This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Gelber, R. P. Articles by Seto, T. B. Search for Related Content PubMed PubMed Citation Articles by Gelber, R. P. Articles by Seto, T. B.

Ethnic Disparities in Breast Cancer Management Among Asian Americans and Pacific Islanders

¹ Massachusetts Veterans Epidemiology Research and Information Center, VA Boston Healthcare System, Harvard Medical School, 150 S. Huntington Avenue, (151MAV), Boston, Massachusetts 02130
² Department of Medicine, University of Hawaii John A. Burns School of Medicine, 1356 Lusitania Street, Honolulu, Hawaii 96813
³ Department of Medicine, Division of General Medicine and Primary Care, Beth Israel Deaconess Medical Center, Harvard Medical School, 330 Brookline Avenue, Rose 139, Boston, Massachusetts 02215
⁴ Hawaii Medical Service Association (an independent licensee of Blue Cross Blue Shield), 818 Keeaumoku Street, Honolulu, Hawaii 96814
⁵ Center for Best Healthcare Practice, Queen’s Medical Center, 1301 Punchbowl Street, Honolulu, Hawaii 96813

Correspondence: Address correspondence and reprint requests to: Rebecca P. Gelber, MD, MPH; E-mail: rgelber{at}hsph.harvard.edu.

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Background: Little is known about breast cancer management among Asian Americans and Pacific Islanders (AAPI).

Methods: We performed a retrospective analysis of 2030 women (935 Japanese, 144 Chi-nese, 235 Filipino, 293 Hawaiian, and 423 white; mean age ± SD, 59 ± 13 years) with a diagnosis of early breast cancer (stages I, II, and IIIA) in Hawaii from 1995 to 2001. We linked data from the Surveillance, Epidemiology, and End Results program’s Hawaii Tumor Registry to administrative health care claims. We evaluated (1) breast-conserving surgery (BCS); (2) radiotherapy after BCS; and (3) chemotherapy for node-positive disease. We used logistic regression to examine the association between AAPI ethnicity and treatment, adjusting for age, year, rural residence, tumor size, grade, nodal status, receptor status, prior cancer, comorbidity index, health plan type, and income.

Results: Overall, 60.3% of women had stage I disease, 36.8% had stage II, and 2.9% had stage IIIA. Only 55.6% received BCS, and 85.1% of these women also received radiation. Of those with nodal involvement (n = 521), 82.7% received chemotherapy. Japanese and Filipino women were significantly less likely than white women to undergo BCS (for Japanese: adjusted odds ratio, 0.62; 95% confidence interval, 0.48–0.80; for Filipinos: adjusted odds ratio, 0.47; 95% confidence interval, 0.33–0.66). Filipinos tended to be less likely than white women to receive radiation after BCS (adjusted odds ratio, 0.80; 95% confidence interval, 0.42–1.49). AAPI women were as likely as white women to receive adjuvant chemotherapy for nodal spread.

Conclusions: We found disparities in the management of early-stage breast cancer among AAPI women, particularly among Japanese and Filipinos. Further study is needed to determine the reasons for the observed disparities and to understand their effect on health outcomes.

Key Words: Breast cancer • Ethnicity • Disease management • Pacific Islanders • Asian Americans

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Among US women, breast cancer is the most common cancer and the second leading cause of cancer-related death.¹ Clinical trials and national recommendations support the use of adjuvant therapies for early breast cancer and the appropriateness of breast-conserving surgery (BCS) for primary resection.^2–5 Despite broad-based guidelines, however, many women still do not receive standard treatments.

This gap between the ideal and the reality of breast cancer care seems to be associated with both patient-and physician-related⁶ factors. In addition to patient age,⁷ comorbidity burden,⁷ geographical location,⁷ socioeconomic status,⁸ and health insurance,⁹ ethnicity has been associated with variation in management.¹⁰

Although Asian American and Pacific Islander (AAPI) women with breast cancer collectively do not seem to differ substantially from white women by stage or survival, differences among ethnic groups exist. Japanese women present with earlier stages, whereas Filipino and Hawaiian women are more likely to present with advanced disease.¹¹ Furthermore, Japanese and Chinese women seem to have overall better survival, and Hawaiian women, worse survival, as compared with white women.¹¹

