Collisions in Clinical Care

Home > Articles > Collisions in Clinical Care
By John T. Brinkmann, MA, CPO/L, FAAOP(D)

Normal ambulation is performed intuitively, without much conscious attention or effort. Relatively minor disruptions in nerve, muscle, and joint function, however, can result in significant impairment, and, as a consequence, walking often requires more focus and energy. O&P clinicians most often assess gait patterns by observation to determine appropriate treatment. Clinical decisions and specific recommendations are dependent on an accurate assessment of how each patient's gait pattern deviates from normal, the underlying cause of those deviations, and the ways in which O&P interventions can address those deficits. Because observation alone does not allow a detailed analysis of many important features of gait, identifying how a gait pattern deviates from normal is often an intuitive process for experienced clinicians. But just as walking requires more thought and focus by the patients with impairments, understanding the biomechanical causes of deviations and potential solutions requires the clinician to have a deeper understanding than simply identifying how a gait pattern deviates from normal in a global sense.

There are multiple ways for individuals to accomplish specific functional goals necessary for locomotion. The complex interactions between the neurological, muscular, and skeletal systems that support human walking create options for individuals when one or more elements of that system are disrupted. We can perform equally complex alternative movements to accomplish mobility and other functional goals when a normal gait pattern is no longer possible. Patients have the option of responding to the changes imposed by a disease process or trauma by adapting how they use their remaining capabilities to achieve functional objectives as safely as possible, even when the resulting gait pattern is not optimal in terms of energy expenditure, comfort, or long-term tissue health. For example, knee hyperextension is an effective short-term solution for addressing knee instability secondary to quadriceps weakness, despite the long-term consequences (e.g., pain and increased energy expenditure) related to a recurvatum deformity.

The complexity involved in walking creates challenges for clinicians evaluating patients' movement patterns and making decisions about structural interventions. For example, it can be difficult to determine which motions should be allowed and which should be limited to optimize mobility given a specific pattern of functional deficits. Researchers and clinicians have developed models of human walking to reduce complexity and gain insight into the underlying mechanics that form the basis of a gait pattern. This article describes how one model of gait can provide insight into specific clinical decisions.

Reducing Complexity

A complex phenomenon can be better understood if a simplified representation can be created or described. Such simplified representations are often called models and can represent patterns of thinking, concepts, or physical objects. In the same way that a model of the Eiffel Tower is significantly less complex than the original but retains enough features of the real thing to be recognizable, models of gait are based on simplified assumptions and remove the variability between different individuals, but still describe the features of gait that are considered most important. By narrowing the focus to the factors that most significantly impact gait, this simplicity can help clinicians understand what they observe and make appropriate recommendations. Because models do not perfectly match reality, Stefania Fatone, PhD, BPO(Hons), professor of physical medicine and rehabilitation at the Northwestern University Prosthetics-Orthotics Center (NUPOC) describes them this way: "A model is a lie that tells the truth." Despite the limitations that come from reduced complexity, models can be valuable tools for exploring the various factors that contribute to typical and atypical gait.

The Six Determinants of Gait

In 1953, Saunders, Inman, and Eberhart published a paper that identified six factors that they believed determined normal and pathological gait.1 Saunders et al. stated that "the adoption of the concept that fundamentally locomotion is the translation of the center of gravity through space along a pathway requiring the least expenditure of energy supplies the necessary unifying principle which permits of qualitative analysis in terms of the essential determinants of gait."1 They identified six features of gait that they claimed achieved this objective during normal gait. These principles and assumptions were taught to several generations of O&P practitioners as the foundation of understanding normal and pathological gait. Research, some of which was performed by Dudley Childress, PhD, and Steven Gard, PhD, at NUPOC, has shown that the determinants do not affect movement of the center of mass (CoM) in the way that the model described. Additionally, heel rise was not included as one of the determinants, but it has been shown to significantly affect the movement of the CoM. Baker describes the limitations of the six determinants model in a YouTube video series titled "Why we walk the way we do."2 In short, the assumption made by Saunders et al. that the maintenance of a smooth trajectory of the CoM sufficiently explains the most important features of normal walking fails to match reality in important ways.

The Inverted Pendulum and Dynamic Walking

One simplified explanation of the movement of the lower limbs during walking describes that motion as a pendulum. During swing phase, each lower limb functions as a pendulum as it swings through space below the attachment point at the pelvis. During stance phase, the limb functions as an inverted pendulum as the CoM moves in a curved path over the planted foot. The dynamic walking model considers these two pendular actions of the lower limbs. In this model, locomotion is generated passively, with minimal force applied to sustain the alternating pendular movement of the two limbs. Passive physical models have been built that can alternate the swing and stance phases of two rigid limbs as the model moves down a slight incline, with gravity providing the necessary energy. More complex models that incorporate additional segments and jointed limbs have also been constructed. These physical representations demonstrate that a reciprocal gait pattern is possible with a minimal input of energy.

In the simplified inverted pendulum model, the point of contact with the ground is one point. A refinement of this model referred to as the rocker-based inverted pendulum model has been developed to more closely approximate how human limbs function during gait. Adding a rocker-shaped foot section helps blend the transitions between the point when the leg contacts the ground and when it leaves the ground as the weight is transitioned to the contralateral side.

Energy Requirements

In the inverted pendulum gait model, the rise and fall of the body's CoM does not theoretically require mechanical work, because kinetic energy and potential energy oscillate out of phase with each other. The analogy of a bike moving over a series of hills can be helpful in understanding this phenomenon. A bike speeds up as it descends an incline and slows down as it ascends the next incline. In the absence of friction, the kinetic energy of going downhill would carry the bike up the next incline. In this way, motion would continue without requiring additional input of energy. This provides insight into the mechanical costs required for moving the body during gait.

While walking on a level surface is highly efficient, some energy is required to initiate motion and changing the direction and speed of movement. Researchers have demonstrated, however, that most of the energy required to walk is related to the loss of energy that occurs during transitions between steps.3,4 Energy is lost when each foot contacts the ground and the CoM is redirected upward. This energy loss is referred to as a collision loss, and this energy loss requires that we put energy into walking to maintain a steady pace.

A Step-by-step Explanation

Imagine someone's right foot contacting the ground during double-limb stance. At this point, the CoM is at the lowest point in the gait cycle. As the right leg moves through stance phase, the CoM moves upward to the highest point (at midstance), and then downward to another low point in terminal stance. The left foot contacts the ground and redirects the CoM upward as the left limb accepts weight and moves through stance phase. The collision of the left foot with the ground is the force that keeps the CoM from continuing downward during terminal stance on the right side and redirects it upward.

This movement conserves energy because the CoM speeds up as it descends from midstance to terminal stance. The kinetic energy resulting from gravity is recovered passively during the subsequent step. Walking is incredibly energy efficient due to this pendulum mechanism, but it requires the input of some energy because we lose energy each time a foot contacts the ground. Without muscle activity to make up for the energy lost as we transition from step to step, forward progression would slow and eventually stop.

Conserving Energy

Kuo and Donelan describe three strategies for offsetting the energy loss due to collisions in their 2010 open-access article "Dynamic Principles of Gait and Their Clinical Implications" ( The first strategy involves leaning the torso forward. The hip extension torque needed to support the torso in this position acts against the limb that is extending, performing positive work, and offsetting energy loss. Another strategy is to push off with the trailing limb just after the collision occurs at the leading limb when it contacts the ground. A third strategy, preemptive push off, reduces the collision and offsets the energy lost during collision by pushing off with the trailing limb just before the leading foot collides with the ground. This slows the descent of the CoM so that its speed is reduced when the collision occurs. This can significantly reduce the energy lost during this transition and reduce the amount of energy needed to maintain a constant speed.

Clinical Implications

The strategies above are essential for normal walking and are disrupted in common pathological gait patterns. When patients lose the ability to effectively manage step-to-step transitions, walking performance suffers, requiring compensations that we observe as gait deviations. O&P interventions can minimize the effect of these disruptions and, in some cases, restore function by restoring the ability to perform these strategies. A partial foot amputation reduces the ability to push off in terminal stance, resulting in a more abrupt collision of the contralateral foot with the ground. We observe shortened contralateral step in these cases and recognize that the overall gait pattern is indicative of an increase in force with which the contralateral heel contacts the ground. The addition of a toe filler in the shoe does not sufficiently restore the lever arm of the foot as only an intervention with an appropriately rigid keel, ankle, and proximal extension can.

Weak plantarflexor muscles limit the ability to push off, and often result in a less efficient compensatory strategy for controlling the CoM. The decision
to provide an AFO with a solid or articulated ankle should be informed by an understanding of how terminal stance will be impacted by either design. If the patient is unable to adequately control dorsiflexion during stance, it is likely that a design that allows this motion will decrease the efficiency of step-to-step transitions, and result in a less desirable gait pattern. Research has demonstrated that dynamic prosthetic alignment involves restoring normal roll-over shape. Orthoses should also be aligned using appropriate footwear modifications to optimize sagittal plane mechanics and facilitate a more normal gait pattern. Positioning of the foot and ankle properly in relation to the shank of the tibia should be standard practice in orthotics as it is in prosthetics. Many of the materials traditionally used in orthotic designs lack the energy storing and return capabilities that we have come to expect in prosthetic foot designs. As advances are made in materials, practitioners should recognize the value of these materials and designs for addressing key deficits in function for our orthotic patients.


Our understanding of normal and pathological gait can be improved by the application of more accurate models. These simplified explanations can provide a framework for O&P designs and improve outcomes for our patients. Clinicians will continue to make many decisions based on their experience and judgement, and a better understanding of these models can improve decisions by giving us a better understanding of the underlying biomechanical principles that impact the key aspects of gait.

Author's note: Special thanks to Matthew Major, PhD, for his insights into normal gait, models of walking, and the dynamic walking model that served as background for this article.

John T. Brinkmann, MA, CPO/L, FAAOP(D), is an assistant professor at Northwestern University Prosthetics-Orthotics Center. He has over 30 years of experience in patient care and education.




1.       Inman, V. T., and H. D. Eberhart. 1953. The major determinants in normal and pathological gait. The Journal of Bone and Joint Surgery 35(3):543-58.

2.       Baker, R. 2013. Why We Walk the Way We Do 1. University of Salford.

3.       Donelan, J. M., R. Kram, and A. D. Kuo. 2002. Mechanical work for step-to-step transitions is a major determinant of the metabolic cost of human walking. The Journal of Experimental Biology 205(Pt 23): 3717-27.

4.       Kuo, A. D., J. M. Donelan, and A. Ruina. 2005. Energetic consequences of walking like an inverted pendulum: Step-to-step transitions. Exercise and Sport Sciences Reviews 33(2), 88-97.

5.       Kuo, A. D., and J. M. Donelan. 2010. Dynamic principles of gait and their clinical implications. Physical Therapy 90(2):157-74.