The Coming of the Foot Kiosk: Plantar Pressure Measurements, Arch Supports, and the Future

Another great article in LER by Cary Groner.

There’s little question that dynamic pressure data have benefit for designing foot orthoses, particularly for diabetic patients, but recent focus on the tech­nology’s limitations has clinicians won­der­ing about cost justification. And what about those ubiquitous drugstore kiosks?

After a recent paper revived the ever-contentious debate over the role of pressure measurement devices in the prescription of foot orthoses, one thing became clear: technological advances have only complicated the issue.1 These days, researchers and clinicians agree on some things, argue about others, and sometimes end up contradicting themselves.

The discussion nevertheless appears to be coming into focus around several interrelated issues: the limitations imposed on technology by the hectic pace of modern clinical practice, and what this implies about diagnostic accuracy and efficiency; the extent to which static measures are useful and may be extrapolated to information about dynamic foot function; and whether new devices may eventually assess not just vertical pressure but shear forces—the Holy Grail of in-shoe measurement for patients with diabetic neuropathy who are at risk for ulceration.

The real world

The aforementioned paper, published in the Journal of the American Podiatric Medical Association (JAPMA) at the end of 2010, reiterated a point that has been made before: namely, that when you bend a flat film pressure sensor to fit the contour of an orthosis, you compromise its ability to gather accurate data because many of its sensors are no longer perpendicular to the vertical ground reaction forces (GRF) they are designed to measure. A sensor on an incline will detect lower levels of GRF, but only because the forces perpendicular to the sensor (which it can measure) are offset by increased shear forces parallel to the sensor (which it can’t measure). The authors noted that decreased pressure recorded in association with an orthosis may just indicate that the orthosis increased some shear forces but did not decrease the total load.

Of course, nothing decreases total load except weight loss, or perhaps holding a large helium balloon; orthoses have always been about redistribution of pressure and redirection of force. The question is whether researchers and clinicians are misinterpreting the data and making errors in orthotic design as a result.

“The major limitation of pressure measurement systems is that we live in a world with three-dimensional forces,” said the lead author of the paper, Simon Spooner, DPM, who practices in Plymouth, England.

Figure 1. Isobar images illustrate the differences between static and dynamic plantar pressure patterns for a pronated foot and a supinated foot. (Provided by Tom McPoil, PT, PhD.)

“Biomechanics research, in general, is performed with people walking in straight lines on horizontal surfaces,” he continued. “But in the real world, people have to deal with inclined surfaces, go up and down stairs, and so forth. If you follow footprints on the beach, you’ll realize that people rarely walk in a straight line. So the inferences made in the laboratory may be accurate for walking on a flat, level surface, but that’s not often the way we operate.”

Spooner’s coauthor, Kevin Kirby, DPM, is in private practice in Sacramento, CA, and is an adjunct associate professor at the California School of Podiatric Medicine at Samuel Merritt College in Oakland.

“The point of the paper wasn’t that we dislike pressure measurement systems,” he said. “It’s that people have to understand that they have limitations. Sensors are only able to resolve vertical force, not medial-lateral shear forces, so there are going to be errors associated with that. It’s better than having no idea what’s going on, but we don’t really know the direction of forces acting on the foot; they may not be going through the joint axes the way we think they are.”


One of the first to point out such problems was Thomas McPoil, PT, PhD, a professor in the School of Physical Therapy at Regis University in Denver, who wrote about the issue more than a decade ago inPhysical Therapy.2 McPoil was on hand for the earliest days of plantar pressure measurement; he was working for the Public Health Service Hospital in Carville, LA, in the late 1970s, when he met British physician Paul Brand. Brand wanted a way to visualize plantar pressures in patients with diabetic neuropathy, so he came up with a device called a podoscope, which was a fancy name for something that basically consisted of a plate of glass over lights and a mirror. It was useful but static, so Brand then built a long podoscope patients could walk on—the first, albeit primitive, dynamic system.

“Pressure measurement was initially driven by the need to find out what was happening under the feet of diabetic patients with limited sensation,” McPoil explained. “Beyond that, we wanted to know what we could find out for patients with plantar pain. Was there a way to map that, to see if we were making a difference, or to find out whether the pain correlated with the location of the pressure?”

McPoil agreed that today’s in-shoe systems are limited in their utility—but that doesn’t make them useless. However, the fact that the pricetag for such systems starts around $8000 does raise questions about costs versus benefits.

“From a research perspective, those data are important,” he said. “Is the cost out of reach for most clinicians? Probably. But if you see a lot of diabetes patients it may be valuable for you.”

Part of the problem, he noted, is that the sensors’ surroundings would be hard on any technology.

Figure 2. 3-D images illustrate the differences between static and dynamic plantar pressure patterns for a pronated foot and a supinated foot. (Provided by Tom McPoil, PT, PhD.)