However, little is known about variations in the management of early breast cancer across AAPI ethnic groups. We therefore linked tumor registry data to health care claims to examine the association between AAPI ethnicity and management of early breast cancer. We hypothesized that treatment would vary among AAPI groups relative to white women.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Data Sources
We linked data from the Hawaii Tumor Registry (HTR) to health care claims and census tract information to create a database that provides complementary information on patient and tumor characteristics and primary treatment. The HTR, a registry of the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) program, collects information on nearly all cases of cancer diagnosed in the state, according to SEER standards.¹² Information collected includes the standard SEER variables on patient characteristics (including age at diagnosis, ethnicity, county, and census tract), tumor characteristics at diagnosis (date of diagnosis, stage and extent of disease, tumor size, histological grade, and hormone receptor status), and primary course of treatment, encompassing the first 4 months after diagnosis (including surgery, radiation, and chemotherapy).^12,13

We obtained administrative claims from the largest health plan in Hawaii, which provides coverage for approximately 50% of residents (Hawaii Blue Cross Blue Shield). The plan includes fee-for-service (FFS), health maintenance organization (HMO), and Medicare FFS. Health care providers submit inpatient and outpatient claims by using the International Classification of Diseases, 9th Revision (ICD-9-CM) for coding diagnoses and procedures, the Health Care Common Procedure Coding System (HCPCS) level I Current Procedural Terminology (CPT) for physician procedures and services, and the HCPCS level II codes for services, procedures, and supplies.

Registry and claims data were linked by using a deterministic algorithm that matches records on name, sex, date of birth, date of diagnosis, and zip code. We further linked records to the 1990 US Census to ascertain the median household income of the census tract of residence. This study was approved by the appropriate institutional review boards.

Study Population
We identified women aged 21 years with a diagnosis of stage I, II, or IIIA infiltrating ductal breast cancer from 1995 to 2001 (International Classifica-tion of Diseases for Oncology, 2nd edition, codes 50.0–50.9; histology code 8500).¹⁴ Of 4139 cases diagnosed in the HTR during the study years, we excluded 264 women whose ethnicity was coded as other than white, Japanese, Chinese, Filipino, or Hawaiian, which are the 5 largest ethnic groups in the state. We further excluded women with more than one breast tumor diagnosed on the same date (n = 32), with two or more previous cancer diagnoses (n = 52), or with bilateral disease (n = 2); those whose reason for lack of treatment was coded as "contraindicated due to other conditions or autopsy only" (n = 2); those noted as "referred elsewhere," "inoperable," "autopsy only," or "death certificate only" (n = 24); and those missing information on tumor size (n = 14). There were 48 women who had breast cancer diagnosed on different dates; for these women, we analyzed treatment for only their first diagnosis. This left 3716 women eligible for linkage.

Of these 3716 women, we further excluded 377 who would have been eligible to receive Medicare if they had received a diagnosis during 1995 to 1996, because the available claims data did not include Medicare patients during these years. Although claims records for most women with more than one type of insurance plan coverage were not available, three women with both FFS and HMO were analyzed as having FFS. Of the remaining 3339 eligible women in the registry, we linked 2038 (61%) to claims data. From this linked database, we excluded eight patients because of missing data on health plan coverage. Therefore, our final analytic cohort included 2030 women.

Women in the linked database (n = 2030) did not differ from unlinked eligible women in the registry (n = 1686) by stage, nodal status, tumor grade, receptor status, cancer history, or rural residence. However, women in the linked database tended to be younger than unlinked women (mean age, 59.0 ± 13.2 years vs. 61.5 ± 13.4 years; P < .001), were more likely to reside in communities with higher incomes (52.4% above the median level vs. 46.8%; P = .001), and had smaller tumors (35.5% with tumor size 1 cm vs. 29.8%; P < .001). Records of Japanese (69.2%) and Chinese (59.5%) women were most likely to be linked to claims data, whereas records of Hawaiian (45.1%), Filipino (53.9%), and white (40.7%) women were less likely to be linked to claims data (P < .001).

