Patterns of Multidisciplinary Care of Head and Neck Squamous Cell Carcinoma in Medicare Patients

Key Points

Question  Is multidisciplinary care routinely practiced in the treatment of Medicare patients with head and neck cancer?

Findings  This cohort study of 28 293 patients with squamous cell carcinoma found modest rates of multidisciplinary care use, especially in the setting of advanced-stage disease, in which most patients received adjuvant therapy. Multidisciplinary care was associated with longer time to initiation of definitive therapy and was most common among patients treated by initial resection.

Meaning  Multidisciplinary care is relatively uncommon among patients treated with initial surgery, regardless of tumor stage, although a significant proportion received adjuvant therapy.

Importance  Multidisciplinary care (MDC) yields proven benefits for patients with cancer, although it may be underused in the complex management of head and neck squamous cell carcinoma (HNSCC).

Objective  To characterize the patterns of MDC in the treatment of HNSCC among elderly patients in the US.

Design, Setting, and Participants  This nationwide, population-based, retrospective cohort study used Surveillance, Epidemiology, and End Results (SEER)–Medicare linked data from January 1, 1991, to December 31, 2011, to identify patients 66 years or older diagnosed with head and neck cancer and determine the dates of diagnosis, oncology consultations, treatment initiation, and speech therapy evaluation in addition to MDC completion. Multidisciplinary care was defined in a stage-dependent manner: localized disease necessitated consultations with radiation and surgical oncologists, and advanced-stage disease also included a medical oncology consultation, all before definitive treatment. Data were analyzed between December 2016 and September 2020.

Main Outcomes and Measures  Rates of MDC across all subsites of head and neck cancer as measured by the presence of an evaluation for each oncologist on the MDC team and its effect on treatment initiation.

Results  This cohort study assessed 28 293 patients with HNSCC (mean [SD] age, 75.1 [6.6] years; 67% male; 87% White) from the SEER-Medicare linked database. The HNSCC subsites included larynx (40%), oral cavity (30%), oropharynx (21%), hypopharynx (7%), and nasopharynx (2%). Overall, the practice of MDC significantly increased over time, from 24% in 1991 to 52% in 2011 (P < .001). For patients with localized (stage 0-II) tumors, 60% received care in the multidisciplinary setting, whereas 28% of those with advanced-stage disease did. A total of 18 181 patients (64%) were treated with initial definitive nonsurgical therapy across all stages. Regardless of stage and subsite, few patients (2%) underwent evaluation by a speech-language pathologist before definitive therapy. Multidisciplinary care prolonged the time to initiation of definitive treatment by 11 days for localized disease and 10 days for advanced disease.

Conclusions and Relevance  This cohort study found that most elderly patients with localized HNSCC received MDC, whereas few patients with advanced-stage disease received such care, although a significant proportion received adjuvant therapy. Multidisciplinary care may prolong time to initiation of definitive treatment with an uncertain impact. Consultation with a speech-language pathologist before definitive therapy was rare.

Multidisciplinary care (MDC) promotes patient-centered care and facilitates the increasingly complex decision-making processes embodied in modern oncology. For many cancers, the MDC approach includes benefits such as increased patient satisfaction, improved diagnostic accuracy, a more precise staging and treatment approach, and improved oncologic outcomes.^1-22 In 2014, the European Partnership for Action Against Cancer published a policy statement defining the core elements of organization, database management, and a patient-centered approach for MDC.²³ However, no formal recommendations exist in the US.

Patients with head and neck squamous cell carcinoma (HNSCC) may benefit considerably from MDC. Potentially curative therapy often requires the collaborative efforts of physician teams (eg, head and neck surgeons, radiation oncologists, medical oncologists, reconstructive surgeons, oral and maxillofacial surgeons, and physiatrists) and specialized ancillary services (eg, speech therapy, physical therapy, nutrition, and dental care). Moreover, a considerable proportion of patients treated with single-modality therapy (localized oral, laryngeal, or oropharyngeal cancers) have more than a single, accepted definitive treatment option with different modalities (surgery or radiation). These curative modalities vary substantially in the nature of their delivery, patient affinity, and adverse effect profiles. Ideally, individualized treatment decisions should be made with equipoise through multidisciplinary team–based care.

