Differential effects of social determinants on cancer survival by age at diagnosis: A population-based study in Alabama.

Identifying clinical and lifestyle factor mediators of the association between socioeconomic status and colorectal cancer mortality

Purpose Higher SES is associated with lower mortality among colorectal cancer (CRC) survivors. We performed a mediation analysis to identify factors that mediate the association between household income and CRC mortality. Methods Data arise from 1038 participants of the prospective Southern Community Cohort Study (SCCS) who developed incident CRC after enrollment (2002-2009); 73% self-identify as Black. Fifty-nine percent self-reported household income <$15,000/year. Mortality outcomes were determined via the National Death Index. Proportional hazards models were used to estimate hazard ratios (HRs) with 95% confidence intervals (CIs) for mortality by income level. Mediation analysis was used to estimate indirect effects of stage, healthcare-access and use, comorbidities and health behaviors on mortality. Results Higher income was associated with lower risk for all-cause mortality (≥$15,000 vs. <$15,000/year: HR = 0.70, 95%CI:0.58-0.84) and CRC-specific mortality (HR = 0.75, 95%CI:0.59-0.94). Differences in stage and surgical resection together explained 33% of the association with CRC-specific mortality and 16% for all-cause mortality. Smoking history explained 11% of the association with all-cause mortality and 9% for CRC-specific mortality. Diet explained 9% of the association with each outcome. Adjustment for all potential mediators explained 26% and 44% of the associations with all-cause and CRC-specific mortality, respectively. Associations between income and mortality were similar for participants with local/regional tumors only. Conclusions Differences in early detection and surgical resection are the strongest mediators of the association between income and mortality. SES-targeted health policies and patient navigation will improve CRC outcomes. Additionally, health policies to address smoking cessation and food availability will create opportunities for interventions to promote health equity.

Cancer prevalence in the United States: trends and sociodemographic disparities based on national health interview survey data (2019–2023)

Introduction Cancer remains a leading cause of mortality in the United States. This study aimed to examine trends and disparities in the prevalence of selected cancers—any cancer, breast, cervical, prostate, and skin cancer—among U.S. adults from 2019 to 2023. Materials and methods This cross-sectional study used data from the National Center for Health Statistics (NCHS) Interactive Summary Health Statistics for Adults. Cancer prevalence was based on self-reported physician diagnoses. Joinpoint regression analysis assessed temporal trends, with annual percentage change (APC) estimates and 95% confidence intervals (CI) reported. Analyses were stratified by year, gender, age, race/ethnicity, nativity, veteran status, employment status, geographic region, metropolitan statistical area (MSA), and Social Vulnerability Index (SVI). Results The overall prevalence of any cancer remained stable from 2019 (9.6%, 95% CI: 9.3–9.9) to 2023 (9.8%, 95% CI: 9.5–10.1). However, notable disparities were observed. In 2023, White adults had the highest prevalence (11.7%, 95% CI: 11.3–12.2), while Asians had the lowest (3.5%, 95% CI: 2.6–4.8). Breast cancer prevalence in females rose slightly from 3.2% (95% CI: 2.9–3.5) in 2019 to 3.5% (95% CI: 3.2–3.8) in 2023, though not significantly. Cervical cancer in females significantly declined from 1.1% (95% CI: 0.9–1.3) to 0.9% (95% CI: 0.8–1.1) (APC: -6.05, 95% CI: -10.87 to -0.99). Prostate cancer in males rose slightly from 2.3% (95% CI: 2.1–2.6) to 2.5% (95% CI: 2.3–2.8), with Black males having the highest prevalence (3.6%, 95% CI: 2.7–4.7). Skin cancer in females significantly increased from 3.0% (95% CI: 2.7–3.3) to 3.3% (95% CI: 3.0–3.6) (APC: 2.25, 95% CI: 1.11 to 3.36), with the highest prevalence among White adults (4.7%, 95% CI: 4.4–5.0). Higher prevalence was generally observed among older adults, residents of non-MSA areas and the Midwest, individuals with low social vulnerability, the unemployed, U.S.-born individuals, and veterans. Conclusion Although overall cancer prevalence remained relatively stable between 2019 and 2023, significant disparities persist across demographic, geographic, and socioeconomic groups. These findings emphasize the need for targeted cancer control strategies to address ongoing inequities.

Practical considerations for social determinant-based disease prediction in the All of Us research program

Growing recognition that social determinants of health (SDoH) strongly influence health outcomes has expanded their inclusion in biomedical research, underscoring the need to evaluate how best to incorporate these data into disease prediction models. The All of Us (AoU) Research Program is a large, diverse biomedical research dataset that includes participants from across the United States and links electronic health records (EHRs) with extensive survey data covering a wide range of health, lifestyle, and social factors. We assessed selection bias in the SDoH surveys by comparing demographic characteristics across cohorts with varying EHR and survey completion requirements. We additionally used a series of logistic regression models to evaluate the predictive utility of SDoH for nine chronic conditions, compared these results to models using only socioeconomic status (SES), self-reported race and ethnicity, or additional area-level SDoH factors, and discussed the associated trade-offs. Here we show that requiring sufficient individual-level SDoH survey data results in significant selection bias and sample reduction in AoU. We also show that SES alone captures a substantial proportion of the predictive signal from individual-level SDoH data while preserving sample size and mitigating selection bias. Moreover, SES measures provide greater predictive utility than self-reported race and ethnicity, without excluding underrepresented groups. We find disease-specific patterns of association with SDoH and that area-level SDoH metrics contribute to disease prediction independently of individual-level measures. Altogether, we emphasize key analytical considerations and disease-specific trade-offs for the integration of SDoH data into disease prediction models in AoU and similar cohorts.