Treatment Outcomes
We examined the use of standard treatment options as defined by guidelines from the National Cancer Institute and National Institutes of Health: (1) use of BCS for primary resection, (2) adjuvant radiotherapy for all women who receive BCS, and (3) adjuvant chemotherapy for women with node-positive disease.^4,5,15 We used registry data to identify women who received BCS, because registries record the most invasive operation performed within 4 months of diagnosis.

We used both registry and claims data to identify women who received radiotherapy and chemotherapy. Although radiotherapy tends to be captured in registry data, chemotherapy is routinely underestimated. Therefore, advantages of linked databases such as the one used here include the ability to identify chemotherapy through health care claims. For both radiotherapy and chemotherapy, we defined women as having received the treatment if it was present in registry records or if they had corresponding claims from 2 weeks before 4 months after the date of diagnosis recorded in the tumor registry.

Receipt of radiotherapy was identified from registry data as beam radiation (alone or in combination with other modalities) or radiation not otherwise specified and from claims data as radiation delivery or management (ICD-9-CM: 92.21–26, 92.29, V58.0, V66.1, and V67.1; CPT: 77401–4, 77406–9, 77411–14, 77416–20, 77425, 77427, 77430–1, 77520, 77522–3, and 77525). Receipt of chemotherapy was identified from registry and claims data as single- or multiple-agent chemotherapy (ICD-9-CM: V58.1, 66.2, 67.2, and 99.25; CPT: 96408, 96410, 96412, 96414, 96520, 96530, 96545, and 99555; HCPCS: J9000–9999 and Q0083–5). Registry cases for which receipt of radiation after BCS (n = 19) or chemotherapy for node-positive disease (n = 6) was unknown but noted to have been recommended were analyzed as having received therapy to allow for estimates of intended care.

Information on Covariates
Information obtained from the tumor registry included ethnicity, age and year of diagnosis, American Joint Committee on Cancer stage (tumor-node-metastasis), tumor size (<1, 1 to <2, 2–5, and >5 cm), lymph node involvement (positive or negative), county of residence (rural defined as non-Oahu), histological grade (well differentiated or moderately well differentiated [i.e., low grade] vs. poorly differentiated or undifferentiated tumors [i.e., high grade]), hormone-receptor status (positive or negative), history of cancer (yes or no), and median income of residence (quintiles rounded to nearest $5000).¹⁶ According to SEER standards, if a patient’s race/ ethnicity is recorded in the medical record as a combination of white and any other race, the individual’s race is coded as the first stated nonwhite race. If a person’s race/ethnicity is recorded as a combination of Hawaiian and any other race(s), the person’s race/ethnicity is coded as Hawaiian.¹³ To capture comorbid conditions, we used claims for services provided from 18 months prior to 6 months after cancer diagnosis and computed a modified Charlson-Deyo comorbidity index (dichotomized as 0 or 1), not including weights for cancer diagnoses.^17–19

Statistical Analysis
We compared patient and tumor characteristics according to ethnicity by using analysis of variance for continuous variables and ² tests for categorical variables. We used logistic regression to examine the association between ethnicity and treatment, considering two multivariable models. Model 1 adjusted for factors generally considered in risk stratification for management decisions, including age, tumor size and grade, and nodal and receptor status. Model 2 further adjusted for patient characteristics that may influence treatment decisions, including year at diagnosis, median income of residence, rural residence, cancer history, Charlson comorbidity index, and health plan type. Analyses were performed by using SAS software, version 8.2 (SAS Institute, Cary, NC). All statistical tests were two sided.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
The characteristics of the 2030 women in the cohort are listed according to ethnicity in Table 1. Most women were Japanese (46.1%), followed by white (20.8%) and Hawaiian (14.4%). Filipino and Hawaiian women generally received a diagnosis at younger ages and with more advanced disease than women of other ethnicities. Hawaiian women also had higher comorbidity levels than other women. Japanese and Chinese women resided in areas with the highest median incomes, whereas Hawaiian and Filipino women resided in areas with the lowest (P < .001).

