So Lancer and Carlos, when a layman thinks about data analytics at 30,000 feet, how would you describe the craft, the science, just from an initial starting point? What is data analytics?
The ability to monetize data, to make money out of information. It’s all about business, process improvement. So you get a set of data, you take a look at it and it tells you what you need to do with your business to make it successful.
And when you think about a data analytics client in healthcare services, what makes for a good data analytics client? What are they looking for from us?
A person that has not been able to sufficiently create a yield out of information. A good client, a good data analytics client is someone that has data and doesn’t know what it means. So part of our job is to make sense, show trends, analyze that data and present it back to them in a manner that is actionable.
What are some of the tools that you use to do that above and beyond your own experience? I’m thinking more in terms of systems, software, solutions. Where do you go to when you walk into a multi-site healthcare based MSO for performance improvement?
So the EHR and PM systems are a goldmine of information and also general ledger and what’s going on in the staffing. So the hours per clinic, hours per staff, wherever there is a dollar sign usually tends to be my lowest hanging foot.
Okay. So let’s follow that train of thought. So you enter a business from a data perspective, a data ecosystem, and you see the existing range of tools, a large EHR system filled with data. What other systems are you used to spending time in? Financial reporting?
Absolutely. Also billing software, HR software. Anything that has a dollar sign value to it. So we were looking at from a revenue perspective, anything that would deal with demand generation on the marketing side, anything that would deal with patient access, scheduling utilization, distribution of services rendered, scheduling, cancellation, provider schedule, so EHR, PM. And on the revenue side, anything that deals with payers, reimbursements, service, revenue cycle. And revenue cycle, you would probably go through denials, medical necessity, authorization, coding levels, clinical documentation. And then cost, again, following the money we would be looking at paid hours, non-labor paid hours, supplies, and any direct or indirect expenses. So purely data-driven financial monetization of information. And patient end user satisfaction. So whether that’s a patient or what have you, because that’s all dollars. So there’s not necessarily a dollar amount associated with that in a system, but those are the things that drive those dollars.
So you’ve got an EHR practice management solution. You’ve got a billing software. You’ve got a payroll HR system. You’ve got other productivity-based tracking reporting. You’ve got Excel reporting and you’ve got financial accounting reporting. And none of them speak very well to each other and none of them produce timely, coherent, comprehensive reports. What do you do about it?
You map out the relationships and you use an integrated development environment to code. We create a computational layer to unify all of that information. That involves normalizing and creating relationships on the data, having it all centralized in one system for insemination and distribution. In addition to that, it’s, again, in layman’s terms, we’re going to take the data, we’re going to standardize it across all the systems, bring it together. We’re going to make it actionable for the end user in a clearly defined visual methodology that is consistent across all reports and consistent over time. The challenge with the Excel spreadsheets is every individual person’s going out creating their own spreadsheets and they don’t match, the data doesn’t match. They come into meetings, nothing works. The data is not necessarily normalized. So it’s not put into a standard format that everybody knows and understands and knows what the definition of that data is. So in layman’s terms, we’re going to take the data, we’re going to put it in a standardized format, we’re going to present it in a way that’s easy to understand, actionable from your perspective, and allows you to drive business changes to improve revenue or whatever other area of the business, whether that’s outcome, satisfaction, what have you. Data’s data. If it’s not actionable, it’s useless. It’s just data.
So I’m assuming that implied in that process is the ability to know which third party tools, cloud-based and other, need to be used to augment what you are faced with day one. Is that fair?
That’s fair. Especially because of the pricing. Different clients will have different pricing, different budget and computing infrastructures.
And different constraints in terms of the quality of what they’re giving you. So every project is bespoke. How important is the need for you both to understand the actual language, the actual operational context to whatever data is being processed?
The business case is the priority and the ability to reconcile the business case to tools and information is really where we succeed. The larger the variance between the business case and the information that we present, the less effective we’ve been. I think the interesting thing about end users and data is oftentimes they don’t know what they don’t know. So there’s a standard set of reports that come out for revenue cycle that everybody uses, everybody sees. But the key and the important distinction that I think that we’re looking to do is to be able to give insight above and beyond what the end user actually thinks they need. So when they come in, they may say, “I need a report that shows me XXX with regards to our revenue cycle, how much in AR,” all this sort of things. And I think the value add that we bring is we take that a step above and we anticipate what they should be looking for and that’s what we’re looking to provide.
So give the user more than they actually think they need and educate them on what’s possible, not the other way around. So how important is it for a data analytics team to be focused exclusively on healthcare services and beyond that just one clinical specialty within healthcare services?
So in my mind, the answer to that question is going to lie on resource and budget constraints. Analytics as a discipline is very multidimensional. I can go into analytics on statistics on any kind of company and value on many different industries and businesses. As far as niche healthcare services, I think it provides us an advantage because it makes us specialize in what we do. From a specialization perspective, it’s easy to meet with an end user and have them tell you what they want. But the real key is having the experience in that area and knowing more than what that person who’s telling you what they want needs. So it’s important to provide them what they need, but it’s important to have that broader knowledge so that you can provide them that step above. So for example, if I just come in as a data analyst, the needs of data of an anesthesiologist are significantly different than an orthopedist. So having that frame of reference and having worked in those specific silos is very, very different. And I think that’s where we can also add that additional layer of value because if you go to a user, they’re going to tell you what they want, but that may not really be what they need. They just may not know it.
