Wednesday, August 16, 2017

Using GIS to Compare Areas of Economic, Health, and Social Risk in Appalachia

A recent article in Health Affairs points to widening health disparities between Appalachia and the remainder of the United States. Singh et al. find that disparities in both life expectancy and infant mortality have increased over the last decade and can be attributed to several factors, including higher rates of chronic disease, unintentional injuries, and drug overdoses.


While it’s important to highlight the disparities between Appalachia and the rest of the U.S., we shouldn’t lose sight of the major disparities that also exist within the Appalachia region. Central and southern Appalachia generally have higher rates of poverty and worse health outcomes than other parts of Appalachia. As shown in the map below, higher rates (in darkest blue) of age-adjusted mortality are concentrated in central Appalachia, particularly in eastern Kentucky and southern West Virginia. McDowell County, WV has mortality rates almost 400 percent higher than Union County, PA. Similar patterns exist for a number of other health outcomes, economic indicators, and social determinants.


ApBlog12017-08-15_1620.png


The map below shows the distribution of Economic Status according to the Appalachian Regional Commission, where counties are categorized as Distressed, At-Risk, Transitional, Competitive, and Attainment based on factors such as poverty, income, and unemployment.  The majority of the Distressed and At-Risk counties are concentrated in central and southern Appalachia.


ApBlog2-2017-08-15_1621.png


In order to explore the relationship between poverty and mortality, we combined the two indicators by quartiles – showing the counties in the highest quartile for both poverty and mortality in blue, and the counties in the lowest quartile for both poverty and mortality in yellow. Similar to the patterns above, blue counties are concentrated in central and southern Appalachia.
ApBlog3-2017-08-15_1621.png
The Health Affairs article addresses the disparities between Appalachia and the rest of the U.S.  However, it is important to also be aware of the disparities within the Appalachia region, where focusing on the highest need counties is likely to have the largest impact.


All the maps above were created using the Appalachia Data Portal, one of several HealthLandscape mapping tools that are in the process of being updated. For more information about the Appalachia Data Portal, see our previous post that focused on the impact of hospital readmission penalties on safety-net hospitals in Appalachia - http://healthlandscape.blogspot.com/2015/09/hospitals-located-in-distressed.html.


Michael Topmiller, PhD

Health GIS Research Specialist

Friday, August 4, 2017

Community HealthView Update Brief #40: AIDS Prevalence and CMS Costs Data

During the week of August 3rd, 2017, HealthLandscape updated Community HealthView with nineteen datasets from the Appalachian Regional Commission, AIDSVu, and Centers for Medicare and Medicaid (CMS) Administrative Claims Data.

The data from the Appalachian Regional Commission included economic data for only those counties in the Appalachia region.

The data from AIDSVu on AIDS prevalence were also added for rates by age, race, sex, and county for 2013. The HIV prevalence and new HIV diagnoses data presented on AIDSVu are collected by state and local health departments, and de-duplicated and processed by the U.S. Centers for Disease Control and Prevention (CDC) to meet data quality standards for comparability and reliability. The data reflect persons living with diagnosed HIV infection or persons living with diagnosed HIV infection.

The remaining fourteen datasets all came from the latest CMS geographic variation file. For data year 2015, these data include costs information for ambulatory surgery centers, durable medical equipment, skilled nursing facilities, hospice, imaging, Federally Qualified Health Centers (FQHC), and long term care hospital costs, among others. CMS has developed these data that enable researchers and policymakers to evaluate geographic variation in the utilization and quality of health care services for the Medicare fee-for-service population.

Access the Community HealthView tool in any of our tools including:
Social Determinants of Health Mapper: https://www.healthlandscape.org/map_SDOH.cfm
The UDS Mapper: https://ww.udsmapper.org

“How to” on Community HealthView in the UDS Mapper via this link: https://www.udsmapper.org/docs/HowtoUsetheCommunityHealthViewToolintheUDSMapper.pdf

Tuesday, August 1, 2017

ACA Uninsurance Rates and Potential ACHA Effects

During the week of July 10th to the 14th, 2017, HealthLandscape attended the annual Esri User Conference in San Diego. Part of the conference was designated to showcasing attendee’s various cartographic products in a map gallery. HealthLandscape submitted two products to this year’s Map Gallery, a small format map depicting the mosquito season in the United States in relation to the Zika outbreak, and an Esri Story Map showing a pre- and post-Affordable Care Act (ACA) comparison of uninsurance rates for the United States.  A story map is an online map that allows a map maker to combine narrative text, images, and video clips into a map, to tell a story. 


This Map Gallery submission focused on uninsurance rates before and after the ACA went into effect on January 1st, 2014. From this date forward it has been a requirement for U.S. Citizens to have health coverage without incurring a penalty. The ACA was enacted to provide cheaper health insurance, expand Medicaid, and support innovative medical care delivery methods. This map shows the county rates of the population who were uninsured in 2013 (left), compared with the same map on the right of the swipe bar, using 2015 rates. This shows the drastic decrease in uninsurance rates as gains were made in Medicaid and Health Insurance Marketplace coverage around this time. Most of the areas in the 2015 map that show the least amount of change since 2013- like Florida, Texas, Missouri, or Alabama- are states that chose not to adopt Medicaid coverage. States like California that did adopt the expansion have decreased their overall rates of uninsured. The example shown in the map is for Los Angeles County, California, where the uninsurance rate dropped from 23.7% in 2013 to 12.5% in 2015, a difference of a million people.

Even though the recent attempts to repeal the ACA have failed for now, there is still a desire to repeal or alter the ACA.  The House proposal would eliminate health coverage for 14 million Americans in 2018 and then an additional 23 million by 2023.1 Under this proposal, called the American Health Care Act (ACHA), states like California, that saw a decrease in uninsurance by one million, is projected to have total coverage losses of upwards of 2.5 million people, through Medicaid, employer-sponsored insurance, and private insurance.1 The map shown below was created using data reported by the Center for American Progress and with estimates from the Congressional Business Office (CBO), and shows a state-by-state breakdown of the number of nonelderly Americans projected to lose health coverage under the ACHA.



1. Gee, E. “CBO-Derived Coverage Losses by State and Congressional District”.  Center for               American Progress. May 25th, 2017. 

Dave Grolling
GIS Strategist
HealthLandscape