Abstract
PURPOSE
Adverse neighborhood contextual factors may affect breast cancer outcomes through environmental, psychosocial, and biological pathways. The objective of this study is to examine the relationship between allostatic load (AL), neighborhood opportunity, and all-cause mortality among patients with breast cancer.
METHODS
Women age 18 years and older with newly diagnosed stage I-III breast cancer who received surgical treatment between January 1, 2012, and December 31, 2020, at a National Cancer Institute Comprehensive Cancer Center were identified. Neighborhood opportunity was operationalized using the 2014-2018 Ohio Opportunity Index (OOI), a composite measure derived from neighborhood level transportation, education, employment, health, housing, crime, and environment. Logistic and Cox regression models tested associations between the OOI, AL, and all-cause mortality.
RESULTS
The study cohort included 4,089 patients. Residence in neighborhoods with low OOI was associated with high AL (adjusted odds ratio, 1.21 [95% CI, 1.05 to 1.40]). On adjusted analysis, low OOI was associated with greater risk of all-cause mortality (adjusted hazard ratio [aHR], 1.45 [95% CI, 1.11 to 1.89]). Relative to the highest (99th percentile) level of opportunity, risk of all-cause mortality steeply increased up to the 70th percentile, at which point the rate of increase plateaued. There was no interaction between the composite OOI and AL on all-cause mortality (P = .12). However, there was a higher mortality risk among patients with high AL residing in lower-opportunity environments (aHR, 1.96), but not in higher-opportunity environments (aHR, 1.02; P interaction = .02).
CONCLUSION
Lower neighborhood opportunity was associated with higher AL and greater risk of all-cause mortality among patients with breast cancer. Additionally, environmental factors and AL interacted to influence all-cause mortality. Future studies should focus on interventions at the neighborhood and individual level to address socioeconomically based disparities in breast cancer.
INTRODUCTION
Women with breast cancer who identify with socially and economically marginalized groups in the United States (ie, low socioeconomic status [SES]) experience higher mortality rates compared with their privileged counterparts (eg, high SES).1,2 Although most research evaluates individual-level factors (eg, insurance status) and clinical characteristics (eg, stage), substantial evidence has suggested neighborhood contextual factors (eg, neighborhood deprivation) also contribute to stage at diagnosis, breast cancer subtypes, and mortality.3,4 The relationship between neighborhood contextual factors and the development and progression of breast cancer may occur through environmental, psychological, social, and biological pathways.5-9 Few studies, however, have examined the relationship between neighborhood contextual factors and biological correlates of stress, operationalized as allostatic load (AL), on cancer outcomes.
CONTEXT
Key Objective
What is the association between allostatic load (AL), neighborhood opportunity, and all-cause mortality in patients with breast cancer?
Knowledge Generated
Patients with breast cancer residing in neighborhoods with low opportunity have a higher AL and worse all-cause mortality compared with those in areas with higher opportunity. Additionally, interactions between AL and the environment domain of the opportunity index suggests patients with high AL living in unfavorable environments (eg, poor walkability and increased pollution) experience a greater risk of all-cause mortality in comparison with their counterparts living in more favorable environments.
Relevance (I. Cheng)
-
The identification of associations of neighborhood opportunity with AL and all-cause mortality among patients with breast cancer points to the importance of the influence of contextual and environmental factors on stress embodiment and breast cancer outcomes.*
*Relevance section written by JCO Associate Editor Iona Cheng, PhD, MPH.
AL, a measure of physiologic wear and tear secondary to perceptions and cognitive emotional responses to social and environmental stimuli, provides a means to explore pathways between neighborhood contextual factors and clinical outcomes.10 In their seminal work on AL, McEwen and Stellar hypothesize that AL measures chronic activation of the hypothalamic-pituitary-adrenal axis and the sympathetic adrenal medullary pathway.11 AL is composed of primary mediators (eg, cortisol) that exert effects on tissues leading to secondary multisystem dysregulation (eg, elevated blood glucose) and tertiary manifestations of disease (eg, diabetes).12 Elevated AL, a marker of worsening multisystem physiologic dysregulation, has been associated with adverse socioenvironmental stressors such as residence in racially segregated and/or low-income neighborhoods.10,13,14 In fact, higher AL has been reported among Black breast cancer survivors, those living in neighborhoods with greater deprivation, larger tumor sizes, and estrogen (ER) receptor–negative breast cancer.15-17
Few studies have examined the association between AL, neighborhood contextual factors, and mortality among patients with breast cancer. Furthermore, no studies have examined the relationship between neighborhood opportunity—a neighborhood contextual factor—and breast cancer outcomes. Neighborhood opportunity is conceptualized as local conditions and resources that promote healthy development.18 In contrast to neighborhood deprivation, where solutions are often external to the community, neighborhood opportunity reframes the discussion focus to what existing assets can be mobilized within the community to enable and empower members to take greater control over their own health.19,20 This more asset-based approach provides stakeholders and researchers new ways of engaging with the community to improve their health and development.20,21 Although several indices have been created to measure deprivation or opportunity, the Ohio Opportunity Index (OOI) is the first Ohio state–specific multidimensional index encompassing more and higher-resolution variables across multiple domains.21,22
This study aims to (1) determine the association between the composite OOI, OOI domains (transportation, education, employment, health, housing, crime, and environment), and AL in patients with breast cancer, (2) evaluate the association between the composite OOI, OOI domains, and all-cause mortality, and (3) examine the relationship between AL, OOI, and all-cause mortality among patients with breast cancer. We hypothesize that lower neighborhood opportunity will correlate with higher AL and all-cause mortality. Additionally, AL will interact with opportunity and its domains to affect all-cause mortality in patients with breast cancer.
METHODS
Data Source
Women age 18 years and older with newly diagnosed stage I-III breast cancer who received surgical treatment between January 1, 2012, and December 31, 2020, at the Ohio State University Comprehensive Cancer Center were identified in the Ohio State University Cancer Registry and electronic medical record (EMR). Surgical management was considered an inclusion criteria as (1) biomarkers collected to calculate AL are part of the preoperative workup and (2) most patients with stage I-III breast cancer receive surgery.23 Exclusion criteria were patients with noninvasive (stage 0) or metastatic (stage IV) breast cancer, recurrent breast cancer, unknown breast cancer molecular subtypes, or those who did not receive surgical management. Missing cancer stage and molecular subtype were noted in 7.0% of patients, 7.7% had missing OOI data, and 30% had at least one biomarker missing (Data Supplement, Fig S1, online only). Ten imputation data sets were created to address missing values.