Breast-Conserving Surgery
Table 3 presents ethnic variations in BCS use for women with registry-linked claims data, first adjusting for factors considered in clinical risk stratification (model 1) and then further adjusting for additional patient characteristics (model 2). Japanese and Filipino women were significantly less likely to undergo BCS compared with white women when adjusting for patient- and tumor-related factors routinely considered in risk stratification for management decisions (model 1). Further adjustment for additional patient characteristics did not substantially change the results (model 2). Japanese women had an adjusted OR (aOR) of 0.62 (95% CI, 0.48–0.80), and Filipino women had an aOR of 0.47 (95% CI, 0.33–0.66), as compared with white women. Hawaiian (aOR, 0.81; 95% CI, 0.58–1.12) and Chi-nese (aOR, 0.74; 95% CI, 0.49–1.13) women also seemed less likely to receive BCS than white women. However, these findings did not achieve statistical significance, partly as a result of the small sample size.

Chemotherapy
Overall, 82.7% of the 521 women with node-positive disease received chemotherapy, without signifi-cant differences according to ethnic group (Table 4). Interpretation of these results, however, is limited by small numbers of women with node-positive disease (for example, only 36 women with Chinese ethnicity had node-positive disease). Although this was not statistically significant, partly as a result of the sample size, Japanese women seemed more likely than white women to receive chemotherapy. In age- and multivariable-adjusted analyses, chemotherapy use among Japanese women increased with further adjustment for tumor and patient characteristics (aOR, 2.10; 95% CI, 0.98–4.50, after adjustment for all measured potential confounders), thus suggesting that Japanese women may receive more chemotherapy despite overall less aggressive tumor characteristics, as compared with white women.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Japanese and Filipino women were markedly less likely to receive BCS for primary resection as compared with white women. These disparities remained significant after adjustment for potential confounders, thus suggesting that these associations are not entirely explained by differences in patient- or tumor-related factors. Among women treated with BCS, radiation use seemed lower among Filipino women. By contrast, chemotherapy use for node-positive disease did not differ substantially across ethnic groups.

Although both mastectomy and BCS remain options for primary resection and the decision to use BCS depends on multiple factors, including tumor size and patient preferences, systematic differences in BCS use between ethnic groups may indicate that women of certain ethnic groups may not be completely exercising their choice of surgery.

Other potential explanations for our observed disparities include unmeasured differences between ethnic groups that may contribute to treatment variation. For example, we do not have information on previous radiotherapy among women with a non–breast cancer history. However, although prior radiation may have affected treatment decisions for breast cancer, cancer history did not vary signifi-cantly among ethnic groups. In addition, ethnic variation in breast size may contribute to differences in BCS use between ethnic groups, because small breast size may lead to cosmetically unacceptable results after BCS. Cultural differences may also explain some of the observed differences in treatment, and this warrants further study.

Other studies have documented widespread variations in standard treatment for early breast cancer. Although BCS use may vary with age, geographical location, and socioeconomic status, most studies note that less than half of women with early breast cancer receive BCS.⁷ Furthermore, up to 20% to 30% of women treated with BCS do not receive radiation, and lower use is associated with advanced age.^7,21 Chemotherapy rates for node-positive disease also vary with age, comorbid illness, and geographical region, with as many as 60% of eligible women not receiving chemotherapy.^21,22

Few studies have addressed disparities among AAPIs, and the available results are conflicting. Our findings differ from a study of BCS use in Hawaii from 1995 to 1998 that also linked registry and claims data.²³ In that study, BCS use was not associated with ethnicity. However, only 32.8% (n = 722) of tumor registry cases were successfully linked to claims data, and those linked were a young population. In contrast, our study linked 61% of eligible cases and included a population more diverse in age.

Other reports including AAPI women, however, have documented ethnicity-related differences in breast cancer management. Studies including AAPI women in California during the early 1990s identified lower rates of BCS use among AAPIs as compared with white women,²⁴ particularly among Chinese and Filipino women.²⁵ Less use of radiation after BCS has also been reported among Chinese women.²⁵ In a recent nationwide sample, AAPI women were less likely to undergo BCS than white women.²⁶ However, the effects of ethnicity-related differences in comorbidity burden, health coverage, and socioeconomic status on treatment estimates have not been examined previously.

