Reliability analysis of the Chinese version of the Functional Assessment of Cancer Therapy - Leukemia (FACT-Leu) scale based on multivariate generalizability theory.

Yuanyuan Xiao
Yuanyuan Xiao
Kunming Medical University
Epidemiology, Biostatistics

Health Qual Life Outcomes 2017 May 4;15(1):93. Epub 2017 May 4.

West China School of Public Health, Sichuan University, Chengdu, Sichuan, 610041, China.

Background: The Functional Assessment of Cancer Therapy-Leukemia (FACT-Leu) scale, a leukemia-specific instrument for determining the health-related quality of life (HRQOL) in patients with leukemia, had been developed and validated, but there have been no reports on the development of a simplified Chinese version of this scale. This is a new exploration to analyze the reliability of the HRQOL measurement using multivariate generalizability theory (MGT). This study aimed to develop a Chinese version of the FACT-Leu scale and evaluate its reliability using MGT to provide evidence to support the revision and improvement of this scale.

Methods: The Chinese version of the FACT-Leu scale was developed by four steps: forward translation, backward translation, cultural adaptation and pilot-testing. The HRQOL was measured for eligible inpatients with leukemia using this scale to provide data. A single-facet multivariate Generalizability Study (G-study) design was demonstrated to estimate the variance-covariance components and then several Decision Studies (D-studies) with varying numbers of items were analyzed to obtain reliability coefficients and to understand how much the measurement reliability could be vary as the number of items in MGT changes.

Results: One-hundred and one eligible inpatients diagnosed with leukemia were recruited and completed the HRQOL measurement at the time of admission to the hospital. In the G-study, the variation component of the patient-item interaction was largest while the variation component of the item was the smallest for the four of five domains, except for the leukemia-specific (LEUS) domain. In the D-study, at the level of domain, the generalizability coefficients (G) and the indexes of dependability (Ф) for four of the five domains were approximately equal to or greater than 0.80 except for the Emotional Well-being (EWB) domain (>0.70 but <0.80). For the overall scale, the composite G and composite Ф coefficients were greater than 0.90. Based on the G coefficient and Ф coefficient, two decision options for revising this scale considering the number of items were obtained: one is a 37-item version while the other is a 45-item version.

Conclusion: The Chinese version of the FACT-Leu scale has good reliability as a whole based on the results of MGT and the implementation of MGT could lead to more informed decisions in complex questionnaire design and improvement.

Download full-text PDF

Source Listing
May 2017
63 Reads
2.120 Impact Factor

Publication Analysis

Top Keywords

fact-leu scale
chinese version
multivariate generalizability
hrqol measurement
assessment cancer
eligible inpatients
variation component
generalizability theory
functional assessment
version fact-leu
estimate variance-covariance
variance-covariance components
demonstrated estimate
design demonstrated
study g-study
g-study design
measurement reliability
understand measurement


(Supplied by CrossRef)

W Chen et al.
CA Cancer J Clin 2016

A Redaelli et al.
Expert Rev Anticancer Ther 2003

AM Stalfelt et al.
Qual Assur Health Care 1993

M Watson et al.
Qual Life Res 1996

F Efficace et al.
Qual Life Res 2014

D Cella et al.
Value Health 2012

DF Cella et al.
J Clin Oncol 1993

NA Gage et al.
Behav Disord 2014

L Berggraf et al.
Psychother Res 2012

C Hall et al.
Neurology 2015

C Iramaneerat et al.
Adv Health Sci Educ Theory Pract 2008

Similar Publications