Jiayuan Wu1, Liren Hu2, Gaohua Zhang2, Qilian Liang3, Qiong Meng4, Chonghua Wan5. 1. Nutritional Department, The Affiliated Hospital of Guangdong Medical College, Zhanjiang, 524001, Guangdong, China. 2. School of Pubic Health, Guangdong Medical College, Zhanjiang, 524023, Guangdong, China. 3. Department of Oncology, The Affiliated Hospital of Guangdong Medical College, Zhanjiang, 524001, Guangdong, China. lianqilian@gdmc.edu.cn. 4. School of Public Health, Kunming Medical University, Kunming, 650500, Yunnan, China. 5. School of Humanities and Management, Research Center on Quality of Life and Applied Psychology, Guangdong Medical College, Dongguan, 523808, Guangdong, China. wanchh@hotmail.com.
Abstract
OBJECTIVES: This research was designed to develop a nasopharyngeal cancer (NPC) scale based on quality of life (QOL) instruments for cancer patients (QLICP-NA). This scale was developed by using a modular approach and was evaluated by classical test and generalizability theories. METHODS: Programmed decision procedures and theories on instrument development were applied to create QLICP-NA V2.0. A total of 121 NPC inpatients were assessed using QLICP-NA V2.0 to measure their QOL data from hospital admission until discharge. Scale validity, reliability, and responsiveness were evaluated by correlation, factor, parallel, multi-trait scaling, and t test analyses, as well as by generalizability (G) and decision (D) studies of the generalizability theory. RESULTS: Results of multi-trait scaling, correlation, factor, and parallel analyses indicated that QLICP-NA V2.0 exhibited good construct validity. The significant difference of QOL between the treated and untreated NPC patients indicated a good clinical validity of the questionnaire. The internal consistency (α) and test-retest reliability coefficients (intra-class correlations) of each domain, as well as the overall scale, were all >0.70. Ceiling effects were not found in all domains and most facets, except for common side effects (24.8 %) in the domain of common symptoms and side effects, tumor early symptoms (27.3 %) and therapeutic side effects (23.2 %) in specific domain, whereas floor effects did not exist in each domain/facet. The overall changes in the physical and social domains were significantly different between pre- and post-treatments with a moderate effective size (standard response mean) ranging from 0.21 to 0.27 (p < 0.05), but these changes were not obvious in the other domains, as well as in the overall scale. Scale reliability was further confirmed by G coefficients and index of dependability, with more exact variance components based on generalizability theory. CONCLUSIONS: QLICP-NA V2.0 exhibited reasonable degrees of validity, reliability, and responsiveness. However, this scale must be further improved before it can be used as a practical instrument to evaluate the QOL of NPC patients in China.
OBJECTIVES: This research was designed to develop a nasopharyngeal cancer (NPC) scale based on quality of life (QOL) instruments for cancerpatients (QLICP-NA). This scale was developed by using a modular approach and was evaluated by classical test and generalizability theories. METHODS: Programmed decision procedures and theories on instrument development were applied to create QLICP-NA V2.0. A total of 121 NPC inpatients were assessed using QLICP-NA V2.0 to measure their QOL data from hospital admission until discharge. Scale validity, reliability, and responsiveness were evaluated by correlation, factor, parallel, multi-trait scaling, and t test analyses, as well as by generalizability (G) and decision (D) studies of the generalizability theory. RESULTS: Results of multi-trait scaling, correlation, factor, and parallel analyses indicated that QLICP-NA V2.0 exhibited good construct validity. The significant difference of QOL between the treated and untreated NPCpatients indicated a good clinical validity of the questionnaire. The internal consistency (α) and test-retest reliability coefficients (intra-class correlations) of each domain, as well as the overall scale, were all >0.70. Ceiling effects were not found in all domains and most facets, except for common side effects (24.8 %) in the domain of common symptoms and side effects, tumor early symptoms (27.3 %) and therapeutic side effects (23.2 %) in specific domain, whereas floor effects did not exist in each domain/facet. The overall changes in the physical and social domains were significantly different between pre- and post-treatments with a moderate effective size (standard response mean) ranging from 0.21 to 0.27 (p < 0.05), but these changes were not obvious in the other domains, as well as in the overall scale. Scale reliability was further confirmed by G coefficients and index of dependability, with more exact variance components based on generalizability theory. CONCLUSIONS: QLICP-NA V2.0 exhibited reasonable degrees of validity, reliability, and responsiveness. However, this scale must be further improved before it can be used as a practical instrument to evaluate the QOL of NPCpatients in China.
Entities:
Keywords:
Classical test theory; Generalizability theory; Nasopharyngeal cancer; Quality of life; Standardized response mean
Authors: Paul Crits-Christoph; Jennifer Johnson; Robert Gallop; Mary Beth Connolly Gibbons; Sarah Ring-Kurtz; Jessica L Hamilton; Xin Tu Journal: Psychother Res Date: 2011-05
Authors: David C Buitenweg; Ilja L Bongers; Dike van de Mheen; Hans A M van Oers; Chijs Van Nieuwenhuizen Journal: Int J Methods Psychiatr Res Date: 2018-05-25 Impact factor: 4.035