Rural-Urban Continuum Codes Research Articles

Male Factor Infertility and the Rural—Urban Continuum

OBJECTIVETo examine the association between male factor infertility and the Rural—Urban continuum. MATERIALS AND METHODSSingle institution retrospective cohort study using the Utah Population Database (UPDB), which combines demographic, medical, and residential data for patients residing in Utah and links to the Subfertility Health Assisted Reproduction and Environment (SHARE) database, which houses fertility data from 1998-2017. The data was divided by metropolitan (metro-) [large, medium, small], and non-metropolitan (non-metro-) [urban, rural] status, based on United States Department of Agriculture (USDA) rural-urban continuum codes (RUCC). RESULTSNon-metro urban/rural males were less likely to be a racial/ethnic minority (91.3% non-Hispanic white vs 85.7%), or use assisted reproductive technology (ART) (13.4% vs 18.5%). Multivariate regression controlling for race/ethnicity, age, semen analysis category (oligozoospermic vs normozoospermic), previous successful fertility outcome, and use of ART, demonstrated complete rurality was associated with decreased likelihood of successful fertility outcome (Hazard Ratio [HR] 0.63, 95% CI 0.44–0.91, n=65). Non-metro urban individuals trended towards lower likelihood of successful fertility outcome ([HR] 0.93, 95% CI 0.85–1.01). Complete rurality was associated with longer time for 50% cohort successful fertility outcome (>60 months vs approximately 34 months both metro). All comparisons p<0.001. CONCLUSIONSAcross the rural—urban continuum, residing in a metro area was associated with higher rates of racial/ethnically diversity, fertility treatment utilization, and successful fertility outcomes (live births). Given approximately 18% of the United States resides with a rural community (12% in Utah), these findings can provide more informed infertility care.

A Demographic Description of Children and Adolescents With Concomitant Autism and Anxiety Along With a Comparison of Receipt of Psychological Services

Objective: (1) Evaluate differences in demographic distributions between those with autism alone and those with autism concomitant with anxiety and (2) assess demographic factors associated with receipt of psychological therapy in patients with autism. Methods: Using deidentified health care claims data, we extracted all encounters among children with the International Classification of Diseases-10-CM for Autism Spectrum Disorder (F84) and flagged those with concomitant anxiety (F41.9). Extracted data included age, sex, race, ethnicity, insurance status, and Rural-Urban Continuum Codes. Current procedural terminology codes for psychological services were extracted. Data were analyzed using Mann–Whitney U tests, χ2 tests of association, and logistic regression. Results: Demographic differences between children/adolescents with autism alone versus autism concomitant with anxiety were noted for age, sex, and race/ethnicity, with a greater prevalence of autism/anxiety identified in older-aged children and children identifying as female and non-Hispanic White residing in an urban area. While 21% of those included in this study had received psychotherapy, the percentages within the autism only and autism/anxiety groups were 12% and 50%, respectively (p &lt; 0.001). Children/adolescents who were older age quartile, resided in an urban residence, and with a concomitant diagnosis of anxiety were more likely to receive psychological services, while non-Hispanic Other were less likely to, compared with the reference of non-Hispanic White. Conclusion: We identified demographic disparities in the diagnosis of autism versus autism with concomitant anxiety, including age, sex, race/ethnicity, and residence location. Age quartile, urban location, and concomitant anxiety were associated with increased likelihood of receipt of psychological services.

Identifying measures for understanding and addressing county-level disparities in adolescent HPV vaccination coverage in North Carolina.

HPV vaccination coverage is characterized by geographic disparities in the US, with national studies finding lower coverage in rural versus nonrural areas. To direct quality improvement efforts in North Carolina, we sought to understand how different rurality measures characterize these disparities. We used separate negative binomial regression models to test associations between 5 dichotomized county-level rurality measures and HPV vaccination coverage (≥1 dose) among North Carolina adolescents, ages 11-12 (n=326,345). Rurality measures were derived from: Office of Management and Budget's Metropolitan Statistical Areas, Rural-Urban Continuum Codes, Index of Relative Rurality, US Census Bureau classifications, and North Carolina Rural Center classifications. Models controlled for Social Vulnerability Index (SVI) percentile and rate of pediatricians per county. Vaccination data came from the North Carolina Immunization Registry. HPV vaccination coverage was 29% across North Carolina's 100 counties (range: 13%, 46%). Agreement between rurality measures ranged from 54% to 93% of counties. In adjusted analyses, none of the 5 rurality measures correlated with HPV vaccination coverage, but higher SVI and higher rate of pediatricians were positively associated with coverage (p< 0.01). Exploratory moderation analyses suggested regional variation in the relationship between rurality and coverage, with a positive association in one region, a negative association in one region, and no association in four regions. County-level rurality measures did not identify disparities in HPV vaccination coverage in North Carolina. Measures related to social vulnerability and access to pediatricians may be better suited for understanding and addressing the state's substantial county-level vaccination disparities.

