Abstract
Introduction:
Mortality from breast cancer among Black women is 60% greater than that of White women in South Carolina (SC). The aim of this study was to assess racial differences in mortality among Black and White breast cancer patients based on variations in social determinants and access to state-based early detection programs.
Methods:
We obtained a retrospective record for breast cancer patients diagnosed between 2002–2010 from the SC Central Cancer Registry. Mortality was the main outcome while race-stratified Cox proportional hazard models were performed to assess disparities in mortality. We assessed effect modification, and we used an automated backward elimination process to obtain the best fitting models.
Results:
There were 3286 patients of which the majority were White women (2186, 66.52%). Compared to married White women, the adjusted hazard ratio (aHR) for mortality was greatest among Black unmarried women (aHR: 2.31, CI: 1.83, 2.91). Compared to White women who lived in the Lowcountry region mortality was greatest among Black women who lived in the Midland (aHR: 2.17 CI: 1.47, 3.21) and Upstate (aHR: 2.96 CI: 1.96, 2.49). Mortality was higher among Black women that were not receiving services in the Best Chance Network (BCN) program (aHR: 1.70, CI: 1.40, 2.04) compared with White women.
Conclusions:
To reduce the racial disparity gap in survival in SC, Black breast cancer patients who live in the Upstate, are unmarried, and those that are not enrolled in the BCN program may benefit from more intense navigation efforts directed at early detection and linkage to breast cancer treatments.
Keywords: breast neoplasms, survival, mortality, racial disparities, South Carolina, Adjuvant hormone treatment, breast cancer, health region, marital status, racial disparity
Introduction
The Black-White disparity in breast cancer mortality is of a higher magnitude in South Carolina (SC). Mortality among White breast cancer patients is 7% lower in SC compared to the national average, however mortality is 29% higher among Black Breast cancer patients in SC[1–3]. Previous work has showed that mortality from breast cancer amongst Blacks in SC is greater than 60% than that of Whites [4]. Studies show that Black women experience worse mortality outcomes after matching for known prognostic factors, a persistent finding over time [5–8].
National statistics in the United States show that the mortality rates among breast cancer patients are significantly higher among Black women compared to White women, thus the nation exhibits strong racial disparities in mortality. While national data shows that breast cancer incidence among Black/non-Hispanic Black women is slightly closer to White/non-Hispanic White women, the percentage of breast cancer mortality among Black women is about 42% higher compared to White women. Mortality gap is over the years showed that breast cancer survival rate has remained lower among White women and has increased over time in Black women. [9]
Factors shown to influence breast cancer survival include age, socioeconomic status, hormone receptor status, health insurance type, stage at diagnosis, type of surgery, complications of surgery, marital status, county of residence (rural vs. urban) and enrolment in programs that link women to timely screening and treatment e.g., the South Carolina’s Breast and Cervical Cancer Early Detection Program (NBCCEDP) termed the Best Chance Network (BCN).[10–27]). The BCN is SC’s federally sponsored program for breast cancer screening for uninsured women. This study contributes to the existing body of literature on racial disparities in breast cancer-specific survival,[28] and overall survival.[29, 30] An assessment of survival differences by region has not been analyzed previously and would inform targeted secondary prevention strategies to the areas of greatest need (South Carolina Department of Health and Environmental Control, SCDHEC, recognizes 4 public health regions in SC). The aims of this study were to describe breast cancer-specific and overall survival for the entire state and by public health region, and to assess racial differences in mortality among Black and White breast cancer patients in SC based on variations in social determinants and access to state-based early detection programs (i.e., the Best Chance Network BCN). We hypothesize that treatment and mortality outcomes will be worse for Black women who live in Pee Dee region, an area characterized by lower socioeconomic status and low access to specialized cancer care, and better in the Lowcountry because of the National Cancer Institute’s designated cancer hospital systems available in this region.
Methods
Data Source
A retrospective cohort study design (2002 to 2010) was used including data on all breast cancer patients derived from linked files from the South Carolina Central Cancer Registry (SCCCR) and Office of Revenue and Fiscal Affairs (which maintains the administrative medical claims data for a private payor insurance plan and Medicaid). The dataset was deidentified, hence the study was exempt from IRB review. The SCCCR conducted further review of the protocol in the Department of Health and Environmental Control (DHEC) prior to data release. Information on all newly diagnosed cancer patients are collected by SCCCR, which is a population-based cancer surveillance system in SC. Data in the SCCCR include information on demographics, diagnosis date, cancer location and histology, treatment and survival. [31] All incident cancer patients are required by law to be reported to SCCCR, a resource established and maintained with funding from Centers for Disease Control’s National Program of Cancer Registries (NPCR) since 1994.
The SCCCR has a history of receiving the highest/gold rating for data completeness (>94%), timeliness and data quality from the North American Association of Central Cancer Registries and NPCR. SCCCR is a member of the CDC National Interstate Data Exchange System (N-IDEAS) so any member state may share resident incident patients with others to ensure the completeness of incident cancer data. A cohort of 3,286 patients from the SCCCR with a diagnosis of female breast cancer from 2002 to 2010 was created. They linked this cohort to the same patients in the private payor insurance plan or Medicaid datasets. The resulting combined dataset was used to conduct all analyses.
Data Linkage and Security
Linkages were made with 3 personal identifiers: name, date of birth and social security number. RFA performed these linkages in partnership with SCCCR. Because of data security issues, only the final de-identified dataset was released to study personnel and investigators for analysis. RFA retained the key in case further we needed data clarifications from the primary record. Once the de-identified data were received, the study data manager performed routine outlier and logic checks. We verified improbable values with RFA or SCCCR and rectified where possible. To create an analytic dataset, datasets from the RFA (Medicaid and private payor plan), Best Chance Network (BCN) and SCCCR were utilized to create a comprehensive record for breast cancer treatment for Black and White women with linked records. SCDHEC’s BCN, the state breast and cervical cancer early detection program, was established in 1991 to improve access to breast and cervical cancer screening and diagnostic services among low-income women in South Carolina [32]. Each participant was assigned a study participant number to facilitate analysis.
