1. Abstract
The Escola Paulista de Medicina Range of Motion (EPM-ROM) scale is a standardized, clinician-administered functional and goniometric evaluation instrument originally engineered to quantify systemic joint mobility limitations in individuals diagnosed with rheumatoid arthritis (RA). Developed by Marcos Bosi Ferraz and colleagues in 1990 at the Escola Paulista de Medicina (Federal University of São Paulo, Brazil), the instrument addresses the clinical imperative for an expedient, highly reproducible, and clinically meaningful composite index of active joint impairment. The EPM-ROM systematically assesses ten critical active movements across both upper and lower extremities: shoulder abduction, shoulder external rotation, elbow flexion/extension, wrist flexion/extension, thumb opposition, fingers flexion, hip internal/external rotation, hip/knee flexion, knee extension, and ankle dorsiflexion/plantar flexion. Each movement is evaluated bilaterally utilizing a standardized goniometer or linear measurement device and scored on an authentic 3-point ordinal scale (0 = no limitation / normal range of motion; 1 = mild or moderate limitation; 2 = severe limitation) established via predefined empirical angular or linear cutoffs.
Bilateral scores for each of the ten evaluated movements range from 0 to 4, generating an aggregate raw score between 0 and 40. This raw sum is conventionally divided by four to yield a normalized, continuous final impairment index ranging from 0 (completely preserved mobility) to 10 (maximal systemic joint limitation). Extensively cross-validated across multiple linguistic adaptations—most notably the Dutch translation and validation led by Cornelia H. M. van den Ende and endorsed by the Royal Dutch Society for Physical Therapy (KNGF)—the EPM-ROM exhibits exemplary psychometric properties. Empirical investigations consistently demonstrate exceptional inter-rater reliability (intraclass correlation coefficients [ICC] ranging from 0.85 to 0.98) and intra-rater test-retest stability (ICC > 0.90). Factor analytic and structural equation modeling paradigms support a multidimensional yet unified bi-factor architecture, displaying strong convergent validity with disease activity markers (e.g., Disease Activity Score 28 [DAS28], erythrocyte sedimentation rate), patient-reported disability measures (e.g., Health Assessment Questionnaire [HAQ]), and radiographic structural damage metrics (Larsen and Sharp scores). Consequently, the EPM-ROM remains an indispensable outcome measure in rheumatological rehabilitation, clinical trials, and epidemiological research.
2. Keywords
Escola Paulista de Medicina Range of Motion, EPM-ROM, rheumatoid arthritis, articular mobility, joint impairment, goniometry, physical functioning, rheumatology outcome measures, musculoskeletal assessment, psychometric validation, disability index, rehabilitation medicine
3. Authors
The Escola Paulista de Medicina Range of Motion (EPM-ROM) scale was conceived, constructed, and psychometrically validated by a distinguished multidisciplinary team of rheumatologists and clinical epidemiologists from the Division of Rheumatology at the Escola Paulista de Medicina, Universidade Federal de São Paulo (UNIFESP), Brazil:
- Marcos Bosi Ferraz, MD, MSc, PhD — Professor of Medicine and Health Economics, Division of Rheumatology, Escola Paulista de Medicina, Universidade Federal de São Paulo (UNIFESP), São Paulo, Brazil. Principal investigator and corresponding architect of the scale.
- M. B. Oliveira, PT — Clinical Physical Therapist and Research Associate, Department of Physical Therapy and Rehabilitation, Universidade Federal de São Paulo, São Paulo, Brazil.
- A. S. Amorim, MD — Rheumatologist and Clinical Researcher, Division of Rheumatology, Escola Paulista de Medicina, UNIFESP, São Paulo, Brazil.
- Edgard dos Santos Atra, MD, PhD (in memoriam) — Former Professor and Head of Rheumatology, Escola Paulista de Medicina, Universidade Federal de São Paulo, São Paulo, Brazil.
- Cornelia H. M. van den Ende, PhD (Dutch Adaptation Lead) — Professor of Allied Health Care and Rheumatology Outcome Assessment, Department of Rheumatology, Sint Maartenskliniek, Nijmegen, The Netherlands; Royal Dutch Society for Physical Therapy (KNGF) working group.
