Nijmegen Gait Analysis List

1. Abstract

The Nijmegen Gait Analysis List (Dutch: Ganganalyselijst Nijmegen; GALN) is a standardized observational gait analysis instrument developed to assess, document, and monitor pathological walking patterns in clinical and research environments. Formulated by Jaap J. Brunnekreef and Cor J.T. van Uden in 2004 at the Sint Maartenskliniek in Nijmegen, the Netherlands, the GALN addresses the historical vulnerability of visual gait analysis—specifically, poor inter-rater agreement and subjective observer bias—by providing a structured observational protocol. The instrument systematically evaluates human locomotion across 5 distinct gait phases: initial contact, loading response/midstance, terminal stance, initial swing, and terminal swing. These functional phases are operationalized through 13 specific biomechanical items per lower extremity, yielding a total of 26 items across both sides. Each item isolates the kinematic alignment and excursion of key anatomical segments: the trunk, pelvis, hip, knee, and ankle-foot complex. Scored primarily via dichotomous and categorical metrics capturing the presence or absence of explicit deviations, the tool aggregates segment-specific alterations into an interpretable profile of functional impairment. Psychometric evaluations demonstrate that the GALN exhibits moderate to substantial inter-rater reliability (Cohen’s kappa values generally ranging between 0.40 and 0.85 depending on joint segment and observer experience) and high intra-rater stability. Criterion-related validity assessments comparing GALN observational ratings to three-dimensional (3D) computerized kinematic motion capture confirm acceptable sensitivity in detecting pathological deviations, particularly in the sagittal plane. The instrument serves as a critical bridge between qualitative visual inspection in daily physical therapy practice and cost-prohibitive laboratory instrumentation, providing an empirical framework for functional diagnosis, rehabilitation prescription, and therapeutic outcome tracking across pediatric, adult, and geriatric populations with musculoskeletal and neuromuscular disorders.

2. Keywords

Nijmegen Gait Analysis List, Ganganalyselijst Nijmegen, observational gait analysis, kinematics, biomechanics, physical therapy, locomotion, lower extremity, rehabilitation, reliability, validity, clinical assessment

3. Authors

The Nijmegen Gait Analysis List was developed and validated by clinical researchers based in the Netherlands:

• Jaap J. Brunnekreef, PhD, PT: Department of Physical Therapy and Research, Sint Maartenskliniek, Nijmegen, the Netherlands. Dr. Brunnekreef is a clinical physical therapist, movement scientist, and researcher focusing on orthopedics, physical rehabilitation, lower extremity biomechanics, and clinical measurement methodology.
• Cor J.T. van Uden, PhD, PT: Department of Physical Therapy and Orthopedic Research, Sint Maartenskliniek, Nijmegen, and Allied Health Care Research, HAN University of Applied Sciences, Nijmegen, the Netherlands. Dr. van Uden has published extensively on clinical gait assessment, orthotic interventions, and functional outcome measures.

Institutional oversight and initial clinical trial implementation were coordinated through the Sint Maartenskliniek (Hengstdal 3, 6522 JV Nijmegen, the Netherlands), an internationally recognized center of excellence in orthopedics, rheumatology, and rehabilitation medicine.

4. Purpose

The primary objective of the Nijmegen Gait Analysis List is to provide physical therapists, physiatrists, orthotists, and movement specialists with a standardized, reliable, and clinically feasible observational tool to evaluate human locomotion. Before the standardization of visual gait assessment protocols, routine clinical gait examinations relied heavily on unstructured, subjective appraisal. Practitioners often observed a walking patient and generated qualitative narrative descriptions (e.g., “patient walks with a noticeable limp and antalgic trunk shift”). Such unstructured observations demonstrate poor diagnostic reliability, exhibit significant susceptibility to confirmation bias, and prevent meaningful outcome comparisons across different clinics or between successive rehabilitation sessions.

