The Community Balance and Mobility Scale (CBMS)

1. Abstract

The Community Balance and Mobility Scale (CB&M or CBMS) is an internationally recognized, clinician-administered performance measure developed to evaluate high-level balance, dynamic postural control, and mobility in ambulatory individuals who exhibit persistent motor impairments following traumatic brain injury (TBI), stroke, and other neurological conditions. Historically, traditional functional balance instruments, most notably the Berg Balance Scale (BBS), have suffered from pronounced ceiling effects when applied to community-dwelling, independent ambulatory patients, thereby failing to capture subtle yet debilitating dynamic balance deficits required for safe community reintegration. The CB&M addresses this critical measurement gap by incorporating complex, multi-directional, multi-tasking motor challenges that emulate real-world physical demands, including dual-task walking, rapid directional transitions, running with controlled stopping, lateral dodging, hopping, and visual-vestibular integration tasks.

Comprising 13 performance tasks (with six items administered bilaterally, yielding 19 total scored operational items), the instrument is scored across a graduated ordinal scale from 0 to 5 per task, with a bonus point available on the stair-descent item, establishing a maximum cumulative score of 96 points. Psychometric evaluations demonstrate exceptional measurement properties. The CB&M exhibits outstanding internal consistency (Cronbach’s alpha = 0.96), exemplary inter-rater reliability (intraclass correlation coefficient [ICC] = 0.977), intra-rater reliability (ICC = 0.977), and robust test-retest reliability across both immediate (ICC = 0.975) and 5-day intervals (ICC = 0.898). Construct and convergent validity are supported through significant associations with laboratory-based spatiotemporal gait parameters (e.g., walking velocity, step length, step width variability), self-reported balance confidence via the Activities-specific Balance Confidence (ABC) Scale (r = 0.60), and real-world participation measures including the Community Integration Questionnaire (CIQ; r = 0.54). Moreover, Rasch measurement analyses confirm the unidimensionality of the construct while mapping items across a wide calibration spectrum, establishing the CB&M as a gold-standard clinical assessment tool for ambulatory rehabilitation.

2. Keywords

Community Balance and Mobility Scale, CB&M, traumatic brain injury, dynamic postural control, neurorehabilitation, outcome measurement, psychometrics, gait velocity, visual-vestibular integration, ceiling effect

3. Authors

The Community Balance and Mobility Scale was developed by a multidisciplinary team of physical therapy researchers, clinical specialists, and biostatisticians affiliated with the Toronto Rehabilitation Institute (now part of the University Health Network) and the University of Toronto, Ontario, Canada.

• Jo-Anne Howe, PT, MSc: Toronto Rehabilitation Institute, University Health Network, Toronto, Ontario, Canada. Email: [email protected]
• Elizabeth L. Inness, PT, PhD: KITE Research Institute, Toronto Rehabilitation Institute, University Health Network; Department of Physical Therapy, University of Toronto, Ontario, Canada. Email: [email protected]
• A. Venturini, PT, MSc: Department of Physical Therapy, University of Toronto; Toronto Rehabilitation Institute, Toronto, Ontario, Canada.
• J. Ivan Williams, PhD: Department of Health Policy, Management and Evaluation, University of Toronto; Institute for Work & Health, Toronto, Ontario, Canada.
• Molly C. Verrier, DipP&OT, MHSc: Department of Physical Therapy, Graduate Department of Rehabilitation Science, University of Toronto, Ontario, Canada.

4. Purpose

Traumatic brain injury represents a leading cause of worldwide morbidity and long-term disability, with hundreds of thousands of new cases recorded annually. A substantial proportion of individuals sustaining mild, moderate, or severe TBI achieve independent community ambulation prior to rehabilitation discharge. However, standard clinical balance evaluations often mask underlying sensorimotor deficits. The primary purpose of the Community Balance and Mobility Scale is to provide a rigorous, objective, and sensitive standardized metric designed specifically to quantify subtle impairments in dynamic equilibrium, motor coordination, and functional mobility in higher-functioning ambulatory individuals attempting full community reintegration.