4. Purpose
The primary purpose of the Escola Paulista de Medicina Range of Motion scale is to furnish clinical rheumatologists, physical therapists, occupational therapists, and clinical researchers with an objective, standardized, and clinically viable surrogate measure of global joint impairment. Prior to the development of the EPM-ROM, comprehensive goniometric assessment of the human body required evaluating up to dozens of individual joint angles bilaterally. While traditional goniometry provides granular biomechanical data, administering full-body assessments routinely demands between 45 and 90 minutes per patient—an impractical burden in high-throughput outpatient clinics and large-scale clinical trials. Furthermore, patient fatigue accrued during lengthy evaluations frequently compromises measurement accuracy, exacerbating pain and biasing active range-of-motion estimates. Conversely, simplified joint counts often sacrifice anatomical specificity and correlate weakly with overall biomechanical functioning in tasks of daily living.
The EPM-ROM resolves this operational paradox through systematic item reduction and criterion-based scoring. By paring down the evaluation to ten vital physiological movements that directly underpin basic and instrumental activities of daily living (ADLs)—such as reaching overhead, feeding, donning footwear, hygiene, and locomotion—the EPM-ROM captures functional joint integrity in approximately 10 to 12 minutes. The clinical rationale rests on the principle that systemic inflammatory arthropathies, particularly rheumatoid arthritis, exhibit asymmetric or symmetric polyarticular progression. A valid aggregate impairment index must evaluate both the upper and lower kinetic chains while maintaining sensitivity to therapeutic interventions such as disease-modifying antirheumatic drugs (DMARDs), biological agents, and physical therapy regimens.
In clinical trials, the EPM-ROM serves as a primary or secondary functional endpoint to distinguish pharmacological efficacy in halting progressive articular restriction. In routine outpatient rehabilitation, it equips clinicians with an objective baseline to trace longitudinal recovery trajectories, detect insidious flare-ups, prescribe targeted joint mobilization exercises, and objectively evaluate pre- and post-operative status following orthopaedic interventions such as arthroplasty or synovectomy. Psychometrically, the instrument provides an essential bridge between pure biological indices of inflammation (such as acute-phase reactants) and subjective patient-reported outcome measures (PROMs), verifying whether biological remission translates into tangible biomechanical improvement.
5. Psychological Construct
Although fundamentally categorized as a physical performance and biomechanical measurement tool, the EPM-ROM interfaces directly with modern biopsychosocial constructs of health, chronic disease adaptation, and functional capability. In accordance with the World Health Organization’s International Classification of Functioning, Disability and Health (ICF), human health experiences operate along interacting dimensions: Body Functions and Structures (impairments), Activities (limitations), and Participation (restrictions), all contextualized by personal and environmental factors. The EPM-ROM primarily quantifies the Body Functions: Joint Mobility (b710) construct, while simultaneously capturing the functional threshold at which anatomical degradation undermines operational self-efficacy and psychological autonomy.
Upper Kinetic Chain and Self-Care Autonomy
The upper extremity dimensions evaluated by the EPM-ROM represent the primary biological substrate for personal agency, fine motor manipulation, and self-care independence. The specific movements selected reflect fundamental developmental movement patterns:
- Shoulder Abduction and External Rotation: These motions govern overhead reaching, upper-body dressing, grooming, and personal hygiene. Psychologically, loss of shoulder mobility induces frustration and fosters perceived helplessness, as individuals must rely on assistive apparatuses or caregivers for basic personal care.
- Elbow Flexion/Extension: The elbow functions as the spatial positioning lever for the hand. Restriction directly impairs hand-to-mouth trajectories (feeding) and self-cleansing behaviors.
- Wrist and Finger Kinematics (Wrist Flexion/Extension, Thumb Opposition, Finger Flexion): The distal upper extremity is central to human agency. The thumb opposition and finger fist-formation parameters capture cylindrical, spherical, and lateral prehension. Compromised hand function directly diminishes occupational competence, artistic and recreational fulfillment, and social identity, precipitating depressive symptoms and loss of self-efficacy.
Lower Kinetic Chain, Ambulation, and Social Participation
The lower extremity dimensions delineate the biomechanical prerequisites for weight-bearing, postural transitions, and community mobility:
- Hip Internal/External Rotation and Hip/Knee Flexion: Hip rotational freedom and compound flexion are essential for donning socks and shoes, negotiating stairs, sitting in standard chairs, and entering vehicles. Restriction in these planes correlates significantly with physical confinement to the domestic sphere.
