Background:

Prior studies of breast cancer disparities have focused primarily on differences between Black and White women, yet contemporary patterns of disparity for other groups are not well understood. We examine breast cancer disparities by stage at diagnosis across nine racial and ethnic groups.

Methods:

The SEER 18 registries identified 841,975 women diagnosed with breast cancer from 2000 to 2017. Joinpoint models assessed trends in diagnosis stage and survival. Multivariable logistic regression evaluated associations between race/ethnicity and diagnosis stage. Multivariable Cox models compared survival of groups by stage and molecular subtype.

Results:

Black, American Indian, Southeast Asian, South Asian, Pacific Islander, and Hispanic women were less likely than white women to be diagnosed with early stage breast cancer. Among those diagnosed at early stage, Hispanic, American Indian, Pacific Islander and Black women were 9%, 14%, 22%, and 39% (respectively) more likely than White women to die from breast cancer, whereas Asian subgroups had lower risk of death. Among those diagnosed at late stage, Black women were 18% more likely than White counterparts to die from breast cancer, and survival disparities for Black women persisted across all subtypes and stages, (except late stage HR/HER2). East Asian women with early stage HR+/HER2 tumors had better survival than White women.

Conclusions:

Persistent disparities in early detection and survival of breast cancer demand further work to address and reduce disparities across the cancer continuum.

Impact:

Results have implications for efforts to reduce entrenched racial and ethnic disparities in breast cancer early detection and survival.

This article is featured in Highlights of This Issue, p. 1137

Over the past two decades, reductions in breast cancer mortality have been attributed to improvements in treatment, early detection by mammography screening, and primary prevention efforts to address risk factors (1). These advances, in addition to improving breast cancer mortality, aim to shift disease detection from late to early stages when treatment is most effective. However, not all women have benefited equally from this progress, as evidenced by variation in in disease burden across race/ethnic groups. For example, Black women are disproportionately affected by more aggressive breast cancer subtypes, such as triple-negative breast cancer, and are more likely to be diagnosed at younger ages and at more advanced stages of the disease compared with other groups (1–3). Despite improvements in early detection and targeted treatment for breast cancer, Black women continue to have the highest breast cancer mortality rate (3, 4).

Monitoring persistent and potentially worsening disparities remains an important focus of efforts to improve equity in breast cancer detection and treatment. Existing studies of breast cancer disparities have focused primarily on differences in outcomes among broad aggregate racial/ethnic categories (4–10), which may inadvertently mask important differences within distinct racial/ethnic subgroups. For example, Asian Americans are typically reported as one aggregated racial category, and often grouped together with Pacific Islanders. Yet, substantial variations in breast cancer outcomes within these groups have been reported (11–14). With the U.S. population growing racially and ethnically diverse, reports of disparity trends using aggregate racial/ethnic groups may no longer be sufficient.

Using the most recent data available from the NCI's Surveillance, Epidemiology, and End Results Program (SEER; ref. 15), we provide a contemporary analyses of breast cancer trends and disparities by stage at diagnosis, grouping cases most likely detected via screening at “early stage” (i.e., stage I–IIIA) separately from those detected at more advanced stage at diagnosis (i.e., stage IIIB–IV), for comparison across nine mutually exclusive racial and ethnic groups. We also consider comparison by molecular subtype given previously reported variability by race and ethnicity (6, 14).

Data source

We used data from the SEER 18 registry research plus database (November 2020 submission; ref. 15). The SEER 18 database includes data from 18 population-based registries in 14 states, capturing approximately 28% of the U.S. population (15). All SEER data were publicly available, deidentified, and therefore, determined to be exempt from Institutional Review Board review. However, a data use agreement submission was required to access the SEER Research Data File (16). Following acceptance of the agreement, SEER*Stat Version 8.3.9 software and data files were downloaded directly from the SEER website.

Cohort selection

Patients diagnosed with first primary breast cancer between 2000 and 2017 were identified, using the Classification of Diseases for Oncology, 3rd edition with site codes C50.0-C50.9. We excluded 2018 data due to recent changes and inconsistencies in cancer coding and staging criteria for patients diagnosed after 2017 (17). Additional exclusion criteria included: male breast cancers, patients age<20 years old at diagnosis, breast cancer that was not the first primary, in situ or stage 0 cases, cancer diagnosis by death certificate only, unknown or missing staging information, and unknown race/ethnicity (Supplementary Fig. S1). For survival analysis, we further excluded cases with unknown or zero survival months.

Measures

The primary outcomes were breast cancer incidence, stage at diagnosis, and breast cancer specific survival. Using the American Joint Committee on Cancer (AJCC) 6th edition criteria, we define early-stage breast cancer as stages I–IIIA and late stage as stages IIIB–IV (18). Breast cancer survival was derived from SEER's cause-specific death classification and vital status. Follow-up time was measured from the date of diagnosis until either death or end of follow-up (December 31, 2017).

Our primary predictor was race/ethnicity, categorized into nine mutually exclusive groups: White, Black, American Indian or Alaska Native (AIAN), Hispanic, Pacific Islander/Hawaiian and four Asian subgroups: East Asian (Chinese, Japanese, Korean), South Asian (Asian Indian or Pakistani), Southeast Asian (Filipino, Vietnamese, Thai, Loatian, Kampuchean, Hmong) and Other Asian. Information on race and ethnicity in the SEER database are obtained from patient medical records. Other demographic variables included age at diagnosis (20–34, 35–49, 50–64, 65–79, ≥80), marital status (married, not married, unknown), county residence (metropolitan, nonmetropolitan, unknown), registry region (Northeast, Midwest, South, West), and median household income (<$55,000, $55,000–$74,999, ≥$75,000, unknown).

Clinical characteristics included year of diagnosis, tumor grade (well differentiated, moderately differentiated, poorly differentiated, unknown), radiation treatment (yes, no or unknown), surgery of primary site (yes, no or unknown), and hormone receptor (HR) status [based on combined estrogen receptor (ER) and progesterone receptor (PR) expression]. Cases classified as HR positive were positive for ER and/or PR expression. Breast cancer subtype was derived from SEER based on joint hormone receptor (HR) and HER2 status, and tumors were classified into four mutually exclusive categories: HR+/HER2; HR+/HER2+; HR/HER2+; and HR/HER2 (triple-negative). The SEER program has recorded HR status since 1990, but did not begin collecting HER2 status data until 2010. As such, our analyses stratified by molecular subtype are limited to women diagnosed in 2010 or later.

To characterize racial and ethnic disparities in breast cancer outcomes, we used a set of summary disparity metrics on both the absolute (range difference) and relative (range ratio) scales. The range difference (RD) is a measure of absolute disparity, defined as the difference between highest and lowest values for each race/ethnicity groups for each outcome. We use the range ratio (RR) to measure disparities on the relative scale, derived by dividing the highest rate value by the lowest rate value. Additional details on the calculation of disparity metrics are described elsewhere (19).

Statistical analysis

We used SEER* stat (version 8.3.5) software to extract raw data from the SEER 18 database. Incidence and 5-year cause-specific survival were estimated using the rate and survival functions in SEER*Stat. Incidence rates were age-adjusted to the 2000 U.S. standard population. We used Joinpoint Trend Analysis software (20) version 4.8.0.1 to analyze trends in the percent of early stage breast cancer 5-year survival, and disparity measures over time, presented as average annual percentage changes (AAPC). Multivariable logistic regression model assessed associations between race/ethnicity and breast cancer stage. Kaplan–Meier survival curves with log-rank statistical tests assessed differences in survival across race/ethnicity and stage at diagnosis. After proportionality of hazards was confirmed, multivariable Cox regression analyses compared breast cancer–specific survival of race/ethnic groups by stage at diagnoses. Using the likelihood ratio test, we tested for effect modification by comparing models including interaction terms formed from stage and race/ethnicity variables with models containing only main effects. Odds ratio, hazard ratio and its 95% confidence intervals (CI) were reported based on logistic regression and Cox regression analysis. In subgroup analyses, we compare differences in breast-cancer specific survival by molecular subtype among women diagnosed in 2010 or later.

All regression models were predictive, not causal. A Bonferroni correction was applied to multivariable regression models, such that only results with P ≤ 0.006 were considered statistically significant. Regression analyses of were performed using STATA version 15.0 (RRID:SCR_012763; ref. 21).

Data availability

The data presented in this study are available publicly on the SEER website and database (https://seer.cancer.gov, RRID:SCR_006902).

Study population

A total of 841,975 women were diagnosed with stage I to IV breast cancer between 2000 and 2017 (70.4% White; 10.6% Black; 10.5% Hispanic; 3% East Asian; 2.6% Southeast Asian; 0.7% South Asian; 0.6% Pacific Islander; 0.5% AIAN; and 1.0% Other Asian; Table 1). The percent of cases diagnosed at early stages (I–IIIA) was highest among East Asian (92.1%) and Other Asian (92.9%) women and lowest among Black (83.9%) women. Most breast cancer cases were ER+/PR+ (63.9%) and the percent of HR+ cases were highest among Other Asian (67.9%) women, followed closely by White (65.8%) and Pacific Islander (65.8%) women, and lowest among Black women (50.2%).

Table 1.

Characteristics of women diagnosed with breast cancer between 2000 and 2017 in the SEER 18 Registries Database.

