| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Robert Wood Johnson Clinical Scholars Program, the Department of Medicine, Yale University School of Medicine, New Haven, Connecticut; and the Clinical Epidemiology Unit, West Haven Veterans Affairs Medical Center, West Haven, Connecticut.
Address reprint requests to: Elizabeth Howell, MD, MPP Robert Wood Johnson Clinical Scholars Yale University School of Medicine IE-61 SHM, 333 Cedar Street New Haven, CT 06520-8025 E-mail: elizabeth.howell{at}yale.edu
 |
Abstract |
|---|
Top
 Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Methods: Using data obtained from the Surveillance, Epidemiology, and End Results (SEER) Program for 1988–1994, we determined the associations between race and stage, and race and treatment. Racial differences in survival for up to 7 years of follow-up were adjusted for age, marital status, SEER location, International Federation of Gynecology and Obstetrics (FIGO) stage of disease, lymph node status, grade, histology, and treatment.
Results: Cumulative mortality was 36% (366 deaths in 1029 women) for black women and 24% (1215 deaths in 5021 women) for white women; unadjusted hazard ratio was 1.60 (95% confidence interval [CI] 1.43, 1.80). Black women were more likely to present with advanced disease than white women (43.8% compared with 34.8%). In a model adjusting for demographics and FIGO stage, the hazard ratio for black women compared with white women decreased to 1.35 (95% CI 1.19, 1.54). Treatment varied by race, with black women receiving surgery less often (33.5% compared with 48.2%, respectively) and radiation therapy more often (35.3% and 25.2%, respectively) than white women. In a comprehensive model including demographic factors, FIGO stage, other tumor characteristics, and treatment, the adjusted hazard ratio for mortality remained high for black women at 1.30 (95% CI 1.14, 1.48).
Conclusion: Race remains an independent predictor of cervical cancer survival after accounting for age, stage of disease, treatment patterns, and other factors. Future studies should assess racial differences in clinical severity of disease, comorbidity, and socioeconomic status.
Racial differences in mortality in the United States exist and the mortality gap between blacks and whites is often large.1 As the third-most-common malignancy of the female genital system in the United States, with approximately 16,000 new cases and 5000 deaths each year,2 cervical cancer contributes to this mortality gap.1,3 Black women are twice as likely to be diagnosed with cervical cancer (14.3 compared with 7.9 per 100,000 person-years) and two to three times more likely to die from their disease (6.9 compared with 2.6 per 100,000).3
Previous studies of racial differences in cervical cancer survival have suggested that the higher mortality in blacks is explained primarily by the more advanced clinical stage at time of diagnosis.4,5 Lack of access to screening services and barriers to care4 have been suggested as explanations for the more advanced stage at presentation and consequently the higher mortality of black women with cervical cancer. Chen et al4 found that black women had significantly lower percentages of early stage diagnoses of cervical cancer compared with white women, and the percentage of cervical cancers diagnosed late decreased from 7.2% to 5.5% between 1976 and 1990 for white women, whereas the percentage diagnosed late for black women increased from 8.7% to 13.6% during the same time period.6
Although age, histology, grade, tumor size, depth of invasion, and nodal status7,8 have also been reported as prognostic factors for cervical cancer survival, those possible prognostic markers and others have not been well studied as contributions to racial differences in survival. For example, other possible explanations for higher mortality rates among black women with cervical cancer include less aggressive treatment patterns, higher comorbidity, greater clinical severity of disease, and lower socioeconomic status.
Investigations of patients with cardiovascular disease have revealed that treatment patterns vary according to race9–11 and contribute to survival differences between whites and blacks.11 The purpose of the current study was to examine the impact of differences in International Federation of Gynecology and Obstetrics (FIGO) stage and treatment patterns on mortality among black and white women with cervical cancer.
|
Materials and Methods |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
The source patient population included all microscopically confirmed cases of primary invasive cervical cancer between 1988 and 1994 (n = 7443). Women who had cervical cancer before 1988 were not included because FIGO staging was not listed in the SEER database. Because the focus of our analysis was on white and black differences, we excluded women of other races (n = 1386) from analysis. Also excluded were patients with cervical cancer identified by autopsy or death certificate only and patients with unknown survival time (n = 5), as were subjects less than 15 years old (n = 2). Therefore the final study sample for our analyses included 6050 women.
Cancer-directed treatment was classified in the following five categories: no therapy, surgery only, radiation only, radiation and surgery combined, and unknown therapy. Surgery included cone biopsy, hysterectomy (radical or any other type), nodal surgery, or pelvic exenteration. Radiation included external beam radiation or intracavitary implants. More than 95% of patients who received radiation therapy received external beam therapy either alone or in combination with intracavitary implants or other forms of therapy (eg, surgery).
