Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Dec 1.
Published in final edited form as: J Community Health. 2015 Dec;40(6):1287–1299. doi: 10.1007/s10900-015-0052-y

Intersection of Race/Ethnicity and Socioeconomic Status in Mortality After Breast Cancer

Salma Shariff-Marco 1,2,3,✉, Juan Yang 1, Esther M John 1,2,3, Allison W Kurian 2,3,4, Iona Cheng 1,3, Rita Leung 1, Jocelyn Koo 1, Kristine R Monroe 5, Brian E Henderson 5, Leslie Bernstein 6, Yani Lu 7, Marilyn L Kwan 8, Richard Sposto 5, Cheryl L P Vigen 9, Anna H Wu 5,10, Theresa H M Keegan 1,2,3, Scarlett Lin Gomez 1,2,3
PMCID: PMC4628564  NIHMSID: NIHMS700445  PMID: 26072260

Abstract

We investigated social disparities in breast cancer (BC) mortality, leveraging data from the California Breast Cancer Survivorship Consortium. The associations of race/ethnicity, education, and neighborhood SES (nSES) with all-cause and BC-specific mortality were assessed among 9372 women with BC (diagnosed 1993–2007 in California with follow-up through 2010) from four racial/ ethnic groups [African American, Asian American, Latina, and non-Latina (NL) White] using Cox proportional hazards models. Compared to NL White women with high-education/high-nSES, higher all-cause mortality was observed among NL White women with high-education/ low-nSES [hazard ratio (HR) (95 % confidence interval) 1.24 (1.08–1.43)], and African American women with low-nSES, regardless of education [high education HR 1.24 (1.03–1.49); low-education HR 1.19 (0.99–1.44)]. Latina women with low-education/high-nSES had lower all-cause mortality [HR 0.70 (0.54–0.90)] and non-significant lower mortality was observed for Asian American women, regardless of their education and nSES. Similar patterns were seen for BC-specific mortality. Individual- and neighborhood-level measures of SES interact with race/ ethnicity to impact mortality after BC diagnosis. Considering the joint impacts of these social factors may offer insights to understanding inequalities by multiple social determinants of health.

Keywords: Breast cancer survival, Racial/ethnic disparities, Socioeconomic disparities, Education, Neighborhood socioeconomic status

Introduction

Racial/ethnic and socioeconomic disparities in mortality after breast cancer (BC) diagnosis are persistent in the United States (U.S.). These disparities remain even after accounting for differences in important prognostic factors including clinical factors (e.g., tumor characteristics, treatment), personal risk factors (e.g., reproductive factors and lifestyle behaviors), sociodemographic characteristics, and health care access [1–3]. Race/ethnicity and socioeconomic status (SES) are highly correlated; however, their complex relations with mortality after BC have been difficult to disentangle given that prior studies have used different individual measures (e.g., education, income) and neighborhood levels (e.g., census block, block group, tract, zip code, county) to represent SES [4, 5]. While some studies have evaluated both individual SES and neighborhood SES (nSES) measures [6–11], only one has included diverse racial/ethnic populations [12].

Measuring SES at multiple levels is important because individual-level SES (e.g., education, income, wealth) may influence survival through material and social resources, including access to and quality of health care, and lifestyle risk factors [13, 14], whereas nSES may influence survival through features of the physical (e.g., goods, services, pollutants) and social (e.g., cohesion, collective efficacy, support, stress, coping) environment [7, 15, 16]. A few studies of BC and other health outcomes suggest that the type and level of SES measure can contribute differentially to health, and that these effects may further differ by race/ ethnicity [12, 17–19]. This work supports an emerging perspective for evaluating social inequalities, known as the ‘‘intersectional approach’’ [19], which emphasizes the interactions among multiple social determinants of health and the analytic approach to consider their joint effects. Such studies, however, require large numbers of population subgroups [1, 20, 21].

We aimed to assess the joint associations of race/ethnicity, education, and nSES with all-cause and BC-specific mortality, leveraging data from the large and diverse cohort of women with BC assembled in the California Breast Cancer Survivorship Consortium (CBCSC) [2].

Methods

Study Population

This analysis included five studies from the CBCSC, which was established in 2011 to better understand racial/ethnic disparities in survival among women with BC, who were diagnosed from 1993 through 2007 [2]. The studies included three case–control studies [Asian American Breast Cancer Study (AABCS), the Women’s Contraceptive and Reproductive Experiences Study (CARE), the San Francisco Bay Area Breast Cancer Study (SFBCS)], and two prospective cohort studies [the California Teachers’ Study (CTS), the Multiethnic Cohort (MEC)]. For the three case–control studies, the mean (standard deviation) years from diagnosis to data collection were 1.6 (0.8) years for AABCS, 0.4 (0.3) years for CARE, and 1.4 (0.6) years for SFBCS. In brief, interview data on prognostic factors were harmonized across the five studies and merged with California Cancer Registry (CCR) data on clinical and tumor characteristics, treatment, vital status, hospital characteristics, and nSES. The protocols for the CBCSC study were approved by the institutional review boards (IRBs) at all participating institutions and the California state IRB (Committee for the Protection of Human Subjects).

A total of 10,521 women with BC were potentially eligible for analysis. We further excluded, in sequence, women with in situ BC (n = 22), women with cancers diagnosed before their invasive BC (n = 779), and women with<30 days of follow-up (n = 19). Finally, we excluded 63 women of races/ethnicities other than non-Latina (NL) White, Latina, African American, and Asian American, and 266 with missing education or nSES, yielding a final study population of 9372 women with BC.

Analytic Variables

CCR data included age and year at diagnosis, American Joint Committee on Cancer (AJCC) stage, histology, grade, tumor size, nodal status, estrogen receptor (ER) and pro-gesterone receptor (PR) status, first course of treatment (surgery, radiation, chemotherapy), subsequent tumors (including time between diagnoses), CCR region, and marital status. CCR data were used to create an indicator of hospital-level SES using percent of cancer cases in the highest nSES quintile based on the distribution of nSES (defined below) among registry cases diagnosed from 1993 through 2007. For each hospital, percent of cases residing in high SES neighborhoods (quintile 5) at the time of diagnosis was calculated and then categorized into statewide quintiles.

Geocoding of case addresses at the time of diagnosis was centralized at the CCR using commercial geocoding vendors. Cases’ addresses were assigned latitude and longitude coordinates and then assigned to a U.S. Census block group and merged with a block group-level SES measure (see detailed description below). We included 97.5 % of the cases with complete addresses or zip codes (zip code plus four digit format) that were accurately matched to unique, valid census block groups. For cases diagnosed prior to 1996, 1990 U.S. Census block group and nSES were assigned. For cases diagnosed from 1996 through 2007, 2000 U.S. Census block groups and nSES were assigned. Of the 8225 unique census block groups that were included in our study, 74 % of the block groups had only one case and 92 % had two or fewer cases.

Questionnaire data were collected via in-person interviews (in case–control studies) or self-administered mail surveys (in cohort studies) using structured questionnaires administered in English, Spanish, Tagalog and/or Chinese (Mandarin and Cantonese). Questionnaire data were harmonized according to common definitions for the following variables: number of full-term pregnancies (0, 1, 2, 3, ≥4), smoking status (never, past, current), alcoholic drinks per week (0, ≤2, >2), pre-diagnosis body mass index (BMI) (<25, 25–29.9, ≥30 kg/m2), and personal history of high blood pressure or diabetes [2, 22]. Race/ethnicity was classified (NL White, African American, Latina, Asian American) according to self-report on the study surveys.

As one dimension of individual-level SES, we used self-reported education, categorized into four levels: less than high school, high school degree or equivalent, vocational/ technical degree or some college, college degree or graduate school. No other individual-level SES indicators were available in the CBCSC.

