County-level Characteristics Research Articles

Emergency department visits and dental procedures: Mission of Mercy, 2013-2016.

Objective: This study compared attendees at Wisconsin Mission of Mercy (MoM) events with and without prior emergency department (ED) visits for dental care in terms of demographic characteristics and dental procedures received.Methods: De-identified archival data available from the America’s Dentists Care Foundation (2013–2016) were analyzed. Summary statistics were calculated based on prior ED visit status for county-level characteristics and type of dental procedures received. Generalized estimating equation models with logistic links were fitted to examine associations between the predictor and independent variables.Results: Most attendees were White, adult females (mean age 37 years). Current dental pain was reported by 61% vs 33% of attendees with and without prior ED visit. Cleaning (57%) was the most common procedure received by attendees with no prior ED visits, while extractions (47%) were the most common for those with prior ED visits. Among MoM attendees, males had higher odds of prior ED visits compared to females. Compared to White, Black had higher odds and Hispanic had lower odds of having prior ED visits. In the multivariable analysis, dental pain (OR: 3.32; 95% CI: 2.75, 4.02) had the strongest association with prior ED visits. Attendees with prior ED visit history had higher odds of receiving extractions and restorative care, compared to cleaning after adjustment for person and county-level characteristics.Conclusion: Among MoM attendees, prior ED visits were associated with not being Hispanic, being male, and having dental pain. Compared to attendees with no prior ED visit, those with prior ED visits had higher odds of receiving restorative care and extractions.

Impact of smoke-free ordinance strength on smoking prevalence and lung cancer incidence.

6578 Background: Smoke-free ordinances (SFO) have been shown to decrease the prevalence of cardiovascular and pulmonary disease, but there is limited data on the impact of such policies on lung cancer incidence. We investigated the relationship between strength of county-level SFO with smoking prevalence and lung cancer incidence in Indiana. Methods: Following IRB approval, we queried the Indiana State Cancer Registry and Indiana Tobacco Prevention and Cessation Commission’s policy database between 1995 and 2016. County-level characteristics included population, income, poverty, education, race/ethnicity, sex, and rurality. Lung cancer diagnosis and stage were also collected. Using generalized estimating equations (GEE) with robust standard errors and accounting for the clustering effect at county level, we performed multivariable analyses of smoking prevalence and age-adjusted lung cancer rates with respect to the strength of smoke-free ordinances at the county level over time. Results: Indiana consists of 92 counties, 24 of which had SFO by 2011. In 2012, Indiana enacted a law mandating at least a moderate state-wide SFO. From 1995 to 2016, 110,935 Indiana residents were diagnosed with lung cancer. Indiana also had an average age-adjusted yearly lung cancer incidence of 76.8 per 100,000 population and average yearly smoking prevalence of 25% during this time. Smoking prevalence was 1.2% (95% CI [-1.88, -0.52]) lower in counties with comprehensive or moderate SFO compared with those with weak or no SFO. Counties that had comprehensive or moderate SFO had an 8.36 (95% CI [-11.45, -5.27]) decrease in new lung cancer diagnosis per 100,000 population per year compared with counties that had weak or no SFO. Conclusions: Stronger municipal smoke-free air ordinances are associated with decreased smoking prevalence and fewer new lung cancer cases. Strengthening smoke-free ordinances is paramount to decreasing lung cancer incidence.

Understanding Where We Are Well: Neighborhood-Level Social and Environmental Correlates of Well-Being in the Stanford Well for Life Study.

