ABSTRACT
Objective@#To evaluate the bidirectional association between periodontitis and Sjögren's syndrome using the Mendelian randomization (MR) method.@*Methods@#Genome-wide association study (GWAS) data of periodontitis (N = 45 563) and Sjögren's syndrome (N = 214 435) were selected to meet the requirements of the same ethnicity and different regions. Inverse variance-weighted (IVW), MR-Egger, and weighted median (WM) tests were used to evaluate the causal effect. Cochran's Q statistics, MR-Egger intercept, MR-PRESSO and leave-one-out analysis were used as sensitivity analyses to assess the stability and reliability of the results.@*Results@#After screening, the GWAS data of Sjögren's syndrome were based on the Finnish region, and the periodontitis GWAS data were based on the UK region, both of which originated from European ancestry. Using IVW (OR = 1.017, 95% CI = 0.956-1.082), MR-Egger (OR = 0.985, 95% CI= 0.956-1.082), and WM (OR =1.021, 95% CI = 0.948-1.099), no causal effect of Sjögren's syndrome on periodontitis was found using any of the three methods. Conversely, no causal effect of periodontitis on Sjögren's syndrome was found (IVW, OR = 1.024, 95% CI = 0.852-1.230; MR-Egger, OR = 0.978, 95% CI = 0.789-1.212; WM, OR = 1.024, 95% CI = 0.846-1.260). The sensitivity analyses indicated that the results were stable and reliable. Cochran's Q test and MR-PRESSO revealed that there was no significant heterogeneity among the instrumental variables, which included single nucleotide polymorphisms (SNPs). The intercept of MR-Egger regression indicated no pleiotropy in the included SNPs. No individual SNP was found that significantly affected the results using the leave-one-out method.@*Conclusion@#This study does not support a bidirectional causal effect between periodontitis and Sjögren's syndrome.
ABSTRACT
Observational study of therapy efficacy comparison has been widely conducted to provide the additional efficacy evidence to support randomized control study. Statistical adjustment for unmeasured confounders is a major challenge in observational study of therapy efficacy comparison. This paper summarizes and evaluates the relative statistical methods. Currently, the most commonly used methods include instrumental variable, difference-in-differences (DiD) model and prior event rate ratio (PERR) adjustment. The instrumental variable method skill fully escapes unmeasured confounders through model structure, but it is not easy to obtain satisfied instrumental variables. Both PERR and DiD require the data prior to exposure which are not always collected in observational studies. Unmeasured confounders could result in new requirements and pose new challenges for statistical methods, which needs further study and improvement.
ABSTRACT
Confounders are difficult to avoid in studies on observational comparative effectiveness. It is often unclear whether the confounders have been completely eliminated after controlling the measured or unmeasured potential confounding effects or if sensitivity analysis is needed when using the specific statistical methods, under given circumstances. This manuscript summarizes and evaluates the confounding sensitivity analysis methods. Based on different studies, sensitivity analyses need to use different approaches. The traditional sensitivity analysis can be applied for the measured confounders. Currently, the relatively systematic sensitivity analyses for unmeasured confounders would include confounding function, bounding factor and propensity score calibration. Additionally, more investigations are associated with Monte Carlo and Bayesian sensitivity analysis. Reliability of the research conclusion thus may largely be improved when the sensitivity analysis results are consistent with the main analysis.
ABSTRACT
Observational comparative effectiveness studies have been widely conducted to provide evidence on additional effectiveness and to support randomized controlled findings in research. Although this type of study becomes more important over time, challenges related to the common biases which stemmed from confounders, are difficult to control. This manuscript summarizes some statistical methods used on adjusting measured confounders that often noticed in research, regarding the observational comparative effectiveness. Useful traditional methods would include stratified analysis, paired analysis, covariate model and multivariable model, etc.. Unconventional adjustment approaches such as propensity score and disease risk score methods may also be used in studies, for matching, stratification and adjustment. A good study design should be able to control confounders. The limitations of all the post hoc statistical adjustment methods should also be fully understood before being appropriately applied in practical events.
ABSTRACT
Disease risk score (DRS) can be used to adjust the confounding effects on data with high dimensions and can reduce related bias through balancing the risk or probability,regarding the development of some specific diseases,between the two compared groups.The DRS approach thus can be applied to studies of pharmacoepidemiology when administrative medical database is used for data analysis.Although DRS functions are similarly to the propensity scores (PS) under many situations,even with some advantages over PS or conventional analytical methods in some special exposure settings,the usage of DRS is far limited than the PS method.Considering the important application value of DRS in pharmacoepidemiologic studies,we are introducing the theory,model,estimation and application of DRS,to present reference for the development of DRS method in the pharmacoepidemiologic studies.
ABSTRACT
Disease risk score (DRS) can be used to adjust the confounding effects on data with high dimensions and can reduce related bias through balancing the risk or probability,regarding the development of some specific diseases,between the two compared groups.The DRS approach thus can be applied to studies of pharmacoepidemiology when administrative medical database is used for data analysis.Although DRS functions are similarly to the propensity scores (PS) under many situations,even with some advantages over PS or conventional analytical methods in some special exposure settings,the usage of DRS is far limited than the PS method.Considering the important application value of DRS in pharmacoepidemiologic studies,we are introducing the theory,model,estimation and application of DRS,to present reference for the development of DRS method in the pharmacoepidemiologic studies.
ABSTRACT
Um dos principais objetivos dos estudos epidemiológicos é a estimação de efeitos causais. E a inferência causal deve ser discutida tanto por estudos experimentais quanto por estudos observacionais. Uma importante limitação na interpretação causal de estudos observacionais é a possível presença de confundidores não observados (hidden bias). Uma estratégia para avaliar o possível efeito de um confundidor não observado é através de uma análise de sensibilidade, que determina quão forte deveriam ser os efeitos de um confundidor não observado de modo a explicar uma aparente associação. A proposta deste artigo é rever e integrar dois métodos independentes de análise de sensibilidade: um devido a Greenland, sob uma perspectiva epidemiológica; e outro devido a Rosenbaum, sob uma perspectiva estatística. A combinação de questões estatísticas e epidemiológicas torna a análise de sensibilidade mais completa e direta, e deve estimular sua necessária difusão e mais aplicações. Como os estudos observacionais são mais sujeitos a vieses e confundimento, uma análise de sensibilidade considerando aspectos estatísticos e epidemiológicos fortalece o processo de inferência causal.
Subject(s)
Epidemiology , Observer VariationABSTRACT
The allele frequencies of markers as well as linkage disequilibrium (LD) can be changed in cases due to the LD between markers and the disease allele, exhibiting spurious associations of markers. To identify the true association, classical statistical tests for dealing with confounders have been applied to draw a conclusion as to whether the association of variants comes from LD with the known disease allele. However, a more direct test considering LD using estimated haplotype frequencies may be more efficient. The null hypothesis is that the different allele frequencies of a variant between cases and controls come solely from the increased disease allele frequency and the LD relationship with the disease allele. The haplotype frequencies of controls are estimated using the expectation maximization (EM) algorithm from the genotype data. The estimated frequencies are applied to calculate the expected haplotype frequencies in cases corresponding to the increase or decrease of the causative or protective alleles. The suggested method was applied to previously published data, and several APOE variants showed association with Alzheimer's disease independent from the APOE epsilon4 variant, rs429358, regardless of LD showing significant simulated p-values. The test results support the possibility that there may be more than one common disease variant in a locus.