First we understand the problem. Then we determine the right course of care.
Joint and Spine Solutions is built on a different starting principle than most musculoskeletal care: stop trying to fix it before you understand it.
Most musculoskeletal care tries to make people better. We try to understand the problem first.
This sounds like word games. It isn’t.
If a clinician’s primary goal is to make the patient better, the natural pull is toward doing something — prescribing an exercise, ordering imaging, scheduling more visits, making a referral. The clinician feels productive. The patient feels heard. Both leave the visit feeling like progress was made.
But action without understanding is where musculoskeletal care gets expensive. The patient who needed a different kind of evaluation gets a prescription instead. The patient with a rapidly resolvable problem who needs one precise movement gets generic exercises that don’t address the direction the problem actually responds to. The patient whose problem would have resolved on its own gets weeks or months of treatment that takes credit for what natural history would have done — and is never taught how to keep the problem from coming back, because no one identified what actually changed it. Multiply that across a population and you get the pattern every organization recognizes: rising cost, longer episodes, and patients who cycle through care without ever getting a clear answer.
So our starting question on every visit is not “what can we do to help this person?” It is “what is actually going on here mechanically, and how is the problem responding?” That question is, frankly, harder to answer — not because it is slow, but because it requires a reliable assessment method that most clinicians were never trained in. When the answer comes quickly, as it often does, it comes with real information about what will help. When it takes a few visits to become clear, those visits are still moving toward the right answer — unlike an early image that names a finding, feels like speed, and sends the patient down a pathway that the finding may have nothing to do with.
The difference isn’t effort or good intentions. Most clinicians are genuinely trying to respond to what is in front of them. The difference is whether what is in front of them was reliably understood in the first place.
The patient’s response to specific movements is the most reliable diagnostic information available.
Most musculoskeletal assessment relies on history, palpation, and special tests. Each of these has some value. None of them, on their own, produce a reliable answer about what is driving a given patient’s problem.
Imaging is often assumed to be the gold standard. It isn’t, for this purpose. MRIs are extraordinary for surgical planning, for ruling out serious pathology, and for documenting structural change. They are not reliable guides to what is driving a particular patient’s pain or what treatment will help. Patients with severe imaging findings often have no symptoms. Patients with disabling symptoms often have unremarkable imaging. And no MRI tells us how this specific body responds to load right now.
What does tell us is the response itself. How the symptoms behave under specific loading conditions. What gets worse and what gets better. What changes location, what stays put, what spreads, what centralizes. Whether the response is fast or slow. Whether it lasts or fades.
This information is gathered through a response-based assessment process that observes the patient’s response to specific movement sequences. The assessment process is the same methodology that has been shown in published research to be reliable across trained clinicians — meaning two trained assessors examining the same patient independently reach the same classification. That kind of reliability is rare in musculoskeletal medicine, and it’s the foundation everything else rests on.
An illustration of what disciplined assessment catches:
A worker from a local construction company came in to see us recently for what he thought was an ankle problem. He had just returned from a cruise. The ankle was red, swollen, and tender. He had full motion and good strength. He was using crutches because the ankle hurt too much to walk on.
Most of what we see is musculoskeletal. This wasn’t. The pattern didn’t fit. There was pitting edema. The history didn’t include a mechanism that would explain the presentation. The findings didn’t cluster into a category we recognized.
We walked him down the hall to the physician’s assistants at his employer’s clinic. His blood pressure was extremely elevated. He was sent directly to the emergency department. He had open heart surgery. Four bypass grafts.
This is a rare case, and most patients who walk in with musculoskeletal complaints have musculoskeletal problems. But it illustrates the discipline behind every assessment we do: we read what the body is actually telling us, and we don’t force a presentation into a category it doesn’t fit. Usually that discipline tells us exactly how to help. Once in a while, it tells us this isn’t ours to treat — and that the patient needs to be somewhere else, immediately.
Roughly nine out of ten musculoskeletal problems can be managed conservatively. Our first job is to figure out which patient we’re looking at.
Across the full range of musculoskeletal complaints we see, roughly ninety-two percent of patients have problems that are conservatively manageable when the right approach is identified at the start.
The other eight percent fall into three groups. Some patients need a surgical consultation, and sometimes surgery, because the structural problem they have is genuinely beyond what conservative care can address. Some patients need interventional pain management — epidurals and the other options that specialty handles — because their problem is beyond what mechanical management and ordinary measures can resolve. And some patients have a problem that is being driven or sustained as much by psychosocial factors as by mechanics; with them, the right move is sometimes to set the pure mechanics aside and manage the psychosocial state while keeping them moving.