“You’ve got a rotten environment in the shoe,” he pointed out. “It’s hot, it’s humid, and FSRs—force-sensing resistors used in pressure systems—tend to be very reactive to heat. To compensate for that, we look at relative values rather than absolute values.”

In practical terms, that means he and his colleagues first have subjects wear the pressure sensor inside a simple shoe with a flat rubber sole. This provides a baseline value that is considered when data are then gathered with contoured orthoses or other shoes.

David Armstrong, DPM, MD, PhD, professor or surgery and director of the Southern Arizona Limb Salvage Alliance at the University of Arizona in Tucson, agreed that relative pressure measurements are crucial to clinical decision-making.

“If you have a whole big mountain range of high pressures along the metatarsals, and it looks like the Himalayas, that may be OK,” he said. “But if you have only one area of high pressure, and it looks like Kilimanjaro coming off the savannah, that’s probably not OK—and you can measure that [relative difference] more accurately than just saying what peak plantar pressure is.”

The coming of the kiosks

McPoil’s description of the early podoscope brings to mind the static pressure measurement devices now offered in drugstore kiosks, on which people stand and then receive a recommendation for one of the available prefabricated insole designs. Most clinicians are frankly dismissive of the machines, but not all.

“There’s no way you’ll be able to predict from that how high the arch piece should be,” McPoil said.

The process just involves too many unknowns, said Michael Gross, PT, PhD, a professor of physical therapy at the University of North Carolina at Chapel Hill.

“First, I don’t know how the data are processed, and second, how did the foot end up such that those were the pressures?” Gross said. “The clinical solution to most effectively address the issues that drive the foot to that position can’t possibly be known by these systems.”

For Spooner, the issue is oversimplification.

“You can’t measure a three dimensional shape from a pressure mat, end of story,” he said. “If we assume lower-limb pathologies are related to dynamic function as opposed to static stance, then the usefulness of such a system is clearly limited.”

But it’s not the end of story, as it happens. Spooner himself went on to point out that in a person whose pain is caused by long hours standing on the job, such a system could prove beneficial.

“If there’s not a lot of ambulation and the problems are caused by standing all day, then in those individuals it might be helpful,” he acknowledged.

A 2010 paper in Diabetes Care noted that patients with peripheral neuropathy often spend twice as much time standing as walking in a given day, which lends credence to the notion that standing may be underestimated as a risk factor for pain or ulceration.3

None other than Armstrong cuts the humble kiosks a little slack, in fact. He likened the machines to a drugstore blood-pressure monitor, except that in this case the device dispenses the equivalent of an antihypertensive drug.

“It gets people thinking about their feet, and I think it will help a lot of patients just by getting them to put something in the shoe,” he said. “But those it doesn’t help should seek professional care, and those professionals might consider stepping up to the technological plate with a quality physical exam that may include a gait-lab assessment. Unfortunately, those products are fantastically expensive and are out of reach for many practitioners.”

Decisions, decisions

Such devices are not only costly; they raise the question of what it’s most important to measure, and for which patients.

“What parameters are you going to look at?” asked Sarah Curran, PhD, a senior lecturer in podiatry at the University of Wales in Cardiff. “Peak pressure? Pressure-time integrals? Different systems produce different values, and it’s hard to compare like-for-like.”

Adam Landsman, DPM, PhD, assistant professor of surgery at Harvard Medical School, believes that the most important measure—at least in diabetes patients—may be rate of loading and the associated tissue deformation.4

“In diabetes, most people focus on pressure distribution, but the rate of tissue deformation is critical because the tissues are highly cross-linked and susceptible to injury,” said Landsman, who is also chief of podiatry at Cambridge Health Alliance and director of research at the California School of Podiatric Medicine.

Decisions about treatment benefit from such data, he noted. In one study of patients with recalcitrant foot ulcers, for example, Landsman reported that reduction in peak pressures at ulcer sites ranged from 70% to 92%, and load rate decreased significantly, when subjects wore ankle-foot orthoses.5 Such assessments depend on in-shoe pressure measurement technology.

Knowledge of load rates can also affect orthotic design and materials, he continued. Open-cell foams dissipate mechanical energy quickly but don’t do well with sustained loads because the materials remain flat under pressure. Closed-cell foams, by contrast, don’t dispel sudden loads well, but hold up much better under sustained ones.

“What you really need is an insole that has open-cell foam at the areas of peak impact—at the heel and across the metatarsals—with a base layer of closed-cell foam,” Landsman said. “When people talk about a trilaminate insole, they mean this laminate of open- and closed-cell foams to provide this multiple component ability to dissipate mechanical forces.”