Race/ethnicity
WhiteBlackAmerican IndianEast AsianSouth AsianSoutheast AsianOther AsianPacific IslanderHispanicTotal
N (%)a 592,874 (70.4) 89,297 (10.6) 4,360 (0.5) 24,964 (3.0) 6,178 (0.7) 22,206 (2.6) 8,623 (1.0) 5,202 (0.6) 88,271 (10.5) 841,975 
AJCC diagnosis stage 
 I 297,466 (50.2) 32,590 (36.5) 1,889 (43.3) 12,606 (50.5) 2,497 (40.4) 9,620 (43.3) 4,279 (49.6) 2,163 (41.6) 35,260 (40.0) 398,370 (47.3) 
 IIA 138,647 (23.4) 22,552 (25.3) 1,113 (25.5) 6,251 (25.0) 1,591 (25.8) 5,707 (25.7) 2,223 (25.8) 1,275 (24.5) 22,571 (25.6) 201,930 (24.0) 
 IIB 59,545 (10.0) 12,027 (13.5) 528 (12.1) 2,679 (10.7) 827 (13.4) 2,802 (12.6) 1,029 (11.9) 648 (12.5) 11,829 (13.4) 91,914 (10.9) 
 IIIA 36,489 (6.2) 7,739 (8.7) 299 (6.9) 1,464 (5.9) 519 (8.4) 1,656 (7.5) 483 (5.6) 416 (8.0) 7,750 (8.8) 56,815 (6.8) 
 IIIB 12,675 (2.1) 3,146 (3.5) 101 (2.3) 441 (1.8) 158 (2.6) 513 (2.3) 125 (1.5) 143 (2.8) 2,364 (2.7) 19,666 (2.3) 
 IIIC 18,333 (3.1) 4,002 (4.5) 172 (3.9) 688 (2.8) 233 (3.8) 832 (3.8) 205 (2.4) 202 (3.9) 3,923 (4.4) 28,590 (3.4) 
 IV 29,719 (5) 7,241 (8.1) 258 (5.9) 835 (3.3) 353 (5.7) 1,076 (4.9) 279 (3.2) 355 (6.8) 4,574 (5.2) 44,690 (5.3) 
Age at diagnosis 
 20–34 9,194 (1.6) 2,989 (3.4) 120 (2.8) 577 (2.3) 342 (5.5) 561 (2.5) 296 (3.4) 140 (2.7) 3,534 (4) 17,753 (2.1) 
 35–49 109,082 (18.4) 22,513 (25.2) 1,136 (26.1) 6,969 (27.9) 2,048 (33.1) 6,210 (28) 2,825 (32.8) 1,265 (24.3) 27,258 (30.9) 179,306 (21.3) 
 50–64 222,267 (37.5) 35,478 (39.7) 1,825 (41.9) 9,162 (36.7) 2,261 (36.6) 9,519 (42.9) 3,425 (39.7) 2,256 (43.4) 33,457 (37.9) 319,650 (38) 
 65–79 187,470 (31.6) 22,180 (24.8) 1,076 (24.7) 6,192 (24.8) 1,336 (21.6) 5,110 (23) 1,717 (19.9) 1,320 (25.4) 19,553 (22.2) 245,954 (29.2) 
 80+ 64,861 (10.9) 6,137 (6.9) 203 (4.7) 2,064 (8.3) 191 (3.1) 806 (3.6) 360 (4.2) 221 (4.2) 4,469 (5.1) 79,312 (9.4) 
Marital Status 
 Married 341,807 (57.7) 31,103 (34.8) 2,020 (46.3) 16,462 (65.9) 4,698 (76) 14,163 (63.8) 5,788 (67.1) 2,871 (55.2) 48,443 (54.9) 467,355 (55.5) 
 Not married 226,616 (38.2) 53,863 (60.3) 1,882 (43.2) 7,759 (31.1) 1,285 (20.8) 7,249 (32.6) 2,419 (28.1) 2,141 (41.2) 35,775 (40.5) 338,989 (40.3) 
 Unknown 24,451 (4.1) 4,331 (4.9) 458 (10.5) 743 (3.0) 195 (3.2) 794 (3.6) 416 (4.8) 190 (3.7) 4,053 (4.6) 35,631 (4.2) 
Household Income 
 ≥$75k 201,048 (33.9) 17,707 (19.8) 1,735 (39.8) 15,191 (60.9) 3,534 (57.2) 11,002 (49.5) 4,374 (50.7) 201,048 (33.9) 17,707 (19.8) 1,735 (39.8) 
 $74,999–$55,000 270,115 (45.6) 41,620 (46.6) 1,405 (32.2) 9,333 (37.4) 2,192 (35.5) 10,559 (47.6) 3,736 (43.3) 270,115 (45.6) 41,620 (46.6) 1,405 (32.2) 
 >$55k 121,640 (20.5) 2,9969 (33.6) 1,219 (28) 439 (1.8) 452 (7.3) 645 (2.9) 513 (5.9) 121,640 (20.5) 29,969 (33.6) 1,219 (28) 
HRa Stat 
 ER+/PR+ 390,327 (65.8) 44,822 (50.2) 2,785 (63.9) 16,511 (66.1) 4,016 (65) 14,095 (63.5) 5,859 (67.9) 390,327 (65.8) 44,822 (50.2) 2,785 (63.9) 
 ER+/PR- 5,750 (1) 1,496 (1.7) 70 (1.6) 292 (1.2) 85 (1.4) 204 (0.9) 87 (1) 5,750 (1) 1,496 (1.7) 70 (1.6) 
 ER−/PR+ 65,540 (11.1) 10,911 (12.2) 455 (10.4) 2,694 (10.8) 611 (9.9) 2,550 (11.5) 891 (10.3) 65,540 (11.1) 10,911 (12.2) 455 (10.4) 
 ER−/PR− 87,793 (14.8) 24,729 (27.7) 805 (18.5) 3,997 (16) 1,154 (18.7) 3,850 (17.3) 1,256 (14.6) 87,793 (14.8) 24,729 (27.7) 805 (18.5) 
 Unknown 43,464 (7.3) 7,339 (8.2) 245 (5.6) 1,470 (5.9) 312 (5.1) 1,507 (6.8) 530 (6.1) 43,464 (7.3) 7,339 (8.2) 245 (5.6) 
Grade 
 Well differentiated 131,830 (22.2) 11,308 (12.7) 855 (19.6) 5,250 (21) 977 (15.8) 3,525 (15.9) 1,743 (20.2) 131,830 (22.2) 11,308 (12.7) 855 (19.6) 
 Moderately differentiated 245,198 (41.4) 30,130 (33.7) 1,709 (39.2) 10,413 (41.7) 2,389 (38.7) 9,244 (41.6) 3,705 (43) 245,198 (41.4) 30,130 (33.7) 1,709 (39.2) 
 Poorly differentiated 176,454 (29.8) 40,624 (45.5) 1,560 (35.8) 7,929 (31.8) 2,410 (39) 8,149 (36.7) 2,700 (31.3) 176,454 (29.8) 40,624 (45.5) 1,560 (35.8) 
 Unknown 39,392 (6.6) 7,235 (8.1) 236 (5.4) 1,372 (5.5) 402 (6.5) 1,288 (5.8) 475 (5.5) 39,392 (6.6) 7,235 (8.1) 236 (5.4) 
County residence 
 Metropolitan 519,449 (87.6) 82,212 (92.1) 2,656 (60.9) 24,174 (96.8) 6,120 (99.1) 21,718 (97.8) 8,508 (98.7) 519,449 (87.6) 82,212 (92.1) 2,656 (60.9) 
 Nonmetropolitan 73,354 (12.4) 7,084 (7.9) 751 (17.2) 789 (3.2) 58 (0.9) 488 (2.2) 115 (1.3) 73,354 (12.4) 7,084 (7.9) 751 (17.2) 
Region 
 Northeast 107,888 (18.2) 14,029 (15.7) 123 (2.8) 1,722 (6.9) 1,894 (30.7) 1,393 (6.3) 534 (6.2) 107,888 (18.2) 14,029 (15.7) 123 (2.8) 
 Midwest 63,377 (10.7) 10,396 (11.6) 141 (3.2) 232 (0.9) 285 (4.6) 171 (0.8) 299 (3.5) 63,377 (10.7) 10,396 (11.6) 141 (3.2) 
 South 133,374 (22.5) 41,254 (46.2) 161 (3.7) 599 (2.4) 635 (10.3) 429 (1.9) 534 (6.2) 133,374 (22.5) 41,254 (46.2) 161 (3.7) 
 West 288,235 (48.6) 23,618 (26.5) 3,935 (90.3) 22,411 (89.8) 3,364 (54.5) 20,213 (91) 7,256 (84.1) 288,235 (48.6) 23,618 (26.5) 3,935 (90.3) 
Diagnosis year 
 2000–2002 98,371 (16.6) 12,011 (13.5) 527 (12.1) 3,435 (13.8) 526 (8.5) 2,628 (11.8) 678 (7.9) 98,371 (16.6) 12,011 (13.5) 527 (12.1) 
 2003–2005 93,018 (15.7) 12,720 (14.2) 619 (14.2) 3,570 (14.3) 651 (10.5) 2,913 (13.1) 907 (10.5) 93,018 (15.7) 12,720 (14.2) 619 (14.2) 
 2006–2008 95,762 (16.2) 14,021 (15.7) 640 (14.7) 3,902 (15.6) 834 (13.5) 3,426 (15.4) 1,141 (13.2) 95,762 (16.2) 14,021 (15.7) 640 (14.7) 
 2009–2011 98,947 (16.7) 15,456 (17.3) 790 (18.1) 4,282 (17.2) 1,037 (16.8) 3,805 (17.1) 1,459 (16.9) 98,947 (16.7) 15,456 (17.3) 790 (18.1) 
 2012–2014 101,714 (17.2) 17,020 (19.1) 879 (20.2) 4,593 (18.4) 1,384 (22.4) 4,434 (20) 1,907 (22.1) 101,714 (17.2) 17,020 (19.1) 879 (20.2) 
 2015–2017 105,062 (17.7) 18,069 (20.2) 905 (20.8) 5,182 (20.8) 1,746 (28.3) 5,000 (22.5) 2,531 (29.4) 105,062 (17.7) 18,069 (20.2) 905 (20.8) 
Vital status at follow-up - n (%) 
 Alive 423,792 (71.5) 58,409 (65.4) 3,168 (72.8) 20,395 (81.7) 5,273 (85.4) 18,151 (81.7) 7,711 (89.4) 3,858 (74.2) 68,858 (78.0) 609,615 (72.4) 
 Due to breast cancer 71,666 (12.1) 17,899 (20.0) 588 (13.5) 2,211 (8.9) 569 (9.2) 2,394 (10.8) 514 (6.0) 694 (13.3) 11,211(12.7) 107,746 (12.8) 
 Due to other cause 97,416 (16.4) 12,989 (14.5) 604 (13.8) 2,358 (9.5) 336 (5.4) 1,661 (7.5) 4,398 (4.6) 650 (12.5) 8,202 (9.3) 124,614 (14.8) 
