Modified Postural Assessment Scale for Stroke

1. Abstract

The Modified Postural Assessment Scale for Stroke (mPASS) is a specialized, clinician-rated performance instrument engineered to evaluate postural control and balance equilibrium among individuals recovering from acute, subacute, or chronic cerebrovascular accidents. Originating as a psychometrically refined adaptation of the original 12-item Postural Assessment Scale for Stroke (PASS) developed by Benaim and colleagues, the mPASS was systematically optimized by Persson and collaborators (2011) to eliminate ambiguities in task administration, clarify behavioral scoring thresholds, and enhance inter-rater and intra-rater concordance. The scale comprises 12 discrete functional items categorized into two core dimensions: Maintaining a Posture (evaluating static balance across lying, sitting, and standing postures, specifically discerning paretic versus non-paretic lower extremity weight-bearing) and Changing a Posture (assessing dynamic transitions, including bed mobility, supine-to-sitting, sitting-to-standing, and reaching downward from an upright stance).

Each item is appraised along a standardized 4-point ordinal rating scale ranging from 0 (inability to execute the task) to 3 (independent execution with optimal biomechanical stability), yielding a cumulative sum score between 0 and 36 points. Unlike classical clinical balance batteries such as the Berg Balance Scale (BBS), the mPASS incorporates vital bed-mobility protocols (e.g., supine to lateral decubitus, rolling toward the affected and unaffected sides), ensuring robust discriminative sensitivity and negligible floor effects among severely impaired, non-ambulatory stroke patients. Extensive psychometric investigations corroborate its high internal consistency (Cronbach’s alpha typically ranging from .95 to .98), exceptional inter-rater reliability (intraclass correlation coefficients [ICC] exceeding .90), and robust convergent validity with functional independence indices. Consequently, the mPASS serves as a gold-standard diagnostic, prognostic, and evaluative tool across clinical neurorehabilitation pipelines.

2. Keywords

Modified Postural Assessment Scale for Stroke, mPASS, postural balance, stroke rehabilitation, cerebrovascular accident, psychometrics, static postural control, dynamic balance, bed mobility, neurorehabilitation

3. Authors

The conceptual modification and systematic psychometric optimization of the Modified Postural Assessment Scale for Stroke (mPASS) were spearheaded by Christina Unger Persson and her multidisciplinary clinical research team at the Institute of Neuroscience and Physiology, Department of Clinical Neuroscience and Rehabilitation, Sahlgrenska Academy, University of Gothenburg, Sweden, alongside the Department of Physical Therapy, Sahlgrenska University Hospital, Gothenburg, Sweden.

The primary contributors to the development, structural modification, and empirical validation of the instrument include:

• Christina Unger Persson, PT, PhD: Lead investigator, clinical physical therapist, and neurorehabilitation researcher specializing in early postural control metrics and post-stroke recovery trajectories.
• Katharina S. Sunnerhagen, MD, PhD: Senior professor of rehabilitation medicine, Sahlgrenska Academy, University of Gothenburg, focusing on longitudinal outcomes and motor epidemiology following stroke.
• Anna Danielsson, PT, PhD: Associate professor and clinical researcher investigating motor learning, balance disturbances, and physiological biomechanics in central nervous system pathologies.

Subsequent cross-cultural adaptations and validation paradigms have emerged across global clinical settings, including the Dutch adaptation published by R. Brouwer and colleagues in 2019, expanding the standardized cross-linguistic utility of the mPASS across European rehabilitation networks.

4. Purpose

Postural instability represents one of the most debilitating sensorimotor sequelae following a stroke, predisposing survivors to secondary physical injuries, fall-related fractures, prolonged immobilization, and persistent deficits in activities of daily living (ADLs). While established instruments such as the Berg Balance Scale (BBS) and the Fugl-Meyer Assessment provide critical insights into motor functioning, they frequently suffer from pronounced floor effects when administered to severely impaired inpatients during the hyperacute or acute stages of post-stroke recovery. Many conventional balance scales require the patient to maintain independent unassisted standing or perform complex seated tasks, rendering them unfeasible for individuals exhibiting profound hemiparesis, severe trunk asymmetry, or flaccid paralysis.

