Nonmetro Areas Research Articles

COVID-19 in Rural America: Impacts of Politics and Disadvantage.

Early in the COVID‐19 pandemic, people living in cities were much more likely than rural residents to become infected with and die from the virus. Beginning in the summer of 2020, however, the proportional increase in virus spread became much greater in rural areas. By March 2021, per capita cases and deaths were greater in nonmetro than in metro areas. Multi‐variate regression analysis found that political views were the most important variable explaining per capita COVID‐19 cases, while measures of disadvantage were the best predictors of COVID‐19 deaths. Counties with high proportions of Trump voters had higher per capita cases, and in nonmetro areas, these counties had higher death rates. Overall, the best predictor of COVID‐19 death rates is low levels of educational attainment, high poverty rates, and high minority populations.

A descriptive exploration of the geographic and sociodemographic concentration of firearm homicide in the United States, 2004–2018

This study examined the population-based incidence of firearm homicide in the United States to identify geographic concentrations and to determine whether such concentrations have changed over time. It further examined the simultaneous associations of urbanization, poverty, and ethnicity/race with firearm homicide incidence. Using county-level data from the National Vital Statistics System and the U.S. Census Bureau for the years 2004–2018, the findings show geographic patterns not commonly recognized, including several lengthy and continuous corridors with a high incidence of firearm homicide, traversing both metro and non-metro areas. While the data clearly show a strongly disproportionate concentration of firearm homicide incidence in a subset of the population defined by geography, they do not suggest increasing concentration over time. The study findings also generally indicate increasing firearm homicide incidence with increasing levels of surrounding poverty, a phenomenon observed for both metro and non-metro areas.

Dollar stores and food deserts

The purpose of this study is to examine the relationship between dollar store expansion and food access. We begin by documenting the growth of dollar stores over time and comparing dollar store expansion between areas with and without poor food access. Using a time-varying measure of food accessibility, we identify areas with persistent food access challenges, where supermarket and large grocery stores have been absent for the entire study period (2000–2017), and track dollar store expansion into these areas. We use a combination of geocoded food retail establishments from Nielsen's TDLinx dataset to identify the precise location and years of entry for dollar stores and create yearly measures of food accessibility. We find that dollar stores are most prevalent in non-metro areas, regardless of food access status. We find that both the presence and entry of dollar stores are not more likely to be associated with areas categorized as “low-income and low-access”; however, we do find evidence that once a dollar store enters a food desert, that area is more likely to remain without access to a supermarket. Our results indicate that dollar stores are similar to other profit-seeking food retailers in that they are more likely to locate in areas where supermarkets exist and competition is high, suggesting that dollar stores are becoming key players within the food retailing environment. Yet, unlike traditional retail formats, dollar stores could be filling a gap in food access in places where no other food retailer will enter.

Disparity in Access to Oncology Precision Care: A Geospatial Analysis of Driving Distances to Genetic Counselors in the U.S.

In the US, the growing demand for precision medicine, particularly in oncology, continues to put pressure on the availability of genetic counselors to meet that demand. This is especially true in certain geographic locations due to the uneven distribution of genetic counselors throughout the US. To assess these disparities, access to genetic counselors of all specialties is explored by geography, cancer type, and social determinants of health. Geospatial technology was used to combine and analyze genetic counselor locations and cancer incidence at the county level across the US, with a particular focus on tumors associated with BRCA mutations including ovarian, pancreatic, prostate and breast. Access distributions were quantified, and associations with region, cancer type, and socioeconomic variables were investigated using correlational tests. Nationally, in 2020, there were 4,813 genetic counselors, or 1.49 genetic counselors per 100,000 people, varying between 0.17 to 5.7 per 100,000 at the state level. Seventy-one percent of U.S. residents live within a 30-minute drive-time to a genetic counselor. Drive-times, however, are not equally distributed across the country – while 82% of people in metropolitan areas are 30 minutes from a genetic counselor, only 6% of people in nonmetro areas live within 30 minutes’ drive time. There are statistically significant differences in access across geographical regions, socioeconomics and cancer types. Access to genetic counselors for cancer patients differs across groups, including regional, socioeconomic, and cancer type. These findings highlight areas of the country that may benefit from increased genetic counseling provider supply, by increasing the number of genetic counselors in a region or by expanding the use of telegenetics a term used to describe virtual genetic counseling consults that occur via videoconference. Policy intervention to allow genetic counselors to bill for their services may be an effective route for increasing availability of genetic counselors’ services However, genetic counselors in direct patient care settings also face other challenges such as salary, job satisfaction, job recognition, overwork/burnout, and appropriate administrative/clinical support, and addressing these issues should also be considered along with policy support. These results could support targeted policy reform and alternative service models to increase access to identified pockets of unmet need, such as telemedicine. Data and analysis are available to the public through an interactive dashboard 1 .