Association between obesogenic environments and childhood overweight/obesity across the United States: Differences by rurality.

Childhood obesity is more prevalent in rural compared to urban communities and may be related to urban-rural differences in environmental factors known to affect obesity. However, understanding of how environmental factors impact childhood obesity in rural settings remains limited. This study aimed to address this gap by exploring the relationship between obesogenic environments and childhood overweight/obesity rates, including variations across the urban-rural continuum. This study analyzed data for counties in the United States (N = 3140). Linear regression models were employed to examine the relationship between the Childhood Obesogenic Environment Index, which consisted of ten variables from a variety of sources associated with physical activity and healthy eating, and childhood overweight/obesity rates estimates derived from the 2016 National Survey of Children's Health. County rurality was categorized using Rural-Urban Continuum Codes and a moderation analysis was conducted to identify potential variations by rurality. There was a significant positive association between the COEI and childhood overweight/obesity rates nationally, with notable variations across the urban-rural continuum for specific index components. Neighborhood walkability showed a significant positive association across rurality, indicating that childhood overweight/obesity rates were higher in less walkable communities. Full-service restaurants exhibited an inverse relationship with childhood overweight/obesity rates across all RUCC levels. These results underscore the obesogenic environmental factors associated with childhood overweight/obesity rates nationally and how they vary across the urban-rural continuum. This study highlights the importance of considering these variations when designing interventions to address childhood obesity.

Building partnerships to improve clinical trial accrual in rural and Native American communities through community health educator interventions.

180 Background: Rural populations and racial minorities are often underrepresented in cancer clinical trials. The low representation is determined by multilevel factors, including lack of geographic access to facilities in rural locations and mistrust in the medical and scientific community. Presented here are partial results of an educational intervention to improve clinical trials-related knowledge among rural residents and racial minorities. Methods: The Center to Reduce Cancer Health Disparities (CRCHD) from the National Cancer Institute (NCI) developed the study materials, that were available in English or Spanish, to be utilized according to participants' preferences. The University of Wisconsin Carbone Cancer Center implemented this intervention in northern Wisconsin, which consisted of a baseline questionnaire, a recorded educational intervention, and a post-educational questionnaire. The chosen community is largely rural and Native. The questionnaires consisted of 11-knowledge- and 6-intention-related questions. After being screened by the rural Community Health Educator (CHE), subjects self-administered the intervention online or received the intervention from the CHE during tribal or other educational group events in rural communities. Results: Between January and May 2024, 62 participants started the intervention. Participants’ mean age was 45.1 (±15.36) years, 64.5% (40/62) were female, and 53.2% (33/62) self-identified as American Indian/ Alaska Native. All resided in rural community, defined as Rural-Urban Continuum Codes 7 or 8. Three participants did not respond to the post-educational questionnaire, and one skipped its intentions section. Although the two questions about informed consent presented the highest percent of correct answers post-intervention, the median total knowledge score pre and post-intervention remained the same (p&gt;0.05). Post-intervention, participants more often agreed it is important for minority groups to participate in clinical trials, reported being likely to search for a clinical trial they might be eligible for, join a clinical trial, and talk to family/friends about joining a clinical trial (p&lt;0.05). Conclusions: The CHE successfully engaged rural community members to participate in the intervention. In addition, the intervention facilitated building trust and partnerships for future cancer-related community driven work.

An Analysis of the Otolaryngology Workforce in Pennsylvania.