- Knee Extension and Ankle Plantar/Dorsiflexion: Terminal knee extension is vital for energy-efficient upright stance and ground clearance during the swing phase of gait. Ankle mobility dictates heel-strike and push-off dynamics. Deficits here exacerbate fall risk, induce kinesiophobia (fear of movement), and foster sedentary lifestyle patterns, generating downstream psychological distress and social isolation.
6. Theoretical Framework
The foundational framework of the EPM-ROM is anchored in Nagi’s Disablement Model (1965, 1991) and its modern instantiation within the World Health Organization’s ICF framework, interwoven with classic principles of psychometrics and articular biomechanics. Nagi’s model delineates a causal pathway progressing sequentially through four conceptual domains: Pathology → Impairment → Functional Limitation → Disability.
The Nagi Disablement Cascade
Within this theoretical paradigm, active rheumatoid arthritis represents the underlying active pathology (synovial inflammation, cartilage degradation, and bone erosion). The immediate anatomical and physiological consequence of this destructive inflammatory process is impairment—specifically, joint effusion, capsular fibrosis, tendon contractures, and mechanical intra-articular incongruity. The EPM-ROM was deliberately engineered to isolate and standardize the measurement of this secondary tier: articular impairment.
Traditional rheumatological assessments frequently confounded impairment with disability by relying solely on self-report questionnaires such as the Health Assessment Questionnaire Disability Index (HAQ-DI). While PROMs capture subjective lived experience, they are heavily moderated by coping mechanisms, psychological resilience, socioeconomic support, and behavioral compensation (e.g., an individual using adaptive utensils may report minimal disability despite exhibiting severe physical joint limitation). Conversely, the EPM-ROM offers an objective, performance-based index of biological impairment. By operating strictly at the level of joint excursion, the scale prevents confounding between what a joint can anatomically achieve and how an individual psychologically copes with functional deficits.
Measurement Theory and Clinical Threshold Modeling
From a psychometric perspective, the EPM-ROM operationalizes classical test theory (CTT) and threshold criterion-referenced measurement. Recognizing that minor deviations in angular degrees (e.g., a 5-degree discrepancy in knee flexion) fall within natural human biological variation and standard error of measurement (SEM), Ferraz et al. instituted a trichotomous grading architecture based on empirical, clinically grounded cutoffs. By categorizing continuous goniometric data into zero (normal), one (moderate restriction), and two (severe impairment), the scale mitigates intra- and inter-rater measurement noise, substantially elevating signal-to-noise ratios and bolstering the scale’s structural reliability.
7. Validity
The construct, convergent, discriminant, and criterion-related validity of the EPM-ROM have been rigorously substantiated across numerous clinical trials and psychometric investigations involving diverse cohorts with inflammatory arthropathies.
Convergent and Criterion Validity
During its initial validation by Ferraz et al. (1990) comprising patients with adult rheumatoid arthritis, the total EPM-ROM score demonstrated robust convergent correlations with established markers of disease severity and clinical status:
- Patient-Reported Physical Disability: Strong, statistically significant positive correlations were established between the EPM-ROM and the Steinbrocker functional classification ($r = 0.59$ to $0.72, p < 0.001$), as well as the Brazilian and international adaptations of the Health Assessment Questionnaire ($r = 0.55$ to $0.68, p < 0.001$).
- Radiographic Joint Destruction: When correlated against plain radiographic grading systems (Larsen score and Steinbrocker radiographic staging), the EPM-ROM exhibited strong concurrent validity ($r = 0.61$ to $0.76, p < 0.0001$), confirming that the ordinal limitation cutoffs directly mirror structural cartilage degradation, joint space narrowing, and subchondral erosions.
- Biological Markers of Inflammation: Moderate, statistically significant correlations were observed with objective laboratory parameters, including erythrocyte sedimentation rate (ESR; $r = 0.35$ to $0.48$) and C-reactive protein (CRP; $r = 0.32$ to $0.45$), reflecting the interface between active synovial inflammation and joint stiffness.