Individual well-being is a complex concept that varies among and between individuals and is impacted by individual, interpersonal, community, organizational, policy and environmental factors. This research explored associations between select environmental characteristics measured at the ZIP code level and individual well-being. Participants (n = 3288, mean age = 41.4 years, 71.0% female, 57.9% white) were drawn from a registry of individuals who completed the Stanford WELL for Life Scale (SWLS), a 76-question online survey that asks about 10 domains of well-being: social connectedness, lifestyle and daily practices, physical health, stress and resilience, emotional and mental health, purpose and meaning, sense of self, financial security and satisfaction, spirituality and religiosity, and exploration and creativity. Based on a nationally-representative 2018 study of associations between an independent well-being measure and county-level characteristics, we selected twelve identical or analogous neighborhood (ZIP-code level) indicators to test against the SWLS measure and its ten constituent domains. Data were collected from secondary sources to describe socio-economic (median household income, percent unemployment, percent child poverty), demographic (race/ethnicity), and physical environment (commute by bicycle and public transit), and healthcare (number of healthcare facilities, percent mammogram screenings, percent preventable hospital stays). All continuous neighborhood factors were re-classified into quantile groups. Linear mixed models were fit to assess relationships between each neighborhood measure and each of the ten domains of well-being, as well as the overall SWLS well-being measure, and were adjusted for spatial autocorrelation and individual-level covariates. In models exploring associations between the overall SWLS score and neighborhood characteristics, six of the twelve neighborhood factors exhibited significant differences between quantile groups (p < 0.05). All of the ten SWLS domains had at least one instance of significant (p < 0.05) variation across quantile groups for a neighborhood factor; stress and resilience, emotional and mental health, and financial security had the greatest number of significant associations (6/12 factors), followed by physical health (5/12 factors) and social connectedness (4/12 factors). All but one of the neighborhood factors (number of Federally Qualified Health Centers) showed at least one significant association with a well-being domain. Among the neighborhood factors with the most associations with well-being domains were rate of preventable hospital stays (7/10 domains), percent holding bachelor’s degrees (6/10 domains), and median income and percent with less than high school completion (5/10 domains). These observational insights suggest that neighborhood factors are associated with individuals’ overall self-rated well-being, though variation exists among its constituent domains. Further research that employs such multi-dimensional measures of well-being is needed to determine targets for intervention at the neighborhood level that may improve well-being at both the individual and, ultimately, neighborhood levels.

Clarifying the Theoretical Tenets of the Symbolic Threat Perspective

Although Tittle and Curran in 1988 found support for the symbolic threat hypothesis, no further empirical investigations of the perspective have occurred to explain the contingencies of juvenile justice decision-making. We argue that past research has neglected this multilevel theory by only focusing on the “symbolic threat” thesis component, mischaracterizing the perspective as an inequality theory, or classifying the model as a derivative of Blalock’s minority group threat perspective. We contend that the original formulation of the symbolic threat hypothesis is a distinctive theory and offers a viable alternative to inequality and other racial/threat theoretical perspectives. The current study empirically tests the perspective in its entirety by examining the interrelationships between county-level characteristics (i.e. youth and minority populations), a juvenile’s race/ethnicity, and engaging in “moral offenses” with juvenile justice decision-making. The findings provide insight into the influence of contextual factors on the social control of youth of different racial/ethnic backgrounds.

The association of county-level socioeconomic factors with individual tobacco and alcohol use: a longitudinal study of U.S. adults

BackgroundPlace-based factors have been implicated as root causes of socioeconomic disparities in risky health behaviors such as tobacco and alcohol use. Yet few studies examine the effects of county-level socioeconomic characteristics, despite the fact that social and public health policies are often implemented at the county level. In this study, we tested the hypothesis that county-level socioeconomic disadvantage was associated with individual tobacco and alcohol use.MethodsThe sample included a panel of participants from the National Longitudinal Survey of Youth (N = 9302). The primary predictors were three time-varying measures of socioeconomic disadvantage in an individual’s county of residence: educational attainment, percent unemployment, and per capita income. We first conducted traditional ordinary least squares (OLS) models, both unadjusted and adjusted for individual-level covariates. We then conducted fixed effects (FE) models to adjust for confounding by unmeasured time-invariant individual-level factors.ResultsOLS and FE models yielded contrasting results: higher county-level per capita income was associated with decreased drinking in OLS models and increased drinking in FE models, while decreased county-level educational attainment was associated with decreased smoking in OLS models and more cigarettes per day in FE models. The findings from FE models suggest that OLS models were confounded by unobserved time-invariant characteristics. Notably, the point estimates for the county-level measures were small, and in many cases they may not represent a clinically meaningful effect except at the population level.ConclusionsThese results suggest that county-level socioeconomic characteristics may modestly influence tobacco and alcohol use. Future work should examine the effects of specific county policies that might explain these findings.

