Episode #207. Healthcare Has a Data Problem, Not a Technology Problem, with Inderpal Kohli, Chief Technology Officer at Healthix

Summary

Risk Never Sleeps Podcast – Episode #207

Title: Healthcare Has a Data Problem, Not a Technology Problem
Guest: Inderpal Kohli, Chief Technology Officer at Healthix
Host: Ed Gaudet
Date: April 13, 2026

---

Episode Overview

This episode explores the challenges and opportunities in health information exchange (HIE), focusing on why healthcare's core issue is data quality and integration—not just technology. Inderpal Kohli, CTO at Healthix, outlines Healthix's transformation into a data intelligence organization, discusses standards like FHIR and TEFCA, and reflects on his journey and broader industry trends. The discussion centers on data messiness, interoperability, building trust, and insights-driven healthcare.

---

Key Discussion Points & Insights

---

Healthix: Scope and Role

[00:40 – 01:45]

• Healthix is one of the largest public HIEs in New York:
  • Connects with over 9,000 entities, 82 health systems, and 27,000 physicians in the NYC metro area.
  • Handles hundreds of millions of patient records, integrating with all major EMRs (Epic, eClinicalWorks, Cerner, etc.).
• Primary mission: Transitioning from pure HIE to a "data intelligence organization".
• Inderpal Kohli’s role:
  • Build platforms delivering actionable insights from messy, complex healthcare data.
  • "My goal is to really build a platform and then aid in the development of insights from the data." (Inderpal Kohli, 00:40)

---

The Challenge: Messy Data, Not Technology

[01:54 – 03:25]

• Healthcare data is inherently "messy" due to:
  • Thousands of disparate entities and "few hundred systems of record."
  • Lack of true standardization even when standards exist.
  • Healthix spends enormous effort on data normalization and identity matching.
• The biggest value-add for participants:
  • Healthix sees the longitudinal patient record—enabling analytics, care gap reporting, and population surveillance.
  • "That's the hardest part with this job today: the messy nature of HIE data...the bulk of our effort goes to normalizing and making sure the data makes sense." (Inderpal Kohli, 02:02)

---

Industry Standards: FHIR, TEFCA, and Beyond

[03:25 – 05:48]

• FHIR (Fast Healthcare Interoperability Resources):
  • "We've been on FHIR for many years...it's one of the standards we're all aspiring for." (Inderpal Kohli, 03:28)
  • Not yet the universal standard—adoption is patchy, but growing.
• TEFCA (Trusted Exchange Framework and Common Agreement):
  • Aims to enable national health information exchange.
  • TEFCA and QHINs (Qualified Health Information Networks) reduce burdens on individual health systems; allows broader, cross-geography data sharing.
  • Current models rely on "pull" (data requested), whereas Healthix operates as a "push" model—delivering real-time alerts and data to care organizations.
  • "The value with TEFCA will be when we really build a critical mass...it may take a few years." (Inderpal Kohli, 04:12)
  • "We are a push model...we receive it, process it, we push it out to all entities in near real time." (Inderpal Kohli, 05:25)

---

Evolving Priorities: Insights and Trust

[06:11 – 07:56]

• Top priorities:
  1. Evolving Healthix into a data intelligence platform.
  2. Maintain and exceed high trust certification—vital for industry credibility.
  3. Build AI/ML-driven insights and applications to leverage the unique data asset.
  4. Impact not just point-of-care but also population health—"We can solve public health problems, help populations, do disease prevalence, surveillance, compliance...a level beyond that is my priority." (Inderpal Kohli, 06:53)
• Trust is foundational:
  • "That's how I convey the confidence to my participants, build trust with them, that we are the guardians of the most important information and we are taking it very seriously." (Inderpal Kohli, 06:41)

---

Threats to HIEs: Business Risks, Not Just Cyber

[07:56 – 09:15]

• Potential business threats:
  • TEFCA could threaten regional HIEs if it achieves a "critical mass" of connected health systems.
  • Loss of participant trust would be disastrous: "If we lose trust...that is where most of the hard work goes." (Inderpal Kohli, 08:19)
  • Healthix's strategy: Layer value-added services (alerts, analytics) on data exchange, focusing on data quality and real-time capabilities.
  • "We don't modify any data, we match and tie it all together from various entities. That's our secret sauce." (Inderpal Kohli, 08:50)