For nSES, we used a composite SES measure created by principal component analysis of Census 1990 or 2000 SES indicator variables at the block group-level that includes an education index (among individuals age ≥25 years: proportion with college, high school, or less than high school weighted by 16, 12 or 9, respectively) [23], proportion with a blue collar job, proportion older than age 16 years without a job, median household income, proportion below 200 % of the poverty line, median rent, and median house value [24]. We were interested in a general indicator of SES for neighborhoods, rather than specific components of SES such as education or poverty, which may have different effects on health outcomes across the diverse population and geographic subgroups in California [17, 25]. This composite nSES index has shown consistent associations with a variety of cancer outcomes and also enables us to compare our results to those of other studies that have used the same index [12, 26–32]. We categorized this nSES index into quintiles based on the statewide distribution.

To implement the intersectional approach, we accounted for race/ethnicity, individual- and neighborhood-level SES in a single, combination variable using binary indicators for education and nSES. Low education was defined as having a high school degree or less, and high education as having at least a vocational/technical degree after high school or some college education; low nSES included quintiles 1–3 and high nSES, quintiles 4–5. These binary cut-points were selected to achieve balanced samples.

The CCR obtains vital status and underlying cause of death through hospital follow-up and linkages to vital statistics, death records, and other databases. BC deaths were identified from the underlying cause of death listed on the death certificate [International Classification of Diseases (ICD)-9 or ICD-10 codes 174–175 and C50, respectively] [33, 34]. Follow-up time was defined as the time from date of diagnosis to study end date (December 31, 2010), last known contact, or death, whichever came first. We had a median follow-up time of 9.4 years (interquartile range 6.3–12.5 years).

Analysis

To assess the joint association of race/ethnicity, education, and nSES with mortality, we fitted Cox proportional hazards multiple regression models, with cluster adjustment for block groups, to compute hazard rate ratios (HR) of dying from any cause or from BC. The sandwich estimator of the covariance structure, applied to Cox proportional hazards regression models, was utilized to account for the intracluster dependence and yields robust standard error estimates even under model misspecification [35]. All Cox models used attained age (in days) as the time scale, and were stratified on stage and study to allow the baseline hazards within each model to vary by stage and study. Women in the case–control studies (AABCS, CARE, SFBCS) survived after diagnosis until the time of interview; thus, their follow-up was left censored since women who died or were lost to follow-up before data collection by the parent study were not included in this study. The assumption of proportional hazards was checked by including interaction terms with time and assessing their significance using likelihood ratio tests, and confirming proportionality for each of the covariates included in the models. Analyses were conducted using SAS (version 9.3, Cary, NC). We also tested for spatial autocorrelation using Moran’s I, and found no evidence of this correlation.

First, we assessed associations between our race/ethnicity, education and nSES variables and mortality in base models that were adjusted for age at diagnosis, year of diagnosis, CCR region, tumor characteristics (histology, grade, ER/PR status, nodal involvement, tumor size), and subsequent tumors. Next, models were further adjusted sequentially for various sets of prognostic factors—treatment including chemotherapy, radiation and surgery (model 1); parity, marital status, smoking status, alcohol intake, BMI (model 2); comorbidities including hypertension and diabetes (model 3); and hospital SES (model 4).

Results

Personal and social characteristics of the 9372 women with BC included in the analysis are presented in Table 1. Relative to other racial/ethnic groups, NL White women were more likely to be past smokers or drink more than two servings of alcohol per week. African American women were more likely than other groups to be divorced or separated, current smokers, or obese. Latina women were more likely than other groups to have four or more children, or be overweight. Asian American women were more likely than other groups to be married, never smokers, non-drinkers, or normal/underweight.

Table 1.

Distribution of personal and social characteristics, California Breast Cancer Survivorship Consortium (CBCSC), 1993–2007

Race/ethnicity
Non-Latina White (N = 4480)
African American (N = 1790)
Latina (N = 1797)
Asian American (N = 1305)
Total (N = 9372)
N % N % N % N % N %
Studya
 AABCS 0 0.0 0 0.0 0 0.0 1075 82.4 1075 11.5
 CARE 532 11.9 539 30.1 85 4.7 0 0.0 1156 12.3
 SFBCS 537 12.0 506 28.3 1048 58.3 0 0.0 2091 22.3
 CTS 3062 68.3 70 3.9 86 4.8 92 7.0 3310 35.3
 MEC 349 7.8 675 37.7 578 32.2 138 10.6 1740 18.6
Neighborhood SES (nSES)b
 Quintile 1-lowest nSES 145 3.2 508 28.4 232 12.9 128 9.8 1013 10.8
 Quintile 2 435 9.7 463 25.9 368 20.5 235 18.0 1501 16.0
 Quintile 3 760 17.0 371 20.7 391 21.8 254 19.5 1776 19.0
 Quintile 4 1215 27.1 289 16.1 401 22.3 333 25.5 2238 23.9
 Quintile 5-highest nSES 1925 43.0 159 8.9 405 22.5 355 27.2 2844 30.3
Education
 <High school 82 1.8 255 14.2 683 38.0 104 8.0 1124 12.0
 High school 321 7.2 444 24.8 424 23.6 167 12.8 1356 14.5
 Some college/technical school 504 11.3 665 37.2 404 22.5 291 22.3 1864 19.9
 College graduate or higher degree 3573 79.8 426 23.8 286 15.9 743 56.9 5028 53.6
Marital status
 Single, never married 503 11.2 345 19.3 245 13.6 172 13.2 1265 13.5
 Married 2821 63.0 717 40.1 1062 59.1 924 70.8 5524 58.9
 Separated/divorced 504 11.3 366 20.4 223 12.4 63 4.8 1156 12.3
 Widowed 582 13.0 305 17.0 225 12.5 125 9.6 1237 13.2
 Unknown 70 1.6 57 3.2 42 2.3 21 1.6 190 2.0
Parity
 Nulliparous 1017 22.7 273 15.3 233 13.0 308 23.6 1831 19.5
 1 Birth 632 14.1 333 18.6 209 11.6 220 16.9 1394 14.9
 2 Births 1485 33.1 396 22.1 399 22.2 398 30.5 2678 28.6
 3 Births 816 18.2 327 18.3 357 19.9 219 16.8 1719 18.3
 >4 Births 481 10.7 447 25.0 590 32.8 149 11.4 1667 17.8
 Unknown 49 1.1 14 0.8 9 0.5 11 0.8 83 0.9
Smoking
 Never 2195 49.0 617 34.5 785 43.7 1024 78.5 4621 49.3
 Past 1417 31.6 429 24.0 277 15.4 191 14.6 2314 24.7
 Current 355 7.9 277 15.5 132 7.3 77 5.9 841 9.0
 Unknown 513 11.5 467 26.1 603 33.6 13 1.0 1596 17.0
Alcohol intake (drinks/week)
 Non-drinker 1422 31.7 1073 59.9 1042 58.0 1069 81.9 4606 49.1
 ≤2 842 18.8 316 17.7 362 20.1 82 6.3 1602 17.1
 >2 2052 45.8 343 19.2 358 19.9 149 11.4 2902 31.0
 Unknown 164 3.7 58 3.2 35 1.9 5 0.4 262 2.8
Pre-diagnosis body mass index (BMI)
 <25 (normal/underweight) 2515 56.1 527 29.4 563 31.3 843 64.6 4448 47.5
 25 to < 30 (overweight) 1184 26.4 591 33.0 619 34.4 347 26.6 2741 29.2
 30+ (obese) 619 13.8 616 34.4 575 32.0 93 7.1 1903 20.3
 Unknown 162 3.6 56 3.1 40 2.2 22 1.7 280 3.0
Open in a new tab
a

AABCS Asian American Breast Cancer Study, CARE Women’s Contraceptive and Reproductive Experiences Study, SFBCS San Francisco Bay Area Breast Cancer Study, CTS California Teachers’ Study, MEC Multiethnic Cohort

b

Neighborhood SES is measured using the Yost SES Index which is a composite measure of 7 Census indicator variables

Clinical and tumor characteristics for the sample are presented in Table 2. Relative to the other racial/ethnic groups, NL White women were more likely to be older at diagnosis, have tumors that were<1 cm, stage 1, grade I or lobular, and treated with radiation and lumpectomy. African American women were more likely than other groups to be seen in a low-SES hospital and have higher grade or ER−/PR− tumors. Latina women were more likely than other groups to be seen in a high-SES hospital and treated with chemotherapy. Asian American women were more likely than other groups to be younger at diagnosis, seen in a low-SES hospital, have a mastectomy, and were less likely to have radiation treatment.