Prior to the establishment of the CB&M, clinicians frequently utilized tools such as the Berg Balance Scale, the Timed Up and Go (TUG) test, and the Dynamic Gait Index. While these measures demonstrate exceptional diagnostic and discriminative utility in frail, elderly, or significantly impaired populations, they exhibit pronounced ceiling effects among active adolescents and adults with neurological trauma. A patient recovering from TBI may attain a perfect score of 56/56 on the Berg Balance Scale yet remain utterly incapable of crossing a busy multi-lane street, navigating uneven terrain, dodging unexpected pedestrians in a crowded shopping mall, or running to catch public transportation. Inability to execute these high-level motor functions leads to premature social isolation, secondary physical deconditioning, loss of employment, and elevated fall risk during unpredictable environmental interactions.

The CB&M was engineered to systematically challenge the postural control system at the limits of biomechanical stability. The operational rationale centers on evaluating the interaction between speed, complex base-of-support alterations, rapid directional changes, visual tracking during movement, and divided attention. Clinical applications encompass initial rehabilitation triage, baseline dynamic postural profiling, sensitive longitudinal tracking during post-acute outpatient therapy, and programmatic discharge evaluation. In research settings, the CB&M serves as a primary outcome measure for clinical trials evaluating novel neurorehabilitation protocols, aerobic exercise regimens, virtual reality interventions, and pharmacotherapy targeting dynamic motor recovery in traumatic brain injury, stroke, vestibular disorders, and cerebral palsy.

5. Psychological Construct

The core construct operationalized by the CB&M is high-level dynamic postural control and community-level mobility. Rather than treating balance as a static state, modern psychometrics and motor control theory conceptualize equilibrium as a multi-system, task-specific regulatory dynamic requiring the continuous integration of sensory input, biomechanical stabilization, anticipatory motor planning, and cognitive-motor execution. The scale operationalizes this multifaceted construct through several core clinical and behavioral dimensions:

Dynamic Center-of-Mass (COM) Control Over a Reduced Base of Support (BOS)

Community ambulation necessitates movement execution over narrow or single-limb contact areas. Tasks such as timed unilateral stance, tandem walking, and forward hopping evaluate the participant’s capacity to maintain the projection of their center of mass safely within an unstable, rapidly shifting base of support. For example, during single-limb hopping, the motor system must generate sufficient vertical and forward propulsion while rapidly attenuating ground reaction forces upon landing without losing equilibrium or touching down the contralateral limb.

Anticipatory and Reactive Postural Adjustments

Effective community mobility requires proactive motor strategies before movement initiation (anticipatory postural adjustments, or APAs) and reactive stabilization following self-induced perturbations. The CB&M tasks—such as lateral foot scooting, rapid lateral dodging, and sudden transitions between forward and backward walking—require individuals to dynamically modulate braking forces, adjust segment momentum, and stabilize the trunk against inertial disruptions. Failure in APA programming manifests as hesitation, loss of movement fluidity, or compensatory upper-extremity reaching.

Visual-Vestibular-Somatosensory Integration

In community environments, gaze stability and head rotation must frequently occur independently of locomotion trajectory. The CB&M directly tests this through dynamic tasks such as “walking and looking over the shoulder” and “walk, look, and carry.” In these tasks, the patient must maintain linear forward progression while executing voluntary horizontal head and ocular rotations toward visual targets. This forces central integration of vestibulo-ocular reflexes (VOR), dynamic cervical proprioception, and peripheral somatosensory feedback, unmasking vestibular hypofunction or visual-motion sensitivity common in post-concussion and TBI populations.

High-Velocity Deceleration and Motor Agility

Most clinical balance tests evaluate self-selected, comfortable walking speeds. The CB&M directly challenges the cardiovascular and neuromuscular reserves of the patient through tasks requiring maximal safe velocity followed by instantaneous, controlled deceleration (e.g., “running with controlled stop”). Stopping abruptly on a specified floor marker demands profound eccentric quadriceps torque, core stability, and rapid posterior COM displacement, replicating the sudden motor demands encountered when responding to traffic signals or urban obstacles.

