Unite Us Live: Program Integrity Amid Policy Change

Welcome, everyone, and thank you for joining us today. I’m Katie Keating, and I lead our Solutions Engineering team here at Unite Us.

As public sector leaders navigate a rapidly evolving policy environment, they’re being asked to do more than ever: strengthen program integrity, demonstrate accountability, and deliver meaningful outcomes for the communities they serve.

These are not simple challenges, and they require both strategic leadership and best-practice approaches to implementation.

Today, I’m excited to be joined by Keith Gardner, our senior vice president of regulatory and government affairs. Keith brings more than two decades of experience working across government and the private sector, including serving as chief of staff to the governor of New Mexico.

Throughout his career, he has helped shape policy, navigate regulatory change, and build partnerships that drive impact at the state and local levels. Today, we’ll explore how leaders are thinking about program integrity in this changing landscape, the priorities emerging across the states, and what it takes to align oversight and compliance with meaningful community impact.

Keith, thanks so much for joining us today.

Hey, thank you, Katie. Excited to be here.

Love it. Alright, so let’s set some table stakes for everyone listening. Program integrity means different things to different people. When you’re talking with state leaders today, how is the conversation evolving beyond fraud prevention to broader questions about accountability and outcomes?

I think it’s interesting because, as you look across the political spectrum—left, right, center, wherever—most rational people believe that those in need need help, and that we, as a compassionate society and a great giving society as Americans, should serve those in need.

I think what’s interesting is we’re starting to see a change in the expectation of those needs being met and how those needs are met. Really, it’s about real results for real people. I think that’s what the public is now demanding. They want to actually see results as part of this.

The conversation has evolved beyond just the fraud portion, which is a problem and has to be addressed. But really, then it goes into waste and abuse. Let’s talk about what the outcomes are—not outputs, but outcomes. Are we really giving the services at the appropriate place, at the appropriate time, with the appropriate amount? These are finite resources, and if you do it the right way, you can help even more people.

I think that’s really interesting. Sometimes government is really focused on what happens inside of a building, but community is really about what happens for a person everywhere. I think some of the exciting conversation going on right now is: if someone has a food need, for instance, are they applying for SNAP? But also, are they going to a food bank? Are they getting support from a family member with a meal on Sunday? What does it really take to serve somebody and make sure that they are getting to that place of self-sufficiency, while also making sure that safety net is there for them?

Yeah, it is funny that we want to do a one-size-fits-all sometimes in this space as well, because government likes that, right? If you can fit it into a nice, clean, tight box. But the truth is, just like all of us have specific needs, so do the individuals that are in need. It’s having that flexibility, but yet attaching accountability on the backside of that flexibility. Again, these resources are finite, and you want to help as many people as we possibly can.

Absolutely. So in today’s landscape, there’s a lot going on, of course. There are new federal policies. There are increased expectations around everything from Medicaid community engagement to rural health initiatives. What do you think are the biggest questions leaders are asking as they think about oversight in this environment?

I think the high-level leaders, the folks who are at the top of this, are really trying to figure out: okay, we have this influx, especially with rural health specifically, but in other areas. There’s an influx of dollars coming into the system. How do we take those dollars and make sure they’re spent in a way that doesn’t increase this ripple effect that we see sometimes with these programs?

The unintended consequences of adding to the siloing within agencies that we see so often in government, or adding to the fragmentation, as we call it, of systems not communicating. You solve for X, but then you hurt Y and Z. I think that’s the biggest question a lot of leaders are asking. We have this coming to us. There are these changes going on. How can you use this as an opportunity to solve well beyond our own tenure?

Most of the folks who are in the chair today won’t actually see the true effect of all this, so it’s putting aside the politically expedient for longer term. The way to do that is data, right?

Yeah.

Isn’t it crazy that in government, we collect all this data? We’re constantly consuming and pulling data. But then it’s like, what are we doing with that? If you ask some of the leaders, you’ll go through one of their intake forms and go, “Okay, well, where does line 16 go? What is done with line 16?” A lot of times, it’s, “I don’t know. It was there two years ago, so we just kept it.”

We collect a lot of data, and we need to do something with that data to better improve and further these programs, make sure there’s integrity in the program, and make sure the right people get served.

I love that policy focuses on data, because as a technologist, I also think of data as my problem. When we have a shared language, I think it goes far in helping us understand what’s actually happening.

The question about where that data goes is really interesting. For so long, data was really focused on who the person is that I have to report up to, either at the state level or what my federal oversight body requires of me. As opposed to, if I am looking at a SNAP program that reports up to USDA and Medicaid, which reports up to CMS, but I have the same person that I’m serving, what does that data look like when I can start to look at it at that person level, as opposed to the program I’m reporting on?