Environmental justice in the oil refinery industry: A panel analysis across United States counties

Environmental injustice due to economic inequality and its correlates and consequences has been a focus for researchers and policy makers alike over many years, with much of that research focusing on race as the major determinant of inequality. In this paper we extend this literature by focussing on emissions from the oil refinery industry at the community level. In particular, based on the use of panel data, we analyse econometrically how the environmental performance of individual petroleum refineries (emissions of benzene and toluene to the atmosphere) associates with determinants of economic inequality at the micro-level in terms of average per capita income at the level of United States counties as well as average county-level unemployment rates. The paper finds evidence of environmental injustice as a result of unemployment levels in areas around refineries and, to a slightly lesser extent, as a result of income inequality. It discusses these results in a wider context, referring amongst other things to the role of county-level community characteristics and the potential for private firms to substitute for the intervention of public institutions, if these are lacking.

Community Characteristics and Regional Variations in Diabetes Prevalence in Texas Counties.

Diabetes has become a critical population health issue in the United States. Although researchers have focused on diabetes prevalence at the individual level, few studies address community contexts of diabetes prevalence at the county level. The objective of this study is to explore the association between community characteristics and diabetes prevalence in 254 counties in Texas. Using 3 countywide data sets in 2012, the authors measured county-level community characteristics, including social, environmental, and health behavior factors, and examined how these county-level factors are associated with diabetes prevalence. Multivariate ordinary least squares models weighted with population of county were used for estimation. Rurality, physical inactivity prevalence, and obesity prevalence were positively associated with diabetes prevalence, while the percentage of nonphysician health care professionals was negatively associated with diabetes prevalence. The findings indicate that environmental and health behaviors are significantly associated with a high prevalence of diabetes, but more nonphysician health care professionals, including nurse practitioners, may mitigate diabetes prevalence. This study highlights the significance of community factors in diabetes prevalence and provides insights for diabetes prevention programs with nonphysician health care professionals.

Heterogeneity of colon and rectum cancer incidence across 612 SEER counties, 2000-2014.

Recent analyses have suggested decreases over time in colorectal cancer incidence at older ages (≥55 years) but increases at younger ages (20-54 years). Understanding the geographic heterogeneity of incidence facilitates resource allocation for potential interventions and advances our knowledge of differential etiologies for these cancers. We performed age-period-cohort analysis using 2000-2014 county-level incidence from the Surveillance, Epidemiology, and End Results (SEER) database, estimating relative risk (RR) and age-adjusted annual percent change (Net Drifts) simultaneously for 612 counties via a hierarchical model, separately for colon and rectum cancer, stratified by age group (20-54 vs. 55-84). We also studied correlates of RR and Net Drift with various county-level characteristics. In all SEER counties, colon and rectum cancer incidence rates increased at ages 20-54, whereas rates decreased at ages 55-84. There was marked heterogeneity in both RR and Net Drift among states and counties for both cancer types. Maps of county RR and Net Drift revealed localized clusters in several states. For both cancer types, counties with high RR and unfavorable Net Drift tended to have higher prevalence of obesity and diabetes and to be of a lower socioeconomic status. Counties with higher overall screening rates tended to have lower Net Drifts for both cancer types. Increasing colorectal cancer incidence in the younger age group is geographically widespread, although there is significant heterogeneity in temporal trends and risk both within and between states. These geographic patterns correlate with different county-level characteristics depending on cancer type and age group.

Delivery of Preventive Oral Health Services by Rurality: A Cross-Sectional Analysis.

Integrating oral health care into primary care has been promoted as a strategy to increase delivery of preventive oral health services (POHS) to young children, particularly in rural areas where few dentists practice. Using a multistate sample of Medicaid claims, we examined a child's odds of receiving POHS in a medical office by county rurality. We used 2012-2014 Medicaid Analytic extract claims data for 6,275,456 children younger than 6 years in 39 states that allowed Medicaid payment for POHS in medical offices. We used county-level characteristics from the Area Health Resources Files, including a 3-level measure of county rurality. We used logistic regression to estimate a child's odds of receiving POHS in a medical office by county rurality, while controlling for other patient and county characteristics. POHS in medical offices were received by 7.8% of children. Rates of POHS in medical offices were higher in metropolitan (metro) counties (8.4%) than nonmetro adjacent to metro (5.8%) and nonmetro not adjacent to metro (4.3%). In adjusted analysis, children living in nonmetro not adjacent to metro (OR = 0.79, 95% CI: 0.64-0.99) and adjacent to metro counties (OR = 0.70, 95% CI: 0.59-0.82) were significantly less likely to receive POHS in medical offices than children living in metro counties. In this study of POHS in medical offices among young Medicaid-enrolled children, we found POHS rates were lowest in nonmetro counties. Given barriers to dental care in rural areas, states should take additional steps beyond allowing Medicaid reimbursement to increase delivery of POHS in medical offices.