The extent to which patients with HNSCC receive MDC in the US has not been previously described. Few studies have evaluated the practice of MDC among patients with HNSCC in the US, and these studies^20,24,25 have been limited to single-institution experiences. A UK retrospective analysis found an association between MDC and improved survival after controlling for patient age, cancer stage, and year of diagnosis and increased use of multimodality therapy options,²⁶ and a recent Triological Society Best Practice guideline²⁷ recommends that “all H&N [head and neck] cancer patients requiring single or multimodality therapy for advanced H&N cancer should be presented at MDT [multidisciplinary team] conference” secondary to all reported literature supporting improved survival with the practice. We sought to characterize the patterns of MDC in US Medicare patients diagnosed with HNSCC.

In this cohort study, we used the Surveillance, Epidemiology, and End Results (SEER)–Medicare linked data set from January 1, 1991, to December 31, 2011. The SEER system represents a collaborative effort of the National Cancer Institute, SEER registries, and Centers for Medicare & Medicaid Services. The 18 registries account for approximately 28% of the US population and collect information related to new cancer diagnoses obtained from multiple sources. Medicare claims data provide information on enrollment and details of inpatient, outpatient, home health, hospice, and skilled nursing facility services use and associated International Classification of Diseases, Ninth Revision (ICD-9) diagnosis and procedure codes. Fox Chase Cancer Center Institutional Review Board and National Cancer Institute approvals were obtained. Informed consent for this database was waived according the SEER Data Use Agreement. The study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline. We identified patients 66 years and older who were diagnosed with nonmetastatic squamous cell carcinoma of the oral cavity (excluding lip), oropharynx, hypopharynx, nasopharynx, and larynx. All patients were enrolled in Medicare Part A and Medicare Part B. Exclusion criteria included non–squamous cell histologic subtype, diagnosis at death or autopsy, discharge to hospice, enrollment in managed care within 1 year before or after diagnosis, or receipt of Medicare benefits because of end-stage renal disease.

Current Procedural Terminology, Healthcare Common Procedure Coding System, ICD-9, and Health Care Finance Administration practitioner specialty codes were used to identify consultation or follow-up visits with a specific practitioner type. Current Procedural Terminology, Healthcare Common Procedure Coding System, and ICD-9 procedure codes were used to identify diagnostic and therapeutic procedures for primary site or lymph nodes, surgical resection and lymphadenectomy, radiation planning and delivery, chemotherapy administration, and speech therapy evaluation and treatment (eTables 1-9 in the Supplement).

The first biopsy or pathology billing code defined the date of diagnosis, if available. Otherwise the first HNSCC ICD-9 diagnosis code used for a given patient defined the diagnosis date. In either circumstance, the diagnosis date had to occur in the same or subsequent month as the SEER month of diagnosis. The time to treatment initiation was based on this date and the start of treatment. To approximate a uniform staging system for comparison, historical staging criteria were converted to American Joint Committee on Cancer’s Cancer Staging Manual, 7th edition criteria.

Data were analyzed between December 2016 and September 2020. Multidisciplinary care was defined as a postpathologic diagnosis, and predefinitive treatment evaluation was performed by an otolaryngologist, oral surgeon, head and neck surgeon, and radiation oncologist for stage 0 to II cancer. For stage III to IV cancer, MDC also required evaluation by a medical oncologist. Because this analysis uses 7th edition staging, stages 0 to II were almost entirely node negative T1/T2 stage designation primary tumors thought to be adequately described as localized tumors. However, given the heterogeneity of the stage III to IV designation (could describe either a T1N1 or T4N3 tumor), we elected to evaluate the application of MDC to stage III and stage IV separately. Although many patients with HNSCC are appropriately treated primarily by an operation, a short window between an operation and the ideal initiation of adjuvant therapy (4-8 weeks) suggests that some degree of planning for adjuvant therapy by both the patient and physician occurs before the operation.²⁶ We separately considered evaluation by a speech-language pathologist (SLP) (B.E.) before definitive therapy in subset analyses (MDC and SLP). Visits could occur on the same or different days. Treatment initiation was defined as surgical resection or initiation of (chemo)radiation therapy (chemotherapy, radiation therapy, or both [chemoradiation therapy]) within 365 days of diagnosis.

