Abstract
Background
In the U.S., poorer breast cancer survival in Black women relative to white women has persisted for over 30 years, with recent data showing a 38% higher mortality in Black women with breast cancer. Stage at diagnosis is the most powerful predictor of survival. However, despite rates of mammographic screening having become approximately equal in the two groups, Black women are more likely to be diagnosed at later stages. We examined whether perceived experiences of discrimination due to race in receipt of health care is associated with a late stage at diagnosis of breast cancer.
Methods
Nested within the prospective Black Women’s Health Study (BWHS), this case-only study included 1,617 self-identified U.S. Black women diagnosed with a first invasive breast cancer in 2003 through 2022. Eligible cases had completed a BWHS questionnaire in 2003 in which participants were asked whether they received differential health care due to their race or insurance status. The primary outcome in the present research was stage at breast cancer diagnosis, obtained from medical records and cancer registry data. Odds ratios (OR) for the association of perceived racial discrimination in health care with later stage at diagnosis were estimated in logistic regression analyses, with each of stages II, III, and IV compared with stage I diagnosis.
Results
In multivariable analyses controlled for age, body mass index, socioeconomic factors, and mammographic screening, perceived racial discrimination in healthcare was associated with an increased odds of breast cancer diagnosis at stage IV versus stage I (OR = 2.10, 95% CI 1.15–3.83). The association was present even among women who reported having a mammogram in the two years before diagnosis. No associations were observed for stage II or III versus stage I.
Conclusions
The findings support reducing healthcare discrimination to alleviate the disproportionate burden of worse prognosis and survival experienced by Black women.
Supplementary Information
The online version contains supplementary material available at 10.1186/s13058-025-02160-0.
Keywords: Breast cancer, Stage at diagnosis, Racial discrimination, Delayed diagnosis
Background
National data from the U.S. indicate that breast cancer mortality is 38% higher in Black women than in white women, even though breast cancer incidence is slightly higher in white women [1]. This mortality disparity is present for every breast cancer subtype, indicating that it is not entirely explained by the higher incidence of triple-negative breast cancer in U.S. Black women [1]. The differential is not consistent across all 50 states: recent data show that breast cancer mortality rates for Black women and white women are nearly identical in several states, including Rhode Island, Oregon, and Washington, whereas mortality is more than 50% higher among Black women than white women in other states, such as Mississippi, Arizona, and the District of Columbia [1]. These statistics suggest that the poorer survival from breast cancer experienced by Black women may, in part, be determined by external modifiable factors. Discrimination in health care is a modifiable factor that particularly affects Black Americans. In multiple studies, experiences of discrimination have been associated with negative patient experience with health services and delay in seeking healthcare [2, 3].
Stage at diagnosis has been shown to be the most powerful predictor of breast cancer survival [4–6]. Black women are more likely to be diagnosed at a later stage than any other U.S racial/ethnic group [1, 7]. It is critical, therefore, to identify the drivers of the advanced stage at diagnosis experienced disproportionately by Black women. We used data from a prospective cohort study of U.S. Black women, the Black Women’s Health Study (BWHS), to assess whether perceived discrimination in healthcare due to race is associated with a late stage at diagnosis of breast cancer.
Methods
Black Women’s Health Study
The BWHS began in 1995 when 59,000 women, aged 21–69 from across the U.S., enrolled by completing a detailed questionnaire about their health history, reproductive history, body size, behaviors, medication use, and usual diet [8]. The questionnaires were mailed to subscribers to Essence magazine who lived in 17 selected U.S. states. Participants who checked the option “Black” from a question on race and ethnicity were included. Follow-up questionnaires have been administered every two years, with options for mailed paper questionnaires or online questionnaires. Follow-up questionnaires ask about new diagnoses of breast cancer and other major health conditions and include questions on social, economic, and psychosocial factors as well as body size, health behaviors, and medication use. Participant addresses are updated every year, geocoded, and linked to national databases to characterize socioeconomic features of participants’ residential neighborhoods. The study protocol has been approved by the Boston University Medical Center Institutional Review Board (IRB) and the IRBs of state cancer registries.
