- Email: [email protected]
Impact of African–American race on presentation, treatment, and survival of head and neck cancer
Oral Oncology 50 (2014) 1177–1181
Contents lists available at ScienceDirect
Oral Oncology journal homepage: www.elsevier.com/locate/oraloncology
Impact of African–American race on presentation, treatment, and survival of head and neck cancer Brandon A. Mahal a,1, Gino Inverso b,1, Ayal A. Aizer c, R. Bruce Donoff b,d, Sung-Kiang Chuang b,d,⇑ a
Harvard Medical School, 25 Shattuck St., Boston, MA 02115, USA Harvard School of Dental Medicine, 188 Longwood Ave., Boston, MA 02115, USA c Harvard Radiation Oncology Program, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114, USA d Department of Oral and Maxillofacial Surgery, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114, USA b
a r t i c l e
i n f o
Article history: Received 18 May 2014 Received in revised form 18 July 2014 Accepted 6 September 2014 Available online 26 September 2014 Keywords: Head and neck cancer Healthcare disparities African Americans Minority health Head and neck cancer specific survival
s u m m a r y Objectives: To determine the associations between African American race and stage at diagnosis, receipt of definitive therapy, and cancer-specific mortality among patients with head and neck cancer. Materials and methods: The Surveillance, Epidemiology and End Results (SEER) database was used to conduct a retrospective study on 34,437 patients diagnosed with head and neck cancer from 2007 to 2010. Multivariable logistic regression analyses were applied to determine the impact of race on cancer stage at presentation (metastatic vs. non-metastatic) and receipt of definitive treatment. Fine and Gray competing-risks regression modeled the association between race and head and neck cancer-specific mortality. Results: African Americans were more likely to present with metastatic cancer compared to non-African Americans (Adjusted Odds Ratio [AOR] 1.76; CI 1.50–2.07; P < 0.001). Among patients with nonmetastatic disease, African Americans were less likely to receive definitive treatment (AOR 0.63; CI 0.55–0.72; P < 0.001). After a median follow-up of 19 months, African Americans with non-metastatic disease were found to have a higher risk of head and neck cancer specific mortality (AHR 1.19; 95% CI 1.09–1.29; P < 0.001). Conclusion: African Americans with head and neck cancer are more likely to present with metastatic disease, less likely to be treated definitively, and are more likely to die from head and neck cancer. The unacceptably high rates of disparity found in this study should serve as immediate targets for urgent healthcare policy intervention. Ó 2014 Elsevier Ltd. All rights reserved.
Introduction There are an estimated 42,440 new cases of oral and pharyngeal cancer diagnosed each year and 8390 deaths annually [1]. The 5-year survival rate for all patients diagnosed with oral and pharyngeal cancer is 64% [1]. When stratified by race, the 5-year survival rate for African Americans drops to 43%. Although there are currently no practical head and neck cancer (HNC) screening or detection strategies in place, cure rates approximate 50% and can reach 90% with early stage detection [2,3]. Still most patients present with advanced stage disease [1]. Previous studies in HNC outcomes research have demonstrated that African American patients face an unequal burden of disease ⇑ Corresponding author at: Harvard School of Dental Medicine, 188 Longwood Ave, Boston, MA 02115, USA. Tel.: +1 617 432 1434; fax: +1 617 432 1897. E-mail address: [email protected] (S.-K. Chuang). 1 Authors had equal 1st author contribution. http://dx.doi.org/10.1016/j.oraloncology.2014.09.004 1368-8375/Ó 2014 Elsevier Ltd. All rights reserved.
due to differences in socioeconomic status, limited access to care, advanced cancer stage, lower receipt of treatment, and comorbidity [4–7]. Nevertheless, the existing literature focusing on racial disparities in HNC outcomes have not been able to account for relevant sociodemographic factors such as insurance, which has been suggested to have a significant impact on cancer outcomes and treatment patterns [2,8,9]. Additionally, the majority of existing studies do not utilize cancer specific mortality as an endpoint and are unable to account for comorbid disease that has been shown to disproportionally contribute to the mortality of African American patients with HNC [7]. Lastly, previous studies focusing on racial disparities in HNC outcomes have examined smaller cohorts restricted to few sites, thus limiting the number of pertinent covariates that could be adjusted for and generalized conclusions that could be made about disparities in HNC [8–11]. The Surveillance, Epidemiology and End Results Program (SEER) program, sponsored by the National Cancer Institute, collects incidence, prevalence, survival and mortality on cancer [12]. The SEER
1178
B.A. Mahal et al. / Oral Oncology 50 (2014) 1177–1181
program captures approximately 97% of incident cancers and the 17 tumor registries encompass approximately 26% of the US population. Additionally, the database provides valuable demographic information, such as county level income level and educational status, residence type, and as of 2007, individual level data on insurance status. The main objective of this study is to utilize the SEER database to examine the associations between African American race and stage at diagnosis, receipt of definitive therapy, and cancer-specific mortality among patients with HNC utilizing robust multivariable models adjusting for the potentially pertinent sociodemographic covariates mentioned above. Material and methods Study population The SEER database was utilized to identify 34,437 patients (3820 African American and 30,617 non-African American) diagnosed with HNC from 2007 to 2010. The inclusion period was limited to 2007–2010, as 2007 represents the year data on insurance status was introduced and 2010 represents the most recent year that full patient information was made available. Study variables The primary covariate of interest was African American race as designated by the SEER database. Other races (non-African American) were classified as non-Hispanic white, non-black Spanish/Hispanic/Latino, Asian or Pacific Islander, Native American, and other as provided by the SEER dataset. The outcomes of interest were cancer stage at presentation (metastatic vs. non-metastatic), receipt of definitive treatment, and cancer-specific mortality. Other-cause mortality (non-HNC) was used as a proxy to adjust for the effect co-morbidity may have had on receipt of definitive treatment. Metastases were recorded according to the TNM system as defined by the American Joint Committee on Cancer Staging (AJCCS) Manual (6e) [13]. Definitive treatment included surgery, radiation, or combination of surgery and radiation in accordance with the National Comprehensive Cancer Network (NCCN) guidelines for Head and Neck