Inclusion and Exclusion criteria
We included Black and White breast cancer cases in this study because both races accounted for 99.28% of all incident cases of breast cancer and 99.52% of all mortality from breast cancer in the state. Inclusion of other races will lead to small cells that will not be viable for analyses and will also lack adequate statistical power.
Variables
Exposure Variable:
The main exposure variable was race, defined as Black or White. Variables considered as covariates or effect modifiers included age, marital status, county of residence, year of diagnosis, hormone receptor status, enrollment in BCN (dichotomized as yes or no), stage of breast cancer at diagnosis and grade of breast cancer at diagnosis.
Outcome Variables:
Overall and breast cancer mortality.
We used total survival time, vital status, and breast cancer cause of death (yes/no) for this investigation. From cause of death information, we examined breast cancer-specific mortality. For overall survival, the outcome is time to all-cause death while for breast cancer specific survival; the outcome is time to breast-cancer death. We defined the censoring time as the follow up time till 12/31/2010.
Exploratory Data Analysis:
Frequencies of age, marital status, urbanicity, year of diagnosis, hormone receptor status, cancer stage, cancer grade, enrolment status on BCN were compared amongst Black and White women utilizing the Chi square. The Kaplan-Meier survival curves with respect to race and log-rank tests, were displayed for breast cancer specific survival Kaplan-Meier survival estimators stratified by race were used to explore 5-year and 12-year survival for breast cancer-specific and all-cause survival at the state and regional levels. This was also stratified by race. The log-rank test was used to assess the Kaplan-Meier plot with respect to the main exposure, race. To test for adequacy of proportional hazard (PH) model, we graphically inspected whether the curves of logarithm of cumulative hazard function were parallel with respect to race. The curves were parallel, satisfying the PH model. We tested other variables of interest utilizing the Schoenfeld Residual.[33] The analysis satisfied the PH assumption for all variables.
Assessment of interactions:
we tested Interactions to assess the relationship between race and mortality. Statistically significant interactions were noted between race and marital status; race and health insurance status; race and region; race and BCN enrollment. Analyses which assessed factors influencing disparity in mortality were stratified by these variables into models. To assess the interactions, the framework suggested by Ward et al. was utilized whereby the subgroup with the best survival was utilized as the referent category e.g., in the interaction between marital status and race, married White women were the referent sub-category while the other 3 sub-categories were compared with the referent. We made similar categorizations for the regions, enrolment on BCN and health insurance type.[34] As per data use agreement, we masked the health insurance type. Although we presented the HR by health insurance in Table 4, we did not draw inferences on this variable.
Table 4:
Table HR and 95% CI for the associations between predictors and Breast cancer specific mortality by race in SC state overall
| Characteristic | Race | Death | Adjusted HR* Breast cancer-specific survival |
|---|---|---|---|
| Overall | No (%) | HR (CI) | |
| White | 338 (15.46) | 1.00 | |
| Black | 301 (27.36) | 1.65 (1.38, 1.97) | |
| Married | |||
| Yes | White | 168 (12.36) | 1.00 |
| Black | 93 (22.30) | 1.58 (1.19, 2.10) | |
| No | White | 170 (20.56) | 1.60 (1.25, 2.04) |
| Black | 208 (30.45) | 2.31 (1.83, 2.91) | |
| Region | |||
| Low Country | White | 51 (11.89) | 1.00 |
| Black | 72 (26.97) | 2.14 (1.43, 3.20) | |
| Midlands | White | 109 (16.59) | 1.51 (1.04, 2.21) |
| Black | 83 (25.08) | 2.17 (1.47, 3.21) | |
| Pee Dee | White | 67 (17.77) | 1.40 (0.93, 2.12) |
| Black | 78 (26.00) | 1.99 (1.34, 2.95) | |
| Upstate | White | 111 (15.35) | 1.47 (1.01, 2.13) |
| Black | 68 (33.66) | 2.96 (1.96, 4.45) | |
| BCN | |||
| No | White | 311 (15.09) | 1.00 |
| Black | 264 (27.44) | 1.70 (1.40 2.04) | |
| Yes | White | 27 (21.60) | 1.38 (0.90, 2.12) |
| Black | 37 (26.81) | 1.68 (1.13, 2.49) | |
| Insurance | |||
| 1 | White | 132 (8.47) | 1.00 |
| Black | 76 (15.67) | 1.71 (1.24, 2.34) | |
| 2 | White | 206 (32.80) | 3.42 (2.66, 4.39) |
| Black | 225 (36.59) | 3.96 (3.08, 5.08) |
All models adjusted for stage, grade, age, diagnosis year, hormone status and urbanicity
Fitting the best model in each stratum:
Relationships between race and mortality (breast-cancer specific and all-cause) were assessed in each stratum via fitting the best Cox PH model through automated backward elimination, starting with all potential covariates like age, year of diagnosis, hormone receptor status, enrollment in BCN (yes or no), stage of breast cancer at diagnosis and grade of breast cancer at diagnosis.
Results
We show descriptive statistics for this study sample in Table 1. Overall, there were 3286 patients of breast cancer patients, most which were White women (2186, 66.52%). In bivariate analyses, there were significant differences between Black and White women for age, urbanicity, year of diagnosis, hormone receptor status, cancer grade, cancer stage, marital status, enrolment in BCN and insurance provider (Table 1). Black women were more likely to be in the youngest age group of less than 45 years, when compared to White women (21.82% vs. 16.19%). White women were more likely to be in the older age group, 55–64 years compared to Black women (38.88% vs. 30.00%). Black women were more likely to be unmarried (62.09%) than White women (37.83%). The proportion of White women living in urban areas (78.18% vs. 69.21%) with hormone receptor positive cancer (43.05% vs 32.28%) is higher than that of Black women. More black women took part in the BCN program than whites (12.55% vs. 5.72%). Figure 1 presents the Kaplan Meier plots of 12-year Breast cancer-specific mortality among black and white women in South Carolina. There was a statistically significant difference between Black women in the cohort. This difference was early as 1 year, continued to widen through 12 years of follow-up.