Dual-Tasking and Cognitive-Motor Interference

Everyday mobility rarely occurs in the absence of secondary manual or cognitive tasks. The inclusion of complex manual manipulation during gait (such as carrying a weighted basket while descending stairs or carrying a loaded tray while scanning the surrounding environment) introduces dual-task demands. In individuals with brain injury, cognitive-motor interference often degrades gait automaticity, resulting in dramatically lowered gait velocity, asymmetric step times, and heightened risk of stumbling.

6. Theoretical Framework

The conceptual foundation of the Community Balance and Mobility Scale is grounded in the Systems Approach to Postural Control, pioneered by biomechanist Nikolai Bernstein and extensively advanced within contemporary neurorehabilitation by Anne Shumway-Cook and Marjorie Woollacott. Under the Systems Approach, postural control is not viewed merely as a collection of hardwired, static brainstem reflexes (the classical hierarchical model). Instead, balance emerges dynamically from the complex self-organization of multiple interacting neurobiological subsystems, constrained by the biomechanical attributes of the human body, the specific goals of the behavioral task, and the physical characteristics of the surrounding environment.

Ecological Dynamics and Environmental Affordances

Complementing the Systems Model, the CB&M draws upon ecological psychology and James J. Gibson’s theory of affordances. In ecological motor control, movement is understood as an ongoing perception-action coupling: individuals continuously perceive opportunities for action within their environment (such as the traversability of a curb, the clearance needed to bypass an obstacle, or the compliance of a surface). Following traumatic brain injury, patients experience distorted perceptual mapping, delayed sensorimotor processing times, and compromised motor execution. The CB&M simulates dynamic ecological constraints within a structured clinical paradigm by manipulating environmental boundaries (e.g., designated floor lines, target circles, step heights, and temporal constraints).

The Dynamic Interaction Model: Individual, Task, and Environment

Within this theoretical framework, the CB&M specifically manipulates task constraints along three primary axes:

• Velocity Demands: Transitioning from steady-state self-selected walking to maximal burst speed, acceleration, and rapid deceleration.
• Multi-Planar Degrees of Freedom: Forcing movement through the sagittal, frontal, and transverse planes simultaneously (e.g., the 180° pivot turn and lateral dodging).
• Sensory Conflict: Intentionally dissociating focal visual direction from the vector of physical locomotion (e.g., looking backward over the shoulder while walking forward).

By systematically varying these parameters, the CB&M exposes motor deficits that remain completely undetectable under resting, single-task, or linear steady-state conditions, providing a robust operationalization of community-level dynamic equilibrium.

7. Validity

The psychometric development of the Community Balance and Mobility Scale involved rigorous empirical validation protocols spanning content, construct, convergent, discriminant, and predictive paradigms.

Content and Face Validity

Content validity was established through an iterative, collaborative consensus process engaging both expert clinical panels and patients recovering from traumatic brain injury. During scale development, 17 specialized physical therapists and clinical neuroscientists reviewed proposed functional motor challenges, refining scoring descriptors to reflect precise biomechanical deviations (such as trunk tilt, foot placement errors, guard contacts, and loss of momentum). In parallel, 7 ambulatory individuals with traumatic brain injury were engaged to evaluate the ecological relevance, perceived difficulty, and face validity of the activities, ensuring the tasks accurately mirrored the real-world obstacles encountered in independent community living.

Construct and Convergent Validity

Construct validity has been verified across multiple biomechanical, self-report, and clinical outcome benchmarks. In the seminal validation study by Inness et al. (2011), CB&M scores were correlated with objective three-dimensional, laboratory-based spatiotemporal gait parameters in individuals with traumatic brain injury:

• Walking Velocity: Highly significant positive correlation with self-selected and fast gait speed (r values ranging from 0.64 to 0.87, p < 0.001).
• Step Length and Step Width: Moderate-to-high correlations with average step length (r = 0.60 to 0.78) and base of support modulation.
• Gait Variability: Marked inverse correlations were demonstrated between CB&M scores and step time variability as well as step length variability (r = -0.46 to -0.70). Because heightened spatiotemporal variability serves as a primary biomechanical marker of dynamic postural instability, these correlations strongly confirm the construct validity of the scale.

