SEARCH

SEARCH BY CITATION

Keywords:

  • colon cancer;
  • survival;
  • mortality;
  • race;
  • ethnicity;
  • racial disparity

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

BACKGROUND.

To the authors' knowledge, few studies have addressed racial disparities in the survival of patients with colon cancer by adequately incorporating treatment and socioeconomic factors in addition to patient and tumor characteristics.

METHODS.

The authors studied a nationwide and population-based, retrospective cohort of 18,492 men and women who were diagnosed with stage II or III colon cancer at age ≥65 years between 1992 and 1999. This cohort was identified from the Surveillance, Epidemiology, and End Results (SEER) cancer registries-Medicare linked databases and included up to 11 years of follow-up.

RESULTS.

A larger proportion (70%) of African-American patients with colon cancer fell into the poorest quartiles of socioeconomic status compared with Caucasians (21%). Patients who lived in communities with the lowest socioeconomic level had 19% higher all-cause mortality compared with patients who lived in communities with the highest socioeconomic status (hazards ratio [HR], 1.19; 95% confidence interval [95% CI], 1.13–1.26; P < .001 for trend). The risk of dying was reduced only slightly after controlling for race/ethnicity (HR, 1.17; 95% CI, 1.10–1.24). Compared with Caucasian patients with colon cancer, African-American patients were 21% more likely to die after controlling for age, sex, comorbidity scores, tumor stage, and grade (HR, 1.21; 95% CI, 1.12–1.30). After also adjusting for definitive therapy and socioeconomic status, the HR of mortality was only marginally significantly higher in African Americans compared with Caucasians for all-cause mortality (HR, 1.10; 95% CI, 1.02–1.19) and colon cancer-specific mortality (HR, 1.16; 95% CI, 1.01–1.33).

CONCLUSIONS.

Lower socioeconomic status and lack of definitive treatment were associated strongly with decreased survival in both men and women with colon cancer. Racial disparities in survival were explained substantially by differences in socioeconomic status. Cancer 2007. © 2007 American Cancer Society.

Colorectal cancer is the third most commonly diagnosed cancer (excluding skin cancer) and the second most common cause of cancer death in the U.S.1–4 It is estimated that 148,610 men and women will be diagnosed and 55,170 men and women will die of cancer of the colon and rectum in 2006.1, 2 Although overall colon cancer mortality has decreased over time,1–4 incidence and mortality still varies greatly by race/ethnicity.1–8 A number of studies have indicated that the increased mortality in African-American patients with colon cancer can be attributed to more aggressive cancer and more advanced stage at diagnosis,9–12 differences in treatment,13–19 provider characteristics,20, 21 screening, and posttreatment surveillance.2, 8, 22–24 For example, several population-based studies demonstrated that African Americans with stage III colon cancer were less likely than Caucasians to undergo definitive surgery and receive adjuvant chemotherapy as part of standard therapy.16–18, 25 The racial/ethnic disparities also have been attributed to the quality and tolerance of treatment and posttreatment surveillance.8 A relatively limited number of studies examined the impact of socioeconomic factors on the treatment and survival of patients with colon cancer,26–40 and fewer studies addressed the racial disparities in survival for colon cancer by adequately incorporating treatment and socioeconomic factors32, 34, 37, 39 in addition to factors on cancer stage, grade, or comorbidity. Among this limited number of studies regarding socioeconomic factors and survival for colon cancer, the findings have been inconsistent.29–38 The results from 3 studies indicated that significant racial/ethnic disparities in survival between African Americans and Caucasians still existed after adjusting for socioeconomic factors,30, 31, 39, 41 whereas findings from 7 other studies indicated that these disparities were not significant after adjustment.29, 32–38, 40, 41 Additional studies that consider key variables, such as treatment and socioeconomic status (SES), conducted in larger population-based settings should increase our understanding of racial disparities. We examined the effect of race/ethnicity, treatment and socioeconomic factors on long-term survival in a large, nationwide, and population-based cohort of men and women who were diagnosed with stage II or III colon cancer at age ≥65 years. This cohort was identified from the Surveillance, Epidemiology, and End Results (SEER) cancer registries-Medicare linked databases and had up to 11 years of follow-up. These linked data not only provided reliable information concerning tumor stage and grade at diagnosis and long-term follow-up of vital status but also allowed the examination of various socioeconomic factors at the Census tract level. In addition, comorbid conditions and adjuvant chemotherapy as a standard care for stage III colon cancer could be identified uniquely and accurately from Medicare claims data. We hypothesized in this study that there was no racial/ethnic difference in the long-term survival of patients with colon cancer after controlling for differences in age, sex, tumor stage, comorbidity, treatment, and SES.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Data Sources

