Community Factors Show Distinct Patterns Across Cancer Types in Nationwide Analysis

Smoking rates emerged as the strongest predictor of lung and colorectal cancer mortality across US counties, while the proportion of non-Hispanic Black residents showed the highest association with breast and prostate cancer deaths, according to a cross-sectional analysis published in JAMA Network Open.1

The study reveals that no single community factor consistently predicts outcomes across all cancer types, suggesting the need for tailored interventions based on specific cancer profiles, lead author Elizabeth Y Rula, PhD, executive director, Harvey L. Neiman Health Policy Institute, in Reston, VA, and colleagues wrote.

Elizabeth Y Rula, PhD

Courtesy of Neiman Health Policy Institute

The research examined 24 community measures, spanning health behaviors, socioeconomic conditions, environmental exposures, and healthcare access, to determine their relative importance in explaining county-level variations in screening, prevalence, and mortality for breast, colorectal, lung, and prostate cancers, the 4 cancer types that account for half of all new cancer diagnoses in the United States.1

The Costs of Cancer

Cancer imposes substantial human and economic costs on the US healthcare system. Approximately 40.5% of Americans will receive a cancer diagnosis during their lifetime, and the disease claims over 600,000 lives annually.2 By 2030, cancer-related healthcare costs could reach $246 billion, 34% higher than 2015 levels, with Medicare and Medicaid bearing 43% of this burden.3

The County Health Rankings model, an evidence-based model for how community characteristics impact health, suggests that 80% to 90% of health outcomes stem from factors outside medical care, including socioeconomic status (40-50%), health behaviors and lifestyle (30-40%), and environmental factors (5-10%).4,5 While previous research has established associations between individual community factors and health outcomes, this study sought to quantify their relative importance for specific cancer types using random forest algorithms, which can capture complex, nonlinear relationships that traditional regression models might miss, investigators explained.

Key Findings Across Cancer Types

The analysis of a nationally representative 5% sample of Medicare fee-for-service beneficiaries (87% aged 65 or older) revealed distinct patterns for each cancer outcome.

For breast cancer, the share of Hispanic residents was most strongly linked to screening rates, while lack of insurance and the proportion of non-Hispanic Black residents were most important for prevalence and mortality, respectively. Lung cancer outcomes were most shaped by environmental and behavioral factors—air pollution and access to primary care influenced screening, limited food access and uninsured rates affected prevalence, and smoking overwhelmingly drove mortality. For colorectal cancer, poverty, unemployment, and smoking emerged as leading factors across screening, prevalence, and mortality. For prostate cancer, environmental risks such as air toxics, indicators of poor physical health, and the proportion of non-Hispanic Black residents were key contributors.

Across cancer types, smoking consistently ranked among the top predictors of mortality, especially for lung and colorectal cancers, with geographic disparities most evident in the South. Socioeconomic hardship, such as poverty and severe housing problems, was also closely tied to lower screening rates and higher prevalence, particularly in the South and parts of the West and East Coast. Environmental exposures (air toxics, air pollution) and racial/ethnic composition (notably Hispanic and non-Hispanic Black population shares) added further context to disparities.

Although access to health care played a smaller overall role than socioeconomic or environmental factors, lack of insurance was notably important for cancer prevalence. Together, these findings underscore that improving cancer outcomes requires addressing the broader social, economic, and environmental conditions of communities—not just access to medical care.

Emerging Themes, Regional Patterns

The authors found that several factors demonstrated importance across multiple cancer types.

• Severe housing problems, eg, overcrowding, high housing costs, and inadequate facilities, ranked highly for prevalence across breast, colorectal, lung, and prostate cancers. Geographic analysis revealed these associations concentrated in Western states and along the East Coast, with particularly strong correlations in the mid-South for breast cancer.
• Insufficient sleep showed unexpectedly high relative importance, exceeding 80% for screening of breast, colorectal, and lung cancers, all four cancer prevalence models, and breast cancer mortality. The authors note that melatonin's role in regulating circadian rhythm may influence tumor development, though research on night-shift workers and breast cancer risk remains inconclusive.
• Environmental factors generally ranked lower than behavioral or socioeconomic measures, consistent with the County Health Rankings model. However, the Environmental Justice Index, which captures cumulative environmental burdens in disadvantaged communities, ranked between second and fourth for all cancer mortality models. Air toxics cancer risk and air pollution showed particular importance for lung and prostate cancer screening.

Among the study's limitations the researchers acknowledged that county-level analysis may mask substantial within-county variation, but smaller geographic units had insufficient sample sizes. The study design cannot establish causality, only associations. The Medicare-based sample limits generalizability to uninsured individuals and those under 65, which translates into missing screening recommendations that begin before Medicare eligibility.

The authors in their discussion reiterated that the importance of any of the individual measures studied varied with "no measure(s) consistently ranking at the top for all cancer types studied. Hence, individual measure importance must be considered uniquely for each outcome–cancer type combination," they stressed. "Our results are consistent with the literature demonstrating the importance of health behaviors and lifestyle, socioeconomic status, and environmental factors on population health"

Moreover, the findings support using both national factor rankings and geospatial mapping to guide cancer prevention efforts, an approach that will allow policymakers to prioritize high-impact factors nationally while addressing region-specific needs revealed through county-level patterns.

References

1. Drake AR, Christensen EW, Ochoa AC , Small Jr W, Scott J, Rula EY. Community factors and county-level screening, prevalence, and mortality. JAMA Netw Open. 2025;8(10):e2537690. doi:10.1001/jamanetworkopen.2025.37690
2. National Cancer Institute. Cancer statistics. Accessed March 6, 2025. https://www.cancer.gov/about-cancer/ understanding/statistics
3. Mariotto AB, Enewold L, Zhao J, Zeruto CA, Yabroff KR. Medical care costs associated with cancer survivorship in the United States. Cancer Epidemiol Biomarkers Prev. 2020;29(7):1304-1312. doi:10.1158/1055-9965. EPI-19-1534
4. Hood CM, Gennuso KP, Swain GR, Catlin BB. County health rankings: relationships between determinant factors and health outcomes. Am J Prev Med. 2016;50(2):129-135. doi:10.1016/j.amepre.2015.08.024
5. Anderson TJ, Saman DM, Lipsky MS, Lutfiyya MN. A cross-sectional study on health differences between rural and non-rural U.S. counties using the County Health Rankings. BMC Health Serv Res. 2015;15:441. doi:10.1186/s12913- 015-1053-3