County and Physician Variation in Benzodiazepine Prescribing to Medicare Beneficiaries by Primary Care Physicians in the USA

Physicians widely prescribe benzodiazepines (BZD) despite well-recognized harms. To determine county and provider characteristics that predict high-intensity BZD prescribing by primary care physicians (PCPs) to Medicare beneficiaries. Cross-sectional analysis of the 2015 Medicare Part D Public Use Files (PUF). n = 122,054 PCPs who prescribed 37.3 billion medication days. Primary outcome was intensity of BZD prescribing (days prescribed/total medication days) at the county- and physician levels. PCP and county characteristics were derived from the Part D PUF, Area Health Resources Files, and County Health Rankings. Logistic regression determined the characteristics associated with high-intensity (top quartile) BZD prescribing. Beneficiaries were prescribed over 1.2 billion days of BZD in 2015, accounting for 2.3% of all medication days prescribed in Part D. Top quartile counties had 3.1 times higher BZD prescribing than the lowest (3.4% vs. 1.1%; F = 3293.8, df = 3, p < 0.001). Adjusting for county-level demographics and health care system characteristics (including supply of mental health providers), counties with more adults with at least some college had lower odds of high-intensity prescribing (per 5% increase, adjusted odds ratio [AOR] 0.80, 99% confidence interval (CI) 0.73-0.87, p < 0.001), as did higher income counties (per US$1000 increase, AOR 0.93, CI 0.91-0.95, p < 0.001). Top quartile PCPs prescribed at 6.5 times the rate of the bottom (3.9% vs. 0.6%; F = 63,910.2, df = 3, p < 0.001). High-intensity opioid prescribing (AOR 4.18, CI 3.90-4.48, p < 0.001) was the characteristic most strongly associated with BZD prescribing. BZD prescribing appears to vary across counties and providers and is related to non-patient characteristics. Further work is needed to understand how such non-clinical factors drive variation.

Access to Abortion in Central Appalachian States: Examining County of Residence and County‐Level Attributes

Studies of how women's individual characteristics and place of residence are related to variability in gestational age at the time of abortion have not examined county of residence and county-level characteristics. The county level is potentially meaningful, given that county is the smallest geographic unit with policy implications. Data on 38,611 abortions that took place in North Carolina, Virginia and West Virginia in 2012 were used to study the relationship between gestational age and county-level attributes (e.g., metropolitan status and poverty). Three-level hierarchical linear models captured individuals nested in county of residence, clustered by state of residence, and adjusted for individual characteristics and distance traveled to care. Eight percent of the variation in gestational age at abortion was attributable to county-level characteristics. Residents of counties characterized by persistent poverty obtained abortions 2.3 days later in gestation than those from counties not characterized by that level of economic hardship. Women living in nonmetropolitan counties obtained abortions 1.7 days later than those living in metropolitan counties, even after distance traveled and county-level poverty were controlled for. County of residence is relevant to gestational age at the time of abortion for women in these three states. Evidence that county-level attributes are related to access adds insight to the consequences for women when the landscape of abortion service delivery shifts. Integrating county of residence into research on access to abortion services may be critical to capturing disparities in access.

Religion and youth educational aspirations: a multilevel approach

ABSTRACTPrevious research has demonstrated that individual religious beliefs and practices are associated with educational outcomes. However, less is known about the manner in which the overall religious cultural influence of a religion can affect an individual’s educational aspirations. Using multilevel analyses on two waves of the National Study of Youth and Religion (NSYR) merged with county-level variables from the US Census, and the Religious Congregations and Membership Study (RCMS), the study shows that a county’s higher Jewish population share is associated with a higher level of educational aspirations for the youth, regardless of being Jewish or not. This Jewish contextual effect remains robust even after controlling a wide range of individual and county-level variables, such as individual academic achievement and religious identity. Consistent with the relevant theory and literature, the negative main effect of county-level conservative Protestant population share on educational aspirations is explained by county-level socioeconomic characteristics. The same mechanism is used in part to explain the initially positive relationship between Catholic population share and educational aspirations. Finally, the study highlights the importance of using a multilevel approach to examine and provide understanding of the relationship between religious influence and educational outcomes.