Race/ethnicity
WhiteBlackAmerican IndianEast AsianSouth AsianSoutheast AsianOther AsianPacific IslanderHispanicTotal
N (%)a 592,874 (70.4) 89,297 (10.6) 4,360 (0.5) 24,964 (3.0) 6,178 (0.7) 22,206 (2.6) 8,623 (1.0) 5,202 (0.6) 88,271 (10.5) 841,975 
AJCC diagnosis stage 
 I 297,466 (50.2) 32,590 (36.5) 1,889 (43.3) 12,606 (50.5) 2,497 (40.4) 9,620 (43.3) 4,279 (49.6) 2,163 (41.6) 35,260 (40.0) 398,370 (47.3) 
 IIA 138,647 (23.4) 22,552 (25.3) 1,113 (25.5) 6,251 (25.0) 1,591 (25.8) 5,707 (25.7) 2,223 (25.8) 1,275 (24.5) 22,571 (25.6) 201,930 (24.0) 
 IIB 59,545 (10.0) 12,027 (13.5) 528 (12.1) 2,679 (10.7) 827 (13.4) 2,802 (12.6) 1,029 (11.9) 648 (12.5) 11,829 (13.4) 91,914 (10.9) 
 IIIA 36,489 (6.2) 7,739 (8.7) 299 (6.9) 1,464 (5.9) 519 (8.4) 1,656 (7.5) 483 (5.6) 416 (8.0) 7,750 (8.8) 56,815 (6.8) 
 IIIB 12,675 (2.1) 3,146 (3.5) 101 (2.3) 441 (1.8) 158 (2.6) 513 (2.3) 125 (1.5) 143 (2.8) 2,364 (2.7) 19,666 (2.3) 
 IIIC 18,333 (3.1) 4,002 (4.5) 172 (3.9) 688 (2.8) 233 (3.8) 832 (3.8) 205 (2.4) 202 (3.9) 3,923 (4.4) 28,590 (3.4) 
 IV 29,719 (5) 7,241 (8.1) 258 (5.9) 835 (3.3) 353 (5.7) 1,076 (4.9) 279 (3.2) 355 (6.8) 4,574 (5.2) 44,690 (5.3) 
Age at diagnosis 
 20–34 9,194 (1.6) 2,989 (3.4) 120 (2.8) 577 (2.3) 342 (5.5) 561 (2.5) 296 (3.4) 140 (2.7) 3,534 (4) 17,753 (2.1) 
 35–49 109,082 (18.4) 22,513 (25.2) 1,136 (26.1) 6,969 (27.9) 2,048 (33.1) 6,210 (28) 2,825 (32.8) 1,265 (24.3) 27,258 (30.9) 179,306 (21.3) 
 50–64 222,267 (37.5) 35,478 (39.7) 1,825 (41.9) 9,162 (36.7) 2,261 (36.6) 9,519 (42.9) 3,425 (39.7) 2,256 (43.4) 33,457 (37.9) 319,650 (38) 
 65–79 187,470 (31.6) 22,180 (24.8) 1,076 (24.7) 6,192 (24.8) 1,336 (21.6) 5,110 (23) 1,717 (19.9) 1,320 (25.4) 19,553 (22.2) 245,954 (29.2) 
 80+ 64,861 (10.9) 6,137 (6.9) 203 (4.7) 2,064 (8.3) 191 (3.1) 806 (3.6) 360 (4.2) 221 (4.2) 4,469 (5.1) 79,312 (9.4) 
Marital Status 
 Married 341,807 (57.7) 31,103 (34.8) 2,020 (46.3) 16,462 (65.9) 4,698 (76) 14,163 (63.8) 5,788 (67.1) 2,871 (55.2) 48,443 (54.9) 467,355 (55.5) 
 Not married 226,616 (38.2) 53,863 (60.3) 1,882 (43.2) 7,759 (31.1) 1,285 (20.8) 7,249 (32.6) 2,419 (28.1) 2,141 (41.2) 35,775 (40.5) 338,989 (40.3) 
 Unknown 24,451 (4.1) 4,331 (4.9) 458 (10.5) 743 (3.0) 195 (3.2) 794 (3.6) 416 (4.8) 190 (3.7) 4,053 (4.6) 35,631 (4.2) 
Household Income 
 ≥$75k 201,048 (33.9) 17,707 (19.8) 1,735 (39.8) 15,191 (60.9) 3,534 (57.2) 11,002 (49.5) 4,374 (50.7) 201,048 (33.9) 17,707 (19.8) 1,735 (39.8) 
 $74,999–$55,000 270,115 (45.6) 41,620 (46.6) 1,405 (32.2) 9,333 (37.4) 2,192 (35.5) 10,559 (47.6) 3,736 (43.3) 270,115 (45.6) 41,620 (46.6) 1,405 (32.2) 
 >$55k 121,640 (20.5) 2,9969 (33.6) 1,219 (28) 439 (1.8) 452 (7.3) 645 (2.9) 513 (5.9) 121,640 (20.5) 29,969 (33.6) 1,219 (28) 
HRa Stat 
 ER+/PR+ 390,327 (65.8) 44,822 (50.2) 2,785 (63.9) 16,511 (66.1) 4,016 (65) 14,095 (63.5) 5,859 (67.9) 390,327 (65.8) 44,822 (50.2) 2,785 (63.9) 
 ER+/PR- 5,750 (1) 1,496 (1.7) 70 (1.6) 292 (1.2) 85 (1.4) 204 (0.9) 87 (1) 5,750 (1) 1,496 (1.7) 70 (1.6) 
 ER−/PR+ 65,540 (11.1) 10,911 (12.2) 455 (10.4) 2,694 (10.8) 611 (9.9) 2,550 (11.5) 891 (10.3) 65,540 (11.1) 10,911 (12.2) 455 (10.4) 
 ER−/PR− 87,793 (14.8) 24,729 (27.7) 805 (18.5) 3,997 (16) 1,154 (18.7) 3,850 (17.3) 1,256 (14.6) 87,793 (14.8) 24,729 (27.7) 805 (18.5) 
 Unknown 43,464 (7.3) 7,339 (8.2) 245 (5.6) 1,470 (5.9) 312 (5.1) 1,507 (6.8) 530 (6.1) 43,464 (7.3) 7,339 (8.2) 245 (5.6) 
Grade 
 Well differentiated 131,830 (22.2) 11,308 (12.7) 855 (19.6) 5,250 (21) 977 (15.8) 3,525 (15.9) 1,743 (20.2) 131,830 (22.2) 11,308 (12.7) 855 (19.6) 
 Moderately differentiated 245,198 (41.4) 30,130 (33.7) 1,709 (39.2) 10,413 (41.7) 2,389 (38.7) 9,244 (41.6) 3,705 (43) 245,198 (41.4) 30,130 (33.7) 1,709 (39.2) 
 Poorly differentiated 176,454 (29.8) 40,624 (45.5) 1,560 (35.8) 7,929 (31.8) 2,410 (39) 8,149 (36.7) 2,700 (31.3) 176,454 (29.8) 40,624 (45.5) 1,560 (35.8) 
 Unknown 39,392 (6.6) 7,235 (8.1) 236 (5.4) 1,372 (5.5) 402 (6.5) 1,288 (5.8) 475 (5.5) 39,392 (6.6) 7,235 (8.1) 236 (5.4) 
County residence 
 Metropolitan 519,449 (87.6) 82,212 (92.1) 2,656 (60.9) 24,174 (96.8) 6,120 (99.1) 21,718 (97.8) 8,508 (98.7) 519,449 (87.6) 82,212 (92.1) 2,656 (60.9) 
 Nonmetropolitan 73,354 (12.4) 7,084 (7.9) 751 (17.2) 789 (3.2) 58 (0.9) 488 (2.2) 115 (1.3) 73,354 (12.4) 7,084 (7.9) 751 (17.2) 
Region 
 Northeast 107,888 (18.2) 14,029 (15.7) 123 (2.8) 1,722 (6.9) 1,894 (30.7) 1,393 (6.3) 534 (6.2) 107,888 (18.2) 14,029 (15.7) 123 (2.8) 
 Midwest 63,377 (10.7) 10,396 (11.6) 141 (3.2) 232 (0.9) 285 (4.6) 171 (0.8) 299 (3.5) 63,377 (10.7) 10,396 (11.6) 141 (3.2) 
 South 133,374 (22.5) 41,254 (46.2) 161 (3.7) 599 (2.4) 635 (10.3) 429 (1.9) 534 (6.2) 133,374 (22.5) 41,254 (46.2) 161 (3.7) 
 West 288,235 (48.6) 23,618 (26.5) 3,935 (90.3) 22,411 (89.8) 3,364 (54.5) 20,213 (91) 7,256 (84.1) 288,235 (48.6) 23,618 (26.5) 3,935 (90.3) 
Diagnosis year 
 2000–2002 98,371 (16.6) 12,011 (13.5) 527 (12.1) 3,435 (13.8) 526 (8.5) 2,628 (11.8) 678 (7.9) 98,371 (16.6) 12,011 (13.5) 527 (12.1) 
 2003–2005 93,018 (15.7) 12,720 (14.2) 619 (14.2) 3,570 (14.3) 651 (10.5) 2,913 (13.1) 907 (10.5) 93,018 (15.7) 12,720 (14.2) 619 (14.2) 
 2006–2008 95,762 (16.2) 14,021 (15.7) 640 (14.7) 3,902 (15.6) 834 (13.5) 3,426 (15.4) 1,141 (13.2) 95,762 (16.2) 14,021 (15.7) 640 (14.7) 
 2009–2011 98,947 (16.7) 15,456 (17.3) 790 (18.1) 4,282 (17.2) 1,037 (16.8) 3,805 (17.1) 1,459 (16.9) 98,947 (16.7) 15,456 (17.3) 790 (18.1) 
 2012–2014 101,714 (17.2) 17,020 (19.1) 879 (20.2) 4,593 (18.4) 1,384 (22.4) 4,434 (20) 1,907 (22.1) 101,714 (17.2) 17,020 (19.1) 879 (20.2) 
 2015–2017 105,062 (17.7) 18,069 (20.2) 905 (20.8) 5,182 (20.8) 1,746 (28.3) 5,000 (22.5) 2,531 (29.4) 105,062 (17.7) 18,069 (20.2) 905 (20.8) 
Vital status at follow-up - n (%) 
 Alive 423,792 (71.5) 58,409 (65.4) 3,168 (72.8) 20,395 (81.7) 5,273 (85.4) 18,151 (81.7) 7,711 (89.4) 3,858 (74.2) 68,858 (78.0) 609,615 (72.4) 
 Due to breast cancer 71,666 (12.1) 17,899 (20.0) 588 (13.5) 2,211 (8.9) 569 (9.2) 2,394 (10.8) 514 (6.0) 694 (13.3) 11,211(12.7) 107,746 (12.8) 
 Due to other cause 97,416 (16.4) 12,989 (14.5) 604 (13.8) 2,358 (9.5) 336 (5.4) 1,661 (7.5) 4,398 (4.6) 650 (12.5) 8,202 (9.3) 124,614 (14.8) 

