Impact of Facility Volume on Overall Survival in Patients With Head and Neck Cancer Undergoing Palliative Treatment

Abstract

Background

Treatment at high-volume facilities (HVF) has been associated with improved prognosis of HNC patients undergoing curative treatment. Whether this systemic factor influences survival outcomes of patients with HNC undergoing palliative treatment is unknown.

Aim

To investigate the impact of palliative treatment facility volume on overall survival (OS) in patients with head and neck cancer (HNC).

Design

The 2004 to 2018 National Cancer Database was queried retrospectively for patients with HNC undergoing palliative treatment.

Setting/participants

Patients were stratified based on treatment facility volume percentile. Multivariable binary logistic and Cox proportional hazards regression models were implemented.

Results

Of 8682 patients included, 1661 (19.1%) underwent palliative therapy at facilities with volume ≥80^th percentile. Among 972 facilities included, 643 (66.2%), 182 (18.7%), 85 (8.8%), 44 (4.5%), and 18 (1.9%) had volume <20^th, 20-40^th, 40-60^th, 60-80^th, and ≥80^th percentiles, respectively. 5-year OS rates of patients undergoing palliative therapy at facilities with volume <20^th, 20-40^th, 40-60^th, 60-80^th, and ≥80^th percentile was 11%, 13%, 11%, 14%, and 23%, respectively (P < .001). Facility volume ≥80^th percentile was associated with higher 5-year OS on multivariable Cox regression (aHR 0.34, 95% CI 0.16-0.69, P < .001). Surgical treatment (aOR 1.34, 95% CI 1.07-1.68, P = .012) was associated with undergoing treatment at facilities with volume ≥80^th percentile.

Conclusions

Undergoing palliative treatment at HVFs is associated with higher OS in HNC. The survival benefit derived from high facility volume should be carefully considered in the context of other patient and facility characteristics in end-of-life management, with specific emphasis on patient-directed goals of care.

Keywords

national cancer database head and neck cancer facility volume survival palliative treatment end-of-life 4

Introduction

Head and neck cancer (HNC) has a global incidence of approximately 890,000 cases per year, and is typically treated with surgery, radiotherapy, and/or chemotherapy.¹ The prognosis of HNC depends on tumor site, stage, and etiology, with an overall 5-year survival rate ranging from 40% to 65%.² However, an estimated 15%-20% of patients with HNC are not amenable to treatment with curative intent, and may benefit from palliative treatment, which focuses on providing symptom relief, comfort, and support to patients and their families.³ Studies across various cancer types have demonstrated that palliative treatment improves quality of life, depression, satisfaction with care, and survival for patients, and offers education on end-of-life caregiving, respite care, and psychological counseling for caregivers.^4-12

HNC imposes significant functional challenges through the disruption of essential functions including eating, breathing, and speaking, potentially rendering patients psychologically and socially distressed.^13-19 Palliative treatment modalities in HNC, including medical management, surgery, chemotherapy, radiation, and combination therapy, are intended to provide airway protection, nutrition,^20-23 tumor debulking, delay complications, and improve the overall end-of-life experience, and may also extend life in some cases.^24-26

Modifiable systemic factors such as treatment facility type and volume also impact patient survival in HNC. Several studies have demonstrated improved prognosis of HNC patients who undergo curative treatment at high-volume facilities (HVF), hinting that multidisciplinary teams with an array of specialists may help better manage the intricacies of HNC.^27-31 However, there is a lack of research assessing whether facility volume is a prognostic factor of HNC patients undergoing palliative treatment. Our study of the National Cancer Database (NCDB) investigates facility volume in palliative treatment management of HNC and associated differences in overall survival (OS).

Materials and Methods

Data Source

The NCDB is a national database jointly sponsored by the Commission on Cancer (CoC) of the American College of Surgeons and the American Cancer Society (ACS) that accounts for >70% of newly diagnosed cancer cases in the United States.^32-36 The NCDB has established criteria to ensure that submitted data meets specified quality benchmarks. The Rutgers New Jersey Medical School and University of Pennsylvania Institutional Review Boards exempted our study because of the de-identified nature of patient data. The ACS and CoC have not verified and are not responsible for the validity of data analysis or conclusions derived herein.

Inclusion Criteria

Our study sample was comprised of adults undergoing palliative treatment for histologically confirmed head and neck cancer (HNC) between 2004-2018 (Figure 1). The NCDB was queried for cases of HNC using International Classification of Diseases for Oncology, Third Edition (ICD-O-3) histology (conventional squamous cell carcinoma was defined as “8070-8072”), behavior (invasive: “3”), and topography (oral cavity: C00.0 [lip], C02.0-3 [anterior, overlapping, NOS tongue], C03 [gingiva], C04 [floor of mouth], C050, C05.8-9 [hard, overlapping, NOS palate], C06 [other, NOS mouth; major salivary glands: C07 [parotid gland], C08 [others, NOS]; sinonasal tract: C11 [nasopharynx], C30.0 [nasal cavity], C31 [accessory sinuses]; oropharynx: C01 [base of tongue], C02.4 [lingual tonsil], C05.1-05.2 [soft palate, uvula], C09 [tonsil],C10 [oropharynx]; hypopharynx: C12 [pyriform sinus], C13 [hypopharynx]; larynx: C32.0 [glottis], C32.1 [supraglottis], C32.2 [subglottis], C32.3, 32.8-9 [other, overlapping, NOS]) codes. Patients were included if they were determined by the NCDB to have undergone palliative treatment, defined as any care intended to “control symptoms, to alleviate pain, or to make the patient more comfortable.”³⁷ Patients were excluded if they underwent palliative treatment at >1 reporting facility or a facility other than the reporting facility. Patients undergoing treatment with curative intent, and those with unknown vital status or survival time were excluded from analysis. Patients referred to palliative treatment but not documented to undergo palliative treatment were also excluded.

Figure 1.

Inclusion criteria. Abbreviations. NCDB, National Cancer Database.

Variables

The following categorical variables were collected for each patient cohort: sex (male, female), race (White, Black, other), primary payer status (no insurance, private insurance, Medicaid, Medicare), median household income of residence zip code (<$40,227, $40,227-50,353, $50,354-63,332, ≥$63,333), population density (metro, urban/rural), facility type (academic/research, non-academic/research), facility location (New England, Middle and South Atlantic, Central, Mountain/Pacific), Charlson-Deyo comorbidity score (CDCS) (0, ≥1), history of prior malignancy, histology (conventional squamous cell carcinoma, other), primary site (oral cavity, major salivary glands, sinonasal tract, oropharynx, hypopharynx, larynx), HPV status (HPV negative, HPV positive, unknown), grade (low, high, unknown), clinical tumor-nodal-metastasis (cTNM) classification, pathologic extranodal extension (pENE), lymphovascular invasion (LVI), and palliative treatment (surgery alone, radiotherapy alone, chemotherapy alone, pain management alone, combination therapy, other). We also reported the following continuous variables: age at diagnosis, palliative treatment facility volume, distance (mi) to surgical facility, and survival time. Median household income was determined by the median household income in a patient’s zip code based on the 2012 American Community Survey data. Population density was categorized as metropolitan (>250,000 individuals) or urban/rural (0-249,999 individuals). A CDCS of 0 indicates no comorbid diseases. An academic/research facility is defined by the NCDB as an institution that records >500 newly diagnosed cancer cases each year, participates in postgraduate medical education and cancer-related research, and offers the full range of diagnostic and treatment services on-site. Non-academic facilities pools together community (records >100 but <500 newly diagnosed cancer cases per year), comprehensive community (records >500 newly diagnosed cancer cases per year), and integrated network (no minimum caseload but owns, operates, leases, or is part of a joint venture offering comprehensive services) facilities.