Exposure and covariate assessment (Supplemental Table 1)
The 2003 follow-up questionnaire included questions about participants’ views on the health care they had received. Respondents were asked to check yes or no to each part of the following questions: “Do you think you receive health care that is different from what others receive because of: 1) your type of insurance and 2) your race?”, hereafter referred to as “perceived discrimination in healthcare (due to race/insurance)”. These questions were developed by the Boston Public Health Commission in 1999–2000 as part of their Women’s Health Questionnaire designed to assess social determinants of women’s health to improve infant mortality. The questionnaire was also used to screen women who participated in the REACH Boston 2010 Breast and Cervical Cancer initiative. We tested reliability of the questions in data from 1,821 BWHS participants who completed the 2003 questionnaire twice, 3–9 months apart, due to being sent a second mailing before their first questionnaire was received in the office. For the question on perceived discrimination due to race, 86% of responses were concordant (kappa 0.59, 95% CI 0.54–0.63); for the insurance status question, 75% were concordant (kappa 0.49, 95% CI 0.45–0.53). For the current work, the primary exposure was response to the question about perceived discrimination due to race. A secondary analysis examined associations with perceived discrimination due to insurance status.
Data on potential confounders were taken from the 2003 questionnaire (1999 for marital status). They included age at questionnaire completion, educational status, marital status, body mass index (BMI), mammographic screening two years prior to questionnaire completion, and neighborhood concentrated disadvantage score. Each BWHS questionnaire asked whether participants had had a mammogram in the previous two years.
Other discrimination questions were also available in BWHS, adapted from questions developed by Williams et al. and included on the 1997 questionnaire [9]. The questions assessed perceived experiences of everyday discrimination (how you are treated by other people) and lifetime experiences of being treated unfairly by the police, in housing, or in employment. As described previously [10, 11], possible responses to the six everyday discrimination questions ranging from 1(never) to 5 (almost every day) were summed to create a score and then categorized into quartiles, The lifetime discrimination variable was classified as no to all domains, yes to one, yes to two, or yes to three.
A variable that represents relative disadvantage of the neighborhood in which participants resided was derived using data from the U.S. Census Bureau and the American Community Survey, as described previously [11]. Briefly, a factor analysis of multiple block-group variables such as percentage of individuals on public assistance and percentage of female-headed households was used to create a neighborhood concentrated disadvantage score for each questionnaire year. The score variable was grouped into quartiles based on the entire BWHS population.
The 2005 BWHS questionnaire included questions from the Brief Cope Scale [12] The questions get at frequency of different forms of coping, represented by two to three sets of questions [12]. For this analysis we used frequency scores on the two questions that represent seeking “instrumental support”, stratifying participants as low or high based on the median score.
Outcome assessment
Incident breast cancers were identified through self-report on biennial questionnaires and by linkage with state cancer registries. Linkage with the 32 cancer registries in states in which > 95% of BWHS participants live was conducted annually to confirm self-reports, identify new cancers that were not self-reported, and obtain data on tumor characteristics. Linkage with the National Death Index was also used to identify breast cancers. The outcome variable for the current study was stage at breast cancer diagnosis.
Analytic study population
The current study was restricted to BWHS participants who were diagnosed with a first invasive breast cancer from October 2003 through May 2022 and had previously completed the 2003 BWHS biennial questionnaire (n = 1,899). We excluded women who had declined to answer the question on perceived discrimination due to race (n = 160) and women with missing data on stage at diagnosis (n = 122), most of whom had been identified through linkage with the National Death Index. The final study population included 1,617 participants, among whom 870 were diagnosed at stage I, 512 at stage II, 170 at stage III, and 65 at stage IV.
Statistical analysis
Logistic regression was used for estimation of odds ratios (OR) and 95% confidence intervals. Odds ratios (OR) were calculated for perceiving a difference in healthcare received due to race relative to the healthcare others receive for each stage of breast cancer diagnosis II, III, and IV relative to stage I. All analyses were adjusted for age in 2003 (continuous). Multivariable analyses were additionally adjusted for state of residence, years of education (≤ 12, 13–15, ≥ 16), BMI (< 25, 25–29, 30–34, ≥ 35 kg/m2), marital status (married or living as married, single, or widowed/divorced/separated), neighborhood concentrated disadvantage score (quartiles), and mammography screening in the two years prior to 2003. Variables for everyday and lifetime discrimination were not included in the models as neither was independently associated with stage at diagnosis (Supplemental Table 2). Multivariable models assessing perceived discrimination due to race were additionally controlled for perceived discrimination due to insurance status and vice versa.
We conducted additional analyses stratified on mammographic screening approximately two years before breast cancer diagnosis. This captured screening mammography rather than diagnostic mammography. We also conducted analyses stratified on age at breast cancer diagnosis (< 55, ≥ 55 years), attained education (< 16, ≥ 16 years), neighborhood concentrated disadvantage (above/below median), coping score (above/below median), everyday and lifetime racism variables, and year of breast cancer diagnosis (2003–2012, 2013–2022). The analysis stratified by coping score excluded women diagnosed with cancer before 2005, which is when those questions were asked. Due to limited numbers in stratified analyses, multivariable models included only age and mammographic screening. Interaction terms were created for perceived discrimination and stratifying variables. P values for interaction were obtained from likelihood ratio tests comparing models with and without the interaction term, using SAS PROC LOGISTIC procedure. All analyses were conducted using SAS version 9.4 (SAS Institute, Cary, NC). Statistical significance was defined as a two-sided P value < 0.05.
