How would you define appropriate care, and how does it incorporate concepts of either underuse or overuse?

I would say that appropriateness is a quantitative expression of rightness. It is the conjoint analysis of cost, quality, and the potential for harm. These parameters cannot be examined independently. Maximizing one necessitates a tradeoff in another. We all want infinite quality, but are we willing to pay infinite cost? And while many people consider cost and quality when they’re assessing health care, they don’t always factor in the potential for harm. To paraphrase Milton Friedman, there is no free lunch in the world of medicine. Every intervention entails some risk of side effects. Every procedure comes with a risk for complications.

Practice guidelines have existed for 20 or 30 years. Those guidelines were evidence-based, in other words, as are measures of appropriate care. So what’s different now?

One word—operationalization. Motive has rendered practice guidelines operational. I once asked the leader of a prominent guidelines-issuing agency how long it usually takes for one of their guidelines to make it into clinical practice. That is, how long after you issue a recommendation stating, “Thou shalt do X,” that X becomes a standard practice? His answer was, “17 years.” While one can quibble about the number of years it takes for a guideline to affect practice, one cannot argue that practice guidelines are an effective operating mechanism. The horse will have long ago left the barn before a practice guideline recommends closing the barn door.

But physicians read the journals. Presumably, they know there are better ways to do things.

Some physicians are diligent readers, but most are not. Even guidelines struggle to remain current. Guidelines that are revised every five years often contain out-of-date information. Even landmark clinical trials take too long to affect practice. It’s disheartening to realize that today’s landmark New England Journal of Medicine article won’t make it into your doctor’s practice for years. It speaks to the futility of practice guidelines. As currently imagined, practice guidelines are not an effective mechanism for improving practice.

So do you have a better solution?

We believe we do. Take a granular element from a practice guideline or from some other highly credible evidentiary source—you know, a large randomized controlled trial published in the Journal of the American Medical Association—it’s unacceptable for that breakthrough to not make it into practice in a reasonable timeframe. That’s one thing that our appropriateness measures do. They start to close that gap in time.

Is it possible to close that gap too fast?

It is, in fact, and we’re careful about that. The measures aren’t put out in real time, nor should they be. There is a two-steps-forward–one-step-back nature to medical progress. You don’t want to be reactive. You want to allow for some swinging of the pendulum, such that there’s a high degree of certainty that the new practice is a better practice. But putting a better practice simultaneously into the hands of payers, insurers, and providers significantly shortens the lead time.

This brings to mind the “range of better practice” included in the Motive Practicing Wisely format. Can you talk a little about that?

The range of better practice, or ROBP, acknowledges that there is a range of acceptable approaches that doctors might take to address a given condition or situation, all of which could be equally valid. The ROBP allows us to differentiate between doctors who are practicing within the realm of what the evidence and experts would consider to be reasonable, and those who are, in fact, outliers.

This leads to the broader question of how we think about and apply evidence now.

We think of evidence in a couple of broad ways. One is a hierarchy of credibility. For example, a large, randomized, controlled trial published in The Lancet should be very credible, but a small case report of a few patients, published in a low-impact journal, probably is not. There’s a hierarchy of evidence, and it’s essential for the consumer—that is, the physician, or any critical appraiser of the evidence—to understand that. It’s the same in every field, and in the lay press; an article in the New York Times will typically carry more weight than one in a small local paper.

Are there exceptions?

Sure; with all generalizations there are exceptions, but the exception doesn’t invalidate the rule. That’s one parameter of evidence—that is, its provenance. Another important, but less widely appreciated, parameter is the rational implementation of that evidence. It’s not easy to take that randomized controlled trial and turn it into something that reliably changes practice. That requires expert curation and methodology.

And that’s what Motive Practicing Wisely Solutions offers?

That’s part of it. Our methodology is statistically rigorous. It’s transparent; it’s independently verifiable; it’s repeatable, scalable, and objective. And it’s overseen by a body of experts who are themselves rated according to quantitative bibliometrics and other validated rating schema that can take that granule of evidence and translate it into something operational. That is no small feat; it requires real skill, and this is one thing that we excel at doing. It’s what Motive is all about. It’s an essential piece of the puzzle that is often overlooked by unsophisticated consumers.

I’ve heard doctors described in many terms, but “unsophisticated” isn’t usually one of them.

Doctors are not unsophisticated in their areas of expertise, but we often assume that this sophistication extends to areas in which we are not expert. And one of the most important deficits in doctors is knowledge management, an understanding of the structure and implementation of knowledge.

So when you talk about the company’s methodology, can you explain what exactly happens in the implementation phase?

First, it’s literature searching—and this is searching published peer-reviewed literature, not anecdotes. It’s done using proprietary, high-throughput techniques that employ library sciences and quantitative bibliometrics. A number of elements contribute to this process, such as machine learning, so that we can process the approximately one million medical articles published annually into an evidence hierarchy. The idea is that if we start picking from this hierarchy—or this pyramid, if you envision it that way—we’re picking from the top. We’re going to pick the most important of those million studies for analysis before we pick the second-most. And we’re not going to work our way very far down the pyramid, because weight of evidence is not distributed linearly. It’s a steep curve, and after the first tier—the 20 of the 80/20 rule, if you will—impact decreases dramatically. Our system makes a very sophisticated first cut of the literature, the evidence.

But you also work with raw data, yes?

Yes, we ingest raw data, as well. And now, instead of peer-reviewed studies, I’m talking about physician performance data, patient data, large databases that encompass tens of millions of patients, maybe over a hundred million patients from Medicare and private data. We analyze it in a two-step process that is, first, statistical, and we use techniques akin to those one would find in the “Methods” section of population-based studies, big-data studies published in the leading medical journals. The process requires unique capabilities. We take the output of that purely statistical step and subject it to expert curation. This transforms raw data into actionable knowledge that can affect care—something that actually makes the jump from the lab to the bedside. And I think this is something that makes us unique. No one approaches these tasks with the rigor that we do.

In terms of that last step in the process, the translation into practice, how does that actually happen?

We have a lot of different outputs for our information. It can be used in electronic medical records or incorporated at the electronic bedside. We work with payers, with at-risk providers, and we’ve worked for nearly every big medical information company in the United States. We’d like innovation to have the greatest impact and for that reason we attack it from every angle. Motive wants to make care better, and to do that, one has to interface with the current state of care. Right now, it involves a lot of different outlets for our information, and if you’re serious about transforming care, you have to take this broad-based approach. It’s naïve to think that a killer app or some similar deus ex machina is going to be the answer. Rather, it’s going to be a component of the answer. And that’s supported by the slow pace of change in clinical medicine. I wish clinical medicine changed more quickly with the times, but medicine is an incorrigibly stodgy profession.

So, is it fair to say you’re applying knowledge in new ways?

We’re operationalizing knowledge. We’re not writing a book to sit unread on a shelf or putting out a guideline to be ignored. We’re working to make knowledge interoperable. Fundamentally, it is simple—and if it’s not simple, it’s not meaningful. We want to make care better.