Yeah, and it’s crazy, right? When you look at these budgets, budgets are tight always. It doesn’t matter how good or bad things are going in the economy; budgets seem to always be tight. You’re having to make priority-based decisions on, okay, if we put X dollars here, what is the real impact?

Most states have a pretty defined base budget. You have to pay for police. You have to pay for fire. You have to pay for incarceration. You have to pay for all these different things. So you take these other resources that are finite and say, okay, if we leverage a few extra dollars here, what does that ROI look like?

One, what’s the impact to the individuals? Because that should be first and foremost. But then, how can we use that to leverage those dollars to do additional savings? Do we front-load that, and does it save us money on the back end that we can then leverage?

You can’t make those decisions without data. I remember back in my days in the legislature and as chief of staff, sometimes we were just hoping. We didn’t have the data set, so you would just go, “Yeah, maybe if we do it here, we think this will go.” Sometimes we got it right, and sometimes we got it wrong.

I think we’re at a place now in 2026 where we have so much more information we could use to leverage to do that. I think that’s what states are seeing, but that’s not comfortable for government completely. It’s kind of a new space for them. It’s exciting. I think that’s what’s so cool about this.

I love that. I would hope we’ve come further than throwing darts at things, but sometimes you just need to pick a place to start.

I also love you talking about outcomes and ROI, because sometimes when people think about government, whether it’s a technology system or a new fire truck, what is the return on investment? Sometimes folks think about public sector spending as just, there’s cash and you spend it. But as you said, if we’re trying to serve more people, if we’re trying to get to a different place that really moves the needle, you have to think about what is the effective and efficient way to spend that money.

Yeah, they’re not infinite dollars, right? That’s the realization. They’re not infinite, and you have to start to look at it more. I hear a lot that government is a business. This is a little bit different. We’re talking about humans here. But we have to look at outcomes.

Government likes big numbers, saying, “We did these things. We interacted with 25 people.” That’s wonderful. But if 24 of those did nothing with the interaction, yeah, it’s incredible for that one, but how do we leverage what we got with that one for the other 24?

It’s about outcomes. Until we change our mindset to saying we don’t care as much about the output, we care about what the outcomes of that are, we’re going to continue to see the potential for some of this waste and fraud.

Yeah. When you think about that outcome and that one versus 24, how do agencies strike the right balance between thinking about how you strengthen oversight, but also maintaining access to services for the people who need them? How do you help people figure that out?

That’s kind of the yin and yang in an agency. The good news is we have incredible people in the field in a lot of these agencies who care greatly about human beings. Not that others don’t, but they’re really focused on that. That gets balanced with what some people call—I know accountants don’t like being called bean counters—but the accounting teams. That balance has to be struck.

I think it goes back to the simple issue of, you’ve talked about this before and I’ve had this conversation a lot, Katie, that program integrity is really about striking that balance between the two. We know people in need have additional issues or struggles. Looking at how we can help them get through or navigate this process is the part that starts to strike when we talk about these individuals individually, as opposed to somehow everybody who’s hungry fits into this box, and everybody who’s homeless fits into that box, and everybody who’s lonely fits into that box.

The truth is, they fit into a lot of different boxes. Really overseeing that system, again, using the data to look at what’s happening on an outcome basis, because candidly, I could make outputs. We can do those all day long. I can fudge the numbers. I’ve done it before in my previous life and other things. You just have a way of doing that, but you can’t fudge outcomes. You just can’t do it. I think that’s the important part.

I love the focus on the individual being served. I also think about caseworkers a lot, because I have been doing a lot of technology system integrations at a time where policy is changing. You have to think about all of the rules it affects, all of the things that have come before. Policy documents are littered with “refer back to this earlier thing.”

I think it’s one of the great opportunities for tools like AI, because if you are a caseworker who is trying to talk to someone and have this really person-centered discussion, they might have kids running around in the office with them, right? You have a lot going on in limited time to help.

How can you get a check on what the policy was at the time? What is the policy now? If you’re going to make a change in their circumstances, what else is that going to affect? How can you have that conversation with them?

I think that creates that nice balance as well between the person being served and the person who is often feeling overwhelmed, overtaxed, or not supported enough to have a conversation that is really balanced. It gets a lot to how you help that staff member take the next best action for someone.

We know, especially for families with complex needs, that it can be hard to know where to start. Also, if you’re a new caseworker and you are just getting into this, you don’t necessarily have 30 years of experience. So how can you prompt folks with tips and tricks for where to go next?

I think, again, it goes back to that data conversation we were having earlier. What happens too often is data gets reported up. It doesn’t get reported down, and it doesn’t get reported across.

Yeah.

