| BackgroundLow level of fruit consumption is one of the major risk factors for cardiovascular disease(CVD),accounting for an estimated 104 million disability-adjusted life-years worldwide in 2010.In the research of western countries,a higher level of fruit consumption has been showed to associate with a lower risk of cardiovascular disease,but little is known about such associations in China,where the eating habits,lifestyle and disease pattern are quite different from western countries.Most studies related to the association of fruit and disease quoted foreign research results.Because of the diet habit,fruit consumption in China is low,and unlike western countries where fruit and vegetable are usually mixed to "salad",fruit in China is usually eaten individually before or after meals.This is an advantage condition to explore the independent relationship of the fruit consumption and cardiovascular disease.In the disease model,the incidence of stroke subtypes in China is different from western countries,and the proportion of hemorrhagic stroke is much higher than western countries.The etiology and pathogenesis are different in ischemic stroke and hemorrhagic stroke,and the influencing factors are also quite different,so it is necessary to discuss the influencing factors of two subtypes of stroke separately.In previous studies,the association of fruit intake with the risk of stroke has been demonstrated,but the association of fruit consumption with two stroke subtypes is not yet clear.In addition,although previous studies have showed that fruit consumption can reduce the CVD risk,the mechanism by which it works is not fully understood.Some studies have pointed out that to eat more fruit can lower the risk of obesity,hypertension and type 2 diabetes,which are important risk factors for CVD.Therefore,these disease factors or disease-related indicators such as waist circumference,BMI,blood pressure,and blood glucose may have some interaction or mediation effect in the relationship between fruit consumption and CVD.However,the current research is still rare related reports.Cardiovascular diseases are affected by socio-economic factors,life behavior factors,genetic factors and other factors together.However,most of the current study,only to study the influence of a factor or a number of factors on cardiovascular disease.While the relative importance of various factors on cardiovascular disease is rarely reported.To deal with the above-mentioned issues,prospective cohort study was designed to assess the assotion of fruit consumption and CVD risk using baseline information of 35,509 subjects and the incidence and death events followed for 8 years.ObjectivesFirstly,to evaluate the association of fruit consumption and risk of CHD incidence,stroke incidence and CVD death.Secondly,to further evaluate the association of fruit consumption and risk of stroke subtypes incidence.Thirdly,to explore the interaction of fruit consumption and waist circumference and BMI on CVD.To explore the mediation of blood pressure and blood glucose in the assotion of fruit consumption and CVD.Fourthly,to explore the relative importance of individual factors,socioeconomic factors,life behavior factors,physiological and biochemical indexes,fruitconsumption on CVD.MethodsFirstly,35509 research objects aged among 35?79 years old were selected from Qingdao project site of CKB(China Kadoorie Biobank).The information of demographic information(age,sex,education,income,occupation),lifestyle(smoking,drinking,dietary consumption,physical activity),history of main chronic disease(hypertension,coronary heart disease,stroke,diabetes),physical indicators(height,weight,waist circumference,blood pressure and random blood glucose)were obtined through the baseline survey.The characteristics of baseline population and fruit consumption were described.Secondly,the incidence and mortality of cardiovascular diseases of the cohort were obtained through the National Disease Surveillance Points system(DSPs)of the Chinese Center for Disease Control and Prevention(CDC),combining population information from the public security system,medical insurance system and community orientation follow-up.Cox proportional hazards regression model was used to assess the association of fruit consumption and risk of CHD,stroke,stroke subtypes,and CVD death.Thirdly,multiplicative model interaction based on Cox regression was conducted to investigate the interaction of fruit consumption and waist circumference and BMI on CVD.Logistic regression using KHB methodwas conducted toinvestigate the mediation effect of blood pressure and blood glucose on the association of fruit consumption and CVD.Fourthly,five scores(individual score,socioeconomic score,life behavior score,physiological and biochemical score and fruit consumption score)were predicted by five different Logisticmodels.Dominance analysis was conducted to evaluate the relative importance of the risk factors on CVD.Results1.Baseline characteristics(1)Basic situation:35509 people were investgated in the baseline survey.Ratio of male to female was 1:1.27,and the average age was(50.3 ± 10.2)years;The education level of the respondents was mainly junior high school(41.2%)and the occupation was mainly workers(40.7%)and the income was mainly 20000?34999 per year(43.3%);26.3%of the respondents were current smokers and 22.0%were regular drinkers;54.2%of the respondents consumed fruit every day,2.6%almost did not eat fruit;13.6%of the respondents had been diagnosed with hypertension,5.2%had been diagnosed with coronary heart disease,0.7%had been diagnosed with stroke,5.4%had been diagnosed with diabetes;5.5%of the respondents had a family history of coronary heart disease and 16.2%had a family history of stroke.