Cancers [14]. Information on chemotherapy is not provided by SEER. Cancer-specific mortality was designated according the SEER dataset. Other variables Age at diagnosis and sex were determined as provided by the SEER database. Income was calculated as median household income, while level of education was categorized as percent of residents P25 years of age with at least a high school diploma; both were determined at the county level by linking to the 2000 United States Census. Residence type was also determined at the county level by linking to the 2003 United States Department of Agriculture rural-urban continuum codes [15]. Insurance status was determined at the patient-level and analyzed as a bivariate value (insured vs. uninsured). Based on the SEER defined tumor sites, HNC was comprised of five sites: Oral cavity, oropharynx, hypopharynx, nasopharynx, and larynx. Tumor stage (according to the TNM system, AJCCS) and grade were determined as provided in the SEER dataset [13]. Statistical analysis Cohort characteristic information was analyzed using the independent samples t-test, v2 test, and Fisher exact test, as appropriate. After adjusting for patient demographics (age at diag-
nosis, sex, marital status) and socioeconomic factors (income, level of education, residence type, and insurance status), multivariable logistic regression analysis was applied to measure the effect African American race had on metastatic vs. localized presentation. Multivariable logistic regression was also used to determine whether there was an association between race and receipt of definitive treatment vs. not for patients with non-metastatic disease, after adjusting for patient demographics (as listed above), socioeconomic factors (as listed above), and tumor characteristics (tumor site, T stage, and grade) [16]. Logistic regression analyses were repeated, site by site. Fine and Gray competing risks regression was then used to model the impact of African American race on mortality due to HNC among patients with non-metastatic disease, after adjusting for the previously listed variables in addition to receipt of definitive therapy [17]. Competing risks regression was repeated site by site. Cumulative incidences of cancer-specific mortality stratified by race (African American vs. non-African American) were generated from the competing-risks regression models described above and displayed graphically [18]. Point estimates and associated confidence intervals (CI) were generated and compared using Gray k-mean P value. All P values were two-sided. A P value 6 0.05 was considered statistically significant. Statistical analyses were performed using STATA 13.0 (StataCorp, College Station, TX) for all analyses. This study was approved by the institutional review board at our institution; a waiver for informed consent was obtained. Results Cohort characteristics Among the study cohort of 34,437 patients, 3820 patients (11.1%) African American and 30,617 (88.9%) were non-African American. Overall, African American patients lived in areas with lower median household incomes, had lower levels of education, were less likely to be insured, and presented with more advanced stage and grade of HNC in comparison to non-African Americans (P < 0.001 for all characteristics) (Table 1). Impact of race on head and neck cancer stage and treatment In comparison to 3.3% of non-African American patients, 5.9% of African American patients presented with metastatic cancer (P < 0.001). Furthermore, after adjusting for patient demographics and socioeconomic factors, multivariable logistic regression analysis revealed that African American patients were significantly more likely to present with metastatic HNC compared to non-African American patients (Adjusted Odds Ratio [AOR] 1.76; 95% CI 1.50– 2.07; P < 0.001). When stratified by site, African American patients had significantly increased odds for metastatic cancer of the oral cavity and oropharynx (AOR 2.20 and 1.93, respectively) (Table 2). After adjusting for tumor characteristics in addition to patient demographics and socioeconomic factors, African American patients with non-metastatic disease were significantly less likely to receive definitive treatment in comparison to non-African American patients (AOR 0.63; 95% CI 0.55–0.72; P < 0.001). When stratified by site, African American patients had decreased odds of receipt of definitive treatment for cancer of the oral cavity, hypopharynx, nasopharynx, and larynx (AOR 0.73, 0.49, 0.55, and 0.67, respectively). After utilizing other-cause mortality (nonHNC death) as a proxy for co-morbidity in the multivariable logistic regression, the odds ratio for receipt of treatment for African American patients did not change (AOR 0.62; 95% CI 0.53–0.73; P < 0.001) (see Table 3).
1179
B.A. Mahal et al. / Oral Oncology 50 (2014) 1177–1181 Table 1 Baseline cohort characteristics.a Characteristic
Non-African American (N = 30,617)
African American (N = 3820)
P-value
Age at diagnosis, years Sex
61.3 ± 12.8
58.8 ± 11.6
<0.001
7635 (24.9) 22,982 (75.1) 46,000 ± 12,000 29.4 ± 7.8
948 (24.8) 2872 (75.2) 42,000 ± 10,000 77.2 ± 6.9
0.87
Insured Uninsured
29,085 (95.0) 1532 (5.0)
3455 (90.5) 365 (9.6)
Oral cavity Oropharynx Hypopharynx Nasopharynx Larynx
18,238 (59.6) 749 (2.5) 1321 (4.3) 1837 (6.0) 8472 (26.7)
1643 (43.0) 165 (4.3) 266 (7.0) 242 (6.3) 1504 (39.4)
Yes No
1023 (3.3) 29,594 (96.7)
224 (5.9) 3596 (94.1)
T1 T2 T3 T4
8841 7181 3224 3891
598 766 575 824
Well Moderately Poorly-undifferentiated
3795 (16.4) 11,461 (49.5) 7881 (34.1)
Female Male b
Income, USD Percent that completed high schoolb Insurance status
Site
<0.001 <0.001 <0.001
<0.001
Metastatic disease
<0.001
Stagec
<0.001 (38.2) (31.0) (13.9) (16.8)
(21.6) (27.7) (20.8) (29.8)
Gradec
<0.001 281 (10.2) 1568 (56.8) 914 (33.1)
Abbreviations: N = Number; USD = United States Dollar. a Continuous data were analyzed using independent-samples t test and presented as mean ± standard deviation. Categorical measurements we analyzed using v2/Fisher exact test and presented as number (percentage). b County-level data. c (%) Among patients with non-metastatic disease and at least clinical stage T1 disease (African American N = 2763; Non-African American N = 23,137).
Table 2 Multivariable (N = 34,437).
logistic
regression
Characteristic
for
presentation
with
metastatic
Multivariable analysisb a
Adjusted OR (95% CI) Race Non-African American African American (Overall) Site Oral cavity Oropharynx Hypopharynx Nasopharynx Larynx
disease
Characteristic P
1.0 (ref) 1.76 (1.50–2.07)
<0.001
2.20 1.93 1.22 1.47 1.15
<0.001 0.038 0.404 0.092 0.370
(1.70–2.84) (1.04–3.6) (0.76–1.98) (0.94–2.33) (0.85–1.55)
Table 3 Multivariable logistic regression for receipt of definitive treatment among patients with non-metastatic disease (N = 26,614).
Abbreviations: OR = Odds Ratio. a OR comparing the rate of metastatic disease. b Multivariable analyses are adjusted for patient demographics and socioeconomic factors.