Other variables were also coded according to SEER Program criteria. Marital status was categorized as single, married, separated or divorced or widowed, or unknown; and geographic location was divided into the nine SEER locations. Tumor grade was classified as well differentiated, moderately differentiated, poorly differentiated, undifferentiated, or unknown. Lymph node status was classified as negative, positive, or unknown. Tumor stage was defined with the FIGO system as IA, IB, IIA, IIB, III, or IV; Stage I not otherwise specified was also included in our analyses. The main outcome variable was overall mortality rate between 1988 and 1994, as verified by the SEER Program vital status determination, including data from local death certificate, voter registration, driver’s license, and medical records.13 Results for cause-specific mortality were similar but are not reported. The SEER database did not contain information on comorbidity, clinical severity of disease, or socioeconomic status.
All statistical analyses were done using PC SAS version 6.12 (SAS Institute Inc., Cary, NC). Bivariate analyses with 
2 or Fisher exact tests were used to evaluate the association between race and stage, and race and type of treatment. Cross-stratification tables were used to assess treatment differences based on race within each stage. In addition, polytomous logistic regression models were used (because of the five-category outcome variable) to assess whether race was an independent predictor of treatment type, adjusting for age and stage.
Bivariate analyses with 2 or Fisher exact tests were used to evaluate the association of race, age, marital status, geographic location, stage of disease, nodal status, grade, histology of tumor, and treatment with overall mortality rate, as a dichotomous variable. Cross-stratification tables were used to examine possible interactions between race and stage in predicting mortality rates. Multivariate regression models were developed to assess the independent impact of race, other patient demographics, tumor characteristics, and type of treatment on overall mortality rate. Because the follow-up duration for each patient was different (mean follow-up time 31.3 months), Cox proportional hazard analysis was used to model survival for the 6050 patients. This approach calculates hazard ratios that approximate relative risks.
|
Results |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Black women were more likely to receive no therapy than white women (Table 1
). Among the 454 patients who did not receive therapy, 45 (10%) refused surgery or radiation, including 27 of 347 (7.8%) white women and 18 of 107 (16.8%) black women. The remaining 409 patients who did not receive therapy had either contraindications to surgery or no cancer-directed surgery for unknown reasons. The SEER public-use tape only indicated radiation therapy refusals but did not indicate other reasons for not receiving radiation.
|
). Black women were less likely to receive surgery alone (33.5% compared with 48.2%, respectively), however, and more likely to receive radiation alone (35.3% compared with 25.2%, respectively) than white women. In the analysis stratified by stage (Table 2), approximately 94% of both black women and white women received surgery in Stage IA. In Stage IB, Stage I not otherwise specified, and Stage IIA, however, black women received surgery less often than white women (50.8% and 63.1%, respectively, for Stage IB; 43.4% and 56.7%, respectively, for Stage I not otherwise specified; and 7.0% and 15.5%, respectively, for Stage IIA). In Stages IIB through IV, most black women and white women received radiation therapy alone or in combination with surgery.
|
Cumulative mortality up to 7 years after diagnosis was 35.6% for black women compared with 24.2% for white women (unadjusted hazard ratio 1.60, 95% confidence interval [CI] 1.43, 1.80). Bivariate associations between patient characteristics and mortality are shown in Table 3. The unadjusted mortality rates were 6.0% in patients treated with surgery, 52.3% with radiation, and 26.8% with combination therapy. The mortality rate increased with increasing tumor grade, positive nodal status, and increasing FIGO stage (from 2.3% for Stage IA to 76.5% for Stage IV). Analyses stratified by stage (data not shown) showed similar quantitative racial differences in mortality rates throughout most FIGO stages, although the differences were statistically significant only for Stage IB and Stage III.
|
). The Cox model demonstrated that the strongest quantitative predictor of death was stage at the time of diagnosis, with more advanced stages having higher mortality rates. Treatment type was also significantly associated with mortality in this model: surgical therapy, radiation therapy, and combination therapy were all protective compared with patients who did not receive therapy, adjusting for the other factors in the model. Patients with adenosquamous cell histology had higher mortality rates than those with squamous cell histology, and patients with unknown nodal status had higher mortality rates than those with negative nodal status.
|
). After adjusting for demographic characteristics, the hazard ratio for the impact of race (black compared with white) on mortality rate decreased from 1.60 to 1.40 (95% CI 1.23, 1.59). When FIGO stage was included, the hazard ratio for race decreased to 1.35 (1.19, 1.54). After other tumor characteristics and treatment were included in the survival model, the hazard ratio for race remained high at 1.30 (95% CI 1.14, 1.48).
|
|
Discussion |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
The present study also found racial variation in treatment patterns for cervical cancer. Black women with cervical cancer were less likely to receive surgery and more likely to receive radiation than white women, even after controlling for stage of disease. These treatment variations were associated with differences in mortality rates between blacks and whites in our analysis, after accounting for the effects of FIGO stage, histology, grade, and nodal status. The treatment pattern, however, only partially accounted for racial differences in survival.
