Why People Should Care About How Health Care Works (Part 1)

In part one of this two-part Breaking Down Health Care conversation, John Hennessy, MBA, and Michael Kolodziej, MD, dive into the topic of big data: what it is, how it came about, and why we’re still struggling to leverage it.

Read the transcript: 

John Hennessy, MBA: Welcome to Breaking Down Health Care, where we'll be discussing evolving topics on health care in the United States. I'm John Hennessey; I'm a principal at Valuate Health Consultancy. And for this series, I'll be in conversation with Michael Kolodziej, an oncologist and currently an advisor to Canopy, an electronic patient-reported outcome company. 

We're also going to talk about his new Substack, Decoding Healthcare. We're using our expertise to dive into some of the nuances of the health care industry in the United States. And Mike, one area we've circled around a little bit and haven't talked about much is big data. And with the explosion of evidence and treatment options, the increase in the use of patient-reported outcomes, electronic medical records, there's a lot of data out there, but that doesn't mean it's easy to access or evaluate or parse. And so maybe we'll start by defining this term big data. I've struggled with that. And maybe some of the histories and challenges around this thing we call big data.

Michael Kolodziej, MD: Yeah, sure. So, you know, people in the data space have a fairly precise definition of what constitutes big data. Traditionally, it's the 5 Vs. Volume, that is, the data sets are big. Value, that is, the data sets have things that you're interested in. Variety, that is, the data sets have a lot of diversity. Velocity, you have to get the data quickly. And veracity, the data must be accurate. And I would say that big data sets, if you hold them to that precise definition, big data sets don't necessarily currently exist in medicine. More like we got medium data sets that are housed in specific organizations or institutions that don't really talk to one another are rarely shared. And they've been around for a very long time, this is not something new. 

So if you look kind of back we’ve got tumor registries, tumer registry data, right? So all of us who practiced for the last 50 years have contributed to tumor registries and tumor registries are usually run by hospitals or states. They're partially funded by the CDC, 46 states have them. A few states decided that they just don't they don't think the government should have that data go figure. I'm you know, I'm not going there. But, you know, tumor registries are a very tiny piece of the puzzle.

Anybody who's tried to use a tumor registry or contributed to one knows that that clearly is not big data, not even close, right? And then, of course, health plans have big data and I'll include Medicare in the health plan category. They have claims data. Now, claims data tell you the diagnosis and what care was delivered based on the diagnosis codes and the cost associated with that care. But there's no clinical detail. There's this misunderstanding, I think, among medical professionals that the health plans know everything. They do not, I guarantee you, they do not know everything. They know, again, a piece of the puzzle. Are they big data? No, not by themselves. I don't think they're big data, but they could contribute to the big data pool.

Medicare is the biggest data of all. But normal human beings can't understand Medicare data files. You need experts who do Medicare data to understand Medicare data files. Now, Medicare developed a research tool called SEER, S -E -E -R. And they tried to marry tumor registry data with Medicare data. So there is some clinical enrichment of Medicare data. But Medicare SEER is a little bit challenging to work with. You've got to apply to use it. It's got to be for a research purpose, got to submit a grant, and you've got to pay for it. So we have a history of having data at our fingertips, but it's not big data. And truthfully, it's answered some small questions, but not really big questions. 

Hennessy: It's fascinating you talk about that having spent time on the health plan side. It really is a limited data set. It's whatever showed up on the claims form and particularly in oncology, things like staging are missing that are absolutely vital if you want to do something with that data. But, you know, we've talked about, you know, the big data or the medium data we just discussed. But, you know, we've got the EMR has become part of this, and we're capturing a whole lot of ones and zeros there. And there's this concept called meaningful use. And maybe we can define, you know, meaningful use, I think maybe from the federal government standpoint. And is that the same as meaningful use in a practice when you're trying to use the darn thing.

Kolodziej, MD: Yeah, yeah. So, electronic medical record data, in my mind, does qualify for the big data moniker. It has the potential to be big data. But we should remember that electronic medical record data fundamentally did not exist before about 29. And what happened was, as everybody probably remembers, there was this huge recession largely related to real estate stuff. And under President Obama, Congress passed a stimulus package, the Recovery Act. And again, without getting political, those kind of big pieces of legislation have all kinds of stuff tacked on that's got nothing to do with economic recovery. And one of the things that was tacked on was the HITECH Act. And the HITECH Act was designed to try to bring medicine into the technologically modern age by promotion and widespread implementation of electronic medical records. An absolutely worthy goal. And the HITECH Act basically initially provided financial rewards for participating by adopting and implementing a qualified electronic medical records.

Those financial incentives ultimately turn into financial penalties over time. But it worked because we went from a universe where less than 10% of doctors were using electronic medical records to one where now more than 90% are using electronic medical records. The HITECH app identify specific characteristics of the electronic medical record that needed to be provided for you to meet the requirements as defined by the government. And those requirements were called meaningful use. Now, if you go back and look at what the requirements were, and you didn't have to do all of them, it was a cafeteria, right? There were about 26 or 28 requirements. And most of them are just, they're fine. I mean, you would look at them as a physician or somebody in health care and say, that should be there, electronic prescribing, right? Electron, medication reconciliation. I mean, stuff that's perfectly logical ought to be in the EMR. But that was only the first phase and what was supposed to happen where they were supposed to move to phase two and phase three because the goal was to get that data into some place where it could be used and then to use that data to improve health care outcomes. And that's where it fell apart. 