---

Personal Journey: Provider vs. HIE Perspective

[09:15 – 10:47]

• Moving from provider (hospital) to HIE:
  • Broadened exposure: "Now I not just see one health system." (Inderpal Kohli, 10:35)
  • Provider work was broader (network, cybersecurity, applications, AI, etc.); at HIE, focus narrows to data exchange and insight, but applies across many organizations.
  • Career started in Columbia’s biomedical informatics, bridging research and real-world deployment.
  • "What I bring here is a knowledge and inherent understanding of what that data may mean at the point of care." (Inderpal Kohli, 09:23)

---

The Next 10 Years: Interoperability and Impact

[10:47 – 12:03]

• Vision for the future:
  • Desire for interoperability to be seamlessly embedded—not a special, separate project.
  • Persistent problem: Hundreds of vendors and differing state/federal regulations complicate interoperability.
  • "That standardization of interoperability...is a big problem, and it may or may not go away. If it goes away, that's good for us as patients, and as a technology solution provider for the healthcare industry." (Inderpal Kohli, 11:29)
  • Even if standardization is achieved, new problems (cleaning and organizing data) will remain.

---

Early Career & Motivation

[12:03 – 13:31]

• Entry into healthcare IT:
  • Started as a programmer in financial software, ended up at Columbia University Biomedical Informatics after relocating to the US.
  • Early exposure to research and hospital solutions gave him a "bench to bedside" perspective.
  • "A year into it I made a deliberate decision to keep doing it...this is the only thing I know." (Inderpal Kohli, 12:25)
  • Passion for mission-driven work: constantly reminds team of healthcare’s impact.
  • "Mission is so important...that's one of the roles I play every time I bring someone on my team, is to connect them with why we are doing it." (Inderpal Kohli, 13:21)

---

Career Reflections and Advice

[13:31 – 18:00]

• Advice to his younger self:
  • Focus not just on building technology, but adoption and real-world utility.
  • "Technology is 60%, 40%, 50% of the whole equation. It's really adoption for which you need a clinician or an operational leader working with you." (Inderpal Kohli, 14:10)
• Career dreams outside health IT:
  • Sees himself as a "solutions and delivery person." Completion and impact are vital—prefers to deliver (not just develop).
• Hobbies:
  • Enjoys golf (not an expert), gardening for relaxation—"It slows everything down, which is nice. You see something grow to keep nurturing something." (Inderpal Kohli, 15:44)
• Risks taken:
  • Returning to school after a ten-year break, while starting a family—"Very rewarding...at that time, it looked risky." (Inderpal Kohli, 16:05)
• Advice to newcomers:
  • Learn healthcare deeply; don’t just write code but understand users’ (clinicians') day-to-day reality.
  • "Take time to learn about the workflows and utilization...technology is not the job for clinicians, it's a tool." (Inderpal Kohli, 16:45)
  • Advocates for vendor immersion visits (like Epic’s) for developers.

---

Notable Quotes & Memorable Moments

• "HIE data is messy data. So the hardest part is...normalizing that data and making sure it makes sense and identity proving that data." – Inderpal Kohli [02:02]
• "Fire is definitely...one of the standards we are all aspiring for...is it a standard all across? Probably not, but I think we're getting there." – Inderpal Kohli [03:28]
• "Tefka, just like other solutions right now, is a pull model...whereas we are a push model. The data is pushed across almost real time." – Inderpal Kohli [05:25]
• "If we lose trust...I think that's where most of the hard work goes." – Inderpal Kohli [08:19]
• "Technology exists to solve a problem or reduce friction or to add a function...it's really adoption for which you need a clinician or an operational leader working with you." – Inderpal Kohli [14:10]
• "I have a feeling if I'm not doing this, I will be doing something like that—where there is a goal, there's a target to achieve and then deliver it." – Inderpal Kohli [14:41]
• "Take time to learn about the workflows and the utilization...technology is not the job (for clinicians), it's a tool." – Inderpal Kohli [16:45]