Note: Numbers in parentheses represent column percentages unless otherwise specified.

aNumber in parentheses represents row percent.

Percent of breast cancers detected at early stage

From 2000 to 2017, the percent of early stage breast cancer cases was generally highest among East Asian or Other Asian subgroup, and lowest among Black women (Fig. 1A; Table 2).

Figure 1.

Trends in breast cancer outcomes by race and ethnicity. A, Trends in percent of early-stage breast cancer diagnosis B, Trends in 5-year breast cancer survival of women diagnosed at early stage.

Figure 1.

Trends in breast cancer outcomes by race and ethnicity. A, Trends in percent of early-stage breast cancer diagnosis B, Trends in 5-year breast cancer survival of women diagnosed at early stage.

Close modal
Table 2.

Trends and disparities in early stage diagnosis and five-year breast cancer survival rates by race and ethnicity.

Percent of women diagnosed with early-stage breast cancera
Diagnosed in 2000 (%)Diagnosed in 2017 (%)AAPC (95% CI)
White 89.42 90.44 0.05 (0.01–0.09) 
Black 82.77 86.09 0.31 (0.23–0.36) 
American Indian 83.33 91.56 0.30 (0.10–0.55) 
East Asian 91.46 91.67 −0.01 (−0.110.10) 
South Asian 84.47 87.66 0.06 (−0.150.28) 
Southeast Asian 86.44 89.63 0.17 (0.07–0.26) 
Other Asian 91.75 93.57 −0.07 (−0.210.08) 
Pacific Islander 90.29 88.21 0.05 (−0.140.25) 
Hispanic 85.50 88.62 0.20 (0.14–0.26) 
 Disparity in 2000 Disparity in 2017 AAPC (95% CI) 
Range Difference (Absolute disparity) 8.98 7.47 3.45 (4.69 to −2.18) 
Range Ratio (Relative disparity) 1.11 1.09 0.37 (0.51 to −0.23) 
Five-year breast cancer survival among women diagnosed at early stagesb 
 Diagnosed in 2000 (%) Diagnosed in 2012 (%) AAPC (95% CI) 
Whites 94.64 95.66 0.10 (0.08–0.13) 
Black 89.24 92.00 0.34 (0.23–0.45) 
American Indian 89.80 95.34 0.29 (−0.070.66) 
East Asian 96.25 96.41 0.14 (0.04–0.25) 
South Asian 94.47 98.03 0.12 (−0.140.38) 
Southeast Asian 95.82 95.66 0.18 (0.02–0.33) 
Other Asian 96.75 98.04 0.06 (−0.07–0.19) 
Pacific Islander 93.44 94.77 0.15 (−0.20–0.50) 
Hispanic 92.54 94.69 0.20 (0.11–0.28) 
 Disparity in 2000 Disparity in 2012 AAPC (95% CI) 
Range Difference (Absolute disparity) 7.92 6.05 3.34 (−4.78 to −1.88) 
Range Ratio (Relative Disparity) 1.09 1.07 0.27 (−0.39 to −0.15) 
Percent of women diagnosed with early-stage breast cancera
Diagnosed in 2000 (%)Diagnosed in 2017 (%)AAPC (95% CI)
White 89.42 90.44 0.05 (0.01–0.09) 
Black 82.77 86.09 0.31 (0.23–0.36) 
American Indian 83.33 91.56 0.30 (0.10–0.55) 
East Asian 91.46 91.67 −0.01 (−0.110.10) 
South Asian 84.47 87.66 0.06 (−0.150.28) 
Southeast Asian 86.44 89.63 0.17 (0.07–0.26) 
Other Asian 91.75 93.57 −0.07 (−0.210.08) 
Pacific Islander 90.29 88.21 0.05 (−0.140.25) 
Hispanic 85.50 88.62 0.20 (0.14–0.26) 
 Disparity in 2000 Disparity in 2017 AAPC (95% CI) 
Range Difference (Absolute disparity) 8.98 7.47 3.45 (4.69 to −2.18) 
Range Ratio (Relative disparity) 1.11 1.09 0.37 (0.51 to −0.23) 
Five-year breast cancer survival among women diagnosed at early stagesb 
 Diagnosed in 2000 (%) Diagnosed in 2012 (%) AAPC (95% CI) 
Whites 94.64 95.66 0.10 (0.08–0.13) 
Black 89.24 92.00 0.34 (0.23–0.45) 
American Indian 89.80 95.34 0.29 (−0.070.66) 
East Asian 96.25 96.41 0.14 (0.04–0.25) 
South Asian 94.47 98.03 0.12 (−0.140.38) 
Southeast Asian 95.82 95.66 0.18 (0.02–0.33) 
Other Asian 96.75 98.04 0.06 (−0.07–0.19) 
Pacific Islander 93.44 94.77 0.15 (−0.20–0.50) 
Hispanic 92.54 94.69 0.20 (0.11–0.28) 
 Disparity in 2000 Disparity in 2012 AAPC (95% CI) 
Range Difference (Absolute disparity) 7.92 6.05 3.34 (−4.78 to −1.88) 
Range Ratio (Relative Disparity) 1.09 1.07 0.27 (−0.39 to −0.15) 

Note: The range difference was defined as the difference between highest and lowest values for each race/ethnic group. The range ratio was derived by dividing the highest rate value by the lowest rate value. Bolded values represent significance (P < 0.05).

aEarly-stage breast cancer includes AJCC stage I to IIIA.

bEstimates of 5-year survival reflect survival experience of patients diagnosed at early stage with 5 years of follow-up since diagnosis (that is, patients diagnosed between 2000 and 2012, n = 514,668).

The greatest increase in the percentage of early stage breast cancer diagnosis occurred among Black (AAPC, 0.31; 95% CI, 0.23–0.36) and AIAN (AAPC, 0.30; 95% CI, 0.10–0.55) women, followed by Hispanic (AAPC, 0.20; 95% CI, 0.14–0.26), Southeast Asian (AAPC, 0.17; 95% CI, 0.07–0.26), and White women (AAPC, 0.05; 95% CI, 0.01–0.09). Percentage of early stage diagnosis remained relatively stable over time among the remaining four race/ethnic groups (East Asian, South Asian, “Other” Asian, and Pacific Islander women (Table 2). Absolute race/ethnic disparities in early detection decreased significantly over time, declining by 3.45% per year (95% CI, 4.69–−2.18). Relative disparities in the percent of early-stage breast cancer cases also decreased over time, by 0.37% per year (95% CI, −0.51 to −0.23).

Five-year breast cancer survival

Five-year breast cancer survival was highest among women in East Asian or “Other” Asian subgroups, and consistently lowest among Black women (Fig. 1B; Table 2). Black women experienced the largest gain in survival over the study period (AAPC, 0.34; 95% CI, 0.23–0.45), followed by Hispanic (AAPC, 0.20; 95% CI, 0.11–0.28), Southeast Asian (AAPC, 0.18; 95% CI, 0.02–0.33), East Asian (AAPC, 0.14; 95% CI, 0.04–0.25), and White women (AAPC, 0.10; 95% CI, 0.08–0.13). Improvements in survival for the remaining racial and ethnic groups did not reach statistical significance. Absolute and relative disparities in 5-year breast cancer survival also narrowed over time, declining by 3.37% (95% CI, −4.78 to −1.88) and 0.27% (95% CI, −0.39 to −0.15) per year, respectively.

Predictors of early-stage breast cancer diagnosis

In multivariable logistic regression analysis, Black (OR, 0.66; P < 0.001), Pacific Islander (OR, 0.67; P < 0.001), American Indian (OR, 0.80; P < 0.001), Hispanic (OR, 0.79; P < 0.001), South Asian (OR, 0.73; P < 0.001), and Southeast Asian (OR, 0.83; P < 0.001) women were less likely to be diagnosed with breast cancer at early stage compared with White women (Table 3). Other factors such as age <35 years, not being married, low socioeconomic status, rural residence, and living in regions of the Midwest or Northeast also predicted lower likelihood of being diagnosed at early stage.

Table 3.

Univariate and multivariable logistic regression analysis of risk factors associated with early-stage breast cancer diagnosis (N = 841,975).

