Background Cardiovascular diseases seriously threaten the health of the elderly in our country and bring a huge economic burden to our country. Accurate screening of risk assessment tools for early cardiovascular diseases can specifically prevent and delay the development of cardiovascular diseases, studies have suggested that commonly used cardiovascular risk scoring tools are not effective in predicting risk in the elderly, but there is a lack of relevant external validation experimental studies, and there is no well-recognized effective risk prediction tool for elderly patients.
Objective To independently validate and compare the efficacy of five commonly used cardiovascular risk assessment tools in China and abroad in predicting 10-year cardiovascular disease risk in our community elderly population, to explore the cardiovascular disease risk assessment tools suitable for our community elderly population, and to provide theoretical basis and support for the prevention and treatment of cardiovascular disease at the grass-roots level.
Methods From January 2012 to December 2013, more than 20 000 people in the Liuyuan Community Health Service Centre and Runda Community Health Service Centre in Suzhou Province took part in community medical examinations. According to the inclusion and exclusion criteria, three hundred and forty-four people were identified, data on age, sex, region of residence, systolic blood pressure, diastolic blood pressure, BMI, waist circumference, total cholesterol, high-density lipoprotein cholesterol, triglyceride, atrial fibrillation, Left ventricular hypertrophy, cardiovascular disease, hypertension, diabetes, and smoking were collected from the start-up medical examination system In the follow-up survey of chronic diseases, the current smoking status and smoking age, whether taking antihypertensive drugs, family history of cardiovascular disease, prevalence of hypertension and drug use of hypertension, prevalence of diabetes mellitus, and whether and when cardiovascular disease occurred from January 2014 to December 2022 were improved. The outcome events were stroke and coronary heart disease, and were divided into positive group and negative group, all subjects were assessed with the Framingham Heart Risk Score (FRS-CVD) , the modified Framingham Stroke Scale (R-FSRS) , the European systemic coronary risk assessment scale 2019 (SCORE) , Chinese 10-year ischemic cardiovascular disease risk (ICVD) , and the China-PAR cardiovascular disease risk assessment, concordance Index (C-index) , Hosmer-Lemeshow χ2 and calibration chart were used to evaluate the relationship between the results and actual cardiovascular disease incidence.
Results Of the 344 enrolled investigators, with a mean age of 62 (56, 68) years, as of December 2022, 169 had developed cardiovascular disease and 175 had not. Two groups of people were carried out 5 kinds of cardiovascular disease risk score, cardiovascular disease risk assessment tool forecast and the actual incidence of the comparative analysis results are as follows, discrimination test: C-index of FRS-CVD was 0.711 (95%CI=0.658-0.764) , C-index of R-FSRS was 0.728 (95%CI=0.675-0.781) , SCORE: C-index was 0.724 (95%CI=0.671-0.777) , ICVD: C-index was 0.727 (95%CI=0.674-0.779) , China-PAR: C-index was 0.735 (95%CI=0.682-0.788) ; Hosmer-Lemeshow test calibration status: FRS-CVD: χ2 =16.789 (P=0.032) , R-FSRS: χ2=11.019 (P=0.201) , SCORE: χ2=20.396 (P=0.002) , ICVD: χ2=24.311 (P=0.001) , China-PAR: χ2=15.149 (P=0.056) ; R-FSRS is the best calibration. In men, model discrimination: FRS-CVD: C-index was 0.642 (95%CI=0.577-0.707) , R-FSRS: C-index was 0.646 (95%CI=0.581-0.710) , SCORE: C-index was 0.646 (95%CI=0.581-0.711) , ICVD: C-index was 0.628 (95%CI=0.563-0.693) , China-PAR: C-index was 0.636 (95%CI=0.571-0.700) ; Hosmer-Lemeshow test calibration status: FRS-CVD: χ2=7.371 (P=0.288) , R-FSRS: χ2=8.470 (P=0.293) , SCORE: χ2=5.146 (P=0.525) , ICVD: χ2=6.103 (P=0.412) , China-PAR: χ2=9.555 (P=0.298) , SCORE was calibrated best in the calibration diagram. Model discrimination among women: FRS-CVD: C-index was 0.698 (95%CI=0.633-0.762) , R-FSRS: C-index was 0.731 (95%CI=0.666-0.795) , SCORE: C-index was 0.733 (95%CI=0.668-0.798) , ICVD: C-index was 0.747 (95%CI=0.682-0.811) , China-PAR: C-index was 0.754 (95%CI=0.689-0.818) ; Hosmer-Lemeshow test calibration status: FRS-CVD: χ2=14.515 (P=0.069) , R-FSRS: χ2=12.175 (P=0.032) , SCORE: χ2=9.611 (P=0.022) , ICVD: χ2=19.349 (P=0.007) , China-PAR: χ2=12.372 (P=0.135) , China-PAR calibration is the best in calibration chart.
Conclusion R-FSRS model has a good performance in predicting the risk of cardiovascular disease in the elderly, especially in the elderly population. SCORE model did better in predicting cardiovascular disease risk in older men, while China-PAR model did better in predicting cardiovascular disease risk in older women. Therefore, the application of these assessment tools will help doctors to accurately predict the risk of cardiovascular disease in the elderly, and formulate corresponding prevention and treatment strategies.