---

Timestamps for Key Segments

• 00:40: Healthix’s Scope and Role
• 02:02: Main Data Challenges in HIE
• 03:28: FHIR Standardization Status
• 04:12: TEFCA’s National HIE Vision
• 05:25: Push vs Pull Data Exchange Models
• 06:11: Evolving Priorities – Intelligence and Trust
• 07:56: Business Threats to HIEs
• 09:23: Transitioning from Provider to HIE
• 10:47: Future of Interoperability
• 12:06: Inderpal’s Entry to Healthcare IT
• 13:36: Advice to Younger Self
• 14:41: Personal Passions and Work Philosophy
• 16:05: Riskiest Career Move
• 16:45: Advice for Healthcare IT Newcomers

---

Conclusion

This episode spotlights the fundamental role of data quality, standardization, and trust in advancing healthcare IT. Inderpal Kohli compellingly argues that while technology frameworks matter, real progress requires relentless focus on clean, connected, and useful data, organizational trust, and understanding real-world clinician workflows. Transforming HIEs into insight-driven platforms promises to benefit not only care providers and systems but, most importantly, the patients themselves.

Transcript

[Inder Kohli]

Foreign.

[Ed Gaudette]

Welcome to Risk Never Sleeps, where we meet and get to know the people delivering patient care and protecting patient safety. I'm your host, Ed Gaudette.

Welcome to the Risk Never Sleeps podcast in which we learn about the people that are on the front lines delivering patient care and protecting patient safety. I'm Ed Gaudet, the host of the program, and today I am pleased to have a guest back to the podcast. Inder Kohli. Welcome.

[Inder Kohli]

Thank you. Thank you. I'm glad to be back here.

[Ed Gaudette]

Give us listeners a quick update on where you are now, a little bit about your organization and your role.

[Inder Kohli]

Sure. I am working with one of the large health information exchange, a public health information exchange in New York called healthex. I'm chief technology officer there, predominantly working on their transformation from a health information exchange to our data intelligence organization. We manage and exchange Data with over 9,000 different entities, and then now with all the data, we have a ton of intelligence. And my goal is to really build a platform and then aid in the development of insights from the data.

[Ed Gaudette]

Now, is that across the nation or within the New York area?

[Inder Kohli]

It's within the New York City metropolitan area, Southern New York.

[Ed Gaudette]

Okay, so a lot of clinics, a

[Inder Kohli]

lot of health systems, 82 health systems, around 27,000 physicians, total 9,000 entities within federally qualified health entities. And it's on a daily basis, we kind of reconcile a couple of hundred million patient medical records across all health systems.

[Ed Gaudette]

Now, do you integrate with things like EPIC and.

[Inder Kohli]

Yes, EPIC and all of them.

[Ed Gaudette]

All of them, yeah.

[Inder Kohli]

You can think about it. From eclinical works to EPIC to cernus, all the systems utilized by our health systems.

[Ed Gaudette]

Got it.

[Inder Kohli]

Okay.

[Ed Gaudette]

What's the hardest part about your job these days? Integration has been a lifelong journey for many people. Trying to get that right. It's hard.

[Inder Kohli]

So the hardest part with this job today is just the messy nature of HIE data. As I mentioned, thousands of entities and across multiple few hundred systems of record, put it that way. There is not really a standard, and then there is a standard and there is a level of sharing which happens across health systems. But then I think the bottom line is an HIE data is messy data. So the hardest part is, which is what we bring to the table here is our knowledge and intelligence and normalizing that data and making sure that data makes sense and identity, proving that data. That is where I think a bulk of our effort goes. And then really the next piece to this is how is this useful for our participants? And that is where we add value added services on top of whether it's alerts, whether it's gaps in care report. Because we have our health system see data within only their four walls or however many hospitals they have. We see a longitudinal patient record, we see their life history across the systems. So that value is what we want to offer to our participants in a way that we can look at from a population perspective, we can look at from a surveillance perspective, we can look at from an analytics perspective and research perspective. I think that's the. It's very novel concept. So hard to do. So hard to do, yeah.

[Ed Gaudette]

Now is fire winning out? Is fire becoming the standard?