The assessment process is built to sort patients into the right group quickly. The ninety-two percent get a focused, response-guided conservative plan. The eight percent get told clearly what they actually need, and helped toward it. We don’t treat patients we can’t help — and identifying them quickly, before months go by, is one of the most valuable things the assessment does.
Sometimes the picture is unclear at the start. The response pattern doesn’t fit a category we recognize yet, or it fits one but isn’t resolving the way that category should. That isn’t a problem; it’s information. It tells us what the patient isn’t, even when it doesn’t yet tell us what they are. We have a specific approach for those cases — a brief trial of varied loading to see if anything produces a clear response, and a clear threshold for moving to imaging if it doesn’t. Recognizing what we don’t yet know is part of the discipline, not an embarrassment to be papered over.
The assessment doesn’t end after the first visit. Every visit is an examination.
Most patients have experienced a familiar pattern in healthcare: the first visit is an evaluation, and the visits after that are treatment. Once a plan is set, follow-up visits are largely about delivering the plan. The clinician asks how things are going, asks about pain levels, perhaps makes minor adjustments, and continues.
That isn’t how we work. Every visit — the second, the third, the seventh, however many it takes — is an assessment. The plan we set last time is information that the patient’s response is testing. If the response confirms the plan, we continue. If the response doesn’t confirm it, we adjust. The plan is never the goal; the patient’s response is.
When a patient comes back, we recap what was decided last time. We ask three questions in sequence, and the third matters as much as the first two.
Did you do the movement or position we agreed on?
Are you better, worse, or no different?
How are you better, worse, or no different?
The third question is the one most providers don’t ask. “I’m better” can mean fifty percent improvement with the leg pain now only intermittent and resolvable with a known movement, or it can mean five percent improvement and a desire to please. Those two responses point to entirely different next moves. We don’t take “better” as an answer. We ask how.
Alongside these questions, we collect specific measurements at each visit: pain ratings, functional measures, and trajectory markers chosen to fit the patient’s problem. These numbers are not collected to fill a chart. They feed directly into the next clinical decision — and into a structural safeguard we’ll describe in a moment.
If a patient comes back worse, we don’t panic. We assess. Worse but in a new location — say, pain that used to run down the leg and is now felt only in the back — can actually be a good sign. It’s called centralization: pain retreating out of the limb toward the spine. It’s one of the most studied responses in mechanical assessment, and when it happens, it tells us two things that matter — that we’re loading the problem in the right direction, and that the patient’s chances of a rapid and complete recovery are excellent. Centralization isn’t the only favorable response; more often we see a directional preference, where a particular direction of movement simply reduces the symptoms and improves the baseline. But the point is the same: we can’t assume what “worse” means. We have to examine it, because “worse” in one location and “worse” in another can point in opposite directions.
If a patient comes back the same, we take that seriously. Better, worse, and no different are all information — but no different requires more investigative work than the others. The first question is whether we missed a rapidly-changeable problem that should have shown response by now. If we’ve ruled that out, the patient may have a slower-changing problem that won’t move much in a few days, and the history should have given us clues about that. Either way, no different is diagnostic. It tells us that either our understanding needs revision, our chosen management may need to change, or the problem type is what it appears to be — and the timeline for response is longer than we initially expected.
If progress stops, we run a structured differential. We don’t continue what we were doing.
When a patient isn’t responding the way we expected, the temptation in most clinical settings is to continue the plan and hope. The plan was reasonable. The patient seems engaged. Time may still produce improvement. Continuing is the path of least resistance.
That path is exactly how patients end up getting unnecessary procedures, unnecessary referrals, and treatment that never resolves the problem — and it is how they become the recurrent cases who cycle back again and again, because the thing actually driving the problem was never identified.
Our approach when progress stops is structured. We work through four questions in order:
Is the mechanical diagnosis still correct? If the patient’s response over time doesn’t fit the category we initially assigned, the reassessment tells us, and we change the diagnosis. Many failures of conservative care are failures of diagnosis, not failures of effort.
Is the management strategy correct for the diagnosis? Sometimes the diagnosis is right but the specific movement, position, or dosage isn’t. The same diagnosis can require a more precisely chosen approach than the one we initially set, and the reassessment is what reveals it.
Is something else going on? A psychosocial factor that needs to be addressed before mechanical work can proceed. An inflammatory component we underweighted. Occasionally a situational one — the patient who is only here because they were required to be, and isn’t truly engaged. Sometimes the right move is to set the mechanics aside and address the person.
