Diminishing returns among Black patients with cancer: the intersection of race and neighborhood socioeconomic status

Abstract

Background.

Despite narrowing racial gaps, disparities persist across cancer types and socioeconomic levels. The “diminishing returns” hypothesis suggests that economic advantage yields fewer health benefits for Black individuals but is largely unexplored in the context of cancer. We examined the diminishing returns among Black and White individuals across cancer types using a nationally representative study population.

Methods.

The study analyzed cancer-specific survival among 5.3 million non-Hispanic Black and White adults diagnosed with primary cancer (2006–2020) using SEER-22. We assessed how race and neighborhood socioeconomic status (SES) jointly affects survival across 21 cancer types. A 10-level race-SES variable was created, using White individuals in the highest SES group as the reference. The main outcome was cancer-specific death. Diminishing returns were defined quantitively and qualitatively as worse survival for Black individuals even at higher SES. Cox models adjusted for demographics and clinical factors, with multiple imputation for missing data. Social gradients were also evaluated.

Results.

Black women showed strong evidence of diminishing returns overall and for seven cancers, especially uterine and breast. A social gradient was also evident in cancers with diminishing returns, except uterine cancer. For Black men, diminishing returns were not observed across all cancers combined but was present in eight cancers—including prostate and colorectal cancer. Most cancers among men exhibited a strong social gradient. Findings were consistent throughout time and after sensitivity analyses restricted to localized and regional disease.

Conclusions.

Higher SES improves cancer survival for White patients but not Black patients, worsening racial disparities for certain cancers.

Introduction

Cancer mortality rates in the United States have declined significantly over the past several decades.¹ Yet, progress has not been evenly distributed, with persistent disparities across racial and ethnic, geographic, and socioeconomic lines. Among these disparities, the most pronounced are those by race, with Black/African American people experiencing among the highest cancer mortality rates compared with other racial and ethnic groups across most major cancer types and for both men and women.¹ Race disparities in cancer mortality are shaped by a complex interplay of social, economic, environmental, and healthcare-related factors and reflect systemic inequities in access to quality healthcare, early detection, and treatment, as well as broader social determinants of health. It is well-established that Black people are more likely than White people to be diagnosed with distant-stage disease, high-grade tumors, and aggressive subtypes of cancers (e.g., triple-negative breast cancer and high-grade non-endometrioid uterine cancers) for which treatment options are limited and the risk of metastasis and recurrence are high.^{1, 2} While unfavorable tumor characteristics, in part, account for race differences in outcomes, a large proportion of the disparity remains unexplained—and there have been few efforts to disentangle the effects of race, socioeconomic status (SES), and place to better understand drivers of these disparities.^{3, 4} Recent data have shown that considering these intersections can yield perplexing results (e.g., weaker associations between SES and health outcomes among Black compared with White individuals), revealing race disparities at even high levels of SES.^{5, 6}

Due to structurally racist laws, policies, and practices, Black Americans are more likely than White Americans to live in deprived, low-socioeconomic neighborhoods.⁷ Thus, Black individuals experience differential access to resources such as high-quality housing, education, employment opportunities, and healthcare. However, they also experience diminished returns upon attaining access to these resources.⁸ The diminishing returns hypothesis (related to the racial non-equivalence of SES)⁹ posits that SES generates fewer health gains for minoritized groups than the majority group.⁸ As such, increasing SES for White individuals results in a clear, linear gain in health and a strong social gradient such that SES and health are strongly correlated (Figure 1). Conversely, there is often a non-linear gain in health with increasing SES for Black individuals, such that health gains level off and sometimes decrease for those of higher SES, resulting in a weaker correlation between SES and health among Black individuals (i.e., a small social gradient) and more pronounced disparities among high SES individuals.^{5, 10} This phenomenon of diminishing returns has been observed across several domains of physical and mental health, including associations of educational attainment and self-rated health,^{9, 11, 12} educational attainment and mortality,^13–16 educational attainment and depressive symptoms,¹⁷ employment status and life expectancy,¹⁸ social contacts and life expectancy,¹⁹ and persistent high income and BMI.²⁰ However, these trends are not well-understood in the context of cancer. Thus, the goal of this study was to explore the hypothesis of diminishing returns among Black compared with White cancer patients using a nationally represenative study population.