Patients were categorized into 6 groups according to treatment received: definitive surgery, surgery and adjuvant radiation therapy, surgery and adjuvant chemoradiation therapy, definitive radiation therapy, definitive sequential chemoradiation therapy, and definitive concurrent chemoradiation therapy. Covariates included age, sex, race/ethnicity, marital status, SEER geographical area, census tract–level education, census tract–level income, tumor site, and tumor stage.

Time to definitive treatment initiation (TTI-D) was defined as the interval between diagnosis and surgical resection or initiation of (chemo)radiation therapy. Time to adjuvant treatment initiation (TTI-A) was defined as that interval from surgical resection to start of adjuvant (chemo)radiation therapy.

Cohort characteristics were summarized using descriptive statistics. Univariable comparisons were made using χ² tests and 2-tailed, paired t tests as appropriate. Logistic regressions were used to investigate factors associated with receipt of MDC and trends in MDC use by year. A 2-sided P < .05 was considered statistically significant. Data management was performed using SAS software, version 9.4 (SAS Institute Inc), and statistical analyses were conducted using Stata software, version 15 (StataCorp LLC).

We identified 28 293 eligible Medicare patients with HNSCC (mean [SD] age, 75.1 [6.6] years; 67% male; 87% White). Patient characteristics are provided in Table 1. The subsites included larynx (40%), oropharynx (21%), oral cavity (30%), hypopharynx (7%), and nasopharynx (2%). A total of 18 181 patients (64%) were treated with initial definitive nonsurgical therapy. Generally, the practice of MDC increased over time, from 24% in 1991 to 52% in 2011 (P < .001) (Figure).

Regardless of stage and subsite, few patients (2%) underwent SLP evaluation or received MDC and SLP care before initiation of definitive therapy (eTables 1-8 in the Supplement). Pretreatment SLP evaluation was most common among patients with HNSCC of the larynx and hypopharynx, but patients who underwent SLP evaluation still represented fewer than 3% of total laryngeal cases and 5% of total hypopharyngeal cases. Ultimately, 17% of patients underwent SLP evaluation within 1 year of treatment initiation, including those who had evaluation before treatment.

A total of 14 486 patients had localized (stage 0-II) tumors, of whom 64% were treated simultaneously by a radiation oncologist and surgeon, whereas 60% received MDC (Table 2). Patients with localized tumors were largely treated with definitive radiation therapy (54%) or primary surgery (36%). Definitive chemoradiation therapy (9%) and sequential chemoradiation therapy (1%) were used less often in this localized cohort. The most common subsites were the larynx (51%) and oral cavity (35%). Patients who had MDC were most likely to undergo primary radiation therapy (93%) and have a primary cancer of the larynx (70%); those who did not experience MDC were predominately treated with surgery (81%) and had a primary cancer of the oral cavity (67%). A total of 36% of all patients with localized tumors were treated with initial surgery; of this subset, 12% received MDC. Of patients with resected tumors, 20% received adjuvant (chemo)radiation therapy.

A total of 5083 patients had localized oral cavity SCC, of whom 81% underwent surgical resection. Among patients with localized oral cavity, 27% consulted a radiation oncologist before definitive therapy, whereas 25% experienced MDC. Ultimately, 17% received adjuvant radiation therapy and 3% received chemoradiation therapy.

Primary laryngeal and oropharyngeal cancers represented 60% of the localized tumor cohort. Of these patients, 88% with laryngeal cancer and 65% with oropharyngeal cancer met with a radiation oncologist before definitive therapy. Surgery was the primary treatment for 9% of the patients with localized cancer of the larynx and 36% of those with localized cancer of the oropharynx, whereas definitive radiation was used in 83% of those with localized cancer of the larynx and 45% of those with localized cancer of the oropharynx. Definitive chemoradiation therapy was also used in 7% of laryngeal tumors and 18% of oropharynx tumors and sequential chemoradiation therapy in 1% of laryngeal tumors and 2% of oropharyngeal tumors. Multidisciplinary care, including collaborative care with surgeons and radiation oncologists, was experienced by 83% of those with laryngeal cancer and 58% with oropharyngeal cancer.