Sociodemographic Factors
Sociodemographic factors included age, race (Black, White, other), ethnicity (Hispanic or non-Hispanic), marital status (single, married/living as married, widow, separated or divorced), insurance (private, Medicaid, Medicare, other), smoking history, and alcohol use at the time of diagnosis. Individuals who identified as Asian, American Indian, Alaskan Native, Native Hawaiians, Other Pacific Islander, or multiracial were collapsed into the other racial category because of small sample sizes. Race/ethnicity is considered a social construct in this study rather than a reflection of genetic ancestry.24 The National Cancer Institute's modified Charlson comorbidity index was used to determine comorbidity burden, which excludes cancer.25
Clinical and Treatment Characteristics
Patient stage (clinical and pathologic) and receptor status (ER, progesterone, human epidermal growth factor 2 [ERBB2] expression) were obtained from the EMR. Patients were then classified into breast cancer subtypes: hormone receptor–negative/ERBB2–positive, hormone receptor–positive/ERBB2–negative, hormone receptor–positive/ERBB2–positive, hormone receptor–negative/ERBB2–negative.26 Surgeries included breast (lumpectomy or mastectomy) and axillary (sentinel lymph node biopsy or axillary lymph node dissection) procedures. Postoperative complications were dichotomized into yes or no and are listed in the Data Supplement (Table S1). Systemic (chemotherapy, hormone therapy) and radiation therapy were similarly included.
Study Measures
OOI
The 2014-2018 OOI is an Ohio-specific multidimensional composite index of 34 measures capturing social and economic opportunities organized into seven domains: transportation, education, employment, health, housing, crime, and environment. Specific measures within each domain are further outlined in the Data Supplement (Table S2). Data were compiled from multiple sources as previously discussed.21 Census tracts were used as the geographic unit following previous recommendations.27,28 Each domain score was summed and divided by seven (representing the seven domains) according to previous studies, resulting in the composite opportunity index ranging between 0 (lowest opportunity) and 100 (highest opportunity).21 The cohort's index median value (83.5) was used to characterize high (>median) and low (≤median) opportunity.21
AL
AL is a composite measure using vital signs, anthropometric measures, and clinical laboratory values to assess multiple physiologic systems.29 No gold standard biomarkers currently exist.29 However, multisystem modeling has found that allostatic factor loadings remain similar regardless of the specific biomarkers used, so long as multiple physiologic systems are included.30 Consequently, this study used biomarkers collected 12 months before or 6 months after biopsy-confirmed breast cancer diagnosis. Specifically, 10 biomarkers from four physiologic systems were used: (1) cardiovascular—heart rate, systolic blood pressure (SBP) and diastolic blood pressure (DBP); (2) metabolic—BMI, alkaline phosphatase (ALP), blood glucose, and albumin; (3) immunologic—WBC count; and (4) renal—creatinine and blood urea nitrogen (BUN). Distributions of each biomarker were obtained and biomarkers in the cohort's worst quartile were assigned a point. For example, values ≥75th percentile for heart rate, SBP, DBP, BMI, ALP, glucose, WBC, creatinine, and BUN, and ≤25th percentile for albumin were each given a point. All assigned points were combined into a composite AL score ranging from 0 to 10. Composite scores were then dichotomized into high and low AL using the cohort's median score (2.0) as the cutoff.31
Study Outcome
The primary study outcome was all-cause mortality. Follow-up period was calculated from the date of cancer diagnosis to the date of study event (death, loss to follow-up, or end of follow-up). Participants were presumed to be alive on date of loss to follow-up or the end of follow-up (December 31, 2020), whichever came first.
Statistical Analysis
Missing values were imputed using multiple imputations by chained equations to create 10 imputed data sets.32 All imputation-corrected parameters and standard errors were combined using the Rubin method.33
Overall and opportunity-stratified (ie, low v high on the basis of the cohort's median) characteristics were summarized using descriptive statistics. Differences between patients with low versus high OOI were compared using the Wilcoxon rank sum test for continuous variables and χ2 or Fisher's exact tests for categorical variables. The OOI domains and AL associations were examined using Pearson's correlation coefficients. Crude and adjusted generalized estimating equations (GEE) models for binary outcomes with exchangeable correlation structure were used to assess the association between high AL as the outcome and OOI status as the exposure. The GEE approach accounted for participants' clustering within census tracts in the OOI measurements. Crude and fully adjusted Cox proportional hazard models were used to test the effect of the OOI on the risk of all-cause mortality. Models were adjusted for age, race, ethnicity, health insurance, marital status, history of alcohol or smoking use, the Charlson comorbidity index, molecular subtype, cancer stage, breast and axillary surgery type, systemic or radiation therapy, and surgical complications. The role of high AL status as an effect modifier to the relationship between OOI and risk of all-cause mortality was examined by adding an OOI by AL interaction term in all adjusted models. Robust sandwich covariance matrixes in all Cox models were used to account for the intracluster dependence of participants within census tracts.34 The proportional hazard assumption was tested by adding a time-dependent function of OOI status to the regression models.
The dose-response relationship between the OOI score in its continuous form and the odds of high AL were evaluated using a three-knot restricted cubic spline in the adjusted GEE models. The three knots were placed at the OOI sum score's 10th, 50th, and 90th percentiles.35 The same approach examined the dose-response relationship between OOI and the risk of all-cause mortality. Wald chi-square tests assessed overall and nonlinear associations between the OOI score percentiles and odds of high AL.
Post hoc exploratory causal mediation analyses (see Data Supplement, Methods) were conducted to assess the role of AL as a potential mediator in the relationship between OOI and all-cause mortality (Data Supplement, Fig S2).36 In addition, we conducted a sensitivity causal mediation analysis considering the separate and joint effects of AL and CCI as multiple mediators in the relationship between the OOI and all-cause mortality (Data Supplement, Fig S3).37-39 All analyses were performed using SAS 9.4 software (SAS Institute, Cary, NC). The Ohio State University Office of Responsible Research Practices approved this study's institutional review board (2021C0114).
