Dr. Casey Kerrigan, 50, a former miler and captain of her track team at the University of Chicago, has devoted her career to studying gait and footwear, initially at Harvard Medical School where she received her M.D. and set up her first gait laboratory. Later, at the University of Virginia she set up the gait laboratory which is now run by Jay Dicharry, and which has one of the most technologically advanced treadmills in the world. For over a decade, Kerrigan has been publishing academic papers showing that virtually any shoe with cushioning in the midsole increases joint torques. "But the increased joint torques are just the tip of the iceberg of how the current shoe designs are flawed," she says. "Shoes have been constructed to cushion impact (and they don’t even do a good job at that) but the major stresses that cause injury, including osteoarthritis and all the common stress fracture sites, occur well after impact, when the foot is fully planted. And so, for a shoe to actually reduce injury, its midsole would have to provide physiological compliance (compression and release) at the precise time that the joint torques and stresses are at their peak. Yet, as simple as that sounds, no shoe has ever been demonstrated in a laboratory to provide compliance at this critical time." So, over the last 10 years, she began developing a radically different shoe that does this, and two years ago she relinquished her tenured position and created a new footwear company named OESH.
"It has not been easy, which is why I’m sure no other athletic shoe company would even attempt it," she continues. "For example, the OESH's midsole involves long filaments of carbon fiber, not some injection molded plasti-goop made in Asia. Just as funding is extremely limited for all the research I did, so is the funding to develop something so very far from the status quo. OESH Shoes is definitely David versus Goliath all over again. I designed the manufacturing equipment (and built a lot of it myself) and set up a factory here in Charlottesville, Virginia. Now with a technologically advanced, computerized automation process that rivals anything in Asia, we’re making and assembling midsoles that actually do for the first time, what a midsole should do – protect the body beyond the foot."
All this sounds exciting and daunting in the same breath. Production is quite limited. Kerrigan, who is married and has three young daughters, explains further: "Right now, we have a durable all purpose training shoe available in all women’s sizes (up to men’s 9 1/2). We are planning to develop the rest of the men’s sizes soon. Next will be some specific running shoe designs but for now this all purpose shoe is perfectly great for running. I’ve been running in OESH prototypes for the past six years (3 to 4 miles a day) with no injury (prior, I was plagued with the usual plantar fasciitis). And two of my daughters have been running in them for the past year."
But there's a lot more to this story than is first apparent. In fact, Kerrigan's interest in why most running shoes cause foot and leg injuries was inspired 30 years ago by something so rudimentary as a plank of plywood and then later by the Harvard Indoor Track which incorporated a plywood structure on top of its existing hard surface. The following article for Zero Drop by Dr. Kerrigan is an amazing account of a much-neglected aspect of the modern running shoe's questionable past. One might call the Harvard Experience a "footnote" with several "what-if?" asterisks.
The Harvard Indoor Track Revisited, by Casey Kerrigan, M.D.
I live in Charlottesville, Virginia, home of the University of Virginia (UVa), where I recently retired as professor and chair of the department of physical medicine and rehabilitation to launch a new shoe company: OESH Shoes. Though I love UVa, I must admit a lot of good things come from dear ol’ Harvard. The inspiration for completely re-inventing the athletic shoe came from my years at Harvard Medical School (where I also received my M.D.), studying gait and footwear. While a lot of the concept for OESH came from my own biomechanical studies on gait and footwear, some of it also came from the work of a fellow Harvard researcher, Thomas McMahon, a biomechanical scientist who I overlapped with slightly. I haven’t heard his work much mentioned lately but think it deserves re-visiting. Any legitimate discussion attempting to determine whether or not an athletic shoe midsole could ever reduce injury north of the foot has to consider his work.
