SURVIVAL OF U.S. BLACK AND WHITE PATIENTS: RESULTS

Posted by James

For the 1,125 black and 2,392 white patients with squamous cell carcinoma of the esophagus, distant stage was more frequent and regional and unknown stages less frequent in blacks than whites (Table 1). The proportions of female and younger patients were higher, and the proportion with cancer-directed surgery lower—among black than white patients—while the proportion with radiotherapy differed little (Table 1). Among patients diagnosed at localized stage, the proportion without cancer-directed surgery was statistically and significantly higher for blacks than whites, while proportions with unknown regional lymph-node status and without pathologic review of lymph nodes were slightly higher among blacks than whites (Table 1). However, only a small proportion of patients (8% of blacks and 12% of whites) had pathologic review of regional lymph nodes. Among localized-stage patients who had cancer-directed surgery, the proportion with total esophagectomy or radical surgery did not differ between blacks (30.8%) and whites (30.9%). Ninety-seven (38.3%o) of 253 localized-stage patients with cancer-directed surgery versus only eight (1.1%) of 719 patients without surgery had pathologic review of regional lymph nodes (data not shown); the latter eight patients had surgery only for the purpose of staging. RSRs for all stages combined were lower in blacks than whites, and 95% CIs for RSRs in blacks and whites did not overlap (Table 2). For RSRs by stage, the largest black-white differences were within localized stage. However, RSRs were lower and black-white differences somewhat smaller among localized-stage patients without pathologic review of regional lymph-node(s) (Table 2). The small numbers of blacks (N=27) and whites (N=78) with pathologic review of regional lymph nodes (Table 1) precluded analysis of RSRs by race, but for the two racial groups combined (105 patients), RSRs were high relative to other localized-stage patients (Table 2).

Table 1. Characteristics of 1,125 Black and 2,392 White Patients Diagnosed With Squamous Cell Esophageal Cancer in 1992-1998, SEER Program

Characteristic	Black (N=1,125) No.	i %	No.	White (N=2,392) %	P value0
All Stages at Diagnosis Stage Localized Regional Distant Unknown	320 286 289 230	28.4 25.4 25.7 20.4	652 690 475 575	27.3 28.8 19.9 24.0	O.001*
Sex Male Female	776 349	69.0 31.0	1453 939	60.7 39.3	O.001*
Age <60 60-69 70-79 80+	461 378 236 50	41.0 33.6 21.0 4.4	469 718 774 431	19.6 30.0 32.4 18.0	O.001*
Marital Status Married Other	360 765	32.0 68.0	1236 1156	51.7 48.3	O.001*
Cancer-Directed Surgery Surgery received No surgeryb	203 922	18.0 82.0	592 1800	24.7 75.3	O.001*
Radiotherapy No Yes	383 742	34.0 66.0	844 1548	35.3 64.7	0.495
Localized Stage Lymph-Node Status Negative Unknown	181 139	56.6 43.4	406 246	62.3 37.7	0.101
Lymph Nodes Examined Pathologically Yes                                      27 No, Unknown                         293		8.4 91.6	78 574	12.0 88.0	0.120
Cancer-Directed Surgery Surgery received No surgeryb	65 255	20.3 79.7	188 464	28.8 71.2	0.006*
a From Chi-square test for association with race.
b Consists mostly of unknown reason for no surgery; also, surgery refused, surgery recommended but unknown if received, and a small number for whom it was unknown if surgery was recommended or received.
*P<0.05.

Within five years after diagnosis, 3,076 (87.5%) of the 3,517 patients had died (90.1% of blacks and 86.2% of whites); underlying cause of death was coded to esophageal cancer for 80.1% of deaths (data not shown). In a Cox proportional hazards regression model for all patients (Table 3, Model 1), the adjusted relative risks for death from any cause for age (increasing risk with rising age), sex (lower risk for women than men), marital status (higher risk for nonmarried versus married), cancer-directed surgery (higher risk for no surgery versus surgery) and radiotherapy (higher risk for patients without radiotherapy) were statistically significant. The adjusted relative risk for black versus white race was only 1.13, although reaching statistical significance for these sample sizes. For localized-stage patients, the variables for cancer-directed surgery and pathologic review of regional lymph nodes were combined into a single variable (Table 3, Model 2), because only eight patients without cancer-directed surgery had pathologic review of lymph nodes. The relative risk was highest for patients who had neither cancer-directed surgery nor pathologic review of lymph nodes and lowest for those with both procedures; the relative risk for blacks versus whites was 1.31, reaching statistical significance (Table 3, Model 2).
Viagra Online Canadian Pharmacy