Association between poverty and lack of insurance on mortality rates at 1, 2, and 5 years in patients with breast cancer metastatic to the spine: a propensity score-matched analysis of 854 patients.

OBJECTIVE Metastatic cancer to the spine presents a growing challenge in the US, with surgery often recommended for stability and palliation. Despite advancements, postoperative complication and mortality rates remain high. In this study, the authors aimed to investigate the impact of poverty and social determinants of health on survival rates among patients with metastatic breast cancer to the spine. METHODS A propensity score-matched retrospective study of patients aged > 18 years receiving a diagnosis of metastatic breast cancer to the spine between 2006 and 2015 was performed using data from the US Surveillance, Epidemiology, and End Results (SEER) sites in the SEER-Medicare database. Poverty census, dual eligibility, race, tumor characteristics (grade, stage, receptor status), community distress score, education level, insurance status at diagnosis, income level, and marital status were evaluated in the model. RESULTS Of 854 patients, 651 resided in zip codes in which 0% to < 20% of the population lived below the poverty line and 203 patients resided in zip codes in which 20%-100% lived below the poverty line. After propensity score matching, patients living in areas with 20%-100% of the population below the poverty line had worse survival at 1 year (HR 1.43, p = 0.009), 2 years (HR 1.45, p = 0.008) and 5 years (HR 1.5, p = 0.007) after diagnosis. Patients with insurance, both non-Medicaid and Medicaid, respectively, had higher survival at 1 year (HR 0.389, p = 0.009) and (HR 0.429, p = 0.006), at 2 years (HR 0.359, p = 0.003) and (HR 0.444, p = 0.008), and at 5 years (HR 0.446, p = 0.01) and (HR 0.513, p = 0.037) compared with patients with no insurance. CONCLUSIONS This is the first study to demonstrate poverty and insurance status as independent predictors of survival at 1, 2, and 5 years in patients with breast cancer metastatic to the spine.

Subjective cognitive decline and impacts among US cancer survivors: Associations with cancer characteristics and social determinants of health.

Trends in U.S. Gynecologic Cancers (2001-2019): Influence of Social Determinants, Geographic Disparities, and Specialist Access on Stage at Diagnosis.

BACKGROUND Disparities in gynecologic cancer stage at diagnosis, driven by race/ethnicity, socioeconomic disadvantage, and geographic barriers to specialty care, contribute to worse outcomes in marginalized populations. The combined influence of social vulnerability and access to gynecologic oncologists on stage at diagnosis remains incompletely characterized at the national level. OBJECTIVE(S) To examine trends in gynecologic cancer stage at diagnosis from 2001 to 2019 and evaluate the effects of race/ethnicity, social vulnerability, and geographic access to gynecologic oncologists on late-stage diagnosis. STUDY DESIGN Population-based cross-sectional study using the United States Cancer Statistics restricted-use database. A total of 1,398,959 individuals diagnosed with endometrial cancer, ovarian cancer, cervical cancer, or uterine sarcoma from 2001 to 2019 were included. Exposures included race/ethnicity, age, and calendar period at the individual level, and Social Vulnerability Index quartile (low, low-medium, medium-high, high) and geographic access to gynecologic oncologists categorized into quintiles via the two-step floating catchment area method at the county level. The primary outcome was late- versus early-stage diagnosis per International Federation of Gynecology and Obstetrics criteria. Multilevel logistic regression estimated adjusted odds ratios and 95% confidence intervals for each cancer type. RESULTS Late-stage diagnosis was significantly more likely among Non-Hispanic Black women across all cancer types (adjusted odds ratio range: 3.22-6.80; all p<0.05) compared to Non-Hispanic White women, as well as among Hispanic women (adjusted odds ratio range: 2.89-3.50) and Asian/Pacific Islander women (adjusted odds ratio range: 2.71-3.22). Residence in high social vulnerability counties was associated with increased odds of late-stage diagnosis (adjusted odds ratio range: 2.64-3.11; all p<0.05) relative to low vulnerability counties. Women in the lowest geographic access quintile, representing counties with no gynecologic oncologist within 100 miles, had higher odds of late-stage diagnosis than those in the highest quintile (adjusted odds ratio range: 2.23-2.86; all p<0.05). Race/ethnicity and social vulnerability interactions were identified for endometrial and cervical cancers, indicating that race/ethnicity associations with late-stage diagnosis were further elevated in higher versus low vulnerability counties. Late-stage cervical cancer diagnoses declined over the study period, while late-stage endometrial cancer and uterine sarcoma diagnoses increased. CONCLUSION(S) Disparities in gynecologic cancer stage at diagnosis are driven by race/ethnicity, social vulnerability, and poor geographic access to specialty care. Targeted strategies, including gynecologic oncology workforce expansion in underserved areas and culturally tailored early detection programs, are needed to reduce late-stage diagnosis and improve outcomes in marginalized communities.