This study aims to analyze the distribution of otolaryngologists between urban and rural counties in Pennsylvania. Retrospective database. American Academy of Otolaryngology-Head and Neck Surgery (AAO-HNS) database and Centers for Medicare and Medicaid Services (CMS) Database. The AAO-HNS database (Entnet.org) and the CMS Database were used to identify otolaryngologists, their primary location, and their state of training. The 2023 Rural-Urban Continuum Codes were used to classify metropolitan (codes 1-3) and nonmetropolitan counties (codes 4-9) and determine county populations. A total of 388 otolaryngologists were identified across 67 Pennsylvania counties (33 metropolitan, 34 nonmetropolitan), 80% of which were male. There were 354 otolaryngologists in metropolitan counties serving 11.2 million people, with an average of 2.4 otolaryngologists per 100,000 people (/100 k). The majority of otolaryngologists (n = 235, 60.6%) are located in counties designated by code 1. There were 34 otolaryngologists in nonmetropolitan counties serving 1.8 million people, with an average of 2.4 otolaryngologists/100 k. Montour County (code 6) is home to a large academic center and accounts for 14/34 otolaryngologists in the nonmetropolitan counties. With the exception of Montour County, the remaining nonmetropolitan counties averaged 1.0 otolaryngologist/100 k. Only 3 metropolitan counties did not have otolaryngologists compared to 19 nonmetropolitan counties without otolaryngologists. Most otolaryngologists received their residency training in Pennsylvania (n = 177). New York (n = 35) and Maryland (n = 24) were the second most common states for training. There is a lack of otolaryngologists in rural counties of Pennsylvania, except in counties home to large academic centers.

Proximity to cancer rehabilitation and exercise oncology by geography, race, and socioeconomic status.

Cancer rehabilitation and exercise oncology (CR/EO) have documented benefits for people living with and beyond cancer. The authors examined proximity to CR/EO programs across the United States with respect to population density, race and ethnicity, socioeconomic status, and cancer incidence and mortality rates. This cross-sectional study was conducted in 2022-2023. Online searches were initiated to identify CR/EO programs. Geocoding was used to obtain latitudinal and longitudinal geospatial coordinates. Demographic data were abstracted from the 2020 5-year American Community Survey. Cancer incidence and mortality data were obtained from the Centers for Disease Control and Prevention. US 2013 Rural-Urban Continuum Code (RUCC) classification was used to define counties as either urban (RUCC 1-3) or rural (RUCC 4-9). Multivariable logistic regression was used to evaluate the association between being far from a program and census-tract level factors. In total, 2133 CR/EO programs were identified nationwide. The distance from a program increased with decreasing population density: rural tracts were 17.68±0.24 miles farther from a program compared with urban tracts (p<.001). Program proximity decreased as the neighborhood deprivation index increased (p<.001). Exercise oncology programs were less common than cancer rehabilitation programs in tracts with a larger proportion of minority residents (p<.001). Prior research has documented that underrepresented populations have worse cancer-related symptoms and higher cancer mortality. Herein, the authors document their findings that these same populations are less likely to have proximity to CR/EO programs, which are associated with improved cancer-related symptoms and cancer mortality outcomes. To realize the positive outcomes from CR/EO programming, efforts must focus on supporting expanded programming and sustainable payment for these services.

Evolving patterns of COVID-19 mortality in US counties: A longitudinal study of healthcare, socioeconomic, and vaccination associations.

The COVID-19 pandemic emphasized the need for pandemic preparedness strategies to mitigate its impacts, particularly in the United States, which experienced multiple waves with varying policies, population response, and vaccination effects. This study explores the relationships between county-level factors and COVID-19 mortality outcomes in the U.S. from 2020 to 2023, focusing on disparities in healthcare access, vaccination coverage, and socioeconomic characteristics. We conduct multi-variable rolling regression analyses to reveal associations between various factors and COVID-19 mortality outcomes, defined as Case Fatality Rate (CFR) and Overall Mortality to Hospitalization Rate (OMHR), at the U.S. county level. Each analysis examines the association between mortality outcomes and one of the three hierarchical levels of the Social Vulnerability Index (SVI), along with other factors such as access to hospital beds, vaccination coverage, and demographic characteristics. Our results reveal persistent and dynamic correlations between various factors and COVID-19 mortality measures. Access to hospital beds and higher vaccination coverage showed persistent protective effects, while higher Social Vulnerability Index was associated with worse outcomes persistently. Socioeconomic status and vulnerable household characteristics within the SVI consistently associated with elevated mortality. Poverty, lower education, unemployment, housing cost burden, single-parent households, and disability population showed significant associations with Case Fatality Rates during different stages of the pandemic. Vulnerable age groups demonstrated varying associations with mortality measures, with worse outcomes predominantly during the Original strain. Rural-Urban Continuum Code exhibited predominantly positive associations with CFR and OMHR, while it starts with a positive OMHR association during the Original strain. This study reveals longitudinal persistent and dynamic factors associated with two mortality rate measures throughout the pandemic, disproportionately affecting marginalized communities. The findings emphasize the urgency of implementing targeted policies and interventions to address disparities in the fight against future pandemics and the pursuit of improved public health outcomes.

