Association Between Dietary Pattern And The Risk Of Fracture In Chinese Adults

Objective:To evaluate the relationship between dietary inflammation index(DII)and Oriental Healthy Diet Score(OHDS)and the risk of fracture in Chinese groups and provide theoretical support and guidance for reducing the risk of fracture.Methods:1.Taking adults>18 years old in the 1997-2015 data of the China Health and Nutrition Survey(CHNS)database as the study subjects.Fracture was the outcome variable and participants with fracture at baseline,lack of energy and nutrient intake information or unbelievable,pregnancy,lactation,disability,myocardial infarction,stroke or any type of tumor history were excluded.Finally,11999 participants were included.3-day,24-hour diet review method was used to obtain the dietary intake and calculated DII and OHDS.2.The Cox proportional hazard model was used to analyze the correlation between DII and OHDS dietary pattern scores and fracture risk.Adjustment for covariates,Model 1:age,sex,residence,education level,household income and marital status;Model 2:Model 1+smoking status,drinking status,physical activity level,BMI,mid-arm muscle circumference and waist-hip ratio;Model 3:Model 2+hypertension,diabetes,osteoporosis risk level(OSTA index level)and calcium intake.3.Based on the Cox proportional hazards model analysis,subgroup analyses were performed according to age,BMI,smoking status,drinking status and abdominal obesity.Interaction tests were performed.Participants with hypertension or diabetes were excluded for sensitivity analysis.Restricted cubic spline(RCS)was used to fit the trend of DII and fracture risk.DII component analysis was performed to correlate DII scores of carbohydrates,total fats,proteins,vitamin A,thiamine,riboflavin,niacin and folic acid with fracture risk alone.Results:1.A total of 11999 participants were included,including 5519 men and6480 women,with an average age of 44.0±14.6 years.At 18 years of follow-up(median follow-up=9.0 years),a total of 902 participants developed fractures,including 463 men and 439 women.The median DII score for the total population was 0.64(-1.74,1.46),0.75(-1.68,1.50)for men and 0.53(-1.79,1.42)for women.The median OHDS score for the total population was6.0(4.0,7.0),6.0(4.0,7.0)for men and 5.0(4.0,7.0)for women.2.Cox proportional risk model analysis showed that DII was positively associated with fracture risk in women but not in men.Among women,after adjusting for covariates,take the lowest DII score group Q1 as a reference,there was no difference in fracture risk in Q2 and Q3,but increased risk in Q4and Q5 with a 1.53-fold and 1.63-fold increase.Fracture risk ratio of Q2,Q3,Q4 and Q5 in women.HR(95%CI)were 1.16(0.73-1.84),1.26(0.84-1.90),1.53(1.04-2.27)and 1.63(1.11-2.41)(P for trend=0.004).RCS analysis showed a significant correlation between fracture risk and DII score in women(P-overall association=0.017).HR increased significantly with DII>0.53.3.Subgroup analysis showed that DII in the subgroup of women aged<50 years,non-smoking,non-drinking and non-abdominal obesity was positively associated with fracture risk.After adjusting for the covariates,take the lowest DII score group Q1 as a reference,the HR(95%CI)was 2.18(1.18-4.02)(P for trend=0.003),1.65(1.11-2.46)(P for trend=0.004),1.53(1.02-2.29)(P for trend=0.015)and 1.94(1.12-3.34)(P for trend=0.031).RCS analysis showed that fracture risk was significantly correlated with DII score in these four subgroups(P-overall association<0.05).HR increased with the increase of DII.However,no association was found in age≥50,smokers,drinkers,abdominal obesity,BMI<24 kg/m~2 or BMI≥24 kg/m~2.4.Sensitivity analysis showed that DII was positively correlated with fracture risk in women with excluded hypertension,excluded diabetes,excluded hypertension and diabetes,excluded hypertension or diabetes.After adjusting for the covariates,take the lowest DII score group Q1 as a reference,the HR(95%CI)was 1.92(1.26-2.94)(P for trend<0.001),1.64(1.11-2.43)(P for trend=0.003),1.98(1.28-3.04)(P for trend<0.001)and 1.61(1.09-2.37)(P for trend=0.004).5.DII component analysis showed that DII scores of thiamine,riboflavin,niacin and folic acid were positively correlated with fracture risk in women.After adjusting for the covariates,take the lowest DII score group Q1 as a reference,the HR(95%CI)was 1.49(1.02-2.17)(P for trend=0.003),1.49(1.01-2.21)(P for trend=0.013),1.65(1.12-2.42)(P for trend<0.001)and1.37(0.95-1.99)(P for trend=0.005).6.Cox proportional risk model analysis showed that OHDS were not correlated with fracture risk.After adjusting the covariates,take the lowest DII score group Q1 as a reference,the HR(95%CI)of Q2,Q3,Q4 and Q5 of the total population were 1.02(0.85-1.21),1.17(0.97-1.41),1.08(0.86-1.35)and0.95(0.73-1.24)(P for trend=0.381);In men,the HR(95%CI)of Q2,Q3,Q4and Q5 were 1.03(0.80-1.34),1.36(1.05-1.77),1.21(0.88-1.67)and 1.23(0.88-1.73)(P for trend=0.731);In women,the HR(95%CI)of Q2,Q3,Q4and Q5 were 1.02(0.79-1.30),1.00(0.76-1.31),0.97(0.70-1.35)and 0.67(0.43-1.06)(P for trend=0.104).Conclusions:1.DII was positively correlated with fracture risk in women.The higher the dietary inflammatory potential,the higher the fracture risk in women.2.DII was positively associated with fracture risk in the subgroup of women aged<50years,non-smoking,non-drinking or non-abdominal obesity.3.DII of thiamine,riboflavin,niacin and folic acid was positively associated with fracture risk in women.4.OHDS is not associated with fracture risk.