Emily Andrew1, Ziad Nehme2, Stephen Bernard3, Karen Smith4. 1. Department of Research and Evaluation, Ambulance Victoria, Melbourne, Australia; Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia. Electronic address: emily.andrew@ambulance.vic.gov.au. 2. Department of Research and Evaluation, Ambulance Victoria, Melbourne, Australia; Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia. Electronic address: ziad.nehme@ambulance.vic.gov.au. 3. Department of Research and Evaluation, Ambulance Victoria, Melbourne, Australia; Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia; Intensive Care Unit, Alfred Hospital, Melbourne, Australia. Electronic address: s.bernard@alfred.org.au. 4. Department of Research and Evaluation, Ambulance Victoria, Melbourne, Australia; Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia; School of Primary, Aboriginal and Rural Health Care, University of Western Australia, Perth, Australia. Electronic address: karen.smith@ambulance.vic.gov.au.
Abstract
INTRODUCTION: Although a number of validated health-related quality of life (HR-QOL) instruments exist for critical care populations, a standardised approach to assessing the HR-QOL of out-of-hospital cardiac arrest (OHCA) survivors has not been developed. We sought to compare the responses of 12-month OHCA survivors to three HR-QOL and functional recovery instruments, and assess instrument validity. METHODS: The Victorian Ambulance Cardiac Arrest Registry invited 12-month OHCA survivors to participate in telephone follow-up between January 2011 and December 2015. Responders provided answers to the 12 Item Short Form Health Survey (SF-12), Three-Level EuroQol-5D (EQ-5D-3L) and the Glasgow Outcome Scale-Extended (GOSE). The SF-12 was also used to derive the SF-6D. Responses were used to assess the interpretability and construct validity of the instruments. RESULTS: A total of 1188 patients and proxies responded. Large ceiling effects were observed for the EQ-5D-3L (patients=46%, proxies=23%). Substantial variability was also observed in SF-6D responses for patients who reported full health according to the EQ-5D-3L. For patient responders, the strongest correlations were observed between the EQ-5D-3L index score and SF-6D (ρ=0.65, p<0.001), and between the SF-6D and SF-12 physical component (ρ=0.69, p<0.001). The distribution of the SF-6D and EQ-5D-3L differed significantly for patients reporting a lower or upper moderate GOSE outcome and lower or upper good recovery (p<0.001 for all comparisons). CONCLUSIONS: The EQ-5D-3L demonstrated limited interpretability due to the presence of ceiling effects. However, the measurement properties of the SF-12, SF-6D and GOSE suggest that these may be useful measures of HR-QOL and functional recovery in OHCA survivors.
INTRODUCTION: Although a number of validated health-related quality of life (HR-QOL) instruments exist for critical care populations, a standardised approach to assessing the HR-QOL of out-of-hospital cardiac arrest (OHCA) survivors has not been developed. We sought to compare the responses of 12-month OHCA survivors to three HR-QOL and functional recovery instruments, and assess instrument validity. METHODS: The Victorian Ambulance Cardiac Arrest Registry invited 12-month OHCA survivors to participate in telephone follow-up between January 2011 and December 2015. Responders provided answers to the 12 Item Short Form Health Survey (SF-12), Three-Level EuroQol-5D (EQ-5D-3L) and the Glasgow Outcome Scale-Extended (GOSE). The SF-12 was also used to derive the SF-6D. Responses were used to assess the interpretability and construct validity of the instruments. RESULTS: A total of 1188 patients and proxies responded. Large ceiling effects were observed for the EQ-5D-3L (patients=46%, proxies=23%). Substantial variability was also observed in SF-6D responses for patients who reported full health according to the EQ-5D-3L. For patient responders, the strongest correlations were observed between the EQ-5D-3L index score and SF-6D (ρ=0.65, p<0.001), and between the SF-6D and SF-12 physical component (ρ=0.69, p<0.001). The distribution of the SF-6D and EQ-5D-3L differed significantly for patients reporting a lower or upper moderate GOSE outcome and lower or upper good recovery (p<0.001 for all comparisons). CONCLUSIONS: The EQ-5D-3L demonstrated limited interpretability due to the presence of ceiling effects. However, the measurement properties of the SF-12, SF-6D and GOSE suggest that these may be useful measures of HR-QOL and functional recovery in OHCA survivors.
Authors: Adam Viktorisson; Katharina S Sunnerhagen; Ulrika Pöder; Johan Herlitz; Åsa B Axelsson Journal: BMJ Open Date: 2018-06-06 Impact factor: 2.692
Authors: Adam Viktorisson; Katharina S Sunnerhagen; Dongni Johansson; Johan Herlitz; Åsa Axelsson Journal: BMJ Open Date: 2019-07-03 Impact factor: 2.692