As someone with bilateral transtibial amputations, Jeffrey Cain, MD, knows firsthand what’s at stake when his prosthetist assigns him a single-digit number.
His K-level, or mobility predictor score, plays a big role in determining what prosthetic devices his insurer will pay for, which could affect his future health outcomes. It’s a rating that many patients with lower-limb amputations aren’t aware of even though it plays an outsized role in their future, he says.
“Most patients, when they are first injured, they don’t even know what a prosthetic foot is, let alone a K-level,” says Cain, a family physician and past chair of the Amputee Coalition board of directors. “I think the general population doesn’t get this at all.”
He encourages patients to educate themselves so they can advocate for their care and for prosthetists to be thoughtful when assigning such an important score.
But experts say that predicting a person’s future mobility and finding the best device for that future functional level is a complex challenge. Studies show that a person’s future mobility is under- or overestimated more than half of the time. To best serve their patients, O&P clinicians need to fully understand the K-level system, find the best devices for their patient within that system and, when necessary, challenge the system on behalf of their patients.
Understanding K-levels
K-levels were established by the Centers for Medicare & Medicaid Services (CMS) in 1995 as a way to gauge the rehabilitation potential of individuals with lower-limb amputations. Ranging from 0-4, K-levels rank patients from immobile to very active. The K-level system is supposed to reflect the mobility potential of patients and match them with a device appropriate for that level.
Before the K-level system, there wasn’t a way of identifying what technology was appropriate for which patient. Payers might have reimbursed a high-level device for a patient who was unable to transfer on his or her own or approved a conventional knee for a patient capable of running a marathon. The system gave the profession a common standard for patient functionality and guidelines for providers of what technologies were appropriate for that functionality.
“The goal, the ideal use of them, is so that a person living with limb loss and difference can have componentry that matches their physical ability, that would be the ideal world,” says Cain. Unfortunately, he says, it’s not an ideal world.
The K-level system, whether intended to or not, has shaped the whole profession, says Vanessa Carvalho, CPO, product owner, Amparo Prosthetics. The levels influence how manufacturers define and develop products, how clinicians structure treatment plans and prescriptions, and how insurers determine what to cover, she says.
“In a way, K-levels create a framework that controls the market, but also limits it,” she says. “The system should be structured and based on patient and performance-based metrics, but it also needs to look beyond the classification and consider what patients truly need in real life.”
When it comes to the K-levels themselves, they’ve been proven to be accurate assessments of a person’s current mobility, says Bob Gailey, PhD, PT, FAPTA, professor at the University of Miami Miller School of Medicine. He was part of the initial conversations when K-levels were being discussed and says studies have shown them to be an accurate measure of functional capabilities when the proper assessment is applied, such as the Amputee Mobility Predictor Assessment Tool.
“The original authors of the Medical Functional Classification Levels or K-levels did a really nice job of stratifying the descriptions for the five functional levels,” Gailey says. However, while the system might be able to assess a person’s functional capacity for that day, it’s much harder to predict a person’s future potential, he says.
“When you talk about potential, there is no one test that can predict it,” Gailey says. Outcome measures can’t predict how functional a patient will be six months from the test, he says. Sometimes patients exceed expectations and would do better on a more complex device than what they were given. In other cases, patients may have comorbidities that slow down their mobility more than expected. Currently, prosthetists and referring physicians use their professional expertise to gauge the functional level a patient will reach. In the future, predictive modeling, based on thousands of patients, will use age, cause of amputation, comorbidities, outcome measures, and many other variables to more accurately project the long-term capabilities of our patients, he says.
Previous research suggests prosthetic device prescription agreement when compared to patients’ current K-level is fairly equally divided between underprescribed, accurate, and overprescribed. While the reasons behind the lack of agreement have not been studied, factors such as slowed or accelerated recovery, prosthetic technology training, payer reimbursement concerns, or lack of systematic objective patient assessment could be indicators. Gailey also suggests that functional potential does not always mean improvement over time; most people will demonstrate a functional decline starting during the sixth decade of life, where maintaining functional capabilities through prosthetic interventions may be a more realistic rehabilitation goal.
Cain says he has seen many cases where patients were underprescribed because their clinicians did not take their future potential into consideration. Many times, they are given a score based on their limited functional level soon after an amputation. This can narrow down the types of devices they will be eligible for.
