• © 2004 by American Society of Clinical Oncology

An Ounce of Prevention or a Pound of Cure? Investing to Improve Breast Cancer Outcomes for African American Women

Recent improvements in breast cancer care have failed to improve outcomes equally for all cancer patients in the United States. Notably, African American women have a lower incidence of breast cancer than white women but are more likely to present with advanced disease and have greater breast cancer mortality than age- and stage-matched white women.¹ These findings parallel disparities noted across the spectrum of medicine^2-4 and reflect persistent unequal treatment. Although effective interventions to improve adherence to breast cancer screening recommendations have been developed and evaluated,^5-7 less effort has been devoted to enhancing the care of African American women following the diagnosis of breast cancer. Given limited health care resources and available data showing that many African American women are not optimally treated,^8,9 it is important to ask whether it is more efficient for society to fund interventions that improve outcomes in African American women with breast cancer or interventions aimed at further increasing the rate of breast cancer screening. In this context, we are intrigued by the findings reported by Mandelblatt et al¹⁰ in this issue of the Journal of Clinical Oncology suggesting that the most cost-effective method to improve breast cancer outcomes for African Americans is to use interventions which target women with a cancer diagnosis, rather than women who may potentially develop a cancer. Using simulation modeling, the authors compare the costs and benefits of increased biennial mammography screening to those associated with improved compliance with National Comprehensive Cancer Network and the St Gallen consensus recommendations for treatment. They found that at investments of up to $6,000 per breast cancer patient, the cost of an intervention designed to enhance treatment is less than $75,000/yr of life saved. In contrast, the cost-effectiveness of using reminder letters or outreach workers to improve rates of mammography among African American women is poor (cost exceeding $120,000/yr of life saved), unless these interventions are specifically targeted toward pockets of very underscreened or high-risk women.

Patients with advanced stage breast cancer have the most to gain with treatment, and our treatments have improved significantly; as such, the relative lack of improvement in outcomes for African American women is disturbing. Mandelblatt et al's simulation analysis suggests we redirect research and health care funding toward improving our understanding of why African American women with breast cancer receive less standard of care treatment than white women and then toward developing interventions that enhance care. Although it seems logical that modest investments could significantly improve outcomes, we suspect that the devil truly is in the details for eliminating disparities in cancer care within vulnerable populations. The question of what investment per patient is sufficient to eliminate disparities begs empirical testing.

A theoretical model never adequately describes clinical practice, which continues to evolve and is dynamically shaped by diverse medical, logistic, and social forces. Mandelblatt et al assume their interventions will be effective and efficient and compare interventions between dissimilar groups. Substantial investments can still have only modest impacts on adherence.^11-13 As we develop and test interventions designed to improve outcomes for African American women with breast cancer, evaluating the costs of these interventions, as well as their effectiveness, is essential to inform policy decisions.

To improve outcomes for African American women diagnosed with cancer, medical oncologists must face the twin challenges of understanding why African American patients are less likely to receive standard treatment and then design and implement interventions that improve treatment adherence in a culturally competent and durable fashion. The reasons why African American women receive less standard care are complex and not entirely clear. Race and geography may not be the optimal descriptors to define the group of Americans with poorer outcomes, but for much of the available data, socioeconomic and cultural variables are not available. When the Surveillance, Epidemiology, and End Results data are linked with Medicaid data (in the Detroit experience), socioeconomic features can be dissected from race, and only poverty emerges as the marker of a group with a poorer outcome.¹⁴ Access to care and problems with provider acceptance of Medicare and Medicaid reimbursement rates are two important barriers impoverished patients face. In addition, patients may choose (and providers may recommend) nonstandard care due to comorbidities, poverty (time away from work), lack of trust or continuity of providers, real or perceived provider or institutional bias, and cultural beliefs and differences.^2,15,16 Addressing these issues is critical if we are to eliminate disparities in access to high quality breast cancer treatment. Our current health care delivery and reimbursement systems may not be structured to support innovations that are designed to increase use of recommended treatments.

Work by Dr Harold Freeman and collaborators at Harlem Hospital^17,18 suggests that patient navigator programs are one way to improve breast cancer care and outcomes for at-risk populations. Navigators function as culturally competent links from the community to the health system. They have been shown to increase access to quality health care at every stage in the cancer diagnosis and treatment process, as well as provide outreach to unscreened subgroups. However, broader funding of interventions such as the navigator program will only occur if we can make a business case to payers that such interventions are cost-effective, increasing the value of oncology services financed.

As more information about the costs and effectiveness of interventions such as the navigator program become available, further analysis will be necessary to determine the most cost-effective programs for improving care of African American women with breast cancer. Although the up-front costs of improving treatment patterns for local-regional disease may be partially offset by reductions in the cost of treating metastatic cancer, we cannot assume that such investments will be cost saving in the long run. The payoff is improved quality of life and survival.