Although the literature shows that either radiation therapy or surgery are both therapeutic options for early stage cervical cancer and have similar survival rates,14 women in this analysis who received radiation had higher mortality rates than women who received surgery. This result, however, is likely biased by the lack of information about other important clinical factors such as tumor size, an important prognostic factor7 that was missing for many of the patients in the SEER public-use tape. In addition, the SEER public-use tape does not include information on the hospital in which therapy was given, the skill of the physician, radiation doses, or other important treatment information.
Several possible explanations could account for the residual differences in mortality rates between black and white women with cervical cancer, after adjusting for demographic, tumor, and treatment factors. One issue is related to comorbid conditions: black women could be sicker at baseline from other medical conditions and die sooner during follow-up. In addition, comorbid conditions could explain the differences in treatment between black and white women, because older patients with comorbid disease are more often treated with radiation therapy than with surgery.15 An example of a comorbid condition that could possibly help explain differences in mortality rates between blacks and whites is human immunodeficiency virus (HIV) status, which has a higher incidence in blacks than whites. The SEER public-use tape, however, does not include information on HIV status or other comorbid conditions.
A second potential explanation of the observed differences in mortality rates between white and black women with cervical cancer is socioeconomic status, a factor that is inversely related to overall mortality rate16 and cervical cancer mortality rate.5 Evidence1 supports the role of both higher comorbidity and lower socioeconomic status in explaining the higher mortality rate for blacks compared with whites.
Another potential explanation for differences in mortality rates could be that black women present later within each FIGO stage, but our analysis relied on FIGO classification to characterize extent of disease. For example, a comprehensive staging system for cervical cancer, including symptoms and comorbid conditions, predicted mortality differences within FIGO stages.15,17 Thus comorbidity, socioeconomic status, and clinical severity are factors that might account for racial differences in cervical cancer survival but are not included in the SEER public-use tape.
Our results do not address directly whether black women receive inappropriate treatment compared with white women. As mentioned earlier, both surgery and radiation are considered reasonable options for early-stage cervical cancer treatment. In this large observational database, however, the mortality rate was higher for radiation than for surgery, and blacks were more likely to receive radiation, despite controlling for FIGO stage. We need more clinical data to better understand these differences before we can draw any conclusions from this observation.
Although limitations of SEER data warrant caution in interpreting results, our findings are consistent with other evidence of racial disparity in medical treatment. For example, black women with advanced ovarian cancer were treated less often with combined surgery and chemotherapy and were half as likely to receive appropriate treatment when compared with white women.18 Racial disparity in treatment is also well documented for invasive cardiac procedures,9–11 and lower cardiac procedure usage rates in the black community are associated with blacks dying of cardiac disease more often than whites.11 Whether clinical factors, patient preferences, or physician decisions are responsible for different therapy choices for blacks and whites warrants further investigation.
Previous analyses8 of SEER data from 1973–1987 did not find race an independent predictor of cervical cancer survival. The prior study8 grouped all nonwhite races together, however, and might not have detected black versus white survival differences. The earlier publication also did not describe how the historic staging system was converted to FIGO staging. Accordingly, we did not attempt to replicate their analysis.
The current study is a comprehensive population-based analysis, providing national representation of whites and blacks, assuring generalizability of results. We observed that black women die more frequently with cervical cancer than white women, adjusting for demographic characteristics, prognostic factors (age, stage, grade, histology, and nodal status), and type of treatment. Our results have important implications as efforts are made to narrow the mortality gap in the United States, because the current focus on increased screening services alone is unlikely to completely eliminate racial differences in cervical cancer mortality rates. More comprehensive data, including measurement of socioeconomic factors, comorbid conditions, clinical, and treatment information are needed to understand fully the survival differences between the races and to improve care for all women with cervical cancer.
|
Footnotes |
|---|
Dr. Concato is supported by a Career Development Award, Department of Veterans Affairs Health Services Research and Development Service.
Received October 14, 1998. Received in revised form February 22, 1999. Accepted February 24, 1999.
|
References |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
2. American Cancer Society. Cancer facts and figures—1996. Atlanta, Georgia: American Cancer Society, 1996.
3. Miller BA, Ries LAG, Hankey BF, Kosary CL, Harras A, Devesa SS, Edwards BK eds. SEER cancer statistics review: 1973–1990. Bethesda, Maryland: National Cancer Institute, 1993.
