Medicaid: Evaluating KanCare's Effect on the State's Medicaid Program [NLPES Award Edition] [August 2019]

In June 2019, LPA was awarded a 2019 Excellence in Research Methods Award from the National Legislative Program Evaluation Society (NLPES) for its report, Medicaid: Evaluating KanCare's Effect on the State's Medicaid Program.  This award is presented to the office or offices that used exemplary research methods in an evaluation or audit report released during a calendar year.  This episode discusses what research methods LPA used to help answer the audit's questions.

Transcript

Andy Brienzo: 0:00

From the Kansas Legislative Division of Post Audit, this is The Rundown, your source for news and updates from LPA, including performance audits recently released to the Kansas Legislature. I'm Andy Brienzo. In June 2019, Legislative Post Audit received the National Legislative Performance Evaluation Society's 2019 Excellence in Research Methods award. LPA won this award for its 2018 audit, Evaluating KanCare's effects on the state's Medicaid system. I'm joined in the studio for this special celebratory episode of the rundown by Matt Etzel, Principal Auditor at Legislative Post Audit who supervised this audit. Welcome to The Rundown, Matt, congratulations to you and your team on winning this award.

Matt Etzel: 0:47

Thank you. Appreciate it.

Andy Brienzo: 0:49

So why don't you start by just briefly reminding our listeners about the questions this audit answered.

Matt Etzel: 0:54

Sure. So the main question in this audit was to evaluate the effect transitioning to a mostly privatized version of Medicaid had on the state's Medicaid costs, service use, and health outcomes of its beneficiaries. So maybe just a little bit of background. In 2013, Kansas transitioned to a mostly privatized version of Medicaid,, which is called KanCare. KanCare had a lot of goals, but one of them was to help control or reduce the state's rising Medicaid costs. One of the ways it was going to achieve that was by increasing the health outcomes of its beneficiaries. So the idea is that KanCare would put a emphasis on preventative services like visits to your primary care doctor and it hoped to increase the health of the state's beneficiaries and reduce their use of emergent services like trips to the emergency room, which are fairly costly, and thus help kind of control the state's Medicaid costs that way. So again our goal was to evaluate KanCare's effect on costs. So it wasn't actually helping to control the state's Medicaid costs, its effects on services. So were beneficiaries receiving more preventative care under KanCare and where the use of emergent services declining and its effect on beneficiaries' health outcomes. So in other words, were beneficiaries becoming healthier as a result of KanCare?

Andy Brienzo: 2:29

So what in general did you find?

Matt Etzel: 2:32

So I'll start with KanCare's effect on costs. What we ended up finding was that the state's Medicaid costs had remained relatively stable over the last several years-about$2.7 billion since KanCare was implemented. But what we found was that although those costs were stable, it wasn't necessarily KanCare that was responsible for that stability. So we were able to do some pretty advanced regression work to try to isolate KanCare's effect on Medicaid costs separate from other variables. And when we did that, we ended up finding was that it was those other variables like changes in the age, race, gender, geographic, location of the state's beneficiaries that were really helping to control the state's Medicaid costs, not so much this transition to KanCare. When it came to looking at KanCare's effect on service utilization, we were a little bit limited, due to some data reliability issues. So we had wanted to evaluate KanCare's effect on 12 different services and it was a mixture of both preventative care services as well as emergent care services, but data reliability issues limited us to looking at just five of those services. We were able to see though was that beneficiaries use of preventative services had increased under KanCare. What we did not see though was a corresponding decline in emergent services. So, for example, we are able to look at a number of inpatient days, so days that a beneficiary would be admitted into hospital. We didn't see a significant decline in that type of emergent care. The last area that we wanted to look at was KanCare's effect on health outcomes. Here, unfortunately, data reliability issues that we ran into completely prevented us from evaluating health outcomes. So we actually had a separate finding in this audit and recommendation in this audit for the Department of Health and Environment, who is responsible for overseeing the state's Medicaid data, to really improve their processes and implement some new controls to help evaluate the reliability of the state's Medicaid data so that they could do these types of analysis going forward to evaluate the effectiveness of KanCare.

Andy Brienzo: 4:50

Now I know NLPES says this award goes to up to three offices each year that have used exemplary research methods in an audit report. So tell me what made your team's methods so special?

Matt Etzel: 5:02

I think there is a couple of things that really made this audit methodology stand out. I think the first is just the breadth and the scope of our work and the data that the team had to collect. So in order to do the types of analyses that we wanted to do, we needed to collect data on every Medicaid beneficiary in Kansas from 2011 to 2016. So that ended up being somewhere around 200 million Medicaid records that we had to collect, stage, and analyze, which was a heavy undertaking. It took us about six months to get all the data that we needed. And as I mentioned before, there were some issues, some data reliability issues that we ran into that caused us to, you know, it required us to continue working with KDHE staff to, you know, rerun queries and constantly evaluate the data. So just getting the data in house was a significant undertaking., but the other part that really made this audit unique was the type of work that we were able to do with that data. So that regression analysis that I mentioned before, to our knowledge, that was the first time something like that had been done on the state's Medicaid data. So being able to isolate KanCare's effects on, for example costs,, from those other variables like age, race, and gender that hadn't been done before. So, you know, reports had been presented to the Legislature and to the public showing that Medicaid costs had remained stable since KanCare's implementation, which was true, but I think what our audit and our methodology was able to show for the first time was that it wasn't necessarily KanCare that was responsible for that stability, right? It was those other changes to the state's Medicaid population that was really contributing to that stability. And that was information that had not been presented before and I think really made our audit unique and impactful.

Andy Brienzo: 7:01

Matt Etzel is a Principal Auditor at Legislative Post Audit. He supervised a 2018 audit, evaluating KanCare's effect on the state's Medicaid system that recently won the 2019 excellence in research methods award from the National Legislative Performance Evaluation Society. Matt, thank you so much for joining me in the studio.

Matt Etzel: 7:21

Thank you.

Andy Brienzo: 7:22

Thank you for listening to The Rundown. To hear more podcasts, subscribe on Spotify or Apple podcasts. For more information about Legislative Post Audit and our audit reports, visit our website ksalpa.org and follow us on Twitter@ksaudit.