The 51/49 Rule
Alex Bendersky on biomarkers, therapeutic alliance theater, and why PT's focus on making patients feel good is killing the profession's claim to expertise.
Alex Bendersky believes physical therapy has become too comfortable with subjective validation and not serious enough about objective measurement. He spent years running a clinical fellowship program and building digital health tools at SPRY, which gave him a front-row view of how reimbursement constraints and therapeutic alliance rhetoric have replaced measurable function as the profession's north star. This conversation is about what PTs actually own, what they pretend to own, and whether the industry can build expertise worth paying for.
ben.barron: I want to ask you about what is the difference between physical function and functional performance. Is there a difference? Is it semantics? Is it functional goals based? Is functional a quality? Those two words always get thrown together?
alex.bendersky: Yeah, so I’m glad you asked. And I’m pulling that mostly from my research background, but what functional performance really references to is the execution of a certain function in a certain way. That is measurable and that could be tested. So the execution of a gait, execution of a transfer. These are functions, these are measurable functions where you can look at temporospatial characteristics of gait, you can look at dual tasking. Can you walk and talk? Can you perform a cognitive function along with a functional mobility function? And those are ways that you would assess and measure functional performance. Physical function is a lot more generic and more of a general term referencing somebody’s ability to perform movement without necessarily specifying what the function is. It’s a much more broad term. And the difference is that if you really want to harp on what somebody is capable to do and their deficiencies and how to quantify those deficiencies and how to be able to standardize those, you have to specify the function. Otherwise, functional performance could also refer to the ability to operate a computer, right, or the ability to write, the ability to cook. Whereas physical function just referencing to all of your movements related disabilities or limitations.
ben.barron: Got it. So it’s more of a qualitative measurement versus a quantitative measurement?
alex.bendersky: I think functional performance is quantitative, physical function is more generic.
ben.barron: Got it, got it. OK. Then I want to actually back up to something that we mentioned earlier and I think it’s probably worth us flushing out a little bit here, and that is kind of you mentioned diagnosis specific care. I tend to agree with you that where things get hairy is when you’re working with people that have more complex comorbidity or who have long-term chronic conditions, both psychological and physical, that probably don’t fit into the model that you’ve mentioned. Does the model that you mentioned work for that population? Is there a different approach that should be applied to that population? How do we reconcile that care for people that might have comorbidities, whether that’s hypertension, diabetes, psychological impairments, mental health related conditions, how do we serve both of those patients?
alex.bendersky: So it’s a tough question because I don’t think there is a short answer. And I don’t think we’re doing that well, overall in the industry, to serve complex patients. Because much of our profession is predicated on acute care, where the physical therapist is able to produce a trajectory of improvement. Whereas in the chronic care, maybe the physical therapist is trying to mitigate the trajectory of decline. And what adds to it is that a lot of us are not trained in population health management. We’re trained in one-on-one, individualized plans of care. So the first one is to answer the first question of what we should be doing is that we can’t use standardized protocols for everyone. I think that’s what that meant. When we create a protocol, there is a specific type of patient and a specific type of intervention that would serve that patient. But once you move into complex patients, it becomes very complicated because you can have a standardized plan of care or you can have a standardized protocols, but then you have to be able to layer on more context, more variables. So if I was to design a protocol for patients with chronic pain and elevated BMI and diabetes and multiple comorbidities and socioeconomic factors and access to care factors, I can still create a standardized plan of care that will guide me through. But then I need to be a better clinician because I need to be able to alter that plan of care based on the volatility, based on all of the other variables that are confounding to the anticipated trajectory of what I’m trying to accomplish. So I think instead of saying that there is a best model of care for that, it’s more of a I don’t think we’re doing enough. I don’t think we’re creating enough rigor, enough sophistication in a clinician’s ability to address those. Some of the higher quality residencies and some higher quality fellowships and some higher quality learning organizations try to do that. I think you may have heard Dr. Andrew Rothschild. He’s trying to do that with his residency at UCHealth. I think people that come out of that residency are probably better prepared to address complex patients. But by and large, we have been relegated to the role of treating acute injuries and maybe helping somebody post-op. That’s a lot of musculoskeletal, orthopedic world. Whereas chronic care is something that I don’t think we own yet. I don’t think we’re that good yet. And I also think that a lot of folks in that complex realm are the ones that would benefit from a direct primary care model or from a patient-centered medical home model where you have a multidisciplinary team treating that patient and a physical therapist is a point of contact on that team.