The NCDB utilizes the following designations for palliative treatment: (1) “surgery (which may involve a bypass procedure) to alleviate symptoms, but no attempt to diagnose, stage, or treat the primary tumor is made;” (2) “radiation therapy to alleviate symptoms, but no attempt to diagnose, stage, or treat the primary tumor is made;” (3) “chemotherapy, hormone therapy, or other systemic drugs to alleviate symptoms, but no attempt to diagnose, stage, or treat the primary tumor is made;” (4) “patient received or was referred for pain management therapy with no other palliative care;” (5) any combination of surgery, radiation therapy, chemotherapy, and pain management therapy; and (6) “palliative care was performed or referred, but no information on the type of procedure is available in the patient record” or palliative treatment was provided but does not fit the aforementioned descriptions. Patients were stratified into 5 cohorts based on the volume percentile of the facility at which they underwent palliative treatment: <20^th, 20-40^th, 40-60^th, 60-80^th, and ≥80^th. The primary outcome of our study was 5-year OS. Survival time was calculated as the time from diagnosis to either death or 5 years of follow-up.

Statistical Analysis

Chi-square and Kruskal–Wallis H tests were used to compare categorical and continuous variables, respectively, by facility volume percentile. Multivariable binary logistic regression was performed to identify patient demographics, facility characteristics, cTNM classification, and treatment independently associated with undergoing palliative treatment at a facility with volume ≥80th percentile. The Hosmer-Lemeshow test was used to assess the calibration of the logistic regression model. 5-years OS was estimated using the Kaplan-Meier method and compared across cohorts using the log-rank test. A multivariable Cox proportional hazards regression adjusting for all significant variables in univariable Cox regression was used to identify factors independently associated with OS. 3 sensitivity analyses were performed: (1) facility volume was binarized at varying percentile thresholds, (2) 2-year and 3-year OS were utilized as primary outcomes to account for censorship and follow-up loss, and (3) sequential landmark survival analysis was performed for patients surviving $\geq$ 3, $\geq$ 6, and $\geq$ 12 months from diagnosis to account for immortal time bias and patients with favorable baseline prognosis having a higher likelihood of undergoing definitive local therapy. The 2-sided threshold for statistical significance was set at P < .05. Multiple comparisons were accounted for with the Bonferroni correction, and thresholds for significance were adjusted accordingly. SPSS version 25 (IBM) and Microsoft Excel (Microsoft) were utilized for statistical analysis.

Results

Patient Factors and Palliative Treatment Modalities

Of 8682 patients satisfying inclusion criteria, 2016 (33.2%), 1444 (16.6%), 1689 (19.5%), 1872 (21.6%), and 1661 (19.1%) underwent palliative therapy at facilities with <20^th, 20-40^th, 40-60^th, 60-80^th, and ≥80^th volume percentiles, respectively (Table 1). <20^th, 20-40^th, 40-60^th, 60-80^th, and ≥80^th percentiles included 643 (66.2%), 182 (18.7%), 85 (8.8%), 44 (4.5%), and 18 (1.9%) facilities, respectively. Median patient age at diagnosis (interquartile range [IQR]) was 65 (56-76) years; 6119 (70.5%) patients were male; 6881 (79.8%) were White; and most (80.5%) were from metropolitan locations. Most patients had no history of prior malignancy (72.2%), conventional squamous cell carcinoma histology (80.1%), unknown HPV status (80.1%), and low-grade malignancies (40.7%). Between 2004 and 2018, utilization of palliative surgery decreased sharply, whereas pain management increased (Figure 2).

Table 1.

Patient Demographics and Clinicopathologic Features by Facility Volume Percentile.