Table 1.
Summary of patients’ characteristics by race
| N (%) | N (%) | N (%) | |||
|---|---|---|---|---|---|
| Characteristic | Total (N=3286) | White N= (2186) | Black N= (1100) | P | |
| Age (mean±SD) | 51.00 (7.64) | 51.44 (7.45) | 50.11 (7.92) | <0.01 | |
| Age categories | Under 45 years old | 594 (18.08) | 354 (16.19) | 240 (21.82) | <0.01 |
| 45–54 years old | 1512 (46.01) | 982 (44.92) | 530 (48.18) | ||
| 55–64 years old | 1180 (35.91) | 850 (38.88) | 330 (30.00) | ||
| Marital status | Not married | 1510 (45.95) | 827 (37.83) | 683 (62.09) | <0.01 |
| Married | 1776 (54.05) | 1359 (62.17) | 417 (37.91) | ||
| Rural/Urban status | Rural | 815 (24.80) | 477 (21.82) | 338 (30.73) | <0.01 |
| Urban | 2471 (75.20) | 1709 (78.18) | 762 (69.21) | ||
| Year of diagnosis | 2002–2004 | 994 (30.25) | 671 (30.70) | 323 (29.36) | 0.04 |
| 2005–2007 | 1020 (31.04) | 701 (32.07) | 319 (29.00) | ||
| 2008–2010 | 1272 (38.71) | 814 (37.24) | 458 (41.64) | ||
| Hormone receptor status | Negative | 512 (15.58) | 271 (12.40) | 241 (21.91) | <0.01 |
| Unknown | 1472 (44.80) | 974 (44.56) | 498 (45.27) | ||
| Positive | 1302 (39.62) | 941 (43.05) | 361 (32.28) | ||
| Stage at Diagnosis | In-situ | 555 (16.89) | 376 (17.20) | 179 (16.27) | <0.01 |
| Local | 1460 (44.43) | 1015 (46.43) | 445 (40.45) | ||
| Regional/Distant | 1232 (37.49) | 774 (35.41) | 458 (41.64) | ||
| Unknown | 39 (1.19) | (0.96) | 18 (1.64) | ||
| Cancer grade | I | 489 (14.88) | 367 (16.79) | 122 (11.09) | <0.01 |
| II | 1099 (33.44) | 799 (36.55) | 300 (27.27) | ||
| III | 1383 (42.09) | 801 (36.74) | 582 (52.91) | ||
| Unknown | 315 (9.59) | 219 (10.02) | 96 (8.73) | ||
| Best Chance Network | No | 3023 (92.00) | 2061 (94.28) | 962 (87.45) | <0.01 |
| Yes | 263 (8.00) | 125 (5.72) | 138 (12.55) | ||
| Health Insurance* | 1 | 2043 (62.17) | 1558 (71.27) | 485 (44.09) | <0.01 |
| 2 | 1243 (37.83) | 628 (28.73) | 615 (55.91) |
As per data use agreement with data providers, the specific health insurance providers were masked
Figure 1:
Kaplan Meier curve for the association between race and 12-year breast cancer specific mortality in South Carolina (p<0.01)
Table 2 shows the 5- and 12- year survival proportion for Breast cancer-specific mortality among black and white women in the whole state and by SCDHEC’s 4 geographic public health regions. There were significant differences in the survival rate (using the Kaplan Meier survival model) between Black and White breast cancer patients in all public health regions in the state. The highest breast cancer specific 5-year survival rate was identified among Black women (80.0%) and the lowest among White women (87.0%) was observed among patients who lived in the Pee Dee region. The Pee Dee region also accounted for the lowest Black-White differences for women in the state (7.0%). The lowest Breast cancer specific 5-year survival rate among black women (73.3%) was seen in the Upstate region. The highest among white Women (91.4%) was observed among patients living in the Low Country region. The highest Black-White difference was observed among patients living in the Upstate region of the state. When Black women were compared across the 4 public health regions overall, there were no significant differences between them. Similarly, when comparing White women across all 4 regions, no statistically significant differences were found among them. (Table 3).
Table 2:
5-year and 12-year survival by region stratified by race in South Carolina
| Breast cancer-specific survival | |||
|---|---|---|---|
| 5YST (%) | 12YST (%) | ||
| South Carolina | Overall | 85.6 | 80.2 |
| Black | 78.2 | 72.2 | |
| White | 89.2 | 84.2 | |
| Black-white | 11.0 | 12.0 | |
| p-value | <0.01 | ||
| Midlands | Overall | 86.1 | 80.0 |
| Black | 79.2 | 74.2 | |
| White | 89.5 | 82.9 | |
| Black-white | 10.3 | 8.7 | |
| p-value | <0.01 | ||
| Low Country | Overall | 86.6 | 82.0 |
| Black | 78.6 | 72.3 | |
| White | 91.4 | 87.3 | |
| Black-white | 12.8 | 15.0 | |
| p-value | <0.01 | ||
| Pee Dee | Overall | 84.0 | 76.1 |
| Black | 80.0 | 71.3 | |
| White | 87.0 | 82.0 | |
| Black-white | 7.0 | 10.7 | |
| p-value | <0.01 | ||
| Upstate | Overall | 85.5 | 81.5 |
| Black | 73.3 | 70.3 | |
| White | 88.9 | 84.6 | |
| Black-white | 15.6 | 14.3 | |
| p-value | <0.01 | ||
Table 3:
5-year and 12-year survival by race stratified by region in South Carolina
| Breast cancer-specific survival | |||
|---|---|---|---|
| 5YST (%) | 12YST (%) | ||
| Black | Overall | 78.2 | 72.2 |
| Midlands | 79.2 | 74.2 | |
| Low Country | 78.6 | 72.3 | |
| Pee Dee | 80.4 | 71.3 | |
| Upstate | 73.3 | 70.3 | |
| p-value | 0.32 | ||
| White | Overall | 89.2 | 84.2 |
| Midlands | 89.5 | 82.9 | |
| Low Country | 91.4 | 87.3 | |
| Pee Dee | 87.0 | 82.0 | |
| Upstate | 88.9 | 84.6 | |
| p-value | 0.27 | ||
Table 4 presents the Cox PH models. Multivariable Cox PH models adjusted for cancer stage, grade, age, diagnosis year, hormone receptor status and urbanicity. Overall, when Black women were compared with White women, the adjusted hazard ratio of mortality was 1.65 (CI: 1.41, 1.94) for overall survival and 1.65 (CI: 1.38, 1.97) for breast cancer specific survival. Utilizing the group of breast cancer patients that were married and white as referent category, the hazard ratio for mortality was greatest among black women that were not married (HR: 2.31 CI: 1.83, 2.91). Utilizing the group of Breast cancer patients who lived in the Low Country region and white as referent category, the hazard ratio for mortality was greatest among black women who lived in the Midlands (HR: 2.17 CI: 1.47, 3.21) and Black women who lived in the Upstate region (HR: 2.96 CI: 1.96, 2.49). For women enrolled in BCN, the sub-group with the highest hazard of mortality was Black women who enrolled in the BCN program (HR: 1.70 CI: 1.40, 2.04) when compared with White women who were also not enrolled in the BCN program.
Discussion
We found that there was an overall statistically significant difference between Black and White women in the cohort. The 5-year and 12-year survival rate for breast cancer specific survival also illuminated significant differences between Black and White women across all four public health regions. Black women in the Upstate region had the highest breast cancer, specific HR when compared with white women in the Low Country region. This was followed by Black women in the Midlands and Low Country region. We also found that Black women who were not married had a higher breast cancer specific HR compared to married white women. Additionally, Black women who were not BCN program participants had a higher HR compared to white women who were also not enrolled in the program.
The crude hazard ratio of mortality for Black women compared to White women was 1.65 (CI: 1.38, 1.97) for breast cancer-specific survival which represents 65% excess risk of death among Black women. Our results are aligned with findings from similar studies i.e., which noted a 60% higher risk of Black women dying from breast cancer compared to white women in SC. These previous studies showed that although the Black-White disparity in breast cancer mortality is seen both at the national level and state level in SC, the disparity is of a higher magnitude in SC [1–4, 35]. Mortality among White breast cancer women is 7% lower in SC compared to the national average and is 29% higher among Black women in SC [1–3]. Although different analytic methods (age adjusted mortality) were used in two prior studies, the mortality differences in SC utilizing the Cox PH model with adjustments (as we did) showed similar findings. A previous study identified a paradoxical relationship of a lower incidence rate of breast cancer, but a higher mortality rate among Black women when age-adjusted rates were utilized [35]. Our study extends previous work, goes a step beyond using age-adjusted data by incorporating Cox Proportional Hazard modeling which allows us to account for important confounding variables such as stage and demographic characteristics while assessing the relationship between mortality and race among breast cancer patients.
We identified the most disadvantaged group based on the HR as black women living in the Upstate regions. This finding may help inform policies or interventions for SC’s breast and cervical cancer screening preventive programs e.g., BCN program which is a federally sponsored National Breast and Cervical Cancer Early Detection Program [3, 36]. At the time of this study (2002–2010) program eligibility criteria for SC women included women aged 47–64 years, residents of SC, underserved/underinsured or do not have insurance and are of low income (less than 200% of the Federal poverty level). This program provides high quality breast and cervical cancer screening, follow-up, diagnostic and treatment services in all 50 US states, territories and several federally recognized Native American tribes at no cost to participants [2, 3, 36]. At the time of this study (2002–2010), recruitment into the BCN program is usually through an active search by federally qualified health centers, media outlets and through outreach carried out by the American Cancer Society. [3, 36]. Beginning from 2015, the BCN program in SC has instituted a multi-pronged collaborative public health-healthcare provider-community approach and partnerships to reduce racial disparities [37]. Besides these current efforts, our findings may help the administrators of the BCN program to plan and focus efforts to continue to reduce racial disparities to certain regions of the state or implement special programs (e.g., patient navigation) for those regions.