[Inder Kohli]

Fire has been, we've been on fire for many years. So fire is definitely, I will not call it the standard is one of the standard we are all aspiring for. Right? And then if you go on a de identified route then there are research standards like OMOP models and common data model by oma. Is it a standard all across? Probably not, but I think we're getting there because now with the tools and the solutions we're trying to build on data, it becomes even more important that behind the scene we have a trustworthy, well connected data. And I think that's the hardest part for an HIE like us to build that trust across our continuum.

[Ed Gaudette]

So a lot of talk about tefca. Where does that play in your world?

[Inder Kohli]

TEFCA is what government has been trying to do for many years and I so hope they succeed in a way of a national health information exchange. So. So I will give you an example. Prior to this I was at Englewood Health in New Jersey. So I was exchanging information with a New Jersey hie. So what it brought to us is a data from within the state of New Jersey or little crossover from New York. But my system did not have any records from beyond the geographical area. For most part TEFCA enables that. Right. If someone is visiting from Utah or somewhere and they see care here, if everything is done right and the data is shared across and they build the critical mass, we have that information. Right. One thing TEFKA is doing different than last time the model was implemented is that we have this qualified HIN model qhin which kind of takes a burden little bit off from a health system. We're already dealing under too many projects and enables it. So there are many Q hints, including EPIC switch. So sharing of data becomes easy. Then they also get around that fact of information blocking compliance. Right. That helps with that. But I think the value with TEFKA will be when we really build a critical mass. So it May take few years and that time we have to see where the value comes. Tefka, just like other solutions is right now is a pull model and that's where places like healthex come in that we are a push model. So the data is pushed across almost real time. We receive it, we process it, we push it out to all entities in near real time.

[Ed Gaudette]

I better probably a better model when it comes to health records.

[Inder Kohli]

Right. From a health record because otherwise like we EPIC has the same thing. EPIC Care Everywhere is a pull model, very valuable tool because all EPIC organizations data is available there. But there's this proactive action that needs to be taken.

[Ed Gaudette]

Yeah, there's a little latency too in that pull. Yeah, exactly. You look out over the next couple years, what are some of your top priorities?

[Inder Kohli]

So my top priorities are number one is really an evolution of my platform to, to support really what I need to do as an evolution of us as an organization to be more a data intelligence side of things. Right. Definitely. HIE is a big piece of it and we are doing it very well. We have perfected it. It's really now the transformation to the next piece of it. So that and then underlying principle remains the same as always which is clean data, well connected data to allow building of the trust with our participants. That's number one. The second is we are a high trust certified organization that's most critical for us. We manage one of the most critical assets from across organizations. So staying on top of it. Staying on top of it and not just meeting the requirements but exceeding it because that is how I convey the confidence to my participants, build trust with them that we are the guardians of the most important information and we are taking it very seriously. So that remains my immediate implementation side now not even saying couple of years, but as soon as I'm able to, the goal is really to now start working toward building some insights. Right. Whether it's an AI ML tool or whether it's another algorithm, whether it's an app on top of it. I think that is definitely going to be my priority. To utilize the asset, we have to really develop some insights. So not just digitization of the data which we have, it's really trying to impact the point of care or trying to impact the population level because we have that benefit of here. We can solve public health problems, we can help the populations, we can do disease prevalence, we can do surveillance, we can do common things like compliance with immunization, all of those things. But now a level beyond that is that what I will put on what

[Ed Gaudette]

are the biggest threats to an hie? Not so much from a cybersecurity perspective, more from a business model perspective.

[Inder Kohli]

Well, some would call TEFCA could be a threat to us. Right. If federal government entity becomes against the critical mass of all health systems and suddenly there's all this data is available, clean, connected and operational, yeah, that could be a threat. But that's where I think we are evolving by adding value, added services in real time alerts, clinical alerts, that insight. Yeah, but even today, clinically we generate thousands and thousands of alerts every hour for health systems. They rely on those things. So I think the to in partly to mitigate it, partly to then evolve into a more market intelligence and insight platform. But yeah, that's one threat there. The other threat always remain beyond the cybersecurity side of it is if we lose trust. Right. And I think that is where I think most of the hard work goes in making sure you maintain that level of trust within our data. And so that is where I think most HIAs spend a lot of time in perfecting our internal algorithms, processes, programming to make sure that we are consistently improving the quality of our data. As we don't modify any data, we match and tie it all together from various entities. And I think that's our secret sauce there.

