Hip disability and Osteoarthritis Outcome Score-Physical function Short form

1. Abstract

The Hip disability and Osteoarthritis Outcome Score-Physical function Short form (HOOS-PS) is a validated, patient-reported outcome measure (PROM) specifically engineered to quantify functional physical limitations in individuals suffering from hip osteoarthritis, secondary joint degeneration, and related musculoskeletal pathologies of the lower extremity. Derived mathematically and psychometrically from the comprehensive, 40-item Hip disability and Osteoarthritis Outcome Score (HOOS), the HOOS-PS consolidates the original Activities of Daily Living (ADL) and Sport and Recreation (Sport/Rec) physical function subscales into a parsimonious, five-item unidimensional instrument. Administered across diverse clinical, epidemiological, and post-surgical contexts—including pre- and post-operative monitoring for total hip arthroplasty (THA)—the instrument captures task-specific biomechanical difficulty encountered over the preceding week. The tool evaluates five discrete activities: descending stairs, getting in or out of a bath or shower, sitting, running, and twisting or pivoting on the loaded leg. Responses are graded on an authentic 5-point Likert scale ranging from 0 (“None”) to 4 (“Extreme”). Raw ordinal summation yields scores from 0 to 20, which are transformed through an established modern psychometric conversion metric derived from Rasch measurement theory into an interval-scaled metric ranging from 0 (perfect physical function, representing no difficulty) to 100 (extreme functional limitation). Rigorous international cross-cultural adaptations and multicenter validation trials have consistently demonstrated the instrument’s exemplary measurement properties, including high internal consistency (Cronbach’s alpha typically ranging from 0.79 to 0.89), robust test-retest reliability (intraclass correlation coefficients [ICC] > 0.80), absence of floor and ceiling effects in clinical cohorts, strict unidimensionality, and pronounced responsiveness to therapeutic interventions. The HOOS-PS substantially minimizes respondent burden while preserving interval-scale measurement integrity, establishing it as an essential tool for longitudinal research, clinical registries, and evidence-based orthopedic healthcare.

2. Keywords

Hip disability and Osteoarthritis Outcome Score-Physical function Short form, HOOS-PS, hip osteoarthritis, patient-reported outcome measures, Rasch analysis, physical functioning, joint arthroplasty, psychometrics, biomechanical limitation, functional status

3. Authors

The developmental and validation framework of the HOOS-PS was conceptualized and executed under the leadership of Professor Alan Tennant, PhD, FRCPEd, an internationally distinguished psychometrician and emeritus professor of rehabilitation studies at the University of Leeds (United Kingdom) and the Swiss Paraplegic Research (Nottwil, Switzerland). Professor Tennant spearheaded the initiative in collaboration with the Osteoarthritis Research Society International (OARSI) and the Outcome Measures in Rheumatology (OMERACT) task forces on standardized physical function assessment.

The original long-form instrument from which the items were extracted was developed by Ewa M. Roos, PT, PhD, Professor in the Department of Sports Science and Clinical Biomechanics at the University of Southern Denmark, Odense, Denmark, alongside clinical collaborators including Leif S. Lohmander, MD, PhD. The primary validation team led by Tennant sought to harmonize orthopedic evaluation scales across international borders, resolving item redundancy through modern Item Response Theory (IRT) methodologies.

4. Purpose

The fundamental purpose of the HOOS-PS is to provide an efficient, psychometrically sound, and minimally burdensome evaluation of physical function in adult and geriatric populations afflicted with hip osteoarthritis, femoroacetabular impingement, avascular necrosis, or systemic musculoskeletal joint deterioration. In modern orthopedic medicine and clinical epidemiology, comprehensive long-form questionnaires—such as the full 40-item HOOS or the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC)—frequently present substantial administrative hurdles. Long survey batteries lead to participant fatigue, missing data points, and administrative overhead in high-throughput clinical centers, multicenter longitudinal randomized controlled trials (RCTs), and national joint replacement registries.

To overcome these systemic barriers without sacrificing measurement precision, the HOOS-PS was engineered to assess physical function specifically as an individual, unidimensional construct. The theoretical rationale rests on the imperative for linear, interval-level quantification of functional limitation. Traditional sum scores derived from classical ordinal scales violate fundamental measurement axioms because the distance between categorical response categories (such as moving from “None” to “Mild” versus “Severe” to “Extreme”) is rarely mathematically equal across disparate physical tasks. Through modern Rasch scaling, the HOOS-PS establishes a calibrated hierarchical progression of functional demand, enabling clinicians and researchers to map patient capacity along an invariant continuum.