Figure 1.

The Diminishing Returns Hypothesis suggests differential effects of SES on health by race with unequal gains with higher SES for non-Hispanic Black vs. non-Hispanic White.

Methods

Study Population and Data Source

The study population consisted of non-Hispanic Black and non-Hispanic White individuals aged ≥20 years who were diagnosed with a first primary cancer between 2006 and 2020 and included in the Surveillance Epidemiology and End Results (SEER)-22 Incidence Specialized Database with census tract attributes (N=6,093,604).²¹ We excluded individuals with incomplete follow-up dates (n=222,908, 3.7%), those with zero days of survival (n=21,019, 0.3%), death certificate or autopsy-only deaths (n=99,718, 1.6%), those missing neighborhood SES data (n=115,645, 2.0%), and those missing stage information (n=350,522, 6.2%). After accounting for exclusions, the analytic sample included 5,283,792 patients (Figure S1).

Exposure Assessment

The primary exposures in this study were race and neighborhood SES. Race and ethnicity data collected by SEER were used to classify patients into two groups: non-Hispanic Black (NHB) and non-Hispanic White (NHW). To examine the role of neighborhood SES on cancer-specific survival, we used the available census tract-based composite measure in the SEER-22 Incidence Specialized Database.²² This measure was constructed following methodologies described by Yu et al.,²³ where a factor analysis derived a neighborhood SES score for each U.S. census tract based on seven indicators from the American Community Survey (ACS) 5-year estimates. These indicators, originally identified by Yost et al.,²⁴ included percent below 150% of the poverty line, median household income, median house value, median rent, percent working class, percent unemployed, and the Education Index.²⁵ Neighborhood SES scores for census tracts across the entire U.S. were calculated for each diagnosis year using corresponding ACS data to reflect patients’ contemporaneous neighborhood conditions and were then categorized into quintiles (Q), with Q1 representing the lowest SES and Q5 the highest SES. Patients were linked to their neighborhood SES quintile by census tract and diagnosis year.

To evaluate joint effects of our exposures of interest, we combined race and neighborhood SES to construct a 10-level exposure variable. Given the relatively superior outcomes observed among White individuals and those with high SES, we designated NHW in Q5 as the reference group for analysis.

Outcome Assessment

The primary outcome of interest was cancer-specific death, defined as death attributable to a particular cancer diagnosis and determined using SEER’s cause-specific death classification.²⁶ Survival time was measured in months from the date of diagnosis to the date of death or the study endpoint on December 31, 2020.

Cancer-specific survival was evaluated for the 15 most prevalent cancers across all race and sex combinations using the methodology outlined in the Annual Report to the Nation on the Status of Cancer.²⁷ Our analysis encompassed 21 malignancies, including cancers of the female breast, prostate, lung and bronchus, colorectal, melanoma of the skin, urinary bladder, non-Hodgkin lymphoma, kidney and renal pelvis, uterine corpus, thyroid, leukemia, pancreas, oral cavity and pharynx, myeloma, liver and intrahepatic bile duct, ovary, brain and other nervous system, stomach, esophagus, cervix uteri, and larynx. We defined diminishing returns as the persistence of racial disparities in cancer survival at higher levels of neighborhood SES (quintiles 4 and 5). To quantify this, we required that both the hazard ratio and the lower bound of the 95% confidence interval for Q5 NHB individuals be greater than the hazard ratio for Q5 NHW individuals. The same criteria were applied for Q4. Diminishing returns were considered present only if these conditions were met for both Q4 and Q5. We also assessed associations for the presence or absence of a social gradient within race groups. Social gradient was defined as a monotonic increase in hazards as neighborhood SES decreases. For quantitative assessment, we fit a weighted linear regression model of the natural logarithm (ln) of the hazard ratios (HR) corresponding to each quintile of neighborhood SES from the multivariable-adjusted Cox models, treating quintile as a numeric predictor, with higher values representing lower neighborhood SES. The regression model was weighted by the inverse of the variance of ln(HR), similar to commonly used meta-regression methods,²⁸ to account for differences in precision across estimates. A statistically significant positive beta coefficient (P < 0.05) was interpreted as evidence of a social gradient.