[Ed Gaudette]

So you left the provider side to go to an exchange. Basically what was the biggest learning through that transition?

[Inder Kohli]

So I think I'm fairly new here, so I'm still learning the big thing. Right. The big learning, I would say is what I bring here is a knowledge and inherent understanding of what that data may mean at the point of care. Okay. From a provider side to this, I think provider side is more broad because I'm managing all aspects of it, not just one application or one emr. It starts from the network to cybersecurity, digital transformation and AI. Here we manage all of those, but they are more internal to us. From a provider side, it allows us to focus on what really they are looking for. And many times it's not the providers directly, it's their organization who are reaching out. And some providers get involved with the level of details we need mostly researchers. I think that has been my world. About 20 years ago I started in Columbia University in the biomedical and practice research. So it's not totally new for me. I totally understand that. But yeah, I think the scope has in some ways narrowed. But it's exciting.

[Ed Gaudette]

It's broad too because you get to see it a number of different systems

[Inder Kohli]

and entities scope in terms of Application and the surface area to cover is different in one respect and very broad in other respect. Because now I not just see one health system.

[Ed Gaudette]

Yeah, it's exciting.

[Inder Kohli]

Yeah.

[Ed Gaudette]

Where do you see this going in 10 years? Where do you see exchanges going or

[Inder Kohli]

maybe changing if I don't wear my hat as a health or health exchange cto, I would like the interoperability not to be such a deliberate effort. It should be part of how we deploy solutions. Right. Can we get there in 10 years? Maybe because we believe it or not, you think about it and you talk about health systems, then you say, well, it's three or four prevalent health solution vendors there. That's not the case. There are hundreds and hundreds of vendors. Every system has their own. So that standardization of interoperability and then they're governed by local state laws and then federal laws.

[Ed Gaudette]

It's a big problem.

[Inder Kohli]

It's a big problem. And it may or may doesn't look like it's going away. But if I step out from my role in healthex, I would like that to go because that's good for us as patients, good for us as a technology solution provider for the healthcare industry. Because think about it, when this problem goes away of data availability across institutions, the kind of solutions we can build will have a real impact. Because now you're making a decision based on all of the data.

[Ed Gaudette]

Yeah.

[Inder Kohli]

And then the other problem comes is who's going to collate all that? Who's going to clean, organize? It will still be a messy data. Yeah.

[Ed Gaudette]

Different problem probably. How did you get into healthcare?

[Inder Kohli]

Oh, I needed a job at one time.

[Ed Gaudette]

We all do.

[Inder Kohli]

No. So my first few years of my career I was a programmer. I did all the application development for financial software industry. Started in India, Southeast Asia, Middle east and then came to US and I was working for a Citibank for a large project and project got over and I needed another one as a consultant and landed at Columbia University Biomedical Informatics. So at that time I didn't realize, but few months into it I understood what goes on there. It gave me a really good. Because we were developing solutions a for research, but then also solutions at New York Presbyterian Hospital which deploy it there. And that kind of gave me that first sense of bench to bed side. Yeah, right. But a year into it I made a deliberate decision to keep doing it. And few years later, I think this is the only thing I know because

[Ed Gaudette]

it's hard to go back.

[Inder Kohli]

I don't know if I. I definitely not cannot go back to what I was doing. In financial software industry. But this is, I've been doing this for well over two decades but that's how I got landed. Mission is so important I think once you get there because I understand that very well now. This is all I know, right. And that's one of the role I play every time I bring in someone on my team is to connect them with why we are doing it. And it was frustrating in the beginning but now I understand how this is how it works in a. Yeah, that's great.

[Ed Gaudette]

If you could go back in time, what would you tell your 20 year old self?