RESULTS
Participant Characteristics
After multiple imputations of patients with missing values were identified, 4,089 patients with breast cancer were included in the study (Table 1). Patients with lower opportunity scores were more likely to be racialized as non-Hispanic Black (low OOI 13.4% v high OOI 4%; P < .0001), identify as single (low OOI 16.54% v high OOI 11.5%; P < .0001), have Medicaid insurance (low OOI 12.3% v high OOI 5.2%; P < .0001), and have a smoking history but no previous alcohol use (P < .0001). There were no significant differences between groups on breast (P = .80) or axillary (P = .71) surgery type, rates of surgical complications (P = .66), or receipt of hormone therapy (P = .19), chemotherapy (P = .80), or radiation therapy (P = .44). Descriptive characteristics of the opportunity index domains by AL status were examined (Data Supplement, Table S3). Evaluations of AL and OOI showed patients with lower opportunity scores were more likely to have high composite AL scores (low OOI 53.6% v high OOI 45.9%; P < .0001; Data Supplement, Table S4).
TABLE 1.
Overview of Sociodemographic and Clinical Characteristics by Opportunity Status
| Patient Characteristic | All (N = 4,089) | Low Opportunity (n = 2,037) | High Opportunity (n = 2,052) | P a |
|---|---|---|---|---|
| Age, years | .87 | |||
| Mean (SD) | 58.2 (12.5) | 58.3 (12.7) | 58.2 (12.3) | |
| Median (Q1-Q3) | 59.0 (49.0-67.0) | 59.0 (49.0-69.0) | 59.0 (50.0-67.0) | |
| Age group, No. (%) | .07 | |||
| ≤39 | 290 (7.1) | 138 (6.8) | 152 (7.4) | |
| 40-49 | 771 (18.9) | 368 (18.1) | 403 (19.6) | |
| 50-59 | 1,076 (26.3) | 576 (28.3) | 500 (24.4) | |
| 60-59 | 1,171 (28.6) | 578 (28.4) | 593 (28.9) | |
| ≥70 | 781 (19.1) | 377 (18.5) | 404 (19.7) | |
| Race/ethnicity, No. (%) | <.0001 | |||
| Hispanic Black | 3 (0.1) | 2 (0.1) | 1 (0.05) | |
| Non-Hispanic Black | 355 (8.7) | 272 (13.4) | 83 (4.0) | |
| Hispanic White | 22 (0.5) | 7 (0.34) | 15 (0.73) | |
| Non-Hispanic White | 3,530 (86.3) | 1,705 (83.7) | 1,825 (88.9) | |
| Hispanic other | 25 (0.6) | 11 (0.54) | 14 (0.68) | |
| Non-Hispanic other | 154 (3.8) | 40 (1.96) | 114 (5.6) | |
| Marital status, No. (%) | <.0001 | |||
| Single | 574 (14.0) | 337 (16.54) | 237 (11.5) | |
| Married/living as married | 2,609 (63.8) | 1,220 (59.89) | 1,389 (67.7) | |
| Widowed, separated, or divorced | 906 (22.2) | 480 (23.56) | 426 (20.8) | |
| Health insurance, No. (%) | <.0001 | |||
| Managed care | 2,408 (58.9) | 1,098 (53.9) | 1,310 (63.8) | |
| Medicare | 1,258 (30.8) | 655 (32.2) | 603 (29.4) | |
| Medicaid | 357 (8.7) | 250 (12.3) | 107 (5.2) | |
| Other | 66 (1.6) | 34 (1.7) | 32 (1.6) | |
| Smoking history, No. (%) | <.0001 | |||
| Never | 2,549 (62.3) | 1,197 (58.8) | 1,351 (65.8) | |
| Current or former | 1,540 (37.7) | 840 (41.2) | 701 (34.2) | |
| Alcohol use, No. (%) | <.0001 | |||
| Never | 1,914 (46.8) | 1,074 (52.7) | 840 (40.9) | |
| Current or former | 2,175 (53.2) | 963 (47.3) | 1,212 (59.1) | |
| ERBB2 summary, No. (%) | .18 | |||
| Negative | 3,446 (84.3) | 1,732 (85.0) | 1,714 (83.5) | |
| Positive | 643 (15.7) | 305 (15.0) | 338 (16.5) | |
| Progesterone summary, No. (%) | .25 | |||
| Negative | 1,259 (30.8) | 644 (31.6) | 615 (30) | |
| Positive | 2,830 (69.2) | 1,393 (68.4) | 1,437 (70) | |
| Estrogen summary, No. (%) | .04 | |||
| Negative | 830 (20.3) | 440 (21.6) | 390 (19.01) | |
| Positive | 3,259 (79.7) | 1,597 (78.4) | 1,662 (80.99) | |
| Molecular subtype,b No. (%) | .02 | |||
| Hormone receptor–negative/ERBB2–positive | 228 (5.6) | 108 (5.3) | 120 (5.8) | |
| Hormone receptor–positive/ERBB2–negative | 2,507 (61.3) | 1,243 (61) | 1,264 (61.6) | |
| Hormone receptor–positive/ERBB2–positive | 754 (18.4) | 354 (17.4) | 400 (19.5) | |
| Hormone receptor–negative/ERBB2–negative | 600 (14.7) | 332 (16.3) | 268 (13.1) | |
| Cancer stage, No. (%) | .04 | |||
| I | 2,571 (62.9) | 1,242 (61) | 1,329 (64.8) | |
| II | 1,260 (30.8) | 656 (32.2) | 604 (29.4) | |
| III | 258 (6.3) | 139 (6.8) | 119 (5.8) | |
| Mastectomy, No. (%) | 1,951 (47.7) | 976 (47.9) | 975 (47.5) | .8 |
| Lumpectomy, No. (%) | 2,116 (51.7) | 1,085 (52.9) | 1,031 (50.6) | .15 |
| Sentinel lymph node biopsy only, No. (%) | 1,330 (32.5) | 657 (32.3) | 673 (32.8) | .71 |
| Axillary lymph node biopsy only, No. (%) | 225 (5.5) | 125 (6.1) | 100 (4.9) | .08 |
| Both sentinel and axillary lymph node biopsies, No. (%) | 1,828 (44.7) | 876 (43.0) | 952 (46.4) | .03 |
| Surgical complications, No. (%) | 337 (8.2) | 164 (8.1) | 173 (8.4) | .66 |
| Hormone therapy, No. (%) | 3,058 (74.8) | 1,505 (73.9) | 1,553 (75.7) | .19 |
| Radiation therapy, No. (%) | 2,449 (59.9) | 1,208 (59.3) | 1,241 (60.5) | .44 |
| Chemotherapy, No. (%) | 1,955 (47.8) | 978 (48.0) | 977 (47.6) | .8 |
| Charlson comorbidity index,b No. (%) | .01 | |||
| 0 | 3,214 (78.6) | 1,581 (77.6) | 1,633 (79.6) | |
| 1-3 | 767 (18.8) | 387 (19) | 380 (18.5) | |
| ≥4 | 108 (2.6) | 69 (3.4) | 39 (1.9) | |
| Allostatic loadc | ||||
| High allostatic load, No. (%) | 2,032 (49.7) | 1,090 (53.6) | 942 (45.9) | <.0001 |
| Mean (SD) | 2.6 (0.03) | 2.8 (0.04) | 2.5 (0.04) | <.0001 |
| Median (Q1-Q3) | 2.0 (1.0-4.0) | 3.0 (2.0-4.0) | 2.0 (1.0-3.0) |
NOTE. High opportunity was defined as total opportunity score greater than the cohort's median.
Abbreviations: AL, allostatic load; ERBB2, human epidermal growth factor 2; SD, standard deviation.
P values from chi-square tests for the association between opportunity status and patient characteristics, and P value from Wilcoxon rank sum tests for continuous variables. Fisher exact chi-square tests were used for cells with expected counts <5.
Using Charlson index weights (excluding cancer).
AL included alkaline phosphatase, albumin, creatinine serum, heart rate, WBC count, BMI, diastolic blood pressure, systolic blood pressure, blood urea nitrogen, and glucose. High AL was defined as a total AL score (range, 0-10) greater than the median.
Relationship Between OOI and AL
Given the lack of previous studies evaluating the relationship between AL and neighborhood opportunity, correlations between the composite AL score and OOI scores were performed (Data Supplement, Table S5). Higher opportunity scores were associated with lower composite AL measures (r = –0.11; 95% CI, –0.14 to –0.08), thus indicating lower physiologic dysregulation. Specifically, lower composite AL score was associated with greater transportation access (r = –0.04; 95% CI, –0.07 to –0.01), higher education (r = –0.15; 95% CI, –0.19 to –0.12), greater employment (r = –0.04; 95% CI, –0.08 to –0.01), improved housing (r = –0.12; 95% CI, –0.15 to –0.09), better health (r = –0.04; 95% CI, –0.07 to –0.01), and lower crime rates (r = –0.07; 95% CI, –0.10 to –0.04). There was no correlation between high AL and the environment domain (r = 0.02; 95% CI, –0.01 to 0.05). On subset analysis, the relationship between the individual AL biomarkers and the OOI domains largely mirrored findings for the composite AL and OOI (Data Supplement, Table S6).
Low opportunity (v high) was associated with 36% greater odds of high AL (odds ratio [OR], 1.36 [95% CI, 1.19 to 1.55]), which remained significant in fully adjusted models (adjusted odds ratio [aOR], 1.21 [95% CI, 1.05 to 1.40]) and when modeled OOI as quartiles (Table 2). In opportunity domain-specific analyses, higher AL was associated with residency in neighborhoods with low educational attainment (OR, 1.51 [95% CI, 1.32 to 1.73]), housing problems (OR, 1.34 [95% CI, 1.17 to 1.54]), poor health (OR, 1.21 [95% CI, 1.05 to 1.39]), and high crime rates (OR, 1.19 [95% CI, 1.04 to 1.37]). In adjusted analyses, only low educational attainment (aOR, 1.23 [95% CI, 1.03 to 1.46]) and poor housing (aOR, 1.23 [95% CI, 1.06 to 1.43]) remained associated with greater odds of high AL. In dose-response analysis using OOI as a continuous variable, relative to the highest (99th percentile) opportunity level, decreased opportunity was associated with a steep increase in odds of high AL up to the 80th percentile, at which point the increase plateaued (Fig 1).
TABLE 2.
Crude and Adjusted ORs of Association Between High AL and the OOI (n = 4,089)
| OOI Domaina | No. of Patients With High AL | Crude OR (95% CI) | Adjusted ORb (95% CI) |
|---|---|---|---|
| Opportunity index score | |||
| Low | 942 | Ref | Ref |
| High | 1,090 | 1.36 (1.19 to 1.55) | 1.21 (1.05 to 1.40) |
| Opportunity index score quartile | |||
| Q1 | 438 | Ref | Ref |
| Q2 | 504 | 1.29 (1.06 to 1.58) | 1.27 (1.03 to 1.57) |
| Q3 | 528 | 1.44 (1.19 to 1.75) | 1.33 (1.08 to 1.63) |
| Q4 | 562 | 1.66 (1.38 to 1.99) | 1.43 (1.16 to 1.75) |
| Opportunity index domains | |||
| Transportation | |||
| Low | 906 | Ref | Ref |
| High | 1,126 | 1.08 (0.93 to 1.26) | 1.10 (0.94 to 1.29) |
| Education | |||
| Low | 956 | Ref | Ref |
| High | 1,076 | 1.51 (1.32 to 1.73) | 1.23 (1.03 to 1.46) |
| Employment | |||
| Low | 994 | Ref | Ref |
| High | 1,038 | 1.06 (0.92 to 1.22) | 0.96 (0.83 to 1.12) |
| Housing | |||
| Low | 948 | Ref | Ref |
| High | 1,084 | 1.34 (1.17 to 1.54) | 1.23 (1.06 to 1.43) |
| Health | |||
| Low | 943 | Ref | Ref |
| High | 1,089 | 1.21 (1.05 to 1.39) | 1.11 (0.96 to 1.28) |
| Crime | |||
| Low | 867 | Ref | Ref |
| High | 1,165 | 1.19 (1.04 to 1.37) | 1.06 (0.92 to 1.23) |
| Environment | |||
| Low | 1,117 | Ref | Ref |
| High | 915 | 0.90 (0.78 to 1.03) | 1.04 (0.90 to 1.20) |
NOTE. AL included alkaline phosphatase, albumin, creatinine serum, heart rate, WBC count, BMI, diastolic blood pressure, systolic blood pressure, blood urea nitrogen, and glucose; high AL was defined as a total AL score (range, 0-10) greater than the median. Results that are statistically significant are highlighted in bold.
Abbreviations: AL, allostatic load; OOI, Ohio Opportunity Index; OR, odds ratio; Ref, reference.
Lower relative to higher opportunity levels, the lower opportunity status was defined as a domain score lower than the median, increase OOI quartiles indicates lower opportunity.
Total opportunity index score adjusted for age group, race, ethnicity, health insurance, marital status, history of alcohol consumption or smoking, Charlson comorbidity index, molecular subtype, cancer stage, mastectomy, lumpectomy, sentinel lymph node biopsy only, axillary lymph node biopsy only, both sentinel and axillary lymph node biopsies, surgical complications, hormone therapy, radiation therapy, and chemotherapy. Domain-specific models included each domain while adjusting for all mentioned variables.
FIG 1.
Adjusted ORs of high allostatic load for OOI scores relative to highest opportunity score (n = 4,089). Models were adjusted for age group, race, ethnicity, health insurance, marital status, history of alcohol consumption or smoking, Charlson comorbidity index, molecular subtype, cancer stage, mastectomy, lumpectomy, sentinel lymph node biopsy only, axillary lymph node biopsy only, both sentinel axillary lymph node biopsies, surgical complications, hormone therapy, radiation therapy, and chemotherapy. OOI, Ohio Opportunity Index; OR, odds ratio.
Relationship Between OOI and All-Cause Mortality
Relative to high opportunity, low opportunity was associated with a 65% increased risk of all-cause mortality (hazard ratio [HR], 1.65 [95% CI, 1.29 to 2.13]; Table 3), which remained significant in fully adjusted models (adjusted hazard ratio [aHR], 1.45 [95% CI, 1.11 to 1.89]) and when modeled OOI as quartiles. Although low education and lower employment were associated with increased risk of mortality in unadjusted analyses, only education (aHR, 1.34 [95% CI, 1.03 to 1.74]) remained significant upon adjustment. Notably, high AL remained a predictor of a higher risk of all-cause mortality in fully adjusted models that included OOI (aHR, 1.78 [95% CI, 1.27 to 2.49]). There was a dose-response relationship in the association between decreasing opportunity and greater risk of all-cause mortality (Fig 2). Relative to the highest (99th percentile) level of opportunity, the risk of all-cause mortality increased steeply up to the 70th percentile, at which point the rate of increase plateaued. The effect of overall low opportunity on increased risk of all-cause mortality did not differ by AL status (P = .12; Table 3). There was, however, an interaction between the environmental domain of the OOI and AL on mortality, whereby a higher hazard of death was seen among patients with high AL residing in lower-opportunity environments (aHR, 1.96 [95% CI, 1.26 to 3.04]), but not in higher-opportunity environments (aHR, 1.02 [95% CI, 0.71 to 1.46]; P = .02).
TABLE 3.
Association Between the OOI and All-Cause Mortality by AL in Patients With Breast Cancer (n = 4,089)
| OOI Domain | No. of Deaths | All | High AL | Low AL | P Interaction | |
|---|---|---|---|---|---|---|
| Crude | Adjusteda | Adjusteda | Adjusteda | |||
| HR (95% CI) | HR (95% CI) | HR (95% CI) | HR (95% CI) | |||
| Opportunity index scoreb | .12 | |||||
| Low | 99 | Ref | Ref | Ref | Ref | |
| High | 162 | 1.65 (1.29 to 2.13) | 1.45 (1.11 to 1.89) | 1.78 (1.27 to 2.49) | 1.04 (0.65 to 1.69) | |
| Opportunity index score quartileb | .39 | |||||
| Q1 | 46 | Ref | Ref | Ref | Ref | |
| Q2 | 53 | 1.11 (0.75 to 1.66) | 1.16 (0.78 to 1.72) | 1.31 (0.75 to 2.28) | 1.05 (0.56 to 1.98) | |
| Q3 | 76 | 1.61 (1.12 to 2.34) | 1.47 (1.03 to 2.10) | 1.98 (1.18 to 3.32) | 0.99 (0.52 to 1.89) | |
| Q4 | 86 | 1.89 (1.33 to 2.73) | 1.69 (1.16 to 2.45) | 2.18 (1.17 to 3.72) | 1.17 (0.60 to 2.29) | |
| Opportunity index domains | ||||||
| Transportation | .26 | |||||
| Low | 121 | Ref | Ref | Ref | Ref | |
| High | 140 | 0.90 (0.78 to 1.26) | 1.02 (0.79 to 1.31) | 1.09 (0.80 to 1.50) | 0.84 (0.52 to 1.36) | |
| Education | .76 | |||||
| Low | 104 | Ref | Ref | Ref | Ref | |
| High | 157 | 1.61 (1.26 to 2.07) | 1.34 (1.03 to 1.74) | 1.50 (1.07 to 2.09) | 1.28 (0.81 to 2.03) | |
| Employment | .63 | |||||
| Low | 125 | Ref | Ref | Ref | Ref | |
| High | 136 | 1.04 (0.82 to 1.33) | 0.94 (0.73 to 1.23) | 0.89 (0.63 to 1.26) | 1.03 (0.64 to 1.65) | |
| Housing | .67 | |||||
| Low | 109 | Ref | Ref | Ref | Ref | |
| High | 152 | 1.40 (1.10 to 1.78) | 1.27 (0.97 to 1.65) | 1.26 (0.89 to 1.77) | 1.32 (0.82 to 2.12) | |
| Health | .62 | |||||
| Low | 117 | Ref | Ref | Ref | Ref | |
| High | 144 | 1.18 (0.93 to 1.51) | 1.08 (0.84 to 1.39) | 1.06 (0.76 to 1.47) | 1.23 (0.78 to 1.96) | |
| Crime | .85 | |||||
| Low | 106 | Ref | Ref | Ref | Ref | |
| High | 155 | 1.20 (0.94 to 1.54) | 1.10 (0.85 to 1.43) | 1.11 (0.81 to 1.53) | 0.99 (0.62 to 1.58) | |
| Environment | .02 | |||||
| Low | 143 | Ref | Ref | Ref | Ref | |
| High | 118 | 0.96 (0.75 to 1.23) | 1.08 (0.83 to 1.40) | 1.38 (0.99 to 1.94) | 0.70 (0.44 to 1.11) | |
NOTE. AL included alkaline phosphatase, albumin, creatinine serum, heart rate, WBC count, BMI, diastolic blood pressure, systolic blood pressure, blood urea nitrogen, and glucose; high AL was defined as a total AL score (range, 0-10) greater than the median.
Abbreviations: AL, allostatic load; HR, hazard ratio; OOI, Ohio Opportunity Index; Ref, reference.
Models were adjusted for high AL, age group, race, ethnicity, health insurance, marital status, history of alcohol consumption or smoking, Charlson comorbidity index, molecular subtype, cancer stage, mastectomy, lumpectomy, sentinel lymph node biopsy only, axillary lymph node biopsy only, both sentinel and axillary lymph node biopsies, surgical complications, hormone therapy, radiation therapy, and chemotherapy.
Lower relative to higher opportunity levels, the lower opportunity status was defined as a domain score lower than the median, increase OOI quartiles indicates lower opportunity.
FIG 2.
Adjusted HRs of all-cause mortality for OOI scores relative to highest opportunity score (n = 4,089). Models were adjusted for high allostatic load, age group, race, ethnicity, health insurance, marital status, history of alcohol consumption or smoking, Charlson comorbidity index, molecular subtype, cancer stage, mastectomy, lumpectomy, sentinel lymph node biopsy only, axillary lymph node biopsy only, both sentinel and axillary lymph node biopsies, surgical complications, hormone therapy, radiation therapy, and chemotherapy. HR, hazard ratio; OOI, Ohio Opportunity Index.
Post hoc mediation casual analysis results are shown in the Data Supplement (Mediation Analysis Results). The total effect of low OOI was associated with 67% increase in the risk of all-cause mortality (HR, 1.67 [95% CI, 1.25 to 2.09]). The natural indirect effect of low versus high OOI on all-cause mortality was (HR, 1.06 [95% CI, 1.02 to 1.09]), which corresponded to a 13.5% (95% CI, 4.7 to 22.5) percentage mediated through AL (Data Supplement, Table S7).
DISCUSSION
Few studies have evaluated the impact of AL on the relationship between neighborhood contextual factors and all-cause mortality in patients with breast cancer.4 We demonstrated that neighborhoods with lower opportunity are associated with higher AL. Specifically, neighborhood educational attainment and housing drove the associations between the OOI and AL. Furthermore, neighborhoods with lower opportunity have an increased risk of all-cause mortality, regardless of AL status. Patients with high AL living in neighborhoods with poorer-opportunity environments, however, have even greater risk of all-cause mortality.
Although no previous studies have evaluated the relationship between neighborhood opportunity and mortality in adults, studies evaluating neighborhood deprivation have repeatedly demonstrated increased risk of mortality in patients with cancer living in more deprived neighborhoods.4,40 A number of studies have also found associations between neighborhood disadvantage and AL, even after accounting for individual level confounders.41,42 Notably, in the study by Shen et al,42 patients with higher deprivation were younger, more likely to be racialized as Black, less likely to have a college degree, and presented with more poorly differentiated disease at more advanced stages, similar to the demographics of our patients who lived in neighborhoods with lower opportunity. Higher deprivation scores in their study had an increased odds of high AL.42 We similarly noted that lower opportunity scores had an increased odds of high AL. However, no patients in the study by Shen et al lived in the highest quartile of deprivation, limiting interpretation of their results.42
Interestingly, there was a dose-dependent relationship between OOI and AL, as well as between OOI and all-cause mortality. Similarly, a European prospective cohort study noted a dose-dependent association between neighborhood deprivation and AL that remained statistically significant after controlling for individual SES.43 Ribeiro et al43 noted that the impact of neighborhood deprivation on AL was even stronger in patients with lower SES, which was attributed to the deprivation amplification hypothesis. The deprivation amplification hypothesis suggests poorer-quality environments compound individual disadvantages.44 Although our results suggest that patients living in neighborhoods with lower opportunity had higher odds of all-cause mortality regardless of AL status, we did find that patients with high AL living in neighborhoods with poorer environmental conditions had an even greater odds of all-cause mortality, corroborating the deprivation amplification hypothesis. Our study further suggests that neighborhood opportunities allowing greater educational achievement may reduce all-cause mortality among patients with breast cancer. Notably, post hoc mediation analysis in this study suggests that OOI independently affects all-cause mortality and may be possibly mediated through AL.
Our study is limited by sample size and data collection from a single institution. Additionally, the relationship between OOI and AL does not incorporate resident mobility. The OOI treats all seven domains equally when other deprivation indices often use domain weights to acknowledge differing contributions to opportunity structure.21 Similarly, our calculation of AL assumes each biomarker contributes equally to physiologic dysregulation.45 Of note, the OOI was calculated from 2014 to 2018; however, our study examined patients initially diagnosed with breast cancer between 2012 and 2020.
In conclusion, to our knowledge, this study is the first to evaluate the relationship between AL, neighborhood opportunity, and all-cause mortality in patients with breast cancer. Lower neighborhoods opportunity was associated with higher AL and worse all-cause mortality. Additionally, patients with high AL living in neighborhoods with poorer environmental conditions have even higher odds of all-cause mortality. While our study investigated multiplicative interactions between OOI and AL, future work is needed to investigate these interactions on the additive scale.
Mohamed I. Elsaid
Honoraria: Chronic Liver Disease Foundation
Research Funding: AstraZeneca (Inst)
Jesse J. Plascak
Other Relationship: National Comprehensive Cancer Network
No other potential conflicts of interest were reported.
SUPPORT
Supported by The Ohio State University Comprehensive Cancer Center Pelotonia Grant. S.O.-G. is funded by the Paul Calabresi Career Development Award (K12 CA133250), Conquer Cancer Breast Cancer Research Foundation Advanced Clinical Research Award for Diversity and Inclusion in Breast Cancer Research, the Society of University Surgeons, and the American Cancer Society (RSG-22-106-01-CSCT).
DATA SHARING STATEMENT
Data sharing requests should be made to the Ohio State University. Requests will be considered by The Ohio State University Office of Responsible Research Practices after publication following review and approval of proposals, and with appropriate data-sharing agreements in place.
AUTHOR CONTRIBUTIONS
Conception and design: J.C. Chen, Mohamed I. Elsaid, Jesse J. Plascak, Barbara L. Andersen, William E. Carson, Timothy M. Pawlik, Samilia Obeng-Gyasi
Administrative support: William E. Carson
Collection and assembly of data: J.C. Chen, Mohamed I. Elsaid, Samilia Obeng-Gyasi
Data analysis and interpretation: J.C. Chen, Mohamed I. Elsaid, Demond Handley, Barbara L. Andersen, Timothy M. Pawlik, Naleef Fareed, Samilia Obeng-Gyasi
Manuscript writing: All authors
Final approval of manuscript: All authors
Accountable for all aspects of the work: All authors
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
Association Between Neighborhood Opportunity, Allostatic Load, and All-Cause Mortality in Patients With Breast Cancer
The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/jco/authors/author-center.
Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).
Mohamed I. Elsaid
Honoraria: Chronic Liver Disease Foundation
Research Funding: AstraZeneca (Inst)
Jesse J. Plascak
Other Relationship: National Comprehensive Cancer Network
No other potential conflicts of interest were reported.
REFERENCES
- 1. DeSantis CE, Ma J, Gaudet MM, et al. Breast cancer statistics, 2019. CA Cancer J Clin. 2019;69:438–451. doi: 10.3322/caac.21583. [DOI] [PubMed] [Google Scholar]
- 2. Silber JH, Rosenbaum PR, Ross RN, et al. Disparities in breast cancer survival by socioeconomic status despite Medicare and Medicaid insurance. Milbank Q. 2018;96:706–754. doi: 10.1111/1468-0009.12355. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Yedjou CG, Sims JN, Miele L, et al. Health and racial disparity in breast cancer. Adv Exp Med Biol. 2019;1152:31–49. doi: 10.1007/978-3-030-20301-6_3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Cheng E, Soulos PR, Irwin ML, et al. Neighborhood and individual socioeconomic disadvantage and survival among patients with nonmetastatic common cancers. JAMA Netw Open. 2021;4:e2139593. doi: 10.1001/jamanetworkopen.2021.39593. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Costanzo ES, Sood AK, Lutgendorf SK. Biobehavioral influences on cancer progression. Immunol Allergy Clin North Am. 2011;31:109–132. doi: 10.1016/j.iac.2010.09.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Huerta EE, Weeks-Coulthurst P, Williams C, et al. Take care of your neighborhood. Breast Cancer Res Treat. 2018;167:225–234. doi: 10.1007/s10549-017-4492-1. [DOI] [PubMed] [Google Scholar]
- 7. Rhee J, Medgyesi DN, Fisher JA, et al. Residential proximity to dioxin emissions and risk of breast cancer in the sister study cohort. Environ Res. 2023;222:115297. doi: 10.1016/j.envres.2023.115297. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Geng Y, Chandrasekaran S, Hsu JW, et al. Phenotypic switch in blood: Effects of pro-inflammatory cytokines on breast cancer cell aggregation and adhesion. PLoS One. 2013;8:e54959. doi: 10.1371/journal.pone.0054959. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Saini G, Ogden A, McCullough LE, et al. Disadvantaged neighborhoods and racial disparity in breast cancer outcomes: The biological link. Cancer Causes Control. 2019;30:677–686. doi: 10.1007/s10552-019-01180-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Thomas MD, Michaels EK, Reeves AN, et al. Differential associations between everyday versus institution-specific racial discrimination, self-reported health, and allostatic load among black women: Implications for clinical assessment and epidemiologic studies. Ann Epidemiol. 2019;35:20–28.e3. doi: 10.1016/j.annepidem.2019.05.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Seeman T, Epel E, Gruenewald T, et al. Socio-economic differentials in peripheral biology: Cumulative allostatic load. Ann N Y Acad Sci. 2010;1186:223–239. doi: 10.1111/j.1749-6632.2009.05341.x. [DOI] [PubMed] [Google Scholar]
- 12. McEwen BS, Seeman T. Protective and damaging effects of mediators of stress. Elaborating and testing the concepts of allostasis and allostatic load. Ann N Y Acad Sci. 1999;896:30–47. doi: 10.1111/j.1749-6632.1999.tb08103.x. [DOI] [PubMed] [Google Scholar]
- 13. Robinette JW, Charles ST, Almeida DM, et al. Neighborhood features and physiological risk: An examination of allostatic load. Health Place. 2016;41:110–118. doi: 10.1016/j.healthplace.2016.08.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Bellatorre A, Finch BK, Phuong Do D, et al. Contextual predictors of cumulative biological risk: Segregation and allostatic load. Soc Sci Q. 2011;92:1338–1362. [Google Scholar]
- 15. Akinyemiju T, Wilson LE, Deveaux A, et al. Association of allostatic load with all-cause and cancer mortality by race and body mass index in the REGARDS cohort. Cancers (Basel) 2020;12:1695. doi: 10.3390/cancers12061695. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Zhao H, Song R, Ye Y, et al. Allostatic score and its associations with demographics, healthy behaviors, tumor characteristics, and mitochondrial DNA among breast cancer patients. Breast Cancer Res Treat. 2021;187:587–596. doi: 10.1007/s10549-021-06102-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Xing CY, Doose M, Qin B, et al. Prediagnostic allostatic load as a predictor of poorly differentiated and larger sized breast cancers among black women in the women's circle of health follow-up study. Cancer Epidemiol Biomarkers Prev. 2020;29:216–224. doi: 10.1158/1055-9965.EPI-19-0712. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Acevedo-Garcia D, McArdle N, Hardy EF, et al. The child opportunity index: Improving collaboration between community development and public health. Health Aff (Millwood) 2014;33:1948–1957. doi: 10.1377/hlthaff.2014.0679. [DOI] [PubMed] [Google Scholar]
- 19. Cassetti V, Powell K, Barnes A, et al. A systematic scoping review of asset-based approaches to promote health in communities: Development of a framework. Glob Health Promot. 2020;27:15–23. doi: 10.1177/1757975919848925. [DOI] [PubMed] [Google Scholar]
- 20. Martin-Kerry J, McLean J, Hopkins T, et al. Characterizing asset-based studies in public health: Development of a framework. Health Promot Int. 2023;38:daad015. doi: 10.1093/heapro/daad015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Fareed N, Swoboda CM, Jonnalagadda P, et al. Visualizing opportunity index data using a dashboard application: A tool to communicate infant mortality-based area deprivation index information. Appl Clin Inform. 2020;11:515–527. doi: 10.1055/s-0040-1714249. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Fareed N, Singh P, Jonnalagadda P, et al. Construction of the Ohio Children's Opportunity Index. Front Public Health. 2022;10:734105. doi: 10.3389/fpubh.2022.734105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Giaquinto AN, Sung H, Miller KD, et al. Breast cancer statistics, 2022. CA Cancer J Clin. 2022;72:524–541. doi: 10.3322/caac.21754. [DOI] [PubMed] [Google Scholar]
- 24. Kittles RA, Weiss KM. Race, ancestry, and genes: Implications for defining disease risk. Annu Rev Genomics Hum Genet. 2003;4:33–67. doi: 10.1146/annurev.genom.4.070802.110356. [DOI] [PubMed] [Google Scholar]
- 25. Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. J Chronic Dis. 1987;40:373–383. doi: 10.1016/0021-9681(87)90171-8. [DOI] [PubMed] [Google Scholar]
- 26. Perou CM, Sørlie T, Eisen MB, et al. Molecular portraits of human breast tumours. Nature. 2000;406:747–752. doi: 10.1038/35021093. [DOI] [PubMed] [Google Scholar]
- 27. Krieger N, Chen JT, Waterman PD, et al. 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. Am J Epidemiol. 2002;156:471–482. doi: 10.1093/aje/kwf068. [DOI] [PubMed] [Google Scholar]
- 28. Krieger N. A century of census tracts: Health & the body politic (1906-2006) J Urban Health. 2006;83:355–361. doi: 10.1007/s11524-006-9040-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Seeman TE, Singer BH, Rowe JW, et al. Price of adaptation—Allostatic load and its health consequences. MacArthur studies of successful aging. Arch Intern Med. 1997;157:2259–2268. [PubMed] [Google Scholar]
- 30. Wiley JF, Gruenewald TL, Karlamangla AS, et al. Modeling multisystem physiological dysregulation. Psychosom Med. 2016;78:290–301. doi: 10.1097/PSY.0000000000000288. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Chen JC, Handley D, Elsaid MI, et al. The implications of racialized economic segregation and allostatic load on mortality in patients with breast cancer. Ann Surg Oncol. 2024;31:365–375. doi: 10.1245/s10434-023-14431-1. [DOI] [PubMed] [Google Scholar]
- 32. Liu Y, De A. Multiple imputation by fully conditional specification for dealing with missing data in a large epidemiologic study. Int J Stat Med Res. 2015;4:287–295. doi: 10.6000/1929-6029.2015.04.03.7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Rubin DB. Inference and missing data. Biometrika. 1976;63:581–592. [Google Scholar]
- 34.Lee EW, Wei LJ, Amato DA, et al. Cox-type regression analysis for large numbers of small groups of correlated failure time observations Klein JP, Goel PK.Survival Analysis: State of the Art Dordrecht, the Netherlands: Nato Science, Springer; 1992, pp 237-247 [Google Scholar]
- 35.Harrell FE. Regression Modeling Strategies: With Applications to Linear Models, Logistic Regression, and Survival Analysis. New York, NY: Springer-Verlag; 2001. [Google Scholar]
- 36. Valeri L, Vanderweele TJ. Mediation analysis allowing for exposure-mediator interactions and causal interpretation: Theoretical assumptions and implementation with SAS and SPSS macros. Psychol Methods. 2013;18:137–150. doi: 10.1037/a0031034. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Efron B, Tibshirani R. An Introduction to the Bootstrap. Boca Raton, FL: Chapman and Hall; 1998. [Google Scholar]
- 38. Vanderweele TJ, Vansteelandt S, Robins JM. Effect decomposition in the presence of an exposure-induced mediator-outcome confounder. Epidemiology. 2014;25:300–306. doi: 10.1097/EDE.0000000000000034. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. VanderWeele TJ, Vansteelandt S. Mediation analysis with multiple mediators. Epidemiol Methods. 2014;2:95–115. doi: 10.1515/em-2012-0010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Sundquist J, Li X, Sundquist K. Neighborhood deprivation and mortality in individuals with cancer: A multilevel analysis from Sweden. Eur J Cancer Prev. 2012;21:387–394. doi: 10.1097/CEJ.0b013e32834dbc2e. [DOI] [PubMed] [Google Scholar]
- 41. Ribeiro AI, Amaro J, Lisi C, et al. Neighborhood socioeconomic deprivation and allostatic load: A scoping review. Int J Environ Res Public Health. 2018;15:1092. doi: 10.3390/ijerph15061092. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Shen J, Fuemmeler BF, Sheppard VB, et al. Neighborhood disadvantage and biological aging biomarkers among breast cancer patients. Sci Rep. 2022;12:11006. doi: 10.1038/s41598-022-15260-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Ribeiro AI, Fraga S, Kelly-Irving M, et al. Neighbourhood socioeconomic deprivation and allostatic load: A multi-cohort study. Sci Rep. 2019;9:8790. doi: 10.1038/s41598-019-45432-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Macintyre S. Deprivation amplification revisited; or, is it always true that poorer places have poorer access to resources for healthy diets and physical activity? Int J Behav Nutr Phys Act. 2007;4:32. doi: 10.1186/1479-5868-4-32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Geronimus AT, Hicken M, Keene D. Bound J: “Weathering” and age patterns of allostatic load scores among blacks and whites in the United States. Am J Public Health. 2006;96:826–833. doi: 10.2105/AJPH.2004.060749. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Data sharing requests should be made to the Ohio State University. Requests will be considered by The Ohio State University Office of Responsible Research Practices after publication following review and approval of proposals, and with appropriate data-sharing agreements in place.