Item response theory modeling and confirmatory factor analysis of the Geriatric Pain Measure (GPM-24) with a North American sample
The Journal of Pain · 2026
DOI: 10.1016/j.jpain.2026.106233 ↗Kurz gefasst
Fachartikel in der Zeitschrift The Journal of Pain, 2026. Verfasst von Kylie A Arsenault und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Psychology and Centre on Aging and Health, University of Regina, 3737 Wascana Parkway, Regina, SK S4S 0A2, Canada. Hauptthemen laut MEDLINE: Pain Measurement, Psychometrics, Geriatric Assessment, Models, Statistical. Die Arbeit nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Saskatchewan Health Research Foundation.
Abstract (Originalsprache)
The Geriatric Pain Measure (GPM-24) is a commonly used self-report pain measurement tool among older adults. To date, the GPM-24 has undergone few psychometric evaluations, with limited application of robust statistical methods such as confirmatory factor analysis (CFA). Item response theory (IRT) analysis is also needed to examine its properties at the item level. The objective of this study was to conduct a psychometric evaluation of the GPM-24 at both the factor- and item-level. In January 2024, data were collected via Qualtrics Panels from 486 older adults who completed the GPM-24. Reliability was examined using McDonald's omega coefficient, structural validity using CFA, and item properties using IRT modeling. The results showed that the overall GPM-24 and its subscales had acceptable-to-excellent reliability (ω =.77-.95). After removing two weak item cross-loadings, the five-factor model bordered on adequate fit: χ2 = 4.28, CFI =.895, TLI =.881, RMSEA =.082 [CI .077-.087], SRMR =.057. Each GPM-24 item was shown to adequately discriminate between respondents with low and high levels of self-reported pain (a >1), though items 13 and 24 were identified as having the lowest discriminatory and predictive efficacy and may warrant refinement. Item 10 was most effective at distinguishing between respondents with varying levels of self-reported pain (a = 4.74). This is the first study to demonstrate that the GPM-24 is an acceptably reliable and valid tool for evaluating self-reported pain among North American older adults. Future investigations should establish cross-validation with clinical and racially diverse samples. PERSPECTIVE: This article presents the psychometric properties and item performance of the Geriatric Pain Measure (GPM-24) when tested with a North American sample of older adults. Validation of this measure can help clinicians and researchers reliably assess the impact of pain on older adults' physical functioning, affect, and quality of life.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Psychology and Centre on Aging and Health, University of Regina, 3737 Wascana Parkway, Regina, SK S4S 0A2, Canada.
- School of Population and Public Health, University of British Columbia, 2206 East Mall, Vancouver, BC V6T 1Z3, Canada.
- Department of Psychology and Centre on Aging and Health, University of Regina, 3737 Wascana Parkway, Regina, SK S4S 0A2, Canada. Electronic address: hadjistt@uregina.ca.
Förderung
- Saskatchewan Health Research Foundation
Zitieren
APA
Arsenault, K. A., MacNab, Y. C., & Hadjistavropoulos, T. (2026). Item response theory modeling and confirmatory factor analysis of the Geriatric Pain Measure (GPM-24) with a North American sample. The Journal of Pain. https://doi.org/10.1016/j.jpain.2026.106233
Vancouver
Arsenault KA, MacNab YC, Hadjistavropoulos T. Item response theory modeling and confirmatory factor analysis of the Geriatric Pain Measure (GPM-24) with a North American sample. The Journal of Pain. 2026. doi:10.1016/j.jpain.2026.106233
BibTeX
@article{Arsenault2026,
author = {Arsenault, Kylie A and MacNab, Ying C and Hadjistavropoulos, Thomas},
title = {Item response theory modeling and confirmatory factor analysis of the Geriatric Pain Measure (GPM-24) with a North American sample},
journal = {The Journal of Pain},
year = {2026},
doi = {10.1016/j.jpain.2026.106233}
}
EUVIMED