	<20^th	20-40^th	40-60^th	60-80^th	$\geq$ 80^th	P Valuea	Total
No. facilities	643 (66.2)	182 (18.7)	85 (8.8)	44 (4.5)	18 (1.9)	-	971 (100)
No. patients	2016 (23.2)	1444 (16.6)	1689 (19.5)	1872 (21.6)	1661 (19.1)	-	8682 (100)
Age at diagnosis, median years (IQR)	68.0 (58.0-79.0)	67.0 (58.0-78.0)	66.0 (58.0-77.0)	64.0 (56.0-74.0)	61.0 (53.0-70.0)	<.001	65.0 (56.0-76.0)
Sex
Male	1421 (70.5)	1013 (70.2)	1128 (66.8)	1334 (71.3)	1223 (72.6)	<.001	6119 (70.5)
Female	595 (29.5)	431 (29.8)	561 (33.2)	538 (28.7)	438 (26.4)		2563 (29.5)
Race
White	1645 (82.1)	1167 (81.4)	1337 (79.8)	1415 (76.2)	1317 (79.5)	<.001	6881 (79.8)
Black	287 (14.3)	212 (14.8)	275 (16.4)	367 (19.8)	239 (14.4)		1380 (16.0)
Other	71 (3.5)	54 (3.8)	63 (3.8)	75 (4.0)	100 (6.0)		363 (4.2)
Primary payer status
No insurance	110 (5.6)	86 (6.2)	90 (5.5)	142 (7.9)	110 (6.8)	<.001	538 (6.4)
Private insurance, managed care	391 (20.0)	305 (21.8)	348 (21.3)	424 (23.5)	456 (28.2)		1924 (22.9)
Medicaid	269 (13.8)	201 (14.4)	274 (16.8)	333 (18.5)	347 (21.5)		1424 (16.9)
Medicare	1185 (60.6)	805 (57.6)	918 (56.3)	905 (50.2)	703 (43.5)		4516 (53.7)
Household income
$<$ $40,227	422 (23.0)	315 (24.6)	410 (26.5)	526 (31.1)	410 (33.3)	<.001	2083 (27.5)
$40,227-50,353	423 (23.1)	350 (27.3)	346 (22.4)	395 (23.3)	297 (24.1)		1811 (23.9)
$50,354-63,332	434 (23.7)	291 (22.7)	382 (24.7)	346 (20.4)	291 (23.7)		1744 (23.0)
$\geq$ $63,333	552 (30.1)	327 (25.5)	408 (26.4)	425 (25.1)	232 (18.9)		1944 (25.6)
Population density
Metro	1618 (82.0)	1128 (80.2)	1378 (83.4)	1518 (84.0)	1162 (72.3)	<.001	6804 (80.5)
Urban, rural	355 (18.0)	279 (19.8)	274 (16.6)	290 (16.0)	446 (27.7)		1644 (19.5)
Facility type
Academic/research	223 (11.2)	393 (27.7)	895 (53.8)	1241 (67.7)	1237 (77.2)	<.001	3989 (46.9)
Non-academic/research	1764 (88.8)	1026 (72.3)	768 (46.2)	591 (32.3)	365 (22.8)		4514 (53.1)
Facility location
New England	123 (6.2)	123 (8.7)	173 (10.4)	77 (4.2)	216 (13.5)	<.001	712 (8.4)
Middle and South Atlantic	739 (37.2)	483 (34.0)	844 (50.8)	745 (40.7)	532 (33.2)		3343 (39.3)
Central	808 (40.7)	650 (45.8)	520 (31.3)	742 (40.5)	759 (47.4)		3479 (40.9)
Mountain, Pacific	317 (16.0)	163 (11.5)	126 (7.6)	268 (14.6)	95 (5.9)		969 (11.4)
Facility volume, median annual cases (IQR)	0.27 (0.20-0.40)	0.70 (0.60-0.87)	1.33 (1.13-1.60)	2.73 (2.33-3.27)	6.47 (5.4-14.67)	<.001	1.40 (0.53-3.20)
Distance to facility, median mi (IQR)	6.5 (2.8-14.4)	8.3 (3.4-19.4)	9.8 (3.9-24.6)	10.1 (4.2-30.1)	16.2 (6.6-44.2)	<.001	9.3 (3.7-23.8)
Charlson-Deyo comorbidity score
0	1423 (70.6)	976 (67.6)	1254 (74.2)	1366 (73.0)	1127 (67.9)	<.001	6146 (70.8)
$\geq$ 1	593 (29.4)	468 (32.4)	435 (25.8)	506 (27.0)	534 (32.1)		2536 (29.2)
History of prior malignancy
No	1446 (72.1)	1048 (73.1)	1222 (72.7)	1332 (71.6)	1186 (71.7)	.850	6234 (72.2)
Yes	559 (27.9)	386 (26.9)	459 (27.3)	529 (28.4)	469 (28.3)		2402 (27.8)
Histology
Conventional squamous cell carcinoma	1601 (79.4)	1143 (79.2)	1361 (80.6)	1462 (78.1)	1390 (83.7)	<.001	6957 (80.1)
Other	415 (20.6)	301 (20.8)	328 (19.4)	410 (21.9)	271 (16.3)		1725 (19.9)
Primary site
Oral cavity	352 (17.5)	255 (17.7)	343 (20.3)	420 (22.4)	374 (22.5)	<.001	1744 (20.1)
Major salivary glands	172 (8.5)	108 (7.5)	101 (6.0)	112 (6.0)	50 (3.0)		543 (6.3)
Sinonasal tract	238 (11.8)	183 (12.7)	206 (12.2)	249 (13.3)	167 (10.1)		1043 (12.0)
Oropharynx	654 (32.4)	441 (30.5)	495 (29.3)	568 (30.3)	562 (33.8)		2720 (31.3)
Hypopharynx	174 (8.6)	143 (9.9)	155 (29.3)	146 (7.8)	133 (8.0)		751 (8.7)
Larynx	426 (21.1)	314 (21.7)	389 (23.0)	377 (20.1)	375 (22.6)		1881 (21.7)
HPV status
HPV negative	247 (12.3)	171 (11.8)	234 (13.9)	257 (13.7)	170 (10.2)	.007	1079 (12.4)
HPV positive	140 (6.9)	92 (6.4)	129 (7.6)	145 (7.7)	146 (8.8)		652 (7.5)
Unknown	1629 (80.8)	1181 (81.8)	1326 (78.5)	1470 (78.5)	1345 (81.0)		6951 (80.1)
Grade
Low	747 (37.1)	565 (39.1)	668 (39.6)	777 (41.5)	773 (46.5)	<.001	3530 (40.7)
High	616 (30.6)	409 (28.3)	475 (28.1)	508 (27.1)	459 (27.6)		2467 (28.4)
Unknown	653 (32.4)	470 (32.5)	546 (32.3)	587 (31.4)	429 (25.8)		2685 (30.9)
cT classification
1	129 (6.4)	100 (6.9)	119 (7.0)	117 (6.3)	121 (7.3)	<.001	586 (6.7)
2	324 (16.1)	242 (16.8)	295 (17.5)	299 (16.0)	292 (17.6)		1452 (16.7)
3	441 (21.9)	293 (20.3)	330 (19.5)	345 (18.4)	330 (19.9)		1739 (20.0)
4	698 (34.6)	530 (36.7)	670 (39.7)	774 (41.4)	691 (41.6)		3364 (38.7)
x	424 (21.0)	279 (19.3)	275 (16.3)	336 (17.9)	227 (13.7)		1541 (17.7)
cN classification
0	558 (27.7)	415 (28.7)	461 (27.3)	526 (28.1)	476 (28.7)	<.001	2436 (28.1)
1	268 (13.3)	202 (14.0)	227 (13.4)	217 (11.6)	202 (12.2)		1116 (12.9)
2	696 (34.5)	525 (26.4)	669 (29.6)	752 (40.2)	686 (41.3)		3328 (28.3)
3	154 (7.6)	94 (6.5)	120 (7.1)	122 (6.5)	112 (6.7)		602 (6.9)
x	340 (16.9)	208 (14.4)	212 (12.6)	255 (13.6)	185 (11.1)		1200 (13.8)
cM classification
0	1168 (57.9)	893 (61.8)	1072 (63.5)	1194 (63.8)	1205 (72.5)	<.001	5532 (63.7)
1	644 (21.9)	420 (29.1)	474 (28.1)	502 (26.8)	305 (18.4)		2345 (27.0)
x	204 (10.1)	131 (9.1)	143 (8.5)	176 (9.4)	151 (9.1)		805 (9.3)
Pathologic extranodal extension
No	77 (72.0)	68 (74.7)	78 (63.9)	105 (65.2)	122 (64.2)	.280	450 (67.1)
Yes	30 (28.0)	23 (25.3)	44 (36.1)	56 (34.8)	689 (35.8)		221 (32.9)
Lymphovascular invasion
No	189 (76.5)	122 (66.3)	172 (76.4)	249 (76.4)	166 (63.1)	<.001	898 (72.1)
Yes	58 (23.5)	62 (33.7)	53 (23.6)	77 (23.6)	97 (36.9)		347 (27.0)
Palliative treatment
Surgery alone	223 (11.1)	179 (12.4)	190 (11.2)	298 (15.9)	462 (27.8)	<.001	1352 (15.6)
Radiotherapy alone	871 (43.2)	610 (42.2)	579 (34.3)	499 (26.7)	249 (15.0)		2808 (32.3)
Chemotherapy alone	357 (17.7)	225 (15.6)	326 (19.3)	258 (13.8)	199 (12.0)		1365 (15.7)
Pain management alone	221 (11.0)	149 (10.3)	251 (14.9)	380 (20.3)	412 (24.8)		1413 (16.3)
Combination therapy	182 (9.0)	148 (10.2)	188 (11.1)	244 (13.0)	189 (11.4)		951 (11.0)
Other	162 (8.0)	133 (9.2)	155 (9.2)	193 (10.3)	150 (9.0)		793 (9.1)

Abbreviations: cTNM, pathologic tumor-nodal-metastasis; IQR, interquartile range.

^a2-sided threshold for statistical significance is P < .005, as calculated with the Bonferroni correction.

Bold indicates statistical significance (P < 0.005)

Figure 2.

Trends in palliative treatment modalities between 2004 and 2018.

Older age (adjusted odds ratio [aOR] 0.98, 95% confidence interval [CI] 0.97-0.99, P < .001), non-academic/research facility (aOR 0.09, 95% CI 0.08-0.11, P < .001), radiotherapy alone (aOR 0.37, 95% CI 0.29-0.47, P < .001), chemotherapy alone (aOR 0.52, 95% CI 0.40-0.68, P < .001), and other palliative therapy (aOR 0.67, 95% CI 0.50-0.90, P < .001), among other factors, were associated with decreased odds of undergoing palliative treatment at a facility with volume ≥80^th percentile on multivariable logistic regression (Hosmer-Lemeshow: chi-square 6.634, P = .577) (Table 2). Undergoing surgery alone (aOR 1.34, 95% CI 1.07-1.68, P = .012), urban/rural population density and CDCS ≥1, among other factors, were associated with increased odds of undergoing palliative treatment at a facility with volume ≥80^th percentile on multivariable logistic regression ().