ben.barron: That makes a lot of sense. I think that answers my question in large part because it sounds like what you’re saying is the model of care that we’re talking about isn’t where we’re failing. Where we’re failing is our ability as an industry to scale the creation of clinicians that have the skill sets in order to be successful in caring for those patients because that requires a very robust form of education, whether that’s through a fellowship or a residency. Safe to say?
alex.bendersky: Yeah, absolutely. And I also, I’m always careful with how I say that because then the second thing is what are we incentivizing? Are we incentivizing volume? Are we incentivizing quality? Are we incentivizing expertise? And more times than not, the business structure incentivizes volume. The organizational structure, the employment structure incentivizes volume. And so what you get is the higher volume of patients seen with less expertise, less specialized care and then less refined care. And so that is a bigger thing. Therein lies rub, right? If we can start defining and reimbursing some of those things that are hard to define, like a threshold of expertise, and how much value there is to that. I think we also incentivize the things we’ve created, which is, like you said, that 45 minute visit, which is a made up number. Somebody’s getting 45 minutes, somebody’s getting 30 minutes, somebody’s getting an hour. Why is that the number? Where did that number come from? Was there any research behind that? I don’t have any reason to believe that 45 minutes is better or worse. It’s just the number that we’ve been able to produce the most benefit within the reimbursement environment.
ben.barron: To your point, it’s a, it’s, it’s a reimbursement driven number and not based upon any sort of clinical diagnosis or impairment or what’s likely to get us to the shared goals that patient and clinician agree to. Those are very different things.
alex.bendersky: Correct.
ben.barron: Yeah, right. Walk me through how you would demonstrate it because I’d imagine you’re talking about some form of outcome measurement in order to do that. The thing that I think is somewhat in conflict with everything we just talked about is the idea that the patient is actually qualified to determine for themselves what health care services they need based upon very limited information and often comes with a whole litany of biases, previous experiences, external influences, whether that’s family, friends, things they saw on social media and in many occasions don’t really know what they don’t know. And as a clinician, I have more information about what is likely to work versus what is not likely to work. And so to hand all of the agency to the patient to say, where do you want to be and what do you want to be able to do? While I understand that that can provide appropriate goals and measurements of success for the patient. It also seems to relieve us of a lot of responsibility to guide that patient towards what would be the right thing to do. That feels like it has an inherent amount of conflict.
alex.bendersky: I think this is a good conflict and I think this is a great question. One, what I referenced was shared decision making and shared goal setting. So it’s not just the patient and it’s not just a clinician. I don’t think that a patient is always right. I also don’t think that a clinician is always right. So what shared decision making tries to accomplish is it’s an ethical framework of autonomy. And so autonomy in this context is that somebody who has some understanding of what they’re getting can consent to what they want. And so what we really do in shared decision making is we’re giving the patient all options of care. Risks and benefits of those options of care and ultimately letting them decide. So in this model, the clinician still has and retains the agency to speak to the expertise.
ben.barron: I think one of the things that I hear you saying here is very relevant to the conversation which I’ve had a lot, which is just being okay with not knowing everything and being comfortable in saying that within the boundaries of what I know, here’s what I can help you with. But that doesn’t mean it’s the totality of the resources that are available to you.
