BACKGROUND: Development of a validated risk prediction model for future cardiovascular disease (CVD) in Australians is a high priority for cardiovascular health strategies. DESIGN: Recalibration of the SCORE (Systematic COronary Risk Evaluation) risk chart based on Australian national mortality data and average major CVD risk factor levels. METHODS: Australian national mortality data (2003-2005) were used to estimate 10-year cumulative CVD mortality rates for people aged 40-74 years. Average age-specific and sex-specific levels of systolic blood pressure, total cholesterol and prevalence of current smoking were generated from data obtained in eight Australian large-scale population-based surveys undertaken from the late 1980s. The SCORE risk chart was then recalibrated by applying hazard ratios for 10-year CVD mortality obtained in the SCORE project. Discrimination and calibration of the recalibrated model was evaluated and compared with that of the original SCORE and Framingham equations in the Blue Mountains Eye Study in Australia using Harrell's c and Hosmer-Lemeshow chi statistics, respectively. RESULTS: An Australian risk prediction chart for CVD mortality was derived. Among 1998 Blue Mountains Eye Study participants aged 49-74 years with neither CVD nor diabetes at baseline, the Harrell's c statistics for the Australian risk prediction chart for CVD mortality were 0.76 (95% confidence interval: 0.69-0.84) and 0.71 (confidence interval: 0.62-0.80) in men and women, respectively. The corresponding Hosmer-Lemeshow chi statistics, the measure of calibration, were 2.32 (P = 0.68) and 7.43 (P = 0.11), which were superior to both the SCORE and Framingham equations. CONCLUSION: This new tool provides a valid and reliable method to predict risk of CVD mortality in the general Australian population.
BACKGROUND: Development of a validated risk prediction model for future cardiovascular disease (CVD) in Australians is a high priority for cardiovascular health strategies. DESIGN: Recalibration of the SCORE (Systematic COronary Risk Evaluation) risk chart based on Australian national mortality data and average major CVD risk factor levels. METHODS: Australian national mortality data (2003-2005) were used to estimate 10-year cumulative CVD mortality rates for people aged 40-74 years. Average age-specific and sex-specific levels of systolic blood pressure, total cholesterol and prevalence of current smoking were generated from data obtained in eight Australian large-scale population-based surveys undertaken from the late 1980s. The SCORE risk chart was then recalibrated by applying hazard ratios for 10-year CVD mortality obtained in the SCORE project. Discrimination and calibration of the recalibrated model was evaluated and compared with that of the original SCORE and Framingham equations in the Blue Mountains Eye Study in Australia using Harrell's c and Hosmer-Lemeshow chi statistics, respectively. RESULTS: An Australian risk prediction chart for CVD mortality was derived. Among 1998 Blue Mountains Eye Study participants aged 49-74 years with neither CVD nor diabetes at baseline, the Harrell's c statistics for the Australian risk prediction chart for CVD mortality were 0.76 (95% confidence interval: 0.69-0.84) and 0.71 (confidence interval: 0.62-0.80) in men and women, respectively. The corresponding Hosmer-Lemeshow chi statistics, the measure of calibration, were 2.32 (P = 0.68) and 7.43 (P = 0.11), which were superior to both the SCORE and Framingham equations. CONCLUSION: This new tool provides a valid and reliable method to predict risk of CVD mortality in the general Australian population.
Authors: Peter Ueda; Mark Woodward; Yuan Lu; Kaveh Hajifathalian; Rihab Al-Wotayan; Carlos A Aguilar-Salinas; Alireza Ahmadvand; Fereidoun Azizi; James Bentham; Renata Cifkova; Mariachiara Di Cesare; Louise Eriksen; Farshad Farzadfar; Trevor S Ferguson; Nayu Ikeda; Davood Khalili; Young-Ho Khang; Vera Lanska; Luz León-Muñoz; Dianna J Magliano; Paula Margozzini; Kelias P Msyamboza; Gerald Mutungi; Kyungwon Oh; Sophal Oum; Fernando Rodríguez-Artalejo; Rosalba Rojas-Martinez; Gonzalo Valdivia; Rainford Wilks; Jonathan E Shaw; Gretchen A Stevens; Janne S Tolstrup; Bin Zhou; Joshua A Salomon; Majid Ezzati; Goodarz Danaei Journal: Lancet Diabetes Endocrinol Date: 2017-01-24 Impact factor: 32.069