Table 2.

Univariable and Multivariable Binary Logistic Regression Models Identifying Factors Associated With Undergoing Palliative Treatment at a Facility With Volume $\geq$ 80th Percentile.

	Univariable		Multivariable
	OR (95% CI)	P Value	aOR^a (95% CI)	P Value
Age at diagnosis, median years (IQR)	0.97 (0.97-0.98)	<.001	0.98 (0.97-0.99)	<.001
Sex
Male	Ref		Ref
Female	0.83 (0.73-0.93)	.002	0.96 (0.81-1.13)	.616
Race
White	Ref		Ref
Black	0.89 (0.76-1.03)	.115	0.79 (0.64-0.98)	.032
Other	1.61 (1.27-2.04)	<.001	2.59 (1.87-3.59)	<.001
Primary payer status
No insurance	0.83 (0.66-1.05)	.113	0.85 (0.63-1.16)	.303
Private insurance, managed care	Ref		Ref
Medicaid	1.04 (0.88-1.22)	.655	0.86 (0.69-1.09)	.212
Medicare	0.59 (0.52-0.68)	<.001	0.98 (0.79-1.22)	.856
Household income
$<$ $40,227	Ref		Ref
$40,227-50,353	0.80 (0.68-0.94)	.008	0.95 (0.78-1.17)	.655
$50,354-63,332	0.82 (0.69-0.97)	.017	1.10 (0.89-1.36)	.396
$\geq$ $63,333	0.55 (0.46-0.66)	<.001	0.69 (0.55-0.86)	.001
Population density
Metro	Ref		Ref
Urban, rural	1.81 (1.59-2.05)	<.001	1.99 (1.63-2.43)	<.001
Facility type
Academic/research	Ref		Ref
Non-academic/research	0.20 (0.17-0.22)	<.001	0.09 (0.08-0.11)	<.001
Facility location
New England	Ref		Ref
Middle and South Atlantic	0.44 (0.36-0.53)	<.001	5.18 (3.00-8.94)	<.001
Central	0.64 (0.54-0.77)	<.001	5.37 (3.10-9.29)	<.001
Mountain, Pacific	0.25 (0.19-0.33)	<.001	2.45 (1.36-4.44)	.003
Distance to facility, mi	1.00 (1.00-1.00)	<.001	1.00 (1.00-1.00)	.260
Charlson-Deyo comorbidity score
0	Ref		Ref
$\geq$ 1	1.19 (1.06-1.33)	.003	1.18 (1.00-1.39)	.045
History of prior malignancy
No	Ref
Yes	1.03 (0.92-1.16)	.596
Histology
Conventional squamous cell carcinoma	Ref		Ref
Other	0.75 (0.65-0.86)	<.001	1.16 (0.92-1.47)	.208
Primary site
Oral cavity	Ref		Ref
Major salivary glands	0.37 (0.27-0.51)	<.001	0.51 (0.33-0.80)	.004
Sinonasal tract	0.70 (0.57-0.85)	<.001	0.69 (0.51-0.92)	.012
Oropharynx	0.95 (0.82-1.11)	.531	0.93 (0.74-1.15)	.492
Hypopharynx	0.79 (0.63-0.98)	.034	0.76 (0.56-1.04)	.083
Larynx	0.91 (0.78-1.07)	.262	0.72 (0.57-0.90)	.004
HPV status
HPV Negative	Ref		Ref
HPV Positive	1.54 (1.21-1.97)	<.001	1.10 (0.76-1.58)	.613
Unknown	1.28 (1.08-1.53)	.005	1.38 (1.08-1.75)	.010
Grade
Low	Ref		Ref
High	0.82 (0.72-0.93)	.002	0.96 (0.80-1.58)	.680
Unknown	0.68 (0.60-0.77)	<.001	0.91 (0.75-1.10)	.339
cT classification
1	Ref		Ref
2	0.97 (0.76-1.23)	.784	1.00 (0.71-1.42)	.983
3	0.90 (0.71-1.14)	.376	1.04 (0.74-1.46)	.826
4	0.99 (0.80-1.23)	.953	1.04 (0.76-1.43)	.804
x	0.66 (0.52-0.85)	.001	0.82 (0.54-1.26)	.365
cN classification
0	Ref		Ref
1	0.91 (0.76-1.09)	.311	0.98 (0.76-1.27)	.904
2	1.07 (0.94-1.22)	.316	0.96 (0.79-1.17)	.698
3	0.94 (0.75-1.18)	.603	0.93 (0.67-1.28)	.649
x	0.75 (0.62-0.90)	.002	0.83 (0.56-1.24)	.371
cM classification
0	Ref		Ref
1	0.54 (0.47-0.62)	<.001	0.70 (0.57-0.85)	<.001
x	0.83 (0.69-1.00)	.051	1.22 (0.86-1.73)	.262
Palliative treatment
Surgery alone	1.26 (1.07-1.48)	.005	1.34 (1.07-1.68)	.012
Radiotherapy alone	0.24 (0.20-0.28)	<.001	0.37 (0.29-0.47)	<.001
Chemotherapy alone	0.42 (0.34-0.50)	<.001	0.52 (0.40-0.68)	<.001
Pain management alone	Ref		Ref
Combination therapy	0.60 (0.50-0.73)	<.001	0.89 (0.68-1.17)	.400
Other	0.57 (0.46-0.70)	<.001	0.67 (0.50-0.90)	.008

Abbreviations: aOR, adjusted odds ratio; CI, confidence interval; cTNM, pathologic tumor-nodal-metastasis; OR, odds ratio.

^aN = 6970; number of uncensored events: 1112.

Bold indicates statistical significance (P < 0.005)

5-Year OS by Facility Volume and Prognostic Factors

5-year OS rates of patients undergoing palliative therapy at facilities with volume <20^th, 20-40^th, 40-60^th, 60-80^th, and ≥80^th percentile was 11%, 13%, 11%, 14%, and 23%, respectively (P < .001) (Table 3). Increasing facility volume percentile was not associated with higher 5-year OS in cM1 disease (6%, 7%, 7%, 6%, 6%, P = 0.149) (Figure 3).

Table 3.

Kaplan-Meier Analysis of 5-Year OS Rate (%) by Facility Volume Percentile.

	<20^th	20-40^th	40-60^th	60-80^th	$\geq$ 80^th	P value^a
Histology
Conventional squamous cell carcinoma	10	13	11	13	23	<.001
Other	14	14	14	15	21	.007
Primary site
Oral cavity	7	8	8	10	20	<.001
Major salivary glands	10	9	14	14	15	.020
Sinonasal tract	12	17	10	13	18	.044
Oropharynx	13	15	15	17	29	<.001
Hypopharynx	9	12	5	8	12	.014
Larynx	13	15	11	15	24	<.001
HPV status
HPV Negative	10	12	9	13	17	.030
HPV Positive	17	22	27	35	46	<.001
Unknown	11	12	10	12	21	<.001
Grade
Low	9	14	11	14	25	<.001
High	12	13	9	13	22	<.001
Unknown	12	12	14	13	19	<.001
cT classification
1	16	21	23	29	32	<.001
2	14	19	15	22	32	<.001
3	11	12	9	13	27	<.001
4	8	10	8	9	14	<.001
x	14	13	13	13	24	<.001
cN classification
0	11	15	11	17	23	<.001
1	13	15	12	15	30	<.001
2	10	12	12	12	22	<.001
3	7	6	7	11	13	.046
x	13	12	13	11	21	<.001
cM classification
0	13	16	12	17	27	<.001
1	6	7	7	6	6	.149
x	18	15	19	15	22	.058
Palliative treatment
Surgery alone	23	31	23	28	34	<.001
Radiotherapy alone	8	8	8	7	6	.600
Chemotherapy alone	10	8	12	9	8	.455
Pain management alone	12	13	12	16	30	<.001
Combination therapy	8	16	9	9	14	.011
Other	16	16	9	15	26	<.001

Abbreviations: cTNM, pathologic tumor-nodal-metastasis; OS, overall survival.