Comparing Palliative Care Knowledge in Metropolitan and Nonmetropolitan Areas of the United States: Results from a National Survey.

Background: Despite recent growth in access to specialty palliative care (PC) services, awareness of PC by patients and caregivers is limited and misconceptions about PC persist. Identifying gaps in PC knowledge may help inform initiatives that seek to reduce inequities in access to PC in rural areas. Objective: We compared knowledge of PC in metropolitan and nonmetropolitan areas of the United States using a nationally representative sample of U.S. adults. Design: We used data from the 2018 Health Information National Trends Survey (HINTS) 5 Cycle 2 to compare prevalence and predictors of PC knowledge and misconceptions in nonmetropolitan and metropolitan areas as defined by the 2013 Urban-Rural Classification (URC) Scheme for Counties. We estimated the association between nonmetro status and knowledge of PC, adjusted for respondent characteristics, using multivariable logistic regression. Results: More respondents reported that they had never heard of PC in nonmetro (78.8%) than metro (70.1%) areas (p < 0.05). Controlling for other factors, nonmetro residence was associated with a 41% lower odds of PC knowledge (odds ratio [OR] = 0.59; 95% confidence interval [CI] = 0.37-0.94), and Hispanic respondents also demonstrated significantly lower odds of PC knowledge conditional on rural status (OR = 0.47; CI = 0.27-0.83). Misconceptions about PC were high in both metro and nonmetro areas. Conclusion: Awareness of PC was lower in rural and micropolitan areas compared with metropolitan areas, suggesting the need for tailored educational strategies. The reduced awareness of PC among Hispanic respondents regardless of rural status raises concerns about equitable access to PC services for this population.

Association of rurality and race with surgical treatment and outcomes for nonmetastatic colon cancer.

e18536 Background: Rural cancer patients face limited access to care due to greater travel distance and lack of specialty cancer care. Little is known about the intersection of rurality with well-documented racial disparities in colon cancer treatment and outcomes. Methods: We used fee-for-service Medicare claims to study patients age 65+ diagnosed with incident colon cancer without evidence of metastases who underwent cancer-directed surgery between 04/01/2016 and 09/30/2018. The primary exposure was rurality of patient’s residence categorized as metropolitan (metro), micropolitan, and small town/rural. Outcomes were non-elective surgery (emergency department visit or transfer within 2 days of surgery), receipt of minimally invasive surgery (laparoscopic or robotic), 90-day surgical complications, and 90-day mortality. Logistic regression adjusted for patient demographics, cancer side (right vs left), comorbidities, and Area Deprivation Index. We assessed effect modification by race/ethnicity. Results: Of 57,710 patients with incident non-metastatic colon cancer, 37,691 (65%) underwent surgery. In this surgical cohort, small town/rural and micropolitan residents were more likely to be older, white, and Medicare-Medicaid dual-eligible than metro residents. After risk adjustment, patients in small town/rural areas had higher odds of non-elective surgery (OR=1.24, 95% CI:1.13-1.36) and lower odds of minimally invasive surgery (OR=0.75, 95% CI:0.71-0.80) than patients living in metro areas. Similar results were seen for micropolitan areas. The association between rurality and 90-day outcomes differed by race/ethnicity (p-interaction=0.001 for surgical complications and mortality, see Table). Hispanics and other races had higher odds of 90-day surgical complications in non-metro versus metro areas but there was no notable difference for white patients. Likewise, compared to metro areas, racial/ethnic minorities had higher odds of 90-day mortality in small town/rural areas but white patients had lower odds. Conclusions: Small town/rural-residing Medicare beneficiaries undergoing surgery for non-metastatic colon cancer were less likely to receive optimal surgical management and worse outcomes, especially among non-white patients. The compounded effect of sociodemographic factors should be further studied to develop targeted policies and improve care for rural cancer patients.[Table: see text]