While optoelectronic three-dimensional motion capture systems, ground reaction force plates, and dynamic electromyography (EMG) serve as the scientific gold standard in biomechanical laboratories, these technologies are characterized by high equipment costs, extensive calibration times, technical expertise requirements, and space constraints that render them impractical for everyday ambulatory clinics. The GALN bridges this clinical divide. It offers a structured checklist that decomposes continuous locomotor performance into discrete, observable kinematic events across specific spatial-temporal intervals.

In clinical practice, the GALN is designed to:

• Systematically identify kinematic deviations at the trunk, pelvis, hip, knee, and ankle-foot complex throughout both the stance and swing phases.
• Establish a standardized diagnostic baseline of gait pathology in patients presenting with orthopedic disorders (e.g., osteoarthritis, anterior cruciate ligament reconstruction, total joint arthroplasty), neurological impairments (e.g., stroke, traumatic brain injury, cerebral palsy), and general musculoskeletal dysfunctions.
• Formulate targeted therapeutic interventions, such as gait retraining, targeted muscle strengthening, joint mobilization, neurodevelopmental re-education, and the prescription of mobility aids or lower limb orthoses.
• Measure objective functional changes over time to verify therapeutic efficacy or document longitudinal disease progression.

In clinical research, the GALN provides an accessible, validated performance-based metric for comparative trials, epidemiologic investigations, and multicenter rehabilitation studies evaluating therapeutic techniques where instrumented gait laboratories are unavailable.

5. Psychological and Biomechanical Construct

The construct assessed by the GALN is human locomotor competence, operationalized through observational gait kinematics. Human walking is a cyclical task requiring the coordinated integration of dynamic balance, motor planning, neuromuscular execution, and skeletal geometry. The GALN captures abnormal functional variations within this motor control loop by breaking down the gait cycle into distinct physiological phases and anatomical levels.

The 5 Gait Phases

1. Initial Contact (Heel Strike): The moment the reference foot touches the ground. The construct focuses on whether the contact is correctly initiated with the heel, or whether premature flat-foot or forefoot contact occurs due to dorsiflexor insufficiency or plantarflexor contractures.
2. Loading Response (Shock Absorption / Weight Acceptance): The interval where the limb decelerates the body mass and accepts full weight bearing. It assesses rapid knee flexion shock-absorption dynamics and pelvic stabilization.
3. Midstance (Single Limb Support): The body center of mass translates forward over the stationary foot. Biomechanical constructs measured here include single-limb stability, hip abductor engagement, knee extension stability, and coronal alignment of the trunk.
4. Terminal Stance / Preswing (Weight Release / Push-Off): The heel rises, and the limb prepares to advance. Evaluated deviations include lack of heel rise, delayed or absent hip extension, inadequate ankle plantarflexion push-off, and compensatory trunk movements.
5. Initial and Terminal Swing (Limb Advancement): The non-weight-bearing limb accelerates, clears the floor, and decelerates prior to the next initial contact. Constructs include active foot clearance (dorsiflexion), knee flexion during early swing, hip flexion excursion, and foot positioning for stable subsequent landing.

Anatomical Segments Evaluated

For each limb across these functional periods, 13 specific kinematic questions examine the following segments:

• Trunk: Compensatory strategies such as excessive lateral leaning (e.g., Duchenne compensation), anterior trunk bending (compensating for quadriceps weakness), or posterior trunk thrust (gluteus maximus gait).
• Pelvis: Maladaptive vertical and transverse drops, such as the Trendelenburg sign (pelvic drop on the contralateral swing limb due to ipsilateral gluteus medius weakness), or pelvic retraction/hike during swing.
• Hip: Impairments including insufficient extension in terminal stance, internal or external rotation abnormalities, and adduction/abduction discrepancies during stance and swing.
• Knee: Pathological deviations including knee hyperextension (genu recurvatum) during midstance, loss of the shock-absorbing flexion wave during loading response, or stiff-knee gait (inadequate swing-phase flexion).
• Ankle and Foot: Deviations such as foot drop during swing, initial contact with the forefoot or flat foot, premature heel lift, pronation/supination collapses, and lack of active terminal plantarflexor propulsion.

