| Vitamin D is an essential vitamin for maintaining bone health in the human body.It promotes the reabsorption of calcium in the renal duct and regulates the absorption of calcium and phosphorus in the intestine.Recently,it is reported to be involved in various biological pathways including regulation in hormone secretion,neuromuscular function,inflammatory systems,and cell apoptosis.Despite its importance,vitamin D deficiency is common globally,and it may increase the risk of adverse health outcomes such as rickets,sarcopenia,cardiovascular disease,and metabolic syndrome.With the increasing trend of population aging in China,new challenges have raised regarding healthy aging.Health issues of the older population,especially the oldest-old people,deserve additional attention.Vitamin D levels are highly correlated with age,and older adults are more susceptible to vitamin D deficiency due to the decreased efficiency of vitamin D synthesis and metabolism in the body.Although recent epidemiological studies have suggested that vitamin D deficiency may lead to weakened physical function and even increased risk of mortality among older population,the conclusions from observational studies were inconsistent while the causal relationship remained unclear,leaving the problem waiting to be solved.Additionally,there is still a scarcity of nationwide statistical data and research on vitamin D levels among older adults in China.Evidence from large-scale cohort studies targeting the Chinese oldest-old is rather limited.It is imperative to conduct analyses based on Chinese older population to provide new evidence for research on vitamin D,which would help address the current knowledge gap and inform evidence-based recommendations for optimal vitamin D intake and management among Chinese older population.Mendelian randomization（MR）is a causal inference method which utilizes single nucleotide polymorphisms（SNPs）as instrumental variables for the exposure being studied.By leveraging the random allocation of alleles,MR effectively"randomizes"the grouping of study participants,allowing for causal inferences to be made between a given exposure and outcome while avoiding confounding factors.Using genetic instrumental variable as proxies for the exposure of interest,MR offers a powerful tool to infer causality in epidemiological studies,particularly when studying complex traits and diseases.The one-sample MR approach examines the relationships of genetic instrumental variables with exposures and outcomes within the same sample population.This minimizes the impact of population differences on causal inference results,making it more suitable for realizing the goal to analyze causal relationships among older populations.ObjectiveBased on a prospective cohort study of the older population in China,this research aimed to explore the key influencing factors of vitamin D deficiency and two health outcomes-all-cause mortality risk and incident risk of disability in activities of daily living（ADL）among oldest-old adults.Utilizing the one-sample MR method,the causal effect of vitamin D on the health status among the oldest old was determined,hoping to provide novel scientific evidence to promote healthy aging among Chinese older population.MethodsThe study was conducted using data from the Healthy Aging and Biomarkers Cohort Study（HABCS）between 2012 and 2018.Based on previous work and the principle of sample representativeness,the project ensured sufficient sample size of oldest-old participants through sampling design while considering the balanced proportion between different age groups and genders at the same time.The survey covered 18 regions in 14provinces around China,collecting information on participants’personal characteristics,lifestyle habits,physical function,cognition and other health status through face-to-face questionnaire interviews.Physical measurements were obtained through physical examinations,and biomarker detection and gene sequencing were conducted using collected biological samples.In this study,the serum 25（OH）D concentration at baseline is considered the exposure factor.Survival status and duration were recorded through follow-up surveys.The Katz scale from the questionnaire was used to assess the ability to perform six basic ADL among participants,and the ADL function was categorized as intact or disabled.In this study,a total of 2,300 oldest-old participants were included and the distribution of baseline characteristics was compared among the two groups stratified by vitamin D deficiency status.Logistic regression analysis was employed to identify the key influencing factors of vitamin D deficiency.Based on the follow-up survey regarding the two outcomes,the subjects were subsequently screened,and the Cox proportional hazards regression models were used to explore the influencing factors of all-cause mortality risk and incident risk of ADL disability among the oldest-old participants.To further investigate the causal relationships between vitamin D levels and the health outcomes,MR analyses were conducted where 1,621 participants were included after matching with the quality-controlled genetic sequencing database.Genetic variant association analyses were performed to identify SNPs associated with vitamin D levels as candidate genetic instrumental variables.The association was validated in different analysis populations,and genetic risk scores（GRS）associated with 25（OH）D levels were constructed.In exploring the associations between vitamin D and health outcomes,Cox proportional hazards models adjusted for multiple covariates were used.Restricted cubic spline models were employed to depict the nonlinear association curves.Mendelian randomization（MR）analysis was subsequently conducted to validate the causal effect of vitamin D.The two-stage least-squares regression（2SLS）method was used to calculate linear MR results,while fractional polynomial methods were applied to describe nonlinear causal relationships.Sensitivity analyses were performed using various two-sample MR models and the repeated analyses excluding participants with follow-up time less than one year.Stratified MR analyses were conducted to explore differences in causal relationships among subgroup population regarding various influencing factors.Cause-specific analyses distinguishing CVD mortality and non-CVD mortality,as well as subgroup analysis stratified by sex,were both conducted to further validate the robustness of the causal relationships between vitamin D and health outcomes.Based on the network MR method,the mediation effect analysis was carried out to explore the potential mediating role of ADL disability in the causal relationship between vitamin D and mortality risk.Results（1）Among 2,300 participants,the average age was 92.8 years and the median and quartiles of serum 25（OH）D level in population were 34.5（25.0,48.4）nmol/L.Most of them were female（64.6%）,Han nationality（91.9%）,lived in rural areas（80.4%）,received less than one year of education（78.1%）,received middle-level income（74.4%）,unmarried（unmarried,divorced or widowed,80.2%）and never smoker（81.61%）.Logistic model analyses showed that age,sex,residence,exercise,dietary diversity,BMI level and baseline health status（including ADL disability and cognitive impairment）were the key factors of vitamin D deficiency among the Chinese oldest old.