Study On The Predictive Value And Clinical Utility Of Handgrip Strength For Cardiovascular Disease, Cancer And Death Risk In Chinese Community Populatio

Background and aimsLower levels of muscle strength,as measured by grip strength,are associated with increased risk of death and cardiovascular disease(CVD).However,most of the previous studies on grip strength and clinical outcomes came from Western countries,and there were few large-scale epidemiological studies in China.In addition,the association of grip strength with cancer incidence remains to be explored.With CVD as the leading cause of death,well-established predictive models have been used to stratify individuals for CVD risk,and whether the addition of grip strength testing as a quick and inexpensive method to the original model would increase predictive performance has not been investigated.In recent years,with the increasing development of predictive models,many scholars have pointed out that evaluating the additional predictive value of biomarkers requires not only evaluating the predictive performance of the model but also evaluating its clinical utility.However,few studies have evaluated both the predictive performance and clinical utility of biomarkers.Therefore,this project aims to describe the distribution characteristics of grip strength values in Chinese adults;explore the association of different grip strength levels with death,CVD and cancer;compare the strength of the association between grip strength and other anthropometric parameters and the risk of CVD and death for the first time;and explore the additional predictive value and clinical utility of grip strength on traditional CVD and mortality risk prediction models to fill relevant research gaps.MethodsThis study relies on the baseline survey part and follow-up part of the PURE China cohort study.PURE China is a large-scale,multi-center,nationally representative cohort study.From the beginning of 2005 to the end of 2009,PURE China researched data from eastern China(Beijing,Jiangsu,Shandong,and Liaoning),central China(Shanxi,Jiangxi,and Inner Mongolia),and western China(Yunnan,Qinghai,Shaanxi,Xinjiang,Sichuan).A sample survey was conducted in 115 communities(including 45 urban communities and 70 rural communities)in 12 provinces.A total of 47677 participants were included in the baseline survey.The follow-up survey is a round of follow-up every three years.Trained researchers go to various centers to conduct questionnaire surveys,clinical event collection,physical examination and telephone follow-ups.The follow-up events used in this study ended in September 2017.Taking into account the lack of judgement of the grip strength of Chinese people,referring to the previous literature,the interquartile range of age-specific gender grouping is used to divide the population into low grip strength,medium grip and high grip strength groups.Restricted cubic splines were used to explore the linear relationship between grip strength and clinical event endpoints.Use Cox shared frailty proportional hazard model to analyze time event data.The proportional hazard hypothesis is checked by the Schoenfeld residual test.All analysis of multi-factor adjustment models adopts three methods of increasing covariate analysis.The nature of the research is still observational,and the E-value method is used to quantify the influence of confounding on the effect value.Standardized processing was used to compare the strength of the association between different anthropometric parameters and clinical outcomes.The predictive value and clinical utility of grip strength on the risk of cardiovascular disease and death were measured using C-index,IDI(Integrated discrimination improvement),continuous NRI(Net reclassification improvement),binary NRI,three-category NRI,DCA(Decision curve analysis,decision curve),Net benefit and EFLYs(Event Free Life Year)were evaluated in the derivation population and the validation population respectively.Finally,the results are visualized by Nomogram(nomogram).ResultsThe baseline analysis included 46,072 baseline participants into the analysis.Among them,26,878 were women and 19,194 were men.The grip strength value and the grip strength BMI are different for different ages and genders,so the research on grip strength should take into account the basic characteristics of the population.After inclusion and exclusion criteria,39,969 participants who had undergone at least one follow-up were included,including 23,271 women and 16,698 men,with a median follow-up time of 8.9 years(interquartile range of follow-up time 6.7-9.9 years).A total of 1249 deaths(3.12%)occurred during this period,including 387(0.97%)CVD deaths and 416 cancer deaths(1.04%);1819 major CVD events(4.55%),including(473 myocardial infarctions)1.18%),1291 strokes(3,23%).Compared with the high grip strength group,the low grip strength group increased the risk of death from all causes by 45%(hazard ratio 1.22,95%CI:1.221.73,Model 3),and increased the risk of CVD death by 66%(risk ratio 1.66,95%CI:1.22-2.27,model 3),the risk of major CVD increased by 50%(hazard ratio 1.50,95%CI:1.34-1.69,model 3),and the risk of myocardial infarction increased by 74%(hazard ratio 1.74,95%CI:1.32)-2.30,model 3),the risk of stroke increased by 33%(hazard ratio 1.33,95%CI:1.13-1.56,model 3);and with the decrease in grip strength,the risk of the above clinical events increased(trend test test P value<0.001).As for the clinical outcome of cancer events,compared with the high grip strength group,there was no significant difference in the risk of the middle grip strength group and the low grip strength group(cancer death risk ratio 1.02,95%CI:0.76-1.38;cancer risk ratio 0.89,95%CI:0.72-1.11).When adjusting for confounding factors,the attributable risk of low grip strength for all-cause death,CVD death,major CVD,myocardial infarction,and stroke populations and their 95%confidence intervals were 7.9%(4.0%-11.6%)and 13.9%(6.5%-20.8%),6.6%(3.8%-9.4%),10.2%(3.9%-15.9%)and 5.9%(2.5%-9.2%),Grip strength values were the strongest associated with other anthropometric predictors in predicting CVD and mortality risk.Take allcause mortality as an example:every one standard deviation decrease in grip strength increased the risk of all-cause mortality by 27%(p<0.001),each one standard deviation increase in systolic blood pressure increased the risk of all-cause death by 18%(p<0.001),and every one standard deviation increase in ABSI Standard deviation 9%increased risk of all-cause mortality(p<0.001),BMI(p=0.379),waist circumference(p=0.666),waist-to-hip ratio(p=0.598)and BRI(p=0.887)associated with all-cause mortality was not statistically significant.Sensitivity analysis based on the E value shows that the relationship between grip strength and cardiovascular death,major cardiovascular events,myocardial infarction and stroke is stable,and is not affected by unknown and unmeasured confounding.The focus of predicting CVD and mortality risk with the addition of grip strength to the base model,in the female derivation population:the predictor C-index improved by 0.007(95%CI:0.006-0.008),and the IDI improved by 1.2%(95%CI:0.008-0.01 8)),the continuous NRI improved by 8,8%(95%CI:0.027-0.135),while the two-category NRI(0.018,95%CI:-0.008-0.032)and the three-category NRI(0.015,95%CI:-0.022-0.050),only the point estimate has increased;the clinical utility index shows that for every 1000 people predicted by the new model of handgrip strength,there will be 0.7 more net benefits,an increase of 0.28 EFLYs(95%CI:0.04-0.43);in men In the modeling population:the predictor C-index improved by 0.014(95%CI:0.013-0.015),the IDI improved by 4.9%(95%CI:0.0380.062),and the continuous NRI improved by 15.5%(95%CI:0.100-0.228)),the twocategory NRI increased by 4.4%(95%CI:0.002-0.077),and the three-category NRI increased by 8.2%(95%CI:0.027-0.122).4.1 more people had a net benefit,which could increase 21.3 EFLYs(95%CI:19.1-23.5).The above conclusions are basically consistent in the validation model.ConclusionThis study is the first to propose that adding grip strength to the base CVD and mortality risk prediction model can significantly improve the prediction performance of the original model,and has certain clinical utility.Grip strength measurement is simple,convenient,low-cost,highly practical,and highly reproducible in clinical practice.Therefore,it is recommended to include grip strength in routine physical examinations as an indicator for screening high-risk groups of cardiovascular disease.These findings may have important public health implications.