Are there place-based disparities in mortality risk? Findings from two longitudinal studies.

Most work on place-based (e.g., rural-urban) health disparities has been conducted with population-level data, which is limited in its capacity for causal inferences about individuals and lifespan health. The present study leverages individual-level longitudinal data, spanning up to 29 years, to understand how rurality-urbanicity predicts risk for all-cause mortality; whether these associations hold above and beyond socioeconomic status (SES); and whether the association between rurality-urbanicity and mortality risk varies by sex, SES, race, ethnicity, and partner status. The present preregistered study uses data from two large longitudinal studies of U.S. Americans (Health and Retirement Study and Midlife in the United States; total N = ∼55,000), who reported on their sociodemographic characteristics, had their addresses linked to geographical indicators (i.e., rural-urban continuum codes), and have data from the National Death Index regarding the vital status and survival time. Using Cox proportional hazards regression models, findings showed that suburban and rural residents were at a 12% and 18% greater risk for earlier mortality compared to urban residents in Health and Retirement Study, but the associations between rurality-urbanicity and mortality risk were nonsignificant in Midlife in the United States. The longitudinal associations between rurality-urbanicity and mortality risk were largely independent of SES. Finally, there was only one statistically significant interaction effect, suggesting the strength and direction of the association between rurality-urbanicity and mortality risk was largely the same across sociodemographic subgroups. There is tentative evidence suggesting that rurality-urbanicity is an important social determinant of longevity, over and above other sociodemographic factors. Future studies should explore how to promote longer and healthier lives among rural residents. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

A closer examination of the "rural mortality penalty": Variability by race, region, and measurement.

Racial health disparities are well documented and pervasive across the United States. Evidence suggests there is a "rural mortality penalty" whereby rural residents experience poorer health outcomes than their urban counterparts. However, whether this penalty is uniform across demographic groups and U.S. regions is unknown. To assess how rural-urban differences in mortality differ by race (Black vs. White), U.S. region, poverty status, and how rural-urban status is measured. Age-standardized mortality rates (ASMRs)/100,000 by U.S. county (2015-2019) were obtained by race (Black/White) from the CDC Wonder National Vital Statistics System (2015-2019) and were merged with county-level social determinants from the US Census Bureau and County Health Rankings. Multivariable generalized linear models assessed the associations between rurality (index of relative rurality [IRR] decile, rural-urban continuum codes, and population density) and race-specific ASMR, overall, and by Census region and poverty level. Overall, average ASMR was significantly higher in rural areas than urban areas for both Black (rural ASMR = 949.1 per 100,000vs. urban ASMR = 857.7 per 100,000) and White (rural ASMR = 903.0 per 100,000vs. urban ASMR = 791.6 per 100,000) populations. The Black-White difference was substantially higher (p < 0.001) in urban than in rural counties (65.1 per 100,000vs. 46.1 per 100,000). Black-White differences and patterns in ASMR varied notably by poverty status and U.S. region. Policies and interventions designed to reduce racial health disparities should consider and address key contextual factors associated with geographic location, including rural-urban status and socioeconomic status.

The contribution of rural/urban residence to incidence and survival in thymoma and thymic carcinoma, a retrospective cohort study of the SEER 2000–2020 database

ObjectiveRural-urban healthcare disparities have been demonstrated throughout the United States, particularly in acquiring oncologic care. In this study, we aim to discern the role of rural-urban health disparities in thymic cancer incidence and uncover potential survival disparities. MethodsThe Surveillance, Epidemiology, and End Results (SEER) 17-State database was queried for all cases of thymoma (ICD-O-3/3 codes: 8580–8585) and thymic carcinoma (8586) located in the thymus (primary site code C37.9) diagnosed between 2000 and 2020. Residence was established using SEER Rural-Urban Continuum Codes. Incidence trend modeling for rural versus urban patients was completed using Joinpoint Regression Software. Chi-square, Kaplan-Meier with log-rank testing, and Cox proportional hazards was completed using SPSS, with significance set to p <0.05. ResultsJoinpoint analysis revealed a significant growth in incidence in the urban population compared to a stagnant incidence among the rural population. Disease specific survival was higher among urban patients on univariate modeling (p = 0.010), and confirmed on multivariate analysis, whereby rural living conferred an adjusted hazard ratio of 1.263 (95 % CI 1.045–1.527; p = 0.016) in comparison to urban patients. ConclusionsThese findings demonstrate differences between thymic cancer incidence and outcomes in patients living in urban versus rural environments and demonstrate an important disparity.

