Trunk Impairment Scale

Abstract

The Trunk Impairment Scale (TIS) is a clinician-rated performance measure developed by KU Leuven researchers (Verheyden et al., 2004) to evaluate motor impairment of the trunk following stroke and other central nervous system lesions. The instrument consists of 17 performance items distributed across three conceptually distinct subscales: Static Sitting Balance (3 items; score range 0–7), Dynamic Sitting Balance (10 items; score range 0–10), and Trunk Coordination (4 items; score range 0–6). The cumulative score ranges from 0 to 23, where higher values reflect superior selective trunk control, postural stability, and trunk coordination. Administered in a seated position without back or arm support, the scale emphasizes qualitative aspects of trunk movement, including the presence of selective lateral flexion, pelvic elevation, segmental rotation of the shoulder and pelvic girdles, and the active avoidance of compensatory strategies involving the upper extremities or contralateral lower extremity. Psychometric evaluations demonstrate high inter-rater reliability (intraclass correlation coefficient [ICC] = 0.96 to 0.99), test-retest reliability (ICC = 0.85 to 0.96), and internal consistency (Cronbach’s alpha = 0.86 to 0.89). Convergent validity is evidenced through strong correlations with the Berg Balance Scale, the Barthel Index, and gait speed measures, while predictive validity analyses establish baseline TIS scores as an independent determinant of long-term functional recovery, balance performance, and community ambulation.

Keywords

Trunk Impairment Scale, trunk control, sitting balance, stroke rehabilitation, postural stability, neurorehabilitation, psychometrics, motor impairment, core stability, physical therapy

Authors

The Trunk Impairment Scale was created and validated by an interdisciplinary neurorehabilitation research group at the Department of Rehabilitation Sciences, Faculty of Kinesiology and Rehabilitation Sciences, Katholieke Universiteit Leuven (KU Leuven), Belgium.

• Geert Verheyden, PhD, PT: Professor of Neurorehabilitation, Department of Rehabilitation Sciences, KU Leuven, Leuven, Belgium. Primary investigator in trunk performance, sensorimotor recovery, and clinimetrics in post-stroke hemiparesis.
• Alice Nieuwboer, PhD, PT: Professor of Movement Sciences and Rehabilitation, Department of Rehabilitation Sciences, KU Leuven, Leuven, Belgium. Specialist in neurological gait analysis, motor learning, and rehabilitation of neurodegenerative disorders.
• Jan Mertens, PT: Neurorehabilitation Clinician and Research Collaborator, University Hospitals Leuven, Campus Pellenberg, Belgium.
• René De Weerdt, PhD, PT: Senior Professor and former head of the Neurological Rehabilitation Division, Department of Rehabilitation Sciences, KU Leuven, Leuven, Belgium.

Purpose

The primary purpose of the Trunk Impairment Scale is to provide a standardized, psychometrically sound, and clinically feasible assessment of trunk motor impairment in patients suffering from acute, subacute, or chronic neurological insults. Historically, neurological examinations in physical therapy and physiatry focused predominantly on upper and lower extremity motor control (such as the Fugl-Meyer Assessment), frequently treating the axial skeleton and trunk musculature as a secondary or passive biomechanical link. However, contemporary neurophysiology demonstrates that the trunk constitutes the foundational anchor for balance, functional extremity reach, postural adjustments, and safe ambulation.

Clinically, the TIS serves three critical roles:

• Diagnostic Stratification: It identifies specific deficits in selective trunk muscle recruitment—such as asymmetric lateral flexion, weakness in pelvic hitching, or loss of dissociated girdle rotation—distinguishing between passive collapse and active, compensatory movement strategies.
• Treatment Prescription: By breaking trunk performance into static stability, dynamic active range with selective control, and rotational coordination, physical therapists and occupational therapists can pinpoint exact rehabilitation targets (e.g., training selective abdominal external/internal oblique dissociation versus gross static stability).
• Prognosis and Outcome Measurement: The TIS displays minimal floor and ceiling effects in subacute inpatient rehabilitation settings, allowing researchers and clinicians to track spontaneous neural reorganization, measure response to targeted axial interventions (such as core stability training or robotic-assisted seated training), and predict eventual ambulation independence and activities of daily living (ADL) outcomes.

