2017 Section 7 Green Book

Original Article

TABLE 5. Risk Factors for Mortality Among 664 Individuals for Whom Health-Related Quality of Life Was Assessed Within 5 Years Before Head and Neck Cancer Diagnosis

HR [95% CI]

Multivariate Analysis a

Characteristic

No. of Patients (%)

Univariate Analysis

Prediagnosis HRQOL score per 10-point increase: Median/IQR Age at diagnosis: Mean 6 SD, y

90.1/74.4-106.0

0.86 [0.82-0.91]

0.91 [0.85-0.97]

75.2 6 7.5

1.04 [1.03-1.06]

1.07 [1.05-1.09]

Sex

Men

466 (70) 198 (30)

REF

Women

1.12 [0.92-1.37]

Smoking status Former/never

405 (68) 190 (32)

REF

REF

Current

1.24 [1.02-1.52]

1.48 [1.15-1.91]

Marital status Married

369 (57) 279 (43)

REF

REF

Not married

1.41 [1.17-1.69]

1.18 [0.92-1.50]

Education

< High school

205 (32) 204 (32) 235 (36) 241 (36) 220 (33) 73 (11) 130 (20)

REF

High school graduate/GED

0.86 [0.68-1.08] 0.77 [0.61-0.96]

> High school

Household income, US$ < $19,999

REF

REF

$20,000-49,999

0.74 [0.60-0.92] 0.50 [0.35-0.71] 0.86 [0.66-1.11]

1.08 [0.83-1.42] 0.64 [0.42-0.97] 0.83 [0.59-1.17]

$50,000

Do not know/missing

Race

White Other

530 (80) 134 (20)

REF

1.14 (0.89-1.44]

No. of comorbidities 0-1

215 (33) 237 (37) 196 (30) 267 (40) 170 (26) 131 (20) 96 (14)

REF

REF

2-3

1.11 [0.88-1.39] 1.46 [1.16-1.82]

0.99 [0.76-1.31] 1.19 [0.89-1.59]

4

Primary site

Larynx, hypopharynx

REF

REF

Oral cavity Oropharynx

1.24 [0.99-1.56] 1.59 [1.25-2.03] 1.48 [0.90-2.41]

0.98 [0.73-1.30] 1.08 [0.79-1.46] 0.88 [0.61-1.28]

Other b

Stage

Early Late

338 (54) 284 (46)

REF

REF

2.30 [1.90-2.79]

2.50 [1.95-3.19]

Radiotherapy No

257 (39) 397 (61)

REF

REF

Yes

1.20 [1.00-1.45]

1.20 [0.94-1.54]

Abbreviations: CI, confidence interval; GED, General Educational Development; HR, hazard ratio; HRQOL, health-related quality of life; IQR, interquartile range; REF, reference category; SD, standard deviation. a The model includes all characteristics for which multivariate analysis results are reported and survey by proxy. b Other sites include the nasopharynx, nasal cavity, paranasal sinuses, and lip.

fore, to counsel patients that overall HRQOL will cer- tainly improve after therapy, as suggested by previous studies, is likely inaccurate; instead, perhaps providers should emphasize the importance of acclimatization to a new standard of emotional and physical health. Further- more, the finding that individuals with the lowest prediag- nosis HRQOL suffer a worse prognosis independent of other prognostic indicators should lend added gravity to treatment decisions for this subset of patients, such as when considering potentially morbid, life-prolonging interventions versus high-quality palliative care.

HNSCC diagnosis has prognostic significance that was not previously appreciated; these large population-based data provide compelling evidence that prediagnosis HRQOL independently predicts survival. These findings have relevance to patient care. Although most studies to date have reported a decline in HRQOL after diagnosis followed by a recovery toward baseline after treatment, 23-25 these data demonstrate that such recovery is not observed after stratification by sur- vival group and actually may represent an artifact of earlier deaths among individuals with lower HRQOL. There-

Cancer

June 15, 2016

144