In the state of SC overall, it has been hypothesized that comorbid chronic disease illnesses and obesity could be the driver of cancer racial disparities because of sharing many of the same risk factors. [38] This is because obesity and comorbid illnesses affect Black residents in SC at rates above the national average. [38, 39] Previously we have shown an increased risk of colorectal cancer in diabetic patients that is most pronounced in Black residents [40]. In a 2018 report, SC had the 10th highest adult obesity rate in the US, up from being the 13th highest in 2016. Comparing these figures to that of 2000 (21.1%), the weight gain problem appears to be on the rise persistently in SC. [38, 39] Some authors have shown that access to healthy foods such as fresh fruits and vegetables and lack of grocery stores contribute to the higher BMI. In these studies, respondents perception of a lack of access to adequate food shopping in their neighborhoods correlated with higher BMI and they also identified shopping frequency, use of community food resources, transportation methods, and shopping distance as key factors that defined food shopping patterns.[41, 42]
Previous study has also shown that severe obesity and high waist-to-hip ratio among Black women have also been found to contribute to racial differences in stage at breast cancer diagnosis, as Black women had a higher likelihood of severe obesity and being in the highest tertile of waist-to-hip ratio [43]. For example, Moorman et al. found that the odds of a woman diagnosed with metastatic breast cancer is 3.07 (CI: 1.31–7.17) compared with a woman who is not severely obese. [43] Obesity has been associated with increased risk of breast cancer and epidemiologic research have shown that there is a higher obesity prevalence among Black women in the United States. [44] Breast cancer is the most common cancer in postmenopausal women in the United States and has been associated with chronic obesity. [44] When the obesity rate in SC is broken down by race, it is 42.1% Black residents (ranking 9th) and 29.6% among White residents (ranking 25th). Specifically in Charleston, SC (the largest city in the Lowcountry), there is a 14.21% difference in the overweight or obesity rate (73.1% among Black residents and 63.4% among White residents). [45]
The regional sub-population of Gullahs also may be one factor affecting the racial disparities in the Lowcountry region of SC. The Gullahs are a unique black sub-population that live and live in the farming and the fishing communities along SC’s coastal areas. [38] The Gullah community are geographically isolated and previous studies show that they experienced limited access to health care and are at a higher risk of cardio metabolic risk factors for diabetes mellitus. [46, 47] SC diabetes rates is the 5th highest (13.4%) in the US and may help explain some of the racial disparities. [39] The Gullah’s limited access to care (for diabetes mellitus for example) may also have affected their receipt of care for breast cancer. Charlot et al. showed that there is positive association between type 2 diabetes and breast cancer. [48] Since there is higher prevalence and earlier onset of type 2 diabetes in Black women, the hypothesis that diabetes has contributed to racial disparities is likely a plausible hypothesis. [48]
Other reports among Black residents noted persistent, systemic barriers to accessing care for Black women as shown by a consistent receipt of lower-than-recommended breast cancer care when compared with their white counterparts [6, 49–53]. This includes lower-than-recommended rates of radiation after surgical treatment for breast cancer [6, 49–51]. Black women were also shown to be 30–40% more likely to receive breast cancer treatments that are not in line with guidelines across all breast cancer subtypes [14]. Previous studies show that there is a tendency among Black individuals not to receive preventative services such as early screening mammograms at relatively young ages from health care providers. [54].
We also found that Black women who were not married had a higher risk of death compared with White women who were married. This information also may be of great utility in identifying demographic variables that may benefit more from navigation to breast cancer preventive services. Given that studies show unmarried women with breast cancer have higher rates of mortality compared to married women [55–57], additional efforts should be made to understand the underlying factors related to marital status and mortality, and how best to support and navigate unmarried women through timely breast cancer diagnosis and treatment. Potential mechanisms that may have led to increased mortality among women with breast cancer are lack of social support, living in low socioeconomic status neighborhoods, higher levels of comorbid conditions e.g., obesity, lower level of breastfeeding and stress associated with singleness and heading homes with children. [55, 58–66] The first possible factor is lack of social support; this may lead to higher mortality among women that are not married as social support is needed to enhance interventions that may prolong life. [59] In one of our research studies, we found that Black participants who had a partner enrolled in the study were about three times more likely than participants without a partner to be retained in the behavioral trial study. [59] Having a partner in this sense may be a crude measure of social support that enhances healthful behavior. [59] However, we do not have data on if the women that were not married had partners or not. Further studies are warranted to better understand this mechanism.
Also, a previous study showed that mortality was highest among unmarried women in low socioeconomic status neighborhoods. [55] Along the line of neighborhood deprivation, some previous studies also show that residential factors are important to decisions about early detection in community-based samples of Black people; collective efficacy and neighborhood satisfaction were associated significantly with decisions about cancer screening and adherence to recommendations for diet and physical activity among Black women. [64, 65] In a post-hoc analyses, the impact of a dietary and physical activity intervention on dietary intake differed by the food access level of the participant’s residential area.[60, 62] In this previous study, we found that participants that lived in areas with limited access to healthy diets and resources for participating in physical activity may have had higher difficulty in making the behavior changes learned through our intervention. Similarly, for those participants making behavior changes, it could be more challenging to maintain these healthier behaviors without a supportive environment [67]. Future studies should seek to understand the complex interaction between neighborhood environments of women who have breast cancer, marital status and mortality.
Another reason that may be responsible for this finding of higher mortality among Black women that were not married is existence of a higher level of comorbid conditions such as obesity among Black women compared to White women who have breast cancer. [63] This problem of higher rate of obesity/weight gain is further compounded by lower rate of breastfeeding seen among Black women. [63] Cohen et al. found that at all levels of parity and breastfeeding, Black women had a higher BMI and weight gain since age of 21 years. [63] Also, the proportion of Black women that did not breastfeed their infant (68.7%) was higher than White women that did not breastfeed their infant (63.5%). Another study showed that parous women who have not breastfed (higher rates seen in Black women) are more likely to have more aggressive forms of breast cancer e.g., triple negative breast cancer, estrogen receptor negative breast cancer. [66]
Another study showed that Black women had increased odds of being unmarried (OR: 4.9 CI: 4.57–5.24) while the mean parity among the unmarried was 1.2 + 1.5. [58] The fact that unmarried Black women with breast cancer may have children might be important for the stress, taking care of family without help of a spouse and lower SES (since similar income will be used for more individuals to take care per women).[58] We also found in a previous work that higher levels of stress among Blacks led to lower retention in a healthy intervention. [59] In a similar work, an association between psychological stress and breast cancer was found especially related to stressful life events. [61] Summarily, lack of social support, living in low socioeconomic status neighborhoods, higher levels of comorbid conditions e.g., obesity, lower level of breastfeeding and stress associated with singleness and heading homes with children may play a role
This study contributes to the existing body of literature on racial disparities, on breast cancer-specific survival, [28] and overall survival [29, 30]. To our knowledge, this is the first study conducting regional cancer survival disparity analysis on breast cancer in SC, showing that disparities are higher in certain public health regions in the state. This study also considers the impact of the state’s CDC funded breast and cervical cancer screening program (BCN) on racial disparity in breast mortality. One of the study’s limitations was the inability to make an assessment of data for the state’s Hispanic population because of a very low subpopulation of Hispanic breast cancer patients because sample size was small and would yield unstable estimates. We included black and White breast cancer cases in this study because both races accounted for 99.28% of all incident cases of breast cancer and 99.52% of all mortality from breast cancer in the state. Inclusion of other races will lead to small cells that will not be viable for analyses and will also lack adequate statistical power. Another limitation is that this data is from the state of SC hence is more relatable to Southern United States. This may limit generalizability. The study cohort was also limited in the number of other biological, patient-, physician-, and healthcare-system-related factors that could be assessed to further examine issues observed.