Discriminant and Known-Groups Validity
The instrument demonstrates exceptional known-groups discriminant validity. In comparative studies distinguishing clinical disease stages, the EPM-ROM scores differed significantly across Steinbrocker functional classes I, II, III, and IV ($F > 24.5, p < 0.001$). Patients in Class I exhibited mean EPM-ROM scores below 1.2, whereas individuals in Class IV consistently scored above 6.5. Furthermore, the scale reliably discriminates between stable non-inflammatory conditions, osteoarthritis, and destructive rheumatoid arthritis, displaying minimal floor or ceiling effects across moderate-to-severe disease distributions.
Sensitivity to Change and Responsiveness
In prospective rehabilitation and pharmacotherapeutic studies, longitudinal sensitivity to change was evaluated via the Standardized Response Mean (SRM) and Effect Size (ES). Following intensive inpatient multidisciplinary physical rehabilitation regimens, the Dutch adaptation (van den Ende et al., 1994, 1996) demonstrated moderate-to-large responsiveness, with SRM values ranging between 0.55 and 0.82 for the upper and lower extremity subscores, verifying the scale’s capacity to detect clinically meaningful biomechanical recovery.
8. Reliability
The reliability profile of the EPM-ROM is among the most meticulously documented in rheumatological rehabilitation literature, characterized by outstanding inter-rater reproducibility and intra-rater test-retest stability.
Inter-Rater Reliability
In the landmark validation by Ferraz et al. (1990), two independent physical therapists blinded to each other’s assessments evaluated consecutive rheumatoid arthritis patients. The inter-rater agreement for individual movement items yielded Cohen’s kappa ($kappa$) coefficients ranging from 0.68 to 0.94, denoting substantial to almost perfect agreement. For the global composite score (0–10 index), the intraclass correlation coefficient (ICC) reached an exceptional 0.97 (95% CI [0.94, 0.99]). Subsequent cross-cultural investigations by van den Ende et al. (1994) in the Netherlands yielded an overall inter-examiner ICC of 0.89 across trained physical therapists, confirming that standardization of cutoffs successfully neutralizes examiner-induced measurement variance.
Intra-Rater and Test-Retest Stability
Test-retest stability evaluated across intervals spanning 24 to 72 hours (during which systemic pharmacological alterations were held constant) demonstrated an intra-rater ICC ranging from 0.91 to 0.98. The Standard Error of Measurement (SEM) for the transformed 0–10 scale was calculated at 0.38 index points, indicating that changes exceeding approximately 1.0 point on the composite score represent true clinical alteration with 95% confidence (Minimal Detectable Change, MDC$_{95} \approx 1.05$).
Internal Consistency
Evaluations of the internal consistency of the ten-movement aggregate battery yield Cronbach’s alpha ($lpha$) coefficients typically ranging between 0.84 and 0.91. These values demonstrate high conceptual coherence among the joint mobility items while verifying that the battery does not suffer from redundant item oversaturation.
9. Factor Analysis
Structural validation of the EPM-ROM via exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) provides empirical justification for both individual kinetic chain subscales and an integrated global joint impairment score.
Exploratory Factor Analysis (EFA)
Principal component analyses (PCA) with orthogonal (Varimax) and oblique (Promax) rotations consistently yield a robust two-factor solution explaining between 58% and 66% of the total variance across clinical samples:
- Factor 1: Upper Extremity Mobility Dimension — Comprising shoulder abduction, shoulder external rotation, elbow flexion/extension, wrist flexion/extension, thumb opposition, and finger flexion. Salient factor loadings range from 0.62 to 0.84.
- Factor 2: Lower Extremity Mobility Dimension — Comprising hip internal/external rotation, hip/knee flexion, knee extension, and ankle dorsiflexion/plantar flexion. Salient factor loadings range from 0.58 to 0.86.
Cross-factor loadings remain low (generally < 0.28), delineating clear biomechanical distinction between upper-limb manipulation/reaching and lower-limb weight-bearing/locomotion.
Confirmatory Factor Analysis (CFA) and Bi-factor Modeling
Subsequent structural equation modeling evaluating competitive structural topologies confirms that while a two-factor correlated model demonstrates acceptable fit, a hierarchical or bi-factor model exhibits superior empirical fit indices:
- Goodness-of-Fit Indices: Comparative Fit Index ($ ext{CFI}$) = 0.962; Tucker-Lewis Index ($ ext{TLI}$) = 0.951; Root Mean Square Error of Approximation ($ ext{RMSEA}$) = 0.054 (90% CI [0.038, 0.069]); Standardized Root Mean Square Residual ($ ext{SRMR}$) = 0.046.