The servant and the master

Landsman’s point, in context, is that the technology must serve the clinician’s experience and contribute to better decisions and designs. This may seem obvious, but most clinicians resist expensive investments in equipment unless the benefits clearly outweigh the costs—and many believe pressure measurement systems cost more than they’re worth.

“I think both doctors and industry have homework to do,” said Armstrong. “Doctors should work harder to get these devices into their clinics and use them, because you can’t manage what you can’t measure. And the industry would do well to create an affordable price point. If the devices are a little less accurate but a lot cheaper, and doctors can make decisions based on the data, that’s fine.”

Kevin Kirby, for example, doesn’t use pressure measurement devices in his clinical practice. The reasons have partly to do with expense and partly with time. On a recent (and typical) day, he saw 29 patients. To add pressure measurement to his clinical routine would mean hiring another staff person in addition to buying the equipment, then spending significantly more time on each case.

“Is the data we’d get really necessary to fine-tune an orthotic?” he asked. “In ninety percent of cases it isn’t.”

But Kirby also emphasized that most of his practice is treating athletes and work-related injuries. For a clinician who had a significant caseload of patients with diabetic neuropathy, he thought pressure measurement devices could be invaluable.

At UNC, Michael Gross, too, trusts his clinical judgment more than the data from a machine.

“In-shoe systems measure the pressures imposed by the support surface on the foot, but they don’t tell you anything about how the person got to the point where those pressures end up being imposed,” he said.

“Say they have excessively high pressures on their first metatarsal head, and that’s the source of their pain,” he continued. “You watch them walk and they’re quickly going over to the medial side of their forefoot and pushing off from there. For the clinician, the question is then: What’s driving them to load the medial side of their forefoot so heavily and what should I do about it?”

Gross noted that such a movement pattern could be caused by a tight triceps surae driving pronation or by tibial varum or genu varus, either of which could drive the distal third of the leg farther off vertical during gait, or by forefoot varus leading to pronation.

“Those are three different things that could cause the weight bearing on the first metatarsal head, and theoretically yield the exact same plantar pressure pattern,” he said. “So it’s not the pressure reading that gives you the information you need to treat the patient effectively—it’s the clinical exam. We never know how they move until they show us.”

Seeking correlations

The time constraints imposed by busy practices pale compared to those faced in other situations, such as the military. Not surprisingly, time limitations in the U.S. Army have pushed its clinicians to seek effective methods of determining what interventions could prevent injuries in troops entering basic training.

These efforts have led to some intriguing findings about the relationship between static and dynamic foot measures. Recent research at Fort Sam Houston in Texas sought correlations between static arch height and dynamic plantar pressure measurements, for example.6 Another study from the same team sought to relate static foot posture to dynamic plantar pressure measurements.7

The papers reported plausible if not powerful associations, and such findings may seem arcane out of context. Lead author Lt. Col. Deydre Teyhen, PT, PhD, was unable to comment because she was in mid-deployment, but McPoil was a coauthor on both papers.

“I think they wanted to know, with two thousand recruits coming through, if there was a quick way to have them walk across a platform, then use those data to get a sense of their foot shape and arch height and get them into the best shoe to prevent basic training injuries,” McPoil said. “We gave them some of the measurement systems, and it’s an expensive process. It’s not something a typical clinician could use, but I can see it in the army.”

Although such data may help make better shoe choices, McPoil pointed out the obvious problem with trying to apply it to orthotics: even predicting 60% of arch height, as Teyhen and her colleagues did, doesn’t say enough about the shape of the foot to suggest an effective orthotic design; it’s another variation of the problem facing the kiosks.

Others have sought correlations between static and dynamic measures to simplify diagnosis and treatment, with results all over the map.

For example, researchers at Pennsylvania State University reported in 1997 that only about 35% of the variance in dynamic plantar pressure could be explained by radiographic measurements of foot structure, and concluded that gait dynamics probably exerted much more influence on plantar pressure than foot structure did.8 Two years later, Australian scientists found no significant relationship between static measures and dynamic motion of the medial longitudinal arch.9 Another Australian study from 2007, however, reported that measurements of arch height and arch-height ratio taken from standing subjects explained between 66% and 83% of the variance associated with those measurements at midstance during walking and running.10

Similarly, McPoil and his colleague, Mark Cornwall, PT, published studies in 2005 and 2007 reporting that longitudinal arch angle (LAA) significantly predicted dynamic foot posture during walking and running, respectively.11,12

In 2010, Danish researchers reported in Gait & Posture that although video sequence analysis accurately quantified midfoot kinematics during walking, the Foot Posture Index (FPI) classification system they used was a relatively poor predictor of navicular height and navicular drop.13

And this year U.K. researchers, acknowledging that LAA had been shown to be a reliable measure of static midstance foot posture, reported no significant differences between static and dynamic LAA with orthosis wear, as measured using an optoelectric movement analysis system.14

Helen Branthwaite, MSc, a research podiatrist at Staffordshire University and a coauthor of that study, said, “As clinicians, we felt that using that arch angle was a simple thing to do in the exam room, so we wanted to know if that could be used as a measure of what the orthotic was actually doing. Our results showed that it can be.”