Overlapping geographic clusters of food security and health: Where do social determinants and health outcomes converge in the U.S?

We identified overlapping geographic clusters of food insecurity and health across U.S. counties to identify potential shared mechanisms for geographic disparities in health and food insecurity. By analyzing health variables compiled as part of the 2014 Robert Wood Johnson Foundation County Health Rankings, we constructed four health indices and compared their spatial patterns to spatial patterns found in food insecurity data obtained from 2014 Feeding America's County Map the Meal Gap data. Clusters of low and high food security that overlapped with clusters of good or poor health were identified using Local Moran's I statistics. Next, multinomial logistic regressions were estimated to identify sociodemographic, urban/rural, and economic correlates of counties lying within overlapping clusters. In general, poor health and high food insecurity clusters, “unfavorable cluster overlaps”, were present in the Mississippi Delta, Black Belt, Appalachia, and Alaska. Overlapping good health and low food insecurity clusters, “favorable cluster overlaps”, were less common and located in the Corn Belt and New England. Counties with higher black populations and higher poverty were associated with an increased likelihood of lying within overlapping clusters of poor health and high food insecurity. Generally consistent patterns in spatial overlaps between food security and health indicate potential for shared causal mechanisms. Identified regions and county-level characteristics associated with being located inside of overlapping clusters may be used in future place-based intervention and policy.

Mapping hot spots of breast cancer mortality in the United States: place matters for Blacks and Hispanics.

The goals of this study were to identify geographic and racial/ethnic variation in breast cancer mortality, and evaluate whether observed geographic differences are explained by county-level characteristics. We analyzed data on breast cancer deaths among women in 3,108 contiguous United States (US) counties from years 2000 through 2015. We applied novel geospatial methods and identified hot spot counties based on breast cancer mortality rates. We assessed differences in county-level characteristics between hot spot and other counties using Wilcoxon rank-sum test and Spearman correlation, and stratified all analysis by race/ethnicity. Among all women, 80 of 3,108 (2.57%) contiguous US counties were deemed hot spots for breast cancer mortality with the majority located in the southern region of the US (72.50%, p value < 0.001). In race/ethnicity-specific analyses, 119 (3.83%) hot spot counties were identified for NH-Black women, with the majority being located in southern states (98.32%, p value < 0.001). Among Hispanic women, there were 83 (2.67%) hot spot counties and the majority was located in the southwest region of the US (southern = 61.45%, western = 33.73%, p value < 0.001). We did not observe definitive geographic patterns in breast cancer mortality for NH-White women. Hot spot counties were more likely to have residents with lower education, lower household income, higher unemployment rates, higher uninsured population, and higher proportion indicating cost as a barrier to medical care. We observed geographic and racial/ethnic disparities in breast cancer mortality: NH-Black and Hispanic breast cancer deaths were more concentrated in southern, lower SES counties.

Individual- and county-level predictors of cervical cancer screening: a multi-level analysis

ObjectivesDespite the gains in cervical cancer screening, there remain persistent socio-economic, geographical, racial, and ethnic disparities. This study examines the combined effect of individual- and county-level characteristics on the use of cervical cancer screening tests such as Papanicolaou (Pap) tests in Texas. Study designCross-sectional study. MethodsIndividual-level information was obtained from 2014–2015 Texas Behavioral Risk Factor Surveillance System (BRFSS). Using the county of residence of the study population, the BRFSS data were linked to the American Community Survey (2010–2014) and the Area Health Resources File (2015). Women aged between 21 and 65 years, with no history of hysterectomy, and residing in 47 counties in Texas were included in the study (n = 4276). Multi-level logistic regression was used to assess the independent influences of individual- and county-level covariates on receipt of a Pap test in the past 3 years. ResultsThe odds of timely Pap testing were lower among women aged greater than 50 years, single women, and those with low education and income (<$25,000). Black women who reside in counties with higher percentages of Hispanics (quartile 4) were less likely to be screened compared with black women living in counties with a low Hispanic population (adjusted odds ratio [OR] = 0.08 [95% confidence interval [CI]: 0.02–0.37]). County-level socio-economic status, although associated with timely screening in bivariate analysis, was not a significant predictor of screening after controlling for individual characteristics. ConclusionsThere are significant disparities in the uptake of cervical cancer screening across Texas counties. Individual-level socio-economic disparities as well as the number of obstetric-gynecologic physicians in a county are predictors of these disparities.