In clinical practice, the tool serves critical diagnostic, evaluative, and prognostic roles:

• Preoperative Baseline Stratification: Documenting the severity of dynamic functional compromise to substantiate surgical indications for total hip arthroplasty (THA) or joint-preserving osteotomy.
• Longitudinal Postoperative Recovery Tracking: Monitoring functional restitution at standardized clinical intervals (e.g., 6 weeks, 3 months, 12 months, and 5 years post-surgery) to evaluate implant performance and identify atypical recovery trajectories.
• Conservative and Rehabilitation Monitoring: Quantifying functional responses to non-pharmacological interventions, such as physical therapy, hydrotherapy, lifestyle modifications, and intra-articular corticosteroid or hyaluronic acid injections.
• Health Services and Comparative Effectiveness Research: Facilitating macro-level evaluations of healthcare quality, population health metrics, and institutional performance metrics across national and international orthopedic registries.

5. Psychological Construct

The HOOS-PS operationalizes the latent psychological and behavioral construct of perceived functional physical limitation resulting from hip joint pathology. Grounded in the World Health Organization’s International Classification of Functioning, Disability and Health (ICF) framework, the instrument focuses on the dynamic intersection between physiological body structures/functions (hip joint stability, pain-free range of motion, periarticular muscular torque) and activity execution in daily living.

Physical function in the context of hip disease is not merely an objective mechanical variable; it represents an integrated psychomotor construct shaped by subjective pain perception, kinesiophobia (fear of movement), learned avoidance behaviors, neuromuscular compensatory strategies, and task self-efficacy. Rather than assessing independent, isolated anatomical motions (such as passive degrees of internal rotation or flexion measured via a goniometer), the HOOS-PS captures the patient’s subjective appraisal of operational challenge encountered across common ecological motor patterns.

The five carefully retained items represent distinct functional archetypes along the functional difficulty continuum:

• Descending Stairs: A high-demand closed-kinetic-chain activity requiring eccentric contraction of the quadriceps and gluteal musculature, demanding substantial hip flexion, abduction stability, and controlled pelvic tilt under direct gravitational loading.
• Getting In/Out of Bath or Shower: An essential self-care mobility task necessitating unilateral weight-bearing balance while flexing, abducting, and internally/externally rotating the contralateral hip over an elevated physical barrier. This task measures both flexibility and protective postural control.
• Sitting: A low-demand, sustained static posture reflecting tolerance to prolonged hip flexion at approximately 90 degrees. Prolonged sitting frequently exacerbates intra-articular pressure, anterior impingement, and periarticular stiffness in compromised joints.
• Running: A high-impact, dynamic plyometric activity requiring high-velocity hip extension, eccentric shock attenuation, dynamic joint stabilization, and cardiovascular endurance. This item occupies the extreme end of the difficulty hierarchy, serving as a critical diagnostic threshold for active younger patients or high-functioning post-arthroplasty recipients.
• Twisting/Pivoting on Loaded Leg: A complex multi-planar rotational challenge that places maximal shear and torsional stresses across the femoral head, acetabular labrum, and supporting capsuloligamentous structures. This task evokes protective motor recruitment and highlights mechanical joint instability or impingement.

Importantly, the HOOS-PS intentionally consolidates these tasks into a single latent dimension. Unidimensionality implies that irrespective of whether the physical action is classified clinically as basic “Activity of Daily Living” or strenuous “Sport and Recreation,” all items function coherently as indicators of a singular underlying latent trait: physical difficulty attributable to the hip.

6. Theoretical Framework

The structural and conceptual foundation of the HOOS-PS is rooted in Modern Measurement Theory, specifically Item Response Theory (IRT) and the mathematical axioms of the Rasch Measurement Model (Rasch, 1960). In traditional Classical Test Theory (CTT), functional scales rely on the arbitrary summation of ordinal category scores. CTT assumes that all items possess equivalent measurement weight and that the distance between discrete Likert categories remains constant across varying items—an assumption routinely invalidated in clinical functional testing.

