Helping Health Plans Improve Analytics Approach to Whole-Person Care
Siftwell Analytics is a prescriptive analytics company that provides analysis of health-related social needs data to health plans. Trey Sutten, co-founder and CEO, recently spoke with Healthcare Innovation about trends in this space and some of the ways health plans are using this data to improve patient outcomes.
Sutten has served as chief financial officer and CEO of a managed care organization and as CFO of the North Carolina Department of Health and Human Services, which is responsible for the state’s Medicaid program. Siftwell raised $5.8 million in the company’s first venture capital funding round.
Healthcare Innovation: Could you give an example of the ways that your company works with health plans?
Sutten: We work in four areas. The first is emerging risk, so emerging conditions or increasing costs, worsening conditions, things like that.
The next one is quality measures. How do we help plans understand their members and drive things like medication adherence, wellness visits, and cancer screenings?
The third area is member retention. Who's going to leave your plan, why are they going to leave your plan, and what can you do about it?
The last area is risk adjustment — making sure that plans understand the acuity of their members, and they're being paid for that. The way that the technology works, and it's pretty consistent across all those use cases, is we get their data set, we marry it with ours, and we've got thousands of data points about individuals across the country. We stitch it together with our data sets, and then we use machine learning to make predictions, and then we interrogate those predictions for related explainable factors or related causal factors.
HCI: Can you give an example of how that works?
Sutten: We had a client who said ‘we'd really like to improve our members’ adherence on cancer screening measures.’ So we run the predictions, and we told the plan, you've got roughly 12,000 women that are unlikely to go for a breast cancer screening. Of that 12,000, it breaks down into a bunch of different cohorts with similar characteristics that are driving that non-compliance — either obstacles or blockers of some sort. Of the 12,000 we'll say, there are 65 cohorts. Here’s one cohort with an 80% chance of non-compliance. The reasons they're unlikely to go for the cancer screening is they live more than 20 miles from a screening facility. They don't have transportation. They need childcare when they go. Also, they're from a socioeconomic bracket that it will be important for you to talk to them about the fact that this is a covered service.
So the client starts to make calls, and they collect the information from those calls. We take the structured data and build psychographic models for everybody else that they haven't yet called. Maybe 8% will say they don't want to go for religious reasons. Another 11% will say, ‘Well, I've heard that this can be dangerous and that it might actually increase my chance of developing cancer, so I don't want to go.' We take that information and say, ‘All right, when you start to reach out to the remaining 12,000 people that you haven't yet contacted, think about using faith-based organizations to drive awareness for this group. Think about radio ads in these areas and billboards in these areas to build awareness around the benefits of them. For the remainder, the standard call campaign is the right way to go.’ Now they're carrying all that context that we originally gave them and they’re using different channels, different modalities, to better engage those members, to drive compliance.
HCI: Do you have some thoughts on the collection of race, ethnicity, and sexual orientation data? New York State had just announced they're going to propose that insurers be required to collect that data. First, how big a lift is it gathering and reporting that kind of data and what are some of the issues that the health plans face in dealing with that data?
Sutten: I think that there are a lot of existing processes in place, a lot of points of information collection that can be leveraged. I think that from a plan’s perspective, it's completely doable, and I think it's the responsible thing to do over time. On the topic of data collection, whether it's race or sexuality or gender identification, you get into some tricky issues. Some people may not want to self-identify if there are questions around sexual identity. When you get over into some of the questions about race, I think there could also be issues with self-identification. If you come from the Black community, there have been instances that are well documented where the system has betrayed folks. You think about North Carolina as an example and the eugenics project.
HCI: So there’s a lack of trust in reporting this?
Sutten: Absolutely. So as an individual, what would be my motivation to self-identify, as opposed to prefer not to say so? In terms of health equity on the plan side, as soon as you know you've got a problem, you got to fix the problem. I think that with that information comes the responsibility to do something about it.
HCI: Before they gather that information, are the health plans kind of flying blind as far as knowing how deep the disparities are amongst different groups?
Sutten: I think a lot of plans are flying blind with regard to a lot of different data. I think this is just one example. I was talking to a plan recently that said they've only got 60% completion factors on some of the information that we're talking about. I don't know a problem that you can solve if you don't fully understand it, so I think this is a really important thing to be marching forward on. But I don’t think that this is some sort of golden architecture that's going to fix some of these problems. I think we all need to come together and commit that this is an issue in our country, and we all need to do our various parts that we've got control of to move the ball down the field and advancing health equity.
HCI: Are there reasons why the health plans might have difficulty, even once they have this data, figuring out what to do with it internally vs. turning to a company like yours to help them?
Sutten: What we do relative to what folks have internally is very different, and I'm speaking from firsthand experience. The technologists that we've got on our team just aren't that available, and particularly they're not available in the healthcare space. Technologists like my co-founder are going to work for Google, Microsoft, and OpenAI. They're incredibly difficult to recruit. When you look at the small regional plans, it's hard to find them and it's hard to to afford them.
Part of Siftwell’s strategy is how do we bring the best technologists, combine it with the best minds in managed care, and bring that kind of cohesive set of skills and experiences to bear for managed care plans. What plans are generally doing is correlation and retrospective analysis that's entirely different than prospective machine learning and real artificial intelligence. Everybody's talking about AI right now, right? But there are those of us that actually do it and those of us that talk about it,. What I know from talking to my peers in the field is that there's not a lot of real data science going on inside of managed care plans right now.
HCI: As the Medicaid managed care plans start to get paid in a different way for addressing whole-person care issues, are we going to see more plans putting greater emphasis on this, just because they're getting paid that way?
Sutten: Yes, when you really see the market move quickly and broadly, it's when there are the right financial incentives in place. I was on the board of the Association of Community Affiliated Plans. These are nonprofits. There is a difference between motivations for your ACAP or nonprofit plans versus your commercial plans. But in all instances, even when you're at a nonprofit plan, it’s “no money, no mission,” and so a lot of these things do come in the form of unfunded mandates. In this instance, I think the regulators are showing how serious they are by weighting certain measures and including funding for some of this as well. And we're seeing that in California, certainly in New York, and it's a big emphasis in North Carolina as well.