The Modified Postural Assessment Scale for Stroke (mPASS) was purposefully formulated to resolve these persistent measurement dilemmas. Grounded in clinical utility and measurement precision, its primary objectives encompass:

• Granular Evaluation of Severely Impaired Patients: By incorporating foundational bed-mobility elements (e.g., rolling onto the affected and unaffected sides) and supported sitting, the mPASS eradicates the administrative floor effect, reliably mapping postural competence even in individuals unable to achieve upright standing.
• Hemispheric Differentiation: The instrument specifically dissociates and contrasts functional performance between the paretic (affected) and non-paretic (unaffected) lower extremities. Items assessing unipedal stance explicitly demand isolated weight-bearing on each limb independently, providing crucial diagnostic insight into unilateral spatial neglect, hemispatial postural asymmetry, and compensatory weight shifts.
• Elimination of Administrative Ambiguity: While the original 12-item PASS pioneered by Benaim et al. (1999) represented a breakthrough, ambiguous scoring criteria occasionally compromised examiner agreement regarding manual physical assistance versus environmental support. Persson et al. (2011) re-engineered the scoring descriptions to establish unyielding, objective behavioral thresholds, standardizing administration protocols across acute and subacute clinical teams.
• Prognostic Utility and Longitudinal Tracking: The mPASS is widely utilized as an early predictive biomarker. Baseline postural scores gathered via the mPASS during the initial days post-stroke demonstrate high predictive capacity regarding prospective ambulatory independence, community reintegration, and ultimate functional motor recovery at six to twelve months post-insult.

5. Psychological Construct

The psychological, neurophysiological, and behavioral construct measured by the mPASS is postural control—defined within motor control paradigms as the act of maintaining, achieving, or restoring a state of balance during any posture or activity. Postural control is not a monolithic reflex; rather, it is a complex behavioral construct emerging from the seamless interaction of sensory integration (vestibular, visual, and proprioceptive inputs), central neurological processing, musculoskeletal alignment, and task-specific neuromuscular execution. In the context of the mPASS, this overarching construct is operationalized through two core clinical dimensions:

1. Maintaining a Posture (Static Postural Control)

Static postural control refers to the neurobehavioral capacity to maintain the bodily center of mass (COM) within the defined base of support (BOS) under minimal external perturbation. In the mPASS, this dimension contains 5 standardized items that systematically elevate biomechanical challenge:

• Sitting without support: Demands adequate tonic activation of the paraspinal and abdominal core musculature, counteracting gravitational collapse without reliance on upper extremity armrests or posterior backrests.
• Standing with support: Evaluates the basic capacity to recruit lower extremity extensor synergies to sustain an upright posture when external environmental stabilizing contact is permitted.
• Standing without support: Appraises the sensorimotor integration necessary to maintain verticality against gravity solely via intrinsic ankle and hip postural strategies.
• Standing on the non-paretic leg: Assesses unipedal stability on the healthy limb, reflecting underlying executive motor planning and physiological compensation.
• Standing on the paretic leg: Directly isolates the sensorimotor capacity of the neurologically compromised limb. This task tests whether the paretic lower extremity can generate sufficient eccentric and concentric force, proprioceptive feedback, and joint stabilization to bear total body weight without buckling or hyper-extending into pathological recurvatum.

2. Changing a Posture (Dynamic Postural Control)

Dynamic postural control reflects the nervous system’s capacity to preserve equilibrium while purposely displacing the center of mass or altering the base of support during goal-directed, volitional transitional movements. This dynamic dimension comprises 7 standardized items:

• Supine to lateral decubitus (toward the paretic side): Evaluates axial trunk rotation and coordinated limb flexion toward the hemiplegic side, assessing spatial awareness and willingness to load the compromised body hemisphere.
• Supine to lateral decubitus (toward the non-paretic side): Measures the capacity to pull the paretic limb across the midline against resistance and gravity.
• Supine to sitting up on the edge of the examination bed: Evaluates complex biomechanical coordination requiring lateral flexion of the neck and trunk, pushing off with the upper limbs, and controlled lowering of the lower extremities over the edge of the mat.
• Sitting down from an upright seated posture to supine: Appraises controlled eccentric trunk decelerative capacity and protective spatial orientation onto the plinth.
• Sitting to standing: Demands substantial forward trunk flexion, anticipatory postural adjustments, anterior weight translation, and bilateral symmetrical knee/hip extensor power.
• Standing to sitting: Measures the eccentric motor control and dynamic stability required to smoothly descend against gravity without uncoordinated falling back into the seat.
• Standing, picking up an object from the floor: Demands extreme anterior and downward displacement of the center of mass beyond the traditional base of support, testing vestibular-proprioceptive calibration and psychological confidence (absence of debilitating kinesiophobia).