^a2-sided threshold for statistical significance is P < 0.005, as calculated with the Bonferroni correction.

Bold indicates statistical significance (P < 0.005)

Figure 3.

5-year overall survival by facility volume percentile. Significance derived from the log-rank test.

Palliative treatment facility volume ≥80^th percentile (adjusted hazard ratio [aHR] 0.79, 95% CI 0.72-0.87), HPV positive status (aHR 0.60, 95% CI 0.52-0.68), and undergoing surgery alone (aHR 0.55, 95% CI 0.50-0.61), among other factors, were associated with higher 5-year OS on multivariable Cox regression (P < .01) (Table 4). Older age, non-private insurance status, CDCS ≥1, history of prior malignancy, and more advanced tumors, among other factors, were associated with worse 5-year OS on multivariable Cox regression (P < .001). After binarizing facility volume at various percentile thresholds ≥50^th, HVFs were associated with higher 5-year OS compared with lower volume facilities (LVFs) (Table 5).

Table 4.

Univariable and Multivariable Cox Proportional Hazard Regression Models.

	Univariable		Multivariable
	HR (95% CI)	P Value	aHR^a (95% CI)	P Value
Age at diagnosis, median years (IQR)	1.02 (1.02-1.02)	<.001	1.02 (1.02-1.02)	<.001
Sex
Male	Ref
Female	1.04 (0.99-1.10)	.111
Race
White	Ref		Ref
Black	1.16 (1.09-1.24)	<.001	1.08 (1.00-1.16)	.057
Other	0.91 (0.80-1.02)	.103	0.89 (0.78-1.02)	.104
Primary payer status
No insurance	1.49 (1.34-1.65)	<.001	1.33 (1.19-1.50)	<.001
Private insurance, managed care	Ref		Ref
Medicaid	1.33 (1.23-1.44)	<.001	1.22 (1.12-1.33)	<.001
Medicare	1.73 (1.63-1.84)	<.001	1.19 (1.10-1.28)	<.001
Household income
$<$ $40,227	Ref		Ref
$40,227-50,353	0.96 (0.90-1.03)	.276	1.01 (0.94-1.09)	.753
$50,354-63,332	0.91 (0.85-0.98)	.010	0.95 (0.88-1.02)	.149
$\geq$ $63,333	0.94 (0.88-1.00)	.050	0.94 (0.87-1.01)	.106
Population density
Metro	Ref
Urban, rural	0.96 (0.90-1.01)	.955
Facility type
Academic/research	Ref		Ref
Non-academic/research	1.09 (1.04-1.14)	<.001	0.98 (0.93-1.05)	.606
Facility location
New England	Ref		Ref
Middle and South Atlantic	1.30 (1.18-1.42)	<.001	0.95 (0.85-1.06)	.332
Central	1.34 (1.22-1.46)	<.001	1.01 (0.90-1.12)	.895
Mountain, Pacific	1.19 (1.07-1.33)	.001	0.94 (0.83-1.07)	.359
Facility volume percentile
<20^th	Ref		Ref
20-40^th	0.92 (0.85-0.98)	.016	0.93 (0.85-1.00)	.058
40-60^th	0.93 (0.86-0.99)	.027	0.97 (0.90-1.05)	.491
60-80^th	0.85 (0.80-0.91)	<.001	0.93 (0.85-1.00)	.062
$\geq$ 80^th	0.62 (0.58-0.67)	<.001	0.79 (0.72-0.87)	<.001
Distance to facility, mi	1.00 (1.00-1.00)	.004	1.00 (1.00-1.00)	.127
Charlson-Deyo comorbidity score
0	Ref		Ref
$\geq$ 1	1.14 (1.08-1.20)	<.001	1.17 (1.11-1.24)	<.001
History of prior malignancy
No	Ref		Ref
Yes	1.21 (1.16-1.28)	<.001	1.21 (1.14-1.28)	<.001
Histology
Conventional squamous cell carcinoma	Ref		Ref
Other	0.88 (0.83-0.93)	<.001	0.74 (0.69-0.81)	<.001
Primary site
Oral cavity	Ref		Ref
Major salivary glands	0.84 (0.76-0.93)	<.001	0.83 (0.73-0.94)	.003
Sinonasal tract	0.88 (0.81-0.96)	.003	0.96 (0.87-1.06)	.447
Oropharynx	0.78 (0.73-0.84)	<.001	0.97 (0.90-1.05)	.478
Hypopharynx	1.01 (0.93-1.11)	.769	1.00 (0.90-1.11)	.982
Larynx	0.83 (0.77-0.89)	<.001	0.92 (0.85-1.00)	.052
HPV status
HPV Negative	Ref		Ref
HPV Positive	0.54 (0.48-0.61)	<.001	0.60 (0.52-0.68)	<.001
Unknown	0.98 (0.92-1.06)	.651	1.07 (0.99-1.15)	.110
Grade
Low	Ref
High	1.04 (0.98-1.10)	.171
Unknown	1.03 (0.97-1.09)	.339
cT classification
1	Ref		Ref
2	1.13 (1.02-1.26)	.025	1.06 (0.94-1.20)	.322
3	1.49 (1.34-1.66)	<.001	1.42 (1.27-1.60)	<.001
4	1.78 (1.61-1.97)	<.001	1.66 (1.48-1.85)	<.001
x	1.52 (1.36-1.69)	<.001	1.34 (1.17-1.55)	<.001
cN classification
0	Ref		Ref
1	1.00 (0.92-1.08)	.904	1.04 (0.96-1.14)	.327
2	1.07 (1.01-1.13)	.021	1.18 (1.10-1.26)	<.001
3	1.34 (1.22-1.47)	<.001	1.41 (1.26-1.57)	<.001
x	1.13 (1.04-1.21)	.002	1.33 (1.17-1.50)	<.001
cM classification
0	Ref		Ref
1	1.45 (1.38-1.53)	<.001	1.47 (1.38-1.57)	<.001
x	1.03 (0.95-1.11)	.532	0.95 (0.84-1.07)	.420
Palliative treatment
Surgery alone	0.68 (0.62-0.74)	<.001	0.55 (0.50-0.61)	<.001
Radiotherapy alone	1.32 (1.23-1.42)	<.001	0.81 (0.74-0.88)	<.001
Chemotherapy alone	1.04 (0.96-1.13)	.335	0.64 (0.59-0.71)	<.001
Pain management alone	Ref		Ref
Combination therapy	1.07 (0.98-1.17)	.127	0.73 (0.66-0.81)	<.001
Other	1.11 (1.01-1.22)	.038	0.88 (0.79-0.98)	.023

Abbreviations: aHR, adjusted hazard ratio; CI, confidence interval; cTNM, pathologic tumor-nodal-metastasis; HR, hazard ratio.

^aN = 7105; number of uncensored deaths: 6146.

Bold indicates statistical significance (P < 0.005)

Table 5.