Results
Characteristics of the study population according to response to the question about perceived discrimination in healthcare due to race are shown in Table 1. Women who responded yes to perceived discrimination were similar to women who responded no in most respects, including age, history of mammographic screening, neighborhood disadvantage score, and body mass index. Women who were married or living as married were more likely to have said no, as were those in the lowest education level. Although perceived discrimination in healthcare due to race was correlated with discrimination due to insurance status (p < 0.001), there was some discordance between the two variables. For example, among women who reported perceived discrimination in healthcare due to race, 30% responded “no” to the question on differential healthcare due to insurance status.
Table 1.
Characteristics of study participants in 2003, according to perceived discrimination in healthcare due to race
| Characteristics at exposure assessment in 2003 | Perceived discrimination in healthcare due to race | |
|---|---|---|
| Yes (n = 335) N (%) | No (n = 1,282) N (%) | |
| Age | ||
| < 40 | 64 (19.1) | 201 (15.7) |
| 40–49 | 110 (32.9) | 429 (33.5) |
| 50–59 | 117 (34.9) | 427 (33.3) |
| 60–75 | 44 (13.1) | 225 (17.5) |
| Years of education | ||
| ≤ 12 | 33 (9.9) | 167 (13.0) |
| 13–15 | 89 (26.6) | 324 (25.3) |
| ≥ 16 | 213 (63.5) | 791 (61.7) |
| Marital status | ||
| Married/living as married | 143 (42.7) | 619 (48.3) |
| Single | 75 (22.4) | 266 (20.8) |
| Divorced/widowed/separated | 116 (34.6) | 396 (30.9) |
| Body mass index | ||
| < 25 | 69 (20.6) | 277 (21.6) |
| 25–29 | 119 (35.5) | 437 (34.1) |
| 30–34 | 76 (22.7) | 292 (22.8) |
| 35+ | 67 (20.0) | 233 (18.2) |
| Neighborhood disadvantage | ||
| Quartile 1 (least) | 79 (23.6) | 301 (23.5) |
| Quartile 2 | 68 (20.3) | 314 (24.5) |
| Quartile 3 | 80 (23.9) | 302 (23.6) |
| Quartile 4 (most) | 75 (22.4) | 267 (20.8) |
| Regular mammographic screening | ||
| Yes | 143 (42.7) | 609 (47.5) |
| No | 18 (5.4) | 43 (3.4) |
| Age < 50 | 174 (51.9) | 630 (49.1) |
| Everyday discrimination | ||
| Quartile 1 | 53 (15.8) | 379 (29.6) |
| Quartile 2 | 55 (16.4) | 288 (22.5) |
| Quartile 3 | 85 (25.4) | 326 (25.4) |
| Quartile 4 | 127 (37.9) | 235 (18.3) |
| Lifetime discrimination | ||
| No to all | 45 (13.4) | 428 (33.4) |
| Yes to 1 sphere | 79 (23.6) | 356 (27.8) |
| Yes to 2 spheres | 92 (27.5) | 251 (19.6) |
| Yes to 3 spheres | 81 (24.2) | 98 (7.6) |
| Instrumental coping | ||
| Low | 92 (33.7) | 437 (42.7) |
| High | 140 (51.3) | 453 (44.2 ) |
| Perceived discrimination in healthcare due to insurance status | ||
| Yes | 217 (64.8) | 374 (29.2) |
| No | 100 (29.9) | 904 (70.5) |
A stage IV diagnosis of breast cancer was reported twice as often in women who had previously responded yes to the question on perceived discrimination due to race (6.6%) as women who responded no (3.3%), while the proportions diagnosed at stages I, II, or III did not differ appreciably by response to that question (Table 2). In age-adjusted analyses, ORs for a response of yes were 2.02 (95% CI 1.18–3.48) for stage IV versus stage I, 1.01 (95% CI 0.67–1.52) for stage III versus stage I, and 0.98 (95% CI 0.75–1.29) for stage II versus stage I. ORs obtained from multivariable models were similar: OR 2.10 (95% CI 1.15–3.83) for stage IV, 1.00 (95% CI 0.64–1.55) for stage III, and 1.04 (95% CI 0.77–1.39) for stage II versus stage I.