We used the SEER-Medicare linked data for patients who were diagnosed with colon cancer from 1992 to 1999. The SEER Program, which is supported by the National Cancer Institute, includes population-based tumor registries in 11 selected geographic areas,1, 3, 4, 42 including the metropolitan areas of San Francisco/Oakland, Detroit, Atlanta, and Seattle; Los Angeles County; the San Jose-Monterey area; and the states of Connecticut, Iowa, New Mexico, Utah, and Hawaii, covering >14% of the U.S. population. Medicare eligibility could be identified for 94% of the individuals aged ≥65 years who appeared in the SEER records. The method of linking these data has been described elsewhere.42 The Committee for Protection of Human Subjects at the University of Texas Health Science Center approved this study (HCS-SPH-05-0455).

Study Population

We used the analytic SEER-Medicare files that excluded patients who did not have full coverage of both Medicare Parts A and B or who were members of a Health Maintenance Organization in the year when their diagnosis was made to ensure the completeness of Medicare claims. The study population consisted of 18,492 men and women who were diagnosed with incident American Joint Committee on Cancer stage II or III colon cancer at age ≥65 years between 1992 and 1999 that was their first and only primary tumor. Of the 18,492 individuals, 15,913 were Caucasians (non-Hispanic whites), 1320 were African Americans (non-Hispanic blacks), and 1259 had other ethnicities, which were combined because of small numbers.

Study Variables

Outcome variables

All-cause mortality was defined as death from any cause, which was identified by the SEER Program through linking SEER data with National Death Index data. Patients who remained alive at their last follow-up were censored. Colon cancer-specific mortality was defined if colon cancer was the underlying cause of death. In this specific analysis, patients who died of causes other than colon cancer or who remained alive at the date of last follow-up were censored. Survival was calculated in months from the date of diagnosis to the date of death or to the date of last follow-up (December 31, 2002).

SES

Three variables at the Census tract level from the 1990 Census available in the SEER-Medicare linked data were used to define SES. These SES variables were recorded and arbitrarily categorized into quartiles, so that the first quartile represents the highest SES and the fourth quartile represents the lowest (or poorest) SES. The 3 variables were: 1) education, ie, the percentage of adults aged ≥25 years who had <12 years of education (first quartile, ≤11.83%; second quartile, 11.84–19.02%; third quartile, 19.03–26.90%; and fourth quartile, ≥26.91%); 2) poverty, ie, the percentage of individuals living below the poverty line (first quartile, ≤3.91%; second quartile, 3.92–7.21%; third quartile, 7.22–13.08%; and fourth quartile, ≥13.09%); 3) income, ie, the median annual household income (first quartile, ≥$43,861; second quartile, $34,030–43,860; third quartile, $25,735–34,029; and fourth quartile, ≤$25,734). For elderly Medicare beneficiaries, the Poverty Level could be the most directly relevant proxy measure of economic status.43 In addition, we created a composite variable using these 3 socioeconomic variables that were weighted equally. A small proportion of individuals (n = 257 patients) who had missing information regarding SES were excluded from the analysis.

Comorbidity Index

Comorbidity was ascertained from Medicare claims by identifying 18 diagnoses or related procedures recorded between 1 year prior to and 1 month after the diagnosis of colon cancer. Details concerning creating a weighted comorbidity score have been previously reported.44

Treatment

Surgery and radiation therapy

Patients were defined as having undergone definitive surgical treatment if they received total colectomy (SEER codes 50–70), less than local colectomy (codes 30–40), or surgical excision with pathology specimen (including polypectomy, snare, or laser surgery; code 20)45 or if there was a Medicare claim for resection (International Classification of Diseases-9th Revision procedure codes46 45.71–45.76, 45.79–45.89, 48,41, and 48,49–48.69 or Common Procedure Terminology codes47 44140–44160 and 45383–45385). Patients were defined as having received radiation therapy if it was indicated either in the SEER data or on Medicare claims.48

Chemotherapy

The detailed methods for the identification of chemotherapy use through Medicare claims have been described previously.49–52 Definitive therapy was defined as the receipt of surgeries, as defined above, for stage II colon cancer or the receipt of surgeries plus chemotherapy for stage III colon cancer.53, 54

Other Characteristics

Patient and tumor characteristics (age at diagnosis, sex, race/ethnicity, tumor stage, and grade), year of diagnosis, and geographic area were available from the SEER data.