6. Theoretical Framework

The conceptual framework of the Nijmegen Gait Analysis List is rooted in the functional biomechanics of human locomotion developed by Dr. Jacquelin Perry at the Rancho Los Amigos National Rehabilitation Center, complemented by dynamic systems theories of motor control. Perry’s paradigm conceptualizes walking not merely as continuous forward motion, but as the execution of three core functional tasks: Weight Acceptance, Single Limb Support, and Swing Limb Advancement. Within this model, deviations observed at a single joint (e.g., knee hyperextension) are never viewed in isolation; they are treated as mechanical adaptations to underlying muscular weakness, joint stiffness, structural deformity, or altered proprioception.

From a psychophysical measurement standpoint, observational gait analysis involves complex visual perception and cognitive categorization. Human visual tracking possesses a limited sampling frequency and struggle with simultaneous multi-joint processing. When an observer watches a patient walk without structured guidance, visual attention shifts unsystematically between high-motion distal segments (e.g., swinging feet) and posture-dominant proximal segments (e.g., trunk swaying). This visual clutter induces high inter-rater variance.

The GALN addresses this psychophysical limitation through hierarchical decomposition. By structuring visual observation into an itemized, phase-by-phase and joint-by-joint assessment, the GALN reduces cognitive load. Clinicians systematically focus visual attention on predefined anatomical markers at explicit temporal intervals. The theoretical assumption is that when visual inspection is decoupled into standardized spatial-temporal components, the reliability and validity of observational ratings approach the diagnostic consistency achieved by expensive motion-capture systems.

7. Validity

Validation of the Nijmegen Gait Analysis List has focused on content, construct, and concurrent criterion validity, primarily in orthopedic and neurological populations.

Content and Face Validity

During its initial development by Brunnekreef and van Uden (2004), the items were derived from extensive clinical literature and refined through consensus among experienced physical therapists and clinical biomechanists. The selection of the 13 specific items per limb covers the most prevalent biomechanical gait abnormalities observed in clinical pathology, establishing strong face and content validity.

Concurrent Criterion Validity

The validity of observational tools is typically established by correlating visual ratings with objective data from three-dimensional computerized gait analysis (3D-CGA) using optoelectronic cameras and passive reflective markers. Studies validating the GALN against 3D-CGA kinematic curves have confirmed:

• Sagittal Plane Concordance: The GALN demonstrates high concurrent validity for sagittal plane movements. Observational ratings of knee hyperextension, absent initial heel contact, and reduced hip extension show strong concordance (sensitivity > 75%, specificity > 80%) when benchmarked against continuous 3D kinematic trajectories.
• Coronal and Transverse Plane Concordance: In line with general findings in human visual biomechanics, items measuring coronal and transverse plane deviations (such as pelvic rotation or subtle tibial internal/external rotation) show lower agreement with 3D-CGA (sensitivity ranging from 55% to 70%). These findings reflect the inherent physical challenge of estimating rotational alignment from standard two-dimensional clinical vantage points.

Construct and Discriminant Validity

The GALN reliably distinguishes between healthy control subjects (who score zero or near-zero across deviation items) and clinical cohorts presenting with known locomotor deficits (e.g., patients with unilateral hip osteoarthritis, stroke hemiparesis, or chronic ankle instability). Furthermore, total deviation scores correlate significantly with functional mobility scales, including the Timed Up and Go (TUG) test, self-selected walking speed, and the Functional Ambulation Categories (FAC).

8. Reliability

The diagnostic utility of an observational gait scale depends on its inter-rater and intra-rater reliability. Brunnekreef and colleagues extensively evaluated these psychometric attributes across various observer experience levels.

Inter-Rater Reliability

Inter-rater agreement for the 26 items of the GALN varies depending on the joint complex, the specific plane of movement, and whether video playback or real-time observation is employed:

• Overall Inter-Rater Reliability: Cohen’s kappa (κ) coefficients for individual items range from fair to substantial (κ = 0.35 to 0.82). The median kappa across all items settles at approximately κ = 0.58.
• Sagittal Plane Reliability: Items evaluating sagittal plane events at the ankle (e.g., initial contact characteristics, foot drop) and knee (e.g., loading response flexion wave, mid-stance hyperextension) exhibit substantial reliability (κ = 0.65 to 0.84; percentage agreement > 85%).
• Transverse and Frontal Plane Reliability: Items assessing pelvic rotation and subtle trunk tilting yield moderate agreement (κ = 0.38 to 0.55), confirming that multi-planar rotation is more challenging to rate consistently via visual inspection.
• Experience Effects: Experienced physical therapists who completed a standardized calibration training module achieved significantly higher inter-rater concordance (κ ≥ 0.68) compared to untrained clinicians (κ ≤ 0.44), underscoring the necessity of familiarization with the tool’s standardized operational definitions.