Table 2.
Perceived discrimination in healthcare due to race in relation to stage at breast cancer diagnosis among 1,617 Black women with invasive breast cancer
| Stage at diagnosis | Perceived discrimination | Age-adjusted | Multivariable | |
|---|---|---|---|---|
| No N (%) | Yes N (%) | odds ratio (95% CI) | odds ratio* (95% CI) | |
| I | 695 (54.3) | 175 (52.2) | Reference | Reference |
| II | 409 (31.9) | 103 (30.8) | 0.98 (0.75–1.29) | 1.04 (0.77–1.39) |
| III | 135 (10.5) | 35 (10.4) | 1.01 (0.67–1.52) | 1.00 (0.64–1.55) |
| IV | 43 (3.3) | 22 (6.6) | 2.02 (1.18–3.48) | 2.10 (1.15–3.83) |
*Adjusted for age in 2003, state of residence, education, body mass index, neighborhood concentrated disadvantage, marital status, mammographic screening, and perceived discrimination in healthcare due to insurance status
In analyses stratified on mammographic screening 2–4 years before diagnosis, the estimated OR for stage IV versus stage I was higher among women who had a mammogram 2.42 (95% CI 1.29–4.53) compared to women who had not had a mammogram 1.41 (95% CI 0.45–4.40), although there was no statistically significant interaction (p-interaction 0.39) (Table 3). Similarly, there was no statistically significant interaction by other stratified factors, although estimated ORs for perceived discrimination in healthcare in association with a stage IV diagnosis were somewhat higher among women aged < 55 (2.90, 95% CI 1.78–7.78), women with ≥ 16 years of education (OR 2.35, 95% CI 1.20–4.59), women who were diagnosed in 2003–2012 (OR 2.70 (95% CI 1.27–5.73), and women below the median on coping score (OR 4.36, 95% CI 1.70–11.19).
Table 3.
Perceived discrimination in healthcare due to race in relation to late stage at breast cancer diagnosis, within strata of other factors
| Stage at diagnosis | Perceived discrimination | Odds ratio* 95% CI | Perceived discrimination | Odds ratio* 95% CI | P for interaction | ||
|---|---|---|---|---|---|---|---|
| No | Yes | No | Yes | ||||
| Age at diagnosis < 55 | Age at diagnosis 55–80 | 0.43 | |||||
| Stage I | 193 | 53 | Reference | 502 | 122 | Reference | |
| Stage IV | 10 | 8 | 2.90 (1.08 –7.78) | 33 | 14 | 1.83 (0.94–3.57) | |
| Recent screening mammogram | No recent mammogram | 0.39 | |||||
| Stage I | 615 | 156 | Reference | 77 | 19 | Reference | |
| Stage IV | 28 | 17 | 2.42 (1.29–4.53) | 15 | 5 | 1.41 (0.45–4.40) | |
| < 16 years of education | ≥ 16 years of education | 0.46 | |||||
| Stage I | 273 | 55 | Reference | 422 | 120 | Reference | |
| Stage IV | 19 | 6 | 1.69 (0.62–4.64) | 24 | 16 | 2.35 (1.20–4.59) | |
| Neighborhood concentrated disadvantage below median | Neighborhood concentrated disadvantage above median | 0.59 | |||||
| Stage I | 343 | 79 | Reference | 306 | 79 | Reference | |
| Stage IV | 16 | 8 | 2.13 (0.87–5.20) | 25 | 12 | 2.07 (0.98–4.38) | |
| Diagnosis year 2003–2012 | Diagnosis year 2013–2022 | 0.34 | |||||
| Stage I | 341 | 87 | Reference | 354 | 88 | Reference | |
| Stage IV | 19 | 13 | 2.70 (1.27–5.73) | 24 | 9 | 1.62 (0.72–3.67) | |
| Low instrumental coping | High instrumental coping | 0.39 | |||||
| Stage I | 243 | 46 | Reference | 240 | 77 | Reference | |
| Stage IV | 11 | 9 | 4.36 (1.70–11.19) | 12 | 9 | 2.61 (1.04–6.56) | |
| Everyday discrimination (Q1 & Q2) | Everyday discrimination (Q3 & Q4) | 0.92 | |||||
| Stage I | 364 | 53 | Reference | 298 | 117 | Reference | |
| Stage IV | 23 | 7 | 2.27 (0.91–5.70) | 19 | 15 | 2.11 (1.03–4.33) | |
| Lifetime discrimination (No to all & yes to 1 sphere) | Lifetime discrimination (Yes to 2 or 3 spheres) | 0.91 | |||||
| Stage I | 399 | 63 | Reference | 212 | 94 | Reference | |
| Stage IV | 26 | 9 | 2.57 (1.14–5.93) | 10 | 12 | 2.63 (1.09–6.36) | |
*Adjusted for age in 2003 and mammographic screening prior to 2003
As shown in Table 4, 37% of cases reported perceived discrimination in healthcare due to insurance status. Multivariable ORs for a yes response were 1.28 (95% CI 0.73, 2.24) for stage IV versus stage I, 1.21 (95% CI 0.85, 1.74) for III versus I, and 0.89 (95% CI 0.70, 1.14) for II versus I.