Branthwaite reiterated the opinions of Mike Gross and others, however, regarding the role of such data in practice.

“Any of this is just a tool for the clinician to use with their expertise and knowledge of diagnosis,” she said.

Shear magic

Because shear forces have so far been virtually impossible to measure inside footwear, they have been a neglected stepchild.

“Shear forces are to podiatrists what dark matter is to astrophysicists,” said David Armstrong. “Just as with dark matter, we can’t measure it directly—but it’s the most important thing no one’s measuring.”

Because shear forces significantly increase ulcer risk in diabetic neuropathy patients, researchers are seeking ways to assess this devilishly elusive entity. One of Armstrong’s colleagues, Bijan Najafi, PhD, associate professor of applied biomechanics at Rosalind Franklin University in Chicago, is developing socks that incorporate fiber optic threads as a means of measuring shear.

“To accurately predict foot ulcers, you need to simultaneously analyze three things at several locations on the foot—change in plantar pressure, value of shear force, and the thermal response to these two values,” Najafi said.

The optical fibers in the experimental “SmartSox” are finer than a human hair but flexible and durable, and the physical principles involved are relatively straightforward.

“When light is propagated through the fibers, any mechanical change in the environment—say, an application of force or a temperature change—affects its wavelength,” Najafi explained. “We can measure those changes and identify alterations in temperature or, based on horizontal deformation, shear forces.”

Najafi and his team have finished in vitro validation and have funding for an international study pending from the Qatar National Research Fund, with which they plan to begin testing in human subjects this fall.

Cary Groner is a freelance writer based in the San Francisco Bay Area.


1. Spooner SK, Smith DK, Kirby KA. In-shoe pressure measurement and foot orthosis research: a giant leap forward or a step too far? J Am Podiatr Med Assoc 2010;100(6):518-529.

2. Orlin MN, McPoil TG. Plantar pressure assessment. Phys Ther 2000;80(4):399-409.

3. Najafi B, Crews RT, Wrobel JS. Importance of time spent standing for those at risk of diabetic foot ulcer patient. Diabetes Care 2010;33(11):2448-2450.

4. Landsman AS, Meaney DF, Cargill RS 2nd, et al. 1995 William J. Stickel Gold Award. Highest strain rate tissue deformation. A theory on the mechanical etiology of diabetic foot ulcerations. J Am Podiatr Med Assoc 1995;85(10):519-527.

5. Landsman AS, Sage R. Off-loading neuropathic wounds associated with diabetes using an ankle foot orthosis. J Am Podiatr Med Assoc 1997;87(8):349-357.

6. Teyhen DS, Stoltenberg BE, Collinsworth KM, et al. Dynamic plantar pressure parameters associated with static arch height index during gait. Clin Biomech 2009;24(4):391-396.

7. Teyhen DS, Stoltenberg BE, Eckard TG, et al. Static foot posture associated with dynamic plantar pressure parameters. J Orthop Sports Phys Ther 2011;41(2):100-107.

8. Cavanagh PR, Morag E, Boulton AJ, et al. The relationship of static foot structure to dynamic foot function. Biomech 1997;30(3):243-250.

9. Cashmere T, Smith R, Hunt A. Medial longitudinal arch of the foot: stationery versus walking measures. Foot Ankle Int 1999;20(2):112-118.

10. Franettovich MM, McPoil TG, Russell T, et al. The ability to predict dynamic foot posture from static measurements. J Am Podiatr Med Assoc 2007;97(2):115-120.

11. McPoil TG, Cornwall MW. Use of the longitudinal arch angle to predict dynamic foot posture in walking.  J Am Podiatr Med Assoc 2005;95(2):114-120.

12. McPoil TG, Cornwall MW. Prediction of dynamic foot posture during running using the longitudinal arch angle. J Am Podiatr Med Assoc 2007;97(2):102-107.

13. Nielsen RG, Rathleff MS, Moelgaard CM, et al. Video based analysis of dynamic midfoot function and its relationship with Foot Posture Index scores. Gait Posture 2010;31(1):126-130.

14. Burn H, Branthwaite H, Chockalingam N, et al. Do foot orthoses replicate the static longitudinal arch angle during mid-stance in walking? Foot 2011; Jan 18 [Epub ahead of print].

David G. Armstrong

Dedicated to amputation prevention, wound healing, diabetic foot, biotechnology and the intersection between medical devices and consumer electronics.

One comment

Leave a Reply