Table 2. Relative Survival Rate (RSR) (%) for Patients Diagnosed With Squamous Cell Carcinoma of the Esophagus Diagnosed in 1992-1998, SEER Program (11 Registries)

Stage, Race	No.	One-Year RSR (95% CI)	Three-Year RSR (95% CI)	Five-Year RSR (95% CI)
All Stages Black White	1,125 2,392	36.7 (33.7-39.6) 42.1 (40.0-44.2)*	12.1 (10.0-14.3) 19.4 (17.5-21.2)*	8.3 (5.9-10.4) 14.0 (12.1-15.8)*
Localized Stage Black White	320 652	52.0 (46.2-57.7) 59.4 (55.3-63.4)	19.2 (14.2-24.1) 33.1 (28.9-37.3)*	14.4 (9.5-19.3) 25.0 (20.5-29.5)*
Localized Stage, With Pathologic Review of Regional Lymph Nodes (RLNs) Both races0                      105              74.3 (65.3-83.3)      51.8 (40.5-63.2)				48.8 (36.0-61.7)
Localized Stage, No Pathologic Review Black                            293 White                             574		of RNLs 50.4 (44.4-56.5) 57.0 (52.6-61.3)	17.9 (12.9-22.9) 29.8 (25.4-34.1)*	13.6 ( 8.6-18.5) 20.9 (16.4-25.4)
Localized Stage, No Pathologic Review Black                            253 White                             458		of RLN, No Cancer- 48.7 (42.2-55.2) 53.2 (48.2-58.1)	Directed Surgery 16.3 (11.0-21.5) 26.6 (21.9-31.3)*	12.1 ( 7.0-17.2) 16.7 (11.9-21.6)
Regional Stage Black White	286 690	39.1 (33.2-45.0) 46.9 (43.0-50.8)	15.1         (10.4-19.8) 19.2         (15.9-22.6)	11.1 (6.4-15.8) 13.4 (10.1-16.7)
Distant Stage Black White	289 475	20.6 (15.7-25.4) 18.1 (14.5-21.7)	3.9 (1.3-6.5) 4.7 (2.4-6.9)	____ b ____ b
° The numbers of black and white patients were too small for analyses.
b Less than 10 patients survived. * Confidence intervals (CIs) for blacks and whites do not overlap.

Table 3. Multivariable Cox Proportional Hazards Regression Models, Predicting Risk of Death Among Patients With Squamous Cell Carcinoma of the Esophagus
buy antibiotics amoxicillin

Characteristic    Total No.		Relative Risk
		(95% CI)
Model 1: All Stages
Race
White	2392	1.00 Referent
Black	1125	1.13 (1.04-1.22)*
Age
<60	930	1.00 Referent
60-69	1096	1.22 (1.11-1.34)*
70-79	1010	1.32 (1.20-1.46)*
80+	481	1.81 (1.60-2.04)*
Sex
Male	2229	1.00 Referent
Female	1288	0.83 (0.77-0.90)*
Marital Status
Married	1596	1.00 Referent
Other	1921	1.12 (1.03-1.20)*
Cancer-Directed Surgery
Yes	795	1.00 Referent
No, unknown	2722	1.92 (1.74-2.13)*
RadioTherapy
Yes	2290	1.00 Referent
No, unknown	1227	1.79 (1.65-1.94)*
Stage at Diagnosis
Localized	972	1.00 Referent
Regional	976	1.48 (1.34-1.63)*
Distant	764	2.38 (2.15-2.65)*
Unknown	805	1.35 (1.22-1.50)*
Model 2: Localized Stage1 b
Cancer-Directed Surgery/
Pathologic Review of Nodesb
No surgery/no review711		1.00 Referent
Surgery/no review	156	0.67 (0.54-0.83)*
Surgery/review	97	0.42 (0.30-0.58)*
Race
White	646	1.00 Referent
Black	318	1.31 (1.11-1.59)*
* Confidence interval (CI) does not include
1.00.
a Model 2 also includes the other variables
shown in model 1.
b Excludes eight patients with no cancer-
directed surgery who had pathologic review
of lymph nodes (see text).