Impact of Neighborhood and Individual Socioeconomic Status on Survival after Breast Cancer Varies by Race/Ethnicity: The Neighborhood and Breast Cancer Study

• ↵^aTotals include patients with unknown education, and therefore row numbers may not add up to the total. Note that those with unknown education were excluded from the analytic sample.

• ↵^bNeighborhood SES was measured using a composite measure of 7 Census indicator measures known as the YOST SES Index (29).

NOTE: Bold font refers to statistically significant associations.

• ↵^aAdjusted for age at diagnosis (continuous), year of diagnosis (continuous), study eligibility [Northern California site of the Breast Cancer Family Registry (NC-BCFR) high risk, NC-BCFR sporadic, SFBCS], nodal involvement (no, yes, unknown), histology (ductal, lobular, other), histologic grade (1–4, unknown), joint ERPR status (ER⁻PR⁻, ER⁺, or PR⁺, unknown), type of surgery (none, lumpectomy, mastectomy, not otherwise specified, unknown), radiation (no, yes, unknown), chemotherapy (no, yes, unknown), subsequent primary tumor (yes, no) and clustering by block group, and stratified by AJCC stage (I–IV, unknown).

• ↵^bAdjusted for covariates of model 1 and clustering by block group (individual education and neighborhood SES in the same model), and stratified by AJCC stage (I–IV, unknown).

• ↵^cEducation levels <high school graduate and high school graduate collapsed as for non-Hispanic whites and Asian-Americans because of small sample sizes.

• ↵^dNeighborhood SES quintiles 1 and 2 were collapsed for non-Hispanic whites and Asian-Americans because of small sample sizes.

• ↵^eNeighborhood SES was measured using a composite measure of 7 Census indicator measures known as the YOST SES Index (29).

• ↵^fHS, high school. Education levels (<high school graduate and high school graduate) collapsed as ≤ high school graduate. Neighborhood SES levels collapsed as Q1 to Q3: low SES; Q4 to Q5: high SES.

NOTE: Bold font refers to statistically significant associations.

• ↵^aAdjusted for age at diagnosis (continuous), year of diagnosis (continuous), study eligibility (NC-BCFR high risk, NC-BCFR sporadic, SFBCS), histology (ductal, lobular, other), histologic grade (1–4, unknown), joint ERPR status (ER⁻PR⁻, ER⁺, or PR⁺, unknown), nodal involvement (none, positive, unknown), type of surgery (none, lumpectomy, mastectomy, NOS, unknown), radiation (no, yes, unknown), chemotherapy (no, yes, unknown), first subsequent primary tumor (yes, no), second subsequent primary tumor (yes, no), days between the dates of diagnosis of study qualifying tumor and the first subsequent tumor (continuous), days between the dates of diagnosis of the first and second subsequent tumor (continuous), benign breast disease (no, yes, unknown), years since last full-term pregnancy (≤2, 2–4, 5±, unknown), prediagnosis hormonal contraception use (never, ever, unknown), prediagnosis hormone therapy use (never, past, recent, unknown) and clustering by block group, and stratified by AJCC stage (I– IV, unknown).

• ↵^bAdjusted for age at diagnosis (continuous), study eligibility (NC-BCFR high risk, NC-BCFR sporadic, SFBCS), histology (ductal, lobular, other), histologic grade (1–4, unknown), joint ERPR status (ER⁻PR⁻, ER⁺, or PR⁺, unknown), nodal involvement (none, positive, unknown), type of surgery (none, lumpectomy, mastectomy, NOS, unknown), radiation (no, yes, unknown), chemotherapy (no, yes, unknown), first subsequent primary tumor (yes, no), second subsequent primary tumor (yes, no), days between the dates of diagnosis of study qualifying tumor and the first subsequent tumor (continuous), days between the dates of diagnosis of the first and second subsequent tumor (continuous), marital status (single/never married, married, separated/divorced, widowed, unknown) and clustering by block group, and stratified by AJCC stage (I–IV, unknown).