In conclusion, mortality was highest among Black women living in the Upstate region of the state; Black women who were not married; and Black women not enrolled in the BCN Program. Navigation and other available programs aimed at reducing racial disparities may benefit from these findings by using them to implement culturally appropriate outreach, education and referral services and resources to Black communities who live in regions most disadvantaged. Other regions in the state may benefit from specific community-oriented interventions like the Lowcountry’s community COMPASS project [38] which might aid in closing this gap. To reduce the racial disparity gap in survival in SC, Black breast cancer patients that live in the Upstate, not married and not enrolled in the BCN program may benefit from more intense navigation efforts directed at early detection and linkage to receipt of breast cancer treatments. Access to social services for a family would help to relieve some of this stress and economic burden. Future studies are also required to identify the potential biological, patient-, physician-, and healthcare-system-related factors underlying our observations to optimize cancer care provided to Black women in SC. Finally, the state of SC could benefit from future studies assessing whether regional disparities exist for other common cancers to identify if this trend is specific to breast cancer or extends to other cancers and to inform future development and implementation of regionally appropriate policies and interventions.
Financial support.
A National Cancer Institute’s F99/K00 Fellowship grant (1 F99 CA 222722) as principal investigator supported Oluwole A. Babatunde.
A National Cancer Institute’s R15 grant supported Swann A. Adams (R15CA179355) as principal investigator.
Samantha Truman is being supported by Interdisciplinary Graduate Training Program in Cancer Disparities (IGniTE-CD) program (GTDR17500160) sponsored by Susan G. Komen.
Footnotes
Publisher's Disclaimer: This Author Accepted Manuscript is a PDF file of a an unedited peer-reviewed manuscript that has been accepted for publication but has not been copyedited or corrected. The official version of record that is published in the journal is kept up to date and so may therefore differ from this version.
Conflicts.
No Author declares a conflict of interest
Disclosure.
Dr. James R. Hébert owns a controlling interest in Connecting Health Innovations LLC (CHI), a company planning to license the right to his invention of the DII™ from the University of South Carolina to develop computer and smart applications for patient counselling and dietary intervention in clinical settings.
Appreciation.
We will like to appreciate Krystal Johnson (Ph.D.), Director of Research & Planning in the Division of Cancer Prevention & Control in S.C. Dept. of Health & Environmental Control for helping to provide feedbacks about the BCN program and editing the paper.
References
- 1.Adams SA, et al. , Breast cancer disparities in South Carolina: early detection, special programs, and descriptive epidemiology. J S C Med Assoc, 2006. 102(7): p. 231–9. [PMC free article] [PubMed] [Google Scholar]
- 2.Adams SA, et al. , Racial differences in follow-up of abnormal mammography findings among economically disadvantaged women. Cancer, 2009. 115(24): p. 5788–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Smith ER, et al. , Breast cancer survival among economically disadvantaged women: the influences of delayed diagnosis and treatment on mortality. Cancer Epidemiol Biomarkers Prev, 2008. 17(10): p. 2882–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Hebert JR, et al. , Mapping Cancer Mortality-to-Incidence Ratios to Illustrate Racial and Sex Disparities in a High-Risk Population. Cancer, 2009. 115(11): p. 2539–2552. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Brown D, et al. , Trends in use of contralateral prophylactic mastectomy by racial/ethnic group and ER/PR status among patients with breast cancer: A SEER population-based study. Cancer Epidemiol, 2016. 42: p. 24–31. [DOI] [PubMed] [Google Scholar]
- 6.Freedman RA, et al. , The association of race/ethnicity, insurance status, and socioeconomic factors with breast cancer care. Cancer, 2011. 117(1): p. 180–9. [DOI] [PubMed] [Google Scholar]
- 7.Jatoi I, et al. , Breast cancer trends among black and white women in the United States. J Clin Oncol, 2005. 23(31): p. 7836–41. [DOI] [PubMed] [Google Scholar]
- 8.Jatoi I, Becher H, and Leake CR, Widening disparity in survival between white and African-American patients with breast carcinoma treated in the U. S. Department of Defense Healthcare system. Cancer, 2003. 98(5): p. 894–9. [DOI] [PubMed] [Google Scholar]
- 9.Yedjou CG, et al. , Health and Racial Disparity in Breast Cancer. Adv Exp Med Biol, 2019. 1152: p. 31–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.American Cancer Society (2015) Cancer Facts and Figures 2015. Atlanta: American Cancer Society; 2015 [cited October 22nd 2017]. Available from https://www.cancer.org/research/cancer-facts-statistics/all-cancer-facts-figures/cancer-facts-figures-2015.html. [Google Scholar]
- 11.Shavers VL, Harlan LC, and Stevens JL, Racial/ethnic variation in clinical presentation, treatment, and survival among breast cancer patients under age 35. Cancer, 2003. 97(1): p. 134–47. [DOI] [PubMed] [Google Scholar]
- 12.Gajdos C, et al. , Stage 0 to stage III breast cancer in young women. J Am Coll Surg, 2000. 190(5): p. 523–9. [DOI] [PubMed] [Google Scholar]
- 13.Newman LA, Breast Cancer Disparities: Socioeconomic Factors versus Biology. Ann Surg Oncol, 2017. [DOI] [PubMed] [Google Scholar]
- 14.Chen L and Li CI, Racial disparities in breast cancer diagnosis and treatment by hormone receptor and HER2 status. Cancer Epidemiol Biomarkers Prev, 2015. 24(11): p. 1666–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Connors SK, et al. , Breast reconstruction after mastectomy at a comprehensive cancer center. Springerplus, 2016. 5(1): p. 955. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Reyes SA, et al. , Factors Affecting the Completion of Adjuvant Chemotherapy in Early-Stage Breast Cancer. Ann Surg Oncol, 2016. 23(5): p. 1537–42. [DOI] [PubMed] [Google Scholar]
- 17.Cancer Statistics. Surveillance, Epidemiology, and End Results. https://seer.cancer.gov/statistics/summaries.html Accessed October 22nd 2017.
