Background: U.S. municipal governments administer many health-affecting programs and policies, yet often lack city-level metrics to characterize population health, relying instead on county-level measures, which have been more readily available. This lack of data parsed to cities’ actual geographic boundaries could result in municipal leaders planning public health programs based on inaccurate measures.

Objectives: To assess potential inaccuracies that could result from using county-level measures to characterize health and health drivers in cities.

Methods: The study included 447 large U.S. cities (populations > 66,000) that are completely contained within--yet not coterminous with--their surrounding counties. We compared four public health and socio-demographic measures­­­ parsed to city boundaries, as presented on the City Health Dashboard, to the same measures calculated for the counties that contain those cities. Measures included: percent children living in poverty, percent non-Hispanic black, age-adjusted cardiovascular disease mortality rate, and life expectancy at birth.

Results: There was substantial variation in the size and direction of city-county differences within and across metrics. County-level metrics tended to differ substantially from city-level measurements, which yielded higher child poverty, percent non-Hispanic black, and cardiovascular disease death rates, though lower life expectancy at birth, than the counties that contained them.

Conclusion: When examining public health in cities, city-level measures may frame the case for municipal-level funding and intervention with valuable precision. Municipal governments and other stakeholders can avail themselves of city-level data from publicly accessible platforms (e.g. City Health Dashboard).

Background: The U.S. Asian American (AA) population is projected to double by 2050, reaching ~43 million. Despite this, the distribution of social determinants of health (SDH) and health measures in cities/neighborhoods with significant AAs/AA subgroup populations has not been characterized.

Objectives: (1) Map the distribution of AAs/AA subgroups at the city- and neighborhood-level in 500 large U.S. cities (population>66,000). (2) Compare SDH and health outcomes in cities/neighborhoods with significant AA/AA subgroup populations to those in cities/neighborhoods with significant non-Hispanic white (NHW) populations.

Methods: Maps were generated using 2017 Census 5-year estimates. SDH and health outcome data were obtained from the City Health Dashboard. T-tests compared SDH (unemployment, high-school graduation rates, childhood poverty, income inequality, segregation, racial/ethnic diversity, percent uninsured) and health outcomes (obesity, frequent mental distress, cardiovascular disease mortality, life expectancy) in cities/neighborhoods with significant AAs/AA subgroup populations to SDH and health outcomes in cities/neighborhoods with significant NHW populations (significant = top population proportion quintile).

Results: The distribution of AAs/AA subgroups varied substantially across and within cities. There were few meaningful differences in SDH and health outcomes when comparing cities with significant AAs/AA subgroup populations versus significant NHW populations. However, many SDH and health outcomes were less desirable in neighborhoods high in AAs/AA subgroups versus neighborhoods high in NHWs.

Conclusion: When comparing cities/neighborhoods with significant AA populations versus significant NHW populations, city-level data obscured substantial variation in neighborhood-level SDH and health outcome measures. Our findings emphasize the importance of granular data in assessing the influence of SDH in AA populations.

Background: The City Health Dashboard, launched in 2018, is a free online resource, providing over 35 measures of health outcomes and drivers of health for US cities and associated census tracts. The Dashboard recently added data for 256 cities (populations 50-66,000) in addition to the 500 cities already displayed on the site. This expansion into smaller cities presented unique challenges in harmonizing newly added census geographies with source datasets.

Methods: To depict new city and sub-city census tract boundaries, the Dashboard’s data/GIS team made multiple analytic decisions, balancing the need to accurately parse secondary datasets to census boundaries with the desire to reflect city stakeholder understandings of their cities and neighborhoods.

Results: Key challenges shaped the team’s analytic decisions, including: (1) Dashboard map boundaries are static, yet source datasets utilize varying census boundary years, necessitating careful census boundary year selection to reduce data discordance; (2) City and tract geographies do not always align, requiring complex GIS analyses for accurate resolution; (3) The Dashboard only presents city geographies understood by the census to have independent municipal governments. Some city geographies, though, are considered to have independent governments in some parts of the county but not others (e.g. townships do not have governments in the Northeast region, but do in the rest of the country). Analytic decisions must accommodate this heterogeneity.

Conclusions: Harmonizing secondary datasets across census boundaries, especially for smaller cities, requires careful attention. Cities with limited resources can obtain city and tract data from public platforms like the Dashboard.

Hospital utilization rates have been used to gauge the healthcare resource needs at local geographic areas that are served by the hospital. Such a perspective relies on the assumption that most patients visit the hospital that is nearest to their place of residence. Though many rural hospitals offer a broad array of services, residents may sometimes choose more distant facilities for inpatient and/or outpatient care services. Depending on the extent of the bypass phenomenon, hospitals may experience financial distress, reduced service offerings, or closure. This study aims to explore trends of hospital bypass visits in Iowa.

County-hospital origin-destination reports obtained from the Iowa Hospital Association were used as the data source for analyses. We distinguished local outpatient visits from bypass outpatient visits to Iowa hospitals using the “source” and “destination” fields provided on the database. Hospitals are classified by their locations, and whether or not they were “critical access” hospitals. Geographical hospital utilization analysis was conducted on bypass visits in 2013 and 2016. The results have been mapped using geospatial mapping of the region in order to show patterns of bypass. The study also includes demographic data and potential motivators for the observed trends.