Gait Retraining in Injured Distance Runners: A Clinical Review

The purpose of this clinical review is to provide an overview of gait retraining to be used by clinicians as part of a comprehensive treatment plan for managing injured runners. Running-related injuries (RRIs) are multifactorial and it is important to consider all contributors to RRIs, including biomechanics, before determining if gait retraining is appropriate for a patient. Wearable devices may enhance aspects of gait retraining by providing external feedback to modify running biomechanics in or outside the clinic.

Key Words: Gait Modification, Overuse Injuries, Wearables, Running Mechanics, Running Medicine

Despite numerous health benefits, runners experience running-related injuries (RRIs) with a reported incidence of 3.2% to 84.9% (22). A recent study evaluating RRIs in 87 countries found that the cumulative injury percent over 18 months in Germany was 41.9% (38). Distance runners experience a higher proportion of overuse than acute injuries (5, 14, 22). Most RRIs occur in the lower extremities 5, 14, 22(), most commonly the knee (5, 14, 22, 46), followed by the lower leg and ankle (14, 22, 24, 46).
While RRIs are multifactorial in etiology, biomechanics have been suggested as an important modifiable risk factor (7, 50). Gait retraining was initially proposed several decades ago (13, 49) and has been more recently applied to alter biomechanics associated with RRIs (12, 15). Principles of motor learning should be applied (54) including use of visual and auditory feedback and/or verbal cues to promote acquisition and transfer of new movement patterns during running (9, 41, 43). Traditionally, gait retraining has been performed in a laboratory setting; however, this method is costly, time-consuming, and may not be feasible for clinicians, athletes, and coaches with limited resources. The emergence of wearable devices, including smartwatches and inertial measurement units (IMUs), allows for broader implementation of gait retraining and longitudinal monitoring outside of the clinic, which can be helpful in monitoring whether desired changes persist.
Gait retraining can change running mechanics and can be included as part of a comprehensive, multifactorial team-based approach with input from physicians, physiotherapists, coaches, and athletes to address multiple factors contributing to injury. The purpose of this clinical review is: 1) to outline common options for gait retraining targeting biomechanical variables associated with running-related injuries and 2) to present strategies to implement gait retraining in a clinical setting.

Gait analysis is traditionally performed on a treadmill; therefore, ensuring that a runner is familiar with treadmill running is important for both safety and accuracy of analyzing mechanics. Alternatively, if a treadmill gait assessment is not possible, overground running mechanics can be assessed with wearable devices (29). Prior protocols for performing gait analysis have been published (33, 40, 48). Below are recommendations and key considerations for running gait analysis.

Gait Analysis
– Allow adequate warm-up and time to familiarize to treadmill prior to capturing data (approximately 3-7 minutes) (19). 
– Collect running mechanics at self-selected speed. 
– For assessment of 2D sagittal and frontal plane kinematics, make sure that camera is aligned perpendicular to the plane of interest.
– Collect video for at least 10-15 seconds to account for stride-to-stride variability (approximately 20 steps have been recommended as the number of steps required to achieve stable spatiotemporal and kinematic mean values in running (422). 
– For 2D video analysis, collect recordings at a minimum of 120 frame rates per second (most smartphones and tablets meet this requirement).
– Wear clothing that allows appropriate landmarks to be 
viewed. 

Other Considerations
Collection of data using different footwear, speeds, and times during a run may provide further insight into potential injury risk and guide gait retraining. Runners may report change in symptoms during some of the listed conditions below. If this is the case, multiple trials should be performed to assess mechanics under different conditions. 
 a.) Running at different running speeds (i.e. long-run pace, race pace). 
 b.) Wearing different footwear (i.e. carbon-fiber plated “performance shoes” for long runs or races, spikes for speed workouts).
 c.) Measuring at both fresh and exerted states to account for changes in mechanics due to fatigue as well as downhill running – if feasible.

Clinicians should perform a comprehensive evaluation that includes: 1) collecting a detailed patient history with targeted questions for runners to assess for other factors that may contribute to injury risk (e.g., nutrition, sleep, training); 2) performing a physical exam to assess mobility, motor control, strength, and functional movements; and 3) completing a running gait analysis. Although some discomfort during running is expected, it is important to monitor pain during a gait assessment, as it can alter running mechanics and prevent an accurate evaluation of the runner‘s typical movement patterns. Additionally, we advise caution when attempting to modify a runner’s gait to conform to a perceived ‚ideal form.‘
Additional factors should be considered when determining if gait retraining is appropriate for an injured runner. These include addressing other modifiable risk factors for RRIs, performing an accurate gait assessment, and considering the timing of competition or training, since current literature suggests that it may take two to eight weeks for a runner to learn or modify their gait pattern (15).

A number of biomechanical variables have been associated with RRIs and vary by type of injury (table 1) (50). Evidence from prospective longitudinal studies suggests that low cadence (25) and increased vertical center of mass (COM) displacement are associated with an increased risk of bone stress injuries (BSIs) in collegiate runners (20, 23). A recent study demonstrated that a higher duty factor in more cushioned shoes reduced the risk of RRI (27), while another study identified lower duty factor as a predictor for BSIs (20, 27). This variable remains largely unexplored as a risk factor for other RRIs. Many of these spatiotemporal variables may be modified with gait retraining (figure 1).
Numerous kinematic variables have also been associated with RRIs. In the frontal plane, a higher peak contralateral pelvic drop and greater rearfoot eversion are risk factors for medial tibial stress syndrome as demonstrated in a prospective longitudinal study of collegiate runners (4). Excessive hip adduction has been associated with patellofemoral pain (PFP) in a cross-sectional study of female recreational runners (53) and other populations (35). Although there is a paucity of longitudinal prospective studies for sagittal plane kinematics, cross-sectional data demonstrates an association between forward trunk lean and PFP (11). Foot strike pattern has also been linked with RRIs. Runners with a rearfoot strike (RFS) pattern may be more likely to experience PFP or anterior knee pain, while running with a non-RFS (forefoot or midfoot) strike may place greater stress on the Achilles tendon and metatarsals, potentially contributing to Achilles tendon and forefoot injuries (16, 18). 
Peak tibial acceleration has been associated with running-related injuries (e.g., tibial bone stress injury) (30). While there is still debate regarding the causal relationship between this variable and RRI (7), current studies performed in healthy runners demonstrate that peak tibial accelerations can be reduced when targeted through gait retraining (15). 
While there are some established population reference values that can be used as a guideline (28), given the heterogeneity of study types and running populations, caution should be exercised in trying to fit individual patients to these norms.

The emergence of wearable technology has allowed runners to track metrics during prolonged activities in their natural running environment. External training loads such as speed, duration, and steps, as well as internal training loads like heart rate can be monitored (10, 39). Quantifying exposure through the use of wearables may provide better insight into RRIs and could potentially be used as a tool to modify training loads to reduce the risk of RRIs. Recent literature highlights the need to incorporate biomechanical and physiological load metrics, alongside training volume (i.e., weekly running mileage), to accurately monitor training stress and assess injury risk (31, 36, 39). For example, biomechanical variables such as cadence, ground contact time, stride length, and tibial acceleration, which can be measured reliably by wearables, offer valuable insight into the mechanical stress imposed on the body. This integrated approach, combining biomechanical data with physiological metrics like heart rate and session RPE, could improve injury risk stratification by accounting for the complex interplay between external and internal loads. With advancements in the development of wearables such as GPS smartwatches and shoe-mounted activity monitors (10), gait mechanics and training load can now be tracked in the field (21, 44, 47, 51). While these have not been validated to measure all biomechanical variables, some variables can be accurately and reliably measured with wearables, such as cadence (1, 29), ground contact time (29), stride length (29), and tibial acceleration (29). Wearables such as inertial measurement units are appealing due to their accuracy in capturing multiple biomechanical variables and ability to generate large amounts of raw data but have limitations including expense and often require proprietary software for data processing (10). The rapid growth of wearables suggests future versions may become available at lower costs and with improved access to meaningful data.

While impairments including mobility, motor control and strength deficits should be addressed through traditional rehabilitation interventions, these may not be adequate to resolve running injuries, and symptoms and time-loss from running may persist. Gait retraining can have a critical role in a comprehensive rehabilitation program for the injured runner in cases where biomechanics have been identified as a contributor to injury and traditional rehabilitation has failed.

Feedback Techniques
Recognizing that no ideal approach to gait retraining can be universally applied to all populations and conditions, several recent studies suggest that a faded feedback technique is superior to constant feedback in producing desired changes in running mechanics (31532). Additionally, the use of external feedback has been shown to be superior to internal feedback for enhancing motor learning acquisition when retraining movement patterns (2, 6). Depending on available equipment and setting (clinic or field), there are several methods for applying external feedback during gait retraining, including visual (e.g., mirror, screen, watch), auditory (e.g., metronome, verbal cues from clinicians), and haptic feedback (e.g., vibration from watch) techniques. A typical course of gait retraining occurs over 8-12 sessions (2-3 sessions per week) (15). Continuous feedback is provided during the first week and is gradually decreased each week until the patient is able to demonstrate gait modifications without external cues (15, 34). Furthermore, running time is expected to gradually progress over the 12 sessions. For example, during initial gait retraining sessions, running time may start at 15 minutes and as the patient nears completion of gait retraining during final sessions, running time has progressed to 30 minutes of continuous running (15, 34). Patients’ symptoms should be monitored, especially early in the program, and appropriate modifications should be implemented based on their response. For example, for a runner who has a tibial bone stress injury and demonstrates a low cadence, a 5% increase in step rate may not elicit a reduction in symptoms but an increase to 10% resolves pain. Thus, clinicians should continually monitor and adapt gait retraining for individual patient symptom response.

Gait Retraining in a Real-World Setting
The rapid development of consumer-based wearables has provided the flexibility to assess gait mechanics in real-world conditions (21, 44). As previously mentioned, while there may be limited biomechanical variables that have been validated in wearables (26), the same principles of faded feedback can be applied but the mode of application may vary. For example, cadence manipulation in the clinic may use a metronome for auditory feedback while in the field a wireless accelerometer can transmit cadence to a running watch to provide either visual, auditory, or haptic feedback in real-time (51). While emerging evidence has demonstrated that in-field gait retraining can be successful (51), there is still an enormous gap in the scientific literature in this area, especially in injured running populations, and thus gait retraining in the real-world warrants continued investigation. An example of gait retraining using wearables in real-world versus clinic is illustrated in figure 2.

The goals of gait retraining are to address impaired biomechanics suspected to contribute to a given RRI and, in theory, reduce risk for subsequent injury. However, factors specific to the injured runner should be accounted for during treatment. Changes in mechanics will alter loads to different structures (i.e. muscles, bones) which will require time to adapt to new repetitive loads experienced during running. To mitigate injury risk during adaptation, following a phase-based approach to gait retraining is recommended. Strengthening exercises can help to build tissue capacity to tolerate increased loads to a specific body region. For example, runners with patellofemoral pain will often respond well to a combined strengthening program and gait retraining that emphasizes shifting the workload to posterior chain musculature (e.g., plantarflexors and hip extensors) resulting in reduced anterior loads on the knee (52). Many of the gait retraining techniques (e.g., reducing step rate or vertical COM displacement) can be performed simultaneously with any strengthening program. Finally, few studies have examined the idea of using gait retraining as an injury prevention technique for runners who do not currently have, or are not currently recovering from, a RRI (8, 34, 45). While results have been positive, further high level research is necessary before we can recommend gait retraining to prevent future RRI in healthy athletes, especially those without a previous injury history.

Patient-reported outcome measures (PROMs) should be included as part of a comprehensive assessment of injured runners. This can be accomplished by measuring changes in symptoms during treatment or improvements in training intensity or duration. Additionally, using PROMs specific to running can be very useful to monitor running patient outcomes. The University of Wisconsin Running Injury & Recovery Index (UWRI), available in several languages including English, German, Spanish, Portuguese, Mandarin, Japanese, French, Turkish and Persian, is a validated 9-item, patient-reported outcome measure that is specific to runners and can be a helpful tool to track a runner’s subjective progress during gait retraining (17, 37).

Individualized gait retraining can be included as part of comprehensive management of injured runners. RRIs are multifactorial in nature and thus gait retraining should be considered within the entire framework of the clinical examination, including patient history, physical assessment, and gait analysis. Integration of wearables allows for practical application of gait retraining in a clinical or field setting without the need for expensive research-grade motion capture systems. Given the paucity of prospective studies, caution should be exercised in utilizing gait retraining for all injured runners or for injury prevention in healthy runners. Furthermore, gait assessment on a treadmill may not represent running outdoors and thus utilization of wearable devices can aid in assessing biomechanics in the field.

Conflict of Interest
The authors have no conflict of interest.

Ethical Approval and Informed Consent
Ethical approval and informed consent were not required for this study as it did not involve direct patient interaction, clinical trials, or the collection of personal data.

This review explores gait retraining as a rehabilitation strategy for injured runners. Given the multifactorial nature of running-related injuries (RRIs), including biomechanics, clinicians should assess all contributing risk factors before implementing gait retraining. Wearable devices can enhance retraining by providing real-time feedback to adjust running mechanics.