(2)Fruit consumption of the respondents:The five frequency categories of never,monthly,1 to 3 days per week,4 to 6 days per week,daily accounted for 2.6%,10.4%,24.0%,8.8%and 54.2%,respectively.Compared with those who never consumed fruit,respondents with higher frequency of fruit consumption were younger,more women,with higher education level and higher income(P<0.05).The frequency of fruit consumption varied with different occupation,the difference was statistically significant(?2=369.72,P<0.001).Among the current smokers and regular drinkers,the proportion of the "daily" frequency group was much lower than that among non-smokers and non-drinkers(smoker vs.non-smoker:37.1%vs.61.5%;drinkers vs.non-drinkers:36.0%vs.59.5%,P<0.05,respectively).2.Fruit consumption and CVD risk(1)Basic situation of the corhort:As of June 30,2014,128 people of the cohort were lost to follow-up with a loss rate of 0.36%,and 1119 people died.The average follow-up time was 8.1 years and a total of 260814.0 person-years of follow-up were observed.(2)Fruit consumption and risk of CHD incidence:As of June 30,2014,283 new cases of CHD were observed.The incidence of CHD was 108.51/100000 person-years.Univariate Cox regression analysis showed that female,higher education and income,more physical activity,more fruit and fish consumption were protective factors for CHD,while older people,smoking,drinking,higher level of waist circumference,blood pressure and blood glucose were risk factors for CHD.Multivariate Cox regression analysis showed that after adjusting age,gender,education level,income,smoking,drinking,physical activity,family history of CHD,consumption of vegetables,meat,fish and dairy products,survey season,increase the frequency of fruit consumption can reduce the risk of CHD incidence.Compared to the respondents who never consumed fruit,respondents who consumed fruit more than 4 days per week had a 46%lower risk of CHD incidence[HR(95%CI):0.54(0.44?0.66),P<0.05],and the risk reduced by 50%and 29%in male and female[HR(95%CI):male 0.50(0.39?0.65),female 0.71(0.52?0.97),P<0.05],respectively.Further adjustment for WC,BMI,SBP and random blood glucose did not change the association.(3)Fruit consumption and risk of stroke incidence:As of June 30,2014,a total of 1011 new cases of stroke were observed,with a stroke incidence of 387.63/100000 person-years.Univariate Cox regression analysis showed that higher income,more physical activity,more fruit consumption were protective factors for stroke,while older people was risk factor for stroke.Multivariate Cox regression analysis showed that fruit consumption had a protective effect on stroke incidence.Compared to the respondents who never consumed fruit,respondents who consumed fruit more than 4 days per week had a 44%lower risk of stroke incidence[HR(95%CI):0.56(0.50?0.62),P<0.05],and the risk reduced by 46%and 42%in male and female[HR(95%CI):male 0.54(0.46?0.64),female 0.58(0.52?0.69),P<0.05],respectively.Further adjustment for WC,BMI,SBP and random blood glucose did not change the association.(4)Fruit consumption and risk of stroke subtypes incidence:Among the 1011 new cases of stroke,873 new cases of ischemic stroke were found with an incidence of 334.72/100000 person-years.138 cases of hemorrhagic stroke were found with an incidence of 52.91/100000 person-years.Univariate Cox regression analysis showed that higher education and income,more physical activity,more fruit and fish consumption were protective factors for ischemic stroke,while older people,more preserved vegetables consumption,higher level of waist circumference,BMI,blood pressure and blood glucose were risk factors for ischemic stroke.Female,higher education and income,more physical activity,more fruit,meat and fish consumption were protective factors for hemorrhagic stroke,while older people,drinking,higher level of waist circumference,blood pressure and blood glucose were risk factors for hemorrhagic stroke.Multivariate Cox regression analysis showed that increasing the frequency of fruit consumption can also reduce the risk of ischemic stroke and hemorrhagic stroke.Compared to the respondents who never consumed fruit,respondents who consumed fruit more than 4 days per week had a 46%lower risk of ischemic stroke incidence[HR(95%CI):0.54(0.48?0.61)P<0.05],and a 41%lower risk of hemorrhagic stroke incidence[HR(95%CI):0.59(0.29?0.84),P<0.05].Further adjustment for WC,BMI,SBP and random blood glucose did not change the association.