- 18.Jemal A, et al. , Annual Report to the Nation on the Status of Cancer, 1975–2014, Featuring Survival. J Natl Cancer Inst, 2017. 109(9). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Kohler BA, et al. , Annual Report to the Nation on the Status of Cancer, 1975–2011, Featuring Incidence of Breast Cancer Subtypes by Race/Ethnicity, Poverty, and State. J Natl Cancer Inst, 2015. 107(6): p. djv048. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Newman LA, et al. , Meta-analysis of survival in African American and white American patients with breast cancer: ethnicity compared with socioeconomic status. J Clin Oncol, 2006. 24(9): p. 1342–9. [DOI] [PubMed] [Google Scholar]
- 21.Akinyemiju T, et al. , Racial disparities in individual breast cancer outcomes by hormone-receptor subtype, area-level socio-economic status and healthcare resources. Breast Cancer Res Treat, 2016. 157(3): p. 575–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Connors SK, et al. , Breast cancer treatment among African American women in north St. Louis, Missouri. J Urban Health, 2015. 92(1): p. 67–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Vandergrift JL, et al. , Time to adjuvant chemotherapy for breast cancer in National Comprehensive Cancer Network institutions. J Natl Cancer Inst, 2013. 105(2): p. 104–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Chevray PM, Timing of breast reconstruction: immediate versus delayed. Cancer J, 2008. 14(4): p. 223–9. [DOI] [PubMed] [Google Scholar]
- 25.Fernandez-Frias AM, et al. , Immediate reconstruction after mastectomy for breast cancer: which factors affect its course and final outcome? J Am Coll Surg, 2009. 208(1): p. 126–33. [DOI] [PubMed] [Google Scholar]
- 26.Mosunjac M, et al. , Time to treatment for patients receiving BCS in a public and a private university hospital in Atlanta. Breast J, 2012. 18(2): p. 163–7. [DOI] [PubMed] [Google Scholar]
- 27.Williams F and Thompson E, Disparity in Breast Cancer Late Stage at Diagnosis in Missouri: Does Rural Versus Urban Residence Matter? J Racial Ethn Health Disparities, 2016. 3(2): p. 233–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Koh HK, et al. , Effect of Time Interval between Breast-Conserving Surgery and Radiation Therapy on Outcomes of Node-Positive Breast Cancer Patients Treated with Adjuvant Doxorubicin/Cyclophosphamide Followed by Taxane. Cancer Res Treat, 2016. 48(2): p. 483–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Haji-Jama S, et al. , Disparities Report: Disparities Among Minority Women With Breast Cancer Living in Impoverished Areas of California. Cancer Control, 2016. 23(2): p. 157–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Yin D, et al. , Effect of misclassified underlying cause of death on survival estimates of colon and rectal cancer. J Natl Cancer Inst, 2011. 103(14): p. 1130–3. [DOI] [PubMed] [Google Scholar]
- 31.Babatunde OA, et al. , Racial disparities in endometrial cancer mortality-to-incidence ratios among Blacks and Whites in South Carolina. Cancer Causes Control, 2016. 27(4): p. 503–11. [DOI] [PubMed] [Google Scholar]
- 32.South Carolina Cancer Alliance. Breast Cancer statistics. THE STATE O F BR EA ST C A NCE R IN SO U TH C AROLINA. Available at http://www.sccancer.org/media/1154/final-report_62816.pdf Downloaded on January 301th, 2019.
- 33.Zhou B, Fine J, and Laird G, Goodness-of-fit test for proportional subdistribution hazards model. Stat Med, 2013. 32(22): p. 3804–11. [DOI] [PubMed] [Google Scholar]
- 34.Ward JB, et al. , How do we assess a racial disparity in health? Distribution, interaction, and interpretation in epidemiological studies. Ann Epidemiol, 2019. 29: p. 1–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Samson ME, et al. , Disparities in Breast Cancer Incidence, Mortality, and Quality of Care among African American and European American Women in South Carolina. South Med J, 2016. 109(1): p. 24–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Best Chance Network (BCN): South Carolina’s Breast and Cervical Cancer Early Detection Program (SCBCCEDP). Available at http://www.scdhec.gov/Health/DiseasesandConditions/Cancer/FreeCancerScreenings/ downloaded 09/17/2017.
- 37.South Carolina Cancer Alliance. Breast Cancer statistics: The State of Breast Cancer in SC. Downloaded on 09/06/2019 Available at https://www.sccancer.org/media/1154/final-report_62816.pdf.
- 38.Bryant DC, et al. , The Community Compass Project: A Community Engagement Model Targeting African-Americans in the Low Country of South Carolina. J Natl Black Nurses Assoc, 2017. 28(1): p. 38–42. [PubMed] [Google Scholar]
- 39.Robert Wood Johnson Foundation. The state of obesity: Better policies for a healthier America. 2018. Retrieved 12/24/2018 from https://stateofobesity.org/wp-content/uploads/2018/09/stateofobesity2018.pdf.
