ikaSystems CEO Joe Marabito on transforming health plan IT systems (transcript)

This is the transcript of my recent podcast interview with Joe Marabito, CEO of ikaSystems. We discuss the complexities of the health plan IT world and how health reform is providing new opportunities for administrative innovation. David Williams: This is David E. Williams from the Health Business Group. I'm speaking today with Joe Marabito. He is president and CEO of ikaSystems. Joe, it's nice to speak to you. Joe Marabito: David, thanks for inviting me. I'm glad to be here today. Williams: We're going to talk about IT systems for health plans. To start us off, can you give me a sense of what the typical IT infrastructure looks like for commercial health plans? Marabito: IT infrastructure in health plans is very large. For large plans, it's 20% of their administrative cost. That's basically due to the IT strategy that is historically pursued, which is to provide best of breed software on a department-by-department basis. That software is hosted and maintained by the health plans in their own facilities. On top of that, most plans customize the software to meet the unique needs of their business. So, as you might imagine the strategy requires a large investment, not just in the software itself, but also in the people, the hardware and the physical space. If you walk into a plan, you'll find is a large staff of people that maintain the custom software and barely knit these disparate systems together, because all those systems weren't really made to go together. It presents a pretty significant challenge for plans because each has its own data model, its own architecture, its own processing cycle, and when you try to tie them all together, you end up with a lost common denominator effect. Williams: Talk a little bit about what some of those departmental systems are. I know we don't have time to go into every last one. Most people would be familiar with a health plan having a system for paying claims. But presumably, things go beyond that. What are some of the other main administrative systems? Marabito: A plan has hundreds of systems. The core ones are claims systems, sales systems, enrollment and membership systems, customer service systems -- the systems that the reps use when people call in or perhaps stop in to a customer service center -- billing systems, care management systems --a combination of managing authorization, managing cases, managing diseases-- and then analytics.   Williams: We hear a lot about what providers are doing with health IT adoption, both as part of the stimulus from when Obama came into office and then the Affordable Care Act. But you don't hear so much about legislatively imposed changes on health plans. But how does what's happening in the provider world in particular affect what health plans have to do with their IT systems? Marabito: First of all, there's tremendous change happening at health plans, not just in the IT group but throughout the entire plan. I think at their core, their business models are being turned upside-down. This is due to a lot of things, stemming not just from regulatory changes and what's happening in the provider world, but also market conditions. Let's take them one by one. The regulatory impact is both direct and indirect. Right now, old plans are spending a lot of money and time remediating their systems for the new ICD-10 code requirements. It's a massive undertaking for health plans. Some of have likened it to the year 2000 change, because you're changing essentially the field length of a variable that is pervasive throughout all of their systems. It’s so huge that some plans have decided not to do it. They decided simply to replace their systems with one like ours. Moving to a new system that’s already ICD-10 ready rather retrofit an old system. The plans are also changing their IT strategies as a result of regulatory requirements that may not touch providers. The mandated loss ratio is a good example. If you have to hit an 80% or 85% mandated medical loss ratio, you really have to rethink your administrative cost, infrastructure and how you do business.  Another great example is changes in CMS Medicare Advantage reimbursement. In the future, reimbursement is going to be based on your performance or your star rating, which is a combination of many performance-based measures. And the overall level of reimbursement is going down. So plans are trying to optimize to play under new rules that have pretty pervasive implications for how they set up infrastructure of their IT support systems. And the delegation of risks to providers -- whether it be through accountable care organizations or bundled payment programs or just old-fashioned "we're going to give you the economic responsibility for a bundle of services on a commercial basis" -- is also requiring plans to invest in a level of integration with providers that they haven't had before and to work collaboratively with them in order to facilitate care to members. So, if I took a step back, I'd say there are probably four major trends that are causing quite a bit of change in health plans. One, the shift from group to individual sales. That's a sea change from where plans were just five years ago when they were geared to and focused on selling to groups rather than individuals. Two, the advent of exchanges. This is another regulatory change that may not affect providers but has pretty dramatic effects on health plans. Because what it does is it not only forces them to sell to individuals, it forces them to sell to individuals in a forum that's designed to facilitate cost-based competition. So they've got to figure out different ways to differentiate themselves in such a forum and maintain their fair share of members. Third, payment for results managing better care, better engaging members. And finally, cost and price regulation. MLR targets are one manifestation, but in many states prices are regulating as well. So they're getting squeezed from a cost perspective as well as a pricing perspective. And they just have to fundamentally find more efficient ways to run their businesses.  