Constructing and searching adjustment sets based on a causal graph model / 四川精神卫生

The purpose of this paper was to introduce the basic knowledge of the causal graph model， the contents of the CAUSALGRAPH procedure and the method of constructing and searching adjustment sets based on the CAUSALGRAPH procedure in SAS/STAT. The causal graph model was the product of the combination of graph theory and probability theory. It could find all possible adjustment sets including the minimum adjustment set based on the action relationship between the variables set by the user. The contents of the CAUSALGRAPH procedure mainly included three identification criteria， two operating modes and one verification checking method. This paper analyzed the causal effect of two instances based on the CAUSALGRAPH procedure in SAS， and explained the output results.

Role of covariates in the analysis of causal mediation effects / 四川精神卫生

The purpose of this paper was to introduce the theoretical basis of the causal mediation effect analysis and the specific method to realize an example by the causal mediation effect analysis with SAS. The theoretical basis of the causal mediation effect analysis included the following two aspects， the basic concept and defining the counterfactual framework of the causal mediation effect. The example was about whether the encouraging environment provided by parents would affect the cognitive development of children. The traditional multiple linear regression analysis， the causal mediation effect analysis without considering covariates and with considering covariates were used， respectively. By comparing the results obtained by the three analysis methods， the following conclusions were drawn： ① when there were the mediation variables in the data， it was not suitable to use traditional multiple linear regression analysis to replace the causal mediation effect analysis； ② when there were covariates in the data， it was not suitable to conduct causal mediation analysis under the condition of ignoring covariates.

Reasonably conduct the multiple linear regression analysis combined with the propensity score analysis / 四川精神卫生

The purpose of this paper was to introduce how to combine the propensity score analysis to reasonably carry out multiple linear regression analysis. Firstly， it introduced 3 basic concepts related to the propensity score analysis. Secondly， it presented the core contents of the propensity score analysis， that was， three matching methods. Thirdly， through an epidemiological survey example， it gave the whole process of how to use SAS software for the analysis. The contents were as follows： ① for the original data set， test whether the difference of covariates between the treatment group and the control group was statistically significant； ② directly implement the multiple linear regression analysis for the original data set； ③ the propensity score analysis was used to generate the matched data set； ④ for the matched data set， test whether the difference of covariates between the treatment group and the control group was statistically significant； ⑤ a reasonable multiple linear regression analysis was used for the matched data set.

Causal analysis of ordinal treatments and binary outcomes under truncation by death.

It is common that in multi-arm randomized trials, the outcome of interest is "truncated by death," meaning that it is only observed or well-defined conditioning on an intermediate outcome. In this case, in addition to pairwise contrasts, the joint inference for all treatment arms is also of interest. Under a monotonicity assumption we present methods for both pairwise and joint causal analyses of ordinal treatments and binary outcomes in presence of truncation by death. We illustrate via examples the appropriateness of our assumptions in different scientific contexts.

Extending the Bayesian Adjustment for Confounding algorithm to binary treatment covariates to estimate the effect of smoking on carotid intima-media thickness: the Multi-Ethnic Study of Atherosclerosis.

We illustrate the application of the Bayesian Adjustment for Confounding (BAC) algorithm when the treatment covariate is binary. Using data from the Multi-Ethnic Study of Atherosclerosis, we estimate the effect of ever smoking on common carotid artery intimal medial thickness among adult Caucasian participants (n=1378). Our novel implementation of the BAC algorithm is performed first from an outcome model perspective and second from a treatment model perspective with both inverse probability weighting and doubly-robust estimation techniques. The BAC results are compared with the results obtained using standard model averaging and full model strategies, giving a range of adjusted estimates between 45.50 and 65.30 µm for increased common carotid artery intimal medial thickness among ever smokers. For both perspectives, we observe that BAC offers similar performance to using the fully specified outcome and/or treatment model (the full outcome model ever smoking effect is 48.61 µm; 95% CI: (0.62, 96.60)). We then redo the analyses for the African American, Hispanic, and Chinese adult participants to study the robustness of these findings with reduced sample size. For the Chinese subcohort, which corresponds to the smallest sample size (n=436), we find that, from a treatment model perspective, BAC reduces the variability of the estimates in comparison with using a full model approach. This suggests that the use of BAC in conjunction with inverse probability weighting and doubly-robust estimation can be advantageous when applied to relatively small sample sizes. This conjecture is subsequently verified on the basis of three simulated experiments.