Kaplan-Meier Analysis and Multivariable Cox Proportional Hazard Regression Models of 5-Year OS After Binarizing Facility Volume at Varying Percentile Thresholds (Total Number of Facilities Included in Analysis: 984).

	No. LVF^a	No. HVF^b	Kaplan-Meier Analysis			Multivariable Cox Proportional Hazard Regression Models
Percentile	No. LVF^a	No. HVF^b	LVF	HVF	P Value	aHR^c (95%CI)	P Value
50^th	887	97	12	16	<.001	0.93 (0.88-0.98)	.009
55^th	906	78	12	17	<.001	0.87 (0.82-0.92)	<.001
60^th	922	62	12	18	<.001	0.86 (0.81-0.91)	<.001
65^th	937	47	12	18	<.001	0.87 (0.82-0.92)	<.001
70^th	946	38	12	19	<.001	0.83 (0.77-0.88)	<.001
75^th	958	26	12	20	<.001	0.80 (0.75-0.86)	<.001
80^th	966	18	12	22	<.001	0.78 (0.72-0.84)	<.001
85^th	973	11	12	24	<.001	0.74 (0.68-0.81)	<.001
90^th	978	6	13	26	<.001	0.69 (0.63-0.77)	<.001
95^th	983	1	13	35	<.001	0.56 (0.49-0.66)	<.001

Abbreviations: aHR, adjusted hazard ratio; CI, confidence interval; HVF, higher volume facility; LVF, lower volume facility; OS, overall survival.

^aLVF is defined as a facility with volume less than the percentile.

^bHVF is defined as a facility with volume greater than or equal to the percentile.

^cReference: LVF.

2-Year and 3-Year OS Sensitivity Analysis

Patients undergoing palliative treatment at facilities with volume <20^th, 20-40^th, 40-60^th, 60-80^th, and ≥80^th percentile had 2-year OS of 20%, 22%, 23%, 25%, and 36%, respectively (P < .001), and 3-year OS of 15%, 17%, 16%, 19%, and 29%, respectively (P < .001). Undergoing palliative treatment at facilities with volume ≥80^th percentile remained associated with higher 2-year (aHR 0.80, 95% CI 0.73-0.88, P < .001) and 3-year OS (aHR 0.81, 95% CI 0.73-0.89, P < .001) on multivariable Cox regression.

3-Month, 6-Month, and 12-Month Landmark Survival Analysis

1899 (21.9%) patients had survival time <3 months. After censoring survival time <3 months, facility volume ≥80^th percentile (aHR 0.87, 95% CI 0.78-0.98, P = .020) remained associated with OS on multivariable Cox regression.

3506 (404%) patients had survival time <6 months. After censoring survival time <6 months, facility volume ≥80^th percentile (aHR 0.94, 95% CI 0.82-1.07, P = .369) was not associated with OS on multivariable Cox regression.

5305 (61.1%) patients had survival time <12 months. After censoring survival time <12 months, facility volume ≥80^th percentile (aHR 1.05, 95% CI 0.88-1.25, P = .630) was not associated with OS on multivariable Cox regression.

Discussion

Palliative treatment involving pain management and comfort care plays an essential role in offering support to patients with terminal illnesses and their families.³⁸ Higher OS has been demonstrated in patients with HNC undergoing curative treatment at HVFs.^27-31 The results of our study suggest that undergoing palliative treatment at HVFs is independently associated with higher OS. Although treatment goals are shifted towards patient comfort in the palliative setting, improved survival may be an added benefit of successfully implemented palliative treatment regimens.

Undergoing palliative treatment at a facility with volume ≥80th percentile remained an independent predictor of improved survival after adjusting for patient and treatment factors. Further sensitivity analyses demonstrated that increased volume was associated with higher OS starting at the 50^th volume percentile. This finding may be explained by the multidisciplinary teams and other resources in HVFs which facilitate complex procedures requiring extended LOS, intensive care, or outpatient care.³⁹ Moreover, the unique challenges of palliative treatment in HNC may be better addressed through coordination of multiple specialists, outpatient support staff, and comprehensive treatment plans. Collaboration among experienced professionals across all treatment modalities may aid in determining the most appropriate treatment for each patient and likely contributes to lower complication rates evident in HVFs.⁴⁰ It is, however, important to recognize that patient outcomes are multifactorial, and our study does not suggest that optimal care is exclusive to HVFs but highlights the potential benefits of improving access to these facilities. Most importantly, patient-directed goals of care should be the measure of success, as comfort, rather than prolonged survival, may be the priority for many patients undergoing palliative treatment.

Palliative management of HNC often involves various treatments to enhance patient quality of life. Our study found that palliative management most frequently involved radiotherapy (32.3%), pain management (16.3%), chemotherapy (15.7%), or surgery (15.6%) alone. Palliative surgeries, such as tracheostomy, enteral feeding procedures, and tumor debulking, aim to control symptoms and improve swallowing, nutrition, and airway in active and self-caring patients.²⁵ Surgical treatment was associated with undergoing treatment at HVFs, possibly attributed to greater access to resident doctors, hospital resources, and diverse multidisciplinary personnel which aid in performing complex palliative surgeries. Non-surgical treatment modalities, including pain management, were associated with decreased odds of undergoing palliative treatment at an HVF, as they may be more standardized, widely accessible, and less resource intensive, making them suitable for a variety of healthcare settings. Our analysis of palliative treatment is likely confounded by stage. Compared with those undergoing pain management, patients undergoing surgery likely had either less advanced disease increasing the likelihood of tumor eradication, or more advanced disease warranting surgical intervention for symptom palliation and restoration of bodily functions such as breathing. Patients undergoing palliative surgery may also have responded to induction therapy, which is not reported in the NCDB.

Our study found that undergoing surgery, radiotherapy, chemotherapy, or combination therapy was associated with higher OS than pain management alone in the palliative management of HNC, consistent with previous studies.^41,42 Of note, a selection bias may account for patients with less advanced disease undergoing surgery, as these patients may have been more amenable to surgery or had already responded to induction therapy, but these factors are not documented in the NCDB.

5-year OS rates of patients undergoing palliative therapy were significantly greater at HVFs. Although HVFs have more experience with surgery, which remains the mainstay of HNC treatment, the survival benefit at HVFs did not persist among patients with distantly metastatic disease, which is generally managed with pain management or systemic therapy as opposed to intensive localized therapy (i.e., surgery or radiation).⁴³ Although speculative, this may suggest that the survival benefit of palliative treatment at HVFs may be most appropriate for patients amenable to localized therapies such as surgical resection, rather than those with metastatic disease.