Analysis

Differences in the distribution of baseline characteristics among the 3 racial/ethnic groups were tested using the chi-square statistic. A Cox proportional-hazards regression model was used for the analysis of survival using the PHREG procedure in the SAS system.55 The proportionality assumption was considered to be satisfied when the log-log Kaplan-Meier curves for survival functions by race/ethnicity or SES were parallel and did not intersect.56, 57 The hazards ratio (HR) generated from the Cox survival model indicated the risk of death in African Americans compared with Caucasians (reference group).41, 55–57 The interaction between race/ethnicity and SES was tested using the product term of these 2 variables in the model. Analyses were adjusted for age, tumor stage, grade, treatment, comorbidity score, year of diagnosis, and geographic area.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Table 1 presents the distribution of patient age, sex, tumor characteristics, and receipt of definitive therapy among 3 racial/ethnic groups in 18,492 patients who were diagnosed with stage II or III colon cancer. The median age at diagnosis was 77 years for Caucasians, 75 years for African Americans, and 76 years for others. Higher proportions of patients were diagnosed at age ≥80 years in Caucasians (39.8%) compared with African Americans (30.9%) and others (32.2%). Slightly larger proportions of Caucasian patients had lower comorbidity scores and poorly differentiated tumor grade compared with African-American patients. The percentage of patients receiving definitive therapy was higher in Caucasians (81.9%) compared with African Americans (77.8%).

Table 1. Comparison of Demographic and Tumor Characteristics and Type of Treatment Among Different Racial/Ethnic Groups in Patients With Colon Cancer*
CharacteristicCaucasians 77 (65–108)African Americans 75 (65–98)Others 76 (65–102)
No.%No.%No.%
  • *

    All chi-square tests for the distribution of other factors among 3 racial groups were significant (P< .01).

Median age (range)
Age, y
 65–69247415.628021.221717.2
 70–74340321.433325.232425.7
 75–79370423.329922.731324.9
 ≥80633239.840830.940532.2
Sex
 Men648040.747936.358946.8
 Women943359.384163.767053.2
Tumor stage
 II949459.77525769154.9
 III641940.35684356845.1
Tumor grade
 Well-differentiated10726.7967.3665.2
 Moderately differentiated10,36165.192870.388270.1
 Poorly differentiated38122421816.527221.6
 Unknown/missing6684.2785.9393.1
Comorbidity score
 0765548.15944567253.4
 1457028.734025.833426.5
 2211913.320415.515011.9
 ≥315699.918213.81038.2
Definitive therapy
 No288218.129322.227021.5
 Yes13,03181.9102777.898978.5
Total15,91310013201001259100

Table 2 presents the distribution of SES among the 3 racial/ethnic groups. Large proportions of African-American patients with colon cancer were in the poorest quartiles of SES compared with Caucasians. Differences between Caucasians and African Americans were statistically significant. For example, 73.1% of African Americans were in the poorest quartile of SES, as measured by the Poverty Level, compared with 20.2% of Caucasians. The distribution of ethnic population was similar when other measures of SES were used (education, income, and composite SES).

Table 2. Comparison of Socioeconomic Status Among Different Racial Groups in Patients With Colon Cancer*
SESCaucasiansAfrican AmericansOthers
No.%No.%No.%
  • SES indicates socioeconomic status.

  • *

    All chi-square tests for the distribution of socioeconomic variables among 3 racial groups were significant (P <.001).

