This meta-analysis examines rigorous longitudinal 21st century studies on the associations of spirituality with harmful or hazardous alcohol and other drug (AOD) use. To synthesize findings from independent studies about spirituality and AOD use and to produce a comprehensive estimate of the overall effect size of the associated risk reduction. Studies previously identified in the Balboni and colleagues review on the association between spiritual exposures (including religion) and alcohol, tobacco, marijuana, or other drugs were pooled. Studies were identified through the search terms spirituality or religion or spiritual* or religio* or faith and also intersected with a long string of terms that captured health outcomes of interest. From an initial retrieval of more than 20 000 articles, a total of 55 spirituality studies (as defined by Puchalski and colleagues) that were (1) published 2000-2022 in the English language, (2) used validated measures of spirituality, (3) examined longitudinal associations between spirituality and AOD use, and (4) were either prospective cohort studies with sample sizes of 1000 or more or randomized clinical trials (eg, public health interventions) with sample sizes of 100 or more, were captured. Preferred Reporting Items for Systematic Reviews and Meta-Analyses ( PRISMA ) reporting guidelines were used for abstracting data and assessing quality and validity. Eligible studies were those that reported quantitative outcomes measuring AOD use in relation to spiritual exposures, provided sufficient data to calculate log-relative risks (log-RR) and associated error terms, and focused on either preventive effect (eg, delayed initiation) or recovery-related outcomes (eg, cessation). Effects extracted were transformed into log-RR based on the type of effect. The primary outcome was the association between spiritual or religious involvement and AOD. Subgroup analyses examined differences by AOD use type (alcohol, tobacco, marijuana, and illicit drugs) and exposure type (spiritual or religious attendance vs broader spiritual exposures). Results from the 55 studies, which collectively included 540 712 participants, documented a significant protective association related to both prevention and recovery between spirituality and AOD use outcomes. Specifically, a consistent 13% risk reduction extended across the studied drugs (RR, 0.87; 95% CI, 0.84-0.91), a figure that reached 18% for individuals engaging in spiritual or religious communities (defined as >weekly religious service attendance; RR, 0.82; 95% CI, 0.75-0.89). Virtually all 134 effects extracted from the studies demonstrated protective, not detrimental, results. Multiple sensitivity analyses confirmed the robustness of evidence. The results of this meta-analysis regarding a protective association between spirituality and AOD use have implications for clinicians and communities regarding future strategies for AOD use prevention and recovery.

Complete-case analysis (CCA) is often criticized on the belief that CCA is only valid if data are missing-completely-at-random (MCAR). Influential papers have thus recommended abandoning CCA in favor of methods that make a weaker missing-at-random (MAR) assumption. We argue for a different view: that CCA with principled covariate adjustment provides a valuable complement to MAR-based methods, such as multiple imputation. When estimating treatment effects, appropriate covariate control can, for some causal structures, eliminate bias in CCA. This can be true even when data are missing-not-at-random (MNAR) and when MAR-based methods are biased. We describe principles for choosing adjustment covariates for CCA, and we characterize the causal structures for which covariate adjustment does, or does not, eliminate bias. Even when CCA is biased, principled covariate adjustment will often reduce the bias of CCA, and this method will sometimes be less biased than MAR-based methods. When multiple imputation is used under a MAR assumption, adjusted CCA thus still constitutes an important sensitivity analysis. When conducted with the same attention to covariate control that epidemiologists already afford to confounding, adjusted CCA belongs in the suite of reasonable methods for missing data. There is thus good justification for resurrecting CCA as a principled method.