Uveal melanoma distant metastasis prediction system: A retrospective observational study based on machine learning.

Uveal melanoma (UM) patients face a significant risk of distant metastasis, closely tied to a poor prognosis. Despite this, there is a dearth of research utilizing big data to predict UM distant metastasis. This study leveraged machine learning methods on the Surveillance, Epidemiology, and End Results (SEER) database to forecast the risk probability of distant metastasis. Therefore, the information on UM patients from the SEER database (2000-2020) was split into a 7:3 ratio training set and an internal test set based on distant metastasis presence. Univariate and multivariate logistic regression analyses assessed distant metastasis risk factors. Six machine learning methods constructed a predictive model post-feature variable selection. The model evaluation identified the multilayer perceptron (MLP) as optimal. Shapley additive explanations (SHAP) interpreted the chosen model. A web-based calculator personalized risk probabilities for UM patients. The results show that nine feature variables contributed to the machine learning model. The MLP model demonstrated superior predictive accuracy (Precision = 0.788; ROC AUC = 0.876; PR AUC = 0.788). Grade recode, age, primary site, time from diagnosis to treatment initiation, and total number of malignant tumors were identified as distant metastasis risk factors. Diagnostic method, laterality, rural-urban continuum code, and radiation recode emerged as protective factors. The developed web calculator utilizes the MLP model for personalized risk assessments. In conclusion, the MLP machine learning model emerges as the optimal tool for predicting distant metastasis in UM patients. This model facilitates personalized risk assessments, empowering early and tailored treatment strategies.

Geographic distance to Commission on Cancer-accredited and nonaccredited hospitals in the United States.

The Commission on Cancer (CoC) establishes standards to support multidisciplinary, comprehensive cancer care. CoC-accredited cancer programs diagnose and/or treat 73% of patients in the United States. However, rural patients may experience diminished access to CoC-accredited cancer programs. Our study evaluated distance to hospitals by CoC accreditation status, rurality, and Census Division. All US hospitals were identified from public-use Homeland Infrastructure Foundation-Level Data, then merged with CoC-accreditation data. Rural-Urban Continuum Codes (RUCC) were used to categorize counties as metro (RUCC 1-3), large rural (RUCC 4-6), or small rural (RUCC 7-9). Distance from each county centroid to the nearest CoC and non-CoC hospital was calculated using the Great Circle Distance method in ArcGIS. Of 1,382 CoC-accredited hospitals, 89% were in metro counties. Small rural counties contained a total of 30 CoC and 794 non-CoC hospitals. CoC hospitals were located 4.0, 10.1, and 11.5 times farther away than non-CoC hospitals for residents of metro, large rural, and small rural counties, respectively, while the average distance to non-CoC hospitals was similar across groups (9.4-13.6 miles). Distance to CoC-accredited facilities was greatest west of the Mississippi River, in particular the Mountain Division (99.2 miles). Despite similar proximity to non-CoC hospitals across groups, CoC hospitals are located farther from large and small rural counties than metro counties, suggesting rural patients have diminished access to multidisciplinary, comprehensive cancer care afforded by CoC-accredited hospitals. Addressing distance-based access barriers to high-quality, comprehensive cancer treatment in rural US communities will require a multisectoral approach.

Geographic Access to Pediatric Orthopedic Surgeons in the United States: An Analysis of Sociodemographic Factors.