• ↵^cAdjusted for age at diagnosis (continuous), study eligibility (NC-BCFR high risk, NC-BCFR sporadic, SFBCS), histology (ductal, lobular, other), histologic grade (1–4, unknown), joint ERPR status (ER⁻PR⁻, ER⁺, or PR⁺, unknown), nodal involvement (none, positive, unknown), type of surgery (none, lumpectomy, mastectomy, NOS, unknown), radiation (no, yes, unknown), chemotherapy (no, yes, unknown), first subsequent primary tumor (yes, no), second subsequent primary tumor (yes, no), days between the dates of diagnosis of study qualifying tumor and the first subsequent tumor (continuous), days between the dates of diagnosis of the first and second subsequent tumor (continuous), grams per day of alcohol in reference year (0, ≤5, 5–9, 10–14, 15±, unknown), prediagnosis BMI (≤25, 25–29, 30±, unknown), recent recreational physical activity (0, Q1/Q2, Q3/Q4, unknown) and clustering by block group, and stratified by AJCC stage (I–IV, unknown).

• ↵^dAdjusted for age at diagnosis (continuous), study eligibility (NC-BCFR high risk, NC-BCFR sporadic, SFBCS), histology (ductal, lobular, other), histologic grade (1–4, unknown), joint ERPR status (ER⁻PR⁻, ER⁺, or PR⁺, unknown), nodal involvement (none, positive, unknown), type of surgery (none, lumpectomy, mastectomy, NOS, unknown), radiation (no, yes, unknown), chemotherapy (no, yes, unknown), first subsequent primary tumor (yes, no), second subsequent primary tumor (yes, no), days between the dates of diagnosis of study qualifying tumor and the first subsequent tumor (continuous), days between the dates of diagnosis of the first and second subsequent tumor (continuous), percent of white patients with cancer in reporting hospital (≤25%, 25–49%, 50–74%, ≥75%, unknown), percent of patients with cancer in highest SES quintile in reporting hospital (≤25%, 25–49%, 50–74%, ≥75%, unknown) and clustering by block group, and stratified by AJCC stage (I–IV, unknown).

• ↵^eEducation levels (<high school graduate and high school graduate) collapsed as ≤ high school graduate.

• ↵^fNeighborhood SES was measured using a composite measure of 7 Census indicator measures known as the Yost SES Index (29).

• ↵^gSES levels collapsed as Q1 to Q3: low SES; Q4 to Q5: high SES.

NOTE: Bold font refers to statistically significant associations.

• ↵^aAdjusted for age at diagnosis (continuous), year of diagnosis (continuous), study eligibility (FRBC high risk, case–control, FRBC sporadic), nodal involvement (no, yes, unknown), histology (ductal, lobular, other), histologic grade (1–4, unknown), joint ERPR status (ER⁻PR⁻, ER⁺, or PR⁺, unknown), type of surgery (none, lumpectomy, mastectomy, NOS, unknown), radiation (no, yes, unknown), chemotherapy (no, yes, unknown), subsequent primary tumor (yes, no) and clustering by block group, and stratified by AJCC stage (I–IV, unknown).

• ↵^bAdjusted for covariates of model 1 and clustering by block group (individual education and neighborhood SES in the same model), and stratified by AJCC stage (I–IV, unknown).

• ↵^cEducation levels (<high school graduate and high school graduate) collapsed as ≤ high school graduate for non-Hispanic whites and Asian-Americans because of small sample sizes.

• ↵^dNeighborhood SES quintiles 1 and 2 were collapsed for non-Hispanic whites and Asian-Americans because of small sample sizes.

• ↵^eNeighborhood SES was measured using a composite measure of 7 Census indicator measures known as the YOST SES Index (29).

• ↵^fHS, high school. Education levels (<high school graduate and high school graduate) collapsed as ≤ high school graduate. Neighborhood SES levels collapsed as Q1 to Q3: low SES; Q4 to Q5: high SES.