UnivariateAdjusted multivariable
OR (95% CI)POR (95% CI)P
Race/Ethnicity 
 White 1.00 (Ref)  1.00 (Ref)  
 Black 0.59 (0.58–0.60) <0.001 0.66 (0.65–0.67) <0.001 
 American Indian 0.82 (0.75–0.90) <0.001 0.80 (0.72–0.88) <0.001 
 East Asian 1.34 (1.27–1.40) <0.001 1.19 (1.14–1.25) <0.001 
 South Asian 0.83 (0.77–0.90) <0.001 0.73 (0.68–0.79) <0.001 
 Southeast Asian 0.93 (0.89–0.97) 0.002 0.83 (0.80–0.87) <0.001 
 Other Asian 1.50 (1.38–1.63) <0.001 1.32 (1.22–1.44) <0.001 
 Pacific Islander 0.73 (0.68–0.80) <0.001 0.67 (0.62–0.72) <0.001 
 Hispanic 0.81 (0.80–0.83) <0.001 0.79 (0.77–0.81) <0.001 
Age at diagnosis 
 20–34 0.63 (0.60–0.66) <0.001 0.68 (0.65–0.71) <0.001 
 35–49 1.02 (0.99–1.04) 0.084 1.02 (1.00–1.04) 0.017 
 50–64 1.00 (Ref)  1.00 (Ref)  
 65–79 1.08 (1.06–1.10) <0.001 1.11 (1.09–1.13) <0.001 
 ≥80 0.67 (0.65–0.68) <0.001 0.74 (0.73–0.76) <0.001 
Marital Status 
 Married 1.00 (Ref)  1.00 (Ref)  
 Not married 0.62 (0.61–0.63) <0.001 0.68 (0.67–0.69) <0.001 
 Missing/unknown 0.70 (0.68–0.72) <0.001 0.74 (0.72–0.76) <0.001 
Household Income 
 ≥$75k 1.00 (Ref)  1.00 (Ref)  
 $74,999–55,000 0.89 (0.87–0.90) <0.001 0.91 (0.90–0.93) <0.001 
 <$55k 0.77 (0.76–0.79) <0.001 0.82 (0.80–0.84) <0.001 
 Missing/Unknown 0.60 (0.34–1.07) 0.083 0.54 (0.29–1.01) <0.001 
County Residence 
 Metropolitan 1.00 (Ref)  1.00 (Ref)  
 Nonmetro 0.90 (0.88–0.92) <0.001 0.96 (0.93–0.98) <0.001 
 Missing/Unknown 1.05 (0.86–1.28) <0.001 1.15 (0.91–1.46) 0.238 
Region 
 Northeast 0.90 (0.88–0.92) <0.001 0.88 (0.86–0.90) <0.001 
 Midwest 0.90 (0.88–0.92) <0.001 0.95 (0.93–0.98) <0.001 
 South 0.85 (0.83–0.87) <0.001 0.98 (0.96–0.99) 0.021 
 West 1.00 (Ref)  1.00 (Ref)  
Diagnosis year 
 2000–2002 1.00 (Ref)  1.00 (Ref)  
 2003–2005 1.01 (0.98–1.02) 0.897 1.01 (0.99–1.04) <0.001 
 2006–2008 1.01 (0.98–1.03) 0.610 1.03 (1.01–1.06) <0.001 
 2009–2011 1.03 (1.01–1.06) 0.014 1.07 (1.04–1.10) <0.001 
 2012–2014 1.07 (1.05–1.10) <0.001 1.12 (1.09–1.15) <0.001 
 2015–2017 1.11 (1.09–1.14) <0.001 1.15 (1.12–1.18) <0.001 
UnivariateAdjusted multivariable
OR (95% CI)POR (95% CI)P
Race/Ethnicity 
 White 1.00 (Ref)  1.00 (Ref)  
 Black 0.59 (0.58–0.60) <0.001 0.66 (0.65–0.67) <0.001 
 American Indian 0.82 (0.75–0.90) <0.001 0.80 (0.72–0.88) <0.001 
 East Asian 1.34 (1.27–1.40) <0.001 1.19 (1.14–1.25) <0.001 
 South Asian 0.83 (0.77–0.90) <0.001 0.73 (0.68–0.79) <0.001 
 Southeast Asian 0.93 (0.89–0.97) 0.002 0.83 (0.80–0.87) <0.001 
 Other Asian 1.50 (1.38–1.63) <0.001 1.32 (1.22–1.44) <0.001 
 Pacific Islander 0.73 (0.68–0.80) <0.001 0.67 (0.62–0.72) <0.001 
 Hispanic 0.81 (0.80–0.83) <0.001 0.79 (0.77–0.81) <0.001 
Age at diagnosis 
 20–34 0.63 (0.60–0.66) <0.001 0.68 (0.65–0.71) <0.001 
 35–49 1.02 (0.99–1.04) 0.084 1.02 (1.00–1.04) 0.017 
 50–64 1.00 (Ref)  1.00 (Ref)  
 65–79 1.08 (1.06–1.10) <0.001 1.11 (1.09–1.13) <0.001 
 ≥80 0.67 (0.65–0.68) <0.001 0.74 (0.73–0.76) <0.001 
Marital Status 
 Married 1.00 (Ref)  1.00 (Ref)  
 Not married 0.62 (0.61–0.63) <0.001 0.68 (0.67–0.69) <0.001 
 Missing/unknown 0.70 (0.68–0.72) <0.001 0.74 (0.72–0.76) <0.001 
Household Income 
 ≥$75k 1.00 (Ref)  1.00 (Ref)  
 $74,999–55,000 0.89 (0.87–0.90) <0.001 0.91 (0.90–0.93) <0.001 
 <$55k 0.77 (0.76–0.79) <0.001 0.82 (0.80–0.84) <0.001 
 Missing/Unknown 0.60 (0.34–1.07) 0.083 0.54 (0.29–1.01) <0.001 
County Residence 
 Metropolitan 1.00 (Ref)  1.00 (Ref)  
 Nonmetro 0.90 (0.88–0.92) <0.001 0.96 (0.93–0.98) <0.001 
 Missing/Unknown 1.05 (0.86–1.28) <0.001 1.15 (0.91–1.46) 0.238 
Region 
 Northeast 0.90 (0.88–0.92) <0.001 0.88 (0.86–0.90) <0.001 
 Midwest 0.90 (0.88–0.92) <0.001 0.95 (0.93–0.98) <0.001 
 South 0.85 (0.83–0.87) <0.001 0.98 (0.96–0.99) 0.021 
 West 1.00 (Ref)  1.00 (Ref)  
Diagnosis year 
 2000–2002 1.00 (Ref)  1.00 (Ref)  
 2003–2005 1.01 (0.98–1.02) 0.897 1.01 (0.99–1.04) <0.001 
 2006–2008 1.01 (0.98–1.03) 0.610 1.03 (1.01–1.06) <0.001 
 2009–2011 1.03 (1.01–1.06) 0.014 1.07 (1.04–1.10) <0.001 
 2012–2014 1.07 (1.05–1.10) <0.001 1.12 (1.09–1.15) <0.001 
 2015–2017 1.11 (1.09–1.14) <0.001 1.15 (1.12–1.18) <0.001 

Prognostic factors for breast cancer–specific survival

In analysis of racial/ethnic differences in breast cancer survival by stage, we found evidence for significant effect modification by stage (Supplementary Fig. S2; Supplementary Table S1). Thus, for ease of interpretation and to highlight the effects among stages at diagnosis for which clinical intervention generally results in better long-term outcomes, we present findings from multivariable Cox proportional hazards regressions for groups diagnosed at early (i.e., I–IIIA) and late stage (i.e., IIIB–IV) separately (Table 4).

Table 4.

Multivariable Cox regression analysis of race/ethnicity and breast cancer–specific survival, stratified by stage at diagnosis.

Early stage (I–IIIA)Late stage (IIIB–IV)
N = 749,029N = 92,946
HR (95% CI)PHR (95% CI)P
Race/Ethnicity 
 White 1.00 (Ref)  1.00 (Ref)  
 Black 1.39 (1.36–1.43) <0.001 1.18 (1.15–1.21) <0.001 
 American Indian 1.14 (1.01–1.29) 0.042 0.90 (0.79–1.03) 0.130 
 East Asian 0.81 (0.77–0.86) <0.001 0.85 (0.79–0.90) <0.001 
 South Asian 0.85 (0.76–0.96) 0.009 0.82 (0.73–0.92) 0.001 
 Southeast Asian 0.92 (0.87–0.97) 0.004 0.89 (0.84–0.95) <0.001 
 Other Asian 0.60 (0.54–0.68) <0.001 0.72 (0.63–0.82) <0.001 
 Pacific Islander 1.22 (1.10–1.36) <0.001 1.10 (0.99–1.23) 0.067 
 Hispanic 1.09 (1.06–1.13) <0.001 0.95 (0.92–0.98) <0.001 
Age at diagnosis 
 20–34 1.4 (1.34–1.47) 0.000 0.91 (0.87–0.96) 0.001 
 35–49 1.08 (1.05–1.10) 0.000 0.89 (0.87–0.91) <0.001 
 50–64 1.00 (Ref)  1.00 (Ref)  
 65–79 1.25 (1.22–1.27) <0.001 1.15 (1.12–1.18) <0.001 
 80+ 2.44 (2.37–2.51) <0.001 1.57 (1.52–1.61) <0.001 
Marital status 
 Married 1.00 (Ref)  1.00 (Ref)  
 Not married 1.26 (1.24–1.28) <0.001 1.17 (1.15–1.20) <0.001 
 Missing/Unknown 1.14 (1.09–1.19) <0.001 1.03 (0.99–1.08) 0.137 
Household income 
 ≥$75k 1.00 (Ref)  1.00 (Ref)  
 $74,999–55,000 1.13 (1.11–1.15) <0.001 1.08 (1.05–1.10) <0.001 
 <$55k 1.19 (1.16–1.23) <0.001 1.12 (1.09–1.16) <0.001 
 Missing/Unknown 1.20 (0.59–2.43) 0.612 0.49 (0.24–1.01) 0.053 
Receptor status 
 ER+/PR+ 1.00 (Ref)  1.00 (Ref)  
 ER+/PR 1.5 (1.40–1.59) <0.001 1.61 (1.50–1.72) <0.001 
 ER/PR+ 1.44 (1.41–1.48) <0.001 1.3 (1.27–1.34) <0.001 
 ER/PR 1.60 (1.57–1.63) <0.001 1.64 (1.60–1.68) <0.001 
 Unknown/Borderline 1.20 (1.16–1.23) <0.001 1.58 (1.54–1.63) <0.001 
Grade 
 Well differentiated 1.00 (Ref)  1.00 (Ref)  
 Moderately differentiated 2.38 (2.30–2.46) <0.001 1.24 (1.18–1.30) <0.001 
 Poorly differentiated 4.31 (4.17–4.46) <0.001 1.59 (1.52–1.67) <0.001 
 Missing/Unknown 2.42 (2.31–2.53) <0.001 1.45 (1.38–1.53) <0.001 
Radiation 
 Yes 1.00 (Ref)  1.00 (Ref)  
 No/unknown 1.15 (1.13–1.17) <0.001 1.21 (1.19–1.24) <0.001 
Surgery 
 Yes 1.00 (Ref)  1.00 (Ref)  
 No/unknown 4.37 (4.23–4.52) <0.001\ 2.85 (2.79–2.91) <0.001 
County residence 
 Metropolitan 1.00 (Ref)  1.00 (Ref)  
 Nonmetropolitan 1.03 (1.00–1.07) 0.025 1.02 (0.98–1.05) 0.318 
 Missing/Unknown 1.13 (0.87–1.48) 0.356 1.6 (1.18–2.16) 0.002 
Region 
 Northeast 0.87 (0.85–0.90) <0.001 0.92 (0.89–0.94) 0.000 
 Midwest 0.95 (0.92–0.98) <0.001 1.05 (1.02–1.09) 0.002 
 South 1.01 (0.98–1.03) 0.739 1.01 (0.98–1.03) 0.558 
 West 1.00 (Ref)  1.00 (Ref)  
Diagnosis year 
 2000–2002 1.00 (Ref)  1.00 (Ref)  
 2003–2005 0.91 (0.89–0.93) <0.001 0.93 (0.90–0.96) <0.001 
 2006–2008 0.83 (0.81–0.85) <0.001 0.87 (0.84–0.89) <0.001 
 2009–2011 0.80 (0.78–0.82) <0.001 0.86 (0.83–0.89) <0.001 
 2012–2014 0.76 (0.74–0.79) <0.001 0.80 (0.78–0.83) <0.001 
 2015–2017 0.76 (0.73–0.79) <0.001 0.75 (0.72–0.78) <0.001 
Early stage (I–IIIA)Late stage (IIIB–IV)
N = 749,029N = 92,946
HR (95% CI)PHR (95% CI)P
Race/Ethnicity 
 White 1.00 (Ref)  1.00 (Ref)  
 Black 1.39 (1.36–1.43) <0.001 1.18 (1.15–1.21) <0.001 
 American Indian 1.14 (1.01–1.29) 0.042 0.90 (0.79–1.03) 0.130 
 East Asian 0.81 (0.77–0.86) <0.001 0.85 (0.79–0.90) <0.001 
 South Asian 0.85 (0.76–0.96) 0.009 0.82 (0.73–0.92) 0.001 
 Southeast Asian 0.92 (0.87–0.97) 0.004 0.89 (0.84–0.95) <0.001 
 Other Asian 0.60 (0.54–0.68) <0.001 0.72 (0.63–0.82) <0.001 
 Pacific Islander 1.22 (1.10–1.36) <0.001 1.10 (0.99–1.23) 0.067 
 Hispanic 1.09 (1.06–1.13) <0.001 0.95 (0.92–0.98) <0.001 
Age at diagnosis 
 20–34 1.4 (1.34–1.47) 0.000 0.91 (0.87–0.96) 0.001 
 35–49 1.08 (1.05–1.10) 0.000 0.89 (0.87–0.91) <0.001 
 50–64 1.00 (Ref)  1.00 (Ref)  
 65–79 1.25 (1.22–1.27) <0.001 1.15 (1.12–1.18) <0.001 
 80+ 2.44 (2.37–2.51) <0.001 1.57 (1.52–1.61) <0.001 
Marital status 
 Married 1.00 (Ref)  1.00 (Ref)  
 Not married 1.26 (1.24–1.28) <0.001 1.17 (1.15–1.20) <0.001 
 Missing/Unknown 1.14 (1.09–1.19) <0.001 1.03 (0.99–1.08) 0.137 
Household income 
 ≥$75k 1.00 (Ref)  1.00 (Ref)  
 $74,999–55,000 1.13 (1.11–1.15) <0.001 1.08 (1.05–1.10) <0.001 
 <$55k 1.19 (1.16–1.23) <0.001 1.12 (1.09–1.16) <0.001 
 Missing/Unknown 1.20 (0.59–2.43) 0.612 0.49 (0.24–1.01) 0.053 
Receptor status 
 ER+/PR+ 1.00 (Ref)  1.00 (Ref)  
 ER+/PR 1.5 (1.40–1.59) <0.001 1.61 (1.50–1.72) <0.001 
 ER/PR+ 1.44 (1.41–1.48) <0.001 1.3 (1.27–1.34) <0.001 
 ER/PR 1.60 (1.57–1.63) <0.001 1.64 (1.60–1.68) <0.001 
 Unknown/Borderline 1.20 (1.16–1.23) <0.001 1.58 (1.54–1.63) <0.001 
Grade 
 Well differentiated 1.00 (Ref)  1.00 (Ref)  
 Moderately differentiated 2.38 (2.30–2.46) <0.001 1.24 (1.18–1.30) <0.001 
 Poorly differentiated 4.31 (4.17–4.46) <0.001 1.59 (1.52–1.67) <0.001 
 Missing/Unknown 2.42 (2.31–2.53) <0.001 1.45 (1.38–1.53) <0.001 
Radiation 
 Yes 1.00 (Ref)  1.00 (Ref)  
 No/unknown 1.15 (1.13–1.17) <0.001 1.21 (1.19–1.24) <0.001 
Surgery 
 Yes 1.00 (Ref)  1.00 (Ref)  
 No/unknown 4.37 (4.23–4.52) <0.001\ 2.85 (2.79–2.91) <0.001 
County residence 
 Metropolitan 1.00 (Ref)  1.00 (Ref)  
 Nonmetropolitan 1.03 (1.00–1.07) 0.025 1.02 (0.98–1.05) 0.318 
 Missing/Unknown 1.13 (0.87–1.48) 0.356 1.6 (1.18–2.16) 0.002 
Region 
 Northeast 0.87 (0.85–0.90) <0.001 0.92 (0.89–0.94) 0.000 
 Midwest 0.95 (0.92–0.98) <0.001 1.05 (1.02–1.09) 0.002 
 South 1.01 (0.98–1.03) 0.739 1.01 (0.98–1.03) 0.558 
 West 1.00 (Ref)  1.00 (Ref)  
Diagnosis year 
 2000–2002 1.00 (Ref)  1.00 (Ref)  
 2003–2005 0.91 (0.89–0.93) <0.001 0.93 (0.90–0.96) <0.001 
 2006–2008 0.83 (0.81–0.85) <0.001 0.87 (0.84–0.89) <0.001 
 2009–2011 0.80 (0.78–0.82) <0.001 0.86 (0.83–0.89) <0.001 
 2012–2014 0.76 (0.74–0.79) <0.001 0.80 (0.78–0.83) <0.001 
 2015–2017 0.76 (0.73–0.79) <0.001 0.75 (0.72–0.78) <0.001 