- 40.Cavicchia PP, et al. , Racial disparities in colorectal cancer incidence by type 2 diabetes mellitus status. Cancer Causes Control, 2013. 24(2): p. 277–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Liese AD, et al. , Food Shopping and Acquisition Behaviors in Relation to BMI among Residents of Low-Income Communities in South Carolina. Int J Environ Res Public Health, 2017. 14(9). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Ma X, et al. , Food Acquisition and Shopping Patterns among Residents of Low-Income and Low-Access Communities in South Carolina. J Acad Nutr Diet, 2018. 118(10): p. 1844–1854. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Moorman PG, et al. , Race, anthropometric factors, and stage at diagnosis of breast cancer. Am J Epidemiol, 2001. 153(3): p. 284–91. [DOI] [PubMed] [Google Scholar]
- 44.Agurs-Collins T, Ross SA, and Dunn BK, The Many Faces of Obesity and Its Influence on Breast Cancer Risk. Front Oncol, 2019. 9: p. 765. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.College of Charleston: Race and Social Justice Initiative. The state of racial disparities in Charleston county, South Carolina 2000–2015. 2017. Retrieved on 12/24/2018 from https://rsji.cofc.edu/wp-content/uploads/2017/01/The-State-of-Racial-Disparities-in-Charleston-County-SC-Rev.−11-14.pdf.
- 46.Hunt KJ, et al. , Cardiovascular risk in Gullah African Americans with high familial risk of type 2 diabetes mellitus: project SuGAR. South Med J, 2014. 107(10): p. 607–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Johnson-Spruill I and Riegel B, The quality of diabetes care to Gullah families of South Carolina. J Natl Black Nurses Assoc, 2008. 19(2): p. 20–7. [PMC free article] [PubMed] [Google Scholar]
- 48.Charlot M, et al. , Diabetes and breast cancer mortality in Black women. Cancer Causes Control, 2017. 28(1): p. 61–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Bickell NA, et al. , Missed opportunities: racial disparities in adjuvant breast cancer treatment. J Clin Oncol, 2006. 24(9): p. 1357–62. [DOI] [PubMed] [Google Scholar]
- 50.Freedman RA, et al. , Trends in racial and age disparities in definitive local therapy of early-stage breast cancer. J Clin Oncol, 2009. 27(5): p. 713–9. [DOI] [PubMed] [Google Scholar]
- 51.Gross CP, et al. , Racial disparities in cancer therapy: did the gap narrow between 1992 and 2002? Cancer, 2008. 112(4): p. 900–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Sheppard VB, et al. , Disparities in breast cancer surgery delay: the lingering effect of race. Ann Surg Oncol, 2015. 22(9): p. 2902–11. [DOI] [PubMed] [Google Scholar]
- 53.Komenaka IK, et al. , Race and ethnicity and breast cancer outcomes in an underinsured population. J Natl Cancer Inst, 2010. 102(15): p. 1178–87. [DOI] [PubMed] [Google Scholar]
- 54.Dehal A, Abbas A, and Johna S, Racial disparities in clinical presentation, surgical treatment and in-hospital outcomes of women with breast cancer: analysis of nationwide inpatient sample database. Breast Cancer Res Treat, 2013. 139(2): p. 561–9. [DOI] [PubMed] [Google Scholar]
- 55.Martinez ME, et al. , Prognostic significance of marital status in breast cancer survival: A population-based study. PLoS One, 2017. 12(5): p. e0175515. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56.Osborne C, et al. , The influence of marital status on the stage at diagnosis, treatment, and survival of older women with breast cancer. Breast Cancer Res Treat, 2005. 93(1): p. 41–7. [DOI] [PubMed] [Google Scholar]
- 57.Wu AH, et al. , The California Breast Cancer Survivorship Consortium (CBCSC): prognostic factors associated with racial/ethnic differences in breast cancer survival. Cancer Causes Control, 2013. 24(10): p. 1821–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58.Amini SB, Catalano PM, and Mann LI, Births to unmarried mothers: trends and obstetric outcomes. Womens Health Issues, 1996. 6(5): p. 264–72. [DOI] [PubMed] [Google Scholar]
- 59.Babatunde OA, et al. , Predictors of Retention among African Americans in a Randomized Controlled Trial to Test the Healthy Eating and Active Living in the Spirit (HEALS) Intervention. Ethn Dis, 2017. 27(3): p. 265–272. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 60.Babatunde OAP, et al. , The impact of a randomized dietary and physical activity intervention on chronic inflammation among obese African-American women. Women Health, 2020: p. 1–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 61.Chiriac VF, Baban A, and Dumitrascu DL, Psychological stress and breast cancer incidence: a systematic review. Clujul Med, 2018. 91(1): p. 18–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 62.Choi S, et al. Food access impacts dietary intervention effect among obese African-American women in 142nd Annual Meeting of the American Public Health Association. 2014. American Public Health Association. [Google Scholar]
- 63.Cohen SS, et al. , Parity and breastfeeding in relation to obesity among black and white women in the southern community cohort study. J Womens Health (Larchmt), 2009. 18(9): p. 1323–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 64.Halbert CH, et al. , Collective efficacy and obesity-related health behaviors in a community sample of African Americans. J Community Health, 2014. 39(1): p. 124–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 65.Halbert CH, et al. , Neighborhood Satisfaction and Colorectal Cancer Screening in a Community Sample of African Americans. J Community Health, 2016. 41(1): p. 38–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 66.Palmer JR, et al. , Parity, lactation, and breast cancer subtypes in African American women: results from the AMBER Consortium. J Natl Cancer Inst, 2014. 106(10). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 67.Services, U.S.D.o.H.a.H., The Surgeon General’s call to action to prevent and decrease overweight and obesity. [Rockville, MD: ]: U.S. Department of Health and Human Services, Public Health Service, Office of the Surgeon General; [2001]. Available from: U.S. GPO, Washington. 2001. [PubMed] [Google Scholar]