Head and neck cancer specific mortality After a median follow-up of 19 months, the cumulative incidence for HNC mortality was significantly higher for African American patients in comparison to patients in the general population (P < 0.001; Fig. 1). The 3-year HNC mortality rate was 46.7% (95% CI 43.2–50.4%) for African American patients and 26.6% (95% CI 25.7–27.5%) for non-African American patients (P < 0.001). After adjusting for patient demographics, socioeconomic factors, tumor characteristics, and receipt of definitive therapy, competing risks regression analysis revealed that African American patients have a higher risk of head and neck cancer-specific mortality (AHR 1.19; 95% CI 1.09–1.29; P < 0.001). When stratified by site, African American patients had significantly increased odds for
Race (2007–2010) Non-African American African American Site Oral cavity Oropharynx Hypopharynx Nasopharynx Larynx Race (2007–2010) Sensitivityc Non-African American African American
Multivariable Analysisb Adjusted OR (95% CI)a
P
1.0 (ref) 0.63 (0.55–0.72)
<0.001
0.73 1.33 0.49 0.55 0.67
(0.58–0.92) (0.74–2.40) (0.32–0.73) (0.31–0.97) (0.55–0.83)
0.007 0.340 0.001 0.040 <0.001
1.0 (ref) 0.67 (0.58–0.78)
<0.001
Abbreviations: OR = Odds Ratio, HNCSM = Head and Neck Cancer Specific Mortality. a OR comparing the rate of receipt of definitive treatment. b Multivariable analyses are adjusted for patient demographics and socioeconomic factors. c Sensitivity analysis using other-cause mortality as a proxy for co-morbidity.
cancer-specific mortality of the oral cavity and hypopharynx (AOR 1.30 and 1.39, respectively) (Table 4). Discussion The main objective of this study was to examine the association between African American race and stage at diagnosis, receipt of definitive therapy, and cancer-specific mortality among patients HNC, and to determine whether any observed disparities are narrowing with time. We found that African American patients were significantly more likely to present with metastatic cancer. Furthermore, among patients with non-metastatic disease, African
1180
B.A. Mahal et al. / Oral Oncology 50 (2014) 1177–1181
Fig. 1. Cumulative incidence of head and neck cancer specific mortality by race (N = 26,614; Gray k-mean P < 0.001).
Table 4 Fine and Gray competing-risks regression for cancer-specific mortality (N = 26,614). Characteristic
Race Non-African American African American Site Oral cavity Oropharynx Hypopharynx Nasopharynx Larynx
Multivariable Fine and Gray Competing-Risks Regression for HNCSMb AHR (95% CI)a
P
1.0 (ref) 1.19 (1.09–1.29)
<0.001
1.30 1.24 1.39 1.06 0.99
<0.001 0.275 0.009 0.759 0.926
(1.15–1.47) (0.84–1.82) (1.08–1.78) (0.72–1.58) (0.86–1.15)
Abbreviations: AHR = Adjusted Hazard Ratio, HNCSM = Head and Neck Cancer Specific Mortality. a OR comparing the rate of receipt of definitive treatment. b Multivariable analyses are adjusted for patient demographics, socioeconomic factors, and socioeconomic factors.
Americans were less likely to receive definitive treatment, and possessed a higher risk of head and neck cancer-specific mortality in comparison to the study cohort. To date, this study represents the largest contemporary examination of disparities in HNC outcomes. The importance of the findings in this study is that they demonstrate substantial disparities that African American patients with head and neck cancer face, namely, later stage at presentation, less treatment, and worse survival. The unacceptably high rates of disparity found in our study should serve as immediate targets for urgent healthcare policy intervention. It has been established that earlier detection of HNC leads to better prognosis and better chances of cure from treatment [3]. Accordingly, a coordinated effort between medical and oral health providers must take precedence in order to make the push for earlier detection and higher rates of treatment among all patients with HNC. This can potentially be accomplished by maximizing access to surveillance across all racial and sociodemographic divides which, in turn, could mitigate the overwhelming unequal burden African Americans face with HNC. Along with initiatives that aim to increase earlier detection and treatment of HNC for all patients, disparities in HNC survival outcomes need prospective validation in order to identify other factors that may be responsible for the unequal rates of HNC mortality observed in this study. It is imperative to ensure that representative populations of African Americans are enrolled in clinical trials to help better guide future care patterns and treatment strategies aimed at reducing disparities [19].
Although this study establishes multileveled disparities found in HNC, the reasons behind disparate HNC outcomes are likely layered and multifactorial. Differences in socioeconomic status, biologic/genetic variation, and disproportionate exposure to alcohol and tobacco products are implicated as potential contributors to the excess incidence of HNC among African American patients [8,10,20,21]. Increasing oral health education has been suggested as a preventative approach to help reduce the environmental risks that African Americans are exposed to [22]. Unfortunately, the opportunity to provide education to a patient begins with access to a healthcare provider, a benefit that some African Americans do not have [23–25]. Access to a healthcare provider also promotes early detection of HNC [22,26,27]. Historically, it has been suggested that African Americans present with advanced stage HNC cancer at a significantly higher frequency than non-African Americans likely due to inadequate access to care [2,8,10,11,22]. One study found that African American patients with head and neck squamous cell carcinoma had a significantly greater incidence of stage IV disease (65.7% vs. 46.6%, P < .0001) and greater incidence of inoperable disease in comparison to white patients (57.1% vs. 31.0%, P < .0001) [10]. The authors attributed differences in stage, treatment, and survival to differences in insurance coverage. Even after adjusting for insurance status and other pertinent sociodemographic factors, our findings revealed that African Americans were 76% more likely to present with metastases in comparison to the general population, so differences in stage at presentation are likely due to more than just differences in insurance status. Of the subset of HNC patients who presented free of metastatic disease, we found that African American patients had a 19% increased risk of cancer-specific mortality compared to the general population. Similar findings were demonstrated in a smaller regional study, where the 5-year cause-specific survival rate of head and neck squamous cell carcinoma was found