Geographic Variation in Gender Disparities in the US Radiologist Workforce

PurposeTo assess geographic variation in gender disparities in the US radiologist workforce. MethodsGender, location, and practice affiliation of all radiologists and gender of all nonradiologists were identified for all providers listed in the Medicare Physician Compare database. Variation in female representation among radiologists was summarized at state, county, and individual practice levels, and associations with a variety of county-level population characteristics were explored. ResultsNationally, 23.1% (7,501 of 32,429) of all radiologists were women versus 46.6% (481,831 of 1,034,909) of Medicare-participating nonradiologists. At the state level, female representation among radiologists was overall highest in the Northeast and Mid-Atlantic regions (Washington DC, 39.3%; Massachusetts, 34.3%; Maryland, 31.5%) and lowest in the West and Midwest (Wyoming, 9.0%; Montana, 10.7%; Idaho, 11.7%). At the county level, female representation varied from 0.0% to 100.0%, with weak positive correlations with county-level population (r = +0.39), median household income (r = +0.25), college education (r = +0.23), English nonproficiency (r = +0.21), mammography screening rates (r = +0.12), Democratic voting in the 2016 presidential election (r = +0.28), and weak negative correlation with county-level rural population percentage (r = −0.32). Among practices with ≥10 members, female representation varied greatly (0.0% to 100.0%). Female representation was higher among academic (32.3%) than nonacademic (20.6%) radiologists, and in states with higher female-to-male relative earnings (r = +0.556). ConclusionCompared with nonradiologists, women are underrepresented in the national radiologist workforce. This underrepresentation is highly variable at state, county, and practice levels and is partially explained by a variety of demographic, socioeconomic, and political factors. These insights could help inform and drive initiatives to reduce gender disparities and more actively engage women in the specialty.

MP21-10 THE VARIABLE IMPACT OF RACE AND MORE IMPORTANT FACTORS ON PROSTATE CANCER MORTALITY

You have accessJournal of UrologyProstate Cancer: Epidemiology & Natural History I1 Apr 2018MP21-10 THE VARIABLE IMPACT OF RACE AND MORE IMPORTANT FACTORS ON PROSTATE CANCER MORTALITY Heidi Hanson, Christopher Martin, Brock O'Neil, Erik Mayer, Claire Leiser, and William Lowrance Heidi HansonHeidi Hanson More articles by this author , Christopher MartinChristopher Martin More articles by this author , Brock O'NeilBrock O'Neil More articles by this author , Erik MayerErik Mayer More articles by this author , Claire LeiserClaire Leiser More articles by this author , and William LowranceWilliam Lowrance More articles by this author View All Author Informationhttps://doi.org/10.1016/j.juro.2018.02.701AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookTwitterLinked InEmail INTRODUCTION AND OBJECTIVES African American race has been linked to increased risk for prostate cancer mortality by both biological and epidemiologic mechanisms. Based on this information, many experts suggest African American race as an important prostate cancer risk factor, potentially exposing these men to variations in screening or treatment. However, some studies that adjust for access to care or other healthcare related variables have failed to find an increase in risk among African Americans. In our study, we intend to measure the relative impact of individual, disease specific, and county level factors on the risk of cancer specific mortality in men with prostate cancer. METHODS Using SEER data linked to the Area Health Resource File we identified men with prostate cancer diagnosed at age 60-79 between 2004-2012. For these patients we analyzed the affect of both individual and county level characteristics and their relative contribution to risk of prostate cancer specific mortality. Stage is based on information collected under the Collaborative Stage Data Collection System (CS) and coded using the CS algorithm. Tumor grade was based on gleason grading system. Patients with unknown clinical values were placed into a separate unknown category. RESULTS Figure 1 shows the variable importance measures. As expected, stage, Gleason score, and PSA are the most important predictors of prostate cancer survival. County level measures of the number of doctors, social vulnerability, and median family income are also important variables in predicting prostate cancer mortality. CONCLUSIONS The major determinants of mortality in prostate cancer were disease specific factors. Although African American race had a measurable impact, factors related to healthcare access and quality of care (factors known to be associated with African American race) had a stronger affect. Our results suggest the elevated risk identified in African Americans is largely driven by confounding factors associated with African American race. Thus, if factors other than biology or tumor characteristics underlie the elevations in risk found in African Americans, reducing the racial disparities in care rather than individualizing care based on race is more likely to improve prostate cancer outcomes. © 2018FiguresReferencesRelatedDetails Volume 199Issue 4SApril 2018Page: e266 Advertisement Copyright & Permissions© 2018MetricsAuthor Information Heidi Hanson More articles by this author Christopher Martin More articles by this author Brock O'Neil More articles by this author Erik Mayer More articles by this author Claire Leiser More articles by this author William Lowrance More articles by this author Expand All Advertisement Advertisement PDF downloadLoading ...

Understanding risky facilities: an analysis of factors associated with jail escapes in eight states

There is a dearth of research on the factors that make some jails more prone to escapes than others. Clarke and Eck’s (Understanding risky facilities, US Department of Justice, Office of Community Oriented Policing Services, Washington, DC, 2007) risky facilities framework posits there are seven key factors which predict a facility’s risk to crime and disorder. Using data from 88 county jails in eight contiguous states, this study empirically tests if facility-level risk factors can account for jail escapes. Additionally, this research examines county-level characteristics to account for the macro-level explanations. Jails which had reported an escape during the study year were relatively larger, had higher populations relative to their rated capacities, and employed fewer correctional officers per inmate than their control facilities which did not report an escape. This research discusses policy implications in light of these findings.

State entry regulation and home health agency quality ratings

There is a substantial literature assessing the impact of entry restrictions created by state certificate-of-need (CON) programs on hospital and nursing home markets, but comparatively little research has focused on CON for home health agencies (HHAs). We assessed the impact of state CON programs for HHAs, and for potential substitute service providers, on quality ratings for HHAs. HHA quality ratings were obtained from the Home Health Compare database developed by the Centers for Medicare and Medicaid Services (CMS) for the last quarter of 2010 through the last quarter of 2013. The HHA-level data were augmented with county-level area characteristics for each HHA in the CMS database. An ordered logit model was used to estimate the association between state CON restrictions and Low, Medium, and High quality categories, adjusted for HHA and area characteristics. The results indicated that HHAs in states with CON for HHAs were less likely to have High quality ratings, and more likely to have Medium quality ratings, compared to agencies in states without CON for home health. Additional research is needed to assess whether the apparent adverse impact of CON on HHA quality is related to diminished competition among HHAs in states with CON.

Tree-based Claims Algorithm for Measuring Pretreatment Quality of Care in Medicare Disabled Hepatitis C Patients.

To help broaden the use of machine-learning approaches in health services research, we provide an easy-to-follow framework on the implementation of random forests and apply it to identify quality of care (QC) patterns correlated with treatment receipt among Medicare disabled patients with hepatitis C virus (HCV). Using Medicare claims 2006-2009, we identified 1936 patients with 6 months continuous enrollment before HCV diagnosis. We ran a random forest on 14 pretreatment QC indicators, extracted the forest's representative tree, and aggregated its terminal nodes into 4 QC groups predictive of treatment. To explore determinants of differential QC receipt, we compared patient-level and county-level (linked AHRF data) characteristics across QC groups. The strongest predictors of treatment included "liver biopsy," "HCV genotype testing," "specialist visit," "HCV viremia confirmation," and "iron overload testing." High QC [n=360, proportion treated (pt)=33.3%] was defined for patients with at least 2 from the above-mentioned metrics. Good QC patients (n=302, pt=12.3%) had either "HCV genotype testing" or "specialist visit," whereas fair QC (n=282, pt=7.1%) only had "HCV viremia confirmation." Low QC patients (n=992, pt=2.5%) had none of the selected metrics. The algorithm accuracy of predicting treatment was 70% sensitivity and 78% specificity. HIV coinfection, drug abuse, and residence in counties with higher supply of hospitals with immunization and AIDS services correlated with lower QC. Machine-learning techniques could be useful in exploring patterns of care. Among Medicare disabled HCV patients, the receipt of more QC indicators was associated with higher treatment rates. Future research is needed to assess determinants of differential QC receipt.