Abstract Background Asian American/ Pacific Islander (AAPI) groups experience different burdens of severe maternal morbidity (SMM) and pregnancy-related comorbidities that contribute to SMM. We sought to estimate counterfactual disparities for selected AAPI groups, separating out the pathway that includes these comorbidities. Methods We used linked birth certificates and hospital discharge data from births in California from 2011-2020. We examined the presence of pregnancy-related comorbidities (gestational hypertension and diabetes, pre-eclampsia) as a mediator between race/ethnicity and severe maternal morbidity. We used marginal structural models to estimate total effects and controlled direct effects (CDEs) (adjusted for maternal characteristics) for each AAPI group with Non-Hispanic Whites as the referent group. Results Our sample included n=1,849,698 births. All AAPI groups had higher prevalences of comorbidities compared to Non-Hispanic Whites (e.g., White: 15.7%, Chinese: 17.9%: Indian: 25.2%, Filipina: 28.8%, Pacific Islander: 23.9%). Filipinas and Pacific Islanders experienced the largest disparities in SMM (e.g., Total effect risk ratios (RR) Chinese: 1.03 (95% Confidence Interval (CI): 1.00, 1.07); Filipina: 1.64 (95% CI: 1.58, 1.70)). Under M=1 conditions, where everyone experienced comorbidities, disparities were eliminated for Chinese and Indian groups and alleviated for Filipinas and Pacific Islanders (e.g., CDE Chinese: 0.75 (95% CI: 0.69, 0.81); Filipina: 1.21 (95% CI: 1.13, 1.29)). Under M=0, disparities remained similar to the total effect (e.g., CDE Chinese: 1.14 (95% CI: 1.09, 1.19); CDE Filipina: 1.64 (95% CI: 1.53, 1.71)) Conclusions Pregnancy-related comorbidities contributed substantially to disparities for AAPI groups. Disparities persisted for Filipinas and Pacific Islanders, suggesting a need for tailored interventions.

Decreasing meat consumption is a critical element of the EAT-Lancet directive to improve human and planetary health, but scalable, effective solutions remain elusive. Plant-based meat analogues are lauded as a promising approach, but their impact on meat demand remains unknown. We tested whether increasing the number of meat analogues on a restaurant menu would decrease meat selection, as well as whether offering a novel chicken-like meat analogue would specifically decrease chicken selection. In a preregistered, randomized, controlled experiment, 4431 English-fluent adults in the U.S. viewed different versions of the menu from Chipotle. Participants in the three arms were shown a Chipotle menu with the pre-existing meat analogue option, "sofritas", removed (0 meat analogues); the standard Chipotle menu (1 meat analogue); or the menu with an added, fictitious, meat analogue, "chick'nitas" (2 meat analogues). Adding one or two meat analogues to the menu did not meaningfully reduce the proportion of participants selecting animal-based meat. Offering one meat analogue versus none produced only a 1.14 percentage point (pp) decrease in meat selection (95% CI [-1.02, 3.30], P = .30). For two meat analogues versus none, the estimated decrease was a negligible 2.14 pp. (95% CI [-0.08, 4.36], P = .06). However, availability of a chicken meat analogue slightly reduced demand for chicken specifically by -3.65 pp. (95% CI [-7.16, -0.15], P = .04). Our findings do not support the hypothesis that expanding meat analogue offerings alone can meaningfully shift consumer choices away from meat.

Missing data are a pervasive problem in epidemiology, with multiple imputation (MI) a commonly used analysis method. MI is valid when data are missing at random (MAR). However, definitions of MAR with multiple incomplete variables are not easily interpretable and descriptions of graphical model-based conditions are not accessible to applied researchers. Previous literature shows that MI may be valid in subsamples, even if not in the full dataset. Practical guidance on applying MI with multiple incomplete variables is lacking. We present an algorithm using directed acyclic graphs to determine when MI will estimate an exposure-outcome coefficient without bias. We extend the algorithm to assess whether MI in a subsample of the data, in which some variables are complete, and the remaining are imputed, will be valid and unbiased for the exposure-outcome coefficient. We apply the algorithm to several simple exemplars, and in a more complex real-life example highlight that only subsample-MI of the outcome would be valid. Our algorithm provides researchers with the tools to decide whether to use MI in practice when there are multiple incomplete variables. Further work could focus on the likely size and direction of biases and the impact of different missing data patterns.

