Health Care Disparities in Breast Cancer: The Economics of Access to Screening, Diagnosis, and Treatment

Abstract

The widespread acceptance of screening mammography for early detection, along with advances in treatment, have combined to decrease the overall mortality rate from breast cancer. However, significant disparities in health outcomes persist. Socioeconomic factors, including the ability to obtain private insurance, income, education, disparities in the quality of healthcare delivery, and race, as well as the ability to access and complete the most advanced breast cancer treatments, form part of a complex constellation of factors that contribute to disparity in breast cancer mortality. Here, we review some of the factors influencing this disparity and discuss some of the methods that have been suggested for closing the gap in breast cancer outcomes, using our perspective as breast imaging physicians serving both a safety-net hospital and tertiary healthcare system.

breast cancer, health care disparity, mammography screening, breast cancer mortality

Key Messages

• Poverty, race, uninsured status, and a lower level of education are significantly associated with higher breast cancer mortality rates.

• Breast cancer patients treated in safety-net hospitals face significant challenges due to presenting at a later stage and not consistently receiving adjuvant treatments that have been shown to increase survival.

• Providing insurance coverage for uninsured women with breast cancer, even if it is only Medicaid, significantly reduces the incidence of advanced-stage breast cancer diagnoses. However, even with access to Medicaid, the incidence of advanced disease is still much higher than with private insurance due to other issues related to the lower socioeconomic status of these patients.

Introduction

Breast cancer is the most frequently diagnosed non-skin cancer and is the second-leading cause of cancer death in women (1). Advances in breast cancer screening and treatment over the last few decades have reduced the breast cancer mortality rate by 40% (1). However, not all women have benefited from these advances, with African American (AA) women being much more likely to die of breast cancer than non-Hispanic White women (NHW). Causes of these breast cancer health disparities fall into three main categories: those related to poverty (low socioeconomic status (SES)), culture, and social injustice. These differences affect the entire breast cancer care continuum from prevention and detection to treatment and survival (2) (Figure 1).

Figure 1.

Causes of healthcare disparities (adapted from (2)).

In this work, we aim to review the economic, cultural, and social injustice factors that may contribute to disparities in breast cancer detection and mortality. Our starting point is information from our county safety-net hospital compared to published data, as well as our experiences as breast imaging radiologists serving in both safety-net and tertiary care hospital settings.

Breast Cancer Economics of Access (Insured vs. Underinsured)

Americans living in poverty have higher incidence rates, inferior five-year survival rates, higher mortality rates for all types of cancers, and are more likely to be diagnosed with higher-stage disease (3). Moreover, among patients with the same disease stage, those living in poverty, who are uninsured or underinsured, have worse cancer outcomes than patients of higher SES with health insurance (4). Because AA and Hispanic women are more likely to live in poverty, they are more likely to be uninsured throughout their lifetime than NHW women and these disparities in health insurance significantly influence access to health care (5,6). Within women with breast cancer, AA (odds ratio (OR), 1.46; 95% CI: 1.40–1.53), American Indian or Alaskan Native (OR, 1.31; 95% CI: 1.07–1.61), and Hispanic (OR, 1.35; 95% CI: 1.30–1.42) women were more likely to receive a diagnosis of locally advanced disease (stage III) compared with NHW women (7). Most uninsured or underinsured women seek healthcare through safety-net hospitals serving predominantly minority patient populations (8,9). These sites are typically challenged by limited resources, with uninsured and Medicaid patients often receiving poorer-quality care and having higher mortality rates (10–14). Parkland Memorial Hospital, best known as the hospital where the assassinated president John F. Kennedy was taken, is the safety-net hospital for residents of Dallas County. Predominantly serving a Hispanic and AA population, breast cancer patients diagnosed in our county safety-net hospital breast imaging facility (Parkland Comprehensive Breast Center) are younger (average age of 54.8 years vs. 68.9 years nationally) and present with higher-stage disease requiring neoadjuvant chemotherapy. At Parkland, the rates of breast cancers that are stage III–IV at diagnosis range from 20.5%–55.6%, and these rates vary depending upon Zip code (Table 1). Nearly all patients manage at least one other chronic condition, such as diabetes, hypertension, or heart disease, and more than half manage more than one of these conditions. The highest mortality rates from breast cancer are localized to three areas in southern Dallas County, with AA women having the highest mortality rate from breast cancer compared to White women (31.8 vs. 18.8/100 000) (Figure 2).

Table 1.

Parkland Health and Hospital System Cancer Registry Data Analysis: Late Stage (Stage III or IV) Breast Cancer Diagnosis in Areas that Have the Highest Rates of Poverty by Zip Code

Zip Codes	Total Cases	Late Diagnosis Count	Percent
75217	54	11	20.5
75060	18	10	55.6
75243	30	10	33.3
75061	28	8	28.6
75041	11	6	54.6
75228	33	6	18.2
75227	40	6	15.0
75216	50	6	12.0
75159	10	5	50.0

Figure 2.