alex.bendersky: And that all comes back to that objectivity of we own the physical function, we own the measurement of it, we own the interpretation of it and we should be able to then educate patients around what that means. So if you tell me as a patient that I want to go back to playing tennis and I test you and I figure out that you don’t have the shoulder range of motion and you don’t have the grip strength and you don’t have the rotational capacity through your core to effectively play tennis without risk, I can say I can help you improve all three. But in order for you to achieve that, you may also need something else. You may need to lose weight, you may need to address your diabetes. Go do those other things. And then what I can provide will be more applicable and more effective. And I think that is where physical therapists are expert. We’re expert in the assessment of human function. And if we can just maintain that rigor, no one can touch us. Chiropractors, athletic trainers, personal trainers. Nobody can touch us when we know that stuff. But we diminish ourselves when we talk about things we are not trained in. We talk about nutrition without having the level of sophistication and training of a dietitian. We talk about pharmacological pain management without understanding the complexity of rebound hyperalgesia. We talk about multimodal treatments, but don’t understand the full context. And then we lose our credibility because we’re sounding like we’re subject matter experts in things that we have no training to be subject matter experts in.
ben.barron: Yeah, I think everything you’re saying, I couldn’t agree with more. I think we as a profession have been far too confident in things we’re not qualified to be confident in and then not nearly confident enough in the things we are. And I think that if we were a lot more rigid in the things that are within our domain and understood our boundaries and were willing to partner with everybody else while being confident that what we bring to the table as it relates to the assessment measurement, improvement of human function is ours to own and we don’t need to have our fingers in a bunch of pies in order to have a seat at the table. I think we’d be a lot better off and we probably see a lot more utilization of the services that we provide in an appropriate matter. That said, let’s get back to something else you mentioned, which is we need to measure the thing that we’re trying to improve. Because I want to get your kind of opinions on what I think the crux of what a lot of the, you can call it arguments or disagreements within the profession are right now, which is what’s the right thing to measure? How do we create a standardized measurement of the thing so that we all know that we’re measuring it in the same way and therefore the outcomes and comparisons are appropriate. Walk me through kind of those pieces. What should we be measuring? How should we be measuring it? How do we know we’re measuring the thing that’s actually important?
alex.bendersky: So this comes down to a little bit of nuance of what you’re trying to measure. Are you trying to measure satisfaction, you’re trying to measure quality or you’re trying to measure efficacy? And there’s pros and cons to all of them. My bias is that you measure efficacy, meaning you measure whether what you’re doing works. And if you measure efficacy, then you figure out what variables you’re looking at. And so most recently, the field of PT has moved a little bit towards this biomarkers literature. So there’s strength, endurance, flexibility, all these things. There’s a movement towards looking at impairment level measurement tools that look at functional characteristics and that look at specific deficits. So if I’m testing your shoulder, I’m not looking at just range of motion, I’m looking at the rate of force development in your shoulder external rotation. I’m looking at the mean max force. I’m looking at the time to peak torque. I’m looking at a battery of tests that tell me what your shoulder is capable of at a very, very, very refined biomarker level. And then from that refined biomarker level, I can also maybe test your functional abilities. So can you take something off the top shelf? Can you reach behind you? Can you sleep on your side? Can you do a pull up, push up? But I need to be able to get very specified biomarkers. And so if you look at a lot of the emerging literature and emerging technology in our space, that’s what’s moving towards. There’s a lot of technology that does a lot of interesting things, right? There’s force plates to measure your jumping and landing. There’s grip test to measure your grip. There’s a cardio pulmonary exercise testing to measure your ability to produce energy. There’s one rep max tests. There’s rate of force development. There’s a whole bunch of things. And I think we should be advocating for all of those things to be measured because it has always frustrated me that we have a profession that tells you how to perform based on your subjective interpretation of how well you performed. So you’re as good as you think you are. But we don’t tell you, but you’re 50% weaker than you were before the injury. Like, nobody tells you that. They’re like, hey, how’s your shoulder? I feel pretty good. I’m about 75, 80% back. But when you test them, they’re like 50% down. This is a major problem in concussion research and concussion rehabilitation. So an athlete with a concussion can present as being completely asymptomatic because they feel fine. But when you measure their dual task gait, when you measure their ability to walk and perform a cognitive function along with gait, that becomes very impaired because they’re symptomatic. So we’re not measuring their symptoms, we’re measuring their function and their function is impaired.