Table 4.
Perceived discrimination in healthcare due to insurance status in relation to breast cancer stage at diagnosis
| Stage at diagnosis | Perceived discrimination in healthcare due to insurance status | Age-adjusted | Multivariable | |
|---|---|---|---|---|
| No N (%) | Yes N (%) | odds ratio (95% CI) | odds ratio* (95% CI) | |
| I | 541 (53.9) | 314 (53.1) | Reference | Reference |
| II | 333 (33.2) | 176 (29.8) | 0.89 (0.70, 1.12) | 0.89 (0.70, 1.14) |
| III | 97 (9.6) | 71 (12.0) | 1.20 (0.86, 1.69) | 1.21 (0.85, 1.74) |
| IV | 33 (3.3) | 30 (5.1) | 1.55 (0.93, 2.60) | 1.28 (0.73, 2.24) |
*Adjusted for age in 2003, state of residence, years of education, body mass index, neighborhood concentrated disadvantage, marital status, mammographic screening, and perceived discrimination in healthcare due to race
Discussion
Data from 20 years of follow-up in the Black Women’s Health Study provided an opportunity to conduct a prospective analysis of perceived discrimination in healthcare in relation to subsequent late stage at diagnosis of breast cancer. The odds of being diagnosed with de novo stage IV breast cancer was twice as high among women who reported discrimination in healthcare due to race versus those who did not. Perceived discrimination was not associated with a higher odds of stage III or stage II breast cancer. Importantly, the association between perceived discrimination in healthcare and late stage at diagnosis remained after adjustment for other factors that may influence the timing of diagnosis, including educational level, living in a disadvantaged neighborhood, marital status, and recent mammographic screening. Further, it does not appear that the association was driven by differential utilization of mammographic screening. In stratified analyses, the association of perceived discrimination in health care with stage IV diagnosis was at least as strong among women who reported a mammogram two years prior to breast cancer diagnosis as among women who did not appear to have regular screening. ORs were somewhat higher among women diagnosed with breast cancer in 2003–2012 than among those diagnosed in 2013–2022. There may have been less misclassification of exposure in that group, whereas women diagnosed in later years would have had more time for adverse experiences in the healthcare setting not captured by the question in 2003. There is now a substantial literature documenting the health effects of discrimination due to race [13]; the present study demonstrates an impact on stage at breast cancer diagnosis, a key determinant of prognosis.
Regarding perceived discrimination due to insurance status, we found little evidence of an association with late stage at diagnosis after control for perceived discrimination due to race.
National data demonstrate that Black women with breast cancer are diagnosed at a later stage than any other population group defined by race or ethnicity [1]. In the most recently available data, 8% of breast cancers in Black women were diagnosed at stage IV, as compared with only 5% of breast cancers in white women [1]. This disparity translates into poorer survival for Black women, given that five-year survival across all populations in the U.S. is 99% if localized at diagnosis versus only 29% if distant [1]. Reducing the disparity in late stage at diagnosis between Black women and white women is a key step in reducing the mortality disparity.
Increasing utilization of mammographic screening has been the primary focus of efforts to reduce delayed diagnosis of breast cancer among Black women. In the past 20 years, mammography use has increased among U.S. Black women such that CDC data now indicate that Black women are as likely to engage in mammographic screening as women from other racial groups [14–16]. Indeed, in the BWHS, over 80% of women reported receiving at least one mammogram in a two-year period that was one-to-two years before their breast cancer diagnosis and therefore not a diagnostic mammogram.
Regular screening mammography, intended to detect very small tumors, is not the only modifiable step towards early diagnosis of breast cancer, however. The time from mammogram to biopsy is a key next step that influences stage at diagnosis. Several studies have shown that Black women experience a longer interval than white women between first symptoms or screening to biopsy-proven diagnosis [17–21]. That time interval can be influenced by when and how communication about the mammographic findings occurs, the ease or difficulty in getting an appointment for a biopsy, and the individual’s understanding and feelings about undergoing a biopsy. Each of these aspects may be influenced by past interpersonal experiences with the healthcare system. Experiences of being treated differently due to race may reduce trust in health care providers and negatively influence uptake of future health care.