Table 2.

Distribution of clinical characteristics, California Breast Cancer Survivorship Consortium (CBCSC), 1993–2007

Race/ethnicity
Non-Latina White (N = 4480)
African American (N = 1790)
Latina (N = 1797)
Asian American (N = 1305)
Total (N = 9372)
N % N % N % N % N %
Age at diagnosis
 < 40 171 3.8 107 6.0 114 6.3 98 7.5 490 5.2
 40 to < 50 550 12.3 353 19.7 391 21.8 379 29.0 1673 17.9
 50 to < 60 1299 29.0 465 26.0 452 25.2 347 26.6 2563 27.3
 60 to < 70 1217 27.2 430 24.0 504 28.0 293 22.5 2444 26.1
 70+ 1243 27.7 435 24.3 336 18.7 188 14.4 2202 23.5
AJCC summary stage
 Stage I 2406 53.7 720 40.2 804 44.7 627 48.0 4557 48.6
 Stage II 1592 35.5 804 44.9 764 42.5 539 41.3 3699 39.5
 Stage III 243 5.4 122 6.8 136 7.6 86 6.6 587 6.3
 Stage IV 91 2.0 50 2.8 31 1.7 19 1.5 191 2.0
 Unknown 148 3.3 94 5.3 62 3.5 34 2.6 338 3.6
Grade
 Grade I 1074 24.0 241 13.5 274 15.2 177 13.6 1766 18.8
 Grade II 1751 39.1 551 30.8 681 37.9 515 39.5 3498 37.3
 Grade III or IV 1194 26.7 763 42.6 643 35.8 505 38.7 3105 33.1
 Unknown 461 10.3 235 13.1 199 11.1 108 8.3 1003 10.7
ER/PR status
 ER+/PR− 2683 59.9 825 46.1 960 53.4 702 53.8 5170 55.2
 ER+/PR− 499 11.1 152 8.5 184 10.2 102 7.8 937 10.0
 ER−/PR+ 66 1.5 56 3.1 40 2.2 36 2.8 198 2.1
 ER−/PR− 582 13.0 396 22.1 365 20.3 175 13.4 1518 16.2
 Unknown 650 14.5 361 20.2 248 13.8 290 22.2 1549 16.5
Histology
 Ductal 107 69.4 1335 74.6 1344 74.8 956 73.3 6742 71.9
 Lobular 953 21.3 240 13.4 275 15.3 196 15.0 1664 17.8
 Other 420 9.4 215 12.0 178 9.9 153 11.7 966 10.3
Nodal involvement
 No nodes 3025 67.5 1055 58.9 1090 60.7 846 64.8 6016 64.2
 Positive nodes 1313 29.3 636 35.5 639 35.6 436 33.4 3024 32.3
 Unknown 142 3.2 99 5.5 68 3.8 23 1.8 332 3.5
Tumor size (cm)
 < 1 974 21.7 203 11.3 272 15.1 246 18.9 1695 18.1
 1 to < 5 3016 67.3 1315 73.5 1310 72.9 916 70.2 6557 70.0
 ≥5 248 5.5 148 8.3 111 6.2 86 6.6 593 6.3
 Unknown 242 5.4 124 6.9 104 5.8 57 4.4 527 5.6
Diagnosis with 1 subsequent primary tumor
 No 3679 82.1 1443 80.6 1531 85.2 1088 83.4 7741 82.6
 Yes 801 17.9 347 19.4 266 14.8 217 16.6 1631 17.4
Diagnosis with 2 subsequent primary tumors
 No 4391 98.0 1756 98.1 1768 98.4 1288 98.7 9203 98.2
 Yes 89 2.0 34 1.9 29 1.6 17 1.3 169 1.8
Chemotherapy
 No 2855 63.7 1026 57.3 935 52.0 703 53.9 5519 58.9
 Yes 1556 34.7 732 40.9 830 46.2 568 43.5 3686 39.3
 Unknown 69 1.5 32 1.8 32 1.8 34 2.6 167 1.8
Radiation
 No 2004 44.7 977 54.6 863 48.0 781 59.8 4625 49.3
 Yes 2476 55.3 813 45.4 934 52.0 524 40.2 4747 50.7
Surgery
 No surgery 90 2.0 82 4.6 29 1.6 17 1.3 218 2.3
 Mastectomy 1638 36.6 734 41.0 826 46.0 698 53.5 3896 41.6
 Lumpectomy 2743 61.2 971 54.2 941 52.4 588 45.1 5243 55.9
 Other 9 0.2 3 0.2 1 0.1 2 0.2 15 0.2
High blood pressure
 Yes 810 18.1 329 18.4 225 12.5 330 25.3 1694 18.1
 No 2932 65.4 421 23.5 563 31.3 824 63.1 4740 50.6
 Unknown 738 16.5 1040 58.1 1009 56.1 151 11.6 2938 31.3
Diabetes
 Yes 119 2.7 85 4.7 97 5.4 104 8.0 405 4.3
 No 3622 80.8 663 37.0 688 38.3 1050 80.5 6023 64.3
 Unknown 739 16.5 1042 58.2 1012 56.3 151 11.6 2944 31.4
Hospital patients of high SES (%)
 < 25 2053 45.8 1142 63.8 799 44.5 798 61.2 4792 51.1
 25 to < 50 1479 33.0 534 29.8 496 27.6 442 33.9 2951 31.5
 50 to < 75 886 19.8 109 6.1 479 26.7 64 4.9 1538 16.4
 75+ 62 1.4 5 0.3 23 1.3 < 5 0.1 91 1.0
Cancer registry region
 Los Angeles County 1447 32.3 1154 64.5 654 36.4 1236 94.7 4491 47.9
 Greater San Francisco Bay Area 1165 26.0 554 30.9 1058 58.9 29 2.2 2806 29.9
 Sacramento and Sierra 344 7.7 8 0.4 8 0.4 9 0.7 369 3.9
 San Diego, Orange, Imperial 650 14.5 25 1.4 26 1.4 16 1.2 717 7.7
 Rest of California 872 19.5 49 2.7 51 2.8 15 1.1 987 10.5
 Unknown < 5 0.0 0 0.0 0 0.0 0 0.0 < 5 0.0
Vital status and cause of death
 Alive 3303 73.7 1053 58.8 1356 75.5 1047 80.2 6759 72.1
 Breast cancer 527 11.8 433 24.2 250 13.9 164 12.6 1374 14.7
 Other cancer 153 3.4 78 4.4 52 2.9 31 2.4 314 3.4
 Cardiovascular diseases 183 4.1 103 5.8 47 2.6 19 1.5 352 3.8
 Diabetes or obesity 13 0.3 15 0.8 10 0.6 < 5 0.2 40 0.4
 Other causes 289 6.5 103 5.8 69 3.8 30 2.3 491 5.2
 Death certificate not available 12 0.3 5 0.3 13 0.7 12 0.9 42 0.4
Open in a new tab

Education and nSES distributions varied by race/ethnicity (Tables 1, 3). Among NL White women, 80 % had a college degree and 70 % lived in high SES (quintiles 4 and 5) neighborhoods, compared to 24 and 25 %, respectively, among African American women; 16 and 45 %, respectively, among Latina women; and 57 and 53 %, respectively, among Asian American women (Table 1). Table 3 shows the distributions of education by nSES for each racial/ethnic group. While individual-level education and nSES are correlated in all racial/ ethnic groups, the extent of correlation differed substantially across the groups, with similar degrees of correlation among Latina and Asian American women, but more clustering in the higher SES neighborhoods regardless of education among NL White women, and more clustering in the lower SES neighborhoods regardless of education among African American women. Notably, African American women with some college/ technical school, high school, and less than high school education had relatively small differences in terms of their nSES.