How do you feel about providing solutions across different MSO departments? So when you hear data analytics based reporting, dashboarding in your marketing department versus your billing department versus finance versus clinical risk versus patient outcomes versus finance and accounting, how do you react to each of those potential engagements? Agnostic? You have your bias, your preferences, and if so, where and what for what reason?
I think that it should be in its independent vertical to support cross-functionally all departments. I do think that when you spin business intelligence and analytics solutions in one department, that department tends to outperform the other departments in the company. And what you create is a larger silo in terms of information and abilities within the organization. So I’m biased. I think that it should be a platform solution as a vertical supporting all departments, whether you’re compliance or marketing. I think each department has obviously its purpose and the value of its data. I think the real challenge is identifying within each organization because some departments in some organizations are stronger than the others and some departments have more opportunity for change and therefore improvement. And that’s really what analytics is about, driving change and creating revenue. How we approach it and why we approach it the way we do is more from agnostic perspective. So we don’t automatically look at marketing and say, we think marketing is the key area. We look at the data and then take that data and say, these are the areas that we can see need improvement. And then we help guide the customer into focusing on those specific areas that are going to generate the greatest change.
I would’ve thought that patient outcomes would be probably the most complex because what you are often dealing with there is different EMRs that don’t speak to each other, plus HIPAA constraints plus clinical technicalities. And so it just feels like a very difficult space in which to try to get cohesion versus a billing software solution that just doesn’t spit out very educated reports.
I think it’s different depending upon the specialty. So if you’re dealing in oncology, your outcomes are going to be very different than something like an orthopedist. So the value that it may provide from an outcomes’ perspective may be very different from an orthopedic perspective. We know that over a period of two years, the outcomes are relatively the same regardless what your care pathways are. From an oncology perspective, I think that varies significantly depending upon your physician, the care pathway that you take, the medications, the treatment, the type of cancer that you may have. So I think it’s more complex. But if we’re looking for what’s going to drive revenue and change within an organization, I don’t know that that’s the number one driver, if that makes sense. To your point, I think it’s far more complicated. I just don’t know that that’s where the real driver of revenue’s going to be.
When you think about the market for data analytics, very busy space, lots of solutions out there, large brands like enormous, multi-billion companies, Cerna, Optum, and very small startups, who do you think of as comparable data analytics providers to Scale? And why would a group still be better off working with us?
The advantage of coming to us is our specialized knowledge in healthcare services, that’s an advantage. Some of the groups you’ve mentioned, for instance, Optum, their capabilities are slightly different. In Optum’s case, it’s a lot of insurance claims and value and risk mitigation and assessment, from the insurance side. From our side, we’re much, much more ingrained and tactical inside the MSO. So I would always think that we will have an advantage on decision services, healthcare operations, fee for service terrains. I think the key differentiator is, if you look at an Epic, Cerner or these groups, what they do is they provide you data analytics in specific charts and graphs and all of that, and they leave it to you to interpret. I think the differentiator is when you come to Scale healthcare, we don’t just give you the charts, we do the interpretation for you. So we’re going to see things that you’re not going to see, and because of our experience, we know where to look. So the differentiator for us is not that we can provide you with different charts or we can provide you with the same charts. We know how to analyze that so that you don’t have to. If I take a look at people who have Epic, for example, every Epic user has the exact same charts across everything. But are certain practices more efficient because of it? Yes, they are. And why? Because they have people on their team that understand how to interpret that and where the drivers of those things are within the business. And I think that’s what we do. That’s the value of our consulting. So it goes above and beyond just pretty charts and graphs and providing data for data’s sake. It’s that interpretation.
How has the craft, the profession and what’s possible changed over the last 10 years and where are we heading over the next few years across the industry from a healthcare service perspective?
In terms of the physician practices being a highly fragmented market, you’re operating now with more modern tools at a lower cost, Moore’s law. With that, it’s tilted the industry into a place where they’re much more sensitive and aware to organic revenue opportunities under data. Now that’s saying that we went from maybe 1985 to somewhere between 1995 and 2000 for these clinics. The industry as a whole known to you guys, in my mind, is tilting more towards that value-based, full-risk, half-risk, and keeping people out of the clinics and being rewarded for that. If you look back 20 years ago, we had essentially canned reports that would come out, no graphs, no charts, nothing pretty, nothing to look at, nothing to really see trends on. It was just a matter of this is where data is today. Where we’ve come from that over the past 20 years is now we have advanced analytics. So we can not only just look at what is the data today, but what has the data been like in the past. And we can do projections into the future. And so for example, one of the things we’re doing with a client right now is we’ve taken the history and we’ve looked at it and we’ve said, this is the number of admits that come in over time, what is it going to look like? And we’ve created a forecast that’s accurate within 15% of how many admits are going to come in by day. So we know every day from now through the end of the year, within a 15% margin of error, exactly what our revenue stream is going to look like, assuming we make no changes and marketing and all of that. That wasn’t available 20 years ago. From an AI perspective and machine learning, we can pull data out of areas that was never possible because they were all standard, hardcore documents. Now with natural language processing, we can search through those documents, pull out relevant pieces of information and actually create a story out of it. Couldn’t do that before. From an imaging perspective, AI can look at an image and almost, better than human accuracy, find issues and problems within radiologic images. So the impact of AI and machine learning and national language processing on analytics has been absolutely massive over the last 10 years especially.