Statistical Analysis

Descriptive statistics were estimated to summarize the characteristics of the study population at diagnosis, both overall and according to quintile of neighborhood SES. Cox proportional hazards regression models were used to estimate age-adjusted and multivariable-adjusted HRs with 95% confidence intervals (CIs) for the association between the joint race-neighborhood SES exposure and cancer-specific survival. Associations were examined for all cancer sites combined and separately for each cancer site.

Multivariable models included measured covariates identified a priori using graphical methods.²⁹ These covariates included age at diagnosis (continuous, in years), marital status (married or domestic partner; divorced, separated, or widowed; single), and rurality (urban, rural). Although stage may be on the causal pathway for certain cancers and thus not a confounder, it is strongly prognostic for cancer mortality. We therefore adjusted for stage (localized, regional, distant) in all models, except for analyses of leukemia and myeloma as nearly all cases were distant stage. Although only cases of malignant cancer were included in this analysis, urinary bladder includes in situ cases because misalignment between behavior and stage is standard for this cancer type.²⁷ For analyses of urinary bladder cancer, in situ and localized stages were analyzed as separate categories. To address missing covariate data, primarily for marital status, which was missing for 21% of patients, we conducted multiple imputation using the ‘mice’ (Multivariate Imputation by Chained Equations) package in R and derived HRs and corresponding 95% CIs by pooling the multivariable-adjusted estimates and standard errors from Cox regression models fit for m=5 imputed datasets.³⁰

All analyses were stratified by sex (female and male) to account for potential sex-based differences in outcomes. Statistical analyses were conducted using R (version 4.4.1).³¹

Results

Study Characteristics

The relative prevalence of all 21 cancer sites, along with the combined “other” category, is presented in Table S1—overall and stratified by race and sex. The five most common cancers overall were female breast (n=822,153, 16%), prostate (n=806,288, 15%), lung and bronchus (n=658,293, 12%), colorectal (n=463,833, 8.8%), and melanoma of the skin (n=246,391, 4.7%). However, racial differences exist, with melanoma representing a small proportion of cancers among NHB patients (0.2% vs. 5.3%). Among females, the distribution of the four most common cancer sites is relatively consistent by race, whereas among males, prostate cancer disproportionately affects NHB men, accounting for 45% of cases compared with 28% among NHW men.

Table 1 details the frequency (%) of key characteristics of the study population both overall and by neighborhood SES quintiles. The mean age at cancer diagnosis was 64 years (standard deviation = 13). NHB patients comprised just over 13% of the study population and were disproportionately concentrated in low-SES neighborhoods. Most cancers were diagnosed at a localized stage (49%), particularly among individuals residing in high-SES neighborhoods. While prostate and female breast cancers were more prevalent in high-SES neighborhoods, likely reflecting differences in screening rates, lung and colorectal cancers were more frequently diagnosed among individuals in low-SES neighborhoods.

Table 1.

Characteristics of the study population at diagnosis, overall and according to quintiles of neighborhood socioeconomic status, SEER 22, 2006–2020.