Convergent validity with patient-reported outcomes was established via significant correlations with the Activities-specific Balance Confidence (ABC) Scale (r = 0.60, p < 0.01), illustrating that an individual’s physical performance on the CB&M directly mirrors their subjective psychological confidence in navigating complex public environments. Furthermore, performance was positively correlated with the Community Integration Questionnaire (CIQ; r = 0.54, p < 0.01), confirming that higher dynamic postural scores correspond to greater vocational, domestic, and social reintegration.

Ceiling Effect Mitigation and Discriminant Validity

A primary psychometric advantage of the CB&M over conventional instruments is its resolution of ceiling effects. In direct comparative clinical trials involving ambulatory TBI cohorts, over 30% to 50% of participants achieved the maximum possible score (56/56) on the Berg Balance Scale. When these identical individuals were evaluated using the CB&M, their scores distributed across a broad, normal continuum (spanning 40 to 85 out of 96), demonstrating high sensitivity to subtle motor deficits. Additionally, Zbarsky et al. (2010) collected age-stratified normative data across healthy populations spanning the second to eighth decades of life, demonstrating that the CB&M captures physiological balance declines associated with normal human aging and reliably differentiates healthy normative performers from brain injury survivors.

8. Reliability

The Community Balance and Mobility Scale exhibits exemplary reliability indices across internal consistency, inter-rater concordance, intra-rater reproducibility, and longitudinal test-retest assessments.

Internal Consistency

The scale demonstrates remarkable internal consistency. In the foundational validation investigation conducted by Howe et al. (2006), Cronbach’s alpha was calculated at 0.96 for the complete 19-item battery. This elevated coefficient confirms that the items function coherently as an integrated operational metric of high-level dynamic balance without displaying excessive item redundancy.

Inter-Rater and Intra-Rater Reliability

Because the CB&M involves direct clinical observation of dynamic, rapid-velocity movements, establishing high observer concordance was paramount. Evaluators trained on standardized operational criteria demonstrated extraordinary reliability:

• Inter-Rater Reliability: Intraclass correlation coefficient (ICC, model 2,1) = 0.977 (95% CI: 0.95–0.99), indicating near-perfect measurement agreement across different trained physical therapists evaluating the same patient cohort simultaneously.
• Intra-Rater Reliability: Intraclass correlation coefficient (ICC) = 0.977 across repeated video-recorded evaluations scored by the same clinician at distinct time points.

Test-Retest Stability and Measurement Error

Test-retest stability was rigorously examined across distinct time horizons:

• Immediate Same-Day Test-Retest: ICC = 0.975, showing exceptional measurement consistency when physiological fatigue and recovery fluctuations are controlled.
• 5-Day Test-Retest Reliability: ICC = 0.898 to 0.92, confirming that the scale provides stable baseline measurement across clinically realistic pre-intervention observation windows.
• Standard Error of Measurement (SEM) & Minimal Detectable Change (MDC): The SEM for the CB&M has been established between 2.7 and 3.5 points on the 96-point scale. The Minimal Detectable Change at the 95% confidence level (MDC₉₅) is approximately 7.5 to 9.6 points in TBI and post-stroke populations, providing clinicians with clear mathematical thresholds to differentiate genuine therapeutic functional recovery from measurement error.

9. Factor Analysis

The internal structural validity, dimensional hierarchy, and item calibration of the CB&M have been subjected to extensive psychometric exploration utilizing both Classical Test Theory (factor analysis) and Modern Psychometrics (Rasch Measurement Theory).

Unidimensionality and Factor Structure

Exploratory and confirmatory factor analytic studies, alongside principal component analysis (PCA) of residuals under Rasch models, have demonstrated that the CB&M is essentially unidimensional. A predominant single factor—representing integrated dynamic balance and complex ambulatory mobility—accounts for the vast majority of variance across test items (variance explained typically exceeding 60% to 70%). The ratio of the first eigenvalue to the second eigenvalue consistently exceeds 4:1, confirming the structural validity of summing individual task items into a single, composite 96-point summary score.