6. Theoretical Framework

The design, structural taxonomy, and clinical operationalization of the mPASS are anchored within modern Systems Theory of Motor Control, originally conceptualized by the pioneering neurophysiologist Nikolai Bernstein, and subsequently expanded into contemporary neurorehabilitation by Anne Shumway-Cook and Marjorie Woollacott. Unlike classical hierarchical or reflex-based paradigms that viewed motor behavior solely as the output of top-down cortical inhibition over primitive spinal reflexes, systems theory posits that motor control is an emergent property resulting from dynamic interactions between the individual, the specific functional task, and the physical environment.

Under this theoretical framework, balance is not a single, isolated neurological entity, but rather a dynamic sensorimotor task requiring:

• Biomechanical Degrees of Freedom Management: Following a focal cerebrovascular lesion, the central nervous system loses its ability to smoothly coordinate multijoint synergies. The mPASS structures its clinical progression from low-degree-of-freedom tasks (e.g., lying supine, seated posture) to high-degree-of-freedom environments (e.g., unassisted standing, unipedal balance), systematically challenging the patient’s capacity to stabilize multiple joints simultaneously.
• Anticipatory and Reactive Postural Adjustments (APAs and RPAs): Transitional items in the mPASS, such as bending forward to retrieve an object or transitioning from sit-to-stand, necessitate feedforward anticipatory postural adjustments. The brain must activate postural trunk and lower extremity musculature milliseconds prior to focal limb movement to prevent biomechanical perturbation.
• Ecological Validity and Functional Task Hierarchies: Systems theory emphasizes that motor assessments must reflect ecologically valid behaviors directly tied to personal survival and functional autonomy. The mPASS embeds essential functional transitions—such as rolling in bed and transitioning from sit-to-stand—directly aligning clinical measurement with real-world functional competence.

7. Validity

The Modified Postural Assessment Scale for Stroke has been rigorously evaluated across multiple psychometric validation investigations, confirming robust construct, concurrent, predictive, and discriminant validity across diverse stroke populations.

Construct and Structural Validity

Persson et al. (2011) examined the construct validity of the mPASS during their foundational optimization trial involving patients evaluated within the first week following acute stroke. Construct validity was corroborated by observing progressive, statistically significant hierarchical gradients in item difficulty. Rasch analysis paradigms applied to the PASS and mPASS structures demonstrate that items calibrate logically along a unidimensional continuum of postural difficulty, with bed-mobility items functioning at the lower continuum of ability and unipedal standing representing the zenith of motor challenge.

Concurrent and Convergent Validity

Concurrent validity of the mPASS has been established through strong bivariate correlations with reference-standard functional and motor measures. In comparative psychometric studies:

• The mPASS exhibits exceptionally high convergent validity with the original 12-item Postural Assessment Scale for Stroke (PASS), displaying Spearman’s rank correlation coefficients ($r_s$) exceeding .95 ($p < .001$).
• Correlations between the mPASS and the Berg Balance Scale (BBS) are consistently robust, ranging from $r = .82$ to $r = .92$, indicating that both instruments probe overlapping domains of equilibrium while the mPASS retains superior resolution at lower ability levels.
• Significant positive associations have been confirmed between mPASS total scores and the Functional Independence Measure (FIM) motor subscale ($r = .75$ to $.86$) and the Barthel Index ($r = .78$ to $.88$), illustrating that higher postural competence directly maps onto elevated functional self-sufficiency.

Predictive and Discriminant Validity

The mPASS demonstrates powerful predictive validity regarding discharge disposition and longitudinal ambulatory recovery. Acute stroke patients scoring above predetermined mPASS threshold values (specifically achieving scores $ge 21/36$ within the first week post-stroke) exhibit dramatically elevated odds of achieving independent functional ambulation at 3 and 6 months post-stroke (Area Under the Curve [AUC] frequently surpassing .85 to .90 in ROC analyses). Furthermore, the scale displays high discriminant validity, cleanly distinguishing between individuals with mild versus moderate-to-severe neurological impairment categorized via the National Institutes of Health Stroke Scale (NIHSS).