Among women diagnosed at early stage, Black, Pacific Islander, AIAN, and Hispanic women were 39%, 22%, 14%, and 9% (respectively) more likely than Whites to die from breast cancer (P < 0.001 for all), whereas Asian women (of all subgroups) had lower risk of death compared with White (Table 4) women. Among women diagnosed at late stage, Black women were 18% more likely than White women to die from breast cancer (P < 0.001).

Breast cancer subtype analysis

Among those diagnosed at early stage, Black women had worse survival for HR+/HER2 tumors compared with White women, as did Hispanic women (Table 5). Black women diagnosed at early stage also showed worse survival for HR+/HER2+ (HR, 1.21; P = 0.023), HR/HER2 (HR, 1.28; P < 0.001) and HR/HER2+ (HR, 1.40; P < 0.001) subtypes. East Asian women diagnosed with HR+/HER2 breast cancer at early stage showed better survival (HR, 0.73; P < 0.001) compared with white women diagnosed with the same stage and subtype. Among women diagnosed at late stage (Table 5), Black women had significantly higher risk of death compared with White women for HR+/HER2 (HR, 1.19; P < 0.001), HR+/HER2+ (HR, 1.34; P < 0.001), and HR/HER2+ (HR, 1.29; P = 0.001) subtypes. Breast cancer survival for the other racial and ethnic groups was similar to that of White women for the HR+/HER2+, HR/HER2, and HR/HER2+ subtypes, regardless of diagnosis stage.

Table 5.

Breast cancer–specific survival by race/ethnicity, stage, and breast cancer subtypes based on multivariable Cox regression model.a

HR+/HER2HR+/HER2+HR/HER2HR/HER2+
HR (95% CI)PHR (95% CI)PHR (95% CI)PHR (95% CI)P
Early stage 
 White 1.00 (Ref)  1.00 (Ref)  1.00 (Ref)  1.00 (Ref)  
 Black 1.41 (1.32–1.50) <0.001 1.21 (1.03–1.40) 0.023 1.28 (1.18–1.38) <0.001 1.40 (1.18–1.67) <0.001 
 American Indian 1.14 (0.84–1.56) 0.393 0.94 (0.42–2.11) 0.889 0.98 (0.62–1.54) 0.918 0.37 (0.09–1.47) 0.157 
 East Asian 0.73 (0.63–0.86) <0.001 0.74 (0.52–1.07) 0.110 0.83 (0.67–1.02) 0.082 0.88 (0.60–1.30) 0.519 
 South Asian 0.81 (0.60–1.08) 0.153 1.09 (0.60–1.97) 0.788 0.79 (0.54–1.16) 0.225 1.39 (0.74–2.62) 0.303 
 Southeast Asian 0.91 (0.79–1.05) 0.202 0.99 (0.73–1.35) 0.970 0.70 (0.53–0.91) 0.009 1.01 (0.71–1.44) 0.946 
 Other Asian 0.56 (0.42–0.73) <0.001 0.48 (0.51–0.67) 0.004 0.55 (0.36–0.85) 0.007 0.56 (0.26–1.19) 0.130 
 Pacific Islander 1.20 (0.93–1.54) 0.154 1.42 (0.80–2.53) 0.227 1.51 (0.99–2.30) 0.057 1.55 (0.80–3.02) 0.195 
 Hispanic 1.08 (1.00–1.16) 0.038 1.20 (1.02–1.42) 0.030 1.03 (0.93–1.13) 0.623 1.22 (1.00–1.49) 0.050 
Total sample (N255,588  36,002  36,327  14,185  
Late stage 
 White 1.00 (Ref)  1.00 (Ref)  1.00 (Ref)  1.00 (Ref)  
 Black 1.19 (1.12–1.27) <0.001 1.34 (1.19–1.51) <0.001 1.07 (0.98–1.16) 0.137 1.29 (1.11–1.50) 0.001 
 American Indian 0.78 (0.57–1.08) 0.146 0.64 (0.32–1.29) 0.210 0.95 (0.59–1.54) 0.846 1.01 (0.50–2.02) 0.995 
 East Asian 0.88 (0.76–1.03) 0.117 1.07 (0.80–1.45) 0.638 0.96 (0.76–1.20) 0.698 0.89 (0.64–1.23) 0.470 
 South Asian 0.87 (0.70–1.08) 0.213 0.64 (0.39–1.06) 0.082 0.87 (0.58–1.29) 0.478 1.30 (0.79–2.14) 0.302 
 Southeast Asian 0.90 (0.79–1.03) 0.143 1.02 (0.78–1.34) 0.857 0.82 (0.65–1.05) 0.112 1.05 (0.78–1.41) 0.737 
 Other Asian 0.69 (0.53–0.91) 0.008 0.69 (0.39–1.22) 0.202 0.65 (0.35–1.21) 0.175 1.02 (0.63–1.63) 0.947 
 Pacific Islander 1.05 (0.85–1.29) 0.649 1.01 (0.64–1.59) 0.978 0.94 (0.62–1.44) 0.791 0.96 (0.53–1.76) 0.901 
 Hispanic 0.95 (0.88–1.02) 0.128 1.04 (0.91–1.20) 0.539 0.91 (0.82–1.02) 0.099 1.14 (0.97–1.34) 0.116 
Total sample (N22,444  6,222  6,009  3,706  
HR+/HER2HR+/HER2+HR/HER2HR/HER2+
HR (95% CI)PHR (95% CI)PHR (95% CI)PHR (95% CI)P
Early stage 
 White 1.00 (Ref)  1.00 (Ref)  1.00 (Ref)  1.00 (Ref)  
 Black 1.41 (1.32–1.50) <0.001 1.21 (1.03–1.40) 0.023 1.28 (1.18–1.38) <0.001 1.40 (1.18–1.67) <0.001 
 American Indian 1.14 (0.84–1.56) 0.393 0.94 (0.42–2.11) 0.889 0.98 (0.62–1.54) 0.918 0.37 (0.09–1.47) 0.157 
 East Asian 0.73 (0.63–0.86) <0.001 0.74 (0.52–1.07) 0.110 0.83 (0.67–1.02) 0.082 0.88 (0.60–1.30) 0.519 
 South Asian 0.81 (0.60–1.08) 0.153 1.09 (0.60–1.97) 0.788 0.79 (0.54–1.16) 0.225 1.39 (0.74–2.62) 0.303 
 Southeast Asian 0.91 (0.79–1.05) 0.202 0.99 (0.73–1.35) 0.970 0.70 (0.53–0.91) 0.009 1.01 (0.71–1.44) 0.946 
 Other Asian 0.56 (0.42–0.73) <0.001 0.48 (0.51–0.67) 0.004 0.55 (0.36–0.85) 0.007 0.56 (0.26–1.19) 0.130 
 Pacific Islander 1.20 (0.93–1.54) 0.154 1.42 (0.80–2.53) 0.227 1.51 (0.99–2.30) 0.057 1.55 (0.80–3.02) 0.195 
 Hispanic 1.08 (1.00–1.16) 0.038 1.20 (1.02–1.42) 0.030 1.03 (0.93–1.13) 0.623 1.22 (1.00–1.49) 0.050 
Total sample (N255,588  36,002  36,327  14,185  
Late stage 
 White 1.00 (Ref)  1.00 (Ref)  1.00 (Ref)  1.00 (Ref)  
 Black 1.19 (1.12–1.27) <0.001 1.34 (1.19–1.51) <0.001 1.07 (0.98–1.16) 0.137 1.29 (1.11–1.50) 0.001 
 American Indian 0.78 (0.57–1.08) 0.146 0.64 (0.32–1.29) 0.210 0.95 (0.59–1.54) 0.846 1.01 (0.50–2.02) 0.995 
 East Asian 0.88 (0.76–1.03) 0.117 1.07 (0.80–1.45) 0.638 0.96 (0.76–1.20) 0.698 0.89 (0.64–1.23) 0.470 
 South Asian 0.87 (0.70–1.08) 0.213 0.64 (0.39–1.06) 0.082 0.87 (0.58–1.29) 0.478 1.30 (0.79–2.14) 0.302 
 Southeast Asian 0.90 (0.79–1.03) 0.143 1.02 (0.78–1.34) 0.857 0.82 (0.65–1.05) 0.112 1.05 (0.78–1.41) 0.737 
 Other Asian 0.69 (0.53–0.91) 0.008 0.69 (0.39–1.22) 0.202 0.65 (0.35–1.21) 0.175 1.02 (0.63–1.63) 0.947 
 Pacific Islander 1.05 (0.85–1.29) 0.649 1.01 (0.64–1.59) 0.978 0.94 (0.62–1.44) 0.791 0.96 (0.53–1.76) 0.901 
 Hispanic 0.95 (0.88–1.02) 0.128 1.04 (0.91–1.20) 0.539 0.91 (0.82–1.02) 0.099 1.14 (0.97–1.34) 0.116 
Total sample (N22,444  6,222  6,009  3,706  