It is unclear how pediatric orthopedic surgeons are geographically distributed relative to their patients. The purpose of this study was to evaluate the geographic distribution of pediatric orthopedic surgeons in the United States. County-level data of actively practicing pediatric orthopedic surgeons were identified by matching several registries and membership logs. Data were used to calculate the distance between counties and nearest surgeon. Counties were categorized as "surgeon clusters" or "surgeon deserts" if the distance to the nearest surgeon was less than or greater than the national average and the average of all neighboring counties, respectively. Cohorts were then compared for differences in population characteristics using data obtained from the 2020 American Community Survey. A total of 1197 unique pediatric orthopedic surgeons were identified. The mean distance to the nearest pediatric orthopedic surgeon for a patient residing in a surgeon desert or a surgeon cluster was 141.9±53.8 miles and 30.9±16.0 miles, respectively. Surgeon deserts were found to have lower median household incomes (P<.001) and greater rates of children without health insurance (P<.001). Multivariate analyses showed that higher Rural-Urban Continuum codes (P<.001), Area Deprivation Index scores (P<.001), and percentage of patients without health insurance (P<.001) all independently required significantly greater travel distances to see a pediatric orthopedic surgeon. Pediatric orthopedic surgeons are not equally distributed in the United States, and many counties are not optimally served. Additional studies are needed to identify the relationship between travel distances and patient outcomes and how geographic inequalities can be minimized. [Orthopedics. 2024;47(4):e204-e210.].

Differences in Rural Versus Urban Patients With Prostate Cancer in Diagnosis and Treatment: An Analysis of a Population-Based Cohort.

Patients living in rural communities have greater barriers to cancer care and poorer outcomes. We hypothesized that rural patients with prostate cancer have less access and receive different treatments compared with urban patients. We used a population-based prospective cohort, the North Carolina Prostate Cancer Comparative Effectiveness and Survivorship Study, to compare differences in prostate cancer diagnosis, access to care, and treatment in patients by geographic residence. The 2013 rural-urban continuum code (RUCC) was used to determine urban (RUCC 1-3) versus rural (RUCC 4-9) location of residence. Patients with rural residence comprised 25% of the cohort (364 of 1,444); they were less likely to be White race and had lower income and educational attainment. Rural patients were more likely to have <12 cores on biopsy (47.1% v 35.7%; P < .001) and less likely (40.8% v 47.6%; P = .04) to receive multidisciplinary consultation. We observed significant differences in treatment between urban and rural patients, including rural patients receiving less active surveillance or observation (22.6% v 28.7%), especially in low-risk cancer (33.2% v 40.7%). On multivariable analysis that adjusted for patient and diagnostic factors, rural residence was associated with less use of active surveillance or observation over radical treatment (ie, surgery or radiation therapy; odds ratio, 0.49 v urban; P < .001) in patients with low-risk cancer. Patients with prostate cancer who live in rural versus urban areas experience several differences in care that are likely clinically meaningful, including fewer cores in the diagnostic biopsy, less utilization of multidisciplinary consultation, less use of active surveillance, or observation for low-risk disease. Future studies are needed to assess the efficacy of interventions in mitigating these disparities.

Abstract PS04-08: Sociodemographic Risk Factors and Prediction of Aromatase Inhibitor Non-Adherence in Women with Breast Cancer Enrolled in SWOG S1105

