Due to the complexity of assembly processes in complex products, even slight deviations can result in quality problems. Quality problems often stem from variations in quality characteristics that, when propagated and superimposed through the assembly flow, exceed acceptable thresholds. Modelling these causal effects at the quality characteristic level and establishing a causal network is an effective strategy for quality control. This paper proposes a data- and knowledge-driven approach for causal modelling and quality control in complex product assembly. First, the propagation paths of quality characteristic variations are extracted as prior knowledge and incorporated into a reinforcement learning algorithm to guide causal graph construction and improve the accuracy of causal modelling. Second, a quality control framework integrating root cause analysis and prediction is developed based on the established causal network. A Bayesian method is applied to provide probabilistic guidance for root cause analysis, while the causal network is used to identify and eliminate characteristics unrelated to the target characteristic, thereby enhancing the accuracy of quality prediction. Finally, the proposed method is validated using an aircraft assembly case study. Experimental results demonstrate its feasibility and effectiveness in enhancing quality control in complex product assembly.

: Background Existing evidence suggested a relevance between insulin resistance (IR) and stroke, but further confirmation is needed. Research priorities encompass large sample size, stratified analysis, advanced IR index, stroke subtype characterization, and analysis of potential mechanisms. Methods We applied multivariable logistic regression using data from National Health and Nutrition Examination Survey 1999-2018 to estimate the correlation between two IR indexes: metabolic score for insulin resistance (METS-IR) and homeostasis model assessment of insulin resistance (HOMA-IR), and stroke. We also performed a two-sample Mendelian randomization (MR) study to detect the causal relationship between IR phenotype, METS-IR, and HOMA-IR as exposures, and stroke as well as its ischemic subtypes as outcomes. Results A total of 15,016 participants representing 147,325,838 individuals after weighting were enrolled. Both METS-IR and HOMA-IR were not significantly correlated with stroke after strict adjustment, but METS-IR was strongly related to stroke in individuals aged 20-40 in the stratification analysis. The MR analysis showed robust causal associations between IR phenotype and any stroke (AS) as well as ischemic stroke (IS). Besides AS and IS, METS-IR also had a causal effect on large artery stroke and small vessel stroke. Conclusion IR was associated with an increased risk of stroke in young adults.

Caffeine, predominantly consumed through coffee and tea, has received growing attention for its potential cardioprotective actions. Evidence suggests a non-linear association between habitual intake and cardiovascular risk, with moderate consumption offering the most favorable profile. Given that caffeine is often ingested within complex beverage matrices, distinguishing the effects of pure caffeine from those of coffee components and preparation methods is essential for accurate interpretation. This study synthesizes findings from epidemiological research, Mendelian randomization analyses, mechanistic experiments, and clinical trials to evaluate caffeine's cardiovascular impact. We assess key outcomes such as coronary artery disease, heart failure, stroke, hypertension, and arrhythmias. Mechanistic pathways are explored, including adenosine receptor antagonism, modulation of autonomic tone, influences on endothelial function and arterial stiffness, anti-inflammatory and antioxidant effects, and indirect metabolic actions on glucose and lipid regulation. Additional analyses examine how genetics, comorbid conditions, and concomitant medications may modify individual responses. Integrated evidence from observational studies demonstrates a non-linear, J-shaped association between coffee and caffeine intake and cardiovascular outcomes, with moderate consumption associated with the lowest observed risk; however, Mendelian randomization analyses generally do not support a clear causal protective effect of caffeine. Pure caffeine and coffee-derived effects diverge in several respects: unfiltered coffee may elevate lipid levels due to diterpenes, whereas filtered coffee generally has neutral lipid effects. Habitual use is shown to attenuate the acute pressor response through tolerance development. Variability in cardiovascular responses is influenced by genetic polymorphisms, baseline blood pressure phenotype, concurrent illnesses, and interacting medications. Overall, moderate caffeine intake generally appears safe for cardiovascular health. The cardioprotective links seen in epidemiological studies may be affected by coffee components, how it is prepared, and residual behavioral confounding, while solid evidence for a direct protective effect of caffeine is still limited. Remaining uncertainties highlight the need for future trials that isolate caffeine's dose-response from coffee matrices, incorporate genotype-based stratification, evaluate baseline hemodynamic phenotypes, and use standardized clinical endpoints. Such work is essential to clarify causal pathways and optimize the clinical relevance of caffeine-related recommendations.