Is this person in the right office at all? If the answer to the first three questions doesn’t produce a change that helps, the honest conclusion may be that the patient needs something we don’t provide — a surgical consultation, interventional pain management, or another kind of care entirely. Recognizing that is part of the differential, not a failure of it.
This differential runs in the room, with the patient. It is not deferred to a future visit. It is not absorbed into vague reassurances about staying patient. When the patient’s response tells us something needs to change, we change it.
Big Ideas,
Real Impact.
Most healthcare organizations do not collect musculoskeletal outcomes. When they do, they use them to publish results, to satisfy contracts, or to demonstrate value to payers. Outcomes function as a rearview mirror — a way to describe what already happened.
We use outcomes differently. Patient response data — pain ratings, functional scores, trajectory markers — is collected at every visit and feeds directly into the next clinical decision. The interpretation of that data depends on what we understand about the patient’s problem. Some problems are rapidly resolvable, and the outcome trajectory should reflect that quickly. Others involve tissue remodeling or other processes that take weeks to resolve even when assessment and management are correct — and the outcome trajectory for those problems looks measurably different.
If the trajectory confirms what we expected for the patient’s specific problem type, the current approach continues. If the trajectory does not confirm what we expected, the approach may change — informed by the structured questions described earlier and, when appropriate, by the grand rounds process described in the next section.
The numbers are not reports on JSS’s performance. They are inputs into whether the current plan should continue, change, or stop. This is what we mean by outcomes used in real time.
And they carry a second function that’s easy to miss. They make the assessment process honest, even when no one is watching. A clinician working alone with a patient can convince themselves the patient is progressing. The data may not agree. When the data and the clinician’s impression diverge, that gap is itself information.
Cases don’t get a second look only when something has gone wrong. They get one automatically, early, before drift can set in.
In most clinical settings, a case gets a second set of eyes only if the treating clinician asks for one — which means it depends on the clinician already sensing that something is off. Our clinicians are highly trained, and working on their own they are as rigorous as any in the field. But no individual clinician, anywhere, has a structural way to guarantee that every case gets reviewed at the right moment, while there is still time to adjust. That is what our model adds.
We learned this from our own experience. When grand rounds depended on clinicians volunteering cases, only a handful of cases ever came forward — not because the others were being handled poorly, but because a clinician who believes a case is on track has no reason to flag it. When we automated the process through the outcomes data, so many cases were surfaced for review that we had to break into multiple review groups rather than one. The cases that came forward weren’t being mishandled. They were simply cases that no one would have thought to bring — and some of them turned out to benefit from a second perspective that would never have happened otherwise.
The trigger is deliberately simple and uniform. Every patient is measured against the same thresholds, regardless of diagnosis: less than 50% improvement by the third visit, or less than 75% by the sixth, flags the case automatically. It does not matter whether the patient is a derangement, a dysfunction, post-surgical, or a case still being classified — the flag fires on the patient’s own reported outcomes, not on anyone’s judgment about whether the case is on track. The trigger is objective on purpose, so that the decision to take a second look is never left to the same impression that is being reviewed.
The threshold is a flag, not a verdict. A flagged case is not a case in trouble. A patient who is 45% better at visit three trips the threshold and gets reviewed — and very often the review takes thirty seconds to confirm the case is doing exactly what it should and to keep it moving. The diagnosis-specific judgment happens in the review, not the trigger. Some problems are simply on a slower timeline: a post-surgical rotator cuff repair heals on its own schedule and won’t be 75% better by visit six no matter how well it is managed. The threshold catches these cases for a look; the review confirms the trajectory fits the case, or identifies a refinement, a missed factor, or a different next move.
The review itself is a structured process led by an experienced clinician — a grand rounds tutor — alongside a small group of trained clinicians and the treating clinician. The conversation is concrete: what is the working diagnosis, what is the management strategy, what does the patient’s response show, and is there anything worth reconsidering? It is never punitive. It is designed to stay out in front of a problem rather than react to one, and it serves three purposes at once. For the organization, it is real quality assurance — a structural guarantee about how care is delivered, and the mechanism that lets the model produce consistent outcomes across many clinicians and many locations, not just in the hands of the most experienced. For the patient, it means multiple trained clinical minds are on their problem during the process, while there is still time to act — not after twelve visits of “I’m a little better.” And for the clinicians, every review is a teaching moment: the less experienced clinicians in the room learn from cases that aren’t their own, so the whole group’s judgment sharpens over time. It is something no individual clinician working alone can replicate, because it is built into the system rather than dependent on any one person.