Additional strengths of our study include the large number of AAPI and younger women (<65 years) and the use of a combined database with linked registry and claims data. Few other studies have used similar state-level linked databases. Advantages of this approach, similar to those described for the national SEER-Medicare database,²⁷ include the ability to ascertain information on comorbid illness. In addition, such linked databases allow for more complete treatment estimates not available with registry data alone, particularly chemotherapy use. Furthermore, we used claims data that include en-rollees in FFS and HMO plans, as well as Medicare, thus allowing for analysis of women <65 years.

Limitations to our study should be considered. First, although we were able to adjust for multiple possible confounders of the association between ethnicity and management, residual confounding remains a potential explanation for our findings. We were not able to contact patients to determine reasons for treatment omission. Patient preferences may differ by ethnicity, and this could explain some of the observed differences in care. However, racial differences in treatment preferences are, in general, not considered to account fully for disparities in health care.²⁸ Physician characteristics and preferences may account for some of the observed disparities and require further investigation.

Second, interpretation of our findings is limited by small numbers of women in subgroup analyses, particularly among Chinese and Filipino women with node-positive disease. Third, a linked database such as ours is not expected to capture all incident cases in the population. Although SEER registries such as HTR have an overall 97% completeness rate in case ascertainment,¹² our analysis was restricted to the 61% of cases linked to claims records. This may have produced selection bias because linkage varied by patient ethnicity. However, we found similar estimates of BCS use when comparing all women in the registry with those with linked claims. This suggests that potential selection bias may not have substantially altered our results for BCS.

Finally, we studied women with breast cancer in a single state enrolled in a single health plan, so our findings may not generalize to other AAPI groups. However, this plan provides coverage for approximately half of the state’s population, and our cohort was diverse in age and ethnicity.

In conclusion, we found significant disparities in the management of early-stage breast cancer among AAPI ethnic groups. These disparities were not entirely explained by differences in patient and tumor characteristics. Our findings highlight the need to consider AAPI groups separately in studies of breast cancer, because they have differences not only in presentation and survival, but also in management. Further studies are needed to determine the reasons for our findings and whether these disparities in care result in variation in health outcomes.

	ACKNOWLEDGMENTS

 
Supported by grant R24 HS11627 from the Agency for Healthcare Research and Quality.

Received for publication August 26, 2005. Accepted for publication December 22, 2005.