There is a lack of research addressing associations of antidiabetic drug combinations with COVID-19 deaths. We examined whether adding common second-line agents to metformin was associated with COVID-19 mortality risk to inform clinical decision-making when escalating diabetes treatment. This is a nationwide retrospective analysis covering the years 2020 and 2021. Data from the National Diabetes Registry (CroDiab) were linked to primary healthcare data, Causes of Death Registry data, and the SARS-CoV-2 vaccination database. Multivariate logistic regression models were developed for each of the combinations to compare the combination with metformin monotherapy. To address confounders, inverse probability of treatment weighting (IPTW) analysis as well as analysis with stabilized weights was performed. Of 141014 analyzed patients, 1268 (0.90%) died of COVID-19 in 2 years. Weighted results of the drug combinations that showed statistically significant associations to COVID-19 death in comparison to metformin alone were metformin+DPP-4 inhibitor (OR 1.182, 95% CI 1.016-1.376), metformin+sulfonylurea (OR 1.195, 95% CI 1.015-1.406), and metformin+GLP-1 agonist (OR 2.992, 95% CI 2.117-4.229). Some combinations of metformin with second-line antidiabetic drugs might require caution in the context of chronic diabetes mellitus type 2 therapy and COVID-19 related deaths. Findings should be interpreted as hypothesis-generating signals from real-world data rather than evidence of causal drug effects. Further research is needed, especially for metformin+GLP-1 agonist, as well as head-to-head comparisons of combinations therapies.

To examine the causal effects of time-varying treatments on survival, structural nested cumulative survival time models (SNCSTMs) are flexible and theoretically promising semiparametric models characterized by causally interpretable parameters. One concern is the prerequisite for uniformly scheduled data collection and complete data for time-varying confounders. For example, in pharmacoepidemiological studies using medical information databases, laboratory test results can be missing due to unscheduled hospital visits or non-compliance with health checkups. Furthermore, missing mechanisms data may be non-ignorable and non-monotone, invalidating the typical missing-data methods that assume ignorable or monotone missing mechanisms. We propose a novel g-estimation method for SNCSTMs with non-ignorable, non-monotonic missing data for time-varying confounders. We augment the g-estimation functions using missing probability and imputation models, incorporating a user-defined selection function, which allows sensitivity analyses to evaluate the departure of missing data from ignorable mechanisms. Using a proper selection function, our estimator is doubly robust in the sense that it is consistent if either model for missing probability or imputation of missing data is correct at each time point and if either model for propensity score or conditional expectation of counterfactual counting processes is correct. Moreover, applying frequentist-type multiple imputation yields a closed-form solution for calculating the estimator, even if time-varying confounders are missing. A simulation study evaluated our proposed method's finite sample performance and the estimator's double robustness. We also conducted sensitivity analyses in a pharmacoepidemiological study using a Japanese medical claims database, assessing the risk of hypoglycemia in sulfonylurea-treated patients with incomplete hemoglobin A1c values.

Community-based physical and occupational therapy (PT/OT) are critical for stroke recovery but are underused. We conducted a secondary analysis of the COMPASS (Comprehensive Post-Acute Stroke Services) study, a pragmatic trial of comprehensive postacute transitional care (TC) to investigate whether TC programs increase PT/OT use. Forty hospitals were randomized to implement COMPASS-TC or maintain usual care for patients with stroke/transient ischemic attack. In a crossover phase, usual care hospitals implemented COMPASS-TC. We linked participants to administrative claims to assess PT/OT use after stroke. Adjusted generalized estimating equations compared COMPASS-TC to usual care within the trial and crossover cohorts on 30-/90-day PT/OT use, time to first visit, number of visits, and receipt of PT and OT versus single therapy. Per protocol analysis estimated complier average causal effects. COMPASS enrolled 8377 patients from July 2016 to March 2019; 5261 were linked to administrative claims. Thirty-day PT/OT ranged from 22.6% in usual care to 37.5% in COMPASS-TC. Therapy use was similar between groups in the trial cohort, and COMPASS-TC was associated with increased use in crossover analysis (9.4% [95% CI, 5.6-13.3%] at 30 days). COMPASS-TC was consistently associated with a shorter time to therapy (mean difference, -0.16 [95% CI, -0.03 to -0.29]). Per protocol results were larger for most outcomes. COMPASS-TC was associated with shorter time to PT/OT and with greater therapy receipt in the crossover, but not the trial, analysis. Inconsistencies may reflect confounding or differences in hospitals that chose to adopt the intervention in Phase 2. Implementation studies to improve care transitions after stroke are needed to enhance use of postacute rehabilitation.

Major depressive disorder (MDD), smoking, and drinking frequently co-occur, with evidence suggesting these relationships may differ by sex. However, the direction of causality and the extent of sex-specific associations remain unclear. We investigated sex-specific genetic relationships between MDD and substance use phenotypes using genome-wide association studies (GWAS) from the UK Biobank and publicly available sex-stratified GWAS for MDD and problematic alcohol use (PAU). Causal effects were assessed using bidirectional, sex-stratified Mendelian randomization (MR). We further applied multivariable MR (MVMR) to evaluate the influence of socioeconomic status (SES). Genetic correlation analyses indicated significant shared genetic architecture between MDD and all substance use traits in sex-combined GWAS. In sex-specific analyses, the correlation between cigarettes per day and MDD was significantly stronger in females, and drinks per week were correlated with MDD only in females. MR analyses showed that genetic liability to MDD increased the risk of smoking initiation and PAU in females, and was associated with reduced alcohol drinking frequency in males. In contrast, no tested substance use trait showed evidence of a causal effect on MDD in either sex. MVMR adjusting for SES attenuated the association between MDD and smoking initiation. The effect on PAU in females remained. In males, the negative association between MDD and drinking frequency became non-significant after SES adjustment. These findings reveal sex-specific genetic and causal relationships between smoking, drinking, and MDD, and highlight the role of SES as a potential confounder. Incorporating sex and socioeconomic context is critical when examining these associations.