Some of the most important work we do is recognizing the patients we can’t help — and getting them to the right next step.
If a patient needs a surgical consultation, we say so. If a patient needs imaging because their presentation doesn’t fit a mechanical pattern, we arrange it. If a patient needs pain management because their problem has moved beyond mechanical resolution, we guide them toward it. If a patient is going to be best served by another type of care entirely, we tell them.
This is one of the easiest things to claim and one of the hardest things to actually do, because there is always a financial incentive to keep a patient in care. Our outcomes data and our grand rounds process are the structural checks against that pull. When a patient isn’t progressing the way they should, the structured reassessment we’ve described — is the diagnosis right, is the strategy right, is something else going on — runs its course. And when that process doesn’t produce a change that helps, that is itself a clear answer: the patient needs something we don’t deliver, and our job is to tell them and guide them to it.
There is also a middle path worth naming. Some patients — often with the back or neck — turn out to be safe, with nothing else going on, but simply on a longer timeline. They aren’t a quick resolution and they don’t need escalation; they need to keep putting the right movement through the system over the next several months, and they steadily improve. Recognizing that patient matters in two ways. It keeps them from being pushed toward imaging or a procedure they don’t need, just because the improvement isn’t fast. And it changes how they live in the meantime: instead of being afraid to move — afraid that the pain means damage — they move the structure deliberately, over and over, and watch it not get worse. That alone reduces the fear that so often keeps these patients stuck, and reducing the fear is part of the recovery.
Here is the part worth being direct about. A conservative trial, when it’s run through a standardized, unbiased, inter-tester-reliable, quality-assured process, is itself a diagnostic test. If a patient goes through that and isn’t getting better at all, we don’t keep going and we don’t hand them to another conservative provider to start over. We know it’s time for a different category of care — a surgical consultation, interventional pain management, or another kind of provider entirely — and we move them toward it without delay.
Most care models are only as good as the individual clinician delivering them. This one isn’t — and that’s the point. Consistent quality comes from the assessment reliability and the training infrastructure, not from hoping every clinician is exceptional.
The methodology underlying JSS has documented inter-tester reliability — two trained clinicians examining the same patient independently reach the same classification. That reliability is one of two foundations. The other is the training and quality assurance that every JSS clinician is held to: formal post-graduate training in the assessment methodology, passing examination at the required threshold, and ongoing enrollment in a data-enabled quality assurance program. The methodology gives the clinician a reliable way to assess; the training and quality assurance are what keep that reliability intact over time, case after case.
Together, these two foundations are what allow the model to be delivered consistently across clinicians, settings, and organizations.
Reliable assessment methodology.
Response-based decision making at every visit.
Outcomes used as real-time clinical inputs.
Structural grand rounds that stays out in front of cases, before they drift.
Quick, confident identification of who needs something other than conservative care.
Each of these is a discipline that can be taught, audited, and held to. Together, they are what allow the model to be delivered consistently — by a single clinician in a single setting, by a small team in a single organization, or by many clinicians across many organizations — without depending on any individual’s personal style or judgment to hold the quality together. And because every flagged case is reviewed by a group, the process does more than protect quality: it sharpens the whole team over time, as clinicians learn from cases that aren’t their own.
That structural foundation is also what allows JSS to make a specific promise to organizations evaluating musculoskeletal care. When the front end of care is actually reliable — when two clinicians examining the same patient reach the same answer — and the process is held to in real time, total cost of care comes down as a natural consequence. Not because anything is being limited, but because what happens at the start is grounded in what each patient actually needs, rather than in whatever the first clinician happened to assess.
Understand the problem. Let response and outcomes guide care. Stay out in front of every case. Recognize who needs something else. Repeat carefully, every time.
That’s the model. It is not complicated. What makes it work is the discipline of running it carefully on every patient — every visit, including the visits that look routine — with structural checks in place so the discipline doesn’t depend on any individual clinician’s vigilance alone.
Most musculoskeletal problems can be solved or substantially improved when care begins with a reliable understanding of what’s driving them. Joint and Spine Solutions is built to deliver that, one patient at a time and across whole populations. Done right, what’s best for the patient and what’s sustainable for the organization turn out to be the same thing.
See what reliable musculoskeletal care could mean for your organization.
If you are evaluating how musculoskeletal care is delivered in your organization — the outcomes, the cost, the consistency — Joint and Spine Solutions welcomes a direct conversation about whether this model fits. It is a discussion for the people responsible for that care, and it starts wherever your questions do.