Intra-Rater (Test-Retest) Reliability

Intra-rater reliability across repeated reviews of recorded video material demonstrates high temporal consistency. Individual item test-retest kappa statistics consistently exceed κ = 0.70, with overall percentage agreement commonly surpassing 90%. When evaluating the total cumulative deviation score, the intraclass correlation coefficient (ICC) exceeds 0.85, confirming that individual raters maintain a stable diagnostic threshold across repeated administrations.

9. Factor Analysis and Structural Dimensionality

Because the GALN was constructed from biomechanical gait theory rather than psychological latent trait modeling, its internal structure has been investigated through exploratory and confirmatory factor analytic techniques adapted for categorical clinical indicators.

Biomechanical Task Clustering

Factor analyses conducted on observational gait parameters derived from the GALN reveal multidimensional constructs that align closely with Perry’s locomotor tasks rather than isolated anatomical joints. Exploratory factor analyses (EFA) utilizing tetrachoric correlation matrices typically yield a 3- to 4-factor solution accounting for over 60% of the total variance:

• Factor 1: Weight Acceptance and Shock Absorption Stability: Heavy factor loadings from items evaluating initial contact (Item 1), loading response knee flexion (Item 2), and contralateral pelvic drop (Item 3). This dimension reflects the neuromuscular capacity to decelerate and stabilize the limb immediately following ground impact.
• Factor 2: Single Limb Support and Stance Phase Progression: Dominated by items measuring mid-stance trunk lean, terminal stance hip extension, and stance-phase knee stability (Items 4, 5, 6, and 7). This factor reflects balance maintenance and the generation of forward propulsion.
• Factor 3: Swing Phase Limb Clearance and Advancement: Characterized by strong loadings on swing-phase knee flexion, ankle dorsiflexion clearance, and forward pelvic rotation (Items 8, 9, 10, 11, and 12). This represents the spatial execution of non-weight-bearing limb transit.
• Factor 4: Postural and Axial Trunk Control: Moderately loaded by bilateral trunk deviation items throughout both stance and swing periods, reflecting systemic postural equilibrium.

Confirmatory factor analytic (CFA) models testing this three-to-four task-based structure demonstrate adequate goodness-of-fit indices (Comparative Fit Index [CFI] > 0.90; Root Mean Square Error of Approximation [RMSEA] < 0.06), verifying that gait deviations do not occur at random, but cluster according to underlying functional control mechanisms.

10. Instrument / Measurement Tool

The Nijmegen Gait Analysis List is structured as a standardized clinical checklist with explicit observational guidelines.

