert  M, Gatti  G, Luketich  J.  The relationship of surgeon and hospital volume to outcome after gastric bypass surgery in Pennsylvania: a 3-year summary. Surgery. 2003;134(4):613-623.
PubMedArticle
13.
Nguyen  NT, Paya  M, Stevens  CM, Mavandadi  S, Zainabadi  K, Wilson  SE.  The relationship between hospital volume and outcome in bariatric surgery at academic medical centers. Ann Surg. 2004;240(4):586-594.
PubMed
14.
Ho  V, Heslin  MJ.  Effect of hospital volume and experience on in-hospital mortality for pancreaticoduodenectomy. Ann Surg. 2003;237(4):509-514.
PubMed
15.
Sosa  JA, Bowman  HM, Gordon  TA,  et al.  Importance of hospital volume in the overall management of pancreatic cancer. Ann Surg. 1998;228(3):429-438.
PubMedArticle
16.
Casson  AG, van Lanschot  JJ.  Improving outcomes after esophagectomy: the impact of operative volume. J Surg Oncol. 2005;92(3):262-266.
PubMedArticle
17.
Sosa  JA, Bowman  HM, Tielsch  JM, Powe  NR, Gordon  TA, Udelsman  R.  The importance of surgeon experience for clinical and economic outcomes from thyroidectomy. Ann Surg. 1998;228(3):320-330.
PubMedArticle
18.
Loyo  M, Tufano  RP, Gourin  CG.  National trends in thyroid surgery and the effect of volume on short-term outcomes. Laryngoscope. 2013;123(8):2056-2063.
PubMedArticle
19.
Gourin  CG, Tufano  RP, Forastiere  AA, Koch  WM, Pawlik  TM, Bristow  RE.  Volume-based trends in thyroid surgery. Arch Otolaryngol Head Neck Surg. 2010;136(12):1191-1198.
PubMedArticle
20.
Cherenfant  J, Gage  M, Mangold  K,  et al.  Trends in thyroid surgery in Illinois. Surgery. 2013;154(5):1016-1023.
PubMedArticle
21.
Kandil  E, Noureldine  SI, Abbas  A, Tufano  RP.  The impact of surgical volume on patient outcomes following thyroid surgery. Surgery. 2013;154(6):1346-1353.
PubMedArticle
22.
Ahuja  N, Chang  D, Gearhart  SL.  Disparities in colon cancer presentation and in-hospital mortality in Maryland: a ten-year review. Ann Surg Oncol. 2007;14(2):411-416.
PubMedArticle
23.
Doubeni  CA, Field  TS, Buist  DS,  et al.  Racial differences in tumor stage and survival for colorectal cancer in an insured population. Cancer. 2007;109(3):612-620.
PubMedArticle
24.
Singh  JA, Lu  X, Rosenthal  GE, Ibrahim  S, Cram  P.  Racial disparities in knee and hip total joint arthroplasty: an 18-year analysis of national Medicare data [published online September 18, 2013]. Ann Rheum Dis. doi:10.1136/annrheumdis-2013-203494.
PubMed
25.
Hravnak  M, Ibrahim  S, Kaufer  A, Sonel  A, Conigliaro  J.  Racial disparities in outcomes following coronary artery bypass grafting. J Cardiovasc Nurs. 2006;21(5):367-378.
PubMedArticle
26.
Yeung  M, Kerrigan  J, Sodhi  S,  et al.  Racial differences in rates of aortic valve replacement in patients with severe aortic stenosis. Am J Cardiol. 2013;112(7):991-995.
PubMedArticle
27.
Martinez  SR, Chen  SL, Bilchik  AJ.  Treatment disparities in Hispanic rectal cancer patients: a SEER database study. Am Surg. 2006;72(10):906-908.
PubMed
28.
Tammemagi  CM.  Racial/ethnic disparities in breast and gynecologic cancer treatment and outcomes. Curr Opin Obstet Gynecol. 2007;19(1):31-36.
PubMedArticle
29.
Lim  II, Hochman  T, Blumberg  SN, Patel  KN, Heller  KS, Ogilvie  JB.  Disparities in the initial presentation of differentiated thyroid cancer in a large public hospital and adjoining university teaching hospital. Thyroid. 2012;22(3):269-274.
PubMedArticle
30.
Radowsky  JS, Helou  LB, Howard  RS, Solomon  NP, Stojadinovic  A.  Racial disparities in voice outcomes after thyroid and parathyroid surgery. Surgery. 2013;153(1):103-110.
PubMedArticle
31.
Zevallos  JP, Xu  L, Yiu  Y.  The impact of socioeconomic status on the use of adjuvant radioactive iodine for papillary thyroid cancer. Thyroid. 2014;24(4):758-763.Article
32.
Harari  A, Li  N, Yeh  MW.  Racial and socioeconomic disparities in presentation and outcomes of well-differentiated thyroid cancer. J Clin Endocrinol Metab. 2014;99(1):133-141.
PubMedArticle
33.
Sosa  JA, Mehta  PJ, Wang  TS, Yeo  HL, Roman  SA.  Racial disparities in clinical and economic outcomes from thyroidectomy. Ann Surg. 2007;246(6):1083-1091.
PubMedArticle
34.
Noureldine  SI, Abbas  A, Tufano  RP,  et al.  The impact of surgical volume on racial disparity in thyroid and parathyroid surgery. Ann Surg Oncol. 2014;21(8):2733-2739.
PubMedArticle
35.
Keppel  KG, Pearcy  JN, Wagener  DK. Trends in Racial and Ethnic-Specific Rates for the Health Status Indicators: United States, 1990-98: Healthy People Statistical Notes, No. 23. Hyattsville, MD: National Center for Health Statistics; January 2002.
36.
Healthcare Cost and Utilization Project. Overview of the National (Nationwide) Inpatient Sample (NIS). Rockville, MD: Agency for Healthcare Research and Quality; 2013. http://www.hcup-us.ahrq.gov/nisoverview.jsp. Accessed February 2, 2014.
37.
Quan  H, Sundararajan  V, Halfon  P,  et al.  Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005;43(11):1130-1139.
PubMedArticle
38.
Lantz  PM, House  JS, Lepkowski  JM, Williams  DR, Mero  RP, Chen  J.  Socioeconomic factors, health behaviors, and mortality: results from a nationally representative prospective study of US adults. JAMA. 1998;279(21):1703-1708.
PubMedArticle
39.
Sorlie  PD, Backlund  E, Keller  JB.  US mortality by economic, demographic, and social characteristics: the National Longitudinal Mortality Study. Am J Public Health. 1995;85(7):949-956.
PubMedArticle
40.
Andrulis  DP.  Access to care is the centerpiece in the elimination of socioeconomic disparities in health. Ann Intern Med. 1998;129(5):412-416.
PubMedArticle
41.
Fiscella  K, Franks  P, Gold  MR, Clancy  CM.  Inequality in quality: addressing socioeconomic, racial, and ethnic disparities in health care. JAMA. 2000;283(19):2579-2584.
PubMedArticle
42.
Liu  JH, Zingmond  DS, McGory  ML,  et al.  Disparities in the utilization of high-volume hospitals for complex surgery. JAMA. 2006;296(16):1973-1980.
PubMedArticle
43.
Rothenberg  BM, Pearson  T, Zwanziger  J, Mukamel  D.  Explaining disparities in access to high-quality cardiac surgeons. Ann Thorac Surg. 2004;78(1):18-25.
PubMedArticle
44.
Epstein  AJ, Gray  BH, Schlesinger  M.  Racial and ethnic differences in the use of high-volume hospitals and surgeons. Arch Surg. 2010;145(2):179-186.
PubMedArticle
45.
Bristow  RE, Zahurak  ML, Ibeanu  OA.  Racial disparities in ovarian cancer surgical care: a population-based analysis. Gynecol Oncol. 2011;121(2):364-368.
PubMedArticle
46.
Chang  CM, Yin  WY, Wei  CK,  et al.  The association of socioeconomic status and access to low-volume service providers in breast cancer. PLoS One. 2013;8(12):e81801.
PubMedArticle
47.
Quan  H, Li  B, Saunders  LD,  et al; IMECCHI Investigators.  Assessing validity of ICD-9-CM and ICD-10 administrative data in recording clinical conditions in a unique dually coded database. Health Serv Res. 2008;43(4):1424-1441.
PubMedArticle
48.
Bell  N, Schuurman  N, Hayes  MV.  Using GIS-based methods of multicriteria analysis to construct socio-economic deprivation indices. Int J Health Geogr. 2007;6:17.
PubMedArticle
×