When they learn—I mean, nothing is more motivating than success.

Yeah.

It feels great.

It does. It doesn’t matter how much you have.

Yes.

It’s probably one of the good addictions in life. It doesn’t matter how much you have, you want a little bit more.

That is a great lean-in to the next question, which is that if we think about public programs that are delivered across different offices, different organizations, different sectors, and how we relate this to the idea of program integrity, how do you think about that in the challenges around visibility, accountability, and how you can better coordinate?

One of the biggest challenges I always saw, at least in my tenure in government, both legislatively and in the governor’s office—and I saw it when I managed physical therapy clinics, same problem—was fragmentation within the system.

Yeah.

Here we are in 2026. I can be on any named computer operating system and go to the internet and have that interaction. But yet, when I have to go to my three different providers for medical records, none of them talk to each other. It’s crazy to me.

I think this fragmentation that we see, that’s why when we see in social care, it was one attraction—there are a lot of attractions to Unite Us when I came over a year and a half ago, mainly the mission of what we do. I just love our mission. But part of that mission is getting rid of fragmentation. How do we unify?

When you see all these competing systems, how do we start to have a single set where the data looks the same, where it can be evaluated, where it’s clean and usable? That’s where we’re going to start seeing, when the health systems communicate with the community benefit organizations and communicate with the different agencies. As you know, government is the worst about interagency communication.

It was crazy to me. We’re in a state of 2 million people where I was chief, and we had 23 cabinet secretaries at one point. I had more later, but you’re like, “Do you guys ever talk? Does anybody ever talk? Do your systems ever talk?” Because they don’t. Everybody’s in their own little world. I think that’s the biggest risk we have. Sorry, long-winded there.

Don’t be sorry. Prior to coming to Unite Us, I was working on a lot of projects with government around bringing together this record of a person across systems. It was one of the things that really impressed me about Unite Us. Instead of trying to bring all of that together, we foundationally start from a single longitudinal record for a person.

So between your health systems, you can share one screening. Between two government agencies as well, you can share what is being worked on for that family to understand whether they are receiving other services. You can send a chat message without having to try and dig through four other systems to find that person’s email address, if you can even find it.

Being able to understand what someone received, when they received it, and whether they have applied for SNAP four times before and the process always falls apart at the interview, or whether something else is the issue, goes a long way when we think about accountability and understanding what has happened before, but also what we need to have happen to get to the outcomes that we’re looking for.

When you talk about that longitudinal record, we start to talk about truly closing the loop in these referrals. Rather than this notion of, somehow, you just say, “You’re hungry? Oh, here’s three food pantries. I wish you well,” pat them on the back, and send them.

The idea that we can actually say, okay, they went, they received, and we saw them receiving over a period of time, but then that stopped. Then they’re back in the system six months later. What happened during that six months?

Having that longitudinal record you referred to, where we can have historicals and pull in, you can do so much more with. From my personal history as a child in a family that struggles, you go through ebbs and flows. There are highs and lows. This isn’t a magic wand where, boom, they’re done.

We wish we could help people that way. But the truth is, we get so much more of a picture from that longitudinal record, rather than saying, “Oh, they were referred out. We just don’t know what happened.” That doesn’t do us any good.

No, and I love that you brought up the closed loop, because even though we talk about it as a way to ensure connections to care and that people aren’t falling through the cracks, it’s also accountability. It is an auditable understanding of: I sent this referral. It was accepted by this organization on this date. This is what happened.

Keith, to your point on the ebbs and flows, when people come back to us, it allows us to look back at what was successful last time. Instead of trying three other things or throwing a dart at the wall, let’s start from what worked last time. Let’s get you reconnected to that person you already trust. Let’s get you reconnected to the organization you’ve already been successful with and go from there.

On the audit and compliance side from the state, they also have an actual, real record of what happened, versus trying to piece together. I think that’s an important part on integrity. We’ve got to give them what they need, because there are issues around integrity. We need to give them the data they need to be able to enforce and do what’s right by the taxpayers as well.

Yeah.

It solves a lot of issues.

It does. It reminds me, too, that prior to Unite Us, I worked with a lot of programs that were serving teenagers, who of course love to text. But that is not auditable in a system anywhere. Being able to have that audit trail of the communication that happened as well goes right back to exactly what happened.

Exactly. Yeah, I love it.

So we know that technology isn’t the whole answer, just like policy is not the whole answer. But when you’re thinking about how to strengthen program integrity, what do you encourage leaders to look for?

I’m looking for the systems that give them the tools they need to do it. We’re not the enforcement folks, and candidly, others who say they can do that, I don’t think there is a system put into place to do enforcement.