A cross-sectional analysis of trust of information and COVID-19 preventative practices among people with disabilities

BackgroundTrust of information shapes adherence to recommended practices and speed of public compliance during public health crises. This is particularly important for groups with higher rates of high-risk health conditions, including those aged 65 and over and people with disabilities. ObjectiveWe examined trust in information sources and associated adherence to COVID-19 public health recommendations among people with disabilities living in metropolitan, micropolitan, and noncore counties. MethodsWe recruited participants using Amazon’s Mechanical Turk (MTurk) and screened for disability status (n = 408). We compared sociodemographic groups with t-tests, Pearson’s correlations, and Chi-square, as appropriate. We used linear regression to examine factors associated with trust in information and compliance with CDC recommended COVID-19 practices. ResultsNonmetro respondents had the lowest trust ratings among all demographic groups, and reported significantly less trust in most information sources. Respondents aged 65 and over reported the highest compliance with CDC recommended practices, while those from nonmetro areas reported the lowest. A regression model for adherence to CDC recommended practices was significant (F = 11.87, P ≤ .001), and explained 33% of the variance. Specifically, increased adherence was associated with being over 65, female, and higher general trust scores. Decreased practices were associated with being nonwhite, nonmetro, higher trust scores in President Trump, and having a communication disability. ConclusionsTrust in information sources is associated with action. It is important to provide clear, consistent, and non-polarizing messages during public health emergencies to promote widespread community action.

Transportation, community quality of life, and life satisfaction in metro and non-metro areas of the United States

This study examines the impacts of a community's transportation system on the quality of life provided to its residents. Other key livability factors are also identified, and their impacts on community quality of life are estimated. These livability indicators are categorized into four main dimensions: social, physical/climate, functional, and safety. The study also examines the impacts of community quality of life on overall life satisfaction for an individual. Data were obtained from a nationwide livability survey. Survey data were analyzed using ordered probit models. The first model estimated ease of travel as a function of community transportation characteristics and individual characteristics. The second model estimated community quality of life as a function of the quality of the livability indicators. The final model estimated life satisfaction as a function of community quality of life and other factors. Separate models were estimated for metro and non-metro areas. Transit quality, the conditions of roads, congestion, and traffic safety were all found to have significant impacts on ease of travel. Ease of travel and walkability, among other factors, were found to impact community quality of life. Results also showed the positive impact that community quality of life has on overall life satisfaction.

The Design of Disease Maps Shapes Perceptions of Threat and Public Policy Preferences

Choropleth disease maps have become the main tool for communicating information about the geography of health threats to the public. These maps have the potential to shape perceptions of threat, preferences about policy, and perhaps even behavior, but they are unfortunately often poorly designed and misleading. In a large survey of residents of the U.S. state of Georgia conducted in June 2020, we randomly assigned respondents to view one of two maps produced in the spring of 2020 by the Georgia Department of Public Health. The first is a map of county-level COVID case counts, which generated the false perception that the COVID threat was concentrated almost exclusively in the Atlanta metro area. The second is a map of the case rate per 100,000 people, which clarified that the virus was widespread in much of Georgia. Those who saw the second map were less likely to consider the virus as an urban problem, and more likely to perceive it as a concern for Georgia and its economy. Moreover, respondents from non-metro areas who saw the case rate map were more concerned that they, their friends, or community members might contract the virus. Respondents who saw the case rate map also expressed greater support for policies aimed at mitigating the virus – an effect driven by self-identified Republicans, who were far more skeptical about public health measures to mitigate the spread of the virus in general.