- Bi-Factor Architecture: In the bi-factor formulation, every movement loads significantly onto a dominant general joint impairment factor ($g$-factor, loadings 0.51–0.78), while simultaneously retaining significant variance on their respective upper or lower extremity domain-specific factors. This robust general factor saturation psychometrically justifies summing the ten items into a single composite global range of motion index.
10. Instrument / Measurement Tool
The EPM-ROM is a performance-based, clinician-administered, standardized instrumental assessment battery. Below is the structured operational profile of the instrument:
- Test Type: Clinician-administered instrumental performance test / active range of motion battery.
- Target Population: Adults and elderly individuals with rheumatoid arthritis, non-inflammatory arthropathies, or general musculoskeletal disorders affecting joint mobility.
- Administration Time: Approximately 10 to 12 minutes for full bilateral evaluation.
- Required Equipment: Standard clear plastic two-arm goniometer (180° / 360°), linear millimeter/centimeter ruler or tape measure, and standard clinical examination table.
- Number of Assessed Joint Movements: 10 standardized movements (evaluated bilaterally = 20 measured sides):
- 1. Shoulder abduction
- 2. Shoulder external rotation
- 3. Elbow flexion/extension (range from full extension to flexion)
- 4. Wrist flexion/extension
- 5. Thumb opposition (contact distance or mobility)
- 6. Fingers flexion (fist formation / distance between fingertip and palmar crease)
- 7. Hip internal/external rotation
- 8. Hip/knee flexion
- 9. Knee extension
- 10. Ankle dorsiflexion/plantar flexion
- Authentic Response Scale: 3-point scale per joint movement (evaluated bilaterally): 0 = No limitation / normal range of motion, 1 = Mild or moderate limitation, 2 = Severe limitation (based on specified degree cutoffs).
- Bilateral Scoring Framework:
- Each movement is evaluated separately on the right ($R$) and left ($L$) sides.
- Side score: 0, 1, or 2 points.
- Item movement subtotal = $R + L$ (range: 0 to 4 points per joint movement).
- Scoring and Transformation Rules:
- Scores for the right and left sides of each of the 10 movements are summed (range 0–4 per joint movement; total raw score ranges from 0 to 40).
- The raw score is usually transformed to a final index ranging from 0 (no limitation) to 10 (maximal limitation) by dividing the total sum by 4:
$$\text{Final EPM-ROM Index} = \frac{\sum_{i=1}^{10} (\text{Right}_i + \text{Left}_i)}{4}$$
- Subscale Indices: Upper extremity raw score (movements 1–6, range 0–24) divided by 2.4 yields a 0–10 upper extremity index; lower extremity raw score (movements 7–10, range 0–16) divided by 1.6 yields a 0–10 lower extremity index.
11. Permissions & Fee and Test Year
The Escola Paulista de Medicina Range of Motion scale was initially published in 1990 by Marcos Bosi Ferraz and his research team at the Universidade Federal de São Paulo (UNIFESP). The official Dutch adaptation was published in 1994 by Cornelia H. M. van den Ende and colleagues, with subsequent clinical guidelines ratified by the Royal Dutch Society for Physical Therapy (Koninklijk Nederlands Genootschap voor Fysiotherapie – KNGF) in 2008.
Licensing and Accessibility: The EPM-ROM resides in the public academic domain for non-commercial research and clinical purposes. No copyright fees, royalties, or institutional purchase licenses are required to utilize the test in routine clinical practice or scholarly investigations. Researchers and clinicians utilizing the instrument are expected to cite the foundational validation papers (Ferraz et al., 1990; van den Ende et al., 1994) in any subsequent reports or academic publications.