（2）A total of 55 SNP loci closely related to serum 25（OH）D concentration were screened out from 1,621 participants who met the quality control requirements after matching the genetic database;In the subsequent MR analysis,several series of GRSs related to vitamin D levels were constructed,and it was verified that these GRSs were still highly correlated with 25（OH）D levels and met the requirements of genetic instrumental variable.（3）The relationship between vitamin D and mortality risk:2,300 subjects were followed up for 6,899 person-years while 1,235 people died,with a cumulative mortality rate of 53.7%and a death density of 179/1000 person-years.MR analyses included 1,621participants with 870 died（mortality rate 53.7%）.The baseline characteristics were similar between the two populations.The results of multivariate Cox proportional hazards model showed that after adjustment for other factors,age,ADL disability and cognitive impairment would increase all-cause mortality among the Chinese oldest old.Women,people with high vitamin D level and those with BMI at normal weight/overweight or obesity levels had lower mortality risk.In the fully adjusted model,in which covariables such as age,sex,nationality,place of residence,education level,marital status,income level,smoking,drinking,exercise,recruiting season,dietary diversity,BMI level,cognitive impairment and chronic disease status were adjusted,a reverse association between vitamin D level and all-cause mortality risk was observed（HR=0.84,95%CI:0.81～0.87,P<0.001）.Linear MR analysis further confirmed that there was a significant negative causal relationship.With the increase of 1 nmol/L of genetically predicted25（OH）D,the all-cause mortality risk decreased by 2.9%(HR=0.971,95%CI:0.963～0.980,P=4.2×10-11).Significant negative causal relationships between vitamin D and CVD mortality and non-CVD mortality were also confirmed.With the increase of 1nmol/L of genetically predicted 25（OH）D,CVD mortality decreased by 5.7%（HR=0.943,95%CI:0.925～0.960）while non-CVD mortality decreased by 3.8%（HR=0.962,95%CI:0.951～0.972）.Sensitivity analyses and gender stratification analyses proved the causal relationship to be robust.No nonlinear causal relationship was found between vitamin D and mortality risk(Pnonlinearity>0.05).Subgroup analysis showed that among women,people over 90 years old,married,overweight or obese,people with baseline ADL disability and cognitive impairment,increased vitamin D showed a more significant protective effect on all-cause mortality risk.（4）The relationship between vitamin D and the risk of incident ADL disability:A total of 1,427 participants aged 80 years and above were included in the association analysis.Over a follow-up period of 4,446 person-years,231 individuals developed ADL disability（16.2%）,with a higher proportion of women experiencing ADL disability compared to men.For the MR analysis,941 individuals were included,of whom 158developed ADL disability（16.8%）.The multivariable Cox regression analysis revealed that older age and impaired cognitive function were risk factors for ADL disability among oldest-old adults,while higher vitamin D levels were protective.In the fully adjusted model,compared to the severe vitamin D deficiency group,the hazard ratios（95%CI）for ADL disability were 0.66（0.47～0.94）for the vitamin D deficiency group,0.44（0.28～0.69）for the insufficiency group,and 0.33（0.13～0.85）for the sufficiency group.Linear MR analysis demonstrated a significant inverse causal association between genetically predicted 25（OH）D levels and the risk of ADL disability.Specifically,for every 1-unit increase in genetically predicted 25（OH）D,the risk of ADL disability decreased by 5.4%（HR=0.946,95%CI:0.921～0.975）.The two-sample MR model results were consistent,indicating robustness of the causal association.No nonlinear causal association was found between vitamin D and the risk of ADL disability(Pnonlinearity>0.05).Stratified analysis revealed that the protective effect of vitamin D on the risk of ADL disability was more pronounced among women,those aged 80～89 years,those who were not married,and those who were overweight or obese.No causal association was found among men.（5）Mediation effect analysis of ADL disability between vitamin D and mortality risk:traditional mediation analysis models reveal that compared to individuals with vitamin D deficiency,those without deficiency have an overall reduced risk of mortality by 42%（HR=0.58,95%CI:0.49～0.69）,among which the indirect effect mediated by ADL disability accounted for a reduction in mortality risk by approximately 10%（HR=0.90,95%CI:0.86～0.94）.The estimated percentage of the mediating effect of ADL disability on the association between vitamin D status and mortality risk was 19.9%（95%CI:9.2%～41.1%）.Additionally,based on network MR,causal mediation analysis demonstrated a significant causal mediating effect of ADL disability in the causal association between vitamin D and mortality risk,and the proportion of the mediating effect was 14.8%（95%CI:3.6%～35.0%）.Conclusion（1）Vitamin D deficiency is rather common among the oldest-old population in China,with key influencing factors including age,gender,residence,exercise habits,dietary diversity,and health status such as daily activity ability and cognitive function.This underscores the need for heightened awareness and promotion of vitamin D supplementation among this population.（2）The causal protective effect of vitamin D was evident on all-cause mortality risk as well as risk of disability in activities of daily living among the oldest-old in China.The effect was more pronounced among women and those with poorer health status.Boosting vitamin D levels may not only improve survival rates but also maintain the quality of life among the oldest-old people,promoting healthy aging in China.（3）A significant causal mediating effect of ADL disability was found in the relationship between vitamin D levels and mortality risk.This suggests that vitamin D deficiency may increase the risk of ADL disability,subsequently elevating the mortality risk among the oldest-old.Therefore,it is crucial to monitor vitamin D levels among this population,particularly among those with lower vitamin D levels and existing ADL disabilities,to closely monitor changes in their health status. |
PDF Full Text Request