Our study found increased comorbidities, prior malignancy, and more advanced tumors were associated with poorer survival, whereas HPV positive disease was associated with improved survival, consistent with previous studies.^44-46 Older age was also a negative prognostic factor for survival, consistent with literature attributing this trend to greater comorbidities and diminished functional status.^47,48 All primary payer statuses portended poorer survival compared with private insurance/managed care, consistent with prior studies in HNC.^49,50 This underscores health insurance status as an important factor influencing healthcare decision-making, and possibly serving as a proxy for other socioeconomic factors impacting survival.⁵¹ For example, tobacco use, a variable not reported in the NCDB, is twice as likely among patients with Medicaid compared with those with private insurance.⁵²

Despite the large sample size, our study is inherently limited by its retrospective study design and inability to establish causal relationships. Most importantly, factors relevant to palliative decision making such as quality of life, feeding dependence, and pain management, extend beyond survival alone and are often not sufficiently captured in large retrospective databases such as the NCDB. Although our study associated HVFs with higher OS, disease-modifying palliative treatments encoded in the NCDB do not capture the full spectrum of palliative care, previously defined as “an organized program providing specialized medical care, drugs or therapies for the management of acute or chronic pain and/or the control of symptoms administered by specially trained physicians and other clinicians; and supportive care services, such as counseling on advanced directives, spiritual care, and social services, to patients with advanced disease and their families.”^38,53-55 Review of the NCDB presents the possibility of inaccurate histologic diagnosis and variable miscoding due to differing reporting standards across healthcare centers. Selection bias is also an inherent limitation, as patients reported in the NCDB undergo treatment in a select group of treatment facilities in the United States. The NCDB does not encode locoregional control and disease-free survival, potentially confounding survival analysis. The NCDB also lacks patient-level socioeconomic data, rather utilizing an ecologic approximation of resident zip code.⁵⁶ Although our multivariable analysis controlled for several variables, unmeasured confounders including socioeconomic status, comorbidities, functional status, and palliative treatment goals not reported in the NCDB likely influence OS, further restricting causal inference. Patients in our cohort may have undergone treatment with curative intent prior to palliative treatment, which is not encoded in the NCDB. Nevertheless, despite these limitations, our study found that HVFs were associated with higher OS in HNC palliative treatment.⁵⁷

Conclusion

Patients undergoing surgery had increased odds of undergoing palliative treatment at a facility with volume ≥80th percentile. After adjusting for patient demographics, facility characteristics, cTNM classification, and treatment, undergoing palliative treatment at a facility with volume ≥80th percentile was associated with higher OS, although this survival benefit was not observed for patients with cM1 disease. Older age, non-private insurance status, and non-surgical treatment were associated with worse OS. Facility volume alone does not determine survival outcomes and should therefore be considered as one of the many variables involved in maximizing quality of life during end-of-life care.

Footnotes

Author Contributions

Praneet C. Kaki: design, analysis, interpretation, manuscript writing; Aman M. Patel: conception, design, analysis, interpretation, manuscript writing; Lucy Revercomb: conception, design, analysis, interpretation, manuscript writing; Russell Maxwell: design, analysis, interpretation, manuscript writing; Robert M. Brody: design, analysis, interpretation, manuscript writing; Jason A. Brant: design, analysis, interpretation, manuscript writing; Ryan M. Carey: design, analysis, interpretation, manuscript writing, final approval.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Praneet C. Kaki

Lucy Revercomb

References

Barsouk

Aluru

Rawla

Saginala

Barsouk

. Epidemiology, risk factors, and prevention of head and neck squamous cell carcinoma. Med Sci. 2023;11(2):42. doi:10.3390/medsci11020042

Siegel

Miller

Jemal

. Cancer statistics, 2017. CA Cancer J Clin. 2017;67(1):7-30. doi:10.3322/caac.21387

Jones

. The untreated patient with squamous carcinoma of the head and neck. Am J Clin Oncol. 1995;18(4):363-368. doi:10.1097/00000421-199508000-00018

Nilsen

Johnson

. Potential for low-value palliative care of patients with recurrent head and neck cancer. Lancet Oncol. 2017;18(5):e284-e289. doi:10.1016/S1470-2045(17)30260-7

Haun

Estel

Rücker

, et al. Early palliative care for adults with advanced cancer. Cochrane Database Syst Rev. 2017;6(6):CD011129. doi:10.1002/14651858.CD011129.pub2

Integration of palliative care into standard oncology care: American Society of Clinical Oncology clinical practice guideline update - PubMed. Accessed June 10, 2024. https://pubmed.ncbi.nlm.nih.gov/28034065/

Shuman

Yang

Taylor

JMG

Prince

. End-of-life care among head and neck cancer patients. Otolaryngol Head Neck Surg. 2011;144(5):733-739. doi:10.1177/0194599810397603

Bakitas

Lyons

Hegel

, et al. Effects of a palliative care intervention on clinical outcomes in patients with advanced cancer: the Project ENABLE II randomized controlled trial. JAMA. 2009;302(7):741-749. doi:10.1001/jama.2009.1198

The experiences and the needs of caregivers of patients with head and neck cancer: an integrative review - PubMed. Accessed June 10, 2024. https://pubmed.ncbi.nlm.nih.gov/32541207/

10.

Impact of patient symptoms and caregiving tasks on psychological distress in caregivers for head and neck cancer (HNC) - PubMed. Accessed June 10, 2024. https://pubmed.ncbi.nlm.nih.gov/30578599/

11.

Alam

Hannon

Zimmermann

. Palliative care for family caregivers. J Clin Oncol. 2020;38(9):926-936. doi:10.1200/JCO.19.00018

12.

Working in silos? - Head & Neck cancer patients during and after treatment with or without early palliative care referral - PubMed. Accessed June 10, 2024. https://pubmed.ncbi.nlm.nih.gov/28069153/

13.

Mayland

Payne

. Head and neck cancer: a marginalised group? Palliat Med. 2018;32(8):1286-1287.

14.

Lokker

Offerman

MPJ

van der Velden

de Boer

Pruyn

JFA

Teunissen

SCCM

. Symptoms of patients with incurable head and neck cancer: prevalence and impact on daily functioning. Head Neck. 2013;35(6):868-876. doi:10.1002/hed.23053

15.

Improving outcomes in head and neck cancers | Guidance | NICE. Published November 24, 2004. Accessed June 10, 2024. https://www.nice.org.uk/guidance/csg6

16.

Osazuwa-Peters

Barnes

Okafor

, et al. Incidence and risk of suicide among patients with head and neck cancer in rural, urban, and metropolitan areas. JAMA Otolaryngol-- Head Neck Surg. 2021;147(12):1045-1052. doi:10.1001/jamaoto.2021.1728

17.

Stigma, depression, and anxiety among patients with head and neck cancer - PubMed. Accessed June 10, 2024. https://pubmed.ncbi.nlm.nih.gov/34533631/

18.

Hammermüller

Hinz

Dietz

, et al. Depression, anxiety, fatigue, and quality of life in a large sample of patients suffering from head and neck cancer in comparison with the general population. BMC Cancer. 2021;21(1):94. doi:10.1186/s12885-020-07773-6

19.

Pichardo

PFA

Desiato

Hellums

Altman

Purdy

Haugen

. Depression and anxiety in patients with head and neck cancer undergoing free flap reconstruction. Am J Otolaryngol. 2024;45(1):104044. doi:10.1016/j.amjoto.2023.104044

20.

Civantos

Prasad

Carey

, et al. Palliative care in metastatic head and neck cancer. Head Neck. 2021;43(9):2764-2777. doi:10.1002/hed.26761

21.

Ramsey

Lee

Curran

, et al. Associations of insurance, urbanity, and comorbidity with types of palliative care received by patients with head and neck cancer. Head Neck. 2021;43(5):1499-1508. doi:10.1002/hed.26648

22.

Fereydooni

Valdez

William

Malik

Mehra

Judson

. Predisposing, enabling, and need factors driving palliative care use in head and neck cancer. Otolaryngol Head Neck Surg. Online ahead of print. doi:10.1002/ohn.819

23.

Analysis of palliative care treatment among head and neck patients with cancer: national perspective - PubMed. Accessed June 10, 2024. https://pubmed.ncbi.nlm.nih.gov/33151575/

24.

An enhanced role for palliative care in the multidisciplinary approach to high-risk head and neck cancer - PubMed. Accessed June 10, 2024. https://pubmed.ncbi.nlm.nih.gov/26505177/

25.