To overcome these limitations, Tennant and colleagues utilized Rasch analysis during the formal derivation of the HOOS-PS. The Rasch model asserts that the probability of a respondent endorsing a specific category of difficulty on an item is a logistic function of the difference between the person’s latent physical ability (or functional limitation level, denoted as θ) and the inherent difficulty calibration of the item (denoted as β):

P(X_ni = 1) = exp(θ_n – β_i) / [1 + exp(θ_n – β_i)]

Within this framework, the physical function domain was systematically scrutinized to ensure adherence to four fundamental psychometric requirements:

1. Unidimensionality: Verifying that all selected items reflect a singular latent continuum without contamination from orthogonal constructs such as psychological depression, generalized systemic fatigue, or contralateral joint pathology.
2. Monotonicity and Category Ordering: Ensuring that as a patient’s functional impairment increases, their probability of selecting higher ordinal difficulty categories (from “None” up to “Extreme”) systematically increases, validating that the 5-point rating response scale operates logically without threshold disordering.
3. Invariance across Subgroups (Absence of Differential Item Functioning – DIF): Guaranteeing that individual items behave identically across demographic and clinical strata. Specifically, the item calibration parameters must remain stable regardless of patient age, biological sex, country of origin, or surgical intervention status, ensuring cross-cultural and cross-demographic comparability.
4. Local Independence: Confirming that after conditioning on the latent trait of physical function, no residual statistical association exists between any pair of items, preventing artificial inflation of internal consistency metrics.

By extracting an optimized subset of five items from the 40-item parent scale that satisfied these rigorous Rasch parameters, the developers succeeded in building an interval-level measurement tool. The Rasch logit positions are converted via a standardized mathematical transformation table into an intuitive 0–100 interval scale, translating complex psychometric theory into direct clinical utility.

7. Validity

The construct, convergent, discriminant, and predictive validity of the HOOS-PS has been extensively documented in peer-reviewed orthopedic and rheumatological literature worldwide.

Construct and Convergent Validity

Convergent construct validity has been established by evaluating correlations between the HOOS-PS and alternative validated instruments measuring musculoskeletal impairment. In primary validation trials, the HOOS-PS showed powerful correlations with the parent HOOS subscales of Activities of Daily Living (HOOS-ADL; Pearson’s r > 0.85 to 0.92) and Sport and Recreation (HOOS-Sport; r > 0.80 to 0.88). Furthermore, the scale demonstrates strong convergent associations with the WOMAC physical function scale (r > 0.84 to 0.90) and the physical functioning subscale of the Medical Outcomes Study 36-Item Short Form Health Survey (SF-36 PF) (typically r = -0.65 to -0.76, reflecting inverted scoring directions).

Discriminant Validity

Discriminant validity is supported by significantly lower correlations between the HOOS-PS and divergent psychological or systemic constructs. For instance, correlations between the HOOS-PS and the SF-36 Mental Health (MH) subscale or the Mental Component Summary (MCS) consistently hover between r = -0.15 and -0.30, confirming that the scale specifically indexes lower-extremity biomechanical disability rather than generalized emotional distress or non-specific somatization.

Known-Groups and Predictive Validity

The HOOS-PS displays robust discriminative capacity across clinically differentiated patient groups. Statistically significant score gradations are observed when stratifying patients across Kellgren-Lawrence radiographic grades (grades 1–4), with severe joint space narrowing and subchondral sclerosis correlating with markedly elevated HOOS-PS impairment scores (p < 0.001). Moreover, the scale demonstrates high predictive validity regarding patient decisions to proceed with elective total hip replacement surgery, showing substantial divergence between conservative management candidates and patients listed for imminent surgical intervention.

Responsiveness and Longitudinal Validity

In prospective arthroplasty cohorts, the HOOS-PS demonstrates high longitudinal responsiveness. The Standardized Response Mean (SRM) and Effect Size (ES) 6 to 12 months post-total hip arthroplasty routinely exceed 1.20 to 1.80, indicating high sensitivity to surgical resolution of joint pathology. The Minimal Clinically Important Difference (MCID) or Minimal Important Change (MIC) has been empirically established in post-arthroplasty cohorts within a range of 9.0 to 14.5 points on the 0–100 metric, providing clear benchmarks for clinical trial success.

8. Reliability

The reliability of the HOOS-PS has been confirmed across diverse linguistic adaptations and clinical environments, satisfying rigorous international standards for individual-level monitoring as outlined by the COSMIN (COnsensus-based Standards for the selection of health Measurement INstruments) initiative.

Internal Consistency

Across validation cohorts, the HOOS-PS exhibits strong internal consistency. In the foundational development study by Tennant et al. (2007), the Person Separation Index (PSI)—the Rasch equivalent of Cronbach’s alpha—was reported at 0.83. Subsequent international validation cohorts (including French, German, Dutch, Swedish, and Portuguese adaptations) have documented Cronbach’s alpha coefficients ranging consistently between 0.79 and 0.89. This range indicates optimal item coherence without excessive informational redundancy.