Education
 First quartile431327.1806.123018.3
 Second quartile423826.613910.524219.2
 Third quartile417826.318013.627021.5
 Fourth quartile31842092169.851741.1
Poverty
 First quartile433627.3644.923919
 Second quartile422126.5987.429723.6
 Third quartile41392619314.628322.5
 Fourth quartile321720.296573.144035
Income
 First quartile419426.4886.734427.3
 Second quartile414726.114110.733226.4
 Third quartile406425.529122.126821.3
 Fourth quartile35082280060.631525
Composite SES
 First quartile428126.9735.527021.5
 Second quartile422626.61229.227121.5
 Third quartile41402620615.628222.4
 Fourth quartile326620.591969.643634.6
Total15,91310013201001259100

Table 3 presents the 3-, 5-, and 10-year observed survival by racial/ethnic group and SES. The 3-year overall survival rate was 62.1% in Caucasians, 56.1% in African Americans, and 64.8% in others. Survival increased with improving SES. For example, the 3-year survival rate was 57.5% for patients who lived in the community with the lowest education level and 65.7% for patients in the highest quartile of education. The improvement in survival associated with higher SES was consistent and statistically significant when other measures of SES were used (poverty, income, and composite SES). This survival pattern in association with race/ethnicity and SES was similar at the 5- and 10-year levels and for colon cancer-specific survival (Table 3).

Table 3. All-cause and Colon Cancer-specific Observed Survival in Patients With Stage II/III Colon Cancer by Race/Ethnicity and Socioeconomic Status
Race/Ethnicity and SESThree-year survival: Patients from 1992 to 1999 (n=18,492), %Five-year survival: Patients from 1992 to 1997 (n=14,020), %Ten-year survival: Patients from 1992 to 1993 (n=4803), %
All causeColon cancer-specificAll causeColon cancer-specificAll causeColon cancer-specific
  1. SES indicates socioeconomic status.

Racial/ethnic group
 Caucasian62.183.649.776.729.771.6
 African American56.181.243.472.328.465.4
 Others64.884.851.974.228.665.9
Education
 First quartile65.784.952.678.131.973
 Second quartile6284.35176.931.571.7
 Third quartile62.383.349.976.329.871.1
 Fourth quartile57.581.544.273.825.267.9
Poverty
 First quartile64.284.95277.731.971
 Second quartile62.483.950.277.330.873.4
 Third quartile62.282.949.675.229.871.3
 Fourth quartile58.782.445.774.825.867.8
Income
 First quartile65.485.853.379.133.673.2
 Second quartile61.383.848.976.928.372.8
 Third quartile618247.97428.668
 Fourth quartile59.982.547.57527.769.5
Composite SES
 First quartile65.885.653.478.733.472.3
 Second quartile61.684.149.477.329.672.7
 Third quartile61.382.549.375.128.871
 Fourth quartile58.881.945.57426.667.6
Total61.983.549.476.229.670.9

Table 4 presents the effect of SES on all-cause mortality and colon cancer-specific mortality adjusted for race/ethnicity and other factors. There was a clear pattern of increased HR of mortality with lower SES. The magnitude and trend in HR were consistent regardless of which socioeconomic variables were used. For example, patients who lived in communities with the lowest SES were 23% more likely to die from all causes than patients who lived in communities with the highest SES without adjusting for patient and tumor factors (HR, 1.23; 95% confidence interval [95% CI], 1.17–1.30) and were 21% more likely to die after adjusting for these factors (HR, 1.21; 95% CI, 1.14–1.27) (Model 1). The risk of all-cause mortality was reduced only slightly after controlling for race/ethnicity (HR, 1.19; 95% CI, 1.12–1.26) (Model 2). Similar patterns were observed for colon cancer-specific mortality in association with socioeconomic factors. Individuals who lived in communities with the lowest SES were considerably more likely to die of colon cancer than individuals who lived in communities with the highest SES (Model 4).

Table 4. Association Between Socioeconomic Status and Mortality in Patients With Colon Cancer
SESMortality: HR (95% CI)*
All-cause mortalityColon cancer-specific mortality
Model 1Model 2Model 3Model 4
  • HR indicates hazards ratio; 95% CI, 95% confidence interval; SES, socioeconomic status.

  • *

    Models 1 and 3 were adjusted for age; sex; comorbidity; tumor stage; grade; year of diagnosis; Surveillance, Epidemiology, and End Results region; optimal therapy; and radiation. Models 2 and 4 were adjusted for the same factors with the addition of race.