1333. A Cross-Sectional Analysis of Inappropriate Outpatient Antibiotic Use in Children Insured by Kentucky Medicaid

BackgroundKentucky (KY) consistently has one of the highest rates of outpatient pediatric antibiotic prescribing in the nation. Previous analyses identified significant variation in volume of antibiotic prescribing by geographic location, patient demographics, and provider type, but less is known about the appropriateness of this prescribing. We describe appropriateness of outpatient antibiotic prescribing in children insured by KY Medicaid.MethodsWe utilized KY Medicaid pharmacy and medical claims from 2017 for children < 20 years. Patient demographic variables were abstracted from Medicaid enrollment data. Antibiotic prescriptions were identified by NDC and matched to medical claims within 3 days prior to fill date to identify corresponding diagnoses via ICD-10 codes. A previously published appropriateness classification scheme (Chua, BMJ 2019) was applied to categorize antibiotic prescriptions as “appropriate”, “potentially appropriate”, “inappropriate” or “not associated with indication”.ResultsOf the 779,751 antibiotic prescriptions included, 19.5% were appropriate, 45.3% were potentially appropriate, 20.8% were inappropriate, and 14.4% were not associated with an indication (Table 1). Inappropriate prescriptions were more common among children 0-2 years (24.4%) and those living in non-metro areas (22.2%). Antibiotics prescribed by general practitioners were also more likely to be inappropriate (22.2%). The most common diagnoses for each category are summarized in Table 2. Amoxicillin was the most commonly prescribed antibiotic in all categories. Azithromycin was more frequently prescribed for inappropriate indications or those not associated with a diagnosis code. Cefdinir was more common for appropriate and potentially appropriate indications.Table 1: Antibiotic Prescription Characteristics, 2017 Table 2: Top Diagnoses for Antibiotic Prescriptions by Category ConclusionInappropriate antibiotic prescribing is more common among young children living in non-metro areas seen by general practitioners. Outpatient antibiotic stewardship interventions should target these patient demographics and provider types. This classification scheme to describe inappropriate prescribing is feasible for use in pediatric Medicaid patients and could serve as a valuable metric for provider feedback reports on antibiotic use.Disclosures Bethany A. Wattles, PharmD, Kentucky Medicaid (Grant/Research Support)Merck (Grant/Research Support) Michael J. Smith, MD, MSC.E, Kentucky Medicaid (Grant/Research Support)Merck (Grant/Research Support)

Exercise routine change is associated with prenatal depression scores during the COVID-19 pandemic among pregnant women across the United States.

The COVID-19 pandemic has negatively affected physical and mental health worldwide. Pregnant women already exhibit an elevated risk for depression compared to the general public, a pattern expected to be exacerbated by the pandemic. Certain lifestyle factors, including moderate exercise, may help support mental health during pregnancy, but it is unclear how the pandemic may impact these associations across different locations. Here, we test whether: (i) reported exercise routine alterations during the pandemic are associated with depression scores; and, (ii) the likelihood of reporting pandemic-related exercise changes varies between women living in metro areas and those in non-metro areas. This cross-sectional study used data from the COVID-19 And Reproductive Effects (CARE) study, an online survey of pregnant women in the United States. Participants were recruited April-June 2020 (n = 1,856). Linear regression analyses assessed whether reported COVID-19-related exercise change was associated with depression score as measured by the Edinburgh Postnatal Depression Survey. Logistic regression analyses tested whether a participant's Rural-Urban Continuum Code classification of "metro" was linked with higher odds of reporting exercise changes compared to a "non-metro" classification. Women who reported exercise changes during the pandemic exhibited significantly higher depression scores compared to those reporting no changes. Moreover, individuals living in metro areas of all sizes were significantly more likely to report exercise changes compared to women living in non-metro areas. These results suggest that the ability to maintain an exercise routine during the pandemic may help support maternal mental health. It may therefore be prudent for providers to explicitly ask patients how the pandemic has impacted their exercise routines and consider altered exercise routines a potential risk factor for depression. An effort should also be made to recommend exercises that are tailored to individual space restrictions and physical health.