Characteristic	Overall, N = 5,283,792	Q5 (highest SES), N = 1,467,970	Q4, N = 1,168,951	Q3, N = 988,204	Q2, N = 887,517	Q1 (lowest SES), N = 771,150
Age at diagnosis, years, mean (SD)	64 (13)	64 (13)	64 (13)	64 (13)	64 (13)	64 (13)
Race/ethnicity and sex
NHB Female	354,179 (6.7%)	35,529 (2.4%)	55,076 (4.7%)	59,231 (6.0%)	75,002 (8.5%)	129,341 (17%)
NHW Female	2,218,540 (42%)	682,355 (46%)	517,587 (44%)	421,900 (43%)	354,587 (40%)	242,111 (31%)
NHB Male	372,564 (7.1%)	39,910 (2.7%)	58,149 (5.0%)	61,693 (6.2%)	76,730 (8.6%)	136,082 (18%)
NHW Male	2,338,509 (44%)	710,176 (48%)	538,139 (46%)	445,380 (45%)	381,198 (43%)	263,616 (34%)
Stage at diagnosis
In situ1	122,402 (2.3%)	36,562 (2.5%)	28,794 (2.5%)	23,490 (2.4%)	19,690 (2.2%)	13,866 (1.8%)
Localized	2,576,625 (49%)	773,899 (53%)	581,437 (50%)	472,705 (48%)	409,522 (46%)	339,062 (44%)
Regional	1,177,200 (22%)	314,302 (21%)	258,557 (22%)	221,843 (22%)	202,588 (23%)	179,910 (23%)
Distant	1,407,565 (27%)	343,207 (23%)	300,163 (26%)	270,166 (27%)	255,717 (29%)	238,312 (31%)
Marital status
Married	2,447,263 (59%)	809,045 (67%)	572,954 (61%)	444,499 (58%)	358,823 (54%)	261,942 (45%)
Single	710,315 (17%)	163,795 (14%)	149,225 (16%)	129,128 (17%)	127,137 (19%)	141,030 (24%)
Divorced/widowed/separated	1,001,910 (24%)	232,854 (19%)	218,543 (23%)	193,298 (25%)	182,310 (27%)	174,905 (30%)
Unknown	1,124,304	262,276	228,229	221,279	219,247	193,273
Rurality
Urban	4,518,515 (86%)	1,456,595 (99%)	1,103,394 (94%)	807,158 (82%)	629,404 (71%)	521,964 (68%)
Rural	765,273 (14%)	11,371 (0.8%)	65,557 (5.6%)	181,046 (18%)	258,113 (29%)	249,186 (32%)
Cancer site
Female Breast	822,153 (16%)	254,405 (17%)	186,578 (16%)	149,121 (15%)	127,379 (14%)	104,670 (14%)
Prostate	806,288 (15%)	246,846 (17%)	180,369 (15%)	145,723 (15%)	126,159 (14%)	107,191 (14%)
Lung and Bronchus	658,293 (12%)	132,394 (9.0%)	134,399 (11%)	130,120 (13%)	131,560 (15%)	129,820 (17%)
Colon and Rectum	463,833 (8.8%)	112,707 (7.7%)	99,071 (8.5%)	89,024 (9.0%)	84,423 (9.5%)	78,608 (10%)
Melanoma of the Skin	246,391 (4.7%)	92,457 (6.3%)	59,490 (5.1%)	42,967 (4.3%)	32,058 (3.6%)	19,419 (2.5%)

Abbreviations: NHB, non-Hispanic Black; NHW, non-Hispanic White; Q, Quintile; SES, Socioeconomic status.

