Racial disparities in outcome among head and neck cancer patients in the United States: An analysis using SEER-Medicare linked database.

Abstract

6051 Background: Despite overall decline in cancer mortality, African Americans suffer from higher mortality in most cancer types including cancers of the head and neck. These differences likely result from a complex interplay of clinical and non-clinical factors. We aim to estimate disparities in overall survival across racial groups in HNSCC in the United States. Methods: This study used SEER-Medicare linked database. We identified all patients aged 66 years or older diagnosed with HNSCC as their first cancer from 1992 to 2011. We excluded those in HMO, diagnosed by death certificate or autopsy, non-SCC, unknown race, and missing month and/or year of diagnosis. Further exclusions included metastatic disease, salivary gland cancers, receiving no treatment in the first 180 days, and unknown stage. Analytic data set included oropharynx, oral cavity, nasopharynx, hypopharynx, and larynx. Primary treatment was defined as any treatment modality received within 180 days after diagnosis. Overall survival (OS) parameters were estimated across ethnic groups by the Cox regression model stratified by site and stage of cancer at diagnosis, adjusted for clinical and demographic characteristics, and propensity score weighted. Results: Our study population included 15, 547 patients. Median OS was 3.5 years (95% CI: 3.4-3.7) across all ethnic groups. African Americans (AA) had inferior outcome with median OS of 2.0 years (95% CI: 1.9-2.3) compared to 3.7 years (95% CI: 3.6-3.8) for Caucasian Americans (CA) (p < 0.0001). This difference was seen despite AA patients receiving comparable treatments and presenting at similar stage of disease, except for cancers of the oral cavity where AA were more likely to present with advanced disease (67% versus 47%; P < 0.001). The difference was most pronounced in the oropharynx where median OS was 1.9 years (95% CI: 1.7-2.1) for AA and 3.8 years (95% CI: 3.5-4.1) in CA (P < 0.0001). AA also had consistently worse OS over time from 1992 to 2011. This study clearly demonstrated AA have inferior outcomes despite similar treatments, comorbidities, age at diagnosis, stage at presentation, tumor location, year of diagnosis and sex. Conclusions: The current study demonstrates inferior overall survival for African American head and neck cancer patients independent of primary site and treatment modalities.

Explaining Racial Disparities in Surgically Treated Head and Neck Cancer

“…Racial disparities in HNSCC survival have been well characterized, 3–11 however, the relative contribution of the factors that drive such disparities have not been fully explored. Here, we found that the disparity in mortality rates is largely explained by differences in oncologic factors at presentation.…”

“…Previous studies have identified various racial disparities in survival for HNSCC 3–11 . For example, a recent study showed that black patients were at a 10% greater risk of mortality within 90 days after treatment of HNSCC 20 .…”

“…Racial disparities are well documented across multiple types of cancers, including head and neck squamous cell carcinoma (HNSCC) 1–4 . Black patients with HNSCC face worse prognoses, are less likely to receive the standard of care therapy, and have worse overall survival 4–11 . Some of the differences may be explained by comorbid conditions, differences in stage at presentation, socioeconomic status, and hospital factors 3,4,12 .…”

Role of physician density in predicting stage and survival for head and neck squamous cell carcinoma

“…We did find that PCP + D density was associated with a lower hazard of cancer specific mortality for overall HNSCC and for laryngeal cancer, a relationship that requires further evaluation given the lack of association on stage of presentation. Given known and significant disparities in the stage of presentation and survival for HNSCC, the availability of physicians has been described as a possible barrier to timely diagnosis 17‐19 . Our results suggest that provider density may be less important than other barriers to care, such as financial access to a physician via insurance status.…”

“…of presentation and survival for HNSCC, the availability of physicians has been described as a possible barrier to timely diagnosis. [17][18][19] Our results suggest that provider density may be less important than other barriers to care, such as financial access to a physician via insurance status.…”