Multiply robust estimation of marginal structural models in observational studies subject to covariate-driven observations.

Electronic health records and other sources of observational data are increasingly used for drawing causal inferences. The estimation of a causal effect using these data not meant for research purposes is subject to confounding and irregularly-spaced covariate-driven observation times affecting the inference. A doubly-weighted estimator accounting for these features has previously been proposed that relies on the correct specification of two nuisance models used for the weights. In this work, we propose a novel consistent multiply robust estimator and demonstrate analytically and in comprehensive simulation studies that it is more flexible and more efficient than the only alternative estimator proposed for the same setting. It is further applied to data from the Add Health study in the United States to estimate the causal effect of therapy counseling on alcohol consumption in American adolescents.

Mapping complex public health problems with causal loop diagrams.

This paper presents causal loop diagrams (CLDs) as tools for studying complex public health problems like health inequality. These problems often involve feedback loops-a characteristic of complex systems not fully integrated into mainstream epidemiology. CLDs are conceptual models that visualize connections between system variables. They are commonly developed through literature reviews or participatory methods with stakeholder groups. These diagrams often uncover feedback loops among variables across scales (e.g. biological, psychological and social), facilitating cross-disciplinary insights. We illustrate their use through a case example involving the feedback loop between sleep problems and depressive symptoms. We outline a typical step-by-step process for developing CLDs in epidemiology. These steps are defining a specific problem, identifying the key system variables involved, mapping these variables and analysing the CLD to find new insights and possible intervention targets. Throughout this process, we suggest triangulating between diverse sources of evidence, including domain knowledge, scientific literature and empirical data. CLDs can also be evaluated to guide policy changes and future research by revealing knowledge gaps. Finally, CLDs may be iteratively refined as new evidence emerges. We advocate for more widespread use of complex systems tools, like CLDs, in epidemiology to better understand and address complex public health problems.

Reverse causation bias: A simulation study comparing first- and second-line treatments with an overlap of symptoms between treatment indication and studied outcome.

BACKGROUND: Reverse causation is a challenge in many drug-cancer associations, where the cancer symptoms are potentially mistaken for drug indication symptoms. However, tools to assess the magnitude of this type of bias are currently lacking. We used a simulation-based approach to investigate the impact of reverse causation on the association between the use of topical tacrolimus and cutaneous T-cell lymphoma (CTCL) in a multinational, population-based study using topical corticosteroids (TCS) as comparator. METHODS: We used a multistate model to simulate patients' use over time of a first- (TCS) and second-line treatment (topical tacrolimus), onset of atopic dermatitis (indication for drugs) and CTCL (the studied outcome). We simulated different scenarios to mimic real-life use of the two treatments. In all scenarios, it was assumed that there was no causal effect of the first- or second-line treatment on the occurrence of CTCL. Simulated data were analysed using Cox proportional hazards models. RESULTS: The simulated hazard ratios (HRs) of CTCL for patients treated with tacrolimus vs. TCS were consistently above 1 in all 9 settings in the main scenario. In our main analysis, we observed a median HR of 3.09 with 95% of the observed values between 2.11 and 4.69. CONCLUSIONS: We found substantial reverse causation bias in the simulated CTCL risk estimates for patients treated with tacrolimus vs. TCS. Reverse causation bias may result in a false positive association between the second-line treatment and the studied outcome, and this simulation-based framework can be adapted to quantify the potential reverse causation bias.

A Causality Assessment Framework for COVID-19 Vaccines and Adverse Events at the COVID-19 Vaccine Safety Research Center.