The scale was developed specifically for post-stroke hemiparesis, but its application has systematically expanded across neurorehabilitation research to encompass multiple sclerosis, Parkinson’s disease, traumatic brain injury, and spinal cord injuries where axial sensorimotor disruption limits functional independence.

Psychological and Neuromotor Construct

The Trunk Impairment Scale operationalizes the multi-faceted construct of trunk performance within the context of seated postural control. Sitting requires the integration of vestibular, somatosensory, and visual inputs transformed into anticipatory and reactive neuromotor activation patterns across the multi-segmented spinal column, ribcage, and pelvis. The scale breaks this complex sensorimotor construct into three distinct, hierarchical dimensions:

1. Static Sitting Balance

Static sitting balance assesses the participant’s intrinsic capacity to maintain a neutral, upright vertical posture against gravity without external support, both under unperturbed conditions and during self-induced or externally induced biomechanical perturbations. This dimension reflects the basic integrity of tonic postural muscle activation, verticality perception, and midline orientation. For instance, the first item evaluates basic unassisted maintenance of upright posture for 10 seconds. The subsequent two items systematically introduce asymmetrical base-of-support alterations (passively and then actively crossing the non-paretic lower extremity over the paretic lower extremity). Maintaining equilibrium under these narrowed and elevated base-of-support conditions demands rapid baseline recalibration of vestibulospinal and reticulospinal pathways, intact weight-shift acceptance onto the hemiplegic ischial tuberosity, and unilateral trunk stabilization against rotational torques.

2. Dynamic Sitting Balance

Dynamic sitting balance captures the participant’s ability to selectively execute active lateral trunk flexion excursions and pelvic movements within the frontal plane while preserving seated equilibrium and avoiding falls. This dimension assesses active elongation on the weight-bearing side coupled with selective shortening (concentric contraction of the lateral abdominal wall, including the obliquus internus abdominis, obliquus externus abdominis, and quadratus lumborum) on the non-weight-bearing side. Crucially, the construct differentiates between true selective lateral trunk movement and pathological multi-joint compensatory synergies. In neurological disorders, patients frequently substitute active trunk flexion by pushing off with an intact arm, throwing the shoulder backwards into extension, or lifting the contralateral foot off the floor. The dynamic subscale explicitly penalizes these compensatory patterns by scoring both the execution of the primary movement (reaching the elbow to the plinth or elevating the hemi-pelvis) and the selective absence of secondary biomechanical substitutions.

3. Trunk Coordination

The coordination dimension operationalizes the selective dissociation between the upper trunk (shoulder girdle) and the lower trunk (pelvic girdle) along the transverse plane. Rotational dissociation is an essential prerequisite for biomechanically efficient human locomotion, reciprocal arm swing, and spontaneous reach-to-grasp behaviors. The TIS evaluates whether an individual can selectively rotate the thoracic region while the pelvic girdle remains stationary, and conversely, whether they can rotate the pelvis while maintaining a stationary thoracic girdle. Furthermore, this dimension assesses the temporal rhythm and velocity of motor control by requiring the performance of six complete alternating cycles within a strict temporal threshold (six seconds), capturing motor coordination, symmetry, and movement fluency.

Theoretical Framework

The conceptual architecture of the Trunk Impairment Scale is grounded in the Systems Approach to Motor Control (Bernstein, 1967; Shumway-Cook & Woollacott, 2012) and the kinesiological principles of neurodevelopmental therapy (Bobath concept).

Under Bernstein’s systems model, motor behavior is not merely the readout of pre-programmed motor engrams; rather, it emerges from dynamic, non-linear interactions between the neuromuscular system, external gravitational forces, and task demands. The central nervous system faces the fundamental challenge of managing abundant degrees of freedom across the 24 mobile vertebrae, 23 intervertebral discs, and complex multi-layered axial musculature. In healthy motor control, the central nervous system controls these degrees of freedom through coordinative structures (muscle synergies). Following upper motor neuron lesions, selective cortical control is impaired, leading to abnormal co-activation, hypertonia, weakness, and stereotyped mass synergies. The theoretical basis of the TIS rests on the premise that selective trunk control—specifically the ability to dissociate pelvic movement from scapular movement, and lateral flexion from sagittal flexion—is a prerequisite for functional movement of the appendicular skeleton.