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 

1. American Cancer Society. Cancer Facts and Figures 2005. Atlanta: American Cancer Society, 2005.
2. Early Breast Cancer Trialists’ Collaborative Group Systemic treatment of early breast cancer by hormonal, cytotoxic, or immune therapy. 133 randomised trials involving 31,000 recurrences and 24,000 deaths among 75,000 women. Lancet 1992; 339:1–15.[Medline]
3. Early Breast Cancer Trialists’ Collaborative Group Effects of radiotherapy and surgery in early breast cancer. An overview of the randomized trials. N Engl J Med 1995; 333:1444–55.[Abstract/Free Full Text]
4. Treatment of Early-Stage Breast Cancer. NIH Consensus Statement 1990 Jun 18–21; 8:1–19. Available: http://consensus.nih.gov/1990/1990EarlyBreastCancer081html.htm [accessed April 5, 2006].
5. Adjuvant therapy for breast cancer. NIH Consensus Statement 2000 November 1–3; 17(4):1–35. Available: http://consensus.nih.gov/2000/2000AdjuvantTherapyBreastCancer114html.htm [accessed April 5, 2006].
6. Mandelblatt JS, Berg CD, Meropol NJ, et al. Measuring and predicting surgeons’ practice styles for breast cancer treatment in older women. Med Care 2001; 39:228–42.[CrossRef][Medline]
7. Morrow M, White J, Moughan J, et al. Factors predicting the use of breast-conserving therapy in stage I and II breast carcinoma. J Clin Oncol 2001; 19:2254–62.[Abstract/Free Full Text]
8. Albain KS, Green SR, Lichter AS, et al. Influence of patient characteristics, socioeconomic factors, geography, and systemic risk on the use of breast-sparing treatment in women enrolled in adjuvant breast cancer studies: an analysis of two intergroup trials. J Clin Oncol 1996; 14:3009–17.[Abstract]
9. Potosky AL, Merrill RM, Riley GF, et al. Breast cancer survival and treatment in health maintenance organization and fee-for-service settings. J Natl Cancer Inst 1997; 89:1683–91.[Abstract/Free Full Text]
10. Joslyn SA. Racial differences in treatment and survival from early-stage breast carcinoma. Cancer 2002; 95:1759–66.[CrossRef][Medline]
11. Li CI, Malone KE, Daling JR. Differences in breast cancer stage, treatment, and survival by race and ethnicity. Arch Intern Med 2003; 163:49–56.[Abstract/Free Full Text]
12. Zippin C, Lum D, Hankey BF. Completeness of hospital cancer case reporting from the SEER Program of the National Cancer Institute. Cancer 1995; 76:2343–50.[CrossRef][Medline]
13. Fritz A, Ries L. The SEER Program Code Manual. 3rd ed. US Department of Health and Human Services: National Cancer Institute, 1998.
14. Percey C, Van Holten V, Muir C, (eds). International Classi-fication of Diseases for Oncology. 2nd ed. Geneva, Switzerland: World Health Organization, 1990.
15. National Cancer Institute. Breast Cancer Physician Data Query. Available: http://www.cancer.gov/cancertopics/pdq/treatment/breast/HealthProfessional [accessed April 5, 2006].
16. Krieger N. Overcoming the absence of socioeconomic data in medical records: validation and application of a census-based methodology. Am J Public Health 1992; 82:703–10.[Abstract/Free Full Text]
17. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987; 40:373–83.[CrossRef][Medline]
18. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical com-orbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol 1992; 45:613–9.[CrossRef][Medline]
19. Klabunde CN, Warren JL, Legler JM. Assessing comorbidity using claims data: an overview. Med Care 2002; 40(8 Sup-pl):IV26–35.
20. Malin JL, Kahn KL, Adams J, Kwan L, Laouri M, Ganz PA. Validity of cancer registry data for measuring the quality of breast cancer care. J Natl Cancer Inst 2002; 94:835–44.[Abstract/Free Full Text]
21. Malin JL, Schuster MA, Kahn KA, Brook RH. Quality of breast cancer care: what do we know? J Clin Oncol 2002; 20:4381–93.[Abstract/Free Full Text]
22. Du XL, Key CR, Osborne C, Mahnken JD, Goodwin JS. Discrepancy between consensus recommendations and actual community use of adjuvant chemotherapy in women with breast cancer. Ann Intern Med 2003; 138:90–7.[Abstract/Free Full Text]
23. Maskarinec G, Dhakal S, Yamashiro G, Issell BF. The use of breast conserving surgery: linking insurance claims with tumor registry data. BMC Cancer 2002; 2:3.[CrossRef][Medline]
24. Morris CR, Cohen R, Schlag R, Wright WE. Increasing trends in the use of breast-conserving surgery in California. Am J Public Health 2000; 90:281–4.[Abstract/Free Full Text]
25. Prehn AW, Topol B, Stewart S, Glaser SL, O’Connor L, West DW. Differences in treatment patterns for localized breast carcinoma among Asian/Pacific Islander women. Cancer 2002; 95:2268–75.[CrossRef][Medline]
26. Goel MS, Burns RB, Phillips RS, Davis RB, Ngo-Metzger Q, McCarthy EP. Trends in breast conserving surgery among Asian Americans and Pacific Islanders, 1992-2000. J Gen Intern Med 2005; 20:604–11.[Medline]
27. Potosky AL, Riley GF, Lubitz JD, Mentnech RM, Kessler LG. Potential for cancer related health services research using a linked Medicare-tumor registry database. Med Care 1993; 31:732–48.[Medline]
28. Institute of Medicine. Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. Washington, DC: National Academy Press, 2002.

This article has been cited by other articles:

				L. K. Jacobs, K. A. Kelley, G. D. Rosson, M. E. Detrani, and D. C. Chang Disparities in Urban and Rural Mastectomy Populations: The Effects of Patient- and County-Level Factors on Likelihood of Receipt of Mastectomy Ann. Surg. Oncol., October 1, 2008; 15(10): 2644 - 2652. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Gelber, R. P. Articles by Seto, T. B. Search for Related Content PubMed PubMed Citation Articles by Gelber, R. P. Articles by Seto, T. B.