Note: some values do not sum to the total due to missing values

¹As per standard convention in SEER, in-situ urinary bladder cases are classified as malignant and included in analyses of invasive tumors

Diminishing returns and social gradient among women

Across all cancers combined, we observed pronounced diminishing returns among NHB women. NHB women residing in high SES neighborhoods (Q5) had notably worse cancer-specific survival than their White high SES counterparts, with hazards in between those of NHW women in Q3-Q4 SES neighborhoods (Figure 2A and Table S2). A robust social gradient in cancer-specific survival was apparent for both NHW and NHB women for all cancers collectively. Diminishing returns were evident for seven malignancies: breast, colorectal, urinary bladder, non-Hodgkin lymphoma, uterine corpus, leukemia, and ovarian cancer (Figure 3A and Table S3). Of these, uterine corpus cancer demonstrated the most pronounced racial disparity for high SES women, followed by breast cancer. Among those cancers exhibiting diminishing returns, all demonstrated a social gradient in cancer-specific survival among NHB women except uterine corpus for which NHB women had poor survival regardless of neighborhood SES. There were six malignancies where diminishing returns were not apparent but there was a robust social gradient among NHB women (Figure 3B): lung, thyroid, pancreas, oral cavity and pharynx, myeloma, and liver. The remaining cancers (melanoma, kidney and renal pelvis, brain and other nervous system, stomach, esophagus, cervix uteri, and larynx) demonstrated neither prominent diminishing returns or a pronounced social gradient among NHB women (Figure 3B).

Figure 2. Multivariable-adjusted¹ hazard ratios for the association of combined race and neighborhood socioeconomic status (referent group: non-Hispanic White and highest SES) with cancer-specific mortality for (A) female and (B) male cancer patients diagnosed between 2006 and 2020.

Abbreviations: NHB, non-Hispanic Black; NHW, non-Hispanic White; SES, socioeconomic status

¹Adjusted for age at diagnosis, stage, marital status, and rurality.

Figure 3. Multivariable-adjusted¹ hazard ratios for the association of combined race and neighborhood socioeconomic status (referent group: non-Hispanic White and highest SES) with cancer-specific mortality among female cancer patients.

Abbreviations: CRC, colorectal; NHL, non-Hodgkin lymphoma; NHB, non-Hispanic Black; NHW, non-Hispanic White; SES, socioeconomic status

¹Adjusted for age at diagnosis, stage, marital status, and rurality.

Diminishing returns and social gradient among men

In contrast to women, we found minimal evidence of diminishing returns among men across all cancers combined with evidence of a strong social gradient for both NHB and NHW men. In fact, except for those in the highest SES stratum (Q5), Black men exhibited similar or better cancer-specific survival than their White counterparts in equivalent neighborhood SES strata (Figure 2B and Table S2). However, evidence of diminishing returns was observed at eight cancer sites: prostate, colorectal, melanoma of the skin, urinary bladder, non-Hodgkin lymphoma, kidney and renal pelvis, oral cavity and pharynx, and leukemia (Figure 4A and Table S4). Among these malignancies, only melanoma of the skin lacked a clear social gradient among NHB men. Diminishing returns were not apparent, but a social gradient was observed, among NHB men for six cancers: lung, pancreas, myeloma, liver, stomach, and esophagus. Only three cancers among men—thyroid, brain, and larynx—did not display diminishing returns or a social gradient among NHB men.

Figure 4. Multivariable-adjusted¹ hazard ratios for the association of combined race and neighborhood socioeconomic status (referent group: non-Hispanic White and highest SES) with cancer-specific mortality among male cancer patients.

Abbreviations: CRC, colorectal; NHL, non-Hodgkin lymphoma; NHB, non-Hispanic Black; NHW, non-Hispanic White; SES, socioeconomic status

¹Adjusted for age at diagnosis, stage, marital status, and rurality.

Our results were similar when restricted to localized and regional cancers only (Table S5, Figure S2–S3) as well among patients with at least 6 months survival (Table S6, Figure S4–S5). We also estimated sex-specific associations by 5-year intervals (2006–2010, 2011–2015, 2016–2020) for all cancers combined and, similarly, found consistent results across time periods (Figure S6–S7).

We investigated multiple potential factors contributing to diminishing returns, including disparities in cancer incidence, mortality, and survival, as well as the influence of screen-detected, smoking-related, and obesity-related cancers (Figures S8–S9). Except for ovarian cancer among women, cancers demonstrating diminishing returns had better 5-year relative survival rates (>50%) than most cancers without diminishing returns. No trends emerged for the other factors we considered.

