September 19, 2021

Health insurance company uses analytics to spot members in need

UnitedHealthcare uses predictive analytics to check patients for social determinants of health needs and then…


UnitedHealthcare uses predictive analytics to check patients for social determinants of health needs and then work with them to help improve their social and health outcomes.

TechRepublic’s Karen Roby spoke with Rebecca Madsen, chief consumer officer for UnitedHealthcare, about how the company is using predictive analytics to improve the social determinants of health. The following is an edited transcript of their conversation.

SEE: COVID vaccination policy (TechRepublic Premium)

Karen Roby: Talk to us a little bit about how this all works together, and in the end, how it’s helping people lead better lives?

Rebecca Madsen: Absolutely. And you hit the nail on the head right there, which is how do you identify people in need of support and then provide them the support, and plug them in to needed low-cost or no-cost resources in the community? And we know that health disparities in what we call social determinants of health, that 80% of what determines your health is not by what happens in the doctor’s office, it’s all the other factors. And we saw that especially during COVID, that this was even, in the pandemic, the past year and a half, even more accelerated. We saw 41% of people avoided medical care; and this was disproportionately so for people who had a social determinant of health need. And so, we have long had a program in our Medicare and Medicaid population, and we are now bringing even more focus to our commercial or our employer-based population.

Karen Roby: And when we talk about the needs that people have, Rebecca, are you saying those that may not have access to the care that they need, to the interactions that they may need? What specifically is it that they’re needing to help for, that they may not be able to reach out and ask for?

Rebecca Madsen: Sure. So some examples of the social barriers to health are having adequate food. Having food insecurity, transportation, proper housing, a safe environment, access to the internet. And then also social support; and we saw, especially in the past year and a half, but it’s been an accelerating trend around mental health needs, and how those needs impact an individual more holistically, in terms of their overall health and wellbeing.

SEE: Vaccine passports have to be consistent so that all countries can recognize them, expert says (TechRepublic) 

Karen Roby: I would definitely think, Rebecca, that this year, more important than ever, especially with that focus on mental health; because people have been in lockdown and separated from loved ones, and our mental health has taken a toll. There’s no question about it that we’ve lived through, and are still living through something that we’ve never had to deal with before.

Rebecca Madsen: Absolutely. And what we also have seen, is this has been an accelerating trend for quite a while now. And people think that there’s some new guidance there, but in general, that we’re on the other side of the pandemic. But your health needs do not change overnight; that if you have insecurity about a number of issues, if you’re anxious, if you’re worried, if you’re concerned about things, there’s long lasting implications for that, especially people that already had mental health needs. And so by doing a lot of this work, we’re able to connect with people in a different way.

And the way that we really do that is we have three different tiers of how the program work, or three different steps. The first is the predictive analytics. This is generating a health risk score, aggregating a lot of information. And one of the things that we did, is we worked with the American Medical Association to develop new ICD-10 codes. And what that means is, you go into a provider’s office, and they can code you if you have a social determinant of health need. So, 23 of those ICD-10 codes, as well as pulling in other information about you. And then we’re able to see through these analytics, that you may be more likely to have a social determinant of health need, and need more support. That then gets fed into an agent dashboard. So, when you call me, I, as the agent talking to you, am more aware that this may be a need. And through thoughtful and sensitive conversation, am just more attuned to that, so I can support you differently.

SEE: Smart hospitals could be the future of healthcare (TechRepublic) 

The second area is what we call active listening; that if you have a conversation with me, and you say things such as, “I’m hungry. I’m scared. I don’t have food.” That we know what those trigger words are, and we train our advocates to be able to then say, “OK. Let’s have a conversation, and here’s the support I can plug you in to.”

And then the last piece of it is a questionnaire, where we ask you a number of questions to identify if you have a need.

Karen Roby: Rebecca, really before predictive analytics, these are people, a lot of them would fall through the cracks, so to speak; they wouldn’t get the care that they need. And now you all can, and in such an easier way, identify them earlier on.

Rebecca Madsen: Absolutely. And that’s exactly right. And we know that you may be food insecure today, but not in three weeks. And that’s why we’re very sensitive about how we use the information we have. We’re not sending you an email, we’re not doing an outreach. All it’s saying is, “You may be more likely to be in this category.” And so we’ll have a conversation with you to be able… And we won’t say, “Hey, I saw you may be food insecure.” We’ll say, “How are things going in your life? Is there any more support you need?” And with a lot of the issues around social determinants of health, it’s very complicated. There’s a lot of shame, where people aren’t willing to talk about it; or they don’t know that they can talk to their insurance provider about it. Or, more importantly, they don’t want their family or friends to know. So by creating a safe space, using predictive analytics, we’re putting the two pieces together.

Also see



Source link