Rasch Analysis and Item Hierarchy

Subsequent psychometric investigations (e.g., Knorr et al., Inness et al.) applied Rasch rating scale models to assess item difficulty hierarchies, category ordering, and targeting in individuals recovering from neurological trauma. Key findings include:

• Item Infit and Outfit Statistics: Mean square infit and outfit statistics for nearly all items fall comfortably within the acceptable psychometric window of 0.6 to 1.4, indicating good fit to the underlying latent construct without disruptive noise or construct underrepresentation.
• Item Difficulty Continuum: Rasch item calibrations reveal that the CB&M spans a broad difficulty spectrum. The easiest items (calibrated at the lower logit spectrum) include basic linear challenges like “Tandem Walking” and static “Unilateral Stance.” The most challenging items (occupying the highest logit calibrations) include “Running with Controlled Stop,” “Lateral Dodging” at rapid pace, “Forward Hopping” on the affected lower extremity, and “Descending Stairs” under dual-task carrying conditions.
• Targeting: Person-item targeting maps indicate excellent alignment with community-dwelling ambulatory patients, completely resolving the low ceiling threshold that degrades ordinal scales like the Berg Balance Scale.

10. Instrument / Measurement Tool

The Community Balance and Mobility Scale is an objective, clinician-administered, direct-observation performance assessment. It requires standard physical therapy testing equipment and a designated, unobstructed ambulatory track.

Test Overview & Specifications

• Instrument Type: Performance-based clinical rating scale / observational functional mobility assessment.
• Administration Format: Direct clinician observation; patient performs standardized physical movements while the clinician scores biomechanical execution, speed, trajectory, and compensatory balance strategies in real time.
• Target Population: Ambulatory individuals with balance impairments following traumatic brain injury, stroke, vestibular pathology, mild-to-moderate cerebral palsy, or active older adults. (Patients must be capable of independent ambulation without physical contact assistance for at least 8 to 10 meters; orthoses are permitted, but ambulatory assistive devices like canes or walkers are generally not utilized during standard testing).
• Number of Core Functional Tasks: 13 distinct motor tasks.
• Number of Scored Items: 19 operational test items (Tasks 1, 4, 5, 8, 10, and 13 are performed bilaterally to assess both right and left extremities/sides).
• Scoring Range per Item: Ordinal scale ranging from 0 (complete inability to perform or requiring physical assist/loss of balance) to 5 (flawless execution meeting all speed, alignment, and biomechanical criteria). Task 12 (descending stairs) includes an additional +1 bonus point for carrying a loaded basket without handrail contact.
• Cumulative Score Range: 0 to 96 points (where higher scores indicate superior dynamic balance and community mobility).
• Time Required: Approximately 20 to 30 minutes for complete administration and scoring.

Required Clinical Equipment and Testing Environment

• An unobstructed 8-meter (or 10-meter) walking track marked on a flat, non-slip floor surface using high-visibility adhesive tape.
• A visible visual target (e.g., a colored circle or alphanumeric symbol, 20 cm in diameter) mounted at eye level at the end of the walking track.
• A stopwatch capable of measuring to the nearest one-hundredth of a second.
• A standard visual scanning target mounted to the side of the track for visual-vestibular tracking tasks.
• A flight of standard stairs (minimum 6–8 steps) equipped with a secure handrail.
• A standardized weighted basket or laundry container (approx. 2.0 to 2.5 kg) for carrying tasks.
• A low step or aerobic step platform (height approximately 15–20 cm) for the rapid step-up task.