Map of breast cancer and mortality rates per 100 000 in Dallas County. The highest mortality rate from breast cancer (22–27/100 000) in Dallas County is localized to the three areas where women have high uninsured rates and less access to healthcare (shaded in red).

Presenting with advanced disease is not the only reason for worse outcomes in women whose breast cancer is treated in safety-net hospitals. In this setting, patients do not consistently receive adjuvant treatments that have been shown to increase survival (15). In a study investigating the use of adjuvant breast cancer therapies in early-stage breast cancer patients in metropolitan New York, “system failures”—defined as treatment recommended, not refused, but did not ensue—were more common among women seen at municipal hospitals than non-municipal hospitals (82% vs. 36%; P < 0.0001) (15). These patients were also more likely to be uninsured, receiving Medicaid, and members of minority groups. Socioeconomic factors and time commitment are among the key challenges faced by safety-net patient population to complete recommended therapies. If employed, these women often do not have paid sick leave, meaning time away from work results in a significant economic loss for the patient and her family. Some women may not have personal transportation and must rely on public transportation or an unreliable means to deliver them to their healthcare site. If they have children, obtaining necessary childcare may be a major obstacle. These competing social and economic demands make it much more difficult for women, who are already more likely to present with advanced disease, to adhere to months of neoadjuvant or adjuvant chemo-radiation.

Health and educational outcomes are closely linked to each other. Education can provide opportunities for people to earn higher incomes, which, in turn, give them social and psychological benefits, encourage healthier lifestyles, and provide them with access to good health care. A 2013 National Research Council report cites educational and income gaps as leading causes for health disparities within the United States (16). Poorer breast cancer prognosis has been suggested to correlate with lower levels of education. In a recent study comparing the breast cancer mortality in NHW and Hispanic women, the Hispanic group had a significantly lower education level (high school graduate or lower, 53.8% vs. 23.9%) and a lower five-year survival rate (82.2% vs. 94.3%; hazard ratio = 2.78) (17). However, this mortality difference was not statistically significant when adjusted for education level. The relationship between education level and the mortality rate for breast cancer is likely complex. In countries where access to breast cancer screening is opportunistic, education is an important factor that indicates a higher rate of job-related insurance prevalence. Lower education levels likely contribute to unhealthy lifestyle choices, such as physical inactivity, obesity, a high-fat diet, and use of hormone replacement therapy, the effects of which are difficult to measure.

Racial Disparity in Breast Cancer

Although the incidence of breast cancer in NHW has remained steady, it has continued to increase among AA and Hispanic women. Currently, the incidence of breast cancer is relatively similar for AA and NHW women (126.7 vs. 130.8/100 000), while the mortality rate in AA women is 40% higher than in NHW (28.2 vs. 20.3 deaths/100 000) (1). Biological factors and prevalent tumor molecular subtype are other disadvantages faced by AA patients with breast cancer. AA women have a higher incidence of breast cancer before the age of 50 and the aggressive triple-negative (TN) (estrogen, progesterone and HER2− expression negative) subtype of breast cancer (1, 18–20). In Dallas County, AA women have the highest mortality rate from breast cancer (31.8 vs. 18.8/100 000) compared to White women (Figure 3).

Figure 3.

Female breast cancer mortality rates by race, Dallas County, 2016.

Like AA women, Hispanic women with breast cancer are younger, have larger tumors of higher histologic grades, and unfavorable breast cancer subtypes, namely TN and HER2-positive cancers (17,21,22). A population-based cross-sectional study of women in a comprehensive, equal-access healthcare system showed that, despite their equivalent access to screening mammography, Hispanic women were more likely to present with stage IV disease, poorly differentiated and larger (>5 cm) tumors, and with unfavorable molecular subtypes compared to NHW. Hence, biologic or genetic factors may be contributing to these disparities (23).

Opportunities to Reduce Disparities

Improve Access to Breast Cancer Care

The 2010 Patient Protection and Affordable Care Act (ACA) expanded and increased Medicaid benefits, with the potential to reach up to 47 million Americans without health insurance. After a Supreme Court ruling, only 32 states fully adopted the expanded Medicare coverage, while in 14 non-expansion states, Medicaid coverage remained unchanged at pre-ACA levels. In a cohort study of 1 796 902 women with primary breast cancer diagnosed from 2007–2016, the states with expanded Medicaid had a 2.5% decrease in advanced-stage breast cancer across all races (P < 0.001), compared to a 0.7% decrease in non-expansion states (P = 0.14) (24). In AA women, late-stage disease incidence decreased by 3% in expansion states and was essentially unchanged in non-expansion states. Additional analysis suggested a reduction in stage III diagnoses as the predominant factor for improvement. In expansion states, among all races and ethnicities, there was a 9.1% decrease in uninsured patients (P < 0.001), compared with a 0.9% decrease in the non-expansion states (P = 0.12). Providing insurance coverage for uninsured women with breast cancer, even if it is only Medicaid, significantly reduces the incidence of advanced-stage diagnosis of breast cancer. However, even with access to Medicaid, the incidence of advanced disease is still much higher than with private insurance. This could mean that diagnosis and treatment with Medicaid is substandard or, as is well known, factors other than insurance also play a significant role in access to care.

