Modified Harris Hip Score as patient-reported outcome measure in osteoarthritic patients: psychometric properties of the Greek version

Sophia Stasi; George Papathanasiou; Afroditi Diochnou; Basiliki Polikreti; Antonios Chalimourdas; George A Macheras

doi:10.1177/1120700020901682

Modified Harris Hip Score as patient-reported outcome measure in osteoarthritic patients: psychometric properties of the Greek version

Hip Int. 2021 Jul;31(4):516-525. doi: 10.1177/1120700020901682. Epub 2020 Jan 27.

Authors

Sophia Stasi¹, George Papathanasiou¹, Afroditi Diochnou¹, Basiliki Polikreti¹, Antonios Chalimourdas¹, George A Macheras^{1

2}

Affiliations

¹ Laboratory of Neuromuscular and Cardiovascular Study of Motion, Physiotherapy Department, Faculty of Health and Care Sciences, University of West Attica, Egaleo, Attica, Greece.
² 4th Orthopaedic Department, "KAT" General Hospital of Attica, Kifissia, Attica, Greece.

PMID: 31984798
DOI: 10.1177/1120700020901682

Abstract

Introduction: This study explored the psychometric properties of the modified Harris Hip Score-Greek version (mHHS-Gr) as a patient-reported outcome (PRO) measure in osteoarthritic hip patients.

Methods: Internal consistency, test-retest reliability and reproducibility were evaluated in 90 patients aged >55 years. Construct validity was tested against Greek versions of the Lower Extremity Functional Scale (LEFS-Greek) and WOMAC Index (WOMAC-Gr), and the Timed Up and Go (TUG) and 9-stairs-ascend/descend (9S-A/D) tests. Known-groups validity was examined using TUG score (cut-off 13.5 s) as an estimate variable. Responsiveness was examined before and 4 weeks after direct anterior minimal invasive surgery.

Results: Reliability: Internal consistency was moderate (Cronbach's a = 0.614, p < 0.001). Test-retest reliability was excellent (ICC = 0.881, 95% CI, 0.824-0.920). Reproducibility: Floor and ceiling effects were both 1.1%; measurement error was 3.54 (p < 0.05); minimal important change was lower than minimal detectable change.Validity: mHHS-Gr correlated strongly with both LEFS-Greek and WOMAC-Gr (Pearson's r 0.801 and -0.783, respectively; p < 0.001). The questionnaire's correlations with TUG and 9S-A/D were also significant but moderate (Spearman's ρ: -0.547 and -0.575, respectively; p < 0.001). Known-groups validity showed that mHHS-Gr scores were significantly higher in participants with TUG < 13.5 seconds than in those with TUG > 13.5 seconds (p < 0.001). In ROC analysis, the cut-off point of 52.5 yielded sensitivity 81% and specificity 71%.Responsiveness: Standardised response mean and Guyatt's responsiveness statistic were greater than 0.8.

Discussion: mHHS-Gr showed significant moderate to excellent reliability, significant moderate to strong validity properties and excellent responsiveness. Overall, mHHS-Gr could be a reliable and valid PRO measure for assessing patients with osteoarthritis of the hip.

Keywords: Osteoarthritis-hip; patient-reported outcome measures; physical-performance measures; reliability; validity.

MeSH terms

Humans
Patient Reported Outcome Measures*
Psychometrics
ROC Curve
Reproducibility of Results
Surveys and Questionnaires