aMultivariable Cox regression model was adjusted for tumor grade, diagnosis age, marital status, household income, county residence, region, radiation treatment, surgery, and diagnosis year; Subgroup analyses limited to patients diagnosed after 2009 (n = 380,483). Results that were statistically significant at P < 0.006 (Bonferroni corrected).

In our analysis of breast cancer outcomes by racial and ethnic groups, we found persistent disparities in both the likelihood of being diagnosed at early stage and also in survival among women diagnosed at early stages. These findings indirectly suggest that efforts in early detection may have had varying impact by race and ethnicity. For example, the percentage of early stage breast cancer from 2000 to 2017 increased among women of all race/ethnicities (except for East Asian and Other Asian women), with Black women showing the largest improvement in early detection and survival after early stage diagnosis. Progress made over time towards reducing inequalities in breast cancer outcomes in the United States was indicated by shrinking measures of both absolute and relative disparities for early stage diagnosis and survival among those diagnosed at early stage, a finding consistent with prior reports (8, 10, 22). However, multivariable logistic regression analyses revealed that, after adjusting for other factors, Black, Hispanic, AIAN, Pacific Islander, South, and Southeast Asian women were still diagnosed less often with early stage breast cancer than White counterparts.

Survival analyses further revealed that even when diagnosed at earlier stages, Black, Hispanic, AIAN, and Pacific Islander women had poorer survival compared to White women diagnosed at the same stage. Notably, breast cancer survival for Black women remained worse than that of White women, regardless of stage at diagnosis. Adjusting for selected sociodemographic and clinical characteristics reduced but did not eliminate these disparities, underscoring the importance of further inquiry about disparities within those diagnosed at the same stage.

When we further characterized differences in survival by molecular subtype, survival for Hispanic, AIAN, Pacific Islander, and Asian women (all subgroups) was equal to or better than that of White women, regardless of breast cancer subtype or stage diagnosis groups. In contrast, survival disparities for Black women persisted across all molecular subtypes and disease stages, with the exception of late stage triple negative breast cancer, which is known to have particularly poor prognosis. This may indicate, at least for triple negative subtype, the intractability of biologically aggressive disease already metastatic at diagnosis regardless of race and ethnicity. The greatest disparities in survival between Black and White women, however, occurred among early stage HR+/HER2− breast cancer, a tumor type considered least aggressive with a range of effective treatment options. Whether the survival disparity for Black women results more from biological or non-biological factors is unknown, but differences in tumor behavior and morphology across racial/ethnic groups are thought to contribute, in part, to differences in both stage at diagnosis and survival (23–25), complicating the assessment racial and ethnic disparities. Still, our findings consistently showed worse survival among Black women, even when diagnosed at early stages with the least aggressive subtype, represent missed opportunities across the entire cancer care continuum.

Our work also highlights the importance of examining trends in breast cancer outcomes among Asian subgroups and Pacific Islanders, separately. For example, when examined separately, Pacific Islanders, South Asian, and Southeast Asian groups were less likely to be diagnosed with at early stage, while women in East Asian or Other Asian subgroup were more likely to be diagnosed at early stage, compared to White counterparts. Women of different Asian subgroups also varied significantly with regard to breast cancer mortality risk by tumor subtype. For example, East Asian women had 27% lower risk of breast cancer mortality from early-stage HR+/HER2 subtype compared to White women, while South Asian and Southeast Asian women had no such survival advantage over White women with early stage HR+/HER2 breast cancer. No significant differences in breast cancer survival between East Asian and White women were observed for other stage and subtype groupings. Although the reason for improved survival in East Asian women is unclear, it is possible that dietary factors are relevant factors, such as the consumption of green tea, isoflavone, or soy products. Various studies have found that green tea and soy isoflavone consumption among Asian women reduced their risks of breast cancer and recurrence, although more research is needed (26, 27).

Racial and ethnic disparities in early diagnosis of breast cancer likely result from a variety of factors. For example, prior mammography screening data found that Black, Hispanic, Asian/Pacific Islander, and AIAN women were less likely to have had routine screening (22). Other previously reported barriers to mammography uptake among minority women relate to healthcare access, such as not having a primary care provider, not having physician-recommend mammography, and lack of insurance (3, 23–25, 28, 29). Greater scrutiny of these and other factors that influence early breast cancer detection, availability of culturally appropriate education, outreach, accessibility, and use of screening services to communities of color, increased awareness of unintended bias, and evaluation of structural racism and other contributors, is needed to improve likelihood of early detection for all groups.

Report of survival differences between groups that remain significant after accounting for disease stage and other tumor characteristics, from us and others (5, 7, 11, 26), strongly suggest that social determinants of health (i.e., nonbiologic factors) contribute importantly to the disproportionate burden of breast cancer for some groups. Nonbiologic factors that influence breast cancer disparities include socioeconomic factors, access to mammography screenings, delivery of treatment modalities, healthcare system mistrust, and health literacy (24, 25, 29). Research has shown that Black women with breast cancer are more likely to experience treatment delays, less likely to receive guideline concordant treatment, and experience lower adherence to treatment plans than their White counterparts (29–32).

Nevertheless, the dynamic mechanisms underlying racial and ethnic disparities in breast cancer remain poorly understood. A growing body of research suggests that health disparities in cancer comprise a complex problem influenced by multiple and multilevel factors (33, 34). These factors span from individual level characteristics (e.g., genetic susceptibility, biologic markers of disease), to social (e.g., built environments, socioeconomic status, and social networks) and systemic (e.g., policies that affect the availability, receipt of, and quality of cancer care) factors (33). Because multilevel interventions may be able to address multiple determinants within these complex contexts, they may be uniquely suited to reducing breast cancer disparities in racial and ethnic minorities. As such, renewed efforts are needed to design and implement interventions that target the multilevel causes of persistent disparities in breast cancer among racial and ethnic groups.

Our study has several notable strengths. Many prior studies of trends in breast cancer by stage at diagnosis have used broad groupings of stages (i.e., I, II, III, IV) without meaningful use of substage categories (e.g., stage IIIA vs. IIIB; refs. 8, 10, 11). Therefore, our “early stage" definition not only provides greater precision than in prior studies, but also revealed persistent disparities in breast cancer survival among women diagnosed with stages of disease for which early detection is intended to result in better survival. We also report outcomes across nine race/ethnic groups, extending on prior studies using aggregated race/ethnic categories. We assessed breast cancer specific cause of death stratified by HR and HER2 status which added detailed and extensive information to depict the racial and ethnic disparities in breast cancer.

Despite these strengths, there are several intrinsic limitations of using SEER data that should be considered when interpreting the result of this study. First, there is evidence of some race/ethnicity misclassification in SEER (35). Additionally, we were unable to differentiate Hispanic subgroups, such as Mexican, Central and South American. Moreover, we did not report outcomes for women with mixed or unknown race/ethnicity. There is also potential for misclassification of race/ethnicity given that these data are abstracted from medical records. In addition, the SEER cancer registries do not routinely collect risk factor information (e.g., comorbidities, family history, insurance status, complete course of treatment, etc.) that could inform study findings (35). Such a limitation precludes more than speculation about the reasons for differences reported (e.g., whether this difference is due to variations in treatment, early detection by mammography, or differences in underlying tumor biology, etc.).

In conclusion, we provide an updated and detailed analysis of trends and disparities in breast cancer outcomes across nine mutually exclusive racial and ethnic groups in the United States, with particular focus on diagnosis at earlier stages for which intervention is most beneficial to patients. While our results indirectly highlight progress toward reducing breast cancer disparities over the past two decades, they call for more work to address and eliminate persistent racial and ethnic disparities across the entire breast cancer continuum – from screening to survival – particularly among those diagnosed with early stage disease.