8. Reliability

The reliability metrics of the mPASS have been thoroughly investigated, confirming that the structural modifications implemented by Persson et al. successfully enhanced measurement consistency and attenuated inter-evaluator variance.

Internal Consistency

The scale demonstrates exceptional internal consistency across all phases of post-stroke recovery. Published investigations report cumulative Cronbach’s alpha coefficients ranging between .95 and .98 for the complete 12-item inventory. The subdimensions likewise display high internal homogeneity, with the Maintaining a Posture subscale yielding alphas between .88 and .94, and the Changing a Posture subscale yielding alphas between .91 and .96. These values confirm that all 12 items reliably measure facets of a coherent, singular latent motor construct.

Inter-Rater and Intra-Rater Reliability

A primary objective of creating the mPASS was resolving minor ambiguities in scoring definitions that previously contributed to variance between raters. In empirical evaluations:

• Inter-Rater Reliability: Evaluated across physical therapists and clinical neurorehabilitation specialists assessing identical patient cohorts, the overall Intraclass Correlation Coefficient (ICC) for the total mPASS score consistently exceeds .94 (95% CI: .90–.98). Individual item concordance evaluated via weighted Cohen’s kappa ($\kappa_w$) ranges from .72 to .93, demonstrating substantial to almost perfect item-level agreement.
• Intra-Rater / Test-Retest Reliability: Serial administrations conducted across stable time intervals (e.g., 24 to 48 hours apart) demonstrate ICC values ranging from .92 to .97, indicating robust temporal stability in the absence of clinical intervention or spontaneous neurorecovery.
• Standard Error of Measurement (SEM) and Minimal Detectable Change (MDC): The SEM for the mPASS has been established at approximately 1.0 to 1.4 points, resulting in an MDC (at the 95% confidence level) of approximately 2.8 to 3.8 points. Clinicians can confidently interpret an improvement of 4 points or greater as genuine physiological recovery rather than measurement artifact.

9. Factor Analysis

Factor analytic inquiries and structural equation modeling paradigms applied to the PASS and mPASS framework substantiate the structural coherence of the instrument while elucidating its underlying latent dimensionality.

Exploratory Factor Analysis (EFA)

Early structural investigations utilizing principal component and exploratory factor analyses consistently uncover a strong, dominant first factor accounting for greater than 65% to 75% of the total variance, accompanied by an eigenvalue well above 7.0. When oblique rotations (e.g., Promax) are applied, a clean two-factor solution reliably emerges that corresponds directly to the theoretical partition designed by its originators:

• Factor 1: Dynamic Postural Transitions (Changing a Posture): Encompasses the 7 items involving dynamic displacement (bed rolling, sit-to-stand, retrieving objects), with primary factor loadings typically ranging between .72 and .91.
• Factor 2: Static Equilibrium and Unipedal Control (Maintaining a Posture): Encompasses the 5 static items, with unipedal stance items demonstrating robust loadings (.78 to .89) on this dimension.

Confirmatory Factor Analysis (CFA) and Modern Psychometrics

Confirmatory factor analytic evaluations have confirmed that while a correlated two-factor model provides an exceptional theoretical match, a bifactor or higher-order model comprising a single broad ‘Postural Control’ construct best reflects the high inter-correlation between the two latent dimensions ($r > .80$). Modern structural fit indices across stroke cohorts routinely confirm acceptable to excellent parameters:

• Comparative Fit Index (CFI): Values typically $ge .96$
• Tucker-Lewis Index (TLI): Values typically $ge .95$
• Root Mean Square Error of Approximation (RMSEA): Values generally $le .06$ to $.08$ (90% CI: .04–.09)

Rasch unidimensional measurement modeling further validates the construct, proving that the 12 items form a hierarchically ordered, cumulative Guttman-like scale with negligible differential item functioning (DIF) across age and sex brackets.

