In Indiana, Data Drives an Improved Approach to Healthcare Delivery

by   |   July 29, 2015 2:03 pm   |   0 Comments

Hospitals are always looking for ways to improve the quality of care they provide their communities. Information, of course, is a key part of that effort, and Community Health Assessment (CHA) reports, required of nonprofit hospitals by the IRS, can equip care providers with important information about what’s working and areas that require greater focus.

In Indiana, a review of CHA reports revealed the crucial role that data systems can play in improving public health.

Sharon Kandris, left, and Karen Comer

Sharon Kandris, left, and Karen Comer

The Indiana Partnership for Healthy Communities – a collaboration of the Richard M. Fairbanks School of Public Health, The Polis Center, and the Indiana Clinical and Translational Sciences Institute, recently completed a review of their 2012-2013 CHA reports. The effort revealed the importance of data systems like the SAVI Community Information System to improving community health.

Karen Comer, director of collaborative research and health geoinformatics at The Polis Center, and Sharon Kandris, director of community informatics and SAVI, discussed how they are using data to identify and address areas and populations in their community with unmet healthcare needs.

Data Informed: What is the SAVI Community Information System?

Sharon Kandris: SAVI stands for Social Assets and Vulnerabilities Indicators. It is one the nation’s largest community information systems. The SAVI Community Information System helps organizations make data-informed, place-based decisions. SAVI develops a large repository of community data about socio-economic conditions of neighborhoods and builds the capacity of organizations to use the data to be more strategic and effective as they plan, implement, and monitor programs and services.

SAVI’s online mapping and visualization tools allow organizations to access thousands of indicators and community assets about the communities most relevant to them – neighborhoods, school corporations, Census tracts, ZIP codes, etc. Visualizing changes over time and geographic patterns help reveal disparities, service gaps, and opportunities.

What did your review of the CHA reports reveal?

Karen Comer: Preliminary findings included that the majority of hospitals are using only county-level data for both their CHA and their health-improvement plans. This suggests the opportunity to empower the CHA and health improvement planning processes with more local data. Also, most did not inventory community resources with geographic service specificity, but rather just included a list of organization or facility names, without address, contact information, or information about the programs and services offered.

What data sources does SAVI use to provide information for the Indiana Partnership for Healthy Communities?

Sharon Kandris: SAVI collects administrative records, which are often confidential, from about 30 national, state, and local government and nonprofit organizations and transforms them into meaningful indicators about local communities. SAVI provides about 10,000 indicators on socio-economic conditions, health, education, public safety, housing, and more.

SAVI also provides information about the programs and locations of community assets such as health and human service programs, schools, and faith-based organizations.

Related Stories

In Africa, Health Workers Deliver Urgent Care Despite Adversity.
Read the story »

Healthcare Industry Poised to Take Giant Leap into the Cloud.
Read the story »

Playing Catch-up: Healthcare Grasps for Advanced Analytics Value.
Read the story »

Healthcare Plans Improve Member Engagement with Analytics.
Read the story »

Some types of data we collect, transform, and make available at several geographic levels are U.S. Census Bureau data: We collect the ACS and decennial Census, process further aggregations of the data, and publish neighborhood-level statistics across a variety of topics. The Indiana State Department of Health: Data on diseases, vital stats, health clinics. The Marion County Health Department: We collect files of all birth and death records, geo-enable them, and develop neighborhood-level health outcome metrics. Connect2Help: We collect health and human service program information from the local 2-1-1 referral agency to show where resources already exist in a community and what service gaps remain.

What kinds of insights did the SAVI data provide?

Karen Comer: Geographic. SAVI allowed us to understand how hospital responses varied across the state. For example, for each county, SAVI was used to calculate and map the percentage of hospitals that collaborated with their local health departments.

What are some of the areas of healthcare and populations that are being addressed with SAVI data?

Karen Comer: Expectant first-time mothers and their newborns and infants up to 2 years old; prenatal care and education; using SAVI to spatially enable patient records to identity chronic disease populations, such as those with diabetes; domestic violence prevention for victims and perpetrators; program and service planning.

What actions were you able to take as a result of having the SAVI data, and what were the results of those actions? How did these actions and results differ from prior years, when these data were not available?

Karen Comer: Via SAVI, Nurse Family Partnership has been able to identify geographic areas of unmet need toward planning their recruitment efforts for program expansion. The ability to compare the eligible population for health interventions, based on socioeconomic data, with program participant data at a local scale to identify areas of unmet need with more geographic specificity via SAVI Advanced is new.

Percentage of Eligible Mothers Enrolled in NFP in 2013 (Click to enlarge)

Percentage of Eligible Mothers Enrolled in NFP in 2013 (Click to enlarge)


Sharon Kandris: SAVI just launched SAVI Advanced, which utilizes exploratory spatial data analysis (ESDA). ESDA is a very powerful way to explore and visualize data for spatial trends and patterns. It has been around for a while, but the use of this technology in an online platform for program planning and community-based research is just emerging. For example, researchers can (use it to) identify spatial patterns and relationships, detect outliers, and formulate more useful hypotheses before conducting in-depth statistical analysis. And nonprofits can quickly identify geographic areas of unmet need, assess their service area, and prioritize community initiatives.

SAVI Advanced also allows you to create data stories and share them online. Storytelling with data helps you engage with stakeholders to justify the need for a grant, demonstrate program impact, and share research findings.

Going forward, what are the goals in terms of using the SAVI data?

Karen Comer: More targeted program recruitment to areas with significant unmet need and a community environment to support successful program implementation and participant retention.

How can other communities establish a data resource such as SAVI?

Sharon Kandris: The Polis Center consults regularly with more than 20 cities nationally about how to create and sustain a community information system. The first steps should include strategic planning to identify the needs and potential use cases for the data, capacity, and potential partners; establishing relationships with stakeholders, data providers, potential users, and funders; and developing a phased implementation plan.

The Urban Institute coordinates the National Neighborhood Indicators Partnership, a collaboration of 34 cities that maintain neighborhood information systems, all similar in purpose, but different in implementation. This collaboration was formed to further the development of neighborhood information systems in local policy making and community building. The Urban Institute has many great resources for start-ups as well. As a past member of the NNIP Executive Committee, I have helped develop some of these resources for start-ups.

Scott Etkin is the managing editor of Data Informed. Email him at Follow him on Twitter: @Scott_WIS.

Subscribe to Data Informed
for the latest information and news on big data and analytics for the enterprise.

Tags: , , ,

Post a Comment

Your email is never published nor shared. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>