Abstract PO-168: Association between geography and cause-specific mortality in gastrointestinal cancers in Georgia

Abstract Background: Cancers of the gastrointestinal system represent a diverse range of diseases with different etiologies, from screening-detectable colorectal cancer (CRC) to pancreatic cancer, often diagnosed at late stage and highly lethal. While racial disparities have been observed for some of these cancers, the impact of geography is less understood. Georgia is an ideal place to examine disparities in geography because it has a diverse setting and population, with approximately 25% of residents living in a rural area and nearly one-third identifying as non-Hispanic Black (NHB), according to the 2010 decennial census. Methods: We obtained clinical (age at diagnosis, tumor stage, tumor size), sociodemographic (e.g., insurance status, marital status, SES index), and treatment information (e.g., receipt of surgery, lymph node biopsy/removal) on nearly 30,000 NHB and non-Hispanic White (NHW) men and women aged 18 years and older, diagnosed with CRC, hepatocellular carcinoma (HCC), pancreatic cancer or gastric cancer between 2009 and 2014 in Georgia from the population-based Surveillance, Epidemiology, and End Results (SEER) Program. Patients were classified as residing in a metro or non-metro county at diagnosis according to 2013 rural-urban continuum codes. Multivariable Cox proportional hazards models were used to calculate the hazard ratios (HRs) and corresponding 95% confidence intervals (95% CI) for the association between residing in a non- metro versus metro county and cause-specific mortality, stratified by patient characteristics. Results: Over 75% of the study population resided in a metro county at diagnosis, with the highest proportion among HCC cases (81%) and the lowest proportion among CRC cases (76%). Overall, NHBs were more likely to reside in metro counties (82%) than NHWs (74%). The average length of follow-up varied by cancer site and metro status, ranging from 35 months (CRC in metro areas) to 4 months (pancreatic cancer in non-metro areas). For all cancer sites, patients in metro counties were more likely to be younger, single, have localized disease (except gastric cancer), and receive surgery of the primary site. By comparison, patients in non-metro counties were more likely to be diagnosed at older ages, have Medicaid, and be widowed. Residing in a non-metro county was associated with higher cause- specific mortality among the youngest patients (age 18-49 years) for pancreatic cancer (HR=1.4, 95% CI: 1.1, 1.9) and gastric cancer (HR=1.8, 95% CI: 1.2, 2.6). We also observed a higher hazard of mortality for residing in a non-metro compared to metro county among NHB HCC patients (HR=1.3, 95% CI: 1.0, 1.7) and never married gastric cancer patients (HR=1.5, 95% CI: 1.1, 2.1). Associations among CRC patients were less robust. Conclusion: Our results suggest residing in a non-metro county is associated with higher mortality among some groups diagnosed with rarer gastrointestinal cancers. Geographic differences in access to diagnosis and treatment may contribute to this disparity. Citation Format: Rebecca Nash, Jasmine M. Miller-Kleinhenz, Maria C. Russell, Lindsay J. Collin, Katherine Ross-Driscoll, Jeffrey M. Switchenko, Lauren E. McCullough. Association between geography and cause-specific mortality in gastrointestinal cancers in Georgia [abstract]. In: Proceedings of the AACR Virtual Conference: Thirteenth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2020 Oct 2-4. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(12 Suppl):Abstract nr PO-168.

