Hosein Ameri1, Mahmood Yousefi2, Mehdi Yaseri3, Azin Nahvijou4, Mohammad Arab1, Ali Akbari Sari1. 1. a Department of Health Management and Economics, School of Public Health , Tehran University of Medical Sciences , Tehran , Iran. 2. b Iranian Center of Excellence in Health Management, School of Management and Medical Informatics, Health Economics Department , Tabriz University of Medical Sciences , Tabriz , Iran. 3. c Department of Epidemiology and Biostatistics, School of Public Health , Tehran University of Medical Sciences , Tehran , Iran. 4. d Cancer Research Center, Cancer Institute , Tehran University of Medical Sciences , Tehran , Iran.
Abstract
INTRODUCTION: Economic evaluation of healthcare interventions usually needs accurate data on utility and health-related quality-of-life scores. The aim of this study is to map QLQ-C30 scale score onto EQ-5D-5L and SF-6D utility values in colorectal cancer (CRC) patients. METHODS: EQ-5D-5L, SF-6D, and QLQ-C30 were completed by 252 patients with CRC who were referred to three cancer centers in Tehran between May and September 2017. Moreover, OLS, Tobit, and CLAD models were used to predict EQ-5D-5L and SF-6D values. The goodness of fit of models was evaluated using Pred R2 and Adj R2. In addition, their predictive performance was assessed by MAE, RMSE, ICC, MID, and Spearman's correlation coefficients between observed and predicted EQ-5D-5L and SF-6D values. Models were validated using a 10-fold cross-validation method. RESULTS: Considering the goodness of fit and predictive ability of models, the OLS Model 2 performed best for EQ-5D-5L (Adj R2 = 58.09%, Pred R2 = 58.93%, MAE = 0.0932, RMSE = 0.129) and the OLS Model 3 performed best for SF-6D (Adj R2 = 54.90%, Pred R2 = 55.62%, MAE = 0.0485, RMSE = 0.0634). CONCLUSION: Our results demonstrated that algorithms developed based on OLS Models 1 and 2 are the best for predicted EQ-5D-5L and SF-6D values, respectively.
INTRODUCTION: Economic evaluation of healthcare interventions usually needs accurate data on utility and health-related quality-of-life scores. The aim of this study is to map QLQ-C30 scale score onto EQ-5D-5L and SF-6D utility values in colorectal cancer (CRC) patients. METHODS: EQ-5D-5L, SF-6D, and QLQ-C30 were completed by 252 patients with CRC who were referred to three cancer centers in Tehran between May and September 2017. Moreover, OLS, Tobit, and CLAD models were used to predict EQ-5D-5L and SF-6D values. The goodness of fit of models was evaluated using Pred R2 and Adj R2. In addition, their predictive performance was assessed by MAE, RMSE, ICC, MID, and Spearman's correlation coefficients between observed and predicted EQ-5D-5L and SF-6D values. Models were validated using a 10-fold cross-validation method. RESULTS: Considering the goodness of fit and predictive ability of models, the OLS Model 2 performed best for EQ-5D-5L (Adj R2 = 58.09%, Pred R2 = 58.93%, MAE = 0.0932, RMSE = 0.129) and the OLS Model 3 performed best for SF-6D (Adj R2 = 54.90%, Pred R2 = 55.62%, MAE = 0.0485, RMSE = 0.0634). CONCLUSION: Our results demonstrated that algorithms developed based on OLS Models 1 and 2 are the best for predicted EQ-5D-5L and SF-6D values, respectively.
Entities:
Keywords:
EQ-5D-5L; QLQ-C30; SF-6D; colorectal cancer; quality of life
Authors: Aurelie Meunier; Alexandra Soare; Helene Chevrou-Severac; Karl-Johan Myren; Tatsunori Murata; Louise Longworth Journal: Appl Health Econ Health Policy Date: 2021-09-23 Impact factor: 3.686