Vu Quynh Mai1, Sun Sun2,3, Hoang Van Minh4, Nan Luo5, Kim Bao Giang6, Lars Lindholm2, Klas Goran Sahlen2. 1. Center for Population Health Sciences, Hanoi University of Public Health, Hanoi, Vietnam. vqm@huph.edu.vn. 2. Department of Epidemiology and Global Health, Umeå University, Umeå, Sweden. 3. Research group Health Outcomes and Economic Evaluation, Department of Learning, Informatics, Management and Ethics, Karolinska Institutet, 171 77, Stockholm, Sweden. 4. Hanoi University of Public Health, Hanoi, Vietnam. 5. Saw Swee Hock School of Public Health, National University of Singapore, Singapore, Singapore. 6. Hanoi Medical University, Hanoi, Vietnam.
Abstract
PURPOSE: The objective of this study was to develop an EQ-5D-5L value set based on the health preferences of the general adult population of Vietnam. METHODS: The EQ-VT protocol version 2.1 was applied. Multi-stage stratified cluster sampling was employed to recruit a nationally representative sample. Both composite time trade-off (C-TTO) and discrete choice experiment (DCE) methods were used. Several modelling approaches were considered including hybrid; tobit; panel and heteroscedastic models. First, models using C-TTO or DCE data were tested separately. Then possibility of combining the C-TTO and DCE data was examined. Hybrid models were tested if it was sensible to combine both types of data. The best-performing model was selected based on both the consistency of the results produced and the degree to which models used all the available data. RESULTS: Data from 1200 respondents representing the general Vietnamese adult population were included in the analyses. Only the DCE Logit model and the regular Hybrid model that uses all available data produced consistent results. As the priority was to use all available data if possible, the hybrid model was selected to generate the Vietnamese value set. Mobility had the largest effect on health state values, followed by pain/discomfort, usual activities, anxiety/depression and self-care. The Vietnam values ranged from - 0.5115 to 1. CONCLUSION: This is the first value set for EQ-5D-5L based on social preferences obtained from a nationally representative sample in Vietnam. The value set will likely play a key role in economic evaluations and health technology assessments in Vietnam.
PURPOSE: The objective of this study was to develop an EQ-5D-5L value set based on the health preferences of the general adult population of Vietnam. METHODS: The EQ-VT protocol version 2.1 was applied. Multi-stage stratified cluster sampling was employed to recruit a nationally representative sample. Both composite time trade-off (C-TTO) and discrete choice experiment (DCE) methods were used. Several modelling approaches were considered including hybrid; tobit; panel and heteroscedastic models. First, models using C-TTO or DCE data were tested separately. Then possibility of combining the C-TTO and DCE data was examined. Hybrid models were tested if it was sensible to combine both types of data. The best-performing model was selected based on both the consistency of the results produced and the degree to which models used all the available data. RESULTS: Data from 1200 respondents representing the general Vietnamese adult population were included in the analyses. Only the DCE Logit model and the regular Hybrid model that uses all available data produced consistent results. As the priority was to use all available data if possible, the hybrid model was selected to generate the Vietnamese value set. Mobility had the largest effect on health state values, followed by pain/discomfort, usual activities, anxiety/depression and self-care. The Vietnam values ranged from - 0.5115 to 1. CONCLUSION: This is the first value set for EQ-5D-5L based on social preferences obtained from a nationally representative sample in Vietnam. The value set will likely play a key role in economic evaluations and health technology assessments in Vietnam.
Entities:
Keywords:
EQ-5D-5L; Generic measures; Utility; Value set
Authors: Anh Toan Ngo; Long Hoang Nguyen; Anh Kim Dang; Men Thi Hoang; Trang Huyen Thi Nguyen; Giang Thu Vu; Hoa Thi Do; Bach Xuan Tran; Carl A Latkin; Roger C M Ho; Cyrus S H Ho Journal: PLoS One Date: 2021-06-08 Impact factor: 3.240
Authors: Jongnam Hwang; Giang Thu Vu; Bach Xuan Tran; Thu Hong Thi Nguyen; Bang Van Nguyen; Long Hoang Nguyen; Huong Lan Thi Nguyen; Carl A Latkin; Cyrus S H Ho; Roger C M Ho Journal: PLoS One Date: 2020-06-25 Impact factor: 3.240