Ep 221: Does it matter how you apply spinal manipulative therapy? With Dr Casper Nim

[00:00:01] Hey there, it's Claire here. I wanted to let you know about a new open access journal. It's called JOSPT Methods. Here you'll find a place that's dedicated to publishing the best new approaches to producing the highest quality evidence that supports your clinical practice in musculoskeletal rehabilitation.

[00:00:19] JOSPT Methods is published by Movement Science Media, the publisher behind the journals JOSPT, JOSPT Open and JOSPT Cases, and this podcast, JOSPT Insights. Head to jospt.org forward slash josptmethods, that's methods with an S, to read all the latest articles with no paywall and to submit your research. And while you're there, check out the front cover of the journal. It's a real beauty.

[00:00:51] Hello, and welcome to JOSPT Insights, the podcast that aims to help you translate quality research to quality practice. I'm Claire Ardern, the editor in chief of the Journal of Orthopedic and Sports Physical Therapy. It's great to have you listening today. There's a range of different interventions at your disposal when managing spine pain. Today, we're focusing on spinal manipulative therapy and asking questions like,

[00:01:19] Does it matter how you apply a manipulation to get the best outcomes for the patient? Answering the questions is Dr. Kasper Nem from the University of Southern Denmark. Kasper's an associate professor, chiropractor and senior researcher at the Spine Center of Southern Denmark. He leads a team that's working hard to bring the best quality research methods to studying manual and manipulative therapy approaches in musculoskeletal rehabilitation.

[00:01:46] Today's discussion touches on the results of a network meta-analysis that Kasper and his team published in JOSPT in February 2025. And as always, you can find links to all the resources we discuss in the show notes. Okay, over to Kasper. Dr. Kasper Nem, welcome to JOSPT Insights. Thank you so much, Claire. Kasper, today we're talking about approaches to treating managing spine pain.

[00:02:12] You're particularly interested in spinal manipulative therapy or SMT. I think it's helpful if we start by defining what actually is spinal manipulative therapy in your view. Across different research papers, you might find different definitions. So, spinal manipulative therapy or if we should just call it SMT because that's, you know, easier. It can consist of many different aspects.

[00:02:35] They all have in common that it's a manual therapy where you deliver some kind of intervention with your hand on instrument to the, well, in this case, spine of your, you know, patients or participants or whatever way you want to define it. For me specifically, and for the paper that we're going to talk about today, we defined it as a high velocity, low amplitude manipulation, which basically means it's the cracking treatment that we're looking at.

[00:03:04] Not just general mobilization, made in mobilization, what have you, which also sometimes can be defined as SMT. There's also a thrust component to this. So thrust versus no thrust. Can you tell us a little bit about that component of SMT too, Kasper? That's the exact point with the cracking noise.

[00:03:22] That's the intent or the goal with having a thrust that you deliver normally with a high amplitude that people would get that initial popping sound of their back or joints or wherever you do it. Whereas a non-thrust mobilization typically comes in oscillations or it's repeated in some other way. Now, where does SMT fit in the context of other interventions?

[00:03:47] I'm thinking of things like pain neuroscience education, therapeutic exercise, maybe mind-body exercise like yoga or tai chi. Specifically, when we're talking about the context of managing back pain or spine pain. They fit in nicely together because they all have, if we're being honest, small to moderate effects when we look at it on a bigger picture. And SMT is no different there. It's very real research, just like many of the other things. Well, there's a lot of, at least a lot of research on it.

[00:04:16] But it fits in the context that it's pretty much, the effect sizes are pretty much comparable whenever you want to manage spinal pain. I like to call it that instead of treating it. So the average estimates is basically comparable to the recommended treatments that you mentioned here, exercise, education, etc. Now, where it might differ is that normally with SMT, we have a tendency that we want to be very specific. It's not like a general exercise, which is also the basis for our review here.

[00:04:45] So some of the things that you want to do with SMT is that it has to be delivered at the correct site, wherever you indicate that there's some kind of dysfunction in the spine that needs to be corrected almost. And that is where the thrust comes in, that little correction. And then at the end, you know, you want to, of course, optimize this function that in turn will optimize patients' functions and decrease their pain with other types of manual therapies. It may be more generalized if there are non-thrust therapies.