Standardized Scoring Protocol

• Single Attempt Rule: Most tasks allow only one trial unless the patient misinterprets instructions, in which case a single standardized re-instruction and repetition is permitted.
• Orthoses vs. Assistive Devices: Ankle-foot orthoses (AFOs) are permitted if regularly worn for community mobility; however, no physical assistance, walking canes, crutches, or rolling walkers are allowed during official scoring.
• Scoring Hierarchy (General 0–5 Continuum):
  • Score 0: Unable to perform the task safely; requires physical contact/guarding; excessive stumbling or loss of balance.
  • Score 1–2: Severely impaired execution; major deviations in movement trajectory; profound compensatory trunk lean; failed speed or duration thresholds.
  • Score 3–4: Moderate performance; meets fundamental task objectives but exhibits mild asymmetry, brief hesitation, compensatory arm posturing, or slight speed deficits.
  • Score 5: Optimal performance; moves smoothly with rapid execution, proper biomechanical alignment, excellent dynamic stability, and complete absence of compensatory strategies.

11. Permissions & Fee and Test Year

The Community Balance and Mobility Scale was originally published in 2006 by Jo-Anne Howe and colleagues through research conducted at the Toronto Rehabilitation Institute in Toronto, Canada. The instrument, its standardized scoring guidelines, and the accompanying comprehensive clinical administration manual are protected by copyright held by the Toronto Rehabilitation Institute – University Health Network (UHN).

For non-commercial clinical practice, academic education, and investigator-initiated non-commercial health research, the CB&M administration manual and score sheets are widely made accessible by the developing team without licensing fees, provided that appropriate bibliographic citation is maintained and the standardized testing guidelines are preserved without modification. Clinical systems, electronic health record (EHR) vendors, and commercial clinical trials seeking to incorporate the CB&M into proprietary digital platforms or commercialized assessment batteries should contact the Toronto Rehabilitation Institute / University Health Network Technology Transfer and Commercialization Office for licensing agreements and permission.

12. References

The psychometric properties and clinical utility of the Community Balance and Mobility Scale are documented across the following peer-reviewed literature:

• Howe, J. A., Inness, E. L., Venturini, A., Williams, J. I., & Verrier, M. C. (2006). The Community Balance and Mobility Scale–a balance measure for individuals with traumatic brain injury. Clinical Rehabilitation, 20(10), 885–895. https://doi.org/10.1177/0269215506070688
• Inness, E. L., Howe, J. A., Niechwiej-Szwedo, E., Jaglal, S. B., McIlroy, W. E., & Verrier, M. C. (2011). Measuring balance and mobility after traumatic brain injury: Validation of the Community Balance & Mobility Scale (CB&M). Physiotherapy Canada, 63(2), 199–208. https://doi.org/10.3138/ptc.2009-54
• Knorr, S., Brouwer, B., & Garland, S. J. (2010). Validity of the Community Balance and Mobility Scale in individuals with chronic stroke. Physical Therapy, 90(4), 527–537. https://doi.org/10.2522/ptj.20090233
• Zbarsky, K., Parsley, D., Clegg, H., Welch, T., Fernandes, C., Jaglal, S. B., Inness, E. L., Williams, J. I., McIlroy, W. E., & Howe, J. A. (2010). Community Balance & Mobility Scale (CB&M): Age-related reference values. Physiotherapy Canada, 62(Suppl 1), 46–47.
• Wright, F. V., Ryan, J., & Brewer, K. (2010). Reliability and validity of the Community Balance and Mobility Scale in young people with cerebral palsy. Physical & Occupational Therapy in Pediatrics, 30(4), 296–311. https://doi.org/10.3109/01942638.2010.493097
• Shumway-Cook, A., & Woollacott, M. H. (2017). Motor Control: Translating Research into Clinical Practice (5th ed.). Wolters Kluwer.
• Powell, L. E., & Myers, A. M. (1995). The Activities-specific Balance Confidence (ABC) Scale. The Journals of Gerontology: Series A, 50A(1), M28–M34. https://doi.org/10.1093/gerona/50a.1.m28
• Willer, B., Rosenthal, M., Kreutzer, J. S., Gordon, W. A., & Rempel, R. (1993). Assessment of community integration following rehabilitation for traumatic brain injury. The Journal of Head Trauma Rehabilitation, 8(2), 75–87.

13. Items of the Scale