Test-Retest Reliability

Temporal stability evaluated among stable clinical patients over intervals ranging from 48 hours to 2 weeks demonstrates high test-retest reliability. The Intraclass Correlation Coefficient (ICC) for agreement in two-way random-effects models consistently spans from 0.81 to 0.94. Standard Error of Measurement (SEM) values are low, typically calculated between 4.2 and 6.8 points on the 0–100 scale. The Smallest Detectable Change (SDC), reflecting measurement error at the individual level (SDC_individual = 1.96 × √2 × SEM), is situated between 11.6 and 18.8 points, confirming that observed changes beyond this threshold reflect authentic alterations in physical functioning.

9. Factor Analysis

The architectural validity of the HOOS-PS was determined through iterative factor analytical and structural equation modeling protocols, confirming its strict unidimensional factor structure.

Exploratory and Confirmatory Factor Analysis

During scale reduction, items from the parent HOOS instrument’s ADL (17 items) and Sport/Recreation (4 items) dimensions were pooled. In initial exploratory factor analyses (EFA), parallel analysis and scree plot inspections revealed a dominant first factor accounting for greater than 60% of the common variance. The ratio of the first-to-second eigenvalue routinely exceeded 4:1 to 5:1, satisfying classic criteria for essential unidimensionality.

Confirmatory Factor Analysis (CFA) applied to the definitive five-item configuration demonstrates model fit across independent cohorts:

• Comparative Fit Index (CFI): Values consistently exceed 0.96 (standard cut-off ≥ 0.95).
• Tucker-Lewis Index (TLI): Values consistently exceed 0.95.
• Root Mean Square Error of Approximation (RMSEA): Estimates range from 0.042 to 0.068 (standard acceptable threshold ≤ 0.08).
• Standardized Root Mean Square Residual (SRMR): Values consistently fall below 0.035.

Standardized Factor Loadings

The standardized factor loadings (λ) for the five items onto the single physical function latent factor are uniformly high (all p < 0.001):

• Descending stairs: λ = 0.76 – 0.84
• Getting in/out of bath or shower: λ = 0.72 – 0.80
• Sitting: λ = 0.62 – 0.74
• Running: λ = 0.78 – 0.86
• Twisting/pivoting on loaded leg: λ = 0.80 – 0.89

Item Fit within the Rasch Model

Item fit statistics in Rasch modeling further substantiate construct homogeneity. The mean square infit and outfit statistics for all five items systematically fall within the psychometrically acceptable corridor of 0.70 to 1.30. Residual correlation matrices demonstrate absence of local item dependency (all residual inter-item correlations r < 0.20), validating that each item makes a distinct, non-redundant contribution to the measurement of functional hip impairment.

10. Instrument / Measurement Tool

The structural characteristics, administration protocols, and metric scoring parameters of the HOOS-PS are detailed below:

• Test Name: Hip disability and Osteoarthritis Outcome Score-Physical function Short form
• Acronym: HOOS-PS
• Measurement Format: Patient-Reported Outcome Measure (PROM); paper-and-pencil self-administered questionnaire, digital clinical tablet, or web-based registry portal.
• Construct Assessed: Perceived functional physical limitation and dynamic disability secondary to hip joint disorders.
• Target Demographics: Adults and elderly populations experiencing hip osteoarthritis, periarticular hip pathomechanics, or undergoing surgical joint replacement.
• Administration Time: Approximately 1 to 2 minutes.
• Item Inventory: Exactly 5 task-specific functional items.
• Recall Period: The past week (the preceding 7 days).
• Response Categories: 5-point Likert scale: 0 = None, 1 = Mild, 2 = Moderate, 3 = Severe, 4 = Extreme.
• Scoring and Transformation Methodology:
  • Raw scores for each of the 5 items are summed, generating a raw ordinal score ranging from 0 (minimum difficulty) to 20 (maximum difficulty).
  • Raw sum scores cannot be evaluated as linear intervals. To enable mathematical comparisons, the raw sum (0–20) is converted using an established Rasch-based transformation metric into an interval score ranging from 0 to 100.
  • Score Interpretation: 0 represents no difficulty (perfect physical function), whereas 100 represents extreme functional limitation (maximum disability).
• Standard Rasch Conversion Table:
  • Raw 0 = Transformed 0.00
  • Raw 1 = Transformed 16.48
  • Raw 2 = Transformed 25.10
  • Raw 3 = Transformed 31.05
  • Raw 4 = Transformed 35.81
  • Raw 5 = Transformed 39.90
  • Raw 6 = Transformed 43.59
  • Raw 7 = Transformed 47.04
  • Raw 8 = Transformed 50.36
  • Raw 9 = Transformed 53.64
  • Raw 10 = Transformed 56.95
  • Raw 11 = Transformed 60.36
  • Raw 12 = Transformed 63.92
  • Raw 13 = Transformed 67.72
  • Raw 14 = Transformed 71.84
  • Raw 15 = Transformed 76.43
  • Raw 16 = Transformed 81.71
  • Raw 17 = Transformed 88.08
  • Raw 18 = Transformed 96.67
  • Raw 19 = Transformed 98.40
  • Raw 20 = Transformed 100.00
• Missing Data Imputation: If more than one item is missing, the total score cannot be calculated. If exactly one item is omitted, developers recommend mean imputation of the four completed items to derive an adjusted raw sum, or utilizing software calibrated for direct Rasch person-parameter estimation.