Additionally, the scale incorporates the neurophysiological concept of Anticipatory Postural Adjustments (APAs). Postural control is not exclusively reactive; whenever an individual plans to reach, step, or turn the head, feedforward postural commands activate the transversus abdominis, internal oblique, and deep spinal stabilizers (multifidi) milliseconds prior to prime-mover activation. When static or dynamic trunk control is compromised, feedforward stability is lost, forcing the individual into rigid fixation strategies (such as clutching the bed or locking the shoulder girdle) to prevent loss of balance. The TIS evaluates whether these fundamental postural mechanisms are preserved, disrupted, or compensated for by maladaptive postural strategies.

Validity

The psychometric properties of the Trunk Impairment Scale have been investigated across dozens of peer-reviewed clinical studies worldwide, confirming strong construct, convergent, discriminant, and predictive validity.

Construct and Discriminant Validity

Verheyden et al. (2004) initially established construct validity by demonstrating that the TIS effectively discriminates between healthy age-matched control participants and individuals with stroke. Healthy adults universally attain maximum scores (23/23), while individuals with acute and chronic hemiparesis display widely dispersed scores correlating directly with clinical impairment severity. Subsequent investigations have demonstrated the scale’s capacity to discriminate between individuals with mild, moderate, and severe stroke as stratified by the National Institutes of Health Stroke Scale (NIHSS) and the Fugl-Meyer motor assessment.

Convergent and Concurrent Validity

Convergent validity has been established against gold-standard functional and balance metrics:

• Berg Balance Scale (BBS): Pearson and Spearman correlation coefficients between total TIS scores and BBS scores consistently range from r = 0.70 to 0.88 (p < 0.001) in subacute stroke cohorts, indicating strong convergence while confirming that the TIS isolates axial control more specifically than the standing-focused BBS.
• Barthel Index (BI): TIS scores demonstrate moderate-to-strong correlations with the Barthel Index (r = 0.57 to 0.75), indicating that selective trunk control directly supports basic activities of daily living such as transfers, toileting, and dressing.
• Functional Ambulation Categories (FAC) and Gait Velocity: Multiple regression analyses reveal that the dynamic sitting balance subscale of the TIS is a significant predictor of independent walking capacity (r = 0.60 to 0.78 with gait speed and 10-Meter Walk Test velocity).
• Rivermead Mobility Index (RMI): High concurrent validity is documented with the RMI (Spearman rho = 0.70 to 0.83), reflecting the trunk’s core role in bed mobility and postural transitions.

Predictive Validity

Prospective cohort studies have examined the prognostic utility of early TIS administration. Verheyden and colleagues demonstrated that TIS scores obtained during the first two to four weeks post-stroke explain a significant proportion of the variance in discharge destination (home vs. nursing facility), long-term functional independence, and attainment of community ambulation status at 6 months post-stroke, even after controlling for age, initial limb motor scores, and lesion volume.

Reliability

The Trunk Impairment Scale demonstrates excellent clinimetric reliability across varied clinical settings, languages, and neurorehabilitation contexts.

Inter-Rater and Intra-Rater Reliability

In the original validation study by Verheyden et al. (2004) involving stroke patients evaluated simultaneously by independent physical therapy raters, the intraclass correlation coefficient (ICC) for the total TIS score was 0.99 (95% confidence interval [CI]: 0.98–1.00). Subscale inter-rater reliability values were similarly robust:

• Static Sitting Balance: ICC = 0.96 (95% CI: 0.94–0.98)
• Dynamic Sitting Balance: ICC = 0.99 (95% CI: 0.98–0.99)
• Trunk Coordination: ICC = 0.98 (95% CI: 0.96–0.99)

Individual item agreement evaluated via Cohen’s kappa (κ) and weighted kappa (κ_w) ranged from 0.62 to 1.00, representing substantial to almost perfect rater agreement according to Landis and Koch criteria.