Multiple studies have assessed the relation of racism to health services use and negative patient experience with health services [2, 3]. Experiences of racism were, in many cases, associated with lack of trust and delay in seeking healthcare [2, 3]. Perceived discrimination has been associated with lower use of preventive health screening, including mammography [22–24], but we are not aware of any prior studies that evaluated perceived discrimination due to race in relation to stage at breast cancer diagnosis. Hernandez et al. [25] reported on a study of 342 breast cancer patients (including 59 Black patients) from Miami who were queried using a validated everyday discrimination scale that asked about frequency of unfair treatment by others in everyday life. A higher score on that scale was associated with a somewhat higher odds of being diagnosed at stages III or IV versus stage I or II in the overall data set, but only 27% of those who reported discrimination attributed it to their race. Other major reasons were gender, national origins, and age. Although the study does not specifically address perceived discrimination due to race among Black women, it does provide valuable information on the association of perceived discrimination with medical mistrust. A higher perceived discrimination score was associated with lower scores on medical mistrust questions such as “you understand everything your doctor tells you”, “you believe your doctor has your best feelings at heart”, and “you trust your doctor”.
A recent study of reports of discrimination in health care by breast cancer patients treated at three U.S. medical centers found that Black women reported discrimination significantly more often than other women [26]. In particular, Black women were four times as likely to report being ignored in the health care setting as compared with white women and were also more likely to report being treated with less respect and with less courtesy. These experiences may translate into medical mistrust, inadequate communication, and delayed care. Previous research not restricted to breast cancer patients indicated that perceived discrimination due to race was associated with an increased odds of patients reporting mistrust [27]. In a study of cultural competency among physicians, adult cancer survivors who were not non-Hispanic white were less likely to report being able to see physicians who shared or understood their culture [28].
It is also possible that there are direct biologic effects from chronic stress stemming from discrimination. Chronic stress has been shown to result in dysregulation of the HPA axis as well as immune and inflammatory dysregulation, which could lead to more aggressive tumors, typically diagnosed at later stages [13, 29, 30]. Previous BWHS work indicated that living in disadvantaged neighborhoods was associated with increased risk of ER-negative but not ER-positive breast cancer [11]. The association remained after control for many lifestyle factors, suggesting that biologic effects of the chronic stress of living in disadvantaged neighborhoods may play a role. On the other hand, that same publication reported no association of racial discrimination with increased risk of ER- breast cancer [11].
A limitation of our study is that it was based on a single question and it did not have data to delineate the specific experiences of discrimination that led to delayed diagnosis; however, smaller qualitative studies have indicated that Black cancer patients report being ignored, treated with less respect, and treated with less courtesy than others [26]. Another limitation is the sample size. Although the overall sample size of 1,617 women with invasive breast cancer is large, only 65 women had been diagnosed with de novo stage IV breast cancer. The observed association was statistically significant despite the small number of stage IV cases, but power was limited for stratified analyses. Of note, a strong association was observed in the subgroup of women with low coping score, but there was not a statistically significant interaction. A major strength of the study was the prospective design, in that all participants reported on their experiences in receipt of health care years before being diagnosed with breast cancer. Data from cohort studies that query women before they develop cancer will be needed to replicate these findings. As an alternative, studies of newly diagnosed breast cancer patients could ask specific questions about how the cancer came to diagnosis, including timing and any factors they can identify as affecting the timing.
In summary, the present findings suggest that past experiences of racial discrimination in interactions with the healthcare system may pose an impediment to early diagnosis of breast cancer for Black women. A new or continued emphasis on education/training for healthcare providers to avoid bias and approach all patients with courtesy and respect may be a critical step towards reducing the Black/white disparity in stage at breast cancer diagnosis, thus leading to better survival and quality of life after a breast cancer diagnosis for all women.
Supplementary Information
Acknowledgements
We thank JudyAnn Bigby, MD, former Secretary of Health and Human Services for the Commonwealth of Massachusetts, who suggested the questions on differential treatment in health care that were included in the 2003 BWHS questionnaire. We also thank the participants and staff of the Black Women’s Health Study for their important contributions. The authors acknowledge the contributions to this study from central cancer registries supported through the Centers for Disease Control and Prevention’s National Program of Cancer Registries and/or the NCI’s Surveillance, Epidemiology, and End Results Program. Central registries may also be supported by state agencies, universities, and cancer centers. Participating central cancer registries include the following: AL, AR, AZ, CA, CO, CT, DE, DC, FL, GA, HI, IA, IL, IN, KY, LA, MD, MA, MI, MO, MS, NE, NJ, NM, NY, NC, OH, OK, OR, PA, SC, TN, TX, VA, WA, and WI.