to be significantly lower for African American patients (52% vs. 74%, P < 0.001) [9]. Despite the evidence showing poor survival among African Americans with non-metastatic disease, our study demonstrated that African American patients receive disparately low rates of definitive treatment compared to non-African American patients. Hypotheses as to why African American patients do not receive appropriate treatment include late stage of diagnosis, lack of access to care, lack of continual care, poor communication between patient and provider, and patient misconceptions about healthcare [22,28,29]. Furthermore, recent literature examining patients undergoing radiotherapy treatment suggests that disparities in HNC survival can be overcome through increased treatment compliance [30]. These hypotheses should be examined in further detail in future studies. Although this study brings to light many interesting findings regarding disparities among patients with HNC, the SEER database offers limited information concerning a large proportion of the United States and the results should be viewed within the limitations of the data. First, the SEER database does not collect information on lifestyle risk factors such as smoking and alcohol status at the patient level, which are implicated as etiologic factors of HNC [31–33]. Biological factors, mainly presence of human papilloma virus (HPV), were also not taken into consideration when analyzing patient risk for HNC. The recent rise in oropharyngeal squamous cell carcinoma has been attributed to HPV, which disproportionately affects white, middle aged males [34]. SEER also does not provide information on chemotherapy or the quality of treatment patients are receiving, which may have contributed to the observed disparities in outcomes. Additionally, although we were fortunate to include medical insurance status within our cohort, dental benefit status is not captured within SEER. As African American patients have been shown to disproportionately under-utilize oral health services, lack of access to an oral health provider and
B.A. Mahal et al. / Oral Oncology 50 (2014) 1177–1181
subsequent surveillance may have greatly contributed to the disproportionate number of African American patients presents with metastases [35–37]. Including medical insurance status also limited our study period (2007–2010) and subsequent median follow-up (16 months). Despite the short follow-up, significant differences in cancer-specific mortality were found. Lastly, SEER does not provide patient information on co-morbidity, which can affect treatment selection. To address this limitation, we utilized other-cause mortality (non-HNC) as a proxy to adjust for the effect co-morbidity and still found disparate receipt of definitive treatment. Despite the potential limitations, this study highlights significant racial disparities in the use of definitive treatment and the presentation of metastatic disease among patients with HNC. These disparities should be urgently studied and addressed, as they may be potentially correctable contributors to excess HNC mortality among African American patients. Furthermore, the unequal burden of HNC outcomes that African Americans face deserves immediate attention at both the provider and policy level. Specifically, HNC surveillance programs must be implemented and studied to determine their effect on stage at presentation and to better design effective surveillance programs for patients in underserved minority populations. Author contributions Brandon A Mahal and Gino Inverso had equal first author contribution. All authors contributed to study concepts, study design, data analysis and interpretation, manuscript preparation, and manuscript editing. Conflict of interest statement None declared. Acknowledgements This study was not funded by any outside sources. Sung-Kiang Chuang DMD MD DMSc had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. References [1] Siegel R, Ma J, Zou Z, Jemal A. Cancer statistics, 2014. CA Cancer J Clin 2014;64:9–29. [2] Goodwin WJ, Thomas GR, Parker DF, et al. Unequal burden of head and neck cancer in the United States. Head Neck 2008;30:358–71. [3] Vokes EE, Weichselbaum RR, Lippman SM, Hong WK. Head and neck cancer. N Engl J Med 1993;328:184–94. [4] Arbes Jr SJ, Olshan AF, Caplan DJ, Schoenbach VJ, Slade GD, Symons MJ. Factors contributing to the poorer survival of black Americans diagnosed with oral cancer (United States). Cancer Causes Control 1999;10:513–23. [5] Ragin CC, Langevin SM, Marzouk M, Grandis J, Taioli E. Determinants of head and neck cancer survival by race. Head Neck 2011;33:1092–8. [6] Molina MA, Cheung MC, Perez EA, et al. African American and poor patients have a dramatically worse prognosis for head and neck cancer: an examination of 20,915 patients. Cancer 2008;113:2797–806. [7] Zakeri K, MacEwan I, Vazirnia A, et al. Race and competing mortality in advanced head and neck cancer. Oral Oncol 2014;50:40–4. [8] Nichols AC, Bhattacharyya N. Racial differences in stage and survival in head and neck squamous cell carcinoma. Laryngoscope 2007;117:770–5. [9] Al-Othman MO, Morris CG, Logan HL, Hinerman RW, Amdur RJ, Mendenhall WM. Impact of race on outcome after definitive radiotherapy for squamous cell carcinoma of the head and neck. Cancer 2003;98:2467–72.
1181
[10] Gourin CG, Podolsky RH. Racial disparities in patients with head and neck squamous cell carcinoma. Laryngoscope 2006;116:1093–106. [11] Moore RJ, Doherty DA, Do KA, Chamberlain RM, Khuri FR. Racial disparity in survival of patients with squamous cell carcinoma of the oral cavity and pharynx. Ethn Health 2001;6:165–77. [12] Surveillance, Epidemiology, and End Results (SEER) Program: Research Data (2007–2010). National Cancer Institute, DCCPS, Surveillance Research Program, Surveillance Systems Branch, based on November 2012 SEER data submission, posted to the SEER web site. February 2014.. [13] American Joint Committee on Cancer. The American Joint Committee on Cancer Staging Manual, 6e. New York, New York: Springer-Verlag; 2002. [14] National Comprehensive Cancer Network. Guidelines for the management of head and neck cancers. [Accessed 30.02.14]. [15] United States Department of Agriculture Rural-Urban Continuum Codes. [accessed 27.02.14]. [16] Cox DR. Regression models and life-tables. J R Stat Soc B 1972;34:187–220. [17] Fine J, Gray R. A proportional hazards model for the subdistribution of a competing risk. J Am Statist Assoc 1999;94:496–509. [18] Coviello V, Boggess M. Cumulative incidence estimation in the presence of competing risks. Stata J 2004;4:103–12. [19] Murthy VH, Krumholz HM, Gross CP. Participation in cancer clinical trials: race-, sex-, and age-based disparities. JAMA 