Table 3.

Distributions of education and neighborhood SES by race/ethnicity, California Breast Cancer Survivorship Consortium (CBCSC), 1993–2007

Neighborhood socioeconomic status (nSES) quintilesa
Q1-low nSES
Q2
Q3
Q4
Q5-high nSES
Total
N % N % N % N % N % N
Non Latina White (n = 4480)
 <High school 10 12.2 13 15.9 20 24.4 28 34.1 11 13.4 82
 High school 17 5.3 59 18.4 70 21.8 82 25.5 93 29.0 321
 Some college/technical school 25 5.0 58 11.5 86 17.1 130 25.8 205 40.7 504
 College graduate or higher degree 93 2.6 305 8.5 584 16.3 975 27.3 1616 45.2 3573
 Total 145 3.2 435 9.7 760 17.0 1215 27.1 1925 43.0 4480
African American (n = 1790)
 <High school 119 46.7 73 28.6 35 13.7 18 7.1 10 3.9 255
 High school 154 34.7 129 29.1 89 20.0 52 11.7 20 4.5 444
 Some college/technical school 169 25.4 180 27.1 146 22.0 120 18.0 50 7.5 665
 College graduate or higher degree 66 15.5 81 19.0 101 23.7 99 23.2 79 18.5 426
 Total 508 28.4 463 25.9 371 20.7 289 16.1 159 8.9 1790
Latina (n = 1797)
 <High school 137 20.1 186 27.2 168 24.6 118 17.3 74 10.8 683
 High school 46 10.8 78 18.4 101 23.8 108 25.5 91 21.5 424
 Some college/technical school 33 8.2 66 16.3 74 18.3 102 25.2 129 31.9 404
 College graduate or higher degree 16 5.6 38 13.3 48 16.8 73 25.5 111 38.8 286
 Total 232 12.9 368 20.5 391 21.8 401 22.3 405 22.5 1797
Asian American (n = 1305)
 <High school 25 24.0 25 24.0 26 25.0 17 16.3 11 10.6 104
 High school 21 12.6 30 18.0 44 26.3 46 27.5 26 15.6 167
 Some college/technical school 17 5.8 55 18.9 72 24.7 77 26.5 70 24.1 291
 College graduate or higher degree 65 8.7 125 16.8 112 15.1 193 26.0 248 33.4 743
 Total 128 9.8 235 18.0 254 19.5 333 25.5 355 27.2 1305
All (n = 9372)
 <High school 291 25.9 297 26.4 249 22.2 181 16.1 106 9.4 1124
 High school 238 17.6 296 21.8 304 22.4 288 21.2 230 17.0 1356
 Some college/technical school 244 13.1 359 19.3 378 20.3 429 23.0 454 24.4 1864
 College graduate or higher degree 240 4.8 549 10.9 845 16.8 1340 26.7 2054 40.9 5028
 Total 1013 10.8 1501 16.0 1776 19.0 2238 23.9 2844 30.3 9372
Open in a new tab
a

Neighborhood SES is measured using the Yost SES Index which is a composite measure of 7 Census indicator variables

Table 4 shows the hazard ratios for the three-way combination variables between race/ethnicity, education, and nSES. For all-cause mortality, compared to NL White women with high education/high-nSES, the following groups had higher mortality in the base models: NL White women with low-nSES, regardless of education (high-education HR 1.34, 95 % CI 1.16–1.54; low-education HR 1.38, 95 % CI 1.06–1.79), African American women with low-nSES, regardless of education (high-education HR 1.56, 95 % CI 1.32–1.85; low-education HR 1.56, 95 % CI 1.31–1.86), and African American women with low-education/high-nSES (HR 1.48, 95 % CI 1.04–2.09). Only one group had statistically significant lower mortality compared to NL White women with high-education/high-nSES: Latina women with low-education/high-nSES (HR 0.75, 95 % CI 0.58–0.95). After adjusting for treatment, individual-level risk factors, comorbidities and hospital SES, associations for NL White women with low-education/low-nSES and African American women with low-education/ high-nSES were no longer observed (see model 2 in Table 4 which shows associations were not observed after adjusting for individual-level factors). Among African American women with low-education/low-nSES, only a marginal association remained after adjustment for hospital SES. In the fully adjusted models, compared to NL White women with high-education/high-nSES, NL White and African American women with high-education/low-nSES had slightly attenuated associations of higher mortality (HR 1.24, 95 % CI 1.08–1.43 and HR 1.24, 95 % CI 1.03–1.49, respectively), while Latina women with low-education/high-nSES had a stronger association of lower mortality (HR 0.70, 95 % CI 0.54–0.90). Lower mortality was observed for Asian American women, regardless of their education and nSES; however, none of the estimates were statistically significant.

Table 4.

Hazard ratios for joint associations of race/ethnicity, education, and neighborhood socioeconomic status with all-cause and breast cancer-specific mortality, California Breast Cancer Survivorship Consortium (CBCSC), 1993–2007