Cocks

Ah-See

Capel

Taylor

. Palliative and supportive care in head and neck cancer: United Kingdom National Multidisciplinary Guidelines. J Laryngol Otol. 2016;130(S2):S198-S207. doi:10.1017/S0022215116000633

26.

Palliative intent treatment for head and neck cancer: an analysis of practice and outcomes - PubMed. Accessed June 10, 2024. https://pubmed.ncbi.nlm.nih.gov/30929645/

27.

Torabi

Benchetrit

Kuo

, et al. Prognostic case volume thresholds in patients with head and neck squamous cell carcinoma. JAMA Otolaryngol-- Head Neck Surg. 2019;145(8):708-715. doi:10.1001/jamaoto.2019.1187

28.

David

Luu

, et al. Treatment at high-volume facilities and academic centers is independently associated with improved survival in patients with locally advanced head and neck cancer. Cancer. 2017;123(20):3933-3942. doi:10.1002/cncr.30843

29.

Goshtasbi

Abiri

Lehrich

Haidar

Tjoa

Kuan

. The influence of facility volume on patient treatments and survival outcomes in nasopharyngeal carcinoma. Head Neck. 2021;43(9):2755-2763. doi:10.1002/hed.26739

30.

Abiri

Pang

Roman

, et al. Facility volume as a prognosticator of survival in locally advanced papillary thyroid cancer. Laryngoscope. 2023;133(2):443-450. doi:10.1002/lary.30280

31.

Rygalski

Huttinger

Zhao

, et al. High surgical volume is associated with improved survival in head and neck cancer. Oral Oncol. 2023;138:106333. doi:10.1016/j.oraloncology.2023.106333

32.

Bilimoria

Stewart

Winchester

. The National Cancer data base: a powerful initiative to improve cancer care in the United States. Ann Surg Oncol. 2008;15(3):683-690. doi:10.1245/s10434-007-9747-3

33.

Merkow

Rademaker

Bilimoria

. Practical guide to surgical data sets: National Cancer Database (NCDB). JAMA Surg. 2018;153(9):850-851. doi:10.1001/jamasurg.2018.0492

34.

Boffa

Rosen

Mallin

, et al. Using the national cancer database for outcomes research: a review. JAMA Oncol. 2017;3(12):1722-1728. doi:10.1001/jamaoncol.2016.6905

35.

Janz

Graboyes

Nguyen

, et al. A comparison of the NCDB and SEER database for research involving head and neck cancer. Otolaryngol Head Neck Surg. 2019;160(2):284-294. doi:10.1177/0194599818792205

36.

Krebbers

Simon

Pilz

Kremer

Winkens

Baijens

LWJ

. Patients with head-and-neck cancer: dysphagia and affective symptoms. Folia Phoniatr Logop. 2021;73(4):308-315. doi:10.1159/000508367

37.

National Cancer Database . ACS. https://www.facs.org/quality-programs/cancer-programs/national-cancer-database/

38.

Goldsmith

Dietrich

Morrison

. Variability in access to hospital palliative care in the United States. J Palliat Med. 2008;11(8):1094-1102. doi:10.1089/jpm.2008.0053

39.

Birkmeyer

Stukel

Siewers

Goodney

Wennberg

Lucas

. Surgeon volume and operative mortality in the United States. N Engl J Med. 2003;349(22):2117-2127. doi:10.1056/NEJMsa035205

40.

Jalisi

Bearelly

Abdillahi

Truong

. Outcomes in head and neck oncologic surgery at academic medical centers in the United States. Laryngoscope. 2013;123(3):689-698. doi:10.1002/lary.23835

41.

Schwam

Burtness

Yarbrough

Mehra

Husain

Judson

. National treatment patterns in patients presenting with Stage IVC head and neck cancer: analysis of the National Cancer Database. Cancer Med. 2015;4(12):1828-1835. doi:10.1002/cam4.546

42.

Farber

Povolotskiy

Bavier

Riccardi

Eloy

Hsueh

. Impact of palliative treatment on survival in sinonasal malignancies. Int Forum Allergy Rhinol. 2019;9(12):1499-1507. doi:10.1002/alr.22432

43.

Patel

Haleem

Choudhry

Brant

Brody

Carey

. Surgical resection improves overall survival in cT4b major salivary gland cancer. Otolaryngol Head Neck Surg. 2024;170(5):1349-1363. doi:10.1002/ohn.686

44.

Zhu

Lin

Zhang

Chen

Rao

. Impact of prior cancer history on the survival of patients with larynx cancer. BMC Cancer. 2020;20(1):1137. doi:10.1186/s12885-020-07634-2

45.

Stordeur

Schillemans

Savoye

, et al.

Comorbidity in head and neck cancer: is it associated with therapeutic delay, post-treatment mortality and survival in a population-based study?

Oral Oncol. 2020;102:104561. doi:10.1016/j.oraloncology.2019.104561

46.

Ang

Harris

Wheeler

, et al. Human papillomavirus and survival of patients with oropharyngeal cancer. N Engl J Med. 2010;363(1):24-35. doi:10.1056/NEJMoa0912217

47.

Raab

Restifo

McBride

, et al. Outcomes following head and neck cancer surgery among older adults as determined by an electronic geriatric assessment. J Geriatr Oncol. 2022;13(2):228-233. doi:10.1016/j.jgo.2021.10.010

48.

Stromberger

Yedikat

Coordes

, et al. Prognostic factors predict oncological outcome in older patients with head and neck cancer undergoing chemoradiation treatment. Front Oncol. 2020;10:566318. doi:10.3389/fonc.2020.566318

49.

Kwok

Langevin

Argiris

Grandis

Gooding

Taioli

. The impact of health insurance status on the survival of patients with head and neck cancer. Cancer. 2010;116(2):476-485. doi:10.1002/cncr.24774

50.

Sittig

Luu

Yoshida

, et al. Impact of insurance on survival in patients < 65 with head & neck cancer treated with radiotherapy. Clin Otolaryngol. 2020;45(1):63-72. doi:10.1111/coa.13467

51.

Gupta

Sonis

Schneider

Villa

. Impact of the insurance type of head and neck cancer patients on their hospitalization utilization patterns. Cancer. 2018;124(4):760-768. doi:10.1002/cncr.31095

52.

Jamal

Phillips

Gentzke

, et al. Current cigarette smoking among adults - United States, 2016. MMWR Morb Mortal Wkly Rep. 2018;67(2):53-59. doi:10.15585/mmwr.mm6702a1

53.

Rygalski

Huttinger

Zhao

, et al. High surgical volume is associated with improved survival in head and neck cancer. Oral Oncol. 2023;138:106333.

54.

Alfieri

Orlandi

Bossi

. The case volume issue in head and neck oncology. Curr Treat Options Oncol. 2017;18(11):65.

55.

Eskander

Merdad

Irish

, et al. Volume-outcome associations in head and neck cancer treatment: a systematic review and meta-analysis. Head Neck. 2014;36(12):1820-1834.

56.

Sullivan

Al-Qurayshi

Anderson

Seaman

Pagedar

. Factors associated with the choice of radiation therapy treatment facility in head and neck cancer. Laryngoscope. 2021;131(5):1019-1025. doi:10.1002/lary.29033

57.

Pando

Patel

Choudhry

Eloy

Goldstein

Liu

. Palliative care effects on survival in glioblastoma: who receives palliative care? World Neurosurg. 2023;170:e847-e857. doi:10.1016/j.wneu.2022.11.143