ben.barron: So if I’m hearing you correctly, the reason we need to focus on measuring at a biomarker level is because what the patient is feeling and what’s actually happening are not the same. And that what we’re ultimately trying to do is improve the underlying thing to the point where it gets back to normal. And by measuring at that level versus measuring whether or not they feel better about it, we’re able to identify if we’re making actual change, which may over time then result in the patient feeling better. But the patient feeling better isn’t a proxy for actual improvement.
alex.bendersky: Correct. I mean, so to give you a less nuanced story and a simple example, if you are an unhappy person and you’re unhappy because you feel unloved. Let’s say you didn’t have a great childhood and you have a little bit of insecurity and you’re not feeling loved and you want to feel loved. And you go to a therapist and that therapist doesn’t give you anything other than makes you feel loved. So you come back for that therapeutic alliance and for that validation and that recognition and support, you will feel happy in that moment. But if therapist doesn’t address the underlying cause, which is your insecurity, your inability to love yourself, your feelings of loneliness or isolation or what is actually happening underneath the hood, they’re making you feel happy, but they’re not giving you skills to address it and they’re not addressing why you’re feeling that way. And so similarly with physical therapy, if you’re coming to us and you feel pain and I only am looking at your pain level, I’m asking you does it hurt? And you’re saying yes. I’m asking you does it still hurt? You say no, but I didn’t do anything to address what drove that pain. So once you do that thing again, you’re going to experience that pain again. And then you’re back in my clinic and I’m making you feel good again. But if I can test you and if I can test you and say you don’t have enough core stability, you don’t have enough shoulder range of motion and here’s your objective deficits, now I can work with you to address those deficits and whether you feel good or don’t feel good is less important than the fact that I am seeing changes in your ability to perform. Because eventually once you can perform, you will start feeling good. So that’s why I think that we need to move towards measuring biomarkers. And this is what frustrates me the most about our profession is that everyone has access to a goniometer and a stopwatch and a grip and manual muscle test. Everyone. Not everyone uses them. And then you look at some organizations, which I call them performative organizations. They market themselves as performative. I’m not going to give you a specific name, but they make millions of dollars selling courses around psychosocial pain neuroscience, biopsychosocial models of caring, but fail to show any objective measurement of changing function. And so I get frustrated by that because it’s like, wait, we are here to measure human movement and we’re here to be able to determine what somebody is capable to do. And if we don’t do that well, then I don’t know what we’re doing.
alex.bendersky: And so if all you’re doing is you’re acting as an emotional support animal without addressing anything that’s measurable, you’re not a physical therapist. You’re providing value, but it’s not clinical value. And so I think that’s why it’s important for us to start looking at measurable things. And that’s why I focus on this objectivity piece so much is that let’s be the expert in things that we are experts in. Let’s acknowledge that we’re not experts in other things. And then let’s get better in the things that we should be expert in, which is measuring and modifying function.
ben.barron: So I have two places that I’d still like to go and one I think is probably the easiest place to go here. I think everything you’re saying makes a lot of sense to me. I feel like we haven’t yet discussed how do the psychosocial components of disability relate because those can’t be measured necessarily with the one Rep Max test. They have much more to do with all the things that we mentioned in terms of access and context and might have more to do with social determinants of health and all those other factors. How do you marry everything we just talked about from, and I know you said they weren’t necessary, but how do you marry force plates and biomarkers and grip tests with the psychosocial aspects of care?
alex.bendersky: So I can give you a popular answer. I can give you an answer that may anger people. Which one do you want?
ben.barron: I mean, I’d rather know just what you think, right? I mean, you tell me.