Dynamics of chronic diseases in metro and non-metro regions of India: evidence from India Human Development Survey I and II

&lt;p class="abstract"&gt;&lt;strong&gt;Background:&lt;/strong&gt; The growth of metropolitan cities had significantly contributed to the process of urbanization in India. About two-fifth of the urban population, out of total India’s urban population, live in 35 metropolitan cities. It is important to look into the disease dynamics in the population of metro and non-metro regions of India. The study aims to find the differences in the distribution of chronic diseases in metro and non-metro regions of India and depicts the contributions of background factors causing a change in the prevalence of chronic diseases in metro and non-metro regions of India.&lt;/p&gt;&lt;p class="abstract"&gt;&lt;strong&gt;Methods:&lt;/strong&gt; Data from India Human Development Survey (IHDS) I and II conducted in 2004 and 2012 respectively have been used. Bivariate analysis has been performed to find the association between independent variables and chronic diseases, and logistic regression has been used to find the effect of predictor variables on chronic diseases by metro and non-metro regions. Fairlie decomposition technique has been used to find the contribution of each predictor variable accounting for differences in chronic diseases between metro and non-metro regions. &lt;/p&gt;&lt;p class="abstract"&gt;&lt;strong&gt;Results:&lt;/strong&gt; Age, sex, socio-economic status (education and wealth), alcohol consumption, tobacco consumption, and body mass index status are significantly associated with chronic conditions in metro regions of India. Age, wealth, and developed regions contributed most to the differences in chronic diseases between metro and non-metro areas.&lt;/p&gt;&lt;p class="abstract"&gt;&lt;strong&gt;Conclusions:&lt;/strong&gt; Metro regions in India suffers from a massive burden of chronic conditions. Metro regions should be given a special focus to tackle the menace of chronic diseases.&lt;/p&gt;&lt;p class="abstract"&gt; &lt;/p&gt;

O7 Fresh Fruit and Vegetable Availability and Cost in SNAP-participating Retail Stores in Mississippi: Differences by Rurality

Background Access to healthy, affordable food is an environmental factor that can influence food security. Research suggests healthy food availability varies by store type, rurality/urbanicity, Supplemental Nutrition Assistance Program (SNAP) participation, and sociodemographic characteristics. Objective The purpose of this study was to investigate differences in variety and cost of fresh fruits and vegetables in convenience stores, grocery stores, and supermarkets in Mississippi across Rural Urban Continuum Codes (RUCCs). Study Design, Setting, Participants A survey design was used to collect data in 461 convenience stores, grocery stores, and supermarkets. Trained surveyors (Extension county agents, Extension nutrition educators, and research team members) selected an assigned number of SNAP-participating stores by county to represent food availability across Mississippi. Measurable Outcome/Analysis Availability and cost data were collected using an adapted version of the Nutrition Environment Measures Survey-Corner Store (NEMS-CS). Metro areas were defined as those with RUCCs of 1 to 3, while non-metro areas were defined as those with RUCCs of 4 to 9. Independent samples t-tests and Mann-Whitney U tests were used to compare means in metro vs non-metro areas within store types. Results Variety of fresh fruits and fresh vegetables was greater in non-metro vs metro areas in convenience and grocery stores but not supermarkets. Cost of 6 most commonly available fruits and vegetables was similar within store types when comparing metro vs non-metro areas. Conclusion Elucidation of food availability patterns in retailers serving SNAP clients is important for planning outreach efforts. Smaller retailers may be important partners and could be a setting for SNAP-Education social marketing messages encouraging healthy food selection behaviors. Involving agents and educators in data collection and in building shared goals with local food retailers may be a promising strategy for improving purchase and consumption of healthy foods in Mississippi. Funding Mississippi State University Extension.