Author contributions
NNX carried out all analyses, prepared the tables, and co-wrote the manuscript with JRP. MEB participated in interpretation of the results, made suggestions for additional analyses, and provided edits for the manuscript. EXH participated in interpretation of the results and provided edits for the manuscript. NK participated in interpretation of the results and provided edits for the manuscript. LR participated in interpretation of the results and provided edits for the manuscript. JRP conceived the hypothesis, designed the analysis, and co-wrote the manuscript with NNX.
Funding
Funding for this research was provided by NIH grant U01 CA164974 and Susan G. Komen Foundation grant SAC220228. Drs. Palmer and Barnard also receive support from the Breast Cancer Research Foundation and the Karin Grunebaum Cancer Research Foundation. The content of this study is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Data availability
The de-identified data used in this article may be shared upon approval of a data request by the Black Women’s Health Study Data Access Committee and with appropriate human subjects approvals and data transfer agreements. For more information contact the corresponding author.
Declarations
Ethics approval and consent to participate
The Black Women’s Health Study (BWHS) has been approved by the Institutional Review Board of the Boston University Medical Center.
Consent for publication
BWHS participants give implied consent by completing biennial questionnaires about their health and habits and returning the questionnaires to the study offices. Each questionnaire is accompanied by a document that explains the elements of informed consent.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Giaquinto AN, Sung H, Newman LA, et al. Breast cancer statistics 2024. CA Cancer J Clin. 2024;74(6):477–95. 10.3322/caac.21863. [DOI] [PubMed] [Google Scholar]
- 2.Ben J, Cormack D, Harris R, Paradies Y. Racism and health service utilisation: a systematic review and meta-analysis. PLoS ONE. 2017;12(12):e0189900. 10.1371/journal.pone.0189900. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Hamed S, Bradby H, Ahlberg BM, Thapar-Björkert S. Racism in healthcare: a scoping review. BMC Publ Health. 2022;22(1):988. 10.1186/s12889-022-13122-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Ellis L, Canchola AJ, Spiegel D, Ladabaum U, Haile R, Gomez SL. Racial and ethnic disparities in cancer survival: the contribution of tumor, sociodemographic, institutional, and neighborhood characteristics. J Clin Oncol. 2018. 10.1200/JCO.2017.74.2049. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Coughlin SS, Richardson LC, Orelien J, et al. Contextual analysis of breast cancer stage at diagnosis among women in the United States, 2004. Open Health Serv Policy J. 2009;2:45–6. [PMC free article] [PubMed] [Google Scholar]
- 6.Miller-Kleinhenz JM, Collin LJ, Seidel R, et al. Racial disparities in diagnostic delay among women with breast cancer. J Am Coll Radiol. 2021;18(10):1384–93. 10.1016/j.jacr.2021.06.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Reeder-Hayes KE, Jackson BE, Baggett CD, et al. Race, geography, and risk of breast cancer treatment delays: a population-based study 2004–2015. Cancer. 2023;129(6):925–33. 10.1002/cncr.34573. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Palmer JR, Wise LA, Horton NJ, Adams-Campbell LL, Rosenberg L. Dual effect of parity on breast cancer risk in African-American women. JNCI J Natl Cancer Inst. 2003;95(6):478–83. 10.1093/jnci/95.6.478. [DOI] [PubMed] [Google Scholar]
- 9.Williams DR, Yan Yu null, Jackson JS, Anderson NB. Racial differences in physical and mental health: Socio-economic Status, Stress and Discrimination. J Health Psychol. 1997;2(3):335–351. 10.1177/135910539700200305 [DOI] [PubMed]
- 10.Cozier YC, Yu J, Coogan PF, Bethea TN, Rosenberg L, Palmer JR. Racism, segregation, and risk of obesity in the Black Women’s Health Study. Am J Epidemiol. 2014;179(7):875–83. 10.1093/aje/kwu004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Barber LE, Zirpoli GR, Cozier YC, et al. Neighborhood disadvantage and individual-level life stressors in relation to breast cancer incidence in US Black women. Breast Cancer Res. 2021;23:108. 10.1186/s13058-021-01483-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Carver CS. You want to measure coping but your protocol’s too long: consider the brief COPE. [DOI] [PubMed]
- 13.Selvarajah S, Corona Maioli S, Deivanayagam TA, et al. Racism, xenophobia, and discrimination: mapping pathways to health outcomes. Lancet. 2022;400(10368):2109–24. 10.1016/S0140-6736(22)02484-9. [DOI] [PubMed] [Google Scholar]