[00:05:16] Exercise is also a bit more, at least it can be more generalized. They can also be very specific and have their own ideas. But this idea about the dysfunctional segment and then you provide the very specific thrust to it to actually improve patients is, I would say, one of the things that separates it, especially from Tai Chi and some of those more generalizable exercises or treatments.

[00:05:40] Casper, given we're talking about managing spine pain, I think it's worth spending a little bit of time on this distinction that people will read about in particularly when they're reading research on different interventions between acute and chronic spine pain. Can you talk with us a little bit about that distinction, what it means for interventions and what it means for understanding the research field more broadly?

[00:06:04] Casper, first of all, trying to split people up into whether they have acute back pain or chronic back pain, it doesn't really make sense anymore. It did not that long ago, 2015, 20 years ago. But now we have an understanding that back pain is often fluctuating and recurrent. And instead, we should probably talk about managing, you know, the differences between managing a persistent pain load and then spikes or acute episodes of pain happening in some kind of episodic or fluctuating pain experience.

[00:06:34] Which is probably the two, if you really want to dichotomize, probably the two distinctions that you can make. And here, I think it makes a lot of difference. Although in research, it's actually quite difficult because most of the studies are actually looking at chronic, meaning long-lasting pain conditions and acute pain. And specifically for SMT, which is the field that I know most about, it hasn't really been researched that much.

[00:07:01] Like compared to placebo, for instance, there's only four trials out there. And the effects were small, I believe, 12 points or something like that. It just reached statistical significance. You could have, of course, argue clinical relevance. And that's very interesting because I think if you go into clinical practice, most people will say, oh, you use this type of treatment for an acute flare-up or an acute back pain situation or something like that. But most of the research indicates that you actually use it for these chronic and persistent pain syndromes.

[00:07:30] And even there, it works just as well or just as poorly as other interventions, such as exercise, which is, I guess, the other thing that you would probably normally recommend for more long-lasting pain conditions. And it all, I think, boils down to that we don't really have a good understanding of, you know, patients' trajectories of pain. So let me build on that point there about better understanding the interventions and understanding for whom they're appropriate.

[00:07:59] And I guess I'm interested in why you thought it was important to study SMT broadly and then to look at the different elements of applying SMT and how it's used as an intervention. If I'm being honest, it came back when I was, you know, still at university and I'm a chiropractor today. I was, you know, training to be a chiropractor in what I believe is a very modern chiropractic education system we have in Denmark.

[00:08:26] And even there, the fact about, you know, finding this function and providing a very specific thrust. And we even had these mechanical models of the bones moving in different directions when you did different vectors and stuff like that on your thrust. It never really got to me. At least I didn't really believe in it. And I had difficulty, I had that difficulty experience in it in practice. Or even when we were practicing on a student, right, I never really got this sense of that.

[00:08:52] So I was always quite, I would say, critical of the term and, but also very curious because I thought it was fascinating, right? You spoke to these old clinicians and they could feel whatever, right? And they could do something and then feel something completely different. So I was lucky enough that I was able to do my PhD in it where we did a randomized trial. I don't want to bore you with that. It's always been published, but we did a randomized trial looking at different target sites and then did a review and turned out that nothing really happened.

[00:09:19] And so we were kind of limited in the prior work that because we could only look at direct comparisons and there are not too many studies that assesses, you know, one type of SMT versus another type of SMT, which is why we, you know, we dove into this world of network meta-analysis. But you could also obtain indirect effects, but we could look at this more broadly. Specifically, we could look at the target. How was the target selected? Where was the thrust delivered? And even in what region it was delivered?

[00:09:49] Because there were some really good, some really good studies coming out of Brazil showing that, you know, even a specific lumbar manipulation at a very specific site in a low back pain population compared to just a completely generalized thrust at T3, 4 levels, which is in the complete opposite end of the spine. Had the exact same effect, even over a long time. I want to take a side tangent just briefly before we jump into the key results.

[00:10:16] The difference between a head-to-head comparison or a traditional meta-analysis and a network meta-analysis and why network meta-analysis was the right thing for you and the research group to choose. So in a normal head-to-head systematic review and meta-analysis, if you want to call it that, you look at the different studies and how there's one comparison compared to another comparison. So that could be SMT versus placebo, for instance.