“They have a K-level set based on where they are at the lowest point in their life and therefore, they get a prosthetic device that doesn’t allow them to be as active and limits what they can do,” says Cain.
Finding the Right Device Within the System
While the K-level system may have its flaws, it is still the system that the profession must use.
“[K-levels] are what they are, and we have to live with them,” says Curt Bertram, CPO, FAAOP, senior manager, clinical affairs, O&P Insight. “Until there are changes, I am pragmatic about using them…. K-levels give us a starting point in determining potential. They’re not always perfect, but they’re what we use in our documentation and coding to meet policy and assist with medical necessity.”
He has seen numerous cases where clinicians were advocating for their patients for certain devices based on their K-levels, but the documentation came up short when it came to proving the patients were qualified for them.
Bertram says the patient’s activities should be specified in documentation. If, for example, a person likes to fish, the documentation should point out the actions that go into that activity, he says. Does fishing mean the patient likes to sit on a pier with a rod, or does it mean they have to step on and off a wobbling boat to troll around a lake?
“When saying they want to play with grandkids, do you mean playing UNO with a deck of cards or playing soccer?” he asks. “There’s a big difference there. You need to be specific and patient-centric.”
It’s not enough to reiterate the policy language back to the insurer, says Stacy Fuller, a certified billing and coding specialist and chief operating officer, CBS Medical Billing and Consulting. If, for example, a K3 patient is required to navigate three to seven stairs, it’s not enough to reiterate that he or she can climb three to seven stairs, she says. Rather, the reason for climbing those stairs needs to be pointed out. Perhaps the patient lives in a second-floor apartment, or has to walk up or down those stairs while carrying a load of laundry.
“Really it comes down to why this patient,” she says. “If I’m documenting about John versus Jane, what does John’s life look like when he’s at home and at work? Because every patient is different. Everyone has a different job and different hobbies. It is our job as clinicians to paint the picture of our patients so that the insurance payer reviewing the documentation can understand why the patient needs this device.”
The documentation needs to be detailed and specific, she says. “The phrase less is more does not apply to documentation,” Fuller says. “More is more. You can always overdocument. It’s hard to justify medical necessity with underdocumentation.”
In 2024, Medicare updated its policy and allowed for certain K2 patients to qualify for high-tech devices like hydraulic, pneumatic, and microprocessor-controlled knees (MPKs). Fuller says it was a positive change for the profession but cautions that clinicians who are filling the prescriptions for K2 patients must ensure all documentation corroborates the referring physician’s and meets the payer’s policy criteria.
When the guidelines first changed, Fuller says she saw a slight uptick in the number of reimbursement claims for MPKs for K2 patients. Then, as more and more were successfully reimbursed, more claims came in. These days, she sees almost as many MPK claims for K2 ambulators as for K3. However, as that number climbs, Fuller predicts audits as Medicare works to ensure that providers are following guidelines.
“I would bet pretty heavily on the fact that we are probably going to see audits coming down the pike,” she says. “Medicare is going to be checking that their program and policies are being followed. I tell everyone to make sure they know that new policy backward and forward and know what needs to be justified for K2 patients.”
To have that justification, clinicians should ensure that the MPKs they select are indicated for K2 ambulators, she says. The devices for K2 patients also need to have integrated technology, such as stumble recovery, which detects when a user trips or stumbles.
“If the technology is only indicated for a functional level 3, you’re not going to get that reimbursed for a K2 level,” she says. “Make sure in your documentation that you’ve confirmed that the knee is indicated for functional level two.”
This focus on documentation doesn’t mean providers shouldn’t be providing the technology, she says. Rather, it means they should provide the technology when needed and ensure their documentation is in order and audit-proof.
“Be cautious and smart when documenting any Medicare claim,” she says. “Always make sure your documentation supports what you’re providing and meets policy criteria.”
The same warning comes for any newer technology, which tends to be the most difficult to get reimbursed, Fuller says. If a newer device still has a miscellaneous code, it can be a tougher, but not impossible, reimbursement case to make to the insurer.
“There’s an assumption that Medicare does not pay for a miscellaneous code,” she says. “They absolutely do as long as you can document it and can show the medical justification for that patient for that item.”