Dr. McMahon studied the effects of imposing a compliant surface between the foot and an otherwise hard ground surface. By compliant, he meant a surface that compressed and released in tune with the rise and fall of the body’s center of mass during running (and jumping). To do this, he first built a workable compliant ground interface that would consistently compress and release in a laboratory environment. After considering a number of different compliant-like surfaces, the one he studied extensively was a simple sheet of plywood draped across 2X wood supports on either side. He had subjects run up and down the middle of the plywood and measured the deflection of the plywood in relationship to the rise and fall of the subjects’ center of mass (measured with markers placed over the subjects’ trunks). He experimented with changing the compliance of the plywood by changing the distance between the wood supports. And he found a perfect window of compliance (the Goldilocks structure - not too compliant and not too stiff) that increased stiffness in the lower extremity while simultaneously reducing foot contact time.
All that was just the groundwork (so to speak) for his next experiment--which may be the best controlled scientific experiment regarding the effect of human ground interface on injury rate ever done. What he did was incorporate his plywood structure into the Harvard Indoor Track. This was completed, along with Dr. Peter Greene, in 1977. Dr. McMahon’s plywood structure was placed on top of the existing hard surfaced indoor track. Essentially, the track comprised plywood draped across wood supports that ran between the lanes (covered by a thin polyurethane layer). This matched the Goldilocks surface from his laboratory work meant to reduce stresses and strains throughout the body. McMahon hypothesized that the compliant surface should not only reduce injury, but also improve running efficiency.
The results were just as he expected. In the 1977-1978 Harvard indoor track season, injuries were reduced by one-half compared to the prior season. Running efficiency also improved as evidenced by faster race times (approximately 3% improvement), not just by members of the Harvard Track Team, but by runners from visiting schools. These results became well known and his same plywood structure, which is still in existence today, was subsequently built at a number of other indoor tracks. The evidence was clear. A compliant surface that is physiologically tuned to the rise and fall of the body’s center of mass reduces injury and improves efficiency.
This was all in 1978. Could the success of the Harvard Indoor Track (officially known as the Harvard Gordon Indoor Track, which is now 30-plus years time tested, have anything to do with thinking that the same success could be achieved by putting a bunch of foam inside the midsole of a running shoe? I don’t know, but it was certainly around that time that the modern day running shoe was developed.
Over the past 30 plus years, the traditional athletic shoe midsole has comprised all types of foam, gel, air bladders, and plastic. To the extent reducing injuries and improving efficiency mattered, I think it may have been assumed that such midsoles would perform like the Harvard Indoor Track. The plywood track compressed and released. Doesn’t foam in a shoe compress and release too? Well yes, but that’s where the similarities end. Analyzed in a sophisticated gait laboratory, the current shoe design in no way measures up to the Harvard Indoor Track.
A typical cushioned shoe, on a person’s foot, behaves nothing like the tried and true Harvard Indoor Track. The first big difference is that the typical midsole cushions “impact”, the very first contact made with the foot. Plywood on the other hand, does not cushion impact. I’m not sure how this detail was ever missed but I don’t think anyone ever realized the distinction between what was occurring at impact versus what was occurring later in stance. Anyhow, we now know just how important that first contact with the ground is for providing feedback to the foot and body. We have shown that cushioning in a shoe reduces reflex muscle activity around impact and alters foot and ankle position (both of which are detrimental). On the other hand, striking a hard object, like a board, has a very good biomechanical effect, up-regulating feedback from the sole of the foot upward.
I’m not exactly sure where the basis for cushioning impact ever came from—but it wasn’t from a legitimate scientific study. The idea of cushioning impact (or at least trying to) has been one of the main cornerstones of athletic shoe design for the past 30 plus years. But I can’t find a single shred of biomechanical evidence to support that it’s actually impact that causes injury. In fact, there is biomechanical evidence to support the opposite -- that impact has nothing to do with injury (discussed in my recent post here). It was clear from McMahon’s data that the plywood surface did not compress at initial contact. Rather the plywood compressed and released much later in the stance phase, in tune with when the foot is fully planted and the body weight forces are at their peak.
Okay, so the first big difference between cushioning in a shoe and a plywood structure is that the plywood does not give at impact like a typical cushioned shoe does. But the second difference, which needs to be emphasized, is that the plywood structure maximally compresses and releases when the foot is fully planted. Specifically, the slopes of the rise and fall of the ground reaction force correspond precisely to the equal and opposite slopes of the compression and release of the plywood surface.
But doesn’t a cushioned midsole also compress and release when the body weight force ascends and descends from its maximum peak? Actually…No! While the athletic shoe industry has been obsessed with trying to cushion impact, any data we may have seen regarding midsole compression relates to what occurs at impact, not what occurs when the body weight force reaches its peak. Foam, gel, air bladders, and plastic are incapable of providing the same type of compression and release that McMahon showed with his plywood structure. Sure, the sole of the shoe may squish underfoot which makes it feel comfortable under different parts of the bottom of the foot, but we’ve never observed the midsole unit compress and release to a magnitude that is remotely on par with what McMahon demonstrated with his plywood track.
Over these last 30-plus years, nothing has really changed in the design of the typical athletic shoe. It has been mostly about cushioning impact combined with varying amounts of trying to control foot pronation (controlling pronation, just like trying to cushion impact, has little biomechanical basis, but that’s another story we’ll get to).
So, with all we now know about how the Harvard Indoor Track works, and how the current athletic shoe works (or more correctly doesn’t work), it is no surprise that the current running shoe design has been shown, if anything, to increase the risk for injury. From a biomechanical standpoint we know that attempting to cushion impact reduces feedback to the body, resulting in altered muscle activity and foot position at contact. It has been shown that making impact with a soft surface can actually increase injury. And most recently we showed that a typical cushioned running shoe increases peak knee joint torques associated with knee osteoarthritis. Not good.
For a shoe midsole to behave like the time tested Harvard Indoor Track, a shoe midsole would have to measurably compress and release in perfect tune with the rise and fall of the peak body weight force. It would have to be a sole that remains stiff at impact but gradually, as the body weight force reaches it peak, compresses. Specifically, the slope of the amount of compression and release of the midsole would have to be equal and opposite to the rise and fall of the peak body weight force (when injury causing forces are at their peak). And just as important, there would have to be no cushioning right at initial impact. The midsole couldn’t be made of foam, gel or air bladders, which compress too early (at impact) and do not provide any substantial compression and release when the body weight force reaches its peak. And it couldn’t be a simple metal spring, which similarly begins compressing at the wrong time, adversely affecting feedback to the body.
Designing a shoe that could be definitively shown in a laboratory to work like the Harvard Indoor Track would take a comprehensive understanding of how the body weight forces are naturally transferred under the foot and how those forces relate to the position of the rest of the body. It would take combining motion data with body weight force data to understand where and when peak stresses occur. And it would take studying gait in many individuals with varying foot and gait types across a number of conditions to understand natural force and movement patterns and to know which patterns are consistent and which are not. Those comprehensive biomechanical studies were never done 30 years ago. Only recently have we been publishing data that is informing what are these natural and consistent patterns. Simultaneously we now have biomechanical standards that can be used to rigorously evaluate the effect of any new potential shoe design…hopefully before it is ever suggested to the public. The very same parameters used to evaluate the effectiveness of a plywood structure (before it was introduced into a running track), can be used to evaluate the effectiveness of a shoe – before it is introduced to people’s feet.
So you might have guessed by now that I’ve figured out such a shoe design – OESH. Meanwhile, my research pointing out the detriments of current athletic shoe design is often cited and is typically used to support that going barefoot is best. I have nothing against that because yes, going barefoot is far better than wearing any current shoe design (yes, even a lot of new minimalist designs still incorporate major structural flaws). But all the meanwhile, I’ve been working on bringing something new to the party...a shoe that "behaves" a lot like the Harvard Indoor Track.
And that’s the story of the Harvard Indoor Track with regards to athletic footwear design. I wonder what Dr. McMahon would have to say about what we now know. He unfortunately passed away in 1999 and I never had a chance to talk to him about the new things we were just beginning to learn then about footwear. He was a brilliant scientist (and a novelist!). But it’s his success with the Harvard Indoor Track that is specifically noted in his obituary – that the track improved efficiency by 3% and reduced injuries by one-half. That cannot be forgotten.