Unlocking the Power of Predictive Analytics in Healthcare with Dr. Paulo Pinho, the Chief Medical and Strategy Officer at Discern Health

[00:00:02] Hey, everybody.

[00:00:03] Welcome back to the Outcomes Rocket, recording live here at HIMSS 2024.

[00:00:08] Today, I have the privilege of hosting Dr. Paolo Pino.

[00:00:11] He is the Chief Medical and Strategy Officer at Discern Health.

[00:00:15] Dr. Pino, thanks for joining me today.

[00:00:17] Thanks for having me.

[00:00:18] I appreciate being here.

[00:00:19] The pleasure.

[00:00:19] So before we kick things off, I'd like to just get a little better understanding about

[00:00:24] Discern Health.

[00:00:25] What do you guys do?

[00:00:26] And who's the main stakeholder that benefits from the work?

[00:00:28] We're actually a predictive analytics company.

[00:00:31] We focus on high-risk patients, particularly those in the older age demographics.

[00:00:37] Really focused on clinical outcomes of complicated medical events.

[00:00:41] For example, diagnosis of frailty, diagnosis of congestive heart failure,

[00:00:46] diagnosis of things that lead to declines in cognition,

[00:00:50] utilization patterns in the hospital system, fall risk.

[00:00:53] So high impact things in health care that tend to be high cost

[00:00:57] that can be identified, predicted and prevented.

[00:01:00] And so that's really what Discern Health does.

[00:01:02] We're a pre-seed startup.

[00:01:05] We're about a year and a half old now.

[00:01:08] And a small growing team of data scientists and clinicians that are really focused on

[00:01:12] not only creating predictive models, but creating predictive models with explainability,

[00:01:16] which I think is really important.

[00:01:18] In terms of stakeholders, there's various.

[00:01:20] Obviously, everyone benefits from health improvement.

[00:01:22] Patients, caregivers, which are often the crosshairs of many of the challenges associated

[00:01:27] with care provision for that demographic.

[00:01:29] But then also payers, Center for Medicare and Medicaid Services, providers.

[00:01:33] So really multi-lens in terms of our desire to improve various stakeholders

[00:01:37] within the health care ecosystem.

[00:01:38] Ricky Fascinating.

[00:01:39] Thanks for sharing that.

[00:01:40] And how do you layer into the workflow and the timing of that in the entire process?

[00:01:46] Dr. Sajid So it depends on what data assets are used

[00:01:49] for a particular customers.

[00:01:50] Obviously, there's data assets that are more real time.

[00:01:52] Clinical data, for example, tends to be much more real time.

[00:01:55] We're exploring even some more up to the minute data sets,

[00:01:58] things in the wearable space, etc.

[00:02:00] Which you can get up to the second kind of insights.

[00:02:03] But certainly clinical data is probably the most real time,

[00:02:06] continuous data feed that we have right now.

[00:02:08] Claims data tends to be beneficial in some regards,

[00:02:11] sometimes particularly not only with the concordance with the clinical data,

[00:02:14] but sometimes even in its discordance.

[00:02:16] There's actually some insights that can be achieved there.

[00:02:18] The challenge with that is that there's often a lag.

[00:02:20] There's often not the granularity that you need.

[00:02:23] So in terms of provider workflows, it's really about giving them insights

[00:02:26] into the fact that someone's at risk.

[00:02:28] Most of our models have a long tail that we're predicting somebody

[00:02:31] that's kind of progressing towards frailty in the next X number of months

[00:02:36] or this kind of year, really to intervene before they actually get down

[00:02:40] on that downward spiral that's going to lead to their decline.

[00:02:44] Same thing with something like fall risk, a big precursor is things like frailty.

[00:02:47] So identifying that someone is declining so that you can say,

[00:02:50] hey, let's intervene and maybe do medication management,

[00:02:53] maybe do correction of some of the clinical outcomes

[00:02:56] that the person has for their chronic diseases, et cetera.

[00:02:59] So that's the way provider workflows are embedded in terms of payer workflows,

[00:03:03] in terms of provider organizations and hospital systems.

[00:03:06] I don't think that real time up to the minute at the point of care is necessary,

[00:03:10] but we're really providing that insight into workflows,

[00:03:13] care management solutions, et cetera, to really drive that insight better for them

[00:03:17] so that you really have people that are working top of their license

[00:03:22] to be actors within the health care ecosystem,

[00:03:24] to be able to exercise change and create impact.

[00:03:26] So that's the way we work.

[00:03:27] Ah, thanks, Dr. Pino.

[00:03:29] And is the focus on a particular population?

[00:03:31] Are you guys focused on Medicare Advantage?

[00:03:34] Are you focused on commercial?

[00:03:36] Tell me about that.

[00:03:36] Yeah, so I think Medicare Advantage is a great entry point, right?

[00:03:39] It's really about right sizing, but not only the HEDIS and risk adjustment

[00:03:44] and things like that are associated with what tend to be very transactional,

[00:03:47] but really around the value based care concept,

[00:03:49] really around the care management strategy,

[00:03:51] really creating that care journey for that individual.

[00:03:54] A hospitalization can be exorbitantly priced, right?

[00:03:57] So how do we prevent that?

[00:03:58] Readmissions, there's challenges with follow up in patients that have mission risk

[00:04:03] and the lack of follow up in 50% of patients within the Medicare Advantage space is huge, right?

