Watch this Ask the Expert session to learn practical examples of how to integrate data and ask new questions to gain better insight into the needs of vulnerable populations living in rural areas, improving their access to services and outcomes.
Watch the Webinar
You will learn:
How data sharing for analytics purposes can also empower care coordination.
Why advanced analytics is a must-have to serve vulnerable populations in rural communities.
Examples of how to use data for good to improve services and demonstrate impact.
The questions from the Q&A segment held at the end of the webinar are listed below and the slides from the webinar are attached.
Q&A
Do you know of any database that lists individuals experiencing homelessness?
Josh: There are a few options. For the United States in particular, the United States HUD Housing and Urban Development has a standardized database called the Homelessness HMIS (Homelessness Management Information System). Every COC (continuum of care) around the country must have HMIS data, so that is a standardized data set.
It generally is not publicly available so that's challenge one. Challenge 2 is HMIS represents people who have at least been screened for getting HUD funded housing support services, leading to a lot of nuances.
If you have partnerships, especially in more of your local governments (county-level, cities, etc.) and some state agencies, there are ways to be able to share that data. I've been able to do that in a couple of places for care coordination trends, monitoring, and being able to see where folks are hitting now.
Again, there's limitations just like any data source. It's the sample that you've got getting types of services. There's new efforts and research to be able to attempt to identify folks who are homeless even when we may not have firm data on homeless status, and those are looking very promising. One of my colleagues also did work on being able to identify how many of those experiencing homelessness are likely to become high-cost health users down the road.
HMIS would be my go-to place because it's consistent and somewhat standardized around the country for the United States. Outside of that, I would say starting to encourage more forms of data collection and trying to infer who is homeless. I did that on individual data sets is in some of the systems I've worked in. For instance, if somebody was homeless, when they were registered in our system, if the address was one of our clinics, we can infer that they were homeless.
Todd: Couple of quick responses. And Josh with that, I think that HUD report is so incredibly important because, when you look at it, especially on the rural aspect, that shows about 87,607 people experiencing homelessness in the rural across the United States. You wonder, “In the county of Los Angeles, it shows about 60,000 based upon some homeless counts and things like that. Across the United States, shows that 87 are rural.” But think about that, how many people again and that's something that you all have to contend within that rural nature is how many people are underreported or not reported at all. That opportunity to start looking at the data, to start gathering it from multiple sources and from a cumulative or a combination of law enforcement (fire, social services, hospitals, transportation hubs) that's where you're going to start to see the picture and you're going to start to see the opportunities. That's really important because that's still kind of an untouched area in my humble opinion.
We are having issues finding reliable and complete data for socially vulnerable populations nationwide. We have looked at ADI and SVI, but neither offer both a lot of rigor and completeness, and both have several extant issues. What nationwide data would you recommend identifying the socially vulnerable?
It is challenging to get publicly available, robust data sets on this kind of content. I personally encourage folks, where possible, especially oversight agencies who have the rights to PHI/PII level data (particularly for research, evaluation, quality improvement, etc.) to integrate data and creatively look at it and identify meaningful metrics.
The standardized data sets may not take the adequate view we all really want. Identifying the clear evaluation (or policy) question is key here in getting the right outcomes. I also encourage folks to aim for the question you ideally want answered rather than starting with what data is available. And then look creatively at how we can at least get close to those questions. However, here's some data sets my colleagues and I have used that are publicly available (in the US): BRFSS, PLACES, CDC Wonder, SEER, AHRQ SDOH (in addition to the SVI).
Recommended Resources
Using Analytics to Improve Community Health
Data for Good: Enhancing the Partnership of Public Service and Mental Health
Analytics to Support Whole Person Care: From Basic Stats to AI and Machine Learning
Dr. Josh Morgan Blog Series
Moving from SAS®9 to SAS® Viya®
Please see additional resources in the attached slide deck.
Want more tips? Be sure to subscribe to the Ask the Expert board to receive follow up Q&A, slides and recordings from other SAS Ask the Expert webinars.
... View more