Early Detection: Breast Cancer Screening in the Underserved

Serving a predominantly AA and Hispanic patient population in our safety-net hostital gives us the opportunity to note the relatively higher rate of patients diagnosed with breast cancer at a young age and lower rates of screening mammography-detected cancers (25). Indeed, earlier access to high-quality screening is even more important for women at risk of having more aggressive forms of breast cancer at an earlier age, such as AA and Hispanic women, than it may be for NHW women of average risk.

In a “Coming Together for the Cure” Survey in South Dallas (a lower SES area) in 2017, 60% of the surveyed women answered affirmatively to the question “Cost affects my ability to get a cancer test.” Forty-two percent did not know if a co-pay was required to get a cancer screening test. However, under the ACA, many insurance plans have no co-pay for an annual physical, which can include a screening mammogram. The findings of this survey underscore the importance of outreach to areas with lower access to care as a result of SES and education, and education for primary care providers to know the screening guidelines and to appropriately order screening mammograms. Automatic entries in the electronic medical record with the date the patient last received an appropriate screening test can also help increase compliance with screening guidelines.

The quality of screening in underserved populations is extremely important. A Health and Human Services study found that women presenting with symptomatic breast cancer after a recent screening mammogram had lower levels of education (P < 0.001) and income (P < 0.001) and were less likely to have health insurance (P < 0.001) (26). They were more likely to have tumors with unfavorable subtypes (P < 0.001) and to present at a higher stage (P < 0.001). Women screened at facilities designated as imaging centers of excellence (P < 0.001) or university-based centers (P < 0.001) were less likely to have symptomatic breast cancer presenting after a recent screening mammogram, suggesting that poorer women may receive screening of lower quality. It is possible that this result is confounded by underserved women preferring neighborhood clinics instead of centers of excellence. In our experience, even those underserved women who have access to essentially free high-quality mammography in a safety-net screening center experience the well-known barriers to care access when recalled for further evaluation of an abnormality. This in turn leads to delays in breast cancer diagnosis and treatment, resulting in higher mortality.

Reducing Time to Breast Cancer Treatment

Differences in breast cancer stage distributions support the theory that mortality disparities are mediated, at least in part, by diagnostic and treatment delays associated with healthcare access barriers. In a retrospective review of tumor registry data at Parkland Hospital from 2018, the time from the date of diagnosis (pathology report) to first treatment of 207 patients with invasive breast cancer was reviewed (Matthew Porembka, M.D., email communication, 6 January 2020). The patients’ charts were examined manually to determine contributing factors associated with delays in treatment. The mean number of days for all patients regardless of stage was 75 days, highest in stage I (83 days) and lowest in stage IV (71 days). Only 12% of patients began treatment within 30 days of diagnosis, 57% at 60–90 days, and 31% at 91–360 days (Figure 4). The significant barriers preventing patients from receiving timely treatment included the need to apply for, and be accepted to, county-covered insurance, returning to the hospital multiple times for physician consultations and additional testing, care coordination delays, and patient preference. Expediting multidisciplinary consultation and coordinated care with same-day service by providers located in one facility is a known successful method for reducing time to breast cancer treatment.

Figure 4.

Distribution of time to first treatment rates in 207 patients diagnosed with invasive breast cancer in Dallas County safety-net hospital. Time to treatment in days, collected in 2018. Only 12% of patients had their first treatment within 30 days of diagnosis.

Conclusion

In summary, lack of any or adequate healthcare insurance is a disparity for women that impacts breast cancer outcomes resulting in delays in detection, diagnosis, and treatment. Universal healthcare potentially is a means to increase access for those who are now underinsured and uninsured. Unfortunately, disparities in breast cancer outcomes often result from not only a lack of access to care, but also longstanding inequalities in socioeconomic status, transportation, housing, and education. These differences in patient outcomes challenge our values as physicians and our sense of fairness and equality for all patients. Awareness of these disparities among both patients and physicians alike may lead to improvements in access to affordable care and adequate breast cancer screening, diagnosis, and treatment for all women.

Funding

None declared.

Conflict of Interest Statement

None declared.