Metro and Non-Metro Variation in Postsecondary Enrollment: The Role of Race, Ethnicity, and Residential Location in Texas

This study used a spatial approach to explore metropolitan and nonmetropolitan variation of college enrollment at the interplay of place and race within the state of Texas. Analyzing Integrated Public Use Microdata Series (IPUMS-USA) data and using population proportions as well as regression methods, we pay particular attention to the racial/ethnic diversity in non-metro (i.e., rural) areas and its relationship with college enrollment. We find geography is a mediating factor in the college enrollment racial/ethnic gaps for rural Texans, but there are considerable regional disparities. These disparities show diverse trends among rural regions where large proportions of Latinxs reside. This indicates a need for more research to explore geographic post-secondary opportunity and college access practices and policy with the racial/ethnic diversity of place in mind, especially regarding rural communities of Color. The data and code for the analysis can be found here on a GitHub repository.

Abstract B099: Rural black cases have worsened outcomes in head and neck cancer

Abstract Background: Racial and socioeconomic disparities in head and neck cancer survival are well documented. However, rural-urban context, especially the interaction between race and rural-urban context, is understudied. The aim of this study is to examine the relationship between race and rural-urban context in HNC survival and determine factors that potentially drive this disparity. Methods: We constructed a retrospective cohort of 158,718 head and neck squamous cell cases from the National Cancer Database with the following sites: hypopharynx (9.54%), larynx (54.42%), and oral cavity (36.04%). Since we are interested in traditional head and neck cancer, oropharyngeal cases were excluded from the cohort. We used Rural-Urban Continuum Codes (RUCC) to classify cases into metro (RUCC 1-3) and non-metro subgroups (RUCC 4- 9). Risk factors were classified into demographic (age and gender), socioeconomic (insurance status), and clinical factors (site, stage, and treatment modality). Kaplan-Meier survival curves and the Cox proportional hazards regression were used to calculate adjusted hazard ratios (HR). The proportional hazards and linearity assumptions were tested and satisfied. Results: The exposure cohorts that were analyzed include black metro (N = 17,256; 10.87%), black non-metro (N = 2,501; 1.57%), white metro (N = 109,842; 69.21%), and white-non-metro cases (N = 29,119; 18.35%). Rural-urban five-year survival differences are larger between black non-metro cases (39.32%; 95% Confidence Interval (CI): 37.21-41.56%) and black metro cases (43.97%; 95% CI: 43.15-44.81%) compared with white non-metro cases (50.32%; 95% CI: 49.68-50.99%) and white metro cases (53.12%; 95% CI: 52.79-53.45%). In the univariable Cox proportional hazards analysis, compared with white metro cases, black non-metro cases have the worst survival (HR: 1.44; 95% CI: 1.36-1.15; p&amp;lt;.001) followed by black metro cases (HR: 1.28; 95% CI: 1.26-1.31; p&amp;lt;.001) and white non-metro cases (HR: 1.0,76 95% CI: 1.06-1.10; p&amp;lt;.001). Interestingly, there is no interaction between race and rural-urban context (p=0.18). After conducting a multivariable analysis controlling for demographic, socioeconomic, and clinical factors, the HRs were attenuated, but black non-metro cases continue to have worse survival compared with white metro cases. Conclusion: Although there is no interaction between rural-urban context and race, black HNC cases, specifically those living in non-metro areas, have worse survival. This disparity still remains after adjusting for insurance status, site, and stage. Further research should be conducted to determine other modifiable risk factors to mitigate this racial and geospatial disparity. Citation Format: Jacob A Clarke, Alyssa M Despotis, Jose P Zevallos, Angela L Mazul. Rural black cases have worsened outcomes in head and neck cancer [abstract]. In: Proceedings of the Twelfth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2019 Sep 20-23; San Francisco, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(6 Suppl_2):Abstract nr B099.

Why do businesses grow faster in urban areas than in rural areas?