Cases Deaths Base modela Model 1: base model + treatmentb Model 2: model 1 + parity, marital status, and behavioral factorsc Model 3: model 2 + comorbidityd Model 4: model 3 + hospital factorse
N = 9372 (%) N = 2613 (%) HR (95 % CI) HR (95 % CI) HR (95 % CI) HR (95 % CI) HR (95 % CI)
All-cause mortality
 Race, education and nSESf
  NL White, high edu, high nSES 2926 (31.2 %) 679 (26.0 %) 1.00 (Ref.) 1.00 (Ref.) 1.00 (Ref.) 1.00 (Ref.) 1.00 (Ref.)
  NL White, high edu, low nSES 1151 (12.3 %) 356 (13.6 %) 1.34 (1.16–1.54) 1.36 (1.18–1.56) 1.28 (1.11–1.47) 1.27 (1.10–1.46) 1.24 (1.08–1.43)
  NL White, low edu, high nSES 214 (2.3 %) 74 (2.8 %) 1.17 (0.90–1.51) 1.18 (0.91–1.52) 1.16 (0.90–1.50) 1.14 (0.88–1.47) 1.12 (0.87–1.45)
  NL White, low edu, low nSES 189 (2.0 %) 68 (2.6 %) 1.38 (1.06–1.79) 1.43 (1.10–1.86) 1.28 (0.99–1.67) 1.26 (0.97–1.64) 1.22 (0.94–1.58)
  Afr Am, high edu, high nSES 348 (3.7 %) 121 (4.6 %) 1.23 (0.99–1.54) 1.25 (1.00–1.57) 1.14 (0.90–1.43) 1.11 (0.88–1.41) 1.07 (0.85–1.36)
  Afr Am, high edu, low nSES 743 (7.9 %) 295 (11.3 %) 1.56 (1.32–1.85) 1.56 (1.31–1.85) 1.34 (1.12–1.60) 1.30 (1.09–1.56) 1.24 (1.03–1.49)
  Afr Am, low edu, high nSES 100 (1.1 %) 46 (1.8 %) 1.48 (1.04–2.09) 1.50 (1.07–2.11) 1.34 (0.95–1.88) 1.32 (0.94–1.85) 1.27 (0.90–1.78)
  Afr Am, low edu, low nSES 599 (6.4 %) 275 (10.5 %) 1.56 (1.31–1.86) 1.57 (1.32–1.87) 1.30 (1.08–1.57) 1.26 (1.05–1.51) 1.19 (0.99–1.44)
  Latina, high edu, high nSES 415 (4.4 %) 74 (2.8 %) 0.79 (0.61–1.02) 0.80 (0.62–1.03) 0.81 (0.63–1.04) 0.80 (0.62–1.04) 0.80 (0.62–1.03)
  Latina, high edu, low nSES 275 (2.9 %) 64 (2.4 %) 0.97 (0.73–1.29) 1.00 (0.76–1.32) 0.96 (0.72–1.27) 0.92 (0.69–1.22) 0.88 (0.66–1.18)
  Latina, low edu, high nSES 391 (4.2 %) 88 (3.4 %) 0.75 (0.58–0.95) 0.77 (0.60–0.98) 0.73 (0.57–0.94) 0.72 (0.56–0.93) 0.70 (0.54–0.90)
  Latina, low edu, low nSES 716 (7.6 %) 215 (8.2 %) 1.12 (0.93–1.35) 1.13 (0.93–1.36) 1.01 (0.83–1.22) 0.98 (0.81–1.20) 0.94 (0.77–1.15)
  Asian Am, high edu, high nSES 588 (6.3 %) 95 (3.6 %) 0.79 (0.55–1.14) 0.79 (0.55–1.13) 0.80 (0.55–1.16) 0.77 (0.53–1.11) 0.77 (0.53–1.11)
  Asian Am, high edu, low nSES 446 (4.8 %) 94 (3.6 %) 0.91 (0.61–1.37) 0.90 (0.60–1.34) 0.88 (0.58–1.32) 0.82 (0.55–1.23) 0.81 (0.54–1.21)
  Asian Am, low edu, high nSES 100 (1.1 %) 20 (0.8 %) 0.84 (0.52–1.35) 0.82 (0.51–1.32) 0.82 (0.50–1.32) 0.84 (0.53–1.35) 0.84 (0.52–1.34)
  Asian Am, low edu, low nSES 171 (1.8 %) 49 (1.9 %) 0.93 (0.61–1.44) 0.93 (0.61–1.43) 0.91 (0.59–1.41) 0.88 (0.57–1.36) 0.87 (0.56–1.34)
N = 9372 (%) N = 1374 (%) HR (95 % CI) HR (95 % CI) HR (95 % CI) HR (95 % CI) HR (95 % CI)
Breast cancer-specific mortality
 Race, education and nSESf
  NL White, high edu, high nSES 2926 (31.2 %) 319 (23.2 %) 1.00 (Ref.) 1.00 (Ref.) 1.00 (Ref.) 1.00 (Ref.) 1.00 (Ref.)
  NL White, high edu, low nSES 1151 (12.3 %) 147 (10.7 %) 1.14 (0.93–1.40) 1.14 (0.93–1.40) 1.11 (0.90–1.37) 1.10 (0.89–1.35) 1.06 (0.86–1.32)
  NL White, low edu, high nSES 214 (2.3 %) 28 (2.0 %) 0.89 (0.61–1.30) 0.90 (0.61–1.33) 0.88 (0.59–1.30) 0.86 (0.58–1.28) 0.85 (0.57–1.27)
  NL White, low edu, low nSES 189 (2.0 %) 33 (2.4 %) 1.22 (0.82–1.82) 1.31 (0.89–1.93) 1.26 (0.85–1.87) 1.25 (0.84–1.85) 1.20 (0.81–1.77)
  Afr Am, high edu, high nSES 348 (3.7 %) 77 (5.6 %) 1.27 (0.96–1.69) 1.29 (0.97–1.72) 1.21 (0.90–1.63) 1.21 (0.90–1.64) 1.16 (0.86–1.57)
  Afr Am, high edu, low nSES 743 (7.9 %) 186 (13.5 %) 1.54 (1.24–1.92) 1.55 (1.24–1.94) 1.46 (1.16–1.85) 1.45 (1.15–1.84) 1.37 (1.07–1.75)
  Afr Am, low edu, high nSES 100 (1.1 %) 28 (2.0 %) 1.63 (1.07–2.48) 1.69 (1.12–2.55) 1.61 (1.06–2.44) 1.61 (1.05–2.46) 1.55 (1.01–2.37)
  Afr Am, low edu, low nSES 599 (6.4 %) 142 (10.3 %) 1.37 (1.08–1.75) 1.42 (1.11–1.81) 1.31 (1.01–1.69) 1.29 (1.00–1.67) 1.21 (0.92–1.58)
  Latina, high edu, high nSES 415 (4.4 %) 54 (3.9 %) 0.86 (0.63–1.18) 0.89 (0.65–1.21) 0.95 (0.69–1.30) 0.94 (0.69–1.29) 0.93 (0.68–1.27)
  Latina, high edu, low nSES 275 (2.9 %) 39 (2.8 %) 0.92 (0.64–1.31) 0.94 (0.66–1.34) 1.01 (0.70–1.45) 0.99 (0.69–1.42) 0.94 (0.65–1.35)
  Latina, low edu, high nSES 391 (4.2 %) 44 (3.2 %) 0.62 (0.44–0.89) 0.65 (0.45–0.92) 0.71 (0.50–1.02) 0.70 (0.49–1.01) 0.68 (0.47–0.98)
  Latina, low edu, low nSES 716 (7.6 %) 113 (8.2 %) 0.98 (0.76–1.27) 1.00 (0.77–1.29) 0.99 (0.76–1.30) 0.98 (0.75–1.29) 0.93 (0.71–1.23)
  Asian Am, high edu, high nSES 588 (6.3 %) 70 (5.1 %) 0.81 (0.48–1.39) 0.81 (0.47–1.38) 0.85 (0.50–1.44) 0.84 (0.49–1.43) 0.85 (0.49–1.45)
  Asian Am, high edu, low nSES 446 (4.8 %) 59 (4.3 %) 0.81 (0.45–1.44) 0.80 (0.44–1.42) 0.82 (0.46–1.47) 0.82 (0.46–1.47) 0.80 (0.45–1.44)
  Asian Am, low edu, high nSES 100 (1.1 %) 8 (0.6 %) 0.59 (0.26–1.36) 0.59 (0.26–1.33) 0.61 (0.27–1.40) 0.64 (0.29–1.43) 0.64 (0.28–1.43)
  Asian Am, low edu, low nSES 171 (1.8 %) 27 (2.0 %) 0.74 (0.40–1.39) 0.74 (0.40–1.38) 0.76 (0.41–1.44) 0.77 (0.41–1.44) 0.75 (0.40–1.41)
Open in a new tab