ABSTRACT Global financial disturbances challenge countries’ ability to maintain stable economic performance, raising questions about the effectiveness of monetary policy frameworks. This study evaluates how inflation-targeting (IT) countries respond to external spillovers compared to non-IT economies. Using entropy balancing combined with a differences-in-differences strategy for 1980–2019, we address self-selection and estimate causal effects on real per capita income, debt-to-GDP, and employment. The results indicate that IT adoption may involve higher employment costs under shifting global conditions, while effects on output and public debt vary across income groups. Overall, the findings suggest that IT economies display heightened vulnerability to external spillovers.

Association of body fat distribution with bone mineral density: evidence from observational and mendelian randomization analyses.

Abstract Objective Observational studies have linked gastroesophageal reflux disease (GERD) to various otolaryngologic diseases (ear, nose, and throat [ENT] disorders), but causal inference is limited by confounding and reverse causation. We used Mendelian randomization (MR) to evaluate causality while minimizing these biases. Methods MR analyses were conducted using genome-wide association study (GWAS) data from individuals of European ancestry. GERD data were derived from a meta-analysis of UK Biobank and QSKIN ( N = 602,604; cases = 129,080), and outcome data were obtained from FinnGen (outcome sample sizes ranged from 157,453 to 404,309). Six complementary MR methods (including IVW and MR-Egger) were applied to assess the causal effect of GERD on seven otolaryngologic diseases: allergic rhinitis (AR), chronic sinusitis (CRS), nasal polyps (NP), vocal cord dysfunction (VCD), sudden idiopathic hearing loss (SIHL), head and neck cancer (HNC), and thyroid carcinoma (THCA). After harmonization, the number of SNPs included in the analyses varied across outcomes, ranging from 65 to 75. Sensitivity analyses, including tests for heterogeneity, multiplicity, and leave-one-out analysis, were conducted to ensure robustness of the results. Results GERD was causally associated with an increased risk of AR–OR 1.17(95% CI 1.05–1.30), CRS–OR 1.25(95% CI 1.15–1.37), VCD–OR 1.52(95% CI 1.31–1.77), and SIHL–OR 1.38(95% CI 1.14–1.67). No causal effect was found for NP–OR 1.09(95% CI 0.95–1.25), HNC–OR 1.13(95% CI 0.90–1.42), or THCA–OR 1.07 (95% CI 0.83–1.39). Conclusion These results provide genetic evidence supporting a causal association between GERD and increased risks of AR, CRS, VCD, and SIHL. If validated in further studies, improved GERD prevention and management strategies could potentially reduce these risks.

Exploring the causal relationship between osteoporosis and age-related macular degeneration: Evidence from observational studies and mendelian randomization.

Systematic reviews of quasi-experimental studies: challenges and considerations.

One major bias source in causal inference for clinical trials is unmeasured confounding. We propose an innovative, practical Bayesian modeling approach to adjust for unmeasured confounding effects and obtain precise causal average treatment effect estimates for two-arm randomized controlled clinical trials. This approach includes model reparameterization and an iterative algorithm, with a causal inference framework incorporated with unmeasured confounders and related statistical distributions. Model non-identifiability resulting from adjusting for unmeasured confounding is a major inferential problem. Reparameterization transforms one or multiple unmeasured confounders into a single reparameterized unmeasured confounder and can remove model non-identifiability from the model specification of unmeasured confounders. The iterative algorithm consists of detailed steps for inference after model reparameterization and can remove model non-identifiability from prior sensitivity to unmeasured confounders. It includes iterating the prior distribution of the reparameterized unmeasured confounder by certain rules, aggregating posterior means and variances over different prior choices, and obtaining posterior estimates for the average treatment effect. Its essential idea is to make unreliable prior information on unmeasured confounders as close to data information as possible. Compared with usual methods, our approach produces robust effect estimates and correctly concludes statistical significance. From an example using real clinical data, this approach effectively adjusts for confounding effects when we do not adjust for measured confounders. Our approach is also generalizable to other clinical study designs and may be beneficial to applications where data collection is difficult for certain variables or causal relationships are not well understood.

Associations of physical activity and sedentary behaviors with mortality: An observational analysis and Mendelian randomization study.

Structural equation modeling for causal effect estimation with machine learning

Survival of Patients With Noncolorectal Non-Neuroendocrine Liver Metastases: A Nationwide Cohort Study From the Danish Liver Cancer Group.

Task-level causal effects of maintenance staffing shortages on aircraft-on-ground (AOG) and reliability

Discharge from mental health service admissions as a short-term causal risk factor for suicide: A case-crossover study.

Estimating the impact of out-of-home placement on health risk behavior in adolescents exposed to maltreatment: An advanced causal inference approach.

Exploring the causal effects of ICT on creative thinking: A perspective of academic performance differences