[Inder Kohli]

Oh, 20 year old self. Since the last 40 years were still very good for software industry I would have said, I would say yeah, you did the right thing. But I think beyond that I would add and this is again I've learned over time, right Technology exists to solve a problem or reduce friction or to add a function. So when I was younger sir, we were just all excited about building a technical solution. We were just less worried about adoption of it and it became very important for healthcare. So I will go back to the basics of why we are doing it, what problem it's solving, is it reducing friction or adding friction? So and who are your champions on the other side who are going to adopt it? Back then I don't think I had a good visibility of that. Now I understand it and that's what I tell everyone. Technology is 60%, 40%, 50% of the whole equation. It's really adoption for which you need a clinician or an operational leader who is working with you. So yeah, that's definitely I will tell that.

[Ed Gaudette]

And if you weren't doing this job and you could retire and do anything, what are you most passionate about? What would you be doing?

[Inder Kohli]

I never thought of it. I think I call myself a solutions and a delivery person. In other words, development is great but I'm all about delivering in the end. So this industry happened to know very well. I have a feeling if I'm not doing this I will be doing something like that where there is a goal, there's a target to achieve and then deliver it rather than. So in other words, I won't be in that bench side research where I'll spend two years and not get anything out of it. But once the research has produced something then I will take it and bring it to people.

[Ed Gaudette]

Nice. Any hobbies?

[Inder Kohli]

Not a whole lot. I am pretty bad at golf but I still like to try and go out and do that.

[Ed Gaudette]

I like you get better the more you play it.

[Inder Kohli]

Yeah. I know. And that's the problem. That balance is a problem. Finding the balance is intentional. It has to be intentional. Otherwise, I do some gardening I like. I mean, not big, not being into vegetables or other things, but flowers and stuff and brushes. And so I like doing it. It slows everything down, which is nice. You see something grow to keep nurturing something.

[Ed Gaudette]

Yeah. You know, it's nice and connected to the earth.

[Inder Kohli]

Yeah.

[Ed Gaudette]

Which is a good thing because we're so unconnected when we're doing our day job.

[Inder Kohli]

And that's exactly what it helps. It's just totally different. Yeah.

[Ed Gaudette]

What's the riskiest thing you've ever done?

[Inder Kohli]

I think you asked me this question last time too.

[Ed Gaudette]

I did, yeah.

[Inder Kohli]

And so between that then and now, the risk hasn't changed. I think it was still risky for what I did. Very rewarding is going back to school after a gap of 10 years. And I've just. We were having our first child, a big financial undertaking. I took with this and I told everyone, I'm going back to school. So there was no way, like I could just keep it in hiding for a few semesters, see how I do. Yeah. So it was risky for that time, but it was very rewarding. I got like one of the best education, made the connections, made professional gains with that. Yeah. But at that time, it looked risky.

[Ed Gaudette]

If you could give any advice to someone coming out of school and wants to get into health care, what advice would you give them?

[Inder Kohli]

I think they need to take time to learn about health care. And that is a challenge with a lot of technology vendors and companies because their staff members have experience as a patient, maybe not depending on which country they are working in. And they don't really quite understand how their technology will be used. And I'm going back 10 years where physicians used to count click. And nothing wrong with that. You're like, my God, I was able to do this way and now it's seven clicks to get to it. And when I will talk to developers, they would not even think that as a problem. But they have to walk the walk with a physician to see if you have added burden to them or a nurse. Then in many ways we are not successful. So that is what I will say. Take time to learn about the workflows and the utilization of it. Because unlike I was in a financial software industry, a trading system is the job all day. But here, technology is not the job. And that distinction is something which people have to learn. So I know EPIC does this immersion visit. Several other probably organizations have tried this. I think it's very valuable for developers to to understand how their products are utilized in real world. I think that connection will help.

[Ed Gaudette]

Sage advice. Yeah, I love that. I love that. This is Ed Gaudette from the Risk Never Sleeps podcast. If you're on the front lines protecting patient safety and delivering patient care, remember to stay vigilant because Risk never sleeps.

Thanks for listening to Risk Never Sleeps. For the show, notes, resources and more information and how to transform the protection of patient safety, Visit [e-mail] that's C-E N S I N E T.com I'm your host, Ed Gaudet. And until next time, stay vigilant because Risk never sleeps.