[00:10:42] And then you pool, you basically combine the different studies and your effect sizes, and then you say, okay, we end up with an effect size of, let's say, 1 out of 10 points or something like that. Now, in a network meta-analysis, we can, if we imagine, and I'll do this the most simple, and it's not truly correct, but the most simple way possible to explain it.

[00:11:03] Imagine we had two studies, one that used SMT type A, and that was compared to placebo, and then another study that used SMT type B, and that was compared to placebo. So if you say type A had an effect size of 1 and type B had an effect size of about half, because we can combine it, they're looking at the same comparator, we can mathematically estimate that the differences between A and B will be approximately half a point.

[00:11:27] Now, again, if there's a biostatistician out there, this is not entirely correct, but for the general understanding, that's what we mean by indirect estimates. So it's not studies directly comparing each other, but indirectly comparing each other, and that way we can estimate differences between all these different types of SMT procedures that we classified in our review.

[00:11:47] We included 161 RCTs looking at 12,000 people, and I think in those 161, there was about almost 200 SMT arms, and we took each of those arms and we declassified them based on, you know, how was the target site determined? How was the thrust delivered? And in what region was it delivered in? And then we did this magical mathematics or statistics in the network meta-analysis and came out with the findings that we did.

[00:12:16] But we had, I would say you can categorize that we had two main results. And the first one was that our review here echoed most of the other systematic reviews. Again, finding small to moderate effect sizes that was comparable to other recommended interventions. You know, slightly better than non-recommended interventions such as traction, slightly better than placebo interventions, and moderately better than doing nothing at all.

[00:12:41] So this is what everyone else also has found, which brings some validity to our findings. But then the other big thing we found was that SMT might work, but the way that it's actually being used, the way that the target is being determined, the way that you deliver the thrust, or even in what region you delivered it. But it didn't, on average, at least across into our analysis, it didn't make any differences at all. So that was the overall result that we had.

[00:13:10] And for clinicians, I mean, first of all, this challenges beliefs in some professions and some uses of manual therapy that has gone back, you know, more than maybe a hundred years about this entire, you know, finding the dysfunction and then treating it right there. And then, you know, fixing it and that will improve the patient.

[00:13:32] And here we probably indicate more that although patients do improve or participants do improve, but the way that it works is probably not because of this dysfunctional segment that you deliver a very precise thrust into. It probably has a more widespread effect. It's probably also a very, very likely to have a lot, you know, a contextual effect around it. It's a very powerful treatment that way.

[00:13:56] You know, you lay people down, you put your hands on them, you touch them, and you give them a physical treatment that they can, you know, feel and even hear. So I think there's a lot of things there that we don't understand yet, but it, you know, brings us into what are some of the effects that we see here. When we talk about clinicians, I like to put it into two perspectives.

[00:14:17] And one is you have young clinicians and inexperienced clinicians, and maybe even students like myself, who might also be listening to this podcast, who are struggling a little bit with understanding all of these different aspects that they are being taught in universities or schools or even courses. You know, that's coming from this highly theoretical standpoint and not really based on any literature. Just like, this is what we have always done is in a book somewhere. I would say our results, they challenge that.

[00:14:46] And hopefully, at least the idea, we think this is very positive because we actually allow for those young, no-wise clinicians or students to be confident in what they're doing. Because what they're doing is probably really good, you know, for the right patient, of course, but it's probably good. And even though that they don't really, you know, might be able to feel exactly what, you know, what the older guy down the street feels, you know, he can, that young clinician can pretty much expect to get the same results just simply based on the intervention perspective.

[00:15:15] So that's one thing that we really like to prioritize with this is that, you know, we want to give clinicians, you know, the power to be confident when delivering this and not worrying about all the other theoretical aspects that are so difficult to learn because in reality, you probably never learned them. You just, you know, you make your own imagination about that you can feel something or change something.

[00:15:38] So I think that's where these results really come into education and young clinicians and experienced clinicians who have used SMT forever and they use a very specific approach. Well, I mean, they can continue to do so. It's not like that's much worse than or not worse at all, maybe than being, you know, completely generalizable with it. So they can just continue to do what they're doing. There is one thing, though, that we didn't look at in this review that I would just highlight, which is, of course, safety and also comfort.