Discussion

In this large, first-of-its-kind study, we identified multiple cancers exhibiting patterns consistent with the diminishing returns hypothesis. Among women, these included breast, colorectal, urinary bladder, non-Hodgkin lymphoma, uterine corpus, leukemia, and ovarian cancer. Among men, the cancers demonstrating diminishing returns were prostate, colorectal, melanoma of the skin, urinary bladder, non-Hodgkin lymphoma, kidney and renal pelvis, oral cavity and pharynx, and leukemia. When analyzing all cancers collectively, the impact of diminishing returns was notably more pronounced among NHB women compared to NHB men likely due to the strong pattern of diminishing returns among female-specific cancers (breast, corpus uteri, and ovary). This difference may be driven by systemic inequities in healthcare access and quality, potentially influenced by the intersection of structural racism and structural sexism.

For those cancers with diminishing returns, we found that NHB patients in the highest SES group often had cancer-specific mortality similar to the lowest or second lowest SES group among NHW patients, even with adjustment for stage. This striking finding is masked when examining race-stratified associations rather than using a common referent approach.^{32, 33} For most cancers exhibiting diminishing returns, it appears both race and SES independently impact cancer outcomes. However, in the most extreme case for uterine corpus, SES appeared to have little influence on outcomes among NHB women. Hazards for high SES NHB women were substantially worse than those of even the lowest SES NHW women with almost no social gradient among NHB women. Therefore, race (or some non-SES factor associated with race) may play a larger role for this cancer type.

Other cancers like lung and liver cancer did not exhibit diminishing returns but had robust social gradients among NHB patients that paralleled those of NHW patients. For these cancers, it appears that neighborhood SES strongly influences cancer survival, with race having less influence. The reason for these opposing patterns is unclear. One difference we found is that cancers with diminishing returns generally have higher 5-year relative survival than those without. For these more prognostically favorable cancers, there may be a greater opportunity for the emergence of racialized differences regardless of SES (e.g. continuity of care, treatment of existing or new comorbidities, financial and logistical toxicities) that impact long term survival. Historically, advances in early detection and treatment have worsened racial disparities—notably in breast following the introduction of mammography and hormone therapy.³⁴ It is also possible that we see stronger relative differences in survival for prognostically favorable cancers because ratios of rare events will inherently be larger than ratios of common events. Patterns of diminishing returns should be further studied, especially in light of recent evidence of racial disparities in the use of novel targeted therapies across multiple cancers^35–40 and reduced eligibility for precision oncology therapies among Black patients.⁴¹

The diminishing returns hypothesis offers a compelling framework for understanding why high-SES Black individuals continue to experience disproportionately higher cancer mortality rates. Despite socioeconomic advancement, Black individuals at higher SES levels are more likely to reside in neighborhoods with increased exposure to environmental hazards (e.g., pollution, toxic waste sites),^{42, 43} limited healthcare resources, and heightened social stressors such as over-policing and discrimination.⁴⁴ In contrast, high-SES White neighborhoods tend to offer greater health-protective benefits, including superior infrastructure, lower crime rates, and better access to high-quality healthcare. Even with higher income or education, Black individuals face systemic barriers within the healthcare system, including implicit bias in medical treatment, a lower likelihood of receiving guideline-concordant care, and disparities in access to advanced screening and treatment.^{45, 46} Insurance coverage does not necessarily mitigate these inequities—Black patients often experience longer wait times, reduced access to specialty care, and more frequent delays in treatment.^{47, 48} The weathering hypothesis further explains how the cumulative effects of chronic stress, including racial discrimination, accelerate biological aging and deteriorate health outcomes among Black individuals, even at high SES levels.^{49, 50} High-achieving Black individuals frequently report greater exposure to workplace discrimination, racialized stress, and psychosocial burdens, which can negatively impact immune function and increase susceptibility to aggressive cancers.⁵¹ These stressors may not only elevate cancer risk but also contribute to disease progression. SES is commonly measured by income or education, yet wealth—assets, homeownership, and generational financial stability—is a more powerful determinant of long-term health outcomes. Due to historical and structural barriers such as redlining, employment discrimination, and inequitable access to financial resources, Black individuals with comparable incomes to White individuals often possess significantly less wealth.^{52, 53} This persistent racial wealth gap limits access to high-quality healthcare, stable living conditions, and other protective factors, ultimately reducing opportunities for optimal health outcomes.