Abstract Background: Non-adherence to aromatase inhibitors (AIs) for breast cancer is common and increases risk of recurrence. Few prospective studies have systematically evaluated non-adherence and non-clinical factors associated with it. We analyzed baseline sociodemographic characteristics and financial factors associated with non-adherence to create a baseline composite risk-score for AI non-adherence. Methods: Patients enrolled in S1105 were required to have been on an AI for ≥30 days at enrollment. Patients were assessed for non-adherence of AIs every 3 months for 36 months, with non-adherence defined as urine AI metabolite assay results satisfying any of the following: &amp;lt; 10 ng/mL, undetectable, specimen submitted outside of the ± 21-day follow-up appointment window, or no submitted specimen. Factors included race, Hispanic ethnicity, age, education, income, health insurance type, and prescription cost. We also included socioeconomic deprivation and rural or urban residence using trial participants’ residential zip codes linked to the Area Deprivation Index (ADI) and Rural Urban Continuum codes (RUCC) respectively. For numeric factors and categorical factors with more than two levels, we used cutpoint analysis to determine the highest risk range by optimizing the Chi-squared statistic. For each resulting binary factor, we performed a logistic regression, adjusting for the study stratification factors and stratifying by treatment arm. Factors associated with adherence at the alpha=0.10 level were combined into a composite risk score, with study participants given one point for each baseline adverse risk factor. Secondarily, we examined a composite risk score including all factors, regardless of statistical significance. Results: In total, 724 patients were registered from 40 institutions between May 2012 and September 2013. The median age was 60.9 years, and 64.5% were on AI &amp;lt; 12 months prior to registration. Only 8% had out of pocket (OOP) cost &amp;gt;$30. Observed adherence at 36 months was 35.9%. Patients living in metro areas with populations ≥ 250,000 (RUCC ≤ 3) were more likely to be non-adherent (37.5% vs 28.3%, OR=1.50, p=.08); as were those who were younger, those without a college education, and those with limited (≤$5) OOP cost. ADI, race and ethnicity were not associated with non-adherence. The composite risk score was strongly associated with 36-month adherence. For each 1 unit increase in risk level, the risk of non-adherence increased 47% (OR=1.47, p&amp;lt; .001); the risk increased 64% for those with &amp;gt;2 vs. ≤2 risk factors (OR=1.64, p=.002). Results using all factors regardless of statistical significance were consistent. Conclusions: A composite model comprised of sociodemographic risk factors can identify patients on AI’s at much greater risk of long-term AI non-adherence. In addition to mitigating side effects, targeted interventions to improve adherence should focus on structural barriers among those at highest risk. Non-adherence (+/- 6 months) = factor + AI duration + AI type, stratified by arm Citation Format: Dawn Hershman, Anna Moseley, Kathryn Arnold, Julie Gralow, Alfred Neugut, Scott Ramsey, N. Lynn Henry, Joseph Unger. Sociodemographic Risk Factors and Prediction of Aromatase Inhibitor Non-Adherence in Women with Breast Cancer Enrolled in SWOG S1105 [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PS04-08.

Incidence Rates and Outcomes of Nodular and Superficial Spreading Melanoma along the Rural-Urban Continuum in Kentucky and Their Relationship to Provider Care Access.

This study compared incidence rates, stage at presentation, and cause-specific mortality of nodular and superficial spreading melanoma along the rural-urban continuum in Kentucky. We compared resulting patterns in our data with sample demographic and other potential factors, including population by county and primary care provider rate. Retrospective patient data were extracted from the Surveillance, Epidemiology, and End Results database from 2010 through 2017. These data were supplemented by environmental, demographic, and socioeconomic data derived from publicly accessible databases. Correlation and χ2 analyses were used to test for significant differences in outcome variables by US Department of Agriculture Rural-Urban Continuum Code (RUCC) categories and other potential predictor variables. Incidence rates by Kentucky county were not associated with RUCC or population; likewise, there was no relationship between stage at presentation and RUCC category. There was, however, a highly significant association between cause-specific mortality and RUCC; patients from rural areas were significantly more likely to die from melanoma than those in urban areas. This overall difference was due to differences in mortality for superficial spreading melanoma. Our results suggest that a disparity in patients' ability or tendency to access primary care and/or specialist providers postdiagnosis may be critical factors in determining the ultimate outcome of a melanoma diagnosis. Further studies should explore the availability of dermatologists and/or treatment options for melanoma in rural areas. Our data also provide additional support for inclusion of melanoma subtype in the American Joint Committee on Cancer guidelines.

Epidemiological approaches to multivariable models of health inequity: A study of race, rurality, and occupation during the COVID-19 pandemic

PurposeMethods for assessing the structural mechanisms of health inequity are not well established. This study applies a phased approach to modeling racial, occupational, and rural disparities on the county level. MethodsRural counties with disparately high rates of COVID-19 incidence or mortality were randomly paired with in-state control counties with the same rural-urban continuum code. Analysis was restricted to the first six months of the pandemic to represent the baseline structural reserves for each county and reduce biases related to the disruption of these reserves over time. Conditional logistic regression was applied in two phases—first, to examine the demographic distribution of disparities and then, to examine the relationships between these disparities and county-level social and structural reserves. ResultsIn over 200 rural county pairs (205 for incidence, 209 for mortality), disparities were associated with structural variables representing economic factors, healthcare infrastructure, and local industry. Modeling results were sensitive to assumptions about the relationships between race and other social and structural variables measured at the county level, particularly in models intended to reflect effect modification or mediation. ConclusionsMultivariable modeling of health disparities should reflect the social and structural mechanisms of inequity and anticipate interventions that can advance equity.