[00:04:08] So how do we nudge those people to nudge the providers,

[00:04:10] nudge the care managers to get these people the necessary care?

[00:04:13] Because that's a huge cost for our health care system

[00:04:15] and fully worsening with an increased volume of that demographic in terms of patient base.

[00:04:21] But then also dealing with some of the supply issues around health care providers

[00:04:26] that are dwindling in numbers in today's health care ecosystem.

[00:04:29] Totally, yeah.

[00:04:30] Being able to tackle those most important situations with the staffing that we have today.

[00:04:36] Absolutely.

[00:04:36] Makes a lot of sense.

[00:04:38] Look, we're here at HIMSS.

[00:04:40] Why did you decide to come?

[00:04:41] Why did the company decide to come?

[00:04:42] And what's one insight that's risen to the top for you in these last few days?

[00:04:46] Yeah, it's exciting.

[00:04:47] There's a lot of people doing a lot of great things, some of it analogous to what we're doing.

[00:04:51] My philosophy is obviously from an egocentric standpoint,

[00:04:54] I'd love to have everybody like choose us,

[00:04:56] but I know that there's other great solutions that are doing similar things,

[00:04:59] maybe even more prescriptive or more accurate in one particular disease state.

[00:05:03] We tend to have models that are a little bit more expansive.

[00:05:05] But I think what you learn from conferences like this is that it really takes a village.

[00:05:09] I think this model of you have an end-to-end solution that does everything,

[00:05:12] that's become the thing of the past, right?

[00:05:15] Really looking at the entire data value chain

[00:05:19] that leads to the predictive models is a huge piece of it.

[00:05:22] How you exercise an impact on that population,

[00:05:26] the suite of products that exist to support,

[00:05:28] hey, you've identified someone that has a risk,

[00:05:30] now how do you deploy health care to that person?

[00:05:32] So you think about organizations that are doing data access,

[00:05:36] data normalization, data quality improvement,

[00:05:39] lending that data quality to predict the models like ours,

[00:05:42] and then organizations that are able to take it to the final mile and say,

[00:05:45] hey, we've identified this person has a risk,

[00:05:47] here's some community resources that are available,

[00:05:50] here's some technology that's available to prevent falls

[00:05:53] or to identify when a fall happens or perhaps fall 911 earlier

[00:05:57] so that we're not stuck with somebody who has a host of complications

[00:06:01] because they were at home and not attended to for a day, right?

[00:06:06] So I think it's exciting to see that,

[00:06:08] and part of our strategy here is not only to share our message,

[00:06:11] where we have a booth here at 3161, we're in the Microsoft booth,

[00:06:15] but then also to really look at the best of breed solutions

[00:06:17] that are doing great things in health care

[00:06:19] that we could partner with for a common strategy

[00:06:22] of helping this very much at need population.

[00:06:24] Yeah, it's a big need.

[00:06:25] You guys are doing fantastic work.

[00:06:27] For all the viewers and listeners out there,

[00:06:30] what would you leave them with as a call to action?

[00:06:33] And then what's the best place they could find you

[00:06:35] and your team at Discern?

[00:06:36] I think in terms of call to action,

[00:06:38] I think we tend to forget that patient centricity

[00:06:41] is a huge part of why we're all here.

[00:06:43] I don't want that to be forgotten.

[00:06:45] I think that health care consumer behavior is challenging.

[00:06:48] Understanding the challenges of accessing the health care system

[00:06:51] in the United States is a very complex issue

[00:06:53] that affects many families, that affects patients.

[00:06:56] Let's make it so that whatever solutions we're providing

[00:06:59] are patient-centered, right?

[00:07:00] And I think as speaking as a provider myself,

[00:07:02] I've seen the ill effects of burnout,

[00:07:05] the ill effects of mental health amongst providers.

[00:07:07] A lot of it is we have the necessary tools

[00:07:10] that we've learned in medical school

[00:07:11] or our professional training.

[00:07:13] There's technology that exists there.

[00:07:15] Really be technology that augments

[00:07:16] what a provider is able to do.

[00:07:18] Provide that high tech to someone

[00:07:21] that really wants to connect with people

[00:07:23] in more of that high touch capacity.

[00:07:25] And that's where I think a lot of these solutions

[00:07:26] will exercise their greatest impact.

[00:07:29] And the more that I think we can align around that focus,

[00:07:33] the more patients have to tell their story less

[00:07:35] because that data is there.

[00:07:36] It's going to pop, right?

[00:07:37] That story is telling itself to a greater extent.

[00:07:40] And then in terms of reaching out to us,

[00:07:42] we're available on social media.

[00:07:44] We have a LinkedIn profile, Discern Health.

[00:07:46] We have an ex or Twitter account, Discern Health.

[00:07:49] We have a website at discernhealth.ai.

[00:07:52] So feel free to reach out to us.

[00:07:54] Glad to talk to anybody that got interested

[00:07:56] in some of the work that we're doing.

[00:07:57] Amazing.

[00:07:57] Folks, a lot of ways to get in touch with Dr. Paolo Pino

[00:08:00] and his team at Discern Health.

[00:08:02] We'll leave everything in the show notes

[00:08:04] so you could get in touch.

[00:08:05] Dr. Pino, thanks for being with us today.

[00:08:07] Thank you for having me.

[00:08:08] Appreciate it.