4. Chen F, Trapido EJ, Davis K. Differences in stage at presentation of breast and gynecologic cancers among whites, blacks, and Hispanics. Cancer 1994;73:2838–42.[Medline]
5. Shelton D, Paturzo D, Flannery J, Gregorio D. Race, stage of disease, and survival with cervical cancer. Ethn Dis 1992;2:47–54.[Medline]
6. Mitchell JB, McCormack LA. Time trends in late-stage diagnosis of cervical cancer. Differences by race/ethnicity and income. Med Care 1997;35:1220–4.[Medline]
7. Sevin BU, Nadji M, Lampe B, Lu Y, Hilsenbeck S, Koechli OR, et al. Prognostic factors of early stage cervical cancer treated by radical hysterectomy. Cancer 1995;76:1978–86.[Medline]
8. Kosary CL. FIGO stage, histology, histologic grade, age and race as prognostic factors in determining survival for cancers of the female gynecological system: An analysis of 1973–87 SEER cases of cancers of the endometrium, cervix, ovary, vulva, and vagina. Semin Surg Oncol 1994;10:31–46.[Medline]
9. Whittle J, Conigliaro J, Good CB, Lofgren RP. Racial differences in the use of invasive cardiovascular procedures in the Department of Veterans Affairs medical system. N Engl J Med 1993;329:621–7.
10. Peterson ED, Wright SM, Daley J, Thibault GE. Racial variation in cardiac procedure use and survival following acute myocardial infarction in the Department of Veterans Affairs. JAMA 1994;271: 1175–80.[Abstract]
11. Peterson ED, Shaw LK, DeLong ER, Pryor DB, Califf RM, Mark DB. Racial variation in the use of coronary-revascularization procedures. Are the differences real? Do they matter? N Engl J Med 1997;336:480–6.
12. Harlan L, Brawley O, Pommerenke F, Wali P, Kramer B. Geographic, age, and racial variation in the treatment of local/regional carcinoma of the prostate. J Clin Oncol 1995;13:93–100.
13. Krongrad A, Lai H, Lamm SH, Lai S. Mortality in prostate cancer. J Urol 1996;156:1084–91.[Medline]
14. Cervical cancer. National Institutes of Health consensus statement 1996. Bethesda, Maryland: National Institutes of Health 1996;14:1–38.
15. Peipert JF, Wells CK, Schwartz PE, Feinstein AR. The impact of symptoms and comorbidity on prognosis in stage IB cervical cancer. Am J Obstet Gynecol 1993;169:598–604.[Medline]
16. Pappas G, Queen S, Hadden W, Fisher G. The increasing disparity in mortality between socioeconomic groups in the United States, 1960 and 1986. N Engl J Med 1993;329:103–9.
17. Peipert JF, Wells CK, Schwartz PE, Feinstein AR. Prognostic value of clinical variables in invasive cervical cancer. Obstet Gynecol 1994;84:746–51.
18. Parham G, Phillips JL, Hicks ML, Andrews N, Jones WB, Shingleton HM, et al. The National Cancer Data Base report on malignant epithelial ovarian carcinoma in African-American women. Cancer 1997;80:816–26.[Medline]
This article has been cited by other articles:
|
|
 |
|
  E. W. Steyerberg, B. Neville, J. C. Weeks, and C. C. Earle Referral Patterns, Treatment Choices, and Outcomes in Locoregional Esophageal Cancer: A Population-Based Analysis of Elderly Patients J. Clin. Oncol., June 10, 2007; 25(17): 2389 - 2396. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. B. Barton Exploring and crossing the disparity divide in cancer mortality. Ann Intern Med, April 18, 2006; 144(8): 614 - 616. [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Bradley, C. W. Given, and C. Roberts Health Care Disparities and Cervical Cancer Am J Public Health, December 1, 2004; 94(12): 2098 - 2103. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. O. Smith, C. R. Qualls, B. A. Prairie, L. A. Padilla, W. F. Rayburn, and C. R. Key Trends in Gestational Choriocarcinoma: A 27-Year Perspective Obstet. Gynecol., November 1, 2003; 102(5): 978 - 987. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. I. O. Garner Cervical Cancer: Disparities in Screening, Treatment, and Survival Cancer Epidemiol. Biomarkers Prev., March 1, 2003; 12(3): 242S - 247. [Full Text]
|
|
|
|
 |
|
 P. B. Bach, D. Schrag, O. W. Brawley, A. Galaznik, S. Yakren, and C. B. Begg Survival of Blacks and Whites After a Cancer Diagnosis JAMA, April 24, 2002; 287(16): 2106 - 2113. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. L. Shavers and M. L. Brown Racial and Ethnic Disparities in the Receipt of Cancer Treatment J Natl Cancer Inst, March 6, 2002; 94(5): 334 - 357. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Racial Differences Exist in Cervical Cancer Mortality Rates Journal Watch (General), October 29, 1999; 1999(1029): 5 - 5. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