During the coronavirus disease 2019 (COVID-19) pandemic, conclusively evaluating possible associations between COVID-19 vaccines and potential adverse events was of critical importance. The National Academy of Medicine of Korea established the COVID-19 Vaccine Safety Research Center (CoVaSC) with support from the Korea Disease Control and Prevention Agency to investigate the scientific relationship between COVID-19 vaccines and suspected adverse events. Although determining whether the COVID-19 vaccine was responsible for any suspected adverse event necessitated a systematic approach, traditional causal inference theories, such as Hill's criteria, encountered certain limitations and criticisms. To facilitate a systematic and evidence-based evaluation, the United States Institute of Medicine, at the request of the Centers for Disease Control and Prevention, offered a detailed causality assessment framework in 2012, which was updated in the recent report by the National Academies of Sciences, Engineering, and Medicine (NASEM) in 2024. This framework, based on a weight-of-evidence approach, allows the independent evaluation of both epidemiological and mechanistic evidence, culminating in a comprehensive conclusion about causality. Epidemiological evidence derived from population studies is categorized into four levels-high, moderate, limited, or insufficient-while mechanistic evidence, primarily from biological and clinical studies in animals and individuals, is classified as strong, intermediate, weak, or lacking. The committee then synthesizes these two types of evidence to draw a conclusion about the causal relationship, which can be described as "convincingly supports" ("evidence established" in the 2024 NASEM report), "favors acceptance," "favors rejection," or "inadequate to accept or reject." The CoVaSC has established an independent committee to conduct causality assessments using the weight-of-evidence framework, specifically for evaluating the causality of adverse events associated with COVID-19 vaccines. The aim of this study is to provide an overview of the weight-of-evidence framework and to detail the considerations involved in its practical application in the CoVaSC.

The causality between leisure sedentary behaviors, physical activity and obstructive sleep apnea: a bidirectional Mendelian randomization study.

Background: Previous observational studies have shown a correlation between leisure sedentary behaviors (LSB) and physical activity (PA) with the incidence of obstructive sleep apnea (OSA). However, the causal associations remain unknown. Therefore, our study used bidirectional two-sample Mendelian randomization (MR) to identify potential causal relationships between LSB/PA and OSA. Methods: We sourced genetic variation data for LSB and PA from the UK Biobank, while data on OSA were collected from the FinnGen study. The primary analysis method employed was the inverse variance weighted (IVW) approach, complemented by the weighted median and MR-Egger methods. For sensitivity analyses, we conducted Cochran's Q test, the MR-Egger intercept test, the MR-PRESSO global test, and the leave-one-out analysis. Results: IVW analyses showed that genetically predicted leisure television watching (odds ratio [OR] = 1.38, 95% confidence interval [CI] = 1.09-1.75, p = 0.007) and computer use (OR = 1.48, 95% CI = 1.15-1.92, p = 0.002) significantly increased the risk of OSA. Conversely, self-reported vigorous physical activity (VPA) (OR = 0.33, 95% CI = 0.11-0.98, p = 0.046) may reduce the risk of OSA. No causal effects on OSA risk were observed for driving or self-reported moderate-to-vigorous physical activity. Furthermore, the reverse MR analysis indicated no significant causal relationship between OSA and any LSB/PA phenotype. Sensitivity tests showed no significant heterogeneity or horizontal pleiotropy. Conclusion: This study suggests that leisurely television watching and computer use are risk factors for OSA, while VPA may be a protective factor. Additionally, OSA does not affect PA or LSB levels. We recommend reducing sedentary activities, particularly television watching and computer use, and prioritizing VPA to reduce the risk of OSA. Further research in diverse populations and settings is needed to validate these findings.

Bidirectional causality between the levels of blood lipids and endometriosis: a two-sample mendelian randomization study.

BACKGROUND: Observational studies have found a correlation between the levels of blood lipids and the development and progression of endometriosis (EM). However, the causality and direction of this correlation is unclear. This study aimed to examine the bidirectional connection between lipid profiles and the risk of EM using publicly available genome-wide association study (GWAS) summary statistics. METHODS: Eligible exposure variables such as levels of triglycerides (TG), total cholesterol (TC), low-density lipoprotein (LDL), and high-density lipoprotein (HDL) were selected using a two-sample Mendelian randomization (MR) analysis method following a series of quality control procedures. Data on EM were obtained from the publicly available Finnish database of European patients. Inverse variance weighted (IVW), MR Egger, weighted median, and weighted mode methods were used to analyze the causal relationship between lipid exposure and EM, exclude confounders, perform sensitivity analyses, and assess the stability of the results. Reverse MR analyses were performed with EM as exposure and lipid results as study outcomes. RESULTS: IVW analysis results identified HDL as a protective factor for EM, while TG was shown to be a risk factor for EM. Subgroup analyses based on the site of the EM lesion identified HDL as a protective factor for EM of the uterus, while TG was identified a risk factor for the EM of the fallopian tube, ovary, and pelvic peritoneum. Reverse analysis did not reveal any effect of EM on the levels of lipids. CONCLUSION: Blood lipids, such as HDL and TG, may play an important role in the development and progression of EM. However, EM does not lead to dyslipidemia.