A total of 13 807 patients had advanced-stage (stage III-IV) tumors, 60% of whom had stage IV disease. Of the advanced-stage cohort, 61% consulted both a surgeon and radiation oncologist, whereas 28% of patients with either stage III or stage IV disease received MDC (Table 2 and eTables 1-8 in the Supplement). Similar proportions of the patients with advanced-stage disease were treated with primary chemoradiation or radiation therapy (of patients with stage III disease, 31% were treated with primary chemoradiation therapy and 33% with radiation therapy, and of patients with stage IV disease, 29% were treated with primary chemoradiation therapy and 27% with radiation therapy). The most common subsites were larynx (27% for stage III disease and 30% for stage IV disease) or oropharynx (38% for stage III disease and 31% for stage IV disease). Patients receiving MDC were rarely treated surgically (5% for stage III disease and 8% for stage IV disease); instead most were treated with primary chemoradiation therapy (63% for stage III disease and 60% for stage IV disease), radiation therapy alone (22% for stage III disease and 21% for stage IV disease), or sequential chemotherapy and radiation therapy (10% for stage III disease and 11% for stage IV disease). For those patients who did not receive MDC, definitive management included definitive surgery (39% for stage III disease and 50% for stage IV disease), primary radiation therapy alone (38% for stage III disease and 29% for stage IV disease), chemoradiation therapy (19% for stage III disease and 18% for stage IV disease), or sequential chemotherapy and radiation therapy (4% for stage III disease and 3% for stage IV disease). Of the 35% of advanced tumors managed with definitive surgery, 23% of patients (22% for stage III disease and 24% for stage IV disease) consulted with a radiation oncologist before resection, and 6% met with both a radiation oncologist and a medical oncologist, thereby meeting the criteria for MDC. Among all patients with advanced-stage disease who were treated with definitive surgery, 66% received adjuvant (chemo)radiation therapy, accounting for 62% of the stage III cohort and 69% of the stage IV cohort.

Oral cavity cancers represented 25% of the advanced-stage cohort; 12% of those with stage III disease and 15% of those with stage IV disease received MDC. Initial resection was pursued in 66% of patients, and 46% of them completed a consultation with a radiation oncologist before surgery. Adjuvant therapy was used in 61% of these patients. Definitive radiation therapy was used in 21%, chemoradiation therapy in 11%, and sequential chemotherapy and radiation therapy in 2% of advanced oral cavity cancers.

Cancers of the larynx and oropharynx represented 63% of patients with advanced-stage tumors. The initial treatment for both laryngeal and oropharyngeal primary tumors collectively was definitive chemoradiation therapy (35%) or definitive radiation therapy (32%). Surgery (27%) and sequential chemotherapy and radiation therapy (6%) were also used but less frequently. Together, 73% of patients with advanced-stage laryngeal and oropharyngeal cancer met with a radiation oncologist before definitive treatment. Rates of MDC were 28% for both stage III and stage IV cancer of the larynx, 32% for stage III cancer of the oropharynx, and 38% for stage IV cancer of the oropharynx.

Factors associated with provision of MDC on multivariable analysis are summarized in Table 3. Increased probability of MDC was associated with male sex (odds ratio [OR], 1.16; 95% CI, 1.08-1.24), age older than 70 years (age of 70-74 years: OR, 1.14; 95% CI, 1.06-1.23), nasopharyngeal primary tumor (OR, 1.34; 95% CI, 1.09-1.65), and Charlson Comorbidity Index greater than 1 (OR, 1.13; 95% CI, 1.05-1.21). Lack of MDC was associated with oropharyngeal (OR, 0.79; 95% CI, 0.70-0.89) and oral cavity primary tumors (OR, 0.13; 95% CI, 0.11-0.14), living outside a big metropolitan area (less urban or rural²⁸: OR, 0.82; 95% CI, 0.73-0.91), and having advanced-stage disease (OR, 0.16; 95% CI, 0.15-0.17) for stage IV disease.

Receipt of MDC was associated with more than a 10-day increase in median TTI-D for all patients (34 days [95% CI, 33.02-34.22 days] vs 24 days [95% CI, 23.79-24.75 days] for the non-MDC cohort). Multidisciplinary care prolonged TTI-D by 11 days for localized disease and 10 days for both stage III and stage IV disease (Table 4). Notably, 9% of patients underwent a simultaneously diagnostic and therapeutic surgical procedure, thus obviating the possibility of delivering MDC.