Statistically significant values (p < 0.05) are given in bold

a

Hazard ratios (HR) and 95 % confidence intervals (CI) were estimated using multivariable Cox proportional hazards regression models with age at diagnosis or interview (which occurred later) and age at last contact (in days) as the fundamental time scale, adjusted for age at diagnosis in years (continuous), logarithm of age at diagnosis in years (continuous), year at diagnosis (continuous), cancer registry region (Region1/8, Region 2/4/5/6, Region 3, Region 7/10, Region 9), race/ethnicity (non-Latino White, African American, Latina, Asian American), histology (ductal, lobular, other), grade (grade I, II, III and IV, unknown), EPPR status (ER+PR+, ER+PR−, ER−PR+, ER−PR−, unknown), nodal involvement (no nodes, positive nodes, unknown), tumor size data availability (available, unavailable), tumor size in centimeters (continuous), had 1+ subsequent tumor (yes, no), had 2+ subsequent tumor (yes, no), days between dates at diagnosis of CBCSC study tumor and the 1st subsequent tumor (continuous), days between dates at diagnosis of 1st and 2nd subsequent tumor (continuous), and clustering by block group, and stratified by study (CTS, MEC or AABCS, CARE, SFBCS—depends on study type) and AJCC stage (stage I, II, III, IV, unknown)

b

Adjusted for covariates of base model, surgery type (no surgery, mastectomy, lumpectomy, other), chemotherapy (none, treatment, unknown), radiation (none, treatment, unknown), and clustering by block group, and stratified by study (CTS, MEC or AABCS, CARE, SFBCS—depends on study type) and AJCC stage (stage I, II, III, IV, unknown)

c

Adjusted for covariates of model 1, parity (nulliparous, 1, 2, 3, 4+ births, unknown), marital status (single, married, separated/divorced, widowed, unknown), smoking (never, past, current, unknown), alcohol drinks per week (non-drinker, ≤2,>2 drinks, unknown), pre-diagnosis BMI (<25, 25–29, 30+, unknown), and clustering by block group, and stratified by study (CTS, MEC or AABCS, CARE, SFBCS—depends on study type) and AJCC stage (stage I, II, III, IV, unknown)

d

Adjusted for covariates of model 2, hypertension (yes, no, unknown), diabetes (yes, no, unknown), and clustering by block group, and stratified by study (CTS, MEC or AABCS, CARE, SFBCS—depends on study type) and AJCC stage (stage I, II, III, IV, unknown)

e

Adjusted for covariates of model 3, percent of cancer patients in highest SES quintile in reporting hospital (<25, 25–49, 50–74, ≥75 %, unknown), and clustering by block group, and stratified by study (CTS, MEC or AABCS, CARE, SFBCS—depends on study type) and AJCC stage (stage I, II, III, IV, unknown)

f

Education levels collapsed as low education (≤high school) and high education (college+); nSES levels collapsed as low nSES (Q1–Q3) and high nSES (Q4–Q5)

We observed similar patterns for BC-specific mortality. Compared to NL White women with high-education/high-nSES, nearly all groups of African American women (except for those with high-education/high-nSES) had higher BC mortality in base models; Latina women with low-education/high-nSES (HR 0.62, 95 % CI 0.44–0.89) had lower BC mortality; and no statistically significant associations were observed for Asian American women. For African American women with low-education/low-nSES, the association was no longer observed in the fully adjusted model (see model 3 in Table 4 which shows the association was not observed after adjusting for comorbidities). Compared to NL White women with high-education/high-nSES, African American women with high-education/low-nSES and African American women with low-education/high-nSES had slightly attenuated associations of higher mortality (HR 1.37, 95 % CI 1.07–1.75 and HR 1.55, 95 % CI 1.01–2.37, respectively), and Latina women with low-education/high-nSES had a slightly attenuated association of lower mortality (HR 0.68, 95 % CI 0.47–0.98) in fully adjusted models.

Discussion

With data on 9372 BC cases, we documented disparities in all-cause and BC-specific mortality accounting for the complex interplay between race/ethnicity, education, and nSES. To our knowledge, no prior study has examined these associations with mortality after BC diagnosis in such a large, diverse group of women with BC.

When simultaneously measuring multiple levels of SES (education, nSES), and race/ethnicity within a single social status variable, we found that disparities existed within and across racial/ethnic groups. One strength of this approach, rather than the stratified approaches, is that comparisons can be made across racial/ethnic and SES groups. We also observed that prognostic factors explained some of the observed disparities in race/ethnicity and SES; however, after adjusting for the full set of prognostic factors, we continued to observe disparities in mortality by race/ethnicity and SES. For all-cause mortality, compared with NL White women with high education and high nSES, NL White and African American women with high education and low nSES had higher mortality, while Latina women with low education and high nSES was the only group to have lower mortality.

Our findings in NL White and African American women for all-cause mortality and in African American women for BC-specific mortality are consistent with prior studies that found higher mortality among women residing in lower SES neighborhoods [9–11, 13, 14]. Furthermore, we observed mortality disparities among groups discordant on their individual- and neighborhood-level SES: NL White and African American women of high education in low SES neighborhoods for all-cause mortality, and African American women of high education in low SES neighborhoods for BC mortality. It has been suggested that discordant individual- and neighborhood-level SES measures may result in worse health through relative deprivation (i.e., those with low education having fewer resources to navigate their high SES neighborhoods which may include higher living costs) or relative standing (i.e., those with low education may have fewer social resources, higher stress, and different coping mechanisms compared to their counterparts in high SES neighborhoods) [36].

In contrast, Latina women with low education in high SES neighborhoods had lower mortality than NL White women with high education and high nSES for both all-cause and BC-specific mortality and reduced mortality did not disappear with adjustment for other prognostic factors. To our knowledge this finding has not been reported previously and was unexpected and warrants confirmation. In our study, the proportion of women who were lost to follow-up differed somewhat across racial/ethnic groups. However, this is unlikely to explain the lower mortality among Latina women as the percentages of women whose date of last follow-up was more than 2 years ago were 1.2 % among NL White women, 2.5 % among African American women, 3.0 % among Latina women, and 4.1 % among Asian American women.

While we did not observe statistically significant associations for Asian American women in our study, prior work has shown significant associations with heterogeneous associations across specific Asian American subgroups [27, 37]. Aggregating Asian American women into a single group may mask these associations.

Applying the intersectional approach, to jointly examine the impact of race/ethnicity, education and nSES, yielded more informative results than the traditional race/ethnicity-stratified approach that assesses independent effects of these SES factors (see Supplemental Table 1). With stratified analyses, we observed no associations for education and mortality after BC diagnosis, and we observed opposite nSES associations for White and African American women.

Studies that have examined the impact of both individual- and neighborhood-level SES on BC survival have found only nSES [8, 9], only individual-level SES [7], both measures [10], or the interactions between the two measures [11, 12] to be associated with mortality. These mixed findings may be due, in part, to the variation across studies in racial/ethnic composition of the study population, as prior studies had limited racial/ethnic diversity, often including NL White and/or African American women only [7, 9, 10]. For example, in a population-based cohort of primarily NL White women from Wisconsin, no associations were observed for individual-level education and income; nSES (census tract-level education) was associated with overall and BC-specific mortality after adjustment for individual-level education and income, and established prognostic factors [9].

Our finding that African American women have higher mortality in low SES neighborhoods regardless of their education warrants further investigation of specific neighborhood factors: these include social, built, and environmental attributes, and how residents within those neighborhoods use and are impacted by their neighborhoods. This line of research can better inform strategies to effectively reduce social inequalities in mortality after BC diagnosis.