2004;29:2720–6. [20] Day GL, Blot WJ, Austin DF, et al. Racial differences in risk of oral and pharyngeal cancer: alcohol, tobacco, and other determinants. J Natl Cancer Inst 1993;85:465–73. [21] Settle K, Taylor R, Wolf J, et al. Race impacts outcome in stage III/IV squamous cell carcinomas of the head and neck after concurrent chemoradiation therapy. Cancer 2009;115:1744–52. [22] Subramanian S, Chen A. Treatment patterns and survival among low-income medicaid patients with head and neck cancer. JAMA Otolaryngol Head Neck Surg 2013;139:489–95. [23] Sohn W, Ismail A, Amaya A, Lepkowski J. Determinants of dental care visits among low-income African–American children. J Am Dent Assoc 2007;138:309–18. [24] Shelley D, Russell S, Parikh NS, Fahs M. Ethnic disparities in self-reported oral health status and access to care among older adults in NYC. J Urban Health 2011;88:651–62. [25] Schrimshaw EW, Siegel K, Wolfson NH, Mitchell DA, Kunzel C. Insurancerelated barriers to accessing dental care among African American adults with oral health symptoms in Harlem, New York City. Am J Public Health 2011;101:1420–8. [26] Johnson S, McDonald JT, Corsten MJ. Socioeconomic factors in head and neck cancer. J Otolaryngol Head Neck Surg 2008;37:597–601. [27] Shavers VL, Harlan LC, Winn D, Davis WW. Racial/ethnic patterns of care for cancers of the oral cavity, pharynx, larynx, sinuses, and salivary glands. Cancer Metastasis Rev 2003;22:25–38. [28] Cykert S, Dilworth-Anderson P, Monroe MH, et al. Factors associated with decisions to undergo surgery among patients with newly diagnosed earlystage lung cancer. JAMA 2010;303:2368–76. [29] Weng Y, Korte JE. Racial disparities in being recommended to surgery for oral and oropharyngeal cancer in the United States. Community Dent Oral Epidemiol 2012;40:80–8. [30] Liu GF, Ranck MC, Solanki AA, et al. Racial parities in outcomes after radiotherapy for head and neck cancer. Cancer 2014;120:244–52. [31] Blot WJ, McLaughlin JK, Winn DM, et al. Smoking and drinking in relation to oral and pharyngeal cancer. Cancer Res 1988;48:3282–7. [32] Lewin F, Norell SE, Johansson H, et al. Smoking tobacco, oral snuff, and alcohol in the etiology of squamous cell carcinoma of the head and neck: a populationbased case-referent study in Sweden. Cancer 1998;82:1367–75. [33] Tuyns AJ, Esteve J, Raymond L, et al. Cancer of the larynx/hypopharynx, tobacco and alcohol: IARC international case-control study in Turin and Varese (Italy), Zaragoza and Navarra (Spain), Geneva (Switzerland) and Calvados (France). Int J Cancer 1988;41:483–91. [34] Pytynia KB, Dahlstrom KR, Sturgis EM. Epidemiology of HPV-associated oropharyngeal cancer. Oral Oncol 2014;50:380–6. [35] Doty HE, Weech-Maldonado R. Racial/ethnic disparities in adult preventive dental care use. J Health Care Poor Underserved 2003;14:516–34. [36] Gilbert GH, Shah GR, Shelton BJ, et al. Racial differences in predictors of dental care use. Health Serv Res 2002;37:1487–507. [37] Fisher MA, Gilbert GH, Shelton BJ. A cohort study found racial differences in dental insurance, utilization, and the effect of care on quality of life. J Clin Epidemiol 2004;57:853–7.
Contents lists available at ScienceDirect
Oral Oncology journal homepage: www.elsevier.com/locate/oraloncology
Impact of African–American race on presentation, treatment, and survival of head and neck cancer Brandon A. Mahal a,1, Gino Inverso b,1, Ayal A. Aizer c, R. Bruce Donoff b,d, Sung-Kiang Chuang b,d,⇑ a
Harvard Medical School, 25 Shattuck St., Boston, MA 02115, USA Harvard School of Dental Medicine, 188 Longwood Ave., Boston, MA 02115, USA c Harvard Radiation Oncology Program, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114, USA d Department of Oral and Maxillofacial Surgery, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114, USA b
a r t i c l e
i n f o
Article history: Received 18 May 2014 Received in revised form 18 July 2014 Accepted 6 September 2014 Available online 26 September 2014 Keywords: Head and neck cancer Healthcare disparities African Americans Minority health Head and neck cancer specific survival
s u m m a r y Objectives: To determine the associations between African American race and stage at diagnosis, receipt of definitive therapy, and cancer-specific mortality among patients with head and neck cancer. Materials and methods: The Surveillance, Epidemiology and End Results (SEER) database was used to conduct a retrospective study on 34,437 patients diagnosed with head and neck cancer from 2007 to 2010. Multivariable logistic regression analyses were applied to determine the impact of race on cancer stage at presentation (metastatic vs. non-metastatic) and receipt of definitive treatment. Fine and Gray competing-risks regression modeled the association between race and head and neck cancer-specific mortality. Results: African Americans were more likely to present with metastatic cancer compared to non-African Americans (Adjusted Odds Ratio [AOR] 1.76; CI 1.50–2.07; P < 0.001). Among patients with nonmetastatic disease, African Americans were less likely to receive definitive treatment (AOR 0.63; CI 0.55–0.72; P < 0.001). After a median follow-up of 19 months, African Americans with non-metastatic disease were found to have a higher risk of head and neck cancer specific mortality (AHR 1.19; 95% CI 1.09–1.29; P < 0.001). Conclusion: African Americans with head and neck cancer are more likely to present with metastatic disease, less likely to be treated definitively, and are more likely to die from head and neck cancer. The unacceptably high rates of disparity found in this study should serve as immediate targets for urgent healthcare policy intervention. Ó 2014 Elsevier Ltd. All rights reserved.
Introduction There are an estimated 42,440 new cases of oral and pharyngeal cancer diagnosed each year and 8390 deaths annually [1]. The 5-year survival rate for all patients diagnosed with oral and pharyngeal cancer is 64% [1]. When stratified by race, the 5-year survival rate for African Americans drops to 43%. Although there are currently no practical head and neck cancer (HNC) screening or detection strategies in place, cure rates approximate 50% and can reach 90% with early stage detection [2,3]. Still most patients present with advanced stage disease [1]. Previous studies in HNC outcomes research have demonstrated that African American patients face an unequal burden of disease ⇑ Corresponding author at: Harvard School of Dental Medicine, 188 Longwood Ave, Boston, MA 02115, USA. Tel.: +1 617 432 1434; fax: +1 617 432 1897. E-mail address: [email protected] (S.-K. Chuang). 1 Authors had equal 1st author contribution. http://dx.doi.org/10.1016/j.oraloncology.2014.09.004 1368-8375/Ó 2014 Elsevier Ltd. All rights reserved.