The TTI-D varied by treatment modality and was most pronounced for patients treated with surgery, for whom the mean (SD) TTI-D was 46 (46) days with MDC compared with 23 (31) days without MDC, corresponding with a mean (SD) increase of 23 (15) days. Other treatment modalities (eg, definitive radiation therapy and definitive chemoradiation therapy) when administered with MDC also had significant TTI-D prolongation (definitive radiation: 32 days [95% CI, 31.48-33.05 days] with MDC vs 28 days [95% CI, 26.50-28.55 days] without) (Table 4), with mean (SD) increases ranging from 4 (3) to 11 (5) days.

In contrast to longer TTI-D, the TTI-A was not changed by receipt of MDC. Of 10 112 patients treated with initial surgery, 4278 patients (42%) received adjuvant (chemo)radiation therapy. Multidisciplinary care was pursued in only 9% of those who received adjuvant therapy. Multidisciplinary care completion did not alter TTI-A for adjuvant radiation therapy (mean [SD] of 36 [19] days with MDC vs 37 [19] days without) or chemoradiation therapy (mean [SD] of 34 [19] days with MDC vs 33 [19] days without) (Table 4). These intervals to treatment initiation are within accepted guidelines²⁹ that necessitate initiation of adjuvant treatment within 4 to 6 weeks.

In this nationwide, population-based, retrospective cohort study, elderly patients with localized HNSCC were likely to be seen by a surgeon and radiation oncologist, but few of those with advanced-stage disease met all members of the MDC team before initiation of definitive therapy. Management of HNSCC in elderly patients, who often have substantial comorbidity, can be challenging. With other cancers, multidisciplinary approaches improve staging, inform therapy selection, and manage treatment and disease sequelae. These approaches also improve patient satisfaction and oncologic outcomes.^1-22 A prospective analysis²⁵ of a head and neck tumor board found that this intervention alters the diagnosis, workup, staging, or treatment in approximately one-quarter of patients. Among this group, two-thirds of the modifications directly affected the treatment plan.²⁵

Our analysis was not able to evaluate treatment changes as a result of MDC but instead sought to characterize the frequency of MDC and trends in use, as demonstrated by new patient consultation visits with oncologic subspecialists before initiation of definitive therapy. Overall, MDC for the elderly US population increased during the 2 decades analyzed, supporting another analysis.²⁶

We found similar proportions of patients with localized and advanced-stage disease who were seen by a surgeon (64%) and radiation oncologist (67%) in the interval between diagnosis and initiation of definitive therapy. Because patients with oral cavity cancer may be primarily treated by the same specialty that undertakes the initial biopsy, whereas patients with pharyngeal or laryngeal cancers may not be treated by primary surgery, we evaluated these 2 situations separately.

Most patients with an oral cavity cancer are initially referred to a surgeon for biopsy and/or management. Consistent with guidelines,²⁹ patients with localized and advanced-stage oral cavity cancer in this series were primarily treated initially with an operation (81% with localized cancer and 66% with advanced-stage cancer). Perhaps these patients would stand to benefit the least from MDC—the initial member of the treatment team represents the specialist best suited to deliver initial therapy, and our data indicate that MDC lengthened treatment initiation by a median of 18 days (range, 17-35 days).

However, head and neck cancer is often managed with multiple modalities: 17% of patients with localized oral cavity cancer and 61% of patients with advanced-stage oral cavity cancer received adjuvant (chemo)radiation therapy, a few of whom (25% of patients with localized disease and 14% of patients with advanced-stage disease) received MDC before the operation. This finding might be expected to be detrimental in head and neck cancer because of prolongation of treatment package time^30-33; however, the receipt of MDC did not reduce TTI-A. The median time to start adjuvant (chemo)radiation therapy suggests that limited impact from MDC is not attributable to altered treatment starts among patients who received MDC but rather extraordinarily timely treatment initiation for patients who did not have MDC before surgery (median of 29 days for [chemo]radiation therapy). Because the interval between simulation and treatment start is typically approximately 14 days, many patients are meeting adjuvant therapy practitioners for the first time and consenting to therapy within 2 weeks after a head and neck operation. With the intention to promptly initiate adjuvant therapy, a patient’s ability to comprehend the risks and benefits of (chemo)radiation therapy may be limited. This occurrence may constitute 1 of the reasons that MDC is associated with improved patient satisfaction^34-36—patients learning for the first time (while still recovering from the immediate effects of an operation) that they would benefit from additional, time-sensitive therapy likely feel more unsettled than those patients who understand that a recommendation for adjuvant therapy may follow an operation.