alex.bendersky: I think the popular answer is that of course it matters. The therapeutic alliance matters and the psychosocial parts matter and understanding the patient matters. And then my own personal belief is that it matters, but not a lot because it matters to a point. But if you notice that some of the most business successful clinicians are the ones that are social and personable and extroverted and know how to build relationships. Not necessarily the ones that understand how to graduate and progress patients based on empirical evidence. The ones that understand the empiricist also are these introverts who don’t understand how to relate to the patient. So it matters. Psychosocial matters because we are consuming a service and anything that’s service driven, anything that’s not objective, you have to align with the patient. But that alignment needs to be grounded in some source of truth. And that point of resource of truth needs to be objective where somebody’s care seeking behavior and care giving behavior, right? So as a provider you have care giving behavior. As a patient you have care seeking behavior can be matched through an aligned shared goal setting. Which is what the psychosocial part should do, where that’s your part where you align with what your expectations are and you align with what your deliverables are as a provider. And then I think that will give you a much better product output out there if you’re capable to do it. Don’t know if I answer that.
ben.barron: No, I think I understand your position, which is you have to go about the psychosocial aspects of things objectively as well. And part of that, the big thing I learned from you there is that the identification of shared goals, alignment on a path to pursue those goals is an objective way to or a way to bring objectivity to something that can be nebulous otherwise. I can get behind that. I think that makes sense.
alex.bendersky: Yeah, for sure. It’s not taken away. It’s not just mechanizing and just completely eliminating any point of relatedness, but it’s also not over relying on it. And I think as an industry we are over reliant on this art of caring. We’re forgetting the science of giving. We’re only focusing on the art. If you need a less anecdote, there’s a book that I really like. It’s called the Bloom Shots. And so he looks into successful business enterprise. And I like his analogy that any successful business enterprise needs an equal amount of soldiers and artists. You will not build a successful business if you only have soldiers because they’re just enforcing and they’re not creating. But you’re also not going to build a successful business if you only have creatives. Creatives are going to continue to create but not really deliver. So you need equal amount of soldiers and artists. And the way it relates to physical therapists is that the art of caring and the science of giving or whatever that APTA model is, you have to be able to align. You have to be personable. But the formula I used to give my residents was that you have to be 51% substance even if you’re 49% ********. Substantiate your care to a fractionally higher degree than all of the other non objective delivery process. Make sure that the ratio of objectivity to non objectivity is 51/49 and it’ll be fine because then everything else is the context. But if that formula tilts the other way and you’re 51% ******** and 49% context, 49% objectivity, it’s fluff. And that’s what we have a lot of times is we have fluff. We have over dignified fluff in the clinical setting because we are able to show that patients like us and patients feel like they’re getting something of value and that patients get better and of course patients get better. There’s a regression to the mean. There’s just a natural life to a lot of these musculoskeletal disorders that tend to be ongoing and tend to be persistent and tend to be chronic and we just, if you happen to catch a patient at the right time, you’re going to be absolutely adorned by them. If you happen to capture a patient at the wrong time, they’re going to say physical therapy didn’t work for me. The key is to identify the things that are under your control. One Rep Max, speed of gait, physiological variables related to a human condition. Objectify it and then do all the other things that you’re doing.
ben.barron: So the last thing I want to talk about I think is related to all this and it has to do with, a lot of conversations that I’ve had on this in this manner with other people similar to yourself. I think what we qualify, what we’ve historically classified as a typical care model was created because that’s what we could get paid for and we ended up at two to three times a week for 45 to, 40 to 60 minutes for six to eight weeks.
alex.bendersky: Mhm.
ben.barron: It probably has some loose correlations to what good care may look like, but it is also very tightly aligned to CPT codes, descriptions that we get reimbursed for and authorizations of care that get handed down from insurance companies. Safe to say that’s not. I didn’t say anything there you wildly disagree with.
alex.bendersky: No, I mean no, because you can go two ways also then, right? Even if you discuss that, you have these standards of practice can drive refinement like in the airplane industry. Airplanes just don’t crash anymore because the standards of practice created safer airplanes. So payers creating constraints like prioritization, is it can be ineffective sludge. It can be an effective friction point. Standards of practice can also deliver our expectation to over deliver care because that three times a week, that dose response or three times a week for six weeks is a non objective variable. Three times a week for six weeks may be appropriate for some, not appropriate for others. How did you derive to that benchmark of performance? How did you derive to that dosage of clinical care? What makes you say that that’s optimal or suboptimal and if you can justify it?