Clinical correlates of negative health events and disparities among adults with epilepsy enrolled in a self-management clinical trial

AimsDespite advances in care, people with epilepsy experience negative health events (NHEs), such as seizures, emergency department (ED) visits and hospitalizations. This analysis using baseline data from an epilepsy self-management clinical trial targeting people from rural regions and other underserved populations assessed the relationship between demographic and clinical variables vs. NHEs. MethodsData to evaluate disparities and clinical correlates was collected using patient surveys from a baseline sample of 94 participants in a larger prospective study of 160 individuals with epilepsy who experienced an NHE within the last six months. Demographic characteristics, mental and physical functional status assessed using 36-Item Short Form Health Survey questionnaire version 2 (SF-36v2), depression assessed with the 9-item Patient Health Questionnaire (PHQ-9), quality of life assessed with the 10-item Quality of Life in Epilepsy Inventory (QOLIE-10), self-efficacy assessed the Epilepsy Self-Efficacy Scale (ESES), social support assessed with the Multidimensional Scale of Perceived Social Support (MSPSS), self-management assessed with the Epilepsy Self-Management Scale (ESMS), and stigma assessed with the Epilepsy Stigma Scale (ESS) were all examined in association with past 6-month total NHE frequency as well as NHE sub-categories of past 30-day and 6-month seizure counts, self-harm attempts, ED visits and hospitalizations. An exploratory evaluation of NHE correlates in relation to the Rural Urban Continuum Code (RUCC) residence classification compared 3 subgroups of increasing rurality. Descriptive statistics were generated for demographic and clinical variables and NHEs, and exploratory analyses compared the distribution of demographic, clinical, and NHE variables by RUCC categorization. ResultsThe mean age was 38.5 years (SD 11.9), predominantly female (N= 62, 66.0%) and white (N=81, 86.2%). Just a little under half (N=43, 45.7%) of participants had annual incomes of less than $25,000, and 40% (N=38) were rural residents (RUCC >3). The past 6-month NHEs count was 20.4 (SD 32.0). Seizures were the most common NHE with a mean 30-day seizure frequency = 5.4 (SD 11.8) and 6-month seizure frequency of 18.7 (SD 31.6). Other NHE types were less common with a past 6-month self-harm frequency of 0.16 (SD 1.55), ED visit frequency of 0.72 (SD 1.10), and hospitalization frequency of 0.28 (SD 1.02). There were few significant demographic and clinical correlates for total and sub-categories of NHEs. Worse physical health status, as measured by the physical component summary (PCS) of the SF-36v2, was significantly associated with 6-month seizure counts (p=.04). There were no significant differences between the 3 RUCC subgroups on demographic variables. However, past 30-day seizure count, past 6-month seizure count and total past 6-month NHE counts were all higher among individuals from more rural settings (p-values <.01 for each). ConclusionsRural adults with epilepsy were more likely to have a greater number of seizures and more epilepsy complications in general. Worse physical health function was also associated with more epilepsy complications. However, this analysis found few other demographic and clinical correlates of cumulative NHEs among adults with epilepsy. Additional efforts are needed to investigate health disparities among people with epilepsy who live in rural regions or who have poor physical health function.

Poor Representation of Rural Counties of the United States in Some Measures of Consumer Broadband.

Telehealth has the potential to mitigate the lack of health care access in rural and underserved communities; however, telehealth is only viable where sufficiently high-speed internet broadband is available to patients. Existing broadband data sets may not accurately reflect the state of broadband, particularly in rural communities. We examined consumer internet speed test data from two organizations to see if the number of tests per 1,000 residents varied across county-level rurality. We analyzed county-level data from Measurement Labs (M-Lab) and Ookla for Good (Ookla fixed and mobile) across the calendar years 2020 and 2021. We used the number of tests conducted per 1,000 residents within United States counties as the outcome variable, and Rural-Urban Continuum Codes (RUCC) as the main independent variable of interest. Using negative binomial models with robust standard errors, we found that the number of fixed speed tests conducted per 1,000 residents was generally lower in rural counties relative to counties with over one million residents. However, we found no associations between any categories of county-level rurality for the number of mobile tests conducted per 1,000 residents. Patterns of association with other covariates emerged as significant in some models and not in others, suggesting key differences among users generating speed tests among these data sources. Our findings demonstrate the poor representation of residents from very rural counties in M-Lab and Ookla fixed data sets of user-generated internet speed tests. Additional data are needed to inform broadband infrastructure investment to identify those communities most left behind by broadband expansion efforts.