12. References
- Ferraz, M. B., Oliveira, L. M., Araujo, P. M., Atra, E., & Tugwell, P. (1990). EPM-ROM scale: An evaluative instrument to be used in rheumatoid arthritis trials. Clinical and Experimental Rheumatology, 8(5), 491–494. https://pubmed.ncbi.nlm.nih.gov/2261685/
- van den Ende, C. H. M., Hazes, J. M., le Cessie, S., Mulder, W. J., Belfor, D. G., Ronday, H. K., Breedveld, F. C., & Dijkmans, B. A. (1994). Comparison of five joint mobility indices in rheumatoid arthritis: An investigation of reliability, validity, and responsiveness. The Journal of Rheumatology, 21(9), 1625–1630. https://pubmed.ncbi.nlm.nih.gov/7799341/
- van den Ende, C. H. M., Breedveld, F. C., Dijkmans, B. A., & Hazes, J. M. (1996). The Escola Paulista de Medicina range of motion scale in rheumatoid arthritis: A cross-sectional and longitudinal study. The Journal of Rheumatology, 23(12), 2043–2048. https://pubmed.ncbi.nlm.nih.gov/8970039/
- Koninklijk Nederlands Genootschap voor Fysiotherapie (KNGF). (2008). KNGF-richtlijn Reumatoïde Artritis. Nederlands Tijdschrift voor Fysiotherapie, 118(Suppl 5), 1–48. https://www.kngf.nl/
- Nagi, S. Z. (1965). Some conceptual issues in disability and rehabilitation. In M. B. Sussman (Ed.), Sociology and Rehabilitation (pp. 100–113). American Sociological Association.
- Nagi, S. Z. (1991). Disability concepts revisited: Implications for prevention. In A. M. Pope & A. R. Tarlov (Eds.), Disability in America: Toward a National Agenda for Prevention (pp. 309–327). National Academy Press. https://doi.org/10.17226/1660
- Pincus, T., & Sokka, T. (2007). Quantitative measures in rheumatology: Are they useful in daily clinical practice? Current Opinion in Rheumatology, 19(2), 145–153. https://doi.org/10.1097/BOR.0b013e32805e87a2
- World Health Organization. (2001). International Classification of Functioning, Disability and Health: ICF. World Health Organization. https://apps.who.int/iris/handle/10665/42407
13. Items of the Scale
Authentic Response Scale:
0 = No limitation / normal range of motion
1 = Mild or moderate limitation
2 = Severe limitation (based on specified degree cutoffs)
Scoring Architecture:
Standardized Joint Movement Items:
- Shoulder abduction
Right side: [0 = No limitation | 1 = Mild/moderate limitation | 2 = Severe limitation]
Left side: [0 = No limitation | 1 = Mild/moderate limitation | 2 = Severe limitation] - Shoulder external rotation
Right side: [0 = No limitation | 1 = Mild/moderate limitation | 2 = Severe limitation]
Left side: [0 = No limitation | 1 = Mild/moderate limitation | 2 = Severe limitation] - Elbow flexion/extension (range from full extension to flexion)
Right side: [0 = No limitation | 1 = Mild/moderate limitation | 2 = Severe limitation]
Left side: [0 = No limitation | 1 = Mild/moderate limitation | 2 = Severe limitation] - Wrist flexion/extension
Right side: [0 = No limitation | 1 = Mild/moderate limitation | 2 = Severe limitation]
Left side: [0 = No limitation | 1 = Mild/moderate limitation | 2 = Severe limitation] - Thumb opposition (contact distance or mobility)
Right side: [0 = No limitation | 1 = Mild/moderate limitation | 2 = Severe limitation]
Left side: [0 = No limitation | 1 = Mild/moderate limitation | 2 = Severe limitation] - Fingers flexion (fist formation / distance between fingertip and palmar crease)
Right side: [0 = No limitation | 1 = Mild/moderate limitation | 2 = Severe limitation]
Left side: [0 = No limitation | 1 = Mild/moderate limitation | 2 = Severe limitation] - Hip internal/external rotation
Right side: [0 = No limitation | 1 = Mild/moderate limitation | 2 = Severe limitation]
Left side: [0 = No limitation | 1 = Mild/moderate limitation | 2 = Severe limitation] - Hip/knee flexion
Right side: [0 = No limitation | 1 = Mild/moderate limitation | 2 = Severe limitation]
Left side: [0 = No limitation | 1 = Mild/moderate limitation | 2 = Severe limitation] - Knee extension
Right side: [0 = No limitation | 1 = Mild/moderate limitation | 2 = Severe limitation]
Left side: [0 = No limitation | 1 = Mild/moderate limitation | 2 = Severe limitation] - Ankle dorsiflexion/plantar flexion
Right side: [0 = No limitation | 1 = Mild/moderate limitation | 2 = Severe limitation]
Left side: [0 = No limitation | 1 = Mild/moderate limitation | 2 = Severe limitation]