• Instrument Type: Performance-based observational rating checklist.
• Administration Format: Direct real-time clinical visual observation or retrospective video-recorded playback analysis. (High-definition two-dimensional video recording from both sagittal and frontal viewpoints is recommended to optimize measurement precision).
• Target Population: Children, adolescents, adults, and elderly individuals exhibiting musculoskeletal, orthopedic, or neurological gait impairments.
• Item Configuration: 26 total items structured as 13 identical biomechanical items assessed bilaterally (13 items for the left lower extremity; 13 items for the right lower extremity).
• Anatomical Coverage:
  • Trunk (lateral and sagittal deviations)
  • Pelvis (tilt, drop, and rotation)
  • Hip (flexion, extension, adduction, abduction, internal/external rotation)
  • Knee (shock absorption flexion, extension stability, swing flexion)
  • Ankle and Foot (initial contact pattern, dorsiflexion clearance, push-off, supination/pronation)
• Response and Scoring Scale:
  • Each item is rated primarily on a dichotomous or categorical ordinal scale indicating deviation status.
  • 0 = Normal / Expected biomechanical pattern (no clinical deviation).
  • 1 = Abnormal / Pathological deviation present (clear visual departure from normative kinematics).
  • Certain sub-items accommodate directional indicators (e.g., excessive internal vs. external rotation, or excessive flexion vs. hyperextension).
• Scoring and Interpretation:
  • Items are totaled per limb (ranging from 0 to 13 per side) and combined for an overall gait deviation score (ranging from 0 to 26).
  • A score of 0 indicates a completely normative, deviation-free gait profile.
  • Higher aggregate scores reflect a greater degree of multi-segmental gait pathology.
  • Clinically, individual item scores provide qualitative guidance for specific therapeutic targets (e.g., prescribed orthotic bracing for an item indicating absent swing dorsiflexion).
• Administration Time: Approximately 10 to 15 minutes for live observational assessment, or 15 to 20 minutes when performing slow-motion video analysis.

11. Permissions, Fee, and Test Year

The Nijmegen Gait Analysis List was formally documented and published by Jaap J. Brunnekreef and Cor J.T. van Uden in 2004 through Dutch physiotherapy literature and specialized clinical measurement documentation at the Sint Maartenskliniek.

• Copyright Status: The conceptual methodology and scoring framework were developed for clinical and educational enhancement within physical therapy practice. The original Dutch documentation, instructional forms (Toelichtingsformulier), and assessment scoring sheets (Meetinstrument) are protected under institutional copyright by the authors and the Sint Maartenskliniek.
• Access and Licensing: The tool is intended for open clinical and research use by licensed healthcare professionals and qualified academic researchers. The forms have historically been distributed via Dutch clinical guideline networks (such as the Royal Dutch Society for Physical Therapy / KNGF measurement tool repositories) for non-commercial diagnostic purposes.
• Commercial Use: Integration into commercial software packages, electronic medical records, or proprietary digital motion-analysis systems requires formal permission from the primary authors and the Sint Maartenskliniek administration.

12. References

The following publications document the development, clinical application, and psychometric validation of the GALN and related observational gait assessment frameworks:

• Brunnekreef, J. J., & van Uden, C. J. T. (2004). Ganganalyselijst Nijmegen (GALN): Handleiding en toelichtingsformulier [Nijmegen Gait Analysis List: Manual and explanation form]. Afdeling Fysiotherapie en Wetenschappelijk Onderzoek, Sint Maartenskliniek Nijmegen.
• Brunnekreef, J. J., van Uden, C. J. T., van Moorsel, S., & Kooloos, J. G. M. (2005). Reliability of videotaped observational gait analysis in patients with orthopedic lower limb impairments. BMC Musculoskeletal Disorders, 6(1), Article 17. https://doi.org/10.1186/1471-2474-6-17
• Perry, J., & Burnfield, J. M. (2010). Gait analysis: Normal and pathological function (2nd ed.). SLACK Incorporated.
• van Uden, C. J. T., & Besser, M. P. (2004). Test-retest reliability of temporal and spatial gait characteristics measured with an instrumented walkway system (GAITRite®). BMC Musculoskeletal Disorders, 5(1), Article 13. https://doi.org/10.1186/1471-2474-5-13
• Wren, T. A., Rethlefsen, S. A., & Kay, R. M. (2005). Validity of visual gait analysis. Journal of Pediatric Orthopaedics, 25(6), 841–842. https://doi.org/10.1097/01.bpo.0000184648.54413.25
• Toro, B., Nester, C. J., & Farren, P. C. (2003). The status of visual gait analysis: A review of the literature. Gait & Posture, 18(2), 171–186. https://doi.org/10.1016/S0966-6362(02)00159-4
• Rathinam, C., Bateman, A., Peirson, J., & Skinner, J. (2014). Observational gait measurement tools in children with neurological conditions: A systematic review. Pediatric Physical Therapy, 26(4), 400–418. https://doi.org/10.1097/PEP.0000000000000087

13. Items of the Scale