The first phase went fine. We never actually really completely got through the second phase and we sure as heck haven't gone to the third phase. So yes, everybody's got an EMR. No, it hasn't really improved health care in America. In fact, it's sort of the law of unintended consequences. Some things have gone wrong. As I know, you're well aware, we can go into greater depth about them. But I think it's fair to say doctors hate them and patients hate them. Doctors hate them because there are just a million surveys out there that show that the electronic medical record adds hours to your day. Hours to your day. It's almost always listed as one of the top 5 causes of physician burnout. And patients hate them because if you go to a doctor who's very busy, what they do is they sit and look at the computer screen, not at you. And they type their note, instead of talking to you. And I find that problematic. I will say that when I was in practice, and that's a while ago now, although we do use an EMR, my volunteer clinic. I talk to the patient, I do my notes later, my notes suck, but that's irrelevant, right? My notes aren't that great. But the patient comes first. And that's what's more important. So it's been problematic.

Hennessy: As you talk about that, I reflect on the one thing that was an advantage out of the EMR was that the, you no longer had all these stacks of charts and doctors desks in their offices and you could actually find a patient chart, which was sort of a benefit in and of itself. But as you describe, you know, the note taking, you know, it can be highly variable. I think you've described a case where you're sort of taking notes so that you have some record of what happened.
But so often with EMRs, the note-taking has less to do with you communicating information, more with capturing information for billing, which probably isn't something that's enhancing the physician-patient relationship either. 

Kolodziej, MD: Yeah, you hit the nail on the head. So as we tick down the list of what went wrong, one thing that went wrong was that the EMR became a primary source of documentation to justify elevated levels of service and therefore more income. Because as everyone probably recalls, historically Medicare has reimbursed for physician services, E&M codes, based on the documentation. And the documentation was supposed to reflect physician work and time. But all the EMR did was make it really easy to document to the highest possible level.

And in fact, after EMR started being widely adopted, Medicare looked at the emergency room bills they were getting. And all of a sudden, they had taken off. And the reason was ERs were just up coding everything, absolutely everything. And what happened then, of course, was that Medicare said, "This is fraud." Now, it wasn't, I don't know, fraud may be a strong word for it, but they said, "We're gonna pay close attention to this," and they did, and they started auditing charts for documentation of level of service. So that's one of the things that got screwed up. EMRs, and you'll hear this a lot from doctors, EMRs became a tool to optimize billing, not to optimize care.

Hennessy: I think that's spot on and I can remember back in my days running a practice that, you know, we had, you know, I think, you know, oncologists were very good at describing what was going on so they could pass that information on to a referring physician and others who, you know, had the pro skills of Faulkner and could somehow turn a consultation into a 5-page note and had to sort of remind them, listen, nobody has time to read a 5-page note, You've got to find a way to find a happy medium in there. But in any event, so we have these EMRs and as we both have said, there's a lot more documentation in there than there used to be. Certainly, it's more readily available to hand between folks taking care of these patients but also to the data guys.

And so is there something in all of this EMR, all of these ones and zeros that is going to say maybe marry the EMR and that moderate to big data and be able to make something out of this. Is there a pot of gold at the end of this rainbow? 

Kolodziej, MD: Yeah. Well, so I'm going to take a half step back because the other thing that the EMR Gold Rush led to was the incredible growth of EMR companies. And so Epic Healthcare, the EMR of choice for hospitals, the largest EMR provider in the United States, is a $12 billion company with $3 billion in annual revenue. And you are not going to find a lot of fans of Epic in the medical community because it's hard, it's very challenging, doesn't fit into workflow, it's expensive at hell, and it's kind of like a black hole because all that data goes somewhere. Presumably, the institution has access to it. The doctors generally do not. And using that data to answer questions seems to be problematic. And a big issue that explains the problem is trying to define who owns the data. Who owns the data? Because, and I go into this in some detail in the blog, the one party that rarely gets mentioned as the owner of the data, but who in fact is definitely the owner of the data, are the patients. Patients own that data. 

So, physicians think they own the data. I put it in there. It must be mine. And you'll hear this and you and I have heard this in the community and the hospital said it. The doctors think they own the data and they think the data is worth something, right? Now, look, they have to do that documentation to get They don't have the option of doing the document. They must do that documentation. So it's very hard for me to see how they have a legitimate claim to the data. I don't see it. Now, the health plan, Medicare, could argue that they should own the data. They're paying the bill, but they're not actually really paying the bill. The employers or the employees or the Medicare beneficiaries are in fact paying the bill, the health plans are administering the benefit. So they don't have a legitimate claim either. 

In fact, at the end of the day, patients have it. But trying to get it on behalf of patients, good luck with that one. We are stuck. We are stuck in a universe where people think that they're sitting on a goldmine and they don't want to share that goldmine. They think that goldmine belongs to them and that has created an impasse in being able to access this big data set to the benefit of patients.

Thank you for watching this installment of Breaking Down Healthcare. We hope you enjoyed the conversation and learned something you didn't know about health care and how it works in the United States. If you have questions or topics you'd like Mike and I to discuss, you can use the Contact Us feature on the website. Tune in for future conversations because we're just getting started.