No disclosures were reported.

K.M. Primm: Conceptualization, data curation, formal analysis, investigation, methodology, writing–original draft, writing–review and editing. H. Zhao: Resources, supervision, methodology, writing–review and editing. D.C. Hernandez: Resources, supervision, writing–review and editing. S. Chang: Conceptualization, resources, supervision, validation, writing–review and editing.

This research and Drs. Primm and Chang are supported by the Cancer Prevention Research Training Program (CPRTP) at MD Anderson Cancer Center and an award from the Cancer Prevention and Research Institute of Texas (CPRIT) for the CPRTP Postdoctoral Fellowship in Cancer Prevention Program (RP 170259, Drs. Shine Chang and Sanjay Shete, Principal Investigators). This research and K. M. Primm and S. Chang are supported by the MD Anderson Cancer Center and an award from the Cancer Prevention and Research Institute of Texas (CPRIT) for the CPRTP Postdoctoral Fellowship in Cancer Prevention Program (RP 170259).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Berry
DA
,
Cronin
KA
,
Plevritis
SK
,
Fryback
DG
,
Clarke
L
,
Zelen
M
, et al
.
Effect of screening and adjuvant therapy on mortality from breast cancer
.
N Engl J Med
2005
;
353
:
1784
92
.
2.
DeSantis
CE
,
Ma
J
,
Goding Sauer
A
,
Newman
LA
,
Jemal
A
.
Breast cancer statistics, 2017, racial disparity in mortality by state
.
CA Cancer J Clin
2017
;
67
:
439
48
.
3.
Wheeler
SB
,
Reeder-Hayes
KE
,
Carey
LA
.
Disparities in breast cancer treatment and outcomes: biological, social, and health system determinants and opportunities for research
.
Oncologist
2013
;
18
:
986
93
.
4.
Akinyemiju
T
,
Moore
JX
,
Ojesina
AI
,
Waterbor
JW
,
Altekruse
SF
.
Racial disparities in individual breast cancer outcomes by hormone-receptor subtype, area-level socio-economic status and healthcare resources
.
Breast Cancer Res Treat
2016
;
157
:
575
86
.
5.
Tao
L
,
Gomez
SL
,
Keegan
TH
,
Kurian
AW
,
Clarke
CA
.
Breast cancer mortality in African-American and non-Hispanic white women by molecular subtype and stage at diagnosis: a population-based study
.
Cancer Epidemiol Biomarkers Prev
2015
;
24
:
1039
45
.
6.
Hardy
D
,
Du
DY
.
Socioeconomic and racial disparities in cancer stage at diagnosis, tumor size, and clinical outcomes in a large cohort of women with breast cancer, 2007–2016
.
J Racial Ethn Health Disparities
2021
;
8
:
990
1001
.
7.
Ma
H
,
Lu
Y
,
Malone
KE
,
Marchbanks
PA
,
Deapen
DM
,
Spirtas
R
, et al
.
Mortality risk of black women and white women with invasive breast cancer by hormone receptors, HER2, and p53 status
.
BMC Cancer
2013
;
13
:
1
11
.
8.
DeSantis
CE
,
Ma
J
,
Jemal
A
.
Trends in stage at diagnosis for young breast cancer patients in the United States
.
Breast Cancer Res Treat
2019
;
173
:
743
7
.
9.
Ko
NY
,
Hong
S
,
Winn
RA
,
Calip
GS
.
Association of insurance status and racial disparities with the detection of early-stage breast cancer
.
JAMA Oncol
2020
;
6
:
385
92
.
10.
Harper
S
,
Lynch
J
,
Meersman
SC
,
Breen
N
,
Davis
WW
,
Reichman
MC
.
Trends in area-socioeconomic and race-ethnic disparities in breast cancer incidence, stage at diagnosis, screening, mortality, and survival among women ages 50 years and over (1987–2005)
.
Cancer Epidemiol Biomarkers Prev
2009
;
18
:
121
31
.
11.
Iqbal
J
,
Ginsburg
O
,
Rochon
PA
,
Sun
P
,
Narod
SA
.
Differences in breast cancer stage at diagnosis and cancer-specific survival by race and ethnicity in the United States
.
JAMA
2015
;
313
:
165
73
.
12.
Ooi
SL
,
Martinez
ME
,
Li
CI
.
Disparities in breast cancer characteristics and outcomes by race/ethnicity
.
Breast Cancer Res Treat
2011
;
127
:
729
38
.
13.
Liu
L
,
Zhang
J
,
Wu
AH
,
Pike
MC
,
Deapen
D
.
Invasive breast cancer incidence trends by detailed race/ethnicity and age
.
Int J Cancer
2012
;
130
:
395
404
.
14.
Tuan
AW
,
Davis Lynn
BC
,
Chernyavskiy
P
,
Yu
M
,
Gomez
SL
,
Gierach
GL
, et al
.
Breast cancer incidence trends by estrogen receptor status among Asian American ethnic groups, 1990–2014
.
JNCI Cancer Spectrum
2020
;
4
:
pkaa005
.
15.
Surveillance, Epidemiology, and End Results Program
.
SEER*Stat Database: Incidence - SEER Research Plus Data, 18 Registries, Nov 2020 Sub (2000–2018) - Linked To County Attributes - Total U.S., 1969–2019 Counties, National Cancer Institute, DCCPS, Surveillance Research Program, released April 2021, based on the November 2020 submission
;
2021.
Available from
: https://www.seer.cancer.gov.
16.
Surveillance, Epidemiology, and End Results Program
.
Data use agreement for the 1975–2018 (November 2020 Submission) SEER research data file
;
2021.
Available from
: https://seer.cancer.gov/data-software/documentation/seerstat/nov2020/seer-dua-nov2020.html.
17.
National Cancer Institute
.
NCI Dictionary of Cancer Terms: Early-stage breast cancer
.
Available from
: https://www.cancer.gov/publications/dictionaries/cancer-terms/def/early-stage-breast-cancer.
18.
Ruhl
JL
,
Callaghan
C
,
Schussler
N
. editors.
Extent of Disease (EOD) 2018 General Coding Instructions
,
Bethesda, MD
:
National Cancer Institute
;
2021
.
19.
Ahn
J
,
Harper
S
,
Yu
M
,
Feuer
EJ
,
Liu
B
,
Luta
G
.
Variance estimation and confidence intervals for 11 commonly used health disparity measures
.
JCO Clin Cancer Inform
2018
;
2
:
1
19
.
20.
Joinpoint Regression Program, Version 4.8.0.1
.
Statistical Methodology and Applications Branch, Surveillance Research Program, National Cancer Institute. Available from
: https://surveillance.cancer.gov/joinpoint/.
21.
StataCorp
.
Stata Statistical Software: Release 15
.
College Station, TX
:
StataCorp LLC
;
2017
. Available from
: https://www.stata.com/.
22.
Trewin
CB
,
Johansson
ALV
,
Hjerkind
KV
,
Strand
BH
,
Kiserud
CE
,
Ursin
G
.
Stage-specific survival has improved for young breast cancer patients since 2000: but not equally
.
Breast Cancer Res Treat
2020
;
182
:
477
89
.
23.
Danforth
DN
.
Disparities in breast cancer outcomes between Caucasian and African American women: a model for describing the relationship of biological and nonbiological factors
.
Breast Cancer Res
2013
;
15
:
1
15
.
24.
Siddharth
S
,
Sharma
D
.
Racial disparity and triple-negative breast cancer in African-American women: a multifaceted affair between obesity, biology, and socioeconomic determinants
.
Cancers
2018
;
10
:
514
.
25.
Arciero
CA
,
Yang
J
,
Peng
L
,
Ward
KC
,
O'Regan
R
,
Sahin
AA
., et al
.
African American patients with breast cancer have worse prognosis than white patients in certain subtypes and stages
.
Breast Cancer Res Treat
2017
;
166
:
743
55
.
26.
Wu
AH
,
Yu
MC
,
Tseng
CC
,
Hankin
J
,
Pike
MC
.
Green tea and risk of breast cancer in Asian Americans
.
Int J Cancer
2003
;
106
:
574
9
.
27.
Selvakumar
P
,
Badgeley
A
,
Murphy
P
,
Anwar
H
,
Sharma
U
,
Lawrence
K
, et al
.
Flavonoids and other polyphenols act as epigenetic modifiers in breast cancer
.
Nutrients
2020
;
12
:
761
.
28.
Schueler
KM
,
Chu
PW
,
Smith-Bindman
R
.
Factors associated with mammography utilization: a systematic quantitative review of the literature
.
Journal of Women's Health
2008
;
17
:
1477
98
.
29.
Smith-Bindman
R
,
Miglioretti
DL
,
Lurie
N
,
Abraham
L
,
Barbash
RB
,
Strzelczyk
J
, et al
.
Does utilization of screening mammography explain racial and ethnic differences in breast cancer?
Ann Intern Med
2006
;
144
:
541
53
.
30.
Prakash
O
,
Hossain
F
,
Danos
D
,
Lassak
A
,
Scribner
R
,
Miele
L
.
Racial disparities in triple negative breast cancer: a review of the role of biologic and non-biologic factors
.
Front Public Health
2020
;
8
:
576964
.
31.
Lund
MJ
,
Brawley
OP
,
Ward
KC
,
Young
JL
,
Gabram
SS
,
Eley
JW
.
Parity and disparity in first course treatment of invasive breast cancer
.
Breast Cancer Res Treat
2008
;
109
:
545
57
.
32.
Fang
P
,
He
W
,
Gomez
D
,
Hoffman
KE
,
Smith
BD
,
Giordano
SH
, et al
.
Racial disparities in guideline-concordant cancer care and mortality in the United States
.
Adv Radiat Oncol
2018
;
3
:
221
9
.
33.
Gorin
SS
,
Badr
H
,
Krebs
P
,
Prabhu Das
I
.
Multilevel interventions and racial/ethnic health disparities
.
J Natl Cancer Inst Monogr
2012
;
2012
:
100
11
.
34.
Warnecke
RB
,
Oh
A
,
Breen
N
,
Gehlert
S
,
Paskett
E
,
Tucker
KL
, et al
.
Approaching health disparities from a population perspective: the National Institutes of Health Centers for Population Health and Health Disparities
.
Am J Public Health
2008
;
98
:
1608
15
.
35.
Jame
BYM
,
Gross
CP
,
Wilson
LD
,
Smith
BD
.
NCI SEER public-use data: applications and limitations in oncology research
.
Oncology
2009
;
23
:
288
.

Supplementary data