due to differences in socioeconomic status, limited access to care, advanced cancer stage, lower receipt of treatment, and comorbidity [4–7]. Nevertheless, the existing literature focusing on racial disparities in HNC outcomes have not been able to account for relevant sociodemographic factors such as insurance, which has been suggested to have a significant impact on cancer outcomes and treatment patterns [2,8,9]. Additionally, the majority of existing studies do not utilize cancer specific mortality as an endpoint and are unable to account for comorbid disease that has been shown to disproportionally contribute to the mortality of African American patients with HNC [7]. Lastly, previous studies focusing on racial disparities in HNC outcomes have examined smaller cohorts restricted to few sites, thus limiting the number of pertinent covariates that could be adjusted for and generalized conclusions that could be made about disparities in HNC [8–11]. The Surveillance, Epidemiology and End Results Program (SEER) program, sponsored by the National Cancer Institute, collects incidence, prevalence, survival and mortality on cancer [12]. The SEER
1178
B.A. Mahal et al. / Oral Oncology 50 (2014) 1177–1181
program captures approximately 97% of incident cancers and the 17 tumor registries encompass approximately 26% of the US population. Additionally, the database provides valuable demographic information, such as county level income level and educational status, residence type, and as of 2007, individual level data on insurance status. The main objective of this study is to utilize the SEER database to examine the associations between African American race and stage at diagnosis, receipt of definitive therapy, and cancer-specific mortality among patients with HNC utilizing robust multivariable models adjusting for the potentially pertinent sociodemographic covariates mentioned above. Material and methods Study population The SEER database was utilized to identify 34,437 patients (3820 African American and 30,617 non-African American) diagnosed with HNC from 2007 to 2010. The inclusion period was limited to 2007–2010, as 2007 represents the year data on insurance status was introduced and 2010 represents the most recent year that full patient information was made available. Study variables The primary covariate of interest was African American race as designated by the SEER database. Other races (non-African American) were classified as non-Hispanic white, non-black Spanish/Hispanic/Latino, Asian or Pacific Islander, Native American, and other as provided by the SEER dataset. The outcomes of interest were cancer stage at presentation (metastatic vs. non-metastatic), receipt of definitive treatment, and cancer-specific mortality. Other-cause mortality (non-HNC) was used as a proxy to adjust for the effect co-morbidity may have had on receipt of definitive treatment. Metastases were recorded according to the TNM system as defined by the American Joint Committee on Cancer Staging (AJCCS) Manual (6e) [13]. Definitive treatment included surgery, radiation, or combination of surgery and radiation in accordance with the National Comprehensive Cancer Network (NCCN) guidelines for Head and Neck Cancers [14]. Information on chemotherapy is not provided by SEER. Cancer-specific mortality was designated according the SEER dataset. Other variables Age at diagnosis and sex were determined as provided by the SEER database. Income was calculated as median household income, while level of education was categorized as percent of residents P25 years of age with at least a high school diploma; both were determined at the county level by linking to the 2000 United States Census. Residence type was also determined at the county level by linking to the 2003 United States Department of Agriculture rural-urban continuum codes [15]. Insurance status was determined at the patient-level and analyzed as a bivariate value (insured vs. uninsured). Based on the SEER defined tumor sites, HNC was comprised of five sites: Oral cavity, oropharynx, hypopharynx, nasopharynx, and larynx. Tumor stage (according to the TNM system, AJCCS) and grade were determined as provided in the SEER dataset [13]. Statistical analysis Cohort characteristic information was analyzed using the independent samples t-test, v2 test, and Fisher exact test, as appropriate. After adjusting for patient demographics (age at diag-
nosis, sex, marital status) and socioeconomic factors (income, level of education, residence type, and insurance status), multivariable logistic regression analysis was applied to measure the effect African American race had on metastatic vs. localized presentation. Multivariable logistic regression was also used to determine whether there was an association between race and receipt of definitive treatment vs. not for patients with non-metastatic disease, after adjusting for patient demographics (as listed above), socioeconomic factors (as listed above), and tumor characteristics (tumor site, T stage, and grade) [16]. Logistic regression analyses were repeated, site by site. Fine and Gray competing risks regression was then used to model the impact of African American race on mortality due to HNC among patients with non-metastatic disease, after adjusting for the previously listed variables in addition to receipt of definitive therapy [17]. Competing risks regression was repeated site by site. Cumulative incidences of cancer-specific mortality stratified by race (African American vs. non-African American) were generated from the competing-risks regression models described above and displayed graphically [18]. Point estimates and associated confidence intervals (CI) were generated and compared using Gray k-mean P value. All P values were two-sided. A P value 6 0.05 was considered statistically significant. Statistical analyses were performed using STATA 13.0 (StataCorp, College Station, TX) for all analyses. This study was approved by the institutional review board at our institution; a waiver for informed consent was obtained. Results Cohort characteristics Among the study cohort of 34,437 patients, 3820 patients (11.1%) African American and 30,617 (88.9%) were non-African American. Overall, African American patients lived in areas with lower median household incomes, had lower levels of education, were less likely to be insured, and presented with more advanced stage and grade of HNC in comparison to non-African Americans (P < 0.001 for all characteristics) (Table 1). Impact of race on head and neck cancer stage and treatment In comparison to 3.3% of non-African American patients, 5.9% of African American patients presented with metastatic cancer (P < 0.001). Furthermore, after adjusting for patient demographics and socioeconomic factors, multivariable logistic regression analysis revealed that African American patients were significantly more likely to present with metastatic HNC compared to non-African American patients (Adjusted Odds Ratio [AOR] 1.76; 95% CI 1.50– 2.07; P < 0.001). When stratified by site, African American patients had significantly increased odds for metastatic cancer of the oral cavity and oropharynx (AOR 2.20 and 1.93, respectively) (Table 2). After adjusting for tumor characteristics in addition to patient demographics and socioeconomic factors, African American patients with non-metastatic disease were significantly less likely to receive definitive treatment in comparison to non-African American patients (AOR 0.63; 95% CI 0.55–0.72; P < 0.001). When stratified by site, African American patients had decreased odds of receipt of definitive treatment for cancer of the oral cavity, hypopharynx, nasopharynx, and larynx (AOR 0.73, 0.49, 0.55, and 0.67, respectively). After utilizing other-cause mortality (nonHNC death) as a proxy for co-morbidity in the multivariable logistic regression, the odds ratio for receipt of treatment for African American patients did not change (AOR 0.62; 95% CI 0.53–0.73; P < 0.001) (see Table 3).