10. Instrument / Measurement Tool

The Modified Postural Assessment Scale for Stroke is a performance-based, clinician-administered observational rating instrument. Below is the structural summary of the assessment apparatus, administrative prerequisites, and operational rules:

• Instrument Type: Clinician-rated functional performance measurement battery.
• Target Clinical Population: Adults and older adults who have sustained an acute, subacute, or chronic stroke (ischemic or hemorrhagic); adaptable across inpatient acute wards, neurorehabilitation units, and outpatient clinics.
• Administration Time: Approximately 10 to 15 minutes, depending on the severity of the patient’s motor impairment.
• Equipment Required:
  • An examination plinth / treatment bed with height adjustability (standardized sitting height where patient’s feet rest flat on the floor).
  • A digital stopwatch or timer (calibrated in seconds).
  • A standardized small object (e.g., pen, pencil, or clinical marker) for the floor retrieval task.
• Total Item Count: 12 discrete functional tasks divided into two clinical subscales:
  • Subscale I: Maintaining a Posture (5 items)
  • Subscale II: Changing a Posture (7 items)
• Response and Scoring Format: Each item is rated on a 4-point ordinal scale ranging from 0 to 3:
  • 0: Cannot perform the task / total inability even with assistance.
  • 1: Performs the task with substantial manual assistance and/or environmental support.
  • 2: Performs the task with minimal assistance, supervision, or modest compensatory strategies (or maintains posture for a defined sub-maximal duration).
  • 3: Performs the task completely independently with normal postural control and safety (achieving full standardized duration thresholds, e.g., unassisted standing for specified periods).
• Cumulative Scoring Range: Total score extends from 0 to 36 points.
  • Maintaining a Posture subscore: 0 to 15 points.
  • Changing a Posture subscore: 0 to 21 points.
  • Higher aggregate scores denote superior postural stability, advanced motor coordination, and functional independence.

11. Permissions & Fee and Test Year

The Modified Postural Assessment Scale for Stroke was formally established and published by Christina Unger Persson and colleagues in 2011, originating from earlier seminal work on the PASS conducted by Benaim et al. in 1999. In 2019, R. Brouwer formulated and published the validated Dutch clinical adaptation.

Licensing and Fee Status: The mPASS is broadly classified as an open-access clinical and academic assessment tool. There are no proprietary commercial purchase fees, software license keys, or per-use royalties mandated for academic research or routine non-commercial clinical deployment. However, the original publications, operationalized scoring guides, and associated manual descriptions remain under academic copyright held by the respective publishing journals and primary authors. Clinicians, health networks, and researchers wishing to integrate the mPASS into commercial clinical trials, electronic medical record (EMR) vendor packages, or proprietary digital platforms should review licensing parameters via the original primary author (Christina U. Persson, University of Gothenburg) or the respective publishing rights holders.

12. References

• Benaim, C., Pérennou, D. A., Villy, J., Rousseaux, M., & Pelissier, J. Y. (1999). Validation of a new posture assessment scale in stroke patients: Postural Assessment Scale for Stroke Patients (PASS). Stroke, 30(9), 1862–1868. https://doi.org/10.1161/01.STR.30.9.1862
• Brouwer, R. (2019). Toelichtingsformulier Meetinstrument: Modified Postural Assessment Scale for Stroke (mPASS) – Nederlandse Versie. KNGF Richtlijn Beroerte meetinstrumenten documentatie.
• Chien, C. W., Hu, M. H., Hsu, W. L., Hsieh, C. L., & Hsueh, I. P. (2007). Validity and responsiveness of the Postural Assessment Scale for Stroke Patients (PASS) in people with subacute stroke. Physical Therapy, 87(12), 1629–1638. https://doi.org/10.2522/ptj.20060341
• Huang, Y. C., Wang, W. T., Li, P. C., Chen, S. F., & Hsieh, C. L. (2016). Postural Assessment Scale for Stroke Patients: A meta-analytic review of psychometric properties. Archives of Physical Medicine and Rehabilitation, 97(6), 1013–1023. https://doi.org/10.1016/j.apmr.2015.10.106
• Mao, H. F., Hsueh, I. P., Tang, P. F., Sheu, C. F., & Hsieh, C. L. (2002). Analysis and comparison of the psychometric properties of three balance measures for stroke patients. Stroke, 33(4), 1022–1027. https://doi.org/10.1161/01.STR.0000012516.63191.C5
• Persson, C. U., Sunnerhagen, K. S., & Danielsson, A. (2011). Responsiveness of a modified version of the Postural Assessment Scale for Stroke Patients and longitudinal change in postural control after stroke. Journal of Rehabilitation Medicine, 43(8), 706–711. https://doi.org/10.2340/16501977-0841
• Shumway-Cook, A., & Woollacott, M. H. (2017). Motor Control: Translating Research into Clinical Practice (5th ed.). Wolters Kluwer.

13. Items of the Scale