Founder Letter: Innovate, Don't Automate: The Slippery Slope of Clinical Support Software

Today, most of us depend on technology to help us tackle some of the more mundane tasks that would otherwise require time and space on our already-too-full schedules. We text instead of call. We order gifts, groceries, and gadgets online instead of waiting in line at the store. And we deposit checks using our phones—if we ever even receive a check (hello, Square Cash and online bill pay)—instead of trekking over to the bank. In essence, we’ve learned to lean on technology in order to bypass the manual (and often tedious) aspects of our lives—and now we’re beginning to do the same at the office.

In fact, there’s no question that technology has greased the tracks when it comes to businesses’ organizational efficiency and financial viability. If you think about it, continuing to manually perform time-consuming, repetitive processes that could be automated drags down productivity and profit margins—and this becomes even more of a setback once your competitors devise methods for automating those processes. And they will. That being said, the push to automate everything creates an interesting quandary for professionals in certain industries—industries such as health care, for example. Despite the fact that technology has advanced significantly over the last decade or so, it still has a long way to go before it can serve as a replacement for the depth of thinking, reasoning, and clinical-decision making necessary for successful clinical practice. And, to be honest, we wouldn’t want it to—because that would effectively reduce our education, knowledge, and skill down to nothing more than an algorithm (albeit a complex one).

Plus, as PTs Larry Benz and Tim Flynn discussed in this this paper, therapists have the ability to dramatically influence patient outcomes based on their interpersonal interactions with each patient—the more positive the therapist’s approach and expectations are, the more positive the patient’s expectations become, and the better that patient fares. In other words, human-to-human connection is still downright crucial.

That’s why I was concerned when I saw a recent ad for physical therapy software that automates the complexity decision-making process necessary to correctly use the new CPT codes for PT and OT evaluations. I know you’ve heard about these new codes already—if you’re a WebPT Member, you’re already using them in the system—but if not, you should watch this webinar (and read this blog post, this one, and this one) ASAP. In that webinar, rehab therapy compliance expert Rick Gawenda and I discussed the importance of clinical decision-making with respect to accurate code selection. Even though the guidelines governing proper code selection are very detailed, they’re still just guidelines—and they can’t account for every unique situation. You and I both know that no two patients are alike—even those who present with the same injury or condition may end up requiring very different evaluations. And I don’t think it’s too much of a stretch to say that the patients who neatly fall into the low, moderate, or high complexity buckets are going to be few and far between. (In fact, we received an avalanche of scenario-specific questions during the webinar and via email in the weeks leading up to the January 1 implementation date.) With this in mind, it’s clear that fully automated, “canned” responses and notes simply cannot account for all the specific details that influence the complexity of a patient evaluation. If they could, providers wouldn’t be clamoring for additional education around coding criteria—and how to apply that criteria.

And that’s not the only reason that total clinical automation is worrisome. After all, it’s a slippery slope. Of course, we should use technology to its fullest extent to help us streamline treatment and documentation processes—for example, technology that guides providers toward appropriate evidence-based outcome measurement tests is great. But we must remember that technology can’t—and shouldn’t—replace the depth of clinical knowledge, experience, and skill necessary to properly complete and document an evaluative episode or create a complete plan of care. That’s what we rehab therapists are here for—it’s what we’ve been trained for.

Furthermore, if we’re not the ones deciding what data to record—because we’re letting an automated system do that for us—then we call the quality of our data into question. And that certainly isn’t going to help us move forward as providers or as a profession. Instead, we should adopt technology that helps us take control of—and improve upon—our data collection efforts, clinical processes, and payments. And that’s especially important now: With third-party payers becoming increasingly vigilant—read: cautious—about the claims they pay out, providers are having to up their game in order to justify their clinical decisions. And things become all sorts of murky when we start automating certain aspects of the care delivery process.

We all know that the problem of patients diagnosing themselves and developing their own plans of care after browsing the web is pervasive—and, quite frankly, dangerous. Many individuals who should be seeing a physical therapist for their neuromuscular issues are instead attempting to heal themselves using information they learned in a WebMD article or YouTube video. Plus, there are countless new apps and exercise programs claiming to solve for common diagnoses we address—think low back, knee, and shoulder pain. If we, as therapists—as highly trained, skilled, and licensed neuromuscular experts—buy into the belief that a tool can make better decisions and take better care of our patients than we can, then we’re perpetuating the commoditization of our profession. We’re running ourselves out of a job—out of our rightful place in the care delivery process. And that’s a darn shame, because there’s no doubt in my mind that without PTs, there would be a lot more patients falling victim to further functional decline.

Now, let me be clear: I’m not suggesting that we boycott clinical decision support (CDS) of all kinds; I’m actually a huge advocate for using technology to augment and improve efficiency. In many cases, this type of technology can actually enhance treatment and plan of care development—because it can suggest appropriate tests and measures, identify red flags, and propose potential treatment routes, among other things. But there’s a big difference between using technology to help you be a better therapist and depending on it to do your job for you. We simply can’t afford to outsource our expertise as skilled healthcare providers to an app—for many reasons, including the safety of our patients and the fact that the strength and accuracy of our data is on the line. Doing so could seriously hinder our industry’s ability to take our rightful place as valued members of collaborative care teams. And to circle back to the discussion of the new evaluation codes, as soon as payers adopt a differential payment system—assuming, that is, that they’re able to amass the data necessary to actually set accurate tiered payment rates—cookie-cutter, canned responses will be nothing more than a giant red flag for a very happy auditor.

There’s plenty to be said for convenience—for making things easier and more efficient. But the easy road isn’t always the right one, and that’s definitely true when it comes to clinical judgment. So, no matter what regulatory changes come next, I urge you to think twice about implementing fully automated compliance solutions. After all, many of these new regulations are designed to further the goal of creating a more value-based healthcare system—one where PTs have a unique opportunity to shine. Automating ourselves out of the equation now would be a very bad move—especially when we, as rehabilitation experts, are crucial for continued innovation in healthcare delivery.

Instead, stick with—or adopt—technology that uses CDS to improve upon your own expertise, knowledge, and skill. In WebPT, you’ll notice all kinds of features and tools that enhance—as opposed to replace—the work that we, as rehab therapists, do, and that’s intentional. We’ve purposely chosen to focus on improving documentation efficiency to reduce the time therapists must dedicate to creating notes. That way, you can spend more time doing what you do best—what do you better than any algorithm ever could—treating your patients.

What do you think about the slippery slope of clinical support software? Tell me your thoughts in the comment section below.