Although some maintain that MDC results in unnecessary consultations for patients with oral cavity cancer destined to be treated with surgery alone, preparing patients for additional treatment that is ultimately determined to be unnecessary may be preferable to explaining shortly after an operation that an additional 6 to 7 weeks of adjuvant therapy is needed. However, the time to complete MDC was longer than necessary in this data set. Multidisciplinary care should not increase treatment initiation by weeks. At Fox Chase Cancer Center, it is common for patients with oral cavity cancer to meet with a radiation oncologist within days of the initial head and neck surgery consultation or, alternatively, the same day as preadmission testing and/or imaging. With this approach, the MDC does not unnecessarily prolong treatment initiation. Alternatively, the increased TTI for patients with MDC may not be referable to the pursuit of MDC consultations but rather to other factors not captured in this analysis, including increased time for surgical planning for more complex surgical cases (that are expected to require adjuvant therapy preoperatively). Although differentiation between the causes of TTI prolongation is beyond the scope of this submission, treatment teams can and should coordinate care so that such delays are eliminated.^37-39

Most patients with pharyngeal and laryngeal cancer initially consult with a surgeon for biopsy and/or disease management. In contrast to oral cavity cancer, many of these patients in this series were primarily treated by radiation therapy, and, consequently, the likelihood of MDC was greater for these disease sites. However, a significant percentage of patients did not receive MDC before treatment initiation, including 43% of patients with localized oropharyngeal cancer and 17% of patients with localized laryngeal cancer. In the stage III to IV setting, the likelihood of MDC actually decreased; 65% of patients with advanced oropharyngeal cancer and 72% of patients with advanced laryngeal cancer did not receive MDC. Whether this is a legitimate finding or an artifact of coding inconsistencies, as is commonly seen in SEER-Medicare analyses, is unclear.^40,41 Even with adjustment for anticipated error in this setting, MDC was used substantially less often than anticipated.

Although the comments regarding patient satisfaction with MDC for oral cancers apply to pharyngeal and laryngeal tumors as well, patient choice also plays a role in the management of pharyngeal and laryngeal tumors. Although randomized comparisons are small,^42-44 data suggest that treatment with surgery or radiation therapy results in similar efficacy but different adverse effect profiles in appropriately selected patients with pharyngeal and laryngeal cancer. Where randomized comparisons are not available, retrospective analyses similarly suggest comparable outcomes for primary surgery and primary radiation therapy again with different adverse effect profiles,⁴⁵ although there are notable exceptions.⁴⁶ Thus, the primary treatment of laryngeal and pharyngeal cancer may be a matter of patient choice rather than application of a preferred management technique. Patients cannot make informed choices between different primary management modalities without meeting appropriate specialists. Patients treated with primary surgery may be surprised to hear that adjuvant radiation therapy courses are similar in length and adverse effects to definitive radiation therapy courses. Patients treated with primary radiation therapy should understand that failure of radiation therapy to eradicate their cancer may result in incurable disease or a recommendation for surgical resection (which will be more technically demanding and morbid in the post–radiation therapy setting than it would have been as primary treatment).

Pretreatment SLP evaluation, provided to only 2% of the patients with HNSCC in this analysis, is a valuable tool that can inform the management of the complex and multifactorial deficits of speech and swallowing dysfunction that can accompany head and neck cancer. Both retrospective and prospective assessments of pretreatment swallowing using videofluoroscopy have found greater dysfunction among patients with head and neck cancers, worse with increasing stage.^47-49 This dysfunction is often asymptomatic, making clinical judgment unreliable and instrumental swallowing evaluation (ie, videofluoroscopic swallowing study or flexible endoscopic evaluation of swallowing) preferable.⁴⁹ Identifying pretreatment dysphagia facilitates accurate patient and family counseling surrounding current and long-term function because these patients have a poorer functional prognosis than those without pretreatment dysphagia.⁴⁷ Furthermore, 2 randomized clinical trials^50,51 found that prophylactic swallowing exercises before and concurrently with definitive treatment for HNSCC can help improve posttreatment function for all patients. Unfortunately, as this analysis demonstrates, few patients undergo pretherapy evaluation of speech and swallowing function with an SLP; however, 17% of patients are ultimately referred for SLP evaluation within 1 year of diagnosis. Assessment with an SLP should be integrated into the comprehensive care of patients with HNSCC.