While this study has several strengths, there are a few limitations. First, we only had one measure of individual SES, education. Second, we defined neighborhoods using administrative boundaries of census block groups (representing on average 1500 residents) which may not reflect how participants define their neighborhoods. However, this is the smallest level of geography for which rich SES data are available, and census block groups are more homogeneous and better represent neighborhoods where individuals reside and practice healthy behaviors, access services and receive health care than larger geographic areas (e.g., census tracts, zip codes, counties) [25]. Second, for heterogeneous racial/ethnic groups such as the Asian American and Latina groups, subgroup differences may confound or modify associations; unfortunately, our sample did not have sufficient statistical power to examine more refined subgroups. We did not have data on length of residency and whether women moved between date of diagnosis and death or censoring date, which may result in some misclassification of nSES. While we had clinical characteristics, we did not have data on BC subtypes beyond ER/PR status, however, this literature has predominantly shown that black-white disparities in BC persist even after accounting for subtype [38, 39]. Lastly, CCR data on treatment are limited to first course of treatment and may lack meaningful detail, yet, our recent work comparing Medicare claims to registry treatment data shows that registry treatment data are relatively complete and percentages of missing data are similar across racial/ ethnic groups [40, 41].

In conclusion, our analysis demonstrates that associations between two different measures of SES—education and nSES—and mortality after BC diagnosis vary across racial/ethnic groups. In addition, we found that the intersectional approach offers insight to understanding inequalities by multiple social determinants of health, including the adverse outcomes experienced by NL White and African American women with discordant individual-and neighborhood-level SES. Our results point to the need to understand the modifiable features of low SES neighborhoods such as higher crime, low walkability, poor food environment, low collective efficacy and low social cohesion that contribute to worse survival, especially for African American women who continue to have higher all-cause and BC-specific mortality.

Supplementary Material

1

Acknowledgments

We are grateful to all the study participants for their contributions in the five California-based studies. The Asian American Breast Cancer Study was supported by the California Breast Research Program (CBCRP) grants 1RB-0287, 3PB-0120, and 5PB-0018. The San Francisco Bay Area Breast Cancer Study was supported by National Cancer Institute grants R01 CA63446 and R01 CA77305, by the U.S. Department of Defense (DOD) grant DAMD17-96-1-6071, and by the CBCRP grants 4JB-1106 and 7PB-0068. The Women’s CARE Study was funded by the National Institute of Child Health and Human Development (NICHD), through a contract with USC (N01-HD-3-3175); and the California Teachers Study was funded by the California Breast Cancer Act of 1993, National Cancer Institute grants (R01 CA77398 and K05 CA136967 to LB), and the California Breast Cancer Research Fund (contract 97-10500). The Multiethnic Cohort Study was supported by National Cancer Institute grants R01 CA54281, R37CA54281, and UM1 CA164973. Clinical and tumor characteristics and mortality data were obtained from the California Cancer Registry (CCR). The collection of cancer incidence data used in this study was supported by the California Department of Health Services as part of the statewide cancer reporting program mandated by California Health and Safety Code Section 103885; the National Cancer Institute’s Surveillance, Epidemiology, and End Results Program under contract HHSN261201000140C awarded to the Cancer Prevention Institute of California, contract HHSN26120100035C awarded to the University of Southern California, and contract HHSN26120100034C awarded to the Public Health Institute; and the Centers for Disease Control and Prevention’s National Program of Cancer Registries, under agreement #1U58 DP000807-01 awarded to the Public Health Institute. The ideas and opinions expressed herein are those of the authors, and endorsement by the State of California, the California Department of Health Services, the National Cancer Institute, or the Centers for Disease Control and Prevention or their contractors and subcontractors is not intended nor should be inferred. This work was supported by grants from the California Breast Cancer Research Program: 16ZB-8001(USC, Wu), 16ZB-8002 (CPIC, Gomez), 16ZB-8003 (COH, Bernstein), 16ZB-8004 (KPNC, Kwan), 16ZB-8005 (USC, Monroe).

Footnotes

Conflict of interest Dr. Scarlett Gomez reports receiving funding support from Genentech unrelated to this manuscript. All other authors declare that they have no conflict of interest.

Electronic supplementary material The online version of this article (doi:10.1007/s10900-015-0052-y) contains supplementary material, which is available to authorized users.

Compliance with Ethical Standards The protocols for the CBCSC study were approved by the institutional review boards (IRBs) at all participating institutions and the California state IRB (Committee for the Protection of Human Subjects). Informed consent was obtained from all individual participants included in the study.

