Psychometric contribution for the evaluation of the quality of life

Authors

Keywords:

quality of life, item response theory, missing data, multiple imputation, latent class mixed model

Abstract

Introduction: Quality of life is important in oncology and is used as a primary variable in controlled cancer clinical trials. Analysis of quality of life data is quite complex, due to the multidimensionality and management of missing data.

Objectives: to formulate a methodology based on the item response theory for dimensionality reduction for categorical data; to propose a stratified imputation method for missing data patterns based on a simple overpopulation regression model; to fit mixed models of item response theory for imputed longitudinal data and to assess the impact of therapeutic vaccine treatments on quality of life.

Methods: This study included a sample of 1,107 patients included in 5 clinical trials to test the effectiveness and safety of the CIMAvaxEGF vaccine in lung cancer.

Results: The item response theory allowed us to reduce the dimensionality of the questionnaires. It was found that multiple imputation based on simple overpopulation regression provided better estimates. It was constructed a joint model of latent variables for longitudinal categorical data.

Conclusions: The dimension of the questionnaires was reduced. It was shown that the multiple imputation method through the simple regression model of overpopulation is the most appropriate. The proposed latent class mixed model allowed a better explanation of the variation in quality of life over time. Quality of life is predictive and prognostic of survival time for lung cancer.

Downloads

Download data is not yet available.

Published

2024-03-21

How to Cite

Viada González, C. E., Bouza Herrera, C. N., Ballesteros Rodríguez, F. J., Fors López, M. M., & Bringas Vega, M. L. (2024). Psychometric contribution for the evaluation of the quality of life. Anales De La Academia De Ciencias De Cuba, 14(1), e1405. Retrieved from https://revistaccuba.sld.cu/index.php/revacc/article/view/1405

Issue

Section

Biomedical sciences