- 14.Sabatino SA. Cancer screening test receipt—United States, 2018. MMWR Morb Mortal Wkly Rep. 2021;70. 10.15585/mmwr.mm7002a1 [DOI] [PMC free article] [PubMed]
- 15.Miller JW, King JA, Trivers KF, et al. Vital signs: mammography use and association with social determinants of health and health-related social needs among women—United States, 2022. MMWR Morb Mortal Wkly Rep. 2024;73(15):351–7. 10.15585/mmwr.mm7315e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.White A. Cancer screening test use—United States, 2015. MMWR Morb Mortal Wkly Rep. 2017. 10.15585/mmwr.mm6608a1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.George P, Chandwani S, Gabel M, et al. Diagnosis and surgical delays in African American and white women with early-stage breast cancer. J Womens Health. 2015;24(3):209–17. 10.1089/jwh.2014.4773. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Gorin SS, Heck JE, Cheng B, Smith SJ. Delays in breast cancer diagnosis and treatment by racial/ethnic group. Arch Intern Med. 2006;166(20):2244–52. 10.1001/archinte.166.20.2244. [DOI] [PubMed] [Google Scholar]
- 19.Molina Y, Silva A, Rauscher GH. Racial/ethnic disparities in time to a breast cancer diagnosis: the mediating effects of healthcare facility factors. Med Care. 2015;53(10):872–8. 10.1097/MLR.0000000000000417. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Press R, Carrasquillo O, Sciacca RR, Giardina EGV. Racial/ethnic disparities in time to follow-up after an abnormal mammogram. J Womens Health. 2008;17(6):923–30. 10.1089/jwh.2007.0402. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Williams DL, Tortu S, Thomson J. Factors associated with delays to diagnosis and treatment of breast cancer in women in a Louisiana urban safety net hospital. Women Health. 2010;50(8):705–18. 10.1080/03630242.2010.530928. [DOI] [PubMed] [Google Scholar]
- 22.Crawley LM, Ahn DK, Winkleby MA. Perceived medical discrimination and cancer screening behaviors of racial and ethnic minority adults. Cancer Epidemiol Biomarkers Prev. 2008;17(8):1937–44. 10.1158/1055-9965.EPI-08-0005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Dailey AB, Kasl SV, Holford TR, Jones BA. Perceived racial discrimination and nonadherence to screening mammography guidelines: results from the race differences in the screening mammography process study. Am J Epidemiol. 2007;165(11):1287–95. 10.1093/aje/kwm004. [DOI] [PubMed] [Google Scholar]
- 24.Jacobs EA, Rathouz PJ, Karavolos K, et al. Perceived discrimination is associated with reduced breast and cervical cancer screening: the study of women’s health across the nation (SWAN). J Womens Health. 2014;23(2):138–45. 10.1089/jwh.2013.4328. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Hernandez AE, Borowsky PA, Lubarsky M, et al. Associations between perceived discrimination, screening mammography, and breast cancer stage at diagnosis: a prospective cohort analysis. Ann Surg Oncol. 2024;31(12):8012–20. 10.1245/s10434-024-15757-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Oshry LJ, Lederman RI, Gagnon H, et al. Racial discrimination among women seeking breast cancer care. NPJ Breast Cancer. 2025;11:4. 10.1038/s41523-025-00717-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Bazargan M, Cobb S, Assari S. Discrimination and medical mistrust in a racially and ethnically diverse sample of California adults. Ann Family Med. 2021;19(1):4–15. 10.1370/afm.2632. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Butler SS, Winkfield KM, Ahn C, et al. Racial disparities in patient-reported measures of physician cultural competency among cancer survivors in the United States. JAMA Oncol. 2020;6(1):152–4. 10.1001/jamaoncol.2019.4720. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Vona-Davis L, Rose DP. The influence of socioeconomic disparities on breast cancer tumor biology and prognosis: a review. J Womens Health (Larchmt). 2009;18(6):883–93. 10.1089/jwh.2008.1127. [DOI] [PubMed] [Google Scholar]
- 30.Saini G, Ogden A, McCullough LE, Torres M, Rida P, Aneja R. Disadvantaged neighborhoods and racial disparity in breast cancer outcomes: the biological link. Cancer Causes Control. 2019;30(7):677–86. 10.1007/s10552-019-01180-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The de-identified data used in this article may be shared upon approval of a data request by the Black Women’s Health Study Data Access Committee and with appropriate human subjects approvals and data transfer agreements. For more information contact the corresponding author.