This study has limitation, including use of a large national database. General inconsistencies are implicit to these data sets, thought to be related to consultation or treatment omissions or inaccurate coding because of a reliance on claims data. These inconsistencies are responsible for 1% to 5% of error in other SEER-Medicare data sets.^40,41 This error rate is consistent with our analysis; a small proportion of patients, ranging from 1% to 5%, received definitive surgery in the absence of consultation with an otolaryngologist or head and neck surgeon. Consultations with practitioners as shown in claims data may not indicate true coordination of care providing the benefits of a multidisciplinary approach. Tumor board discussions, another form of MDC, are not captured by consultations. Other limitations include that the Medicare population consists of only individuals older than 65 years and only pathologic staging is available for patients treated with surgery, which obscures the need for MDC because the preoperative clinical stage is unknown. The localized and advanced-stage groups certainly include surgically treated patients with a different clinical and pathologic stages. In addition, we defined different stage-dependent MDC rules, requiring that a medical oncologist have consultations with all patients with advanced-stage disease before initiating definitive therapy to meet MDC criteria. The SEER database does not include detailed radiation therapy information, such as number of fractions, dosage, or technique used, and does not detail chemotherapy regimens and cycles. For all treatment modalities, we assumed potentially curative intent for treatment occurring within 1 year of the diagnostic procedure establishing a cancer diagnosis.

Most elderly patients with localized HNSCC are seen by a surgeon and radiation oncologist, but few patients with advanced-stage disease meet all members of the MDC team (including a medical oncologist) before initiation of definitive therapy. In this study, MDC was relatively uncommon among patients treated with initial surgery, regardless of tumor stage, although 20% of patients with localized disease and 66% with advanced-stage disease received adjuvant therapy. Patients with oropharyngeal or oral cavity tumors are less likely to receive MDC. Consultation with an SLP before definitive therapy for HNSCC is rare. Although this examination identified an encouraging increase of MDC use in the time frame evaluated, the moderate rate of MDC at the end of the study period (52%) and the infrequency of pretreatment evaluation by an SLP in particular represent an opportunity to improve the use of MDC for future patients with HNSCC.

Accepted for Publication: August 21, 2020.

Published Online: October 22, 2020. doi:10.1001/jamaoto.2020.3496

Corresponding Author: Thomas J. Galloway, MD, Department of Radiation Oncology, Fox Chase Cancer Center, 333 Cottman Ave, Philadelphia, PA 19111 (thomas.galloway@fccc.edu).

Author Contributions: Dr Galloway had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Hansen, Egleston, Leachman, Churilla, Liu, Ridge, Galloway.

Acquisition, analysis, or interpretation of data: Hansen, Egleston, Leachman, DeMora, Ebersole, Bauman, Liu, Ridge, Galloway.

Drafting of the manuscript: Hansen, Egleston, Leachman, Liu, Galloway.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Hansen, Egleston, Leachman, DeMora.

Obtained funding: Egleston, Ridge, Galloway.

Administrative, technical, or material support: Hansen, Egleston, Galloway.

Supervision: Liu, Ridge, Galloway

Conflict of Interest Disclosures: Dr DeMora reported receiving personal fees from the American College of Radiology outside the submitted work. Dr Bauman reported receiving financial support from Pfizer, AstraZeneca, Kura, and Bayer and grants from BMS outside the submitted work. Dr Galloway reported receiving personal fees from Varian Medical Systems outside the submitted work. No other disclosures were reported.

Meeting Presentations: Presented in part as a poster at the American Society for Radiation Oncology Annual Meeting; September 24-27, 2017; San Diego, California; Multidisciplinary Head and Neck Symposium; February 15-17, 2018; Scottsdale, Arizona; and American Society for Radiation Oncology Annual Meeting; October 21-24, 2018; San Antonio, Texas.