A Rasch Analysis of Assessments of Morning and Evening Fatigue in Oncology Patients Using the Lee Fatigue Scale

Anners Lerdal, Anders Kottorp, Caryl Gay, Bradley Aouizerat, Kathryn A. Lee, Christine Miaskowski

Research output: Contribution to journalArticle

Abstract

Context: To accurately investigate diurnal variations in fatigue, a measure needs to be psychometrically sound and demonstrate stable item function in relationship to time of day. Rasch analysis is a modern psychometric approach that can be used to evaluate these characteristics. Objectives: To evaluate, using Rasch analysis, the psychometric properties of the Lee Fatigue Scale (LFS) in a sample of oncology patients. Methods: The sample comprised 587 patients (mean age 57.3 ± 11.9 years, 80% women) undergoing chemotherapy for breast, gastrointestinal, gynecological, or lung cancer. Patients completed the 13-item LFS within 30 minutes of awakening (i.e., morning fatigue) and before going to bed (i.e., evening fatigue). Rasch analysis was used to assess validity and reliability. Results: In initial analyses of differential item function, eight of the 13 items functioned differently depending on whether the LFS was completed in the morning or in the evening. Subsequent analyses were conducted separately for the morning and evening fatigue assessments. Nine of the morning fatigue items and 10 of the evening fatigue items demonstrated acceptable goodness-of-fit to the Rasch model. Principal components analyses indicated that both morning and evening assessments demonstrated unidimensionality. Person-separation indices indicated that both morning and evening fatigue scales were able to distinguish four distinct strata of fatigue severity. Conclusion: Excluding four items from the morning fatigue scale and three items from the evening fatigue scale improved the psychometric properties of the LFS for assessing diurnal variations in fatigue severity in oncology patients.

Original languageEnglish (US)
JournalJournal of Pain and Symptom Management
DOIs
StateAccepted/In press - 2016

Fingerprint

Fatigue
Psychometrics
Gastrointestinal Neoplasms
Principal Component Analysis
Reproducibility of Results
Lung Neoplasms
Breast

Keywords

  • Cancer
  • Diurnal pattern
  • Fatigue
  • Psychometrics
  • Quality of life
  • Rasch analysis
  • Symptoms

ASJC Scopus subject areas

  • Anesthesiology and Pain Medicine
  • Clinical Neurology
  • Nursing(all)

Cite this

A Rasch Analysis of Assessments of Morning and Evening Fatigue in Oncology Patients Using the Lee Fatigue Scale. / Lerdal, Anners; Kottorp, Anders; Gay, Caryl; Aouizerat, Bradley; Lee, Kathryn A.; Miaskowski, Christine.

In: Journal of Pain and Symptom Management, 2016.

Research output: Contribution to journalArticle

@article{050e4d53173847fb90613364718950ee,
title = "A Rasch Analysis of Assessments of Morning and Evening Fatigue in Oncology Patients Using the Lee Fatigue Scale",
abstract = "Context: To accurately investigate diurnal variations in fatigue, a measure needs to be psychometrically sound and demonstrate stable item function in relationship to time of day. Rasch analysis is a modern psychometric approach that can be used to evaluate these characteristics. Objectives: To evaluate, using Rasch analysis, the psychometric properties of the Lee Fatigue Scale (LFS) in a sample of oncology patients. Methods: The sample comprised 587 patients (mean age 57.3 ± 11.9 years, 80{\%} women) undergoing chemotherapy for breast, gastrointestinal, gynecological, or lung cancer. Patients completed the 13-item LFS within 30 minutes of awakening (i.e., morning fatigue) and before going to bed (i.e., evening fatigue). Rasch analysis was used to assess validity and reliability. Results: In initial analyses of differential item function, eight of the 13 items functioned differently depending on whether the LFS was completed in the morning or in the evening. Subsequent analyses were conducted separately for the morning and evening fatigue assessments. Nine of the morning fatigue items and 10 of the evening fatigue items demonstrated acceptable goodness-of-fit to the Rasch model. Principal components analyses indicated that both morning and evening assessments demonstrated unidimensionality. Person-separation indices indicated that both morning and evening fatigue scales were able to distinguish four distinct strata of fatigue severity. Conclusion: Excluding four items from the morning fatigue scale and three items from the evening fatigue scale improved the psychometric properties of the LFS for assessing diurnal variations in fatigue severity in oncology patients.",
keywords = "Cancer, Diurnal pattern, Fatigue, Psychometrics, Quality of life, Rasch analysis, Symptoms",
author = "Anners Lerdal and Anders Kottorp and Caryl Gay and Bradley Aouizerat and Lee, {Kathryn A.} and Christine Miaskowski",
year = "2016",
doi = "10.1016/j.jpainsymman.2015.12.331",
language = "English (US)",
journal = "Journal of Pain and Symptom Management",
issn = "0885-3924",
publisher = "Elsevier Inc.",