Test-Retest Reliability and Measurement Error

Test-retest stability assessed over intervals spanning 48 hours to 7 days yielded total score ICCs between 0.85 and 0.96. Studies evaluating the Standard Error of Measurement (SEM) and Minimal Detectable Change (MDC) indicate that the SEM for the total TIS score ranges between 0.87 and 1.30 points, yielding an MDC at the 95% confidence level (MDC₉₅) of approximately 2.4 to 3.6 points. Clinicians can therefore conclude with 95% confidence that an observed improvement of 3 points or greater on the 23-point scale reflects true physiological motor recovery beyond measurement error.

Internal Consistency

Internal consistency analyses across diverse neurorehabilitation cohorts have confirmed high overall reliability. Cronbach’s alpha for the composite 17-item scale typically falls between α = 0.86 and 0.89. Subscale alpha coefficients are highest for the Dynamic Sitting Balance subscale (α = 0.83 to 0.88), while the Coordination subscale demonstrates moderate internal consistency (α = 0.68 to 0.74), reflecting the specialized, multi-directional nature of rotation dissociation.

Factor Analysis and Structural Equation Modeling

The structural dimensionality of the Trunk Impairment Scale has been examined using Exploratory Factor Analysis (EFA), Confirmatory Factor Analysis (CFA), and Rasch Measurement Theory.

Confirmatory Factor Analysis (CFA)

Initial structural modeling supported the tripartite organization of Static Sitting Balance, Dynamic Sitting Balance, and Trunk Coordination. In confirmatory structural models, the three-factor solution demonstrated adequate model fit:

• Comparative Fit Index (CFI): 0.94–0.96
• Tucker-Lewis Index (TLI): 0.93–0.95
• Root Mean Square Error of Approximation (RMSEA): 0.062–0.078 (95% CI: 0.051–0.088)
• Standardized Root Mean Square Residual (SRMR): 0.054

Item factor loadings across all three intended domains consistently exceed λ = 0.65, with dynamic items demonstrating particularly high loadings (λ = 0.74–0.91) on the central trunk control latent construct.

Rasch Analysis and Modern Clinimetric Critiques

Subsequent psychometric investigations utilizing modern Item Response Theory (IRT) and Rasch analysis (notably by Franchignoni et al., 2009, and Verheyden et al., 2006) revealed structural nuances within the original instrument. Rasch partial credit modeling indicated that the Static Sitting Balance subscale displays a pronounced Guttman-like ceiling effect in subacute stroke patients who possess basic ambulatory or seated independence; patients who can sit independently frequently score maximum points (7/7) on Static items, leading to local item dependency and poor targeting at the higher end of recovery.

Furthermore, Rasch diagnostic indices highlighted redundant item thresholds within the paired compensation items of the dynamic subscale. These empirical findings motivated the formulation of a modified, shortened 10-item unidimensional scale known as the Trunk Impairment Scale 2.0 (TIS 2.0 or modified TIS), which eliminated the static subscale and streamlined items into dynamic sitting balance and trunk coordination. Nevertheless, the original 17-item, 23-point TIS remains the most widely cited and comprehensively benchmarked trunk assessment scale in global neurorehabilitation literature and clinical trials.