Just like with all claims, clinicians need to provide patient-centric documentation backed up by case studies that prove that is the best device.
“Be cautious and smart with the documentation and make sure you’re following that policy,” she says.
Getting a New K-level
The stakes are high if a prosthetic device doesn’t reflect a person’s potential, says Cain. If the devices provided are not matched with patients’ functional abilities, they may never be as active as they are able, he says.
“How are you supposed to increase your activity level if the device you’ve been supplied by your insurance company limits your ability to be active in the world,” he asks. “You’re literally handicapped by your prosthetic device.”
Even if patients are reassessed for a new K-level, it can still be difficult for them to move up a level because of that initial device.
“During the time with that device you’re struggling, you’re hurting, you’re having a hard time getting to that level, and it may block you from being able to get there.”
That’s why the initial assessment is so important, says Carvalho.
“Clinicians need to go beyond the standard assessment and try to understand what the patient really does in daily life,” she says. “Our decisions have a direct impact on comfort, independence, and long-term outcomes.”
She says clinicians should think of K-levels as an ongoing evaluation rather than a one-time classification.
Bertram says there is room in the system to make adjustments to the K-level and the prescribed device when necessary.
“We always do our best to provide the right technology through our evaluation with our multidisciplinary team, but our patients change over time. In those instances, we need to reassess their K-level to reflect their new potential and modify the treatment plan.”
While it can be a challenge to change a functional level, it’s not impossible, Fuller says. She says CMS doesn’t give a specific timeline for how long a patient must have a device before changing his or her functional level but rather says a “reasonable” amount of time.
“It would be reasonable to expect that someone who, before amputation, was functioning at a K3 or K4 level that, with the right technology, could absolutely get back to that functional level.”
However, if a patient’s functional level increases and he or she requires new componentry, Fuller says the documentation has to prove two points: Why the patient would perform best with the new device and why the patient is underserved by their current device.
“The most important documentation I would be looking for is how their current prosthesis is not allowing them to reach their full potential,” she says. For example, if the patient is capable of performing certain tasks but the current prosthesis does not allow him or her to do them, she says, it should be documented. “The goal here is to prove that this new device is going to give them the potential to reach that functional level, and the current device that they have is not allowing that potential.”
It’s not easy, Cain says. Even with his extensive knowledge of the documentation needed to prove medical necessity for the devices he qualifies for, he’s still been denied on multiple occasions.
“And I come from a place with a high level of knowledge,” he says. “The average person doesn’t have my fortunate background to be able to understand insurance companies. If I can’t get it, what kind of chance does the average Joe have?”
Looking to the Future
Cain would love to see a full-scale up-grade to the 30-year-old K-level system. The system that stands now does not accurately reflect modern technology and safety features, he says.
“I have a 30-year-old car and a modern car,” he says. “The 30-year-old car doesn’t have antilock brakes, it doesn’t have GPS, it doesn’t have the ability to do hands-free phone. Using a 30-year-old rule to see what qualifies for safety in a car would be weird, but we’re doing it for prosthetic technology.”
He has been a part of advocacy efforts to update the system but says the process is slow and challenging.
“So right now what we have to do is work around it,” he says.
Gailey would like to see a system where each prosthetic device has a different standardized score for specific criteria in the areas of mobility, agility, and safety. The clinician could then assess the patient’s greatest needs in those categories and prescribe devices that best address those needs. Gailey acknowledges that the prosthetic profession is far from creating a system like this. Manufacturers and clinicians would all need to agree on the criteria and metrics for standardized scores.
“We’re not there yet,” he says. “Hopefully we’ll get there one day.”
Carvalho agrees that the system needs changing, but until then, it’s the clinicians’ job to successfully evaluate their patients, educate themselves on the latest devices, componentry, and technology and find a solution to best meet the needs of their patients.
“K-levels are useful, but they are not sufficient on their own,” Carvalho says. “As an industry, we have an opportunity to evolve toward more objective, patient-centered approaches that reflect real-world function and long-term potential. Doing so would help ensure that care decisions are guided not just by classification, but by what truly enables people to live fuller, more independent lives.”
Maria St. Louis-Sanchez can be contacted at msantray@yahoo.com.