The causal effect of atopic dermatitis on lung cancer: A Mendelian randomization study.

BACKGROUND: Growing evidence has shown that atopic dermatitis (AD) may decrease lung cancer (LC) risk. However, the causality between the two diseases is inconsistent and controversial. Therefore, we explored the causal relationship between AD and different histological subtypes of LC by using the Mendelian randomization (MR) method. MATERIALS AND METHODS: We conducted the MR study based on summary statistics from the genome-wide association studies (GWAS) of AD (10,788 cases and 30,047 controls) and LC (29,266 cases and 56,450 controls). Instrumental variables (IVs) were obtained after removing SNPs associated with potential confounders. We employed inverse-variance weighted (IVW), MR-Egger, and weighted median methods to pool estimates, and performed a comprehensive sensitivity analysis. RESULTS: The results of the IVW method suggested that AD may decrease the risk of developing lung adenocarcinoma (LUAD) (OR = 0.91, 95% CI: 0.85-0.97, P = 0.007). Moreover, no causality was identified between AD and overall LC (OR = 0.96, 95% CI: 0.91-1.01, P = 0.101), lung squamous cell carcinoma (LUSC) (OR = 1.04, 95% CI: 0.96-1.036, P = 0.324), and small cell lung carcinoma (SCLC) (OR = 0.95, 95% CI: 0.82-1.10, P = 0.512). A comprehensive sensitivity test showed the robustness of our results. CONCLUSION: The present study indicates that AD may decrease the risk of LUAD in the European population, which needs additional investigations to identify the potential molecular mechanisms.

Causal relationships between height, screen time, physical activity, sleep and myopia: univariable and multivariable Mendelian randomization.

Background: This study aims to investigate the independent causal relation between height, screen time, physical activity, sleep and myopia. Methods: Instrumental variables (IVs) for exposures and outcome were obtained from the largest publicly available genome-wide association studies (GWAS) databases. First, we performed a bidirectional univariate MR analysis using primarily the inverse variance weighted method (IVW) with height, screen time, physical activity and sleep as the exposure and myopia as the outcome to investigate the causal relationship between exposures and myopia. Sensitivity analysis was used to demonstrate its robustness. Then the multivariable MR (MVMR) and MR-based mediation approach was further used to estimate the mediating effect of potential confounders (education and time outdoors) on causality. Results: The results of univariate MR analysis showed that taller height (OR = 1.009, 95% CI = 1.005-1.012, p = 3.71 × 10-7), longer time on computer (OR = 1.048, 95% CI = 1.029-1.047, p = 3.87 × 10-7) and less moderate physical activity (OR = 0.976, 95% CI = 0.96-0.991 p = 2.37 × 10-3) had a total effect on the increased risk of developing myopia. Meanwhile our results did not have sufficient evidence to support the causal relationship between chronotype (p = 0.637), sleep duration (p = 0.952) and myopia. After adjusting for education, only taller height remains an independent risk factor for myopia. After adjusting for education, the causal relationship between height, screen and myopia still had statistical significance. A reverse causal relationship was not found in our study. Most of the sensitivity analyses showed consistent results with those of the IVW method. Conclusion: Our MR study revealed that genetically predicted taller height, longer time on computer, less moderate physical activity increased the risk of myopia. After full adjustment for confounders, only height remained independently associated with myopia. As a complement to observational studies, the results of our analysis provide strong evidence for the improvement of myopia risk factors and provide a theoretical basis for future measures to prevent and control myopia in adolescents.

Causal inference in the absence of positivity: The role of overlap weights.

How to analyze data when there is violation of the positivity assumption? Several possible solutions exist in the literature. In this paper, we consider propensity score (PS) methods that are commonly used in observational studies to assess causal treatment effects in the context where the positivity assumption is violated. We focus on and examine four specific alternative solutions to the inverse probability weighting (IPW) trimming and truncation: matching weight (MW), Shannon's entropy weight (EW), overlap weight (OW), and beta weight (BW) estimators. We first specify their target population, the population of patients for whom clinical equipoise, that is, where we have sufficient PS overlap. Then, we establish the nexus among the different corresponding weights (and estimators); this allows us to highlight the shared properties and theoretical implications of these estimators. Finally, we introduce their augmented estimators that take advantage of estimating both the propensity score and outcome regression models to enhance the treatment effect estimators in terms of bias and efficiency. We also elucidate the role of the OW estimator as the flagship of all these methods that target the overlap population. Our analytic results demonstrate that OW, MW, and EW are preferable to IPW and some cases of BW when there is a moderate or extreme (stochastic or structural) violation of the positivity assumption. We then evaluate, compare, and confirm the finite-sample performance of the aforementioned estimators via Monte Carlo simulations. Finally, we illustrate these methods using two real-world data examples marked by violations of the positivity assumption.

Assessing causal associations of bile acids with obesity indicators: A Mendelian randomization study.

Maintaining a balanced bile acids (BAs) metabolism is essential for lipid and cholesterol metabolism, as well as fat intake and absorption. The development of obesity may be intricately linked to BAs and their conjugated compounds. Our study aims to assess how BAs influence the obesity indicators by Mendelian randomization (MR) analysis. Instrumental variables of 5 BAs were obtained from public genome-wide association study databases, and 8 genome-wide association studies related to obesity indicators were used as outcomes. Causal inference analysis utilized inverse-variance weighted (IVW), weighted median, and MR-Egger methods. Sensitivity analysis involved MR-PRESSO and leave-one-out techniques to detect pleiotropy and outliers. Horizontal pleiotropy and heterogeneity were assessed using the MR-Egger intercept and Cochran Q statistic, respectively. The IVW analysis revealed an odds ratio of 0.94 (95% confidence interval: 0.88, 1.00; P = .05) for the association between glycolithocholate (GLCA) and obesity, indicating a marginal negative causal association. Consistent direction of the estimates obtained from the weighted median and MR-Egger methods was observed in the analysis of the association between GLCA and obesity. Furthermore, the IVW analysis demonstrated a suggestive association between GLCA and trunk fat percentage, with a beta value of -0.014 (95% confidence interval: -0.027, -0.0004; P = .04). Our findings suggest a potential negative causal relationship between GLCA and both obesity and trunk fat percentage, although no association survived corrections for multiple comparisons. These results indicate a trend towards a possible association between BAs and obesity, emphasizing the need for future studies.

SLE: a cognitive step forward-a synthesis of rethinking theories, causality, and ignored DNA structures.

Systemic lupus erythematosus (SLE) is classified by instinctual classification criteria. A valid proclamation is that these formally accepted SLE classification criteria legitimate the syndrome as being difficult to explain and therefore enigmatic. SLE involves scientific problems linked to etiological factors and criteria. Our insufficient understanding of the clinical condition uniformly denoted SLE depends on the still open question of whether SLE is, according to classification criteria, a well-defined one disease entity or represents a variety of overlapping indistinct syndromes. Without rational hypotheses, these problems harm clear definition(s) of the syndrome. Why SLE is not anchored in logic, consequent, downstream interdependent and interactive inflammatory networks may rely on ignored predictive causality principles. Authoritative classification criteria do not reflect consequent causality criteria and do not unify characterization principles such as diagnostic criteria. We need now to reconcile legendary scientific achievements to concretize the delimitation of what SLE really is. Not all classified SLE syndromes are "genuine SLE"; many are theoretically "SLE-like non-SLE" syndromes. In this study, progressive theories imply imperative challenges to reconsider the fundamental impact of "the causality principle". This may offer us logic classification and diagnostic criteria aimed at identifying concise SLE syndromes as research objects. Can a systems science approach solve this problem?

Incorporating nonparametric methods for estimating causal excursion effects in mobile health with zero-inflated count outcomes.

In mobile health, tailoring interventions for real-time delivery is of paramount importance. Micro-randomized trials have emerged as the "gold-standard" methodology for developing such interventions. Analyzing data from these trials provides insights into the efficacy of interventions and the potential moderation by specific covariates. The "causal excursion effect," a novel class of causal estimand, addresses these inquiries. Yet, existing research mainly focuses on continuous or binary data, leaving count data largely unexplored. The current work is motivated by the Drink Less micro-randomized trial from the UK, which focuses on a zero-inflated proximal outcome, i.e., the number of screen views in the subsequent hour following the intervention decision point. To be specific, we revisit the concept of causal excursion effect, specifically for zero-inflated count outcomes, and introduce novel estimation approaches that incorporate nonparametric techniques. Bidirectional asymptotics are established for the proposed estimators. Simulation studies are conducted to evaluate the performance of the proposed methods. As an illustration, we also implement these methods to the Drink Less trial data.

Revealing brain cell-stratified causality through dissecting causal variants according to their cell-type-specific effects on gene expression.

The human brain has been implicated in the pathogenesis of several complex diseases. Taking advantage of single-cell techniques, genome-wide association studies (GWAS) have taken it a step further and revealed brain cell-type-specific functions for disease loci. However, genetic causal associations inferred by Mendelian randomization (MR) studies usually include all instrumental variables from GWAS, which hampers the understanding of cell-specific causality. Here, we developed an analytical framework, Cell-Stratified MR (csMR), to investigate cell-stratified causality through colocalizing GWAS signals with single-cell eQTL from different brain cells. By applying to obesity-related traits, our results demonstrate the cell-type-specific effects of GWAS variants on gene expression, and indicate the benefits of csMR to identify cell-type-specific causal effect that is often hidden from bulk analyses. We also found csMR valuable to reveal distinct causal pathways between different obesity indicators. These findings suggest the value of our approach to prioritize target cells for extending genetic causation studies.

Investigating causal associations between pneumonia and lung cancer using a bidirectional mendelian randomization framework.

BACKGROUND: Pneumonia and lung cancer are both major respiratory diseases, and observational studies have explored the association between their susceptibility. However, due to the presence of potential confounders and reverse causality, the comprehensive causal relationships between pneumonia and lung cancer require further exploration. METHODS: Genome-wide association study (GWAS) summary-level data were obtained from the hitherto latest FinnGen database, COVID-19 Host Genetics Initiative resource, and International Lung Cancer Consortium. We implemented a bidirectional Mendelian randomization (MR) framework to evaluate the causal relationships between several specific types of pneumonia and lung cancer. The causal estimates were mainly calculated by inverse-variance weighted (IVW) approach. Additionally, sensitivity analyses were also conducted to validate the robustness of the causalty. RESULTS: In the MR analyses, overall pneumonia demonstrated a suggestive but modest association with overall lung cancer risk (Odds ratio [OR]: 1.21, 95% confidence interval [CI]: 1.01 - 1.44, P = 0.037). The correlations between specific pneumonia types and overall lung cancer were not as significant, including bacterial pneumonia (OR: 1.07, 95% CI: 0.91 - 1.26, P = 0.386), viral pneumonia (OR: 1.00, 95% CI: 0.95 - 1.06, P = 0.891), asthma-related pneumonia (OR: 1.18, 95% CI: 0.92 - 1.52, P = 0.181), and COVID-19 (OR: 1.01, 95% CI: 0.78 - 1.30, P = 0.952). Reversely, with lung cancer as the exposure, we observed that overall lung cancer had statistically crucial associations with bacterial pneumonia (OR: 1.08, 95% CI: 1.03 - 1.13, P = 0.001) and viral pneumonia (OR: 1.09, 95% CI: 1.01 - 1.19, P = 0.037). Sensitivity analysis also confirmed the robustness of these findings. CONCLUSION: This study has presented a systematic investigation into the causal relationships between pneumonia and lung cancer subtypes. Further prospective study is warranted to verify these findings.

Estimating the causal effects of income on health: how researchers' definitions of "income" matter.

BACKGROUND: There is a well-established cross-sectional association between income and health, but estimates of the causal effects of income vary substantially. Different definitions of income may lead to substantially different empirical results, yet research is often framed as investigating "the effect of income" as if it were a single, easily definable construct. METHODS/RESULTS: The aim of this paper is to introduce a taxonomy for definitional and conceptual issues in studying individual- or household-level income for health research. We focus on (1) the definition of the income measure (earned and unearned; net, gross, and disposable; real and nominal; individual and household; relative and absolute income) and (2) the definition of the causal contrast (amount, functional form assumptions/transformations, direction, duration of change, and timing of exposure and follow-up). We illustrate the application of the taxonomy to four examples from the published literature. CONCLUSIONS: Quantified estimates of causal effects of income on health and wellbeing have crucial relevance for policymakers to anticipate the consequences of policies targeting the social determinants of health. However, much prior evidence has been limited by lack of clarity in distinguishing between different causal questions. The present framework can help researchers explicitly and precisely articulate income-related exposures and causal questions.

Guide to Decomposition of Causal Effects Into Mediation, Interaction, and Direct Effects: Case Study on Aerobic Exercise and Parkinson Disease.

Mediation analysis can be applied in medical research with the aim of understanding the pathways that operate between an exposure and its effects on an outcome. This method can help to improve our understanding of pathophysiologic mechanisms and may guide the choice of potential treatment strategies. Traditional mediation analysis decomposes the total effect of an intervention on the outcome into 2 effects: (1) an indirect effect, from exposure using a mediator to the outcome, and (2) a direct effect, directly from exposure to outcome. A limitation of this method is that it assumes no interaction between the exposure and the mediator, which can either lead to an over- or underestimation of clinically relevant effects. The "4-way decomposition" method has the advantage of overcoming this limitation. Specifically, the total effect of an exposure on the outcome is decomposed into 4 elements: (1) reference interaction (interaction only), (2) mediated interaction (mediation and interaction), (3) the pure indirect effect (mediation but not interaction), and (4) the direct effect (no mediation and no interaction). We provide a guide to select the most appropriate method to investigate and decompose any causal effect given the research question at hand. We explain the application of the 4-way decomposition and illustrate this with a real-world example of how aerobic exercise may influence motor function in persons with Parkinson disease.

Causal effects for neurodegenerative diseases on the risk of myocardial infarction: a two-sample Mendelian randomization study.

Several studies have demonstrated a correlation between neurodegenerative diseases (NDDs) and myocardial infarction (MI), yet the precise causal relationship between these remains elusive. This study aimed to investigate the potential causal associations of genetically predicted Alzheimer's disease (AD), dementia with Lewy bodies (DLB), Parkinson's disease (PD), and multiple sclerosis (MS) with MI using two-sample Mendelian randomization (TSMR). Various methods, including inverse variance weighted (IVW), weighted median (WM), MR-Egger regression, weighted mode, and simple mode, were employed to estimate the effects of genetically predicted NDDs on MI. To validate the analysis, we assessed pleiotropic effects, heterogeneity, and conducted leave-one-out sensitivity analysis. We identified that genetic predisposition to NDDs was suggestively associated with higher odds of MI (OR_IVW=1.07, OR_MR-Egger=1.08, OR_WM=1.07, OR_weighted mode=1.07, OR_simple mode=1.10, all P<0.05). Furthermore, we observed significant associations of genetically predicted DLB with MI (OR_IVW=1.07, OR_MR-Egger=1.11, OR_WM=1.09, OR_weighted mode=1.09, all P<0.05). However, there was no significant causal evidence of genetically predicted PD and MS in MI. Across all MR analyses, no horizontal pleiotropy or statistical heterogeneity was observed (all P>0.05). Additionally, results from MRPRESSO and leave-one-out sensitivity analysis confirmed the robustness of the causal effect estimations for genetically predicted AD, DLB, PD, and MS on MI. This study provides further support for the causal effects of AD on MI and, for the first time, establishes robust causal evidence for the detrimental effect of DLB on the risk of MI. Our findings emphasize the importance of monitoring the cardiovascular function of the elderly experiencing neurodegenerative changes.

Causality Investigation between Gut Microbiome and Sleep-Related Traits: A Bidirectional Two-Sample Mendelian Randomization Study.