We should be on the side of really helping folks, but providing the data to the state that they need about what really is going on on the ground. That’s the partnership here. I think that’s what’s important as you hear people talking in this space.

It was always frustrating to me when you get sold this, “We can do this,” and then you realize, yeah, you’re doing part of it, but you’re not doing it all. I was so excited about the closed loop as I learned more of it. I think I told our CEO, Taylor Justice, “Dude, I would have given my left arm, candidly, to have the data set that is given today back when I was having to make those decisions and advising my boss on where to go.”

Leaders are begging for this kind of information. There’s always, in the market, a lot of chatter. It’s important that they really look at what is a true full-wrap solution for social care. Having the data—not only who is in need and that they’ve been sent somewhere for that need, but what the outcome of that need was and whether we changed that.

That’s why we see great results in certain states, where you’re seeing ROIs of over $150 per member per month. The ROIs are insane, but it’s because you take that data. You don’t depersonalize what you’re doing, but you take the personal component, merge it with the data, and you can change people’s stars. Candidly, that’s what I wake up every morning to do. Let’s just change a star or two. It makes people’s lives better.

It is a good way to think about the day. I love that because it’s wild that the numbers are so great, but it is that ability to take that data out and combine it with people’s medical information to understand services that were paid for, and how investing a little bit upfront in paying for a food box means that people don’t go back to the emergency department and don’t inappropriately use the ED as primary care if they know who to go to for help. That’s amazing, that combining that data allows them to see the kind of impact that’s happening.

Alright, so in the interest of time, we’re getting nearer to the end of our time here, unfortunately. I know you spend a lot of time out traveling, talking to state leaders. Medicaid directors are my favorite right now. As you’re out talking to them, what’s one thing you’d encourage them to prioritize over the next 12 months?

For me, it’s interesting to have those conversations because we’re in such a reactive environment, especially with some of the things going on in the country. You’ll see where there’s been investigation, whatever, so a lot of states tend to react in that environment. It’s shifting that from a leadership perspective and saying, okay, we understand some of these states may even be dealing with some of those issues.

But changing from reactive oversight to proactive management and finding the tools that mesh well with what they do. How do we get rid of this information lag? Especially in health care, we’ll see a lot of proactive work in other agencies, but health care tends to be a little bit behind what others do for some reason, and social care tends to sit in that same area.

Really looking at leveraging the technology—my advice is leverage the technology to become a proactive manager, as opposed to a reactive fixer. It’s a lot less stressful, I can tell you. Watching the change you can see in your system, where you don’t become the agency that is always asking for extra money, where you can actually go, “Hey, we’re doing good this quarter, Governor. Thank you.” Shocking, right? It’ll make your chief of staff very happy, I can tell you.

I love that. I always think about when I first learned that there were whole divisions related to going out and pulling back money from Medicaid fraud. An important part about implementing new policy is really designing for what you want to have happen.

Making program integrity part of your program design work right up front: what are the guardrails you’re going to build in? What are the tools you’re going to give people to ensure that happens? All the way down to, I really love when government focuses on the resident experience.

For so long, I think it was really an afterthought. We were worried about implementing policy, eligibility systems, and sometimes caseworkers if we were lucky. But improving that resident experience as well can make it easier for people to get help. They might only have to come through once. You don’t have to match data across systems if they got all the way through the first time.

Being able to make it clear to them what the expectations are and guiding people throughout the way. We joke here all the time that the word you’re looking for when you’re looking for a resource is not the word government calls it. We need to be able to take language they use and provide it in the language they speak at home, if that’s what’s required.

That is not counter to designing with program integrity up front. They really complement each other and get to the same goals. They are not competing, but can really reinforce one another, I think.

I think the fragmentation issue we talked about has a huge effect on program integrity. If you don’t have fragmentation, you have more light, and with light, there’s less fraud and fewer other issues that come in.

The other part of it is the ease of getting in and out of the system makes a huge difference on whether or not people feel like they can defraud it. Believe it or not, it’s counterintuitive at times, but if you make it easier to get in and out of the system, there’s less risk to the system for fraud, waste, and abuse.

I think about it as someone like me who doesn’t read instructions. I put on my man hat when I go to build something, and halfway through, I have to take it apart. There’s an easier way to make mistakes. When the system is simplified and easier to understand, there’s less risk of somebody from the outside trying to take advantage of that complex system. That’s usually where these folks do that. They take complex systems and take advantage of those complex systems. When it’s simple, it’s a lot harder to do.

Program integrity, as you said, has to be from the word go. You have to think about what you’re building in as you go. I think it’s a great point.

Amazing. Keith, thank you so much for spending your lunchtime with me today, and thank you all for joining us. We hope you have a great afternoon.

Thank y’all.