I'd go so far as to say health plans are in crisis right now. A lot of the crisis is hitting the IT departments. Williams: It's a daunting set of changes and challenges that you listed out there. Usually I would fault somebody for saying that an industry’s in crisis or the models being turned upside-down, but it really is not an understatement in this case. I’ll pick out one part that you mentioned: rules about medical loss ratio. A couple of years ago, you might look to spend a dollar in administrative cost in order to save a dollar or two on the medical cost side. But now, the equation is really totally different and you need to focus specifically on the administrative cost, perhaps above all else. But let me back up. When somebody's under threat, they are also sometimes conservative in what they do. And certainly, I could imagine that as much as a plan is worried about how they're going to save money, they may be also worried about what's going to happen if they swap out their legacy infrastructure for something new and risk the whole business going down. So, to what extent does that fear exist within the industry? And how does somebody like an ikaSystems -- who isn't as well know as let's say a TriZetto --  how do you step up and address that? Marabito: David, I think your observation is exactly right. Plans are insurance companies first and foremost. So they are conservative by nature and they are concerned about risk. And I think if you look at system replacements across the industry, historically, the failure rate has been fairly high. And the reason for that is they tend to be massive projects, they tend to be multi-year projects and they tend to have cost in the hundreds of millions of dollars.  And so, a lot of them never succeed. And when I say a lot of them, I think the overall failure rate is probably in the 30% to 40% range. And if you're going to replace your systems and spend that much money, you want to be darn sure that you can do so successfully and that the investment is going to pay off. And so there is a lot of caution as plans approach replacing their systems. We've also thought differently about that. One of the reasons those projects have so much risk is because they tend to take a Big Bang approach. They build the system on the side, build all the pieces and parts and on one fine day, flip the switch, close their eyes and hope everything goes well switching over the processing from one system to another. We think that's a bad idea. Number one, you do introduce risk if you do that, because you're just piling on the number of things that have to go right in order to be successful. And two, you slow your speed to payback. You want the systems in as fast as you can put them in because they will start accruing business benefit for you. So, we take an incremental approach. We have 24 modules, but we will never implement all 24 together. We'll recommend to our clients that they implement them in clusters, in a shorter time frame as they go along. And what this does is it significantly de-risks the implementations and brings greater speed and value to the project. So it's been an approach that has worked quite well in the past and also has the added benefit of letting plans adapt their implementation approach and their choices as they go along. Because each of these implementations changes the business in a fundamental way. A great example is that most plans are making pretty significant accommodations to make up for the limitations of the legacy systems that they're running. They're sub-optimal and the work isn't flowing as it should. They're spending more money than they should to execute their businesses. If you take a system like ours that was recently built and is very, very flexible, it liberates them to conduct business in a different way. And so, what traditionally happens is they rethink how they approach their businesses. They rethink the jobs, the workflow, the positions, and the number of people that they need to execute a certain task. And that fundamentally changes their business. And so, as they go into an implementation phase by phase rather than Big Bang, the lessons learned from the prior phase and for the next phase makes them smarter and  better as they go along. So that's the biggest change that plans face and that's how we've addressed it. But plans also face other challenges. One is knowledge. And when I say this, you might laugh, but most plans don't have good knowledge of their own systems and their own data. And you'll ask, "Well, how can that be?" It's because the system that they're using has traditionally been around, usually been around for 20, 30, even 40 years. And if you had a system that you've owned and maintained for that number of years, you haven't documented everything that goes into it. And there have been so many changes that had been made cumulatively it would be impossible to sift through that documentation even if you had it. So, what we find when we go through replacing a system is that plans don't know what they have. And that gets discovered along the way. For example, it's not unusual for a client to provide specification for a data conversion to us several times. Because as they go through it and we go through it, we both learn more about the true structure of the data as they convert it. Another challenge is that you're talking about a pretty radical change in the approach to IT. Historically, you've had licensed software that's been customized and maintained by the plan from a plethora of different vendors that is hosted