1179
B.A. Mahal et al. / Oral Oncology 50 (2014) 1177–1181 Table 1 Baseline cohort characteristics.a Characteristic
Non-African American (N = 30,617)
African American (N = 3820)
P-value
Age at diagnosis, years Sex
61.3 ± 12.8
58.8 ± 11.6
<0.001
7635 (24.9) 22,982 (75.1) 46,000 ± 12,000 29.4 ± 7.8
948 (24.8) 2872 (75.2) 42,000 ± 10,000 77.2 ± 6.9
0.87
Insured Uninsured
29,085 (95.0) 1532 (5.0)
3455 (90.5) 365 (9.6)
Oral cavity Oropharynx Hypopharynx Nasopharynx Larynx
18,238 (59.6) 749 (2.5) 1321 (4.3) 1837 (6.0) 8472 (26.7)
1643 (43.0) 165 (4.3) 266 (7.0) 242 (6.3) 1504 (39.4)
Yes No
1023 (3.3) 29,594 (96.7)
224 (5.9) 3596 (94.1)
T1 T2 T3 T4
8841 7181 3224 3891
598 766 575 824
Well Moderately Poorly-undifferentiated
3795 (16.4) 11,461 (49.5) 7881 (34.1)
Female Male b
Income, USD Percent that completed high schoolb Insurance status
Site
<0.001 <0.001 <0.001
<0.001
Metastatic disease
<0.001
Stagec
<0.001 (38.2) (31.0) (13.9) (16.8)
(21.6) (27.7) (20.8) (29.8)
Gradec
<0.001 281 (10.2) 1568 (56.8) 914 (33.1)
Abbreviations: N = Number; USD = United States Dollar. a Continuous data were analyzed using independent-samples t test and presented as mean ± standard deviation. Categorical measurements we analyzed using v2/Fisher exact test and presented as number (percentage). b County-level data. c (%) Among patients with non-metastatic disease and at least clinical stage T1 disease (African American N = 2763; Non-African American N = 23,137).
Table 2 Multivariable (N = 34,437).
logistic
regression
Characteristic
for
presentation
with
metastatic
Multivariable analysisb a
Adjusted OR (95% CI) Race Non-African American African American (Overall) Site Oral cavity Oropharynx Hypopharynx Nasopharynx Larynx
disease
Characteristic P
1.0 (ref) 1.76 (1.50–2.07)
<0.001
2.20 1.93 1.22 1.47 1.15
<0.001 0.038 0.404 0.092 0.370
(1.70–2.84) (1.04–3.6) (0.76–1.98) (0.94–2.33) (0.85–1.55)
Table 3 Multivariable logistic regression for receipt of definitive treatment among patients with non-metastatic disease (N = 26,614).
Abbreviations: OR = Odds Ratio. a OR comparing the rate of metastatic disease. b Multivariable analyses are adjusted for patient demographics and socioeconomic factors.
Head and neck cancer specific mortality After a median follow-up of 19 months, the cumulative incidence for HNC mortality was significantly higher for African American patients in comparison to patients in the general population (P < 0.001; Fig. 1). The 3-year HNC mortality rate was 46.7% (95% CI 43.2–50.4%) for African American patients and 26.6% (95% CI 25.7–27.5%) for non-African American patients (P < 0.001). After adjusting for patient demographics, socioeconomic factors, tumor characteristics, and receipt of definitive therapy, competing risks regression analysis revealed that African American patients have a higher risk of head and neck cancer-specific mortality (AHR 1.19; 95% CI 1.09–1.29; P < 0.001). When stratified by site, African American patients had significantly increased odds for
Race (2007–2010) Non-African American African American Site Oral cavity Oropharynx Hypopharynx Nasopharynx Larynx Race (2007–2010) Sensitivityc Non-African American African American
Multivariable Analysisb Adjusted OR (95% CI)a
P
1.0 (ref) 0.63 (0.55–0.72)
<0.001
0.73 1.33 0.49 0.55 0.67
(0.58–0.92) (0.74–2.40) (0.32–0.73) (0.31–0.97) (0.55–0.83)
0.007 0.340 0.001 0.040 <0.001
1.0 (ref) 0.67 (0.58–0.78)
<0.001
Abbreviations: OR = Odds Ratio, HNCSM = Head and Neck Cancer Specific Mortality. a OR comparing the rate of receipt of definitive treatment. b Multivariable analyses are adjusted for patient demographics and socioeconomic factors. c Sensitivity analysis using other-cause mortality as a proxy for co-morbidity.
cancer-specific mortality of the oral cavity and hypopharynx (AOR 1.30 and 1.39, respectively) (Table 4). Discussion The main objective of this study was to examine the association between African American race and stage at diagnosis, receipt of definitive therapy, and cancer-specific mortality among patients HNC, and to determine whether any observed disparities are narrowing with time. We found that African American patients were significantly more likely to present with metastatic cancer. Furthermore, among patients with non-metastatic disease, African
1180
B.A. Mahal et al. / Oral Oncology 50 (2014) 1177–1181
Fig. 1. Cumulative incidence of head and neck cancer specific mortality by race (N = 26,614; Gray k-mean P < 0.001).
Table 4 Fine and Gray competing-risks regression for cancer-specific mortality (N = 26,614). Characteristic
Race Non-African American African American Site Oral cavity Oropharynx Hypopharynx Nasopharynx Larynx
Multivariable Fine and Gray Competing-Risks Regression for HNCSMb AHR (95% CI)a
P
1.0 (ref) 1.19 (1.09–1.29)
<0.001
1.30 1.24 1.39 1.06 0.99
<0.001 0.275 0.009 0.759 0.926
(1.15–1.47) (0.84–1.82) (1.08–1.78) (0.72–1.58) (0.86–1.15)
Abbreviations: AHR = Adjusted Hazard Ratio, HNCSM = Head and Neck Cancer Specific Mortality. a OR comparing the rate of receipt of definitive treatment. b Multivariable analyses are adjusted for patient demographics, socioeconomic factors, and socioeconomic factors.
Americans were less likely to receive definitive treatment, and possessed a higher risk of head and neck cancer-specific mortality in comparison to the study cohort. To date, this study represents the largest contemporary examination of disparities in HNC outcomes. The importance of the findings in this study is that they demonstrate substantial disparities that African American patients with head and neck cancer face, namely, later stage at presentation, less treatment, and worse survival. The unacceptably high rates of disparity found in our study should serve as immediate targets for urgent healthcare policy intervention. It has been established that earlier detection of HNC leads to better prognosis and better chances of cure from treatment [3]. Accordingly, a coordinated effort between medical and oral health providers must take precedence in order to make the push for earlier detection and higher rates of treatment among all patients with HNC. This can potentially be accomplished by maximizing access to surveillance across all racial and sociodemographic divides which, in turn, could mitigate the overwhelming unequal burden African Americans face with HNC. Along with initiatives that aim to increase earlier detection and treatment of HNC for all patients, disparities in HNC survival outcomes need prospective validation in order to identify other factors that may be responsible for the unequal rates of HNC mortality observed in this study. It is imperative to ensure that representative populations of African Americans are enrolled in clinical trials to help better guide future care patterns and treatment strategies aimed at reducing disparities [19].
