This Article  • PDF  • Send to a friend  • Save in My Folder  • Save to citation manager  • Permissions  Citing Articles  • Citing articles on HighWire  • Citing articles on Web of Science (4)  • Contact me when this article is cited  Related Content  • Similar articles in JAMA  Topic Collections  • Oncology  • Breast Cancer  • Women's Health  • Women's Health, Other  • Alert me on articles by topic  Social Bookmarking   What's this?

To the Editor: The incidence of breast cancer among Native Americans in the Southwest has nearly doubled in the last 30 years,¹ and in the 1990s breast cancer represented the most common cancer overall among all members of the Navajo Nation.² Native Americans with breast cancer have significantly lower 5-year survival rates when compared with the general population and other minority groups,^3-4 even when adjusting for stage of cancer presentation. It is unclear whether these worse outcomes for Native American women are related to the process of care delivery or to underlying tumor biology. We studied patterns of breast cancer care at Indian Health Service (IHS) hospitals on the Navajo Nation and evaluated their association with adverse outcome.

Methods
We conducted a retrospective, chart review study of all patients whose written records were retrievable from IHS facilities. Administrative records at each facility identified all patients with IHS medical record numbers from the Navajo Nation area who were diagnosed with breast cancer from 1991-1999. Our sample represented 53% of IHS patients identified with breast cancer. We included patients if they were Navajo women diagnosed or treated for breast cancer at IHS hospitals during the study period. Patients were excluded from this study if they were not Navajo or had completed no part of the cancer diagnosis or care at one of the index hospitals.

We measured the rate of receipt of "standard cancer care," defined as evidence in the medical record of resection of the primary tumor, as well as radiotherapy or chemotherapy, or both, if appropriate for the stage of disease. We also measured a composite outcome of death or recurrence of either local or regional disease. Kaplan-Meier estimates were used to compare the aggregate end point of recurrence or death using the log-rank and Wilcoxon test. We used a Cox model to assess the association between receipt of standard care and rate of recurrence or death. Permission for this study was obtained from the Navajo Nation research review board and human subjects committee.

Results
Of 134 patients with breast cancer (mean [SD] age, 53 [13.5] years), 45 (33.6%) did not receive standard cancer care by our definition. Among these patients, the components of care that were not received included no therapy beyond diagnosis (31.5%), no receipt of indicated chemotherapy (34.2%), no receipt of indicated radiotherapy (13.2%), and prolonged time lags in care (2.6%). More than 1 of these missing components was found in 18.4% of patients. Surgical therapy was indicated in all patients, but 11.2% did not receive it. Breast conservation therapy (eg, lumpectomy, axillary node dissection with radiotherapy) was intended in 19.8% of patients but completed in only 56% of those. Among patients for whom adjuvant therapy was intended, 70.7% did not receive it.

The rate of 5-year recurrence or death (Figure 1) was 9.2% among those whose care met standard care patterns, compared with 19.7% for those whose care did not meet standard care patterns (P = .03). The hazard ratio for recurrence or death was 3.5 times higher (95% confidence interval, 1.3-9.4) for those who did not receive standard care when controlling for age, cancer stage, and receptor status.

View larger version (17K): [in this window] [in a new window] Figure. Recurrence or Death After Breast Cancer Diagnosis, by Receipt of Standard Care Follow-up on 125 patients with complete follow-up to 5 years.

Comment
Nearly one third of Navajo patients with breast cancer in this sample did not receive standard cancer care. Chart notes indicated that in some patients this was related to structural, geographic, and cultural barriers to effective cancer care in this environment.

We acknowledge several limitations to our study. Chart abstraction in the Navajo IHS is problematic because records are not computerized and are difficult to obtain once the patient has died or once many volumes have been generated. As a result, the final sample of charts reviewed represents a convenience sample that may be biased to include patients who were alive after diagnosis. Thus, the resulting bias would tend to overstate the number of patients who received appropriate cancer care. One result of this bias, however, is that the overall estimate of important clinical outcomes such as death and recurrence among Navajo persons is limited to the population surveyed and cannot be said to represent rates among all Navajo patients. Finally, because this is an observational study and not a controlled trial, there may be unmeasured confounders that could explain the different outcomes between groups. Given the identified disparities in both process of care and outcomes, further exploration of barriers to care may represent opportunities for improvement in the care of similar populations.

David R. Flum, MD, MPH
Department of Surgery
University of Washington
Seattle

Susan Stuart, MD
Gallup Indian Medical Center
Indian Health Service-Navajo Nation
Gallup, NM

Michael Wilcox, MD
Tuba City Medical Center
Indian Health Service-Navajo Nation
Tuba City, Ariz

1. US Department of Health and Human Services. Regional Differences in Indian Health 1997. US Dept of Health and Human Services:1997:57. Available at: http://www.ihs.gov/PublicInfo/Publications/trends97/region97.asp. Accessibility verified September 2, 2003. 2. Nutting PA, Freeman WL, Risser DR, et al. Cancer incidence among American Indians and Alaska Natives, 1980 through 1987. Am J Public Health. 1993;83:1589-1598. FREE FULL TEXT 3. Samet JM, Key CR, Hunt WC, Goodwin JS. Survival of American Indian and Hispanic cancer patients in New Mexico and Arizona, 1969-82. J Natl Cancer Inst. 1987;79:457-463. WEB OF SCIENCE | PUBMED 4. Frost F, Tollestrup K, Hunt WC, Gilliland F, Key CR, Urbina CE. Breast cancer survival among New Mexico Hispanic, American Indian, and non-Hispanic white women (1973-1992). Cancer Epidemiol Biomarkers Prev. 1996;5:861-866. ABSTRACT

Letters Section Editor: Stephen J. Lurie, MD, PhD, Senior Editor.

JAMA. 2003;290:1996-1997.

CiteULike    Connotea    Del.icio.us    Digg    Reddit    Technorati    Twitter     What's this?

THIS ARTICLE HAS BEEN CITED BY OTHER ARTICLES

Lost Opportunities: Physicians' Reasons and Disparities in Breast Cancer Treatment
Bickell et al.
JCO 2007;25:2516-2521.
ABSTRACT | FULL TEXT