ben.barron: Yeah, I fully agree with you and I think that’s one of the things that I’ve always had an issue with is that we default to these things because it’s what our business’s operational model is based upon or it’s what’s written on a physician prescription or it’s what I did with the last 13 patients and therefore the 14th patient needs the same thing. How, you mentioned a lot as we’ve talked for the last hour around utilization of technology and digital tools in order to drive care. If you are back in your teaching days of fellowship again, how would you be working with clinicians to identify the right model of care for the patient in front of them, given all the options that are available? Fully in care, in clinic, synchronous, one-on-one, all those sorts of things to go do this on your own. You never need to come back to me. What do you think the paradigm is there as we’re figuring out how do we determine the frequency, duration and length of care, whether that care happens in clinic or happens in a hybrid manner, or it happens purely via telehealth, or maybe it doesn’t. Maybe it’s just leaving patients on their own. What do you think the decision-making tree needs to look like there?
alex.bendersky: So availability of resources and availability of finances are our biggest determinant of that. I think designing a risk stratification tool for a legacy provider for a patient population that was migratory. We looked into the access to variables like whether someone has access to viable in person care and whether someone has enough funds to support that and then we built that into the other stratification tools which then looked into the demographics and into the severity into the comorbidities and everything else. And then you create a little bit more complex formula, the complex adaptive formula based on those variables and see what resources you have available to you. If I could do things differently now, I would reject a lot more patients because I think over delivery of care creates iatrogenesis, but I will also empower a lot of patients. Iatrogenesis. It’s an iatrogenic effect. It’s like we are actually making people sick sometimes because we’re over delivering care. But that’s like you’re indoctrinated into that three times a week for six weeks formula.
ben.barron: It creates what? I’m sorry.
alex.bendersky: And you’re distilling a belief that that is the only way to consume care, which then creates a negative loop. So to make the long story short, I think I would be comfortable turning people away, but also be comfortable having this discussion with people. That you have option A, B and C. Option A is I give you a viable program and it’s gonna be on you to enforce it. Option B would be for you to use me as a reference point, and option C would be for you to come in three times a week for six weeks. And then it becomes more of a consumer driven behavior, right? Consumer can make a choice whether to do something on their own, to do something under loose supervision or to do something under direct supervision.
ben.barron: I like it. I think it’s probably a good place for me to shut up. Is there anything that you feel like we haven’t touched on or anything you want to make sure we talk about or we get out there on the World Wide Web in order to help you out or plug anything you want to or anything like that.
alex.bendersky: No, dude, I mean I appreciate what you guys are doing. I think that you’re fulfilling a very important part of delivery process. I think more clinicians need to use patient reported outcomes. More clinicians need to understand how to interpret patient reported outcomes. There’s PROMs and PROMs, right? The M is the important part. More clinicians need to look into that and just it’s we’re all part of the same industry dude. And I think if we can just get away from labeling one organization or the other and try to build this industry up, I do have a high level of optimism into what we can be if we only do this thing right.
ben.barron: Great. Well, look, I really appreciate the time, Alex. This has been great just to get to know you and hear your perspective. I think you have a very unique take in the private practice segment of the market. At times it does feel like everybody’s saying the same thing all the time, and I just appreciate people that have differing opinions than maybe what everybody else is saying all the time, and they’re happy to have a conversation about it. And what I think is the most refreshing is that anyone who follows you on LinkedIn or has a conversation with you knows that none of the opinions that you have have been pulled out of thin air and haven’t been well thought out, haven’t been well researched and I think that’s unique and refreshing. So I just appreciate that and I appreciate your time today.
alex.bendersky: Thanks man, I appreciate you and thank you for this.
Ben Barron | SVP @ Net Health | LinkedIn
Alex Bendersky | Head of Clinical Innovation @ SPRY | LinkedIn
Fantastic articulation of the objectivity problem. The 51/49 formula cuts right through all the handwaving about "holistic care" that often just masks not measuring anything. I ran into this exact issue workingwith clinics where patient satisfaction was high but functional outcomes weren't being tracked, which made it impossible to seperate placebo from actual intervention.