Education
 First quartile1.001.001.001.00
 Second quartile1.05 (0.99–1.10)1.05 (0.99–1.11)1.08 (0.98–1.19)1.08 (0.98–1.19)
 Third quartile1.07 (1.01–1.13)1.07 (1.01–1.13)1.13 (1.02–1.25)1.12 (1.01–1.24)
 Fourth quartile1.20 (1.14–1.27)1.19 (1.12–1.25)1.29 (1.17–1.42)1.25 (1.13–1.38)
Poverty
 First quartile1.001.001.001.00
 Second quartile1.03 (0.98–1.09)1.03 (0.98–1.09)1.01 (0.92–1.12)1.01 (0.91–1.11)
 Third quartile1.08 (1.02–1.15)1.08 (1.02–1.14)1.11 (1.00–1.22)1.09 (0.98–1.20)
 Fourth quartile1.17 (1.10–1.23)1.14 (1.07–1.21)1.14 (1.03–1.27)1.08 (0.97–1.21)
Income
 First quartile1.001.001.001.00
 Second quartile1.08 (1.03–1.14)1.08 (1.02–1.14)1.11 (1.01–1.23)1.10 (1.00–1.22)
 Third quartile1.13 (1.07–1.19)1.12 (1.06–1.18)1.24 (1.12–1.37)1.21 (1.10–1.34)
 Fourth quartile1.20 (1.13–1.27)1.17 (1.10–1.25)1.24 (1.12–1.38)1.18 (1.05–1.32)
Composite SES
 First quartile1.001.001.001.00
 Second quartile1.08 (1.02–1.14)1.08 (1.02–1.14)1.07 (0.97–1.18)1.06 (0.96–1.17)
 Third quartile1.13 (1.07–1.19)1.12 (1.06–1.18)1.20 (1.09–1.33)1.19 (1.07–1.31)
 Fourth quartile1.21 (1.14–1.27)1.19 (1.12–1.26)1.26 (1.14–1.39)1.21 (1.09–1.35)

The effects of race/ethnicity on the risk for all-cause mortality and colon cancer-specific mortality are presented in 6 different statistical models in Table 5. Compared with Caucasian patients who had colon cancer, African-American patients were 21% more likely to die of all causes after controlling for age, sex, comorbidity scores, tumor stage, grade, year of diagnosis, and geographic area (HR, 1.21; 95% CI, 1.12–1.30). After adjusting for definitive therapy, the risk of dying was reduced slightly in African Americans compared with Caucasians for all-cause mortality (HR, 1.18; 95% CI, 1.10–1.27) and colon cancer-specific mortality (HR, 1.26; 95% CI, 1.11–1.43) (Models 2 and 5). When also adjusting for SES without treatment factors, greater reductions in the risk of mortality were observed for all-cause mortality (HR, 1.12; 95% CI, 1.04–1.21) and colon cancer-specific mortality (HR, 1.17; 95% CI, 1.02–1.34). And, when controlling for both SES and treatment, the HR of mortality was only marginally significantly higher in African Americans compared with Caucasians for all-cause mortality (HR, 1.10; 95% CI, 1.02–1.19) and colon cancer-specific mortality (HR, 1.16; 95% CI, 1.01–1.33) (Models 3 and 6). Patients with other ethnicity had a significantly lower risk of all-cause mortality but an insignificantly reduced risk of colon cancer-specific mortality.

Table 5. Effect of Race/Ethnicity and Socioeconomic Status on Mortality in Patients With Colon Cancer
Race/Ethnicity and other factorsMortality: HR (95% CI)*
All-cause mortalityColon cancer-specific mortality
Model 1Model 2Model 3Model 4Model 5Model 6
  • HR indicates hazards ratio; 95% CI, 95% confidence interval; SES, socioeconomic status.

  • *

    Models 1 and 4 were adjusted for the variables age; sex; comorbidity; tumor stage; grade; year of diagnosis; Surveillance, Epidemiology, and End Results region; optimal therapy; and radiation.; Models 2 and 5 were adjusted the same factors with the addition of treatment; and Models 3 and 6 were adjusted for the same factors with the addition of treatment and composite SES.