Although this study establishes multileveled disparities found in HNC, the reasons behind disparate HNC outcomes are likely layered and multifactorial. Differences in socioeconomic status, biologic/genetic variation, and disproportionate exposure to alcohol and tobacco products are implicated as potential contributors to the excess incidence of HNC among African American patients [8,10,20,21]. Increasing oral health education has been suggested as a preventative approach to help reduce the environmental risks that African Americans are exposed to [22]. Unfortunately, the opportunity to provide education to a patient begins with access to a healthcare provider, a benefit that some African Americans do not have [23–25]. Access to a healthcare provider also promotes early detection of HNC [22,26,27]. Historically, it has been suggested that African Americans present with advanced stage HNC cancer at a significantly higher frequency than non-African Americans likely due to inadequate access to care [2,8,10,11,22]. One study found that African American patients with head and neck squamous cell carcinoma had a significantly greater incidence of stage IV disease (65.7% vs. 46.6%, P < .0001) and greater incidence of inoperable disease in comparison to white patients (57.1% vs. 31.0%, P < .0001) [10]. The authors attributed differences in stage, treatment, and survival to differences in insurance coverage. Even after adjusting for insurance status and other pertinent sociodemographic factors, our findings revealed that African Americans were 76% more likely to present with metastases in comparison to the general population, so differences in stage at presentation are likely due to more than just differences in insurance status. Of the subset of HNC patients who presented free of metastatic disease, we found that African American patients had a 19% increased risk of cancer-specific mortality compared to the general population. Similar findings were demonstrated in a smaller regional study, where the 5-year cause-specific survival rate of head and neck squamous cell carcinoma was found to be significantly lower for African American patients (52% vs. 74%, P < 0.001) [9]. Despite the evidence showing poor survival among African Americans with non-metastatic disease, our study demonstrated that African American patients receive disparately low rates of definitive treatment compared to non-African American patients. Hypotheses as to why African American patients do not receive appropriate treatment include late stage of diagnosis, lack of access to care, lack of continual care, poor communication between patient and provider, and patient misconceptions about healthcare [22,28,29]. Furthermore, recent literature examining patients undergoing radiotherapy treatment suggests that disparities in HNC survival can be overcome through increased treatment compliance [30]. These hypotheses should be examined in further detail in future studies. Although this study brings to light many interesting findings regarding disparities among patients with HNC, the SEER database offers limited information concerning a large proportion of the United States and the results should be viewed within the limitations of the data. First, the SEER database does not collect information on lifestyle risk factors such as smoking and alcohol status at the patient level, which are implicated as etiologic factors of HNC [31–33]. Biological factors, mainly presence of human papilloma virus (HPV), were also not taken into consideration when analyzing patient risk for HNC. The recent rise in oropharyngeal squamous cell carcinoma has been attributed to HPV, which disproportionately affects white, middle aged males [34]. SEER also does not provide information on chemotherapy or the quality of treatment patients are receiving, which may have contributed to the observed disparities in outcomes. Additionally, although we were fortunate to include medical insurance status within our cohort, dental benefit status is not captured within SEER. As African American patients have been shown to disproportionately under-utilize oral health services, lack of access to an oral health provider and
B.A. Mahal et al. / Oral Oncology 50 (2014) 1177–1181
subsequent surveillance may have greatly contributed to the disproportionate number of African American patients presents with metastases [35–37]. Including medical insurance status also limited our study period (2007–2010) and subsequent median follow-up (16 months). Despite the short follow-up, significant differences in cancer-specific mortality were found. Lastly, SEER does not provide patient information on co-morbidity, which can affect treatment selection. To address this limitation, we utilized other-cause mortality (non-HNC) as a proxy to adjust for the effect co-morbidity and still found disparate receipt of definitive treatment. Despite the potential limitations, this study highlights significant racial disparities in the use of definitive treatment and the presentation of metastatic disease among patients with HNC. These disparities should be urgently studied and addressed, as they may be potentially correctable contributors to excess HNC mortality among African American patients. Furthermore, the unequal burden of HNC outcomes that African Americans face deserves immediate attention at both the provider and policy level. Specifically, HNC surveillance programs must be implemented and studied to determine their effect on stage at presentation and to better design effective surveillance programs for patients in underserved minority populations. Author contributions Brandon A Mahal and Gino Inverso had equal first author contribution. All authors contributed to study concepts, study design, data analysis and interpretation, manuscript preparation, and manuscript editing. Conflict of interest statement None declared. Acknowledgements This study was not funded by any outside sources. Sung-Kiang Chuang DMD MD DMSc had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. References [1] Siegel R, Ma J, Zou Z, Jemal A. Cancer statistics, 2014. CA Cancer J Clin 2014;64:9–29. [2] Goodwin WJ, Thomas GR, Parker DF, et al. Unequal burden of head and neck cancer in the United States. Head Neck 2008;30:358–71. [3] Vokes EE, Weichselbaum RR, Lippman SM, Hong WK. Head and neck cancer. N Engl J Med 1993;328:184–94. [4] Arbes Jr SJ, Olshan AF, Caplan DJ, Schoenbach VJ, Slade GD, Symons MJ. Factors contributing to the poorer survival of black Americans diagnosed with oral cancer (United States). Cancer Causes Control 1999;10:513–23. [5] Ragin CC, Langevin SM, Marzouk M, Grandis J, Taioli E. Determinants of head and neck cancer survival by race. Head Neck 2011;33:1092–8. [6] Molina MA, Cheung MC, Perez EA, et al. African American and poor patients have a dramatically worse prognosis for head and neck cancer: an examination of 20,915 patients. Cancer 2008;113:2797–806. [7] Zakeri K, MacEwan I, Vazirnia A, et al. Race and competing mortality in advanced head and neck cancer. Oral Oncol 2014;50:40–4. [8] Nichols AC, Bhattacharyya N. Racial differences in stage and survival in head and neck squamous cell carcinoma. Laryngoscope 2007;117:770–5. [9] Al-Othman MO, Morris CG, Logan HL, Hinerman RW, Amdur RJ, Mendenhall WM. Impact of race on outcome after definitive radiotherapy for squamous cell carcinoma of the head and neck. Cancer 2003;98:2467–72.
1181
[10] Gourin CG, Podolsky RH. Racial disparities in patients with head and neck squamous cell carcinoma. Laryngoscope 2006;116:1093–106. [11] Moore RJ, Doherty DA, Do KA, Chamberlain RM, Khuri FR. Racial disparity in survival of patients with squamous cell carcinoma of the oral cavity and pharynx. Ethn Health 2001;6:165–77. [12] Surveillance, Epidemiology, and End Results (SEER) Program: Research Data (2007–2010). National Cancer Institute, DCCPS, Surveillance Research Program, Surveillance Systems Branch, based on November 2012 SEER data submission, posted to the SEER web site. February 2014.