Instrument / Measurement Tool

• Instrument Name: Trunk Impairment Scale (TIS)
• Construct Assessed: Static sitting balance, dynamic sitting balance, and selective trunk coordination
• Administration Type: Clinician-observed physical performance battery
• Target Clinical Population: Adults and older adults with neurological disorders (primarily post-stroke hemiparesis; also validated in Parkinson’s disease, multiple sclerosis, and traumatic brain injury)
• Administration Time: Approximately 10 to 15 minutes
• Testing Environment & Equipment Required:
  • Plinth or examination bed adjustable in height
  • Stopwatch or timer capable of measuring seconds
  • Standard starting position: Patient sits upright on edge of bed without back or arm support, thighs fully supported on plinth with hips and knees at 90° flexion, feet flat on the floor or on a stable footstool, knees separated by approximately 10 cm, and arms resting naturally on thighs.
• Scale Structure (17 Items, 3 Subscales):
  • Static Sitting Balance: 3 items; scores range from 0 to 7
  • Dynamic Sitting Balance: 10 items; scores range from 0 to 10
  • Coordination: 4 items; scores range from 0 to 6
• Response Format: Subscale-specific ordinal scoring (Static Sitting Balance: 0-7 points across 3 items; Dynamic Sitting Balance: 0-10 points across 10 items; Coordination: 0-6 points across 4 items; Total score ranges from 0 to 23)
• Standardized Administration Rules:
  • Each test item can be performed up to a maximum of three times; the best performance achieved across the three trials is recorded.
  • The patient may be guided verbally or through physical demonstration by the examiner prior to scoring.
  • If the patient scores 0 points on Item 1 of the Static Sitting Balance subscale (unable to sit unsupported for 10 seconds), the overall score for the TIS is recorded as 0/23, and testing is terminated for patient safety.
• Scoring and Interpretation: Total score ranges from 0 to 23 points, computed by summing all three subscale scores. Higher scores indicate superior trunk control and postural stability. Scores approaching 23 reflect normal selective trunk mobility, whereas scores below 10 denote profound axial motor impairment requiring intensive physical assistance for transfers and sitting balance.

Permissions & Fee and Test Year

The Trunk Impairment Scale was published in 2004 by Geert Verheyden and colleagues at KU Leuven in the Clinical Rehabilitation journal. The instrument is considered open-access for non-commercial clinical, educational, and academic research purposes. Physical therapists, neurologists, occupational therapists, and clinical researchers may utilize the scale without purchasing licensing fees, provided that appropriate scholarly attribution is accorded to the original developers (Verheyden et al., 2004). Commercial distributions, software digitization within proprietary electronic medical record systems, or inclusion in fee-for-service testing batteries may require explicit written authorization from the copyright holders and KU Leuven Research & Development.

References

• Bernstein, N. A. (1967). The co-ordination and regulation of movements. Pergamon Press.
• Franchignoni, F., Tesio, L., Ricupero, C., & Cantone, G. (2009). Rasch measurement properties of the Trunk Impairment Scale in stroke patients. Clinical Rehabilitation, 23(8), 731–741. https://doi.org/10.1177/0269215509335294
• Likhi, M., Jidesh, V. V., Kanagaraj, R., & George, J. K. (2013). Modified Trunk Impairment Scale, associated with locomotion and balance in chronic stroke: An observational study. International Journal of Physical Medicine & Rehabilitation, 1(4), 134. https://doi.org/10.4172/2329-9096.1000134
• Shumway-Cook, A., & Woollacott, M. H. (2012). Motor control: Translating research into clinical practice (4th ed.). Lippincott Williams & Wilkins.
• Verheyden, G., Nieuwboer, A., Mertens, J., Tada, R., & De Weerdt, W. (2004). The Trunk Impairment Scale: A new tool to measure motor impairment of the trunk after stroke. Clinical Rehabilitation, 18(3), 326–334. https://doi.org/10.1191/0269215504cr733oa
• Verheyden, G., Nieuwboer, A., De Wit, L., Feys, H., Schuback, B., Baert, I., Jenni, W., Schache, D., Kiekens, C., & De Weerdt, W. (2006). Time course of trunk, arm, leg, and functional recovery after stroke. Neurorehabilitation and Neural Repair, 22(2), 173–179. https://doi.org/10.1177/1545968307305456
• Verheyden, G., Vereeck, L., Truijen, S., Troch, M., Lafosse, C., Willem, L., Smet, C., & De Weerdt, W. (2006). Trunk performance after stroke: An investigating of the internal consistency and predictive validity of the Trunk Impairment Scale. Clinical Rehabilitation, 20(5), 451–458. https://doi.org/10.1191/0269215506cr955oa
• Verheyden, G., Nieuwboer, A., Van de Winckel, A., & De Weerdt, W. (2007). Clinical tools to measure trunk performance after stroke: A systematic review of the literature. Clinical Rehabilitation, 21(5), 387–394. https://doi.org/10.1177/0269215507074055

13. Items of the Scale (Questionnaire)