}

TY - JOUR

T1 - A Rasch Analysis of Assessments of Morning and Evening Fatigue in Oncology Patients Using the Lee Fatigue Scale

AU - Lerdal, Anners

AU - Kottorp, Anders

AU - Gay, Caryl

AU - Aouizerat, Bradley

AU - Lee, Kathryn A.

AU - Miaskowski, Christine

PY - 2016

Y1 - 2016

N2 - Context: To accurately investigate diurnal variations in fatigue, a measure needs to be psychometrically sound and demonstrate stable item function in relationship to time of day. Rasch analysis is a modern psychometric approach that can be used to evaluate these characteristics. Objectives: To evaluate, using Rasch analysis, the psychometric properties of the Lee Fatigue Scale (LFS) in a sample of oncology patients. Methods: The sample comprised 587 patients (mean age 57.3 ± 11.9 years, 80% women) undergoing chemotherapy for breast, gastrointestinal, gynecological, or lung cancer. Patients completed the 13-item LFS within 30 minutes of awakening (i.e., morning fatigue) and before going to bed (i.e., evening fatigue). Rasch analysis was used to assess validity and reliability. Results: In initial analyses of differential item function, eight of the 13 items functioned differently depending on whether the LFS was completed in the morning or in the evening. Subsequent analyses were conducted separately for the morning and evening fatigue assessments. Nine of the morning fatigue items and 10 of the evening fatigue items demonstrated acceptable goodness-of-fit to the Rasch model. Principal components analyses indicated that both morning and evening assessments demonstrated unidimensionality. Person-separation indices indicated that both morning and evening fatigue scales were able to distinguish four distinct strata of fatigue severity. Conclusion: Excluding four items from the morning fatigue scale and three items from the evening fatigue scale improved the psychometric properties of the LFS for assessing diurnal variations in fatigue severity in oncology patients.

AB - Context: To accurately investigate diurnal variations in fatigue, a measure needs to be psychometrically sound and demonstrate stable item function in relationship to time of day. Rasch analysis is a modern psychometric approach that can be used to evaluate these characteristics. Objectives: To evaluate, using Rasch analysis, the psychometric properties of the Lee Fatigue Scale (LFS) in a sample of oncology patients. Methods: The sample comprised 587 patients (mean age 57.3 ± 11.9 years, 80% women) undergoing chemotherapy for breast, gastrointestinal, gynecological, or lung cancer. Patients completed the 13-item LFS within 30 minutes of awakening (i.e., morning fatigue) and before going to bed (i.e., evening fatigue). Rasch analysis was used to assess validity and reliability. Results: In initial analyses of differential item function, eight of the 13 items functioned differently depending on whether the LFS was completed in the morning or in the evening. Subsequent analyses were conducted separately for the morning and evening fatigue assessments. Nine of the morning fatigue items and 10 of the evening fatigue items demonstrated acceptable goodness-of-fit to the Rasch model. Principal components analyses indicated that both morning and evening assessments demonstrated unidimensionality. Person-separation indices indicated that both morning and evening fatigue scales were able to distinguish four distinct strata of fatigue severity. Conclusion: Excluding four items from the morning fatigue scale and three items from the evening fatigue scale improved the psychometric properties of the LFS for assessing diurnal variations in fatigue severity in oncology patients.

KW - Cancer

KW - Diurnal pattern

KW - Fatigue

KW - Psychometrics

KW - Quality of life

KW - Rasch analysis

KW - Symptoms

UR - http://www.scopus.com/inward/record.url?scp=84961912077&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84961912077&partnerID=8YFLogxK

U2 - 10.1016/j.jpainsymman.2015.12.331

DO - 10.1016/j.jpainsymman.2015.12.331

M3 - Article

JO - Journal of Pain and Symptom Management

JF - Journal of Pain and Symptom Management

SN - 0885-3924

ER -