11. Permissions & Fee and Test Year

The Hip disability and Osteoarthritis Outcome Score-Physical function Short form (HOOS-PS) was originally formulated and published in 2007 by Professor Alan Tennant and the OARSI/OMERACT physical function working group. The parent HOOS instrument was developed in 2003 by Professor Ewa M. Roos.

Copyright and Accessibility: The HOOS-PS is distributed in the public domain for non-commercial clinical, academic, and scientific research use. No licensing fees or royalty payments are required for academic investigators, hospital clinics, or independent researchers. Official electronic versions, multi-language translations, and standardized user manuals are maintained and freely downloadable through academic clearinghouses, including the official HOOS repository (www.koos.nu).

Commercial Utilization: For-profit pharmaceutical trials, software companies integrating the HOOS-PS into proprietary commercial health applications, and commercial clinical research organizations (CROs) may require formal licensing clearance or authorization from the copyright holders. Investigators are expected to maintain the exact wording, structural presentation, and Rasch conversion rules without unauthorized modifications to preserve psychometric comparability.

12. References

Davis, A. M., Perruccio, A. V., Canizares, M., Tennant, A., Hawker, G. A., Conaghan, P. G., Roos, E. M., Jordan, J. M., Maillefert, J. F., Dougados, M., & Lohmander, L. S. (2008). Comparative, validity and responsiveness of the HOOS-PS and KOOS-PS to the WOMAC physical function subscale in total joint replacement for osteoarthritis. Osteoarthritis and Cartilage, 16(Suppl 4), S205–S206. https://doi.org/10.1016/S1063-4584(08)60515-5

de Groot, I. B., Reijman, M., Terwee, C. B., Bierma-Zeinstra, S. M., Favejee, C., Roos, E. M., & Verhaar, J. A. (2008). Validation study of the Dutch Hip disability and Osteoarthritis Outcome Score (HOOS). Osteoarthritis and Cartilage, 16(2), 209–215. https://doi.org/10.1016/j.joca.2007.06.012

Klassbo, M., Larsson, E., & Mannevik, E. (2003). Hip disability and Osteoarthritis Outcome Score. An extension of the Western Ontario and McMaster Universities Osteoarthritis Index. Scandinavian Journal of Rheumatology, 32(1), 46–51. https://doi.org/10.1080/03009740310000409

Rasch, G. (1960). Probabilistic models for some intelligence and attainment tests. Danish Institute for Educational Research.

Roos, E. M., & Lohmander, L. S. (2003). The Knee injury and Osteoarthritis Outcome Score (KOOS): From joint injury to osteoarthritis. Health and Quality of Life Outcomes, 1, Article 64. https://doi.org/10.1186/1477-7525-1-64

Tennant, A., Conaghan, P. G., & OARSI-OMERACT Physical Function Working Group. (2007). Derivation of the Hip disability and Osteoarthritis Outcome Score – Physical function Short form (HOOS-PS). Osteoarthritis and Cartilage, 15(Suppl B), B116–B117.

van den Akker-Scheek, I., Stevens, M., Groothoff, J. W., Bulstra, S. K., & Zijlstra, W. (2007). Preoperative considerations for total hip or knee arthroplasty: Assessing physical functioning. Physical Therapy, 87(6), 700–710. https://doi.org/10.2522/ptj.20060086

13. Items of the Scale