[00:16:06] That's something we now change the topic into, being safe and, you know, making the intervention as comfortable as possible for not only the patient, but also for the provider. You know, being specific shouldn't impact that. So, for instance, if you are worried about manipulating the cervical spine, for instance, you could just as well do the thoracic spine. And then if you really want to touch the neck, just do something else that's not so aggressive, for instance, and provide manipulation at the thoracic spine.

[00:16:34] And if there are, there could be other instances, like say you had a patient with a disc herniation. Instead of going over that very sensitive segment and saying that is exactly the thing that needs to be treated, this could also go for any nonspecific case. Just go a little bit off it, let it simmer down. And then before you start to build it up, I mean, just use your manipulation at somewhere else and more or less expect the results to be the same.

[00:16:58] If I can just add one more thing, I would also say, because I get this, you know, by people who are frustrated by the results, I often get the questions, well, you know, couldn't you just then manipulate the foot or something? And I would say it's probably still important in a one-to-one clinical encounter to actually fulfill the patient's experience, especially if they have expectations or if they have tried it earlier. If they're, you know, go ahead and do it in the lumbar spine if that's what they're used to instead of going, you know, across the spine for no apparent reason.

[00:17:26] But if you're worried there, then just go somewhere else and you can provide it with a perfect explanation that it likely has a widespread mechanistic effect rather than a segment specific. You've got this very powerful intervention. I also wonder, too, for people who have invested a lot in their education, they might have done fellowships in manual and manipulative therapy.

[00:17:48] There's also, as you're saying, a lot of value in your skill set, in your skills around diagnosing and in figuring out what else is going on. So my suggestion, respectful suggestion, is we don't throw the baby out with the bathwater and say, forget about your excellent fellowship training. It's about recognizing when is the most appropriate time to choose which intervention. And I wonder what your thoughts are there, Kasper.

[00:18:14] I think you're touching on a really important point here because that education is not wasted at all. This is just some of the classical aspects we looked at. There's something we haven't looked at yet and that's because there's no research on it. So it's difficult to look at it. But we're slowly building that up now. But other stuff like, you know, force time characteristics. How fast is the thrust? You know, how hard is it? When do you use a lot of force? When do you not use a lot of force? And I have my doubts whether that will impact clinical outcomes in the long run.

[00:18:43] But it will definitely impact patient satisfaction, likely comfort. I think as someone who, you know, been in a manual therapy program, I can say that there's a big difference between being, you know, manipulated or treated by one person compared to another who really knows what they're doing. So it's not like that we can just say, oh, technique doesn't matter anymore. It does, even on a mechanical level. That's not the way that we always think about it in this theoretical, almost ideological approach.

[00:19:12] From a clearly biomechanical aspect. There are some really important things of actually teaching this, teaching psychomotor skills, as we call it. And postgraduate education there is definitely a main key. Besides, of course, all the diagnostics and even the indications, who do you give this to? Who doesn't you give it to? Who is more likely to be a responder? You know, stuff like that we don't exactly know from the literature. But I could imagine postgrad education, you know, teachers have a clinical good idea about, you know, who should we provide this to and who shouldn't we.

[00:19:41] So, no, I completely agree with you. We shouldn't throw the technique baby out with the specific bathwater. Because some things might still be important. We just don't really have any good arguments for saying anything about it now. And here we have to go with theory, which is the big difference from now that we can throw some of the other stuff out. Because now we do actually have some evidence that can debunk the theory. So, we can de-implement stuff now. Making room for other more important stuff. I do also want to talk about the certainty in the evidence here.

[00:20:10] You talked about, I think, 160 different randomized controlled trials. 12,000 participants in those trials. And yet, the challenge with all of the research we do or the field does on these manual therapy techniques. And this is not just a problem for SMT. This is an issue for all sorts of manual therapy and exercise interventions. Frankly, a lot of the interventions we have at our fingertips as musculoskeletal rehabilitation clinicians.

[00:20:38] And that is, it's really hard to blind someone. So, how do we overcome or I guess how do you think about overcoming some of the methods challenges to really shore up that evidence base? Are we there yet? Blinding here and the same goes for exercise. It's basically impossible. And people have tried here multiple times, right? And even some studies report that people were blinded. Sometimes to a degree where I don't really, you know, trust the office.