on a mainframe or a client server farm in a basement somewhere and is backed up somewhere else by the plan itself. You are now shifting to a platform that is essentially cloud-based. All you need is a browser to run our system. Our software is a single version of a program that's a same version that's run by other plans but customized for your unique needs through configuration rather than unique code. It is hosted by Ika and backed up by Ika, and ika provides all the security. That's a pretty radical change if you're a CIO and you've been managing your health plan for the last 10, 15 years in the same way. You now have to think about totally different approaches to managing your stuff and dealing with your vendors. Williams: So, Joe, I'm hearing all of these challenges that the health plans have trying to migrate from legacy systems. I understand that what you're doing is a helpful way to do it and it takes the risk out, and an incremental approach can work. But I'm wondering about whether there are opportunities in the market for startup health plans that are not burdened by these legacy systems. In other industries, you would typically see some competitors come up with the next generation of an approach that might be systems-based. I’m thinking back to the days of AT&T and MCI. MCI for a long time was able to offer a 'friends and family' approach that AT&T couldn't do because of the limitations of their systems. Somehow, we don't see that with health plans. It must be that either the opportunity isn't so great or there are other factors like regulation that hold back the startups. Marabito: I think we're about to see it. We're going to see it in the form of ACOs. In some ways, an ACO can be described as a small health plan, because ACOs are coming in to existence today and they're targeting memberships of anywhere from 20,000 to 100,000 members. Some of the largest ones may project 300,000 members or so. But most of them are local delivery systems whose catchment area isn't all that large. And they're not expecting more than, say, 50,000 members. So that's nothing more than a small startup health plan. If you're in that situation, you have to use a model like ours. Why? Because you can't afford the upfront cost. You can't afford to invest in an old legacy system. You can’t afford three years to get it in place. You need something that is economical to implement but also to operate. You need something that's flexible, because if you've seen one ACO, you've seen one ACO. They have many different incarnations. Third, you need something you can get in right away. You're not going to wait three years to implement something. And so, this kind of a model does take away some of the scale advantages of larger plans. Does that mean the world is going to turn to small plans? No, I don't think so. After all there are other scale advantages that large plans have, mainly as it relates to data and analysis. But it has become much easier for smaller plans to compete and new plans to emerge based on the kind of technology that we run. Williams: So, Joe, last question. With all the change that's going on throughout the health plan world, where is ikaSystems pointing its own resources both in terms of business development and product development? What are the sweet spots for your business? Marabito: First of all, our vision is to transform the business of health care. We want to enable the transformation of health plans. We see what they need to do. We see where they need to go and we want to facilitate their getting there. How are we going to do that? One, we want to substantially lower their cost structure. And so we target a 50% cost reduction. It's a pretty audacious goal but it's one that we've achieved in practice and our customers will attest to that. And how do we do that? We do it through continuous improvement of our systems, automating as much as we can, integrating the modules as best as we can; delegating work to constituents where it makes sense to delegate that work. We also provide the flexibility that plans need to adapt to their business. A lot of different business models are emerging and plans are all having different reactions to the economic and regulatory stresses that they're under. That means we need to provide them with ultimate flexibility. How do we do that? We design in more and more configurations for them and we give them the self-service to do so. And finally, we are investing a lot in facilitating competition in the government sector. Medicare and Medicaid plans are a pretty big area of change. You know about the dual Medicare-Medicaid pilots. Many plans are interested in that business. Most platforms today don't facilitate that business. Ours is ready and able to do that, so we're getting a lot of inquiries in that regard. Medicare is changing, so we have a lot of software that helps plans not only comply with what have become more and more stringent rules that CMS has applied to Medicare Advantage plans, but also to drive economically under a different set of revenue rules in Medicare Advantage. And then, we're also an ideal platform for ACOs. That is a completely new market for us, a whole new set of buyers. We're making sure our system can adapt to their needs and specifically, that means multi-tenancy. That means being able to run multiple customers of ACOs, off of the same hardware-software footprint, which further economizes on their cost structure. So those are the areas where we're investing. It falls under the general umbrella of transforming health plans. I've been speaking today with Joe Marabito, president and CEO of ikaSystems. We've been talking about the challenges facing health plans, the role of information technology and what ikaSystems is doing to help address these issues. Joe, thank you so much. Marabito: Thanks for your time today, David.