Race/ethnicity
 Caucasian1.001.001.001.001.001.00
 African American1.21 (1.12–1.30)1.18 (1.10–1.27)1.10 (1.02–1.19)1.29 (1.13–1.46)1.26 (1.11–1.43)1.16 (1.01–1.33)
 Others0.97 (0.89–1.06)0.97 (0.89–1.06)0.94 (0.86–1.03)1.16 (1.00–1.34)1.17 (1.01–1.35)1.12 (0.97–1.30)
Age, y
 65–691.001.001.001.001.001.00
 70–741.22 (1.14–1.31)1.21 (1.12–1.29)1.20 (1.12–1.29)1.11 (0.99–1.25)1.10 (0.98–1.23)1.09 (0.98–1.23)
 75–791.62 (1.52–1.74)1.56 (1.46–1.67)1.56 (1.46–1.67)1.29 (1.16–1.45)1.23 (1.10–1.38)1.23 (1.10–1.38)
 ≥802.58 (2.42–2.75)2.30 (2.16–2.45)2.29 (2.15–2.45)1.79 (1.61–1.98)1.57 (1.41–1.75)1.57 (1.41–1.74)
Sex
 Men1.001.001.001.001.001.00
 Women0.86 (0.83–0.89)0.86 (0.82–0.89)0.85 (0.82–0.88)0.99 (0.93–1.07)0.99 (0.92–1.06)0.99 (0.92–1.06)
Tumor stage
 II1.001.001.001.001.001.00
 III1.53 (1.47–1.59)1.18 (1.13–1.24)1.19 (1.13–1.25)2.25 (2.10–2.41)1.76 (1.62–1.92)1.77 (1.63–1.92)
Tumor grade
 Well-differentiated1.001.001.001.001.001.00
 Moderately differentiated0.97 (0.90–1.05)0.98 (0.91–1.06)0.98 (0.91–1.06)0.93 (0.81–1.08)0.95 (0.82–1.09)0.95 (0.82–1.09)
 Poorly differentiated1.23 (1.14–1.34)1.24 (1.15–1.35)1.24 (1.15–1.35)1.56 (1.34–1.81)1.57 (1.36–1.82)1.58 (1.36–1.83)
 Unknown/missing1.34 (1.20–1.49)1.30 (1.16–1.45)1.29 (1.15–1.44)1.67 (1.38–2.02)1.61 (1.33–1.95)1.60 (1.33–1.94)
Comorbidity score
 01.001.001.001.001.001.00
 11.35 (1.30–1.42)1.34 (1.28–1.40)1.34 (1.28–1.40)1.02 (0.95–1.11)1.02 (0.95–1.11)1.02 (0.95–1.10)
 21.67 (1.58–1.76)1.64 (1.56–1.74)1.64 (1.55–1.73)0.99 (0.89–1.11)0.98 (0.88–1.09)0.98 (0.88–1.09)
 ≥32.38 (2.25–2.53)2.30 (2.16–2.44)2.28 (2.15–2.42)1.16 (1.03–1.32)1.11 (0.98–1.26)1.10 (0.97–1.25)
Definitive treatment
 No1.001.001.001.00
 Yes0.59 (0.56–0.63)0.59 (0.56–0.63)0.60 (0.55–0.66)0.60 (0.55–0.66)
Composite SES
 First quartile1.001.00
 Second quartile1.08 (1.02–1.14)1.06 (0.96–1.17)
 Third quartile1.12 (1.06–1.18)1.19 (1.07–1.31)
 Fourth quartile1.19 (1.12–1.26)1.21 (1.09–1.35)

Table 5 also presents the effects of age, sex, tumor stage, grade, comorbidity, and definitive therapy on mortality. The HR of all-cause and colon cancer-specific mortality, as expected, increased significantly with age and comorbidity. Women had a risk of colon cancer-specific mortality similar to that of men but had a significantly lower risk of dying from all causes. Patients who received the recommended definitive therapy were 49% less likely to die of all causes and 40% less likely to die specifically of colon cancer after adjusting for demographic factors, sex, tumor stage, grade, comorbidity, year of diagnosis, and geographic area. These HRs remained unchanged after also adjusting for socioeconomic factors. In addition, there were no significant interactions between race/ethnicity and socioeconomic variables and no significant interactions between race/ethnicity and definitive therapy on the risk of all-cause and colon cancer-specific mortality.

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

We observed that lower SES was associated significantly with decreased survival, even after controlling for race/ethnicity, other patient/tumor characteristics, and definitive treatment. Although there were racial/ethnic disparities in survival, these differences were reduced substantially and were elevated only marginally after controlling for treatment and socioeconomic factors.

The differences in survival between African-American and Caucasian patients with cancer have been attributed to numerous factors.9–25 Although racial/ethnic differences are most likely multifactorial, it is known that access to quality care, the receipt of definitive treatment, and socioeconomic factors have played major roles.26–40 Several studies have demonstrated that, if patients had equal access to quality health care, then the outcomes would be similar among different racial groups.31, 33, 36, 38, 40 However, other studies have indicated that racial disparities still exist, even after controlling for socioeconomic factors and for access to equitable care and treatment.30, 32, 34, 37, 39, 58 Our results indicated that, among patients who had the same coverage of Medicare fee-for-service insurance, African Americans were more likely to present with higher grade cancer. Even though definitive treatment was associated with significantly increased survival, racial disparities in survival in this cohort were reduced only slightly after adjusting for treatment. This likely was because racial differences in the receipt of definitive treatment were small. We observed that the receipt of definitive treatment varied only slightly among racial/ethnic groups (81% for Caucasians, 78% for African-Americans, and 78% for others). This small difference in receiving the standard of care among ethnic groups with colon cancer, unlike the larger racial gap in other cancers or other diseases,19 is encouraging at a time when our society is making efforts toward eliminating racial disparities in health care.

The HRs showed a greater reduction after controlling for socioeconomic factors, indicating that SES is a strong confounding factor in the association between race/ethnicity and outcome/survival.59–64 Several studies have indicated65–68 that race/ethnicity may be only a surrogate measure of SES in which a large majority of African-American patients lived in the areas with poorer SES. Therefore, taking socioeconomic factors into account in scientific endeavors becomes critical for addressing racial disparities. Furthermore, treatment and socioeconomic differences are 2 of the major barriers to achieving equal outcomes for men and women with colon cancer and are modifiable factors, whereas race/ethnicity is not modifiable. Hence, making efforts to provide the standard of care; to achieve equal opportunity for education, employment, and health insurance; to eliminate discrimination based on race/ethnicity and low social class; and to educate individuals for a healthy lifestyle and better health knowledge may have great public health implications.

Our study had a number of strengths. First, it was a large, nationwide, and population-based study that covered patients with incident colon cancer in the 11 SEER regions across the U.S. Diagnoses were confirmed pathologically by the SEER registries, which are among the most authoritative data sources regarding cancer. SEER registries also provided reliable information concerning tumor stage, grade, and long-term follow-up of vital status. Medicare claims enabled us to identify information regarding patient comorbidity, which is a strong confounder of survival. In addition, we were able to identify the receiptof chemotherapy uniquely and reliably from Medicare claims data.50 Information concerning chemotherapy is critically important in examining survival, because adjuvant chemotherapy after resection is the standard of care for patients with stage III colon cancer.53, 54 Consequently, definitive therapy for both stage II and stage III colon cancer can be established reliably.

There were several limitations to the current study. First, socioeconomic variables were based on the Census tract level. It is possible that residual confounding from individual SES was not controlled in our analyses. To address the limitation of using a single measure of SES,60 we used 3 different measures. The similarities of results across these measures strengthened the study findings. Second, the small remaining differences in survival between racial groups after adjusting for socioeconomic factors may have been caused by racial differences in unmeasured factors, such as provider and physician characteristics.24 For example, a recent study reported that ethnic disparities in receiving poorer quality health care may be affected by the chosen physicians.21 Third, our findings may have been affected by the lack of information regarding patient's personal health beliefs, lifestyle factors, colon cancer screening, and posttreatment surveillance in particular, because these factors are associated with ethnicity and survival, as indicated in a health disparity model.8 Fourth, although Medicare claims for chemotherapy have been validated externaly,50 it still may be possible that some chemotherapy claims might not have been captured by Medicare data. More recently, oxaliplatin-containing chemotherapy became part of standard care for stage III colon cancer,69 but our study patients with this stage disease were diagnosed several years before the new guideline. Fourth, we only studied Medicare beneficiaries aged ≥65 years. The results for the association between race/ethnicity and survival may not be generalizable to younger patients.

In conclusion, lower SES and the lack of definitive treatment were associated strongly with decreased survival in both men and women with colon cancer. Among these modifiable factors, the racial/ethnic disparities in survival between African Americans and Caucasians were explained substantially by ethnic differences in SES rather than by race itself. The current findings have important public health implications for achieving the goals of Healthy People 2010,70 because socioeconomic and treatment factors are modifiable through public health and clinical interventions. Further studies may be needed to address whether this association is true in younger populations and in different areas.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

We acknowledge the efforts of the National Cancer Institute; the Centers for Medicare and Medicaid Services; Information Management Systems; and Surveillance, Epidemiology, and End Results registries in the creation of the database that was used in the current study.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES