Facebook Pixel
Brookbush Institute Logo

Tuesday, June 6, 2023

Integrated Hip Abduction Training Improves Running Mechanics and Decreases Pain

Brent Brookbush

Brent Brookbush


Research Review: Integrated Hip Abduction Training Improves Running Mechanics and Decreases Pain

By Stefanie DiCarrado DPT, PT, NASM CPT & CES

Edited by Brent Brookbush DPT, PT, COMT, MS, PES, CES, CSCS, ACSM H/FS

Original Citation: Noehren, B., Scholz, J., Davis, I. (2011) The effects of real-time gait retraining on hip kinematics, pain, and function in subjects with patellofemoral pain syndrome. Br Journal of Sports Medicine. 45:691-696 - ARTICLE

LEFT: proper stance mechanics during stance phase of running RIGHT: excessive valgus during stance phase of running

Why is this relevant?: Running is a popular form of exercise within the United States and is an essential component of many sports. According to research cited by the author's of the study, 50-85% of runners will endure an injury, and it is likely to be patellofemoral pain syndrome (PFPS). PFPS has a propensity to become chronic due to underlying biomechanical mal-alignments that tend to go unaddressed during therapeutic intervention. Typical treatment protocols focusing on hip and knee strength alone may miss an important component of rehabilitation - the integration of a re-established movement patterns into functional activity. In essence, if you affect mobility and muscle activity but do not reinforce the use of this change in recruitment strategy and movement, you may not optimize the transference of your intervention to activity. This article evaluates the effectiveness of training the motor pattern and neuromuscular control of hip abduction during the task of running itself using real time feedback.

Study Summary

Study DesignExperiment Design - Observational
Level of EvidenceLevel IV: Evidence from a well designed experimental study
Subject Demographics
  • Age: 23.3 + 5.8
  • Gender: 10 females
  • Characteristics
    • Height (m) 1.67 + 0.05
    • Mass (kg) 57.4 + 4.1
    • Weekly mileage (miles) 16.1 + 5.5
    • Duration of pain (months) 75.7 + 76.7

  • Inclusion Criteria: anterior knee pain for greater than 2 months unrelated to trauma; participation in 3+ runs/week; minimum weekly mileage of 6 miles;  PFPS as defined by pain around or under the patella with severity of at least 4/10 during running, and as determined by a Physical Therapist evaluation, pain reproduced with patella compression or palpation of the retro-patellar space; visual running analysis positive for knee valgus or pelvic drop; kinematic running analysis with motion tracking positive for excessive hip adduction > 1 standard deviation above the average for healthy recreational runners
  • Exclusion Criteria: cardiovascular disease or impairment that would impair ability to run, other injury or pain beside knee pain, any concurrent treatment
Outcome Measures
  •  Measured at initial visit, final visit, 1 month follow-up
    • During Running
      • Peak hip adduction (PHA)
      • Hip internal rotation (HIR)
      • Contralateral pelvic drop (CPD)
      • Pain - Visual analogue scale (VAS)
      • Lower extremity function index (LEFI)
      • Vertical loading rates (VLR)
        • Vertical impact peak (VIP)
        • Vertical average load rates (VALR)
        • Vertical instantaneous load rates (VILR)

    • During Single Leg Squat (SLS)
      • Peak hip adduction (PHA)
      • Hip internal rotation (HIR)
      • Contralateral pelvic drop (CPD)

  •  During Running
    • Peak hip adduction (PHA):
      • Final visit: 23% decrease, significant reduction
      • Follow up: 3% increase from final

    • Hip internal rotation (HIR):
      • Final visit: 23% decrease, not statistically significant
      • Follow up: 8% increase from final

    • Contralateral pelvic drop (CPD):
      • Final visit: 24% decrease, significant reduction
      • Follow up: 15% increase from final

    • Pain - Visual analogue scale (VAS)
      • Final visit: 86% decrease
      • Follow up: 100% decrease from final

    • Lower extremity function index (LEFI)
      • Final visit: 18% increase in function, significant reduction
      • Follow up: 10% increase in function from final

    • Vertical loading rates (VLR)
      • VIP: 10% decrease
      • VALR: 18% decrease
      • VILR: 20% decrease

  • During Single Leg Squat (SLS)
    • Peak hip adduction (PHA)
      • Final visit: 18% decrease
      • Follow up: 5% decrease from final

    • Hip internal rotation (HIR)
      • Final visit: 3% increase, not significant
      • Follow up: 3% increase from final

    • Contralateral pelvic drop (CPD)
      • Final visit: 160% increase, not considered significant
      • Follow up: 5% increase from final

ConclusionsIntegrating a corrective exercise strategy to decrease hip adduction during a functional task (running) can improve performance, resolve movement dysfunction, and improve pain even when feedback is removed; results will persist 1 month post training.  The improved pattern reduced vertical loading during stance which reduces risk of future injury.
Conclusions of the ResearchersReal time feedback during running that addressed underlying mechanical dysfunction reduced pain and improved function in individuals with patellofemoral pain syndrome.  The corrected mechanics may have led to the decrease in vertical loading forces which further reduces pain and injury risk.

Normal timing and strength of the Gluteus Medius vs inhibited Gluteus Medius

Review & Commentary:

The current study resulted in some interesting results: - hip internal rotation (HIR) and contralateral pelvic drop (CPD) both increased slightly during a single leg squat (SLS) after gait training. The authors noted these changes were insignificant; however, this may indicate that choosing an optimal integrated, whole body exercise or intervention for a motor learning approach may require further study and comparison of various techniques. Subjects in this study trained during running with no additional intervention resulting in improvements in all measurement of running with some carry over into a different task. The authors noted another interesting result - decreased vertical loading rates after training, likely the result of improved muscle recruitment and subsequent joint alignment. The improved mechanics and decreased vertical load forces all contributed to a very impressive decrease in pain after only 2 weeks and elimination of pain after 1 month.

The strongest aspect of the study is the in-depth evaluation to determine subject selection, as patellofemoral pain syndrome (PFPS) can be a vague term. The researchers used visual observation of a pelvic drop while running and kinematic software with motion trackers placed on specific landmarks for evaluation; the use of such technology eliminates human error and visual observation eliminates possible technology error. The authors provided clear and specific details related to the placement of motion trackers, as well as the purpose of all tasks and assessments. The authors clearly defined their criteria for determination of PFPS as performed by a licensed physical therapist, defined the hip adduction and hip internal rotation criterion by which each subject was measured, and cited studies validating the use of the kinematic device used for measurement.

The researchers created a standardized methodology across all subjects involving data collection during 5 single leg squats to 60º followed by 10 - 2 minute running trials (at 3.35m/s or 7.5mph) where 3D hip motion was captured for further analysis to determine peak hip adduction (PHA), hip internal rotation (HIR), and contralateral pelvic hip drop (CPD) of the stance leg. Following those trials, subjects ran for 30 minutes and recorded pain on a visual analogue scale (0 = no pain; 10 = worst pain) every 5 minutes, stopping if their pain reached a 7/10. In individuals with bilateral knee pain, researchers used the more painful side for the study. The researchers trained all subjects on the real-time feedback system which used motion tracking to display the subjects' hip adduction as they ran. Subjects were instructed to contract their gluteal muscles , to keep their knee straight and pelvis level as they ran using the visual system to monitor their success. Subjects trained for a total of 8 sessions, averaging 4 visits over the course of 2 weeks and gradually increased their run time from 15 minutes to 30 minutes -- the authors did not specify criteria for increasing the run time. Subjects received constant visual feedback during the first 4 sessions. During the last 4 sessions, researchers gradually removed the visual feedback so that subjects could better retain the feeling of correct motion rather than being dependent on feedback. Subjects were instructed to run only during these training sessions for the 2 week period.

The authors noted the biggest limitation of the study was the use of specialized motor analysis equipment that may not be available in most rehabilitation clinics or gyms. Further, the effects at the patellofemoral joint were inferred based on hip biomechanics and previous research, without actual imaging or 3D analysis at that joint. Lastly, the small sample size, lack of randomization and lack of a control group provides only observational evidence. These limitations do not detract from the information ascertained but suggest potential areas for enhancing future studies. The use of motion analysis software is essential for research purposes whether it is available for rehabilitation use or not. Information pertaining to patellofemoral biomechanical changes as a result of this training would have been an interesting addition. Future studies should aim for a larger sample size, a control group, randomization to strengthen the evidence provided, and potentially an effort to monitor patellofemoral biomechanics. It would be interesting to expand the subject evaluation and assessment to include hip abduction strength measures or to compare this protocol with an isolated strengthening protocol, and a combined isolated strengthening with real time gait training protocol. Further, the use of EMG on prime movers, stabilizers, and synergists would have add additional data on the effect training has on muscle activity and recruitment strategies.

Why is this study important?

This study demonstrates the success and importance of integrating a functional task with cuing into a corrective intervention.

How does it affect practice?

Clinicians must understand the concepts of the SAID principle (Specific Adaptation to Imposed Demands) as it applies to movement. Addressing muscle activity and mobility with release, mobilization, lengthening and activation techniques is part of a rehabilitation/corrective exercise program that should include training on functional tasks. If an individual desires improved performance in a certain task that is limited by a movement impairment - the task itself must be addressed along with the various maladaptions in individual structures.

How does it relate to Brookbush Institute Content?

The concept of motor control training is typically associated with neurological rehabilitation but it is worth addressing in any individual presenting with impaired movement patterns, regardless of diagnosis. This style of training focuses on muscles being recruited in optimal sequence and with appropriate force as controlled by the nervous system, rather than the strength of individuals muscles. It is important to recognize the nervous system controls muscle recruitment, and this role requires attention, training and retraining to achieve optimal motion. Previous research by Hodges & Richardson (1996) , Hides & Richardson (1996) , Cholewicke et al (2005) , and Hungerford et al. (2003) has demonstrated delays in muscle activation and recruitment in individuals with pain and/or injury - implying the delay may be the cause, or the result of pain and pathology (1,2,3,4). If left unchecked, this delayed firing can lead to decreased strength of inhibited muscles, as it is used less or is demoted as a prime mover and replaced with the over-activity of synergists (synergistic dominance) - a common example, is inhibition of the Gluteus Medius, and synergistic dominance of the Tensor Fascia Lata (TFL) . The TFL will create movement dysfunction that shows as increased hip adduction, internal rotation, and likely a drop of the contralateral pelvic - just as the subjects within this study demonstrated.

Janet Carr and Roberta Shepherd introduced The Motor Relearning Technique in the 1970s which focused on teaching individuals post stroke how to better normalize their movement (5). This technique involved certain principles that can be applied to the subject group in this article:

  1. Identification of a goal: decrease knee pain during running
  2. Inhibition of unnecessary activity: decrease hip adduction during stance phase of running
  3. React to effects of gravity and adjust balance accordingly while shifting weight: maintain level pelvis during single limb stance phase of running
  4. Appropriate body alignment: keep knee pointing forward and pelvis level
  5. Practice (physical and mental): 4 days/week for 2 weeks
  6. Motivation: subjects internal desire to be pain free and improve performance
  7. Feedback and knowledge of results: visual feedback provided (then slowly removed), pain is also feedback (5)

The Brookbush Institute promotes an orthopedic based framework for correction of movement impairment that involves: release & lengthening techniques for overactive and/or short muscles, joint mobilization for joint mal-alignment, isolated activation and reactive integration techniques for under-active or long muscles and full body integration techniques for movement pattern training. Individuals such as the subjects within this study would likely present with the "knees cave in" sign on the Overhead Squat Assessment used by the Brookbush Institute as a dynamic postural assessment. This study does not investigate ankle mobility, muscle activity or contribution of altered hip motion, so it is not possible to note if the knee valgus is related to Lumbo-Pelvic Hip Dysfunction (LPHCD) or Lower Leg Dysfunction (LLD) (which is commonly the next step in the Brookbush Institute's assessment protocols). Following isolated corrective exercise to address specific muscle and joint dysfunction, the individual must use their newly gained range of motion or improved motor unit recruitment in a functional, full body exercise to improve the pattern itself. We do not move throughout our day in isolated muscle activation patterns -- people move in a variety of planes, with varied speeds and must react to the environment.

As an intermediate step to specific task integration, a sophisticated approach to whole body/integrated exercise selection is the use of core subsystem integration techniques. In short, subsystems are synergistic relationships between core muscles that work synergistically during all movement patterns. The videos below demonstrate subsystem integration exercises one can use as the first step toward integrated movement pattern training for runners with LPHCD and LLD .

Anterior Oblique Subsystem Integration (Step Up to Chest Press):

Squat to Row - Posterior Oblique Subsystem Integration:

Static Lunge to Row POS Integration Progressions:

Side Step and Step up with Shoulder Series for Lateral Subsystem Integration:


1. Review: Hodges, P., Richardson, C. (1996). Inefficient Muscular Stabilization of the Lumbar Spine Associated With Low Back Pain: A Motor Control Evaluation of Transverse Abdominis. Spine,21(22), 2640-2650.

2. Review: Cholewicki, J., Silfies, S., Shah, R., Greene, H., Reeves, N. Alvi, K., Goldberg, B. (2005). Delayed trunk muscle reflex responses increase the risk of low back injuries. Spine. 30(23), 2614-2620.

3. Review: Hides, J. A., Richardson, C. A., & Jull, G. A. (1996). Multifidus Muscle Recovery Is Not Automatic After Resolution of Acute, First‐Episode Low Back Pain.Spine, 21(23), 2763-2769.

4. Review: Hungerford, B., Gilleard, W., Hodges, P. (2003) Evidence of altered lumbopelvic muscle recruitment in the presence of sacroiliac joint pain. Spine 28(14), 1593-1600

5. Montgomery, P.C., Connolly, B.H. (2003). Clinical applications for motor control. Thorofare, NJ: SLACK Incorporated

© 2015 Brent Brookbush

Questions, comments, and criticisms are welcomed and encouraged -