Importance:This meta-analysis examines rigorous longitudinal 21st century studies on the associations of spirituality with harmful or hazardous alcohol and other drug (AOD) use. Objective:To synthesize findings from independent studies about spirituality and AOD use and to produce a comprehensive estimate of the overall effect size of the associated risk reduction. Data Sources:Studies previously identified in the Balboni and colleagues review on the association between spiritual exposures (including religion) and alcohol, tobacco, marijuana, or other drugs were pooled. Studies were identified through the search terms spirituality or religion or spiritual* or religio* or faith and also intersected with a long string of terms that captured health outcomes of interest. Study Selection:From an initial retrieval of more than 20 000 articles, a total of 55 spirituality studies (as defined by Puchalski and colleagues) that were (1) published 2000-2022 in the English language, (2) used validated measures of spirituality, (3) examined longitudinal associations between spirituality and AOD use, and (4) were either prospective cohort studies with sample sizes of 1000 or more or randomized clinical trials (eg, public health interventions) with sample sizes of 100 or more, were captured. Data Extraction and Synthesis:Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) reporting guidelines were used for abstracting data and assessing quality and validity. Eligible studies were those that reported quantitative outcomes measuring AOD use in relation to spiritual exposures, provided sufficient data to calculate log-relative risks (log-RR) and associated error terms, and focused on either preventive effect (eg, delayed initiation) or recovery-related outcomes (eg, cessation). Effects extracted were transformed into log-RR based on the type of effect. Main Outcomes and Measures:The primary outcome was the association between spiritual or religious involvement and AOD. Subgroup analyses examined differences by AOD use type (alcohol, tobacco, marijuana, and illicit drugs) and exposure type (spiritual or religious attendance vs broader spiritual exposures). Results:Results from the 55 studies, which collectively included 540 712 participants, documented a significant protective association related to both prevention and recovery between spirituality and AOD use outcomes. Specifically, a consistent 13% risk reduction extended across the studied drugs (RR, 0.87; 95% CI, 0.84-0.91), a figure that reached 18% for individuals engaging in spiritual or religious communities (defined as >weekly religious service attendance; RR, 0.82; 95% CI, 0.75-0.89). Virtually all 134 effects extracted from the studies demonstrated protective, not detrimental, results. Multiple sensitivity analyses confirmed the robustness of evidence. Conclusions and Relevance:The results of this meta-analysis regarding a protective association between spirituality and AOD use have implications for clinicians and communities regarding future strategies for AOD use prevention and recovery.

Background There is a lack validated outcome measures to assess wound severity in epidermolysis bullosa simplex (EBS).Objectives To assess the reliability and validity of the Investigator's Global Assessment (IGA) scale and a newly developed palms/soles subscale through in-clinic scoring and review of patient-submitted photographs.Methods Seven eligible patients with biopsy-proven or genetically confirmed diagnosis of EBS were recruited. Seven expert raters graded 30 wounds at a 1-day in-clinic scoring exercise using an IGA scale that was previously developed for assessing wound severity in EBS clinical trials. A new palms/soles subscale was developed by the authors for palms and soles wounds. Reliability and construct validity with the Epidermolysis Bullosa Disease Activity and Scarring Index-Activity (EBDASI-Activity) subscore, target wound blister count, the Quality of Life in Epidermolysis Bullosa (QOL-EB) questionnaire, itch and pain measures were assessed. Photographs of the same wounds provided by the patients were sent to the same raters to evaluate intra-rater reliability through a virtual exercise.Results Seven clinicians with EB expertise assessed 30 target wounds. Patients (female n = 5/7) had a mean age of 28.3 years (range 7.1-48.8 years). Strong inter-rater reliability was observed for the in-clinic scoring (Kendall's coefficient of concordance, 0.77; intraclass coefficient, 0.70) with excellent agreement (weighted quadratic kappa, 0.86). A positive association with target wound blister count, EBDASI-Activity, and itch and pain scores were observed. Home photograph scoring also showed strong inter-rater reliability and excellent agreement (Kendall's coefficient of concordance, 0.78; weighted quadratic kappa, 0.84; intraclass coefficient, 0.66). When home-photograph scores were compared to in-clinic scores rendered on corresponding wounds, strong intra-rater and substantial agreement was demonