Association of Type of Treatment Facility With Overall Survival After a Diagnosis of Head and Neck Cancer | Head and Neck Cancer | JAMA Network Open | JAMA Network
[Skip to Navigation]
Sign In
Figure 1.  Factors Associated With Overall Survival on Multivariable Analysis
Factors Associated With Overall Survival on Multivariable Analysis

Calculated using Cox proportional hazards regression analysis. Education is calculated as the percentage of adults without a high school diploma in the patient’s zip code. Hazard ratios (HRs) less than 1.00 represent improved overall survival.

Figure 2.  Univariable Log-Rank Analyses by Facility Type for Patients Receiving Surgery and Radiotherapy
Univariable Log-Rank Analyses by Facility Type for Patients Receiving Surgery and Radiotherapy

Hazard ratios (HRs) of less than 1.00 represent improved overall survival. ACCP indicates academic comprehensive cancer program; CCCP, comprehensive community cancer program; CCP, community cancer program; and INCP, integrated network cancer program.

Figure 3.  Factors Associated With Receiving Treatment at Academic Comprehensive Cancer Programs (ACCPs) and Integrated Network Cancer Programs (INCPs)
Factors Associated With Receiving Treatment at Academic Comprehensive Cancer Programs (ACCPs) and Integrated Network Cancer Programs (INCPs)

Odds ratios (ORs) of less than 1.00 indicate lower odds of being diagnosed with head and neck cancer at ACCPs or INCPs compared with community cancer programs (CCPs) or comprehensive community cancer programs (CCCPs). Education is calculated as the percentage of adults without a high school diploma in the patient’s zip code. CDC score indicates Charlson/Deyo comorbidity score.

Table.  Demographic Information for All Participantsa
Demographic Information for All Participantsa
1.
Siegel  RL, Miller  KD, Jemal  A.  Cancer statistics, 2017.  CA Cancer J Clin. 2017;67(1):7-30. doi:10.3322/caac.21387PubMedGoogle ScholarCrossref
2.
Dimick  JB, Cowan  JA  Jr, Colletti  LM, Upchurch  GR  Jr.  Hospital teaching status and outcomes of complex surgical procedures in the United States.  Arch Surg. 2004;139(2):137-141. doi:10.1001/archsurg.139.2.137PubMedGoogle ScholarCrossref
3.
Eskander  A, Merdad  M, Irish  JC,  et al.  Volume-outcome associations in head and neck cancer treatment: a systematic review and meta-analysis.  Head Neck. 2014;36(12):1820-1834. doi:10.1002/hed.23498PubMedGoogle ScholarCrossref
4.
Lassig  AA, Joseph  AM, Lindgren  BR,  et al.  The effect of treating institution on outcomes in head and neck cancer.  Otolaryngol Head Neck Surg. 2012;147(6):1083-1092. doi:10.1177/0194599812457324PubMedGoogle ScholarCrossref
5.
Inverso  G, Mahal  BA, Aizer  AA, Donoff  RB, Chuang  SK.  Health insurance affects head and neck cancer treatment patterns and outcomes.  J Oral Maxillofac Surg. 2016;74(6):1241-1247. doi:10.1016/j.joms.2015.12.023PubMedGoogle ScholarCrossref
6.
Kubicek  GJ, Wang  F, Reddy  E, Shnayder  Y, Cabrera  CE, Girod  DA.  Importance of treatment institution in head and neck cancer radiotherapy.  Otolaryngol Head Neck Surg. 2009;141(2):172-176. doi:10.1016/j.otohns.2009.03.019PubMedGoogle ScholarCrossref
7.
Au  AG, Padwal  RS, Majumdar  SR, McAlister  FA.  Patient outcomes in teaching versus nonteaching general internal medicine services: a systematic review and meta-analysis.  Acad Med. 2014;89(3):517-523. doi:10.1097/ACM.0000000000000154PubMedGoogle ScholarCrossref
8.
Papanikolaou  PN, Christidi  GD, Ioannidis  JP.  Patient outcomes with teaching versus nonteaching healthcare: a systematic review.  PLoS Med. 2006;3(9):e341. doi:10.1371/journal.pmed.0030341PubMedGoogle Scholar
9.
Puram  SV, Bhattacharyya  N.  Quality indicators for head and neck oncologic surgery: academic versus nonacademic outcomes.  Otolaryngol Head Neck Surg. 2016;155(5):733-739. doi:10.1177/0194599816654689PubMedGoogle ScholarCrossref
10.
von Elm  E, Altman  DG, Egger  M, Pocock  SJ, Gøtzsche  PC, Vandenbroucke  JP; STROBE Initiative.  Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies.  BMJ. 2007;335(7624):806-808. doi:10.1136/bmj.39335.541782.ADPubMedGoogle ScholarCrossref
11.
Fritz  AG, Percy  C, Jack  A,  et al, eds.  International Classification of Diseases for Oncology, Third Edition. Geneva, Switzerland: World Health Organization; 2013.
12.
Carey  RM, Godovchik  J, Workman  AD,  et al.  Patient, disease, and treatment factors associated with overall survival in esthesioneuroblastoma.  Int Forum Allergy Rhinol. 2017;7(12):1186-1194. doi:10.1002/alr.22027PubMedGoogle ScholarCrossref
13.
Carey  RM, Parasher  AK, Workman  AD,  et al.  Disparities in sinonasal squamous cell carcinoma short- and long-term outcomes: analysis from the national cancer database.  Laryngoscope. 2018;128(3):560-567. doi:10.1002/lary.26804PubMedGoogle ScholarCrossref
14.
Rubin  SJ, Cohen  MB, Kirke  DN, Qureshi  MM, Truong  MT, Jalisi  S.  Comparison of facility type outcomes for oral cavity cancer: analysis of the national cancer database.  Laryngoscope. 2017;127(11):2551-2557. doi:10.1002/lary.26632PubMedGoogle ScholarCrossref
15.
D’Cruz  AK, Vaish  R, Kapre  N,  et al; Head and Neck Disease Management Group.  Elective versus therapeutic neck dissection in node-negative oral cancer.  N Engl J Med. 2015;373(6):521-529. doi:10.1056/NEJMoa1506007PubMedGoogle ScholarCrossref
16.
Rogers  SN, Brown  JS, Woolgar  JA,  et al.  Survival following primary surgery for oral cancer.  Oral Oncol. 2009;45(3):201-211. doi:10.1016/j.oraloncology.2008.05.008PubMedGoogle ScholarCrossref
17.
Woolgar  JA.  Histopathological prognosticators in oral and oropharyngeal squamous cell carcinoma.  Oral Oncol. 2006;42(3):229-239. doi:10.1016/j.oraloncology.2005.05.008PubMedGoogle ScholarCrossref
18.
Khan  MN, Konuthula  N, Parasher  A,  et al.  Treatment modalities in sinonasal undifferentiated carcinoma: an analysis from the National Cancer Database.  Int Forum Allergy Rhinol. 2017;7(2):205-210. doi:10.1002/alr.21861PubMedGoogle ScholarCrossref
19.
Luryi  AL, Chen  MM, Mehra  S, Roman  SA, Sosa  JA, Judson  BL.  Positive surgical margins in early stage oral cavity cancer: an analysis of 20,602 cases.  Otolaryngol Head Neck Surg. 2014;151(6):984-990. doi:10.1177/0194599814551718PubMedGoogle ScholarCrossref
20.
Göllnitz  I, Inhestern  J, Wendt  TG,  et al.  Role of comorbidity on outcome of head and neck cancer: a population-based study in Thuringia, Germany.  Cancer Med. 2016;5(11):3260-3271. doi:10.1002/cam4.882PubMedGoogle ScholarCrossref
21.
Brookhart  MA, Stürmer  T, Glynn  RJ, Rassen  J, Schneeweiss  S.  Confounding control in healthcare database research: challenges and potential approaches.  Med Care. 2010;48(6)(suppl):S114-S120. doi:10.1097/MLR.0b013e3181dbebe3PubMedGoogle ScholarCrossref
22.
Ward  E, Jemal  A, Cokkinides  V,  et al.  Cancer disparities by race/ethnicity and socioeconomic status.  CA Cancer J Clin. 2004;54(2):78-93. doi:10.3322/canjclin.54.2.78PubMedGoogle ScholarCrossref
23.
Binazzi  A, Ferrante  P, Marinaccio  A.  Occupational exposure and sinonasal cancer: a systematic review and meta-analysis.  BMC Cancer. 2015;15:49. doi:10.1186/s12885-015-1042-2PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    1 Comment for this article
    EXPAND ALL
    Response to statement regarding accuracy of NCDB data
    Katherine Mallin, PhD | National Cancer Database, Commission on Cancer, American College of Surgeons
    I read with interest the article by Carey, RM et al, “Association of type of treatment facility with overall survival after a diagnosis of Head and Neck Cancer.”(1) This study used data from the Participant User Files (PUF) of the Commission on Cancer (CoC) National Cancer Database (NCDB). In the Limitations section of the paper, the authors stated, “the NCDB has potential issues with accuracy and confounding. The data are gathered from multiple centers, each with their own standards for data collection and reporting.” I would like to correct this latter statement.

    All CoC accredited facilities collect and submit data to the NCDB following the rules and instructions in the Standards for Oncology Registry Entry (STORE) manual for cases diagnosed in 2018 and later (2), and prior diagnosis years complied with definitions in prior manuals.(3) All data is evaluated through data validation (edits) process before submission to the NCDB. Any record that does not meet specified data quality criteria must be corrected or is rejected. In addition, the CoC Standards for hospital accreditation include requirements for Cancer Registry Education, Cancer Registrar Credentials, Data Submission, and Accuracy of Data.(4)
    Providing accurate and standardized data items to the NCDB is key to the success of the Quality Programs of the CoC. The CoC continues to monitor the data accuracy and completeness of the data submitted to the NCDB in order to improve the quality of cancer care.

    1. Carey RM, Fathy R, Shah RR, et al. Association of type of treatment facility with overall survival after a diagnosis of Head and Neck Cancer.” JAMA Network Open 2020;3(1):e1919697.doi.10.1001/jamnetworkopen.2019.19697
    2. Standards for Oncology Registry Entry. STORE 2018. https://www.facs.org/-/media/files/quality-programs/cancer/ncdb/store_manual_2018.ashx Accessed 1/29/2020
    3. Facility Oncology Registry Data Standards (FORDS): Revised for 2016. Facs.org/quality-programs/cancer/ncdb/call-for-data/fordsmanual accessed 1/29/2020
    4. Commission on Cancer. A Quality Program of the American College of Surgeons. 2020 Standards and Resources. https://www.facs.org/quality-programs/cancer/coc/standards/2020 Accessed 1/31/2020.
    CONFLICT OF INTEREST: None Reported
    READ MORE
    Original Investigation
    Oncology
    January 24, 2020

    Association of Type of Treatment Facility With Overall Survival After a Diagnosis of Head and Neck Cancer

    Author Affiliations
    • 1Department of Otorhinolaryngology–Head and Neck Surgery, Perelman School of Medicine, University of Pennsylvania, Philadelphia
    • 2currently a medical student at Perelman School of Medicine, University of Pennsylvania, Philadelphia
    • 3Department of Healthcare Policy and Research, Weill Cornell Medicine, New York, New York
    JAMA Netw Open. 2020;3(1):e1919697. doi:10.1001/jamanetworkopen.2019.19697
    Key Points español 中文 (chinese)

    Question  Is the type of treatment facility associated with overall survival in patients with head and neck cancer?

    Findings  In this cohort study of 525 740 patients diagnosed with malignant neoplasms of the head and neck in the National Cancer Database from 2004 to 2016, treatment at academic comprehensive cancer programs, integrated network cancer programs, and comprehensive community cancer programs was associated with better overall survival rates than treatment at community cancer programs.

    Meaning  This study’s findings suggest that improved understanding of socioeconomic differences and oncologic treatment disparities may improve clinical outcomes in head and neck cancer.

    Abstract

    Importance  Patients with head and neck cancer receive care at academic comprehensive cancer programs (ACCPs), integrated network cancer programs (INCPs), comprehensive community cancer programs (CCCPs), and community cancer programs (CCPs). The type of treatment facility may be associated with overall survival.

    Objective  To examine whether type of treatment facility is associated with overall survival after a diagnosis of head and neck cancer.

    Design, Setting, and Participants  This population-based retrospective cohort study included patients from the National Cancer Database, a prospectively maintained, hospital-based cancer registry of patients treated at more than 1500 US hospitals. Participants were diagnosed with malignant tumors of the head and neck from January 1, 2004, through December 31, 2016. Data were analyzed from May 1 through November 30, 2019.

    Exposures  Treatment at facilities classified as ACCPs, INCPs, CCCPs, or CCPs.

    Main Outcomes and Measures  Overall survival after diagnosis and treatment of head and neck cancer was the primary outcome. The secondary outcome was the odds of receiving treatment at ACCPs and INCPs vs CCCPs and CCPs. Multivariable Cox proportional hazards regression and univariable and multivariable logistic regression models were used for analysis.

    Results  A total of 525 740 patients (368 821 men [70.2%]; mean [SD] age, 63.3 [14.0] years) were diagnosed with malignant tumors of the head and neck during the study period. Among them, 36 595 patients (7.0%) were treated at CCPs; 174 658 (33.2%), at CCCPs; 232 867 (44.3%), at ACCPs; and 57 857 (11.0%), at INCPs. The median survival for patients with aerodigestive cancers was 69.2 (95% CI, 68.6-69.8) months; salivary gland cancers, 107.2 (95% CI, 103.9-110.2) months; and skin cancers, 113.2 (95% CI, 111.4-114.6) months. Improved overall survival was associated with treatment at ACCPs (hazard ratio [HR], 0.89; 95% CI, 0.88-0.91), INCPs (HR, 0.94; 95% CI, 0.92-0.96), and CCCPs (HR, 0.94; 95% CI, 0.92-0.95) compared with CCPs. Compared with patients with private insurance, those with government insurance (odds ratio [OR], 1.35; 95% CI, 1.29-1.41), no insurance (OR, 1.12; 95% CI, 1.09-1.16), or Medicaid (OR, 1.17; 95% CI, 1.14-1.20) were more likely to receive treatment at ACCPs and INCPs, whereas patients with Medicare were less likely to receive treatment at ACCPs and INCPs (OR, 0.95; 95% CI, 0.94-0.97). Compared with white patients, black (OR, 1.55; 95% CI, 1.52-1.59) and Asian (OR, 1.56; 95% CI, 1.49-1.63) patients were more likely to receive care at ACCPs and INCPs. Compared with patients from lower-income areas, patients from high-income areas were more likely to receive treatment at ACCPs and INCPs (OR, 1.25; 95% CI, 1.22-1.28).

    Conclusions and Relevance  These findings suggest that treatment at ACCPs and INCPs was associated with a better overall survival rate in patients with head and neck cancer. Key social determinants of health such as race/ethnicity, socioeconomic status, and type of insurance were associated with receiving treatment at ACCPs and INCPs.

    Introduction

    Malignant tumors of the head and neck account for approximately 4% of all cancers in the United States and involve collaboration between multiple specialties for optimal treatment.1 Owing to the highly specialized management of head and neck cancer, treatment factors such as hospital volume and teaching status are often thought to contribute to variation in patient outcomes.2-4 Physician and institutional variables such as technical ability, treatment modalities offered, and multidisciplinary support and patient factors such as ethnicity, socioeconomic status, and insurance status can vary widely between types of institutions, all of which may affect outcomes.2-6 Despite these disparities, no consensus exists on the differences between the quality of care delivered at teaching and nonteaching hospitals.7,8

    A study from Puram and Bhattacharyya9 using data from the Nationwide Inpatient Sample analyzed quality metrics for patients undergoing surgery for head and neck cancer from academic and nonacademic institutions. Their study found that academic institutions had a greater proportion of patients with a history of radiotherapy, high-acuity procedures, and greater comorbidity scores. Controlling for these variables and others showed a slightly increased length of stay and wound infection rates at academic hospitals. A separate study from the Nationwide Inpatient Sample by Dimick et al2 found that teaching hospitals had lower operative mortality rates for complex surgical procedures (esophageal, hepatic, or pancreatic resections); however, they found no difference in operative mortality after controlling for hospital volume on multivariate analysis. Eskander et al3 conducted a meta-analysis comparing outcomes in head and neck cancer at high- and low-volume centers. They demonstrated better overall survival among patients treated by high-volume hospitals and surgeons than among patients treated by low-volume hospitals and surgeons. However, to our knowledge, no studies have broadened the comparison of treatment facility type and outcomes beyond teaching vs nonteaching facilities.

    Therefore, the primary objective of this study was to use data from the National Cancer Database (NCDB) to evaluate factors that contribute to overall survival in patients with head and neck cancers by comparing outcomes for academic comprehensive cancer programs (ACCPs), integrated network cancer programs (INCPs), comprehensive community cancer programs (CCCPs), and community cancer programs (CCPs). Academic comprehensive cancer programs were facilities that participated in postgraduate medical education in at least 4 program areas (including general surgery and internal medicine) and had more than 500 newly diagnosed cancer cases each year. Integrated network cancer programs were defined by having a unified cancer committee with coordinated practice locations and health care professionals, and training of resident physicians was optional. Comprehensive CCPs and CCPs were facilities where training of resident physicians was optional and included more than 500 and 100 to 500 newly diagnosed cancer cases each year, respectively. Furthermore, we investigated demographic and socioeconomic factors for their association with the type of facility where treatment was administered.

    Methods
    Study Sample

    The NCDB data set is a joint project of the Commission on Cancer of the American College of Surgeons and the American Cancer Society. The data used in this study are derived from a deidentified NCDB file. This study was determined to be exempt by the University of Pennsylvania institutional review board and did not require informed consent for the use of deidentified data. This report follows the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.10

    Data were obtained from the NCDB from January 1, 2004, through December 31, 2016, and analyzed from May 1 through November 30, 2019. The NCDB uses the International Classification of Diseases for Oncology, Third Edition,11 which is similar but not identical to the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision. The NCDB was queried for primary site codes in the head and neck, excluding thyroid (eTable in the Supplement). Site codes were further divided into aerodigestive, salivary gland, and skin. Only cases designated as “malignant neoplasms stated or presumed to be primary” (behavior code of 3) were included, and no histology type was excluded. To avoid confounding of different surgical procedures and ensure that the surgical procedures were on the primary site, cases were excluded for surgery at a distant site. Cases were also excluded if diagnosis and treatment were performed at different facilities or if they were missing data on mortality status at follow-up.

    Primary and Secondary Outcome Variables

    The primary outcome of interest was overall survival. Demographic and socioeconomic factors contributing to receiving treatment at ACs and INCPs vs CCCPs and CCPs served as the secondary outcome, calculated as odds ratios (ORs). Descriptive statistics were calculated.

    Statistical Analysis

    Factors associated with overall survival were evaluated using multivariable Cox proportional hazards regression models. Variables to include in the models were selected a priori based on consensus of the contributing authors and included age, sex, race/ethnicity, educational level, median income, housing area (ie, urban vs rural), geographic area, insurance type, facility location, facility type, Charlson/Deyo comorbidity score, and year of diagnosis. Definitions of the variables have been described in prior studies.12,13 Housing areas were defined based on the size of each facility’s county (metropolitan, >250 000 residents; urban, 2500-250 000 residents; and rural, <2500 residents). The population educational level was determined from the 2012 American Community Survey based on the percentage of adults in the patient’s zip code who did not have a high school diploma. The primary insurance provider at the time of cancer diagnosis was used for determining insurance status.

    Additional analyses of overall survival by facility type for subsets of patients receiving surgical therapy or radiotherapy were also performed. To determine associations of demographic and socioeconomic factors with type of facility, multivariable logistic models with the same variables described above were used to compare the combined group of ACCPs and INCPs with the combined group of CCPs and CCCPs. Statistical analyses were performed with R, version 3.4.1 statistical software (R Project for Statistical Computing), via RStudio, version 1.1.23 statistical software (RStudio, Inc). Missing data were removed from survival and logistic models. Two-sided P < .05 indicated statistical significance.

    Results
    Baseline Characteristics of the Sample

    A total of 581 726 patients met facility and behavior criteria with complete data for follow-up. Of these, 28 470 were excluded for surgery at a distant site, and 27 516 were excluded for not receiving treatment at the facility where they were diagnosed. After exclusions, 525 740 participants (368 821 men [70.2%] and 156 919 women [29.8%]; mean [SD] age, 63.3 [14.0] years) were included in the final analysis. Among these, 389 495 patients (74.1%) had aerodigestive cancers; 36 700 (7.0%), cancers of the salivary gland; and 99 545 (18.9%), skin cancers. Most skin cancers had aggressive histologic subtypes, including approximately 80% melanoma subtypes and 6% Merkel cell carcinoma. The median survival for patients with aerodigestive cancers was 69.2 (95% CI, 68.6-69.8) months; salivary gland cancers, 107.2 (95% CI, 103.9-110.2) months; and skin cancers, 113.2 (95% CI, 111.4-114.6) months. The Table lists the various demographic variables included in the study. During the study period, 36 595 patients (7.0%) were treated at CCPs; 174 658 (33.2%), at CCCPs; 232 867 (44.3%), at ACCPs; and 57 857 (11.0%), at INCPs.

    Results From Multivariable Models

    Several variables were significantly associated with overall survival on multivariable analysis. Specifically, treatment at ACCPs (hazard ratio [HR], 0.89; 95% CI, 0.88-0.91), INCPs (HR, 0.94; 95% CI, 0.92-0.96), and CCCPs (HR, 0.94; 95% CI, 0.92-0.95) were associated with improved overall survival on multivariable analysis compared with CCPs (Figure 1 and Table). Results of univariable analyses by facility type for subsets of patients receiving surgical therapy with or without radiotherapy and for patients receiving radiotherapy with or without surgery are shown in Figure 2. Multivariable subanalysis for patients receiving surgical therapy did not demonstrate statistically significant differences in overall survival at different facility types. Multivariable analysis for patients receiving radiotherapy demonstrated improved overall survival at ACCPs (HR, 0.95; 95% CI, 0.93-0.97), INCPs (HR, 0.95; 95% CI, 0.93-0.98), and CCCPs (HR, 0.96; 95% CI, 0.94-0.98).

    Multiple factors were associated with receiving care at ACCPs and INCPs (Figure 3 and Table). Compared with private insurance, having Medicaid (OR, 1.17; 95% CI, 1.14-1.20), no insurance (OR, 1.12; 95% CI, 1.09-1.16), and other government insurance (OR, 1.35; 95% CI, 1.29-1.41) were associated with greater odds of receiving treatment at ACCPs and INCPs, whereas having Medicare was associated with decreased odds (OR, 0.95; 95% CI, 0.94-0.97). Black patients (OR, 1.55; 95% CI, 1.52-1.59) and Asian patients (OR, 1.56; 95% CI, 1.49-1.63) were more likely to receive care at ACCPs and INCPs compared with white patients. Compared with the lowest income bracket, patients with an annual income of $63 000 or greater were more likely to receive treatment at ACCPs and INCPs (OR, 1.25; 95% CI, 1.22-1.28). Patients with the lowest education levels were more likely to receive treatment at ACCPs and INCPs (OR, 1.05; 95% CI, 1.03-1.08). Patients from urban (OR, 0.64; 95% CI, 0.63-0.65) and rural (OR, 0.53; 95% CI, 0.51-0.55) communities were less likely to receive care at ACCPs and INCPs compared with individuals from metropolitan areas.

    There was a small but consistent and statistically significant trend toward increased diagnosis and treatment at ACCPs and INCPs compared with CCCPs and CCPs over time that appeared to be contributed to primarily by findings for increased diagnosis and treatment at ACCPs. Further analysis of this trend was beyond the scope of the current report.

    Discussion

    In this large national database analysis, we found that patients with head and neck cancers who were diagnosed and treated at ACCPs, INCPs, and CCCPs had better overall survival rates compared with patients who received treatment at CCPs. This difference was also seen in the subset of patients receiving radiotherapy as part of their treatment. Specifically, patients who received radiotherapy at ACCPs, INCPs, and CCCPs had improved overall survival compared with those receiving radiotherapy at CCPs. The second key finding of this study was the association between social determinants of health, such as race/ethnicity, income, educational level, and community type, and where one receives treatment.

    Univariate analysis of facility type for the subset of patients receiving surgical therapy demonstrated improved overall survival at ACCPs, INCPs, and CCCPs compared with CCPs; however, no significant survival difference was found on multivariable analysis. A previous NCDB analysis of oral cavity cancer by Rubin et al14 found that patients treated at ACCPs were more likely to receive surgical treatment and had improved overall survival compared with patients treated at CCPs and CCCPs. Furthermore, those investigators found that patients treated at ACCPs were more likely to undergo neck dissection even after controlling for tumor stage. Based on the work by D’Cruz et al,15 Rubin et al14 hypothesized that the improved overall survival seen in the ACCP group may have been at least partially affected by the survival benefits from elective neck dissection compared with therapeutic neck dissection in early-stage oral cavity cancer.

    Obtaining negative surgical margins is another important principle of head and neck oncologic surgery because it affects the overall prognosis in many cancers.16-18 The NCDB analysis on oral cavity cancer by Luryi et al19 found that positive surgical margins were associated with treatment at nonacademic cancer centers and institutions with lower oral cancer case volume. The present study did not investigate associations between margin status and treatment facility, although these factors may have affected survival.

    Treatment of head and neck cancer often requires multiple specialists and significant resources, including facilities capable of delivering radiotherapy. One retrospective study of patients with head and neck cancer diagnosed at a single academic center6 found that patients who received their radiotherapy at nonacademic centers were more likely to have earlier-stage cancer and to receive radiotherapy alone instead of chemoradiotherapy. However, there was no difference in recurrence rates or overall survival between the academic and nonacademic treatment groups.6

    Another study4 evaluated 388 patients with mucosal head and neck cancer treated with primary and adjuvant radiotherapy at academic or community centers and found that patients treated at ACCPs had more advanced disease, decreased rates of smoking, a higher median income, and a higher percentage of oropharyngeal tumors. Of note, the 5-year survival rates were higher in patients treated at ACCPs compared with community centers (53.2% vs 32.8%; P < .001).4 These investigators found no differences in the rate of treatment completion between academic and community centers.

    Many studies have compared teaching with nonteaching hospitals,2,3,7-9 but few have specifically compared ACCPs, INCPs, CCCPs, and CCPs. This distinction is important but difficult to fully interpret because there may be confounding between resident teaching status and hospital case volume. Because resident training at INCPs, CCCPs, and CCPs was optional, it is unclear what proportion of facilities participated in resident training. Both CCCPs and ACCPs were higher-volume facilities compared with CCPs, whereas the number of newly diagnosed cancer cases at INCPs was not explicitly defined in the NCDB.

    During the present study, 44.3% of patients were treated at ACCPs and 11.0% at INCPs compared with 33.2% treated at CCPS and 7.0% at CCPs. This study’s findings indicate that black and Asian patients, patients from metropolitan areas, and patients in the lowest quartile of educational level were more likely to be diagnosed at ACCPs and INCPs. These findings may reflect a proximity bias and could be related to the demographic characteristics of individuals who most often live in areas where ACCPs and INCPs tend to be located.

    Our multivariable analysis showed worse overall survival for patients with Medicaid, Medicare, no insurance, and other government insurance compared with private insurance. These findings are supported by a study of head and neck cancer by Inverso et al,5 who showed that uninsured patients were more likely to present with metastatic disease and had a higher risk of head and neck cancer–specific mortality. We found that, compared with having private insurance, having Medicaid, no insurance, or other government insurance were associated with greater odds of receiving treatment at ACCPs and INCPs, whereas having Medicare was associated with decreased odds. These findings may again reflect a proximity bias of certain facilities or may be owing to a greater willingness of ACCPs and INCPs to treat patients regardless of their insurance status.

    Patients in the highest income bracket were more likely to be diagnosed and treated at ACCPs and INCPs. Patients with higher incomes may have had greater means to seek out care at larger tertiary research centers; however, the present analysis could not make that determination. A higher Charlson/Deyo comorbidity score has been shown to be a strong risk factor for poor overall survival in head and neck cancer,20 which was again demonstrated in this study.

    These findings suggest improved outcomes for patients with head and neck cancer who receive their treatment at teaching institutions and/or higher-volume facilities. However, socioeconomic and health disparities affect where patients ultimately receive their treatment. Improved access to care for patients from lower socioeconomic status may ultimately help improve these individuals’ outcomes.

    Limitations

    Our results must be interpreted within consideration of several limitations. First, the NCDB has potential issues with accuracy and confounding. The data are gathered from multiple centers, each with their own standards for data collection and reporting. Confounding is a known issue with data sets such as the NCDB owing to missing clinically relevant variables that cannot be included in analyses.21 For example, the NCDB does not include information on tobacco smoking, which is more common in lower socioeconomic classes22 and has an important association with head and neck cancer.23 Moreover, our multivariable analyses only controlled for the variables that we incorporated into the statistical models. We attempted to control for advances in the treatment of head and neck cancer during the study period by including the year of diagnosis in the multivariable models. However, we were not able to account for differences in treatment techniques such as transoral robotic surgery or intensity-modulated radiotherapy. Also, the primary outcome of interest, overall survival, is prone to confounding due to various patient and disease factors.

    Conclusions

    This study’s findings suggest that social factors such as race/ethnicity, income, educational level, and community type were associated with where patients received treatment. Where patients received treatment was associated with their outcomes because patients with head and neck cancers who received treatment at ACCPs, INCPs, and CCCPs had better overall survival compared with those who received treatment at CCPs. Future studies are necessary to improve our understanding of these socioeconomic differences, reduce the disparities that exist in oncologic treatment, and improve overall outcomes.

    Back to top
    Article Information

    Accepted for Publication: November 25, 2019.

    Published: January 24, 2020. doi:10.1001/jamanetworkopen.2019.19697

    Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Carey RM et al. JAMA Network Open.

    Corresponding Author: Ryan M. Carey, MD, Department of Otorhinolaryngology–Head and Neck Surgery, Perelman School of Medicine, University of Pennsylvania, 3400 Spruce St, Ravdin Building, Fifth Floor, Philadelphia, PA 19104 (ryan.carey@uphs.upenn.edu).

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

    Concept and design: Carey, Shah, Rajasekaran, Cannady, Newman, Brant.

    Acquisition, analysis, or interpretation of data: Carey, Fathy, Shah, Cannady, Newman, Ibrahim, Brant.

    Drafting of the manuscript: Carey, Shah, Ibrahim, Brant.

    Critical revision of the manuscript for important intellectual content: Carey, Fathy, Shah, Rajasekaran, Cannady, Newman, Brant.

    Statistical analysis: Fathy, Shah, Brant.

    Administrative, technical, or material support: Cannady, Newman, Brant.

    Supervision: Rajasekaran, Cannady, Newman, Ibrahim, Brant.

    Conflict of Interest Disclosures: Dr Newman reported being a paid consultant to Medtronic and Boulder Surgical and receiving personal fees as a member of the medical board for Castle Biosciences, Inc, all outside the submitted work. Dr Brant reported receiving compensation for serving on a surgical advisory board for MED-EL Corporation and a medical advisory board for for TympoBio, LLC and receiving research funding from the Department of Veterans Affairs and cochlear implant companies outside the submitted work. No other disclosures were reported.

    Disclaimer: The American College of Surgeons and the Commission on Cancer have not verified and are not responsible for the analytic or statistical methods used or the conclusions drawn from these data by the investigators.

    References
    1.
    Siegel  RL, Miller  KD, Jemal  A.  Cancer statistics, 2017.  CA Cancer J Clin. 2017;67(1):7-30. doi:10.3322/caac.21387PubMedGoogle ScholarCrossref
    2.
    Dimick  JB, Cowan  JA  Jr, Colletti  LM, Upchurch  GR  Jr.  Hospital teaching status and outcomes of complex surgical procedures in the United States.  Arch Surg. 2004;139(2):137-141. doi:10.1001/archsurg.139.2.137PubMedGoogle ScholarCrossref
    3.
    Eskander  A, Merdad  M, Irish  JC,  et al.  Volume-outcome associations in head and neck cancer treatment: a systematic review and meta-analysis.  Head Neck. 2014;36(12):1820-1834. doi:10.1002/hed.23498PubMedGoogle ScholarCrossref
    4.
    Lassig  AA, Joseph  AM, Lindgren  BR,  et al.  The effect of treating institution on outcomes in head and neck cancer.  Otolaryngol Head Neck Surg. 2012;147(6):1083-1092. doi:10.1177/0194599812457324PubMedGoogle ScholarCrossref
    5.
    Inverso  G, Mahal  BA, Aizer  AA, Donoff  RB, Chuang  SK.  Health insurance affects head and neck cancer treatment patterns and outcomes.  J Oral Maxillofac Surg. 2016;74(6):1241-1247. doi:10.1016/j.joms.2015.12.023PubMedGoogle ScholarCrossref
    6.
    Kubicek  GJ, Wang  F, Reddy  E, Shnayder  Y, Cabrera  CE, Girod  DA.  Importance of treatment institution in head and neck cancer radiotherapy.  Otolaryngol Head Neck Surg. 2009;141(2):172-176. doi:10.1016/j.otohns.2009.03.019PubMedGoogle ScholarCrossref
    7.
    Au  AG, Padwal  RS, Majumdar  SR, McAlister  FA.  Patient outcomes in teaching versus nonteaching general internal medicine services: a systematic review and meta-analysis.  Acad Med. 2014;89(3):517-523. doi:10.1097/ACM.0000000000000154PubMedGoogle ScholarCrossref
    8.
    Papanikolaou  PN, Christidi  GD, Ioannidis  JP.  Patient outcomes with teaching versus nonteaching healthcare: a systematic review.  PLoS Med. 2006;3(9):e341. doi:10.1371/journal.pmed.0030341PubMedGoogle Scholar
    9.
    Puram  SV, Bhattacharyya  N.  Quality indicators for head and neck oncologic surgery: academic versus nonacademic outcomes.  Otolaryngol Head Neck Surg. 2016;155(5):733-739. doi:10.1177/0194599816654689PubMedGoogle ScholarCrossref
    10.
    von Elm  E, Altman  DG, Egger  M, Pocock  SJ, Gøtzsche  PC, Vandenbroucke  JP; STROBE Initiative.  Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies.  BMJ. 2007;335(7624):806-808. doi:10.1136/bmj.39335.541782.ADPubMedGoogle ScholarCrossref
    11.
    Fritz  AG, Percy  C, Jack  A,  et al, eds.  International Classification of Diseases for Oncology, Third Edition. Geneva, Switzerland: World Health Organization; 2013.
    12.
    Carey  RM, Godovchik  J, Workman  AD,  et al.  Patient, disease, and treatment factors associated with overall survival in esthesioneuroblastoma.  Int Forum Allergy Rhinol. 2017;7(12):1186-1194. doi:10.1002/alr.22027PubMedGoogle ScholarCrossref
    13.
    Carey  RM, Parasher  AK, Workman  AD,  et al.  Disparities in sinonasal squamous cell carcinoma short- and long-term outcomes: analysis from the national cancer database.  Laryngoscope. 2018;128(3):560-567. doi:10.1002/lary.26804PubMedGoogle ScholarCrossref
    14.
    Rubin  SJ, Cohen  MB, Kirke  DN, Qureshi  MM, Truong  MT, Jalisi  S.  Comparison of facility type outcomes for oral cavity cancer: analysis of the national cancer database.  Laryngoscope. 2017;127(11):2551-2557. doi:10.1002/lary.26632PubMedGoogle ScholarCrossref
    15.
    D’Cruz  AK, Vaish  R, Kapre  N,  et al; Head and Neck Disease Management Group.  Elective versus therapeutic neck dissection in node-negative oral cancer.  N Engl J Med. 2015;373(6):521-529. doi:10.1056/NEJMoa1506007PubMedGoogle ScholarCrossref
    16.
    Rogers  SN, Brown  JS, Woolgar  JA,  et al.  Survival following primary surgery for oral cancer.  Oral Oncol. 2009;45(3):201-211. doi:10.1016/j.oraloncology.2008.05.008PubMedGoogle ScholarCrossref
    17.
    Woolgar  JA.  Histopathological prognosticators in oral and oropharyngeal squamous cell carcinoma.  Oral Oncol. 2006;42(3):229-239. doi:10.1016/j.oraloncology.2005.05.008PubMedGoogle ScholarCrossref
    18.
    Khan  MN, Konuthula  N, Parasher  A,  et al.  Treatment modalities in sinonasal undifferentiated carcinoma: an analysis from the National Cancer Database.  Int Forum Allergy Rhinol. 2017;7(2):205-210. doi:10.1002/alr.21861PubMedGoogle ScholarCrossref
    19.
    Luryi  AL, Chen  MM, Mehra  S, Roman  SA, Sosa  JA, Judson  BL.  Positive surgical margins in early stage oral cavity cancer: an analysis of 20,602 cases.  Otolaryngol Head Neck Surg. 2014;151(6):984-990. doi:10.1177/0194599814551718PubMedGoogle ScholarCrossref
    20.
    Göllnitz  I, Inhestern  J, Wendt  TG,  et al.  Role of comorbidity on outcome of head and neck cancer: a population-based study in Thuringia, Germany.  Cancer Med. 2016;5(11):3260-3271. doi:10.1002/cam4.882PubMedGoogle ScholarCrossref
    21.
    Brookhart  MA, Stürmer  T, Glynn  RJ, Rassen  J, Schneeweiss  S.  Confounding control in healthcare database research: challenges and potential approaches.  Med Care. 2010;48(6)(suppl):S114-S120. doi:10.1097/MLR.0b013e3181dbebe3PubMedGoogle ScholarCrossref
    22.
    Ward  E, Jemal  A, Cokkinides  V,  et al.  Cancer disparities by race/ethnicity and socioeconomic status.  CA Cancer J Clin. 2004;54(2):78-93. doi:10.3322/canjclin.54.2.78PubMedGoogle ScholarCrossref
    23.
    Binazzi  A, Ferrante  P, Marinaccio  A.  Occupational exposure and sinonasal cancer: a systematic review and meta-analysis.  BMC Cancer. 2015;15:49. doi:10.1186/s12885-015-1042-2PubMedGoogle ScholarCrossref
    ×