Recent research has highlighted associations between sleep and microbial taxa and pathways. However, the causal effect of these associations remains unknown. To investigate this, we performed a bidirectional two-sample Mendelian randomization (MR) analysis using summary statistics of genome-wide association studies (GWAS) from 412 gut microbiome traits (N = 7738) and GWAS studies from seven sleep-associated traits (N = 345,552 to 386,577). We employed multiple MR methods to assess causality, with Inverse Variance Weighted (IVW) as the primary method, alongside a Bonferroni correction ((p < 2.4 × 10-4) to determine significant causal associations. We further applied Cochran's Q statistical analysis, MR-Egger intercept, and Mendelian randomization pleiotropy residual sum and outlier (MR-PRESSO) for heterogeneity and pleiotropy assessment. IVW estimates revealed 79 potential causal effects of microbial taxa and pathways on sleep-related traits and 45 inverse causal relationships, with over half related to pathways, emphasizing their significance. The results revealed two significant causal associations: genetically determined relative abundance of pentose phosphate decreased sleep duration (p = 9.00 × 10-5), and genetically determined increase in fatty acid level increased the ease of getting up in the morning (p = 8.06 × 10-5). Sensitivity analyses, including heterogeneity and pleiotropy tests, as well as a leave-one-out analysis of single nucleotide polymorphisms, confirmed the robustness of these relationships. This study explores the potential causal relationships between sleep and microbial taxa and pathways, offering novel insights into their complex interplay.

Sensitivity analysis for principal ignorability violation in estimating complier and noncomplier average causal effects.

An important strategy for identifying principal causal effects (popular estimands in settings with noncompliance) is to invoke the principal ignorability (PI) assumption. As PI is untestable, it is important to gauge how sensitive effect estimates are to its violation. We focus on this task for the common one-sided noncompliance setting where there are two principal strata, compliers and noncompliers. Under PI, compliers and noncompliers share the same outcome-mean-given-covariates function under the control condition. For sensitivity analysis, we allow this function to differ between compliers and noncompliers in several ways, indexed by an odds ratio, a generalized odds ratio, a mean ratio, or a standardized mean difference sensitivity parameter. We tailor sensitivity analysis techniques (with any sensitivity parameter choice) to several types of PI-based main analysis methods, including outcome regression, influence function (IF) based and weighting methods. We discuss range selection for the sensitivity parameter. We illustrate the sensitivity analyses with several outcome types from the JOBS II study. This application estimates nuisance functions parametrically - for simplicity and accessibility. In addition, we establish rate conditions on nonparametric nuisance estimation for IF-based estimators to be asymptotically normal - with a view to inform nonparametric inference.

Causal relationship between sleep disorders and the risk of Alzheimer's disease: A Mendelian randomization study.

BACKGROUND AND OBJECTIVE: Epidemiological studies have shown that sleep disorders are risk factors for Alzheimer's disease (AD), but the causal relationship between sleep disorders and AD risk is unknown. We aim to assess the potential genetic causal association between sleep characteristics and AD, which may contribute to early identification and prediction of risk factors for AD. METHODS: Seven sleep-related traits and the outcome phenotype AD were selected from published genome-wide association studies (GWASs). These sleep-related characteristics and instrumental variables (IVs) for AD were extracted. Two-sample and multivariate Mendelian randomization (MR) analyses were performed to assess the causal relationships between sleep characteristics and AD. The inverse variance weighted (IVW), weighted median (WME), weighted mode (WM), MR-Egger regression (MR-Egger) and simple mode (SM) models were used to evaluate causality. The existence of pleiotropy was detected and corrected by MR-Egger regression, MR pleiotropy residuals and outliers. RESULTS: A two-sample MR study revealed a positive causal association between sleep duration and the onset of AD (OR = 1.002, 95 % CI: 1.000-1.004), and the risk of AD increased with increasing sleep duration. The MR-Egger regression method and MR-PRESSO were used to identify and correct pleiotropy, indicating that there was no horizontal pleiotropy. Heterogeneity was evaluated by Cochran's Q, which indicated no heterogeneity. In a multivariate MR study with seven sleep characteristics corrected for each other, we found that sleep duration remained causally associated with AD (OR = 1.004, 95 % CI: 1.000-1.007). Moreover, we found that after mutual correction, daytime napping had a causal relationship with the onset of AD, and daytime napping may reduce the risk of AD (OR = 0.995, 95 % CI: 0.991-1.000). CONCLUSION: This study is helpful for the early identification and prediction of risk factors for AD, long sleep durations are a risk factor for AD, and daytime napping can reduce the risk of AD.