References

  • 1.Albano JD, Ward E, Jemal A, et al. Cancer mortality in the United States by education level and race. Journal of the National Cancer Institute. 2007;99(18):1384–1394. doi: 10.1093/jnci/djm127. [DOI] [PubMed] [Google Scholar]
  • 2.Wu AH, Gomez SL, Vigen C, et al. The California Breast Cancer Survivorship Consortium (CBCSC): prognostic factors associated with racial/ethnic differences in breast cancer survival. Cancer Causes and Control. 2013;24(10):1821–1836. doi: 10.1007/s10552-013-0260-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.McKenzie F, Jeffreys M. Do lifestyle or social factors explain ethnic/racial inequalities in breast cancer survival? Epidemiologic Reviews. 2009;31:52–66. doi: 10.1093/epirev/mxp007. [DOI] [PubMed] [Google Scholar]
  • 4.Bassett MT, Krieger N. Social class and black–white differences in breast cancer survival. American Journal of Public Health. 1986;76(12):1400–1403. doi: 10.2105/ajph.76.12.1400. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Byers TE, Wolf HJ, Bauer KR, et al. The impact of socioeconomic status on survival after cancer in the United States: Findings from the National Program of Cancer Registries Patterns of Care Study. Cancer. 2008;113(3):582–591. doi: 10.1002/cncr.23567. [DOI] [PubMed] [Google Scholar]
  • 6.Yao L, Robert SA. Examining the racial crossover in mortality between African American and white older adults: A multilevel survival analysis of race, individual socioeconomic status, and neighborhood socioeconomic context. Journal of Aging Research. 2011;2011:132073. doi: 10.4061/2011/132073. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Steenland K, Henley J, Calle E, Thun M. Individual- and area-level socioeconomic status variables as predictors of mortality in a cohort of 179,383 persons. American Journal of Epidemiology. 2004;159(11):1047–1056. doi: 10.1093/aje/kwh129. [DOI] [PubMed] [Google Scholar]
  • 8.Major JM, Doubeni CA, Freedman ND, et al. Neighborhood socioeconomic deprivation and mortality: NIH-AARP diet and health study. PLoS One. 2010;5(11):e15538. doi: 10.1371/journal.pone.0015538. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Sprague BL, Trentham-Dietz A, Gangnon RE, et al. Socioeconomic status and survival after an invasive breast cancer diagnosis. Cancer. 2011;117(7):1542–1551. doi: 10.1002/cncr.25589. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Dasgupta P, Baade PD, Aitken JF, Turrell G. Multilevel determinants of breast cancer survival: Association with geographic remoteness and area-level socioeconomic disadvantage. Breast Cancer Research and Treatment. 2012;132(2):701–710. doi: 10.1007/s10549-011-1899-y. [DOI] [PubMed] [Google Scholar]
  • 11.Bentley R, Kavanagh AM, Subramanian SV, Turrell G. Area disadvantage, individual socio-economic position, and premature cancer mortality in Australia 1998 to 2000: A multilevel analysis. Cancer Causes and Control. 2008;19(2):183–193. doi: 10.1007/s10552-007-9084-7. [DOI] [PubMed] [Google Scholar]
  • 12.Shariff-Marco S, Yang J, John EM, et al. Impact of neighborhood and individual socioeconomic status on survival after breast cancer varies by race/ethnicity: The Neighborhood and Breast Cancer Study. Cancer Epidemiology Biomarkers & Prevention. 2014;23(5):793–811. doi: 10.1158/1055-9965.EPI-13-0924. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Klassen AC, Smith KC. The enduring and evolving relationship between social class and breast cancer burden: a review of the literature. Cancer Epidemiology. 2011;35(3):217–234. doi: 10.1016/j.canep.2011.02.009. [DOI] [PubMed] [Google Scholar]
  • 14.Vona-Davis L, Rose DP. The influence of socioeconomic disparities on breast cancer tumor biology and prognosis: A review. Journal of Women’s Health (Larchmt) 2009;18(6):883–893. doi: 10.1089/jwh.2008.1127. [DOI] [PubMed] [Google Scholar]
  • 15.Robert SA, Strombom I, Trentham-Dietz A, et al. Socioeconomic risk factors for breast cancer: Distinguishing individual- and community-level effects. Epidemiology. 2004;15(4):442–450. doi: 10.1097/01.ede.0000129512.61698.03. [DOI] [PubMed] [Google Scholar]
  • 16.Meijer M, Rohl J, Bloomfield K, Grittner U. Do neighborhoods affect individual mortality? A systematic review and meta-analysis of multilevel studies. Social Science and Medicine. 2012;74(8):1204–1212. doi: 10.1016/j.socscimed.2011.11.034. [DOI] [PubMed] [Google Scholar]
  • 17.Braveman PA, Cubbin C, Egerter S, et al. Socioeconomic status in health research: one size does not fit all. JAMA. 2005;294(22):2879–2888. doi: 10.1001/jama.294.22.2879. [DOI] [PubMed] [Google Scholar]
  • 18.Williams DR, Sternthal M. Understanding racial-ethnic disparities in health: Sociological contributions. Journal of Health and Social Behavior. 2010;51(Suppl):S15–S27. doi: 10.1177/0022146510383838. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Williams DR, Kontos EZ, Viswanath K, et al. Integrating multiple social statuses in health disparities research: The case of lung cancer. Health Services Research. 2012;47(3 Pt 2):1255–1277. doi: 10.1111/j.1475-6773.2012.01404.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Kawachi I, Daniels N, Robinson DE. Health disparities by race and class: Why both matter. Health Affairs (Millwood) 2005;24(2):343–352. doi: 10.1377/hlthaff.24.2.343. [DOI] [PubMed] [Google Scholar]
  • 21.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 Epidemiology Biomarkers & Prevention. 2009;18(1):121–131. doi: 10.1158/1055-9965.EPI-08-0679. [DOI] [PubMed] [Google Scholar]
  • 22.Kwan ML, John EM, Caan BJ, et al. Obesity and mortality after breast cancer by race/ethnicity: The california breast cancer survivorship consortium. American Journal of Epidemiology. 2014;179(1):95–111. doi: 10.1093/aje/kwt233. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Liu L, Cozen W, Bernstein L, Ross RK, Deapen D. Changing relationship between socioeconomic status and prostate cancer incidence. Journal of the National Cancer Institute. 2001;93(9):705–709. doi: 10.1093/jnci/93.9.705. [DOI] [PubMed] [Google Scholar]
  • 24.Yost K, Perkins C, Cohen R, Morris C, Wright W. Socioeconomic status and breast cancer incidence in California for different race/ethnic groups. Cancer Causes and Control. 2001;12(8):703–711. doi: 10.1023/a:1011240019516. [DOI] [PubMed] [Google Scholar]
  • 25.Krieger N, Chen JT, Waterman PD, Soobader M, Subramanian SV, Carson R. Geocoding and monitoring of US socioeconomic inequalities in mortality and cancer incidence: Does the choice of area-based measure and geographic level matter? The Public Health Disparities Geocoding Project. American Journal of Epidemiology. 2002;156(5):471–482. doi: 10.1093/aje/kwf068. [DOI] [PubMed] [Google Scholar]
  • 26.Clarke CA, Miller T, Chang ET, Yin D, Cockburn M, Gomez SL. Racial and social class gradients in life expectancy in contemporary California. Social Science and Medicine. 2010;70(9):1373–1380. doi: 10.1016/j.socscimed.2010.01.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Gomez SL, Clarke CA, Shema SJ, Chang ET, Keegan TH, Glaser SL. Disparities in breast cancer survival among Asian women by ethnicity and immigrant status: A population-based study. American Journal of Public Health. 2010;100(5):861–869. doi: 10.2105/AJPH.2009.176651. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Gomez SL, Press DJ, Lichtensztajn D, et al. Patient, hospital, and neighborhood factors associated with treatment of early-stage breast cancer among Asian American women in California. Cancer Epidemiology Biomarkers & Prevention. 2012;21(5):821–834. doi: 10.1158/1055-9965.EPI-11-1143. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Keegan TH, John EM, Fish KM, Alfaro-Velcamp T, Clarke CA, Gomez SL. Breast cancer incidence patterns among California Hispanic women: Differences by nativity and residence in an enclave. Cancer Epidemiology Biomarkers & Prevention. 2010;19(5):1208–1218. doi: 10.1158/1055-9965.EPI-10-0021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Keegan TH, Quach T, Shema S, Glaser SL, Gomez SL. The influence of nativity and neighborhoods on breast cancer stage at diagnosis and survival among California Hispanic women. BMC Cancer. 2010;10:603. doi: 10.1186/1471-2407-10-603. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Livaudais JC, Hershman DL, Habel L, et al. Racial/ ethnic differences in initiation of adjuvant hormonal therapy among women with hormone receptor-positive breast cancer. Breast Cancer Research and Treatment. 2012;131(2):607–617. doi: 10.1007/s10549-011-1762-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Telli ML, Chang ET, Kurian AW, et al. Asian ethnicity and breast cancer subtypes: A study from the California Cancer Registry. Breast Cancer Research and Treatment. 2011;127(2):471–478. doi: 10.1007/s10549-010-1173-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.International Classification of Diseases. Ninth revision. Geneva, Switzerland: World Health Organization; 1980. [Google Scholar]
  • 34.International Classification of Diseases. 10th revision. Geneva, Switzerland: World Health Organization; 1992. [Google Scholar]
  • 35.Lin DY, Wei LJ. The robust inference for the Cox proportional hazards model. Journal of the American Statistical Association. 1989;84(408):1074–1078. [Google Scholar]
  • 36.Winkleby M, Cubbin C, Ahn D. Effect of cross-level interaction between individual and neighborhood socioeconomic status on adult mortality rates. American Journal of Public Health. 2006;96(12):2145–2153. doi: 10.2105/AJPH.2004.060970. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Li CI, Malone KE, Daling JR. Differences in breast cancer stage, treatment and survival by race and ethnicity. Archives of Internal Medicine. 2003;163:49–56. doi: 10.1001/archinte.163.1.49. [DOI] [PubMed] [Google Scholar]
  • 38.Keegan TH, Press DJ, Tao L, et al. Impact of breast cancer subtypes on 3-year survival among adolescent and young adult women. Breast Cancer Research. 2013;15(5):R95. doi: 10.1186/bcr3556. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Kroenke CH, Sweeney C, Kwan ML, et al. Race and breast cancer survival by intrinsic subtype based on PAM50 gene expression. Breast Cancer Research and Treatment. 2014;144(3):689–699. doi: 10.1007/s10549-014-2899-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Cooper GS, Yuan Z, Stange KC, Dennis LK, Amini SB, Rimm AA. Agreement of Medicare claims and tumor registry data for assessment of cancer-related treatment. Medical Care. 2000;38(4):411–421. doi: 10.1097/00005650-200004000-00008. [DOI] [PubMed] [Google Scholar]
  • 41.Cooper GS, Virnig B, Klabunde CN, Schussler N, Free-man J, Warren JL. Use of SEER-Medicare data for measuring cancer surgery. Medical Care. 2002;40(8 Suppl):IV-43–8. doi: 10.1097/00005650-200208001-00006. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS700445-supplement-1.doc (82KB, doc)

RESOURCES