Several limitations should be considered when interpreting our findings. Cause-of-death information obtained from death certificates, may be incomplete or potentially misclassified, which could bias cancer-specific survival estimates.⁵⁴ However, all-cause mortality would not be appropriate across the diverse cancer sites in our analysis, as it would disproportionately reflect non-cancer causes of death for prognostically favorable cancers.⁵⁴ Furthermore, SEER lacks key patient-level data such as comorbidities, insurance status, and detailed treatment information, which are both prognostic and socially patterned. Future studies with more detailed data should explore their mediating role.

We relied on a single measure of neighborhood socioeconomic deprivation, findings from which may not generalize across various indices. However, a recent study showed that this measure performed similarly to other commonly used measures (Area Deprivation Index and Neighborhood Deprivation Index) in estimating breast cancer mortality, supporting the robustness of our approach.⁵⁵ Because this SEER database does not include geographic identifiers, we were unable to account for spatial clustering of cases within a census tract.²² This would most likely impact estimates for all sites combined and across all years.²² However, sensitivity analyses showed similar results for analyses of all sites across time intervals. We observed imprecision in estimates for select cancer sites due to small sample sizes, particularly among high SES Black individuals. The limited representation of Black individuals in affluent neighborhoods constrained our ability to assess cancer mortality patterns within this subgroup, potentially underestimating the presence of diminishing returns. Finally, while neighborhood deprivation indices capture multiple dimensions of socioeconomic conditions, they do not allow us to disentangle which specific neighborhood factors— built, chemical, food, physical, or social environments—are driving the observed associations with cancer mortality. The use of composite indices may obscure the relative contributions of these individual components, limiting the precision of our conclusions. Similarly, race is an ill-defined exposure and serves as a proxy for structural inequities in addition to biologic factors such as ancestry. Future research should aim to isolate these specific neighborhood and race characteristics to better inform interventions.

This study is the first to analyze trends in cancer-specific survival by both race and neighborhood SES across all cancer sites, offering critical insight into the amplified racial disparities observed in high-SES neighborhoods. While living in more affluent areas is generally associated with markedly improved cancer outcomes for White individuals, Black residents derive little to no comparable advantage for some cancers. Our findings support a dual model of disparities—one driven by differential access to resources among low-SES Black individuals and the other by diminishing returns for their high-SES counterparts. Moreover, it suggests that increasing neighborhood SES alone is insufficient to eliminate racial gaps in cancer outcomes..

Context Summary.

Key objective:

We examined whether the “diminishing returns” hypothesis applies to cancer survival, testing if higher neighborhood socioeconomic status (SES) confers fewer survival benefits for Black compared with White adults across 21 cancers using SEER-22 data from 2006–2020.

Knowledge generated:

Black women exhibited diminished survival gains at higher neighborhood SES overall and for seven cancers, particularly breast and uterine. Black men showed diminishing returns in eight cancers, including prostate and colorectal, while most cancers demonstrated a social gradient across SES within race groups.

Relevance (written by Stephanie Wheeler):

Socioeconomic status confers less advantage to Black adult populations with cancer compared with White adult populations with cancer. Understanding and intervening upon the social mechanisms underlying these differences is critical to achieving equity in cancer outcomes.