[00:21:07] I mean, if you want to follow the standards for doing, you know, good research set out by the Cochrane Group, which is the, you know, leading experts on and making the guidelines on how to do this, we're going to run into trouble. Just like we did in our review. That was one of the main reasons why we downgraded into low certainty evidence, as it's called. Because the high risk of bias, as most of these studies were, you know, had difficulty blinding.

[00:21:33] And when you can't blind and you're comparing an active intervention that we know has some kind of effect versus doing nothing and participant has to be told about this, otherwise it would be unethical. Like that is going to introduce a major bias to the studies. And I don't think there's anything specifically we can do about that, which is probably why that a lot of studies right now are being published, which is, we call this A plus B versus A methodology, right? Where it's, you know, the interventions always wins, right?

[00:22:01] Because both groups get something and then one group gets a little bit extra. But that is the exact same issue. Getting that a little bit extra comes with a lot of contextual effects as well, more placebo effects as well. Like there's just higher odds of that group doing better. And we see that across all the trials. So we are really at a point now where we even make the case here that, you know what, stop doing more of these small study SMT versus another type of interventions or even SMTs against each other.

[00:22:29] Because the results are going to come out the same way, right? Small to moderate effect sizes. And the better you do it, like the less of a difference there is because then you control for most of these factors. But it's something that we need to understand. And we have to, you know, simply rethink the way we do trials. And there's a lot of stuff coming out. Even the OSBT published a paper the other day about something, you know, a preference randomized trial.

[00:22:52] And there's a bunch of stuff that we have to explore here because we're not going to get any further doing what we do now using the tools that we do now to critically appraise these studies. And I think this is a really nice place to acknowledge the maturing of our research field, of our musculoskeletal rehabilitation field in terms of the research. And there's, you know, lots of wonderful people who have been at the vanguard of leading that change and bringing more methodological rigor into the work that we do.

[00:23:22] And I guess the other point to make here is you mentioned that some of your comparisons were to a placebo and some were to wait and see. And that the size of the effect for SMT versus, let's say, wait and see was more than the size of the effect between, let's say, SMT and exercise therapy. When you compare to doing absolutely nothing, you will see an effect. But it's not really a fair comparison, right?

[00:23:48] Because the idea about doing nothing and then doing some kind of intervention, that intervention is linked, you know, a clinician showing empathy and, you know, a bunch of other factors surrounding that intervention that you won't get into, you know, wait and see period. Which just means that, you know, for most people that will be included in these types of trials, which is more than likely people with, you know, long lasting pain who has a spike of some kind currently in their maybe persistent, non-persistent pain state.

[00:24:18] Yes, they should probably go see someone and then whatever is being done there is probably fine. Like, the other argument you can make was that how come that the effect when compared to placebo is smaller than when compared to doing nothing? That should basically be the same thing. But again, the placebo effect accounts for maybe one third of the total intervention effects or at least one fourth. Whereas, you know, wait and see is about 50% or contextual effects is about 50%.

[00:24:45] And then we had that, you know, one third, one fourth lift actually described to the intervention specifically. For clinicians, that's important to understand, right? That, you know, touching on or enhancing these contextual effects is probably important because that's really what drives most of the improvement, based on the evidence at least. And then the interventions is an add-on that you do to get that little bit of boost, right? And then, of course, the idea as a clinician is you need to figure out who do you want to give that little bit of boost to. It's such a nuanced discussion.

[00:25:15] Kasper, I'm so grateful to have had you guide us through the nuance today. And to you and the team, a big thank you for all of the ongoing work you're doing in this field to try to help us all do better when we're working with people, supporting people who have back pain. Thanks for joining me on JOSPT Insights. You're welcome. Thank you for inviting me. Thanks for listening to this episode of JOSPT Insights.

[00:25:41] For more discussion of the issues in musculoskeletal rehabilitation that are relevant to your practice, subscribe to JOSPT Insights on Apple Podcasts, Spotify, TuneIn, Stitcher, Google or your favourite podcast app. If you like JOSPT Insights, help others find us. Tell your friends and colleagues and rate and review us. To keep up to date with all the latest JOSPT content, be sure to follow us on Twitter, we're at JOSPT and Facebook, we're JOSPT Official.

[00:26:10] Talk with you next time.