We document that the growth rate of business earnings among young firms is significantly higher in metro areas than in non-metro areas. Agglomeration economies and firm selection (less productive firms are more likely to exit in metro areas) are known to explain a part of the productivity growth in urban areas, but less is known about the role of borrowing constraints. By developing a firm-dynamics model with a location choice, we show borrowing constraints interact with growth and location choices of firms, and contribute to a substantial part of the observed growth-rate difference between urban and rural young firms. Our model suggests the distortion in location choice due to borrowing constraints can induce non-trivial welfare loss.

Irrational Economics and Regional Cultural Life

Irrational Economics and Regional Cultural Life

Race, Residence, and Underemployment: Fifty Years in Comparative Perspective, 1968–2017

High underemployment has been a chronic structural feature of the rural United States for decades. In this paper, we assess whether and how inequalities in underemployment between metropolitan (metro) and nonmetropolitan (nonmetro) areas have changed over the course of the last five decades. Drawing on data from the March Current Population Survey (CPS) from 1968 to 2017, we analyze inequality in the prevalence of underemployment between metro and nonmetro areas of the United States, paying special attention to differences between white, black, and Hispanic workers. Our results show that the underlying risk of underemployment has increased in both metro and nonmetro areas over the last fifty years. Nonmetro workers have consistently faced greater employment hardship compared to their metro counterparts, and these differences cannot be fully explained by differences in population characteristics. Nonmetro ethnoracial minorities have experienced particularly poor labor market outcomes. The disadvantage of ethnoracial minority status and rural residence is especially pronounced for nonmetro black workers, among whom underemployment has remained persistently high with only modest convergence with other workers. Hispanic workers also face elevated risk of underemployment, but we observe a unique convergence between metro and nonmetro workers within this population.

Urban/rural disparities in stage at presentation of colorectal cancer among young adults in the United States, 2007-2015.

3598 Background: The overall incidence rate of colorectal cancer (CRC) has declined in recent decades but is rising in young adults (YA). Disparities also exist in CRC presentation and geographical residence. We examined associations between urban/rural residence and CRC stage at presentation among YA 18-50 years old, using the Surveillance, Epidemiology, and End Results (SEER) 1973-2015 registry. Methods: Retrospective cohort study using SEER patients (pts) diagnosed with CRC between 2007-2015, aged 18-50. Urban/rural status was defined at the county level as large metro, small metro, urban non-metro, and rural. Pts were grouped by age: 18-30, 31-40, and 41-50 years old. Stage was defined as in-situ/localized and regional/distant. We used multivariable logistic regression to describe associations between urban/rural status with stage at presentation, adjusting for tumor location, histology, grade, and patient attributes (e.g. insurance status). Results: 27,198 CRC pts were analyzed: 62.2% large metro, 27.4% small metro, 9.4% urban non-metro, and 67.1% regional/distant stage. In multivariable analysis, YA in urban non-metro counties had lower odds of regional/distant stage at presentation compared to YA in large metro counties (OR = 0.87, 95% CI = 0.79-0.97). Associations between small metro and rural status with stage at presentation were not significant (OR = 0.94, 95% CI = 0.89-1.01 and OR = 0.84, 95% CI = 0.64-1.10, respectively). YA with Medicaid or no insurance had higher odds of regional/distant CRC (OR = 1.28, 95% CI = 1.17-1.39 and OR = 1.34, 95% CI = 1.20-1.51, respectively) compared to YA privately insured. Other factors in YA associated with higher odds of regional/distant CRC included signet-ring histology, poorly-, moderately-, and undifferentiated grade, and younger age (18-30 and 31-40 years old). Conclusions: YA in urban non-metro areas had lower odds of regional/distant CRC at presentation compared to YA in large metro areas. YA with Medicaid, no insurance, signet-ring histology, poorly-, moderately-, and undifferentiated grade, and younger age had higher odds of regional/distant CRC at presentation. Further research is needed to explore the etiology of these differences.