(5)Fruit consumption and risk of CVD mortality:As of June 30,2014,359 cases of cardiovascular death were observed,with a mortality rate of 137.65/100000 person-years.Univariate Cox regression analysis showed that female,higher education and income,more physical activity,more fruit,meat and fish consumption were protective factors for CVD mortality,while older people,smoking,higher level of waist circumference,blood pressure and blood glucose were risk factors for CVD mortality.Multivariate Cox regression analysis showed that fruit consumption had obvious protective effect on CVD mortality.Compared to the respondents who never consumed fruit,respondents who consumed fruit more than 4 days per week had a 57%lower risk of CVD mortality[HR(95%CI):0.43(0.35?0.53),P<0.05],and the risk reduced by 64%and 34%in male and female[HR(95%CI):male 0.36(0.27?0.48),female 0.66(0.44?0.98),P<0.05],respectively.Further adjustment for WC,BMI,SBP and random blood glucose did not change the association.3.Interaction analysis and mediation analysis(1)Interaction analysis:Fruit consumption and waist circumference had interaction effect by multiplication model on CHD incidence and CVD mortality.After adjusting the confounding factors,the effect values of the multiplicative interaction items were 0.71(0.55?0.91),0.65(0.49?0.85),respectively.Fruit consumption and BMI also had interaction effect by multiplication model on CHD incidence and CVD mortality.After adjusting the confounding factors,the effect values of the multiplicative interaction items were 0.84(0.70?1.00),0.82(0.67?0.99),respectively.(2)Mediation analysis:Blood pressure played a part of the mediation effect in the association of fruit consumption and CHD incidence,stroke incidence and CVD mortality.After adjusting the confounding factors,the proportion of mediating effect to total effect was 4.24%,13.29%and 9.26%,respectively.Blood glucose also played a part of the mediation effect in the association of fruit consumption and CHD incidence,stroke incidence and CVD mortality.After adjusting the confounding factors,the proportion of mediating effect to total effect was 9.96%,9.88%and 11.26%,respectively.4.Dominance analysis on influencing factors of CVD(1)Dominance analysis on influencing factors of CHD:The most important predictor for CHD was individual factors(47.2%),following by life behavior factors(17.0%),physiological and biochemical indexes(15.1%),socioeconomic factors(12.9%)and fruit consumption(7.8%).Fruit consumption can explain 7.8%of all the variation on CHD.(2)Dominance analysis on influencing factors of stroke:The most important predictor for stroke was individual factors(39.5%),following by socioeconomic factors(22.5%),physiological and biochemical indexes(20.1%),life behavior factors(14.9%)and fruit consumption(3.0%).Fruit consumption can explain 3.0%of all the variation on stroke.(3)Dominance analysis on influencing factors of CVD mortality:The most important predictor for CVD mortality was individual factors(46.2%),following by socioeconomic factors(17.9%),physiological and biochemical indexes(17.5%),lifebehavior factors(12.3%)and fruit consumption(6.2%).Fruit consumption can explain 6.2%of all the variation on CVD mortality.Conclusions1.Fruit consumption could reduce the risk of CVD.Increasing fruit consumption to above 4 times per week could reduce the risk of CHD incidence,stroke incidence and CVD mortality by 46%,44%,and 57%,respectively.2.Fruit consumption could reduce the risk of stroke subtypes.Increasing fruit consumption to above 4 times per week could reduce the risk of ischemic stroke and hemorrhagic stroke by 46%and 41%,respectively.3.Fruit consumption and waist circumference,BMI had interaction effects by multiplication model on CHD incidence and CVD mortality.Blood pressure and blood glucose played a part of the mediation effects in the association of fruit consumption and CHD incidence,stroke incidence and CVD mortality.4.Among the factors included in this study,the individual factor(age,sex,family history of cardiovascular disease)was the primary factor on CVD.The proportion of fruit consumption on CHD,stroke and CVD death were 7.8%,3.0%and 6.2%respectively.Innovation and Significance1.We explored the association of fruit consumption and stroke subtypes separately,to obtain the evidence to prove the protective effect of fruit consumption on ischemic stroke and hemorrhagic stroke.2.We use dominance analysis to evaluate the relative importance of the different factors on CVD.Few studies investigated the relative importance of of the different factors on CVD,which can provid important guiding significance for the strategy formulation of public health.3.We explored the interaction and mediation effects of CVD factors.Most previous studies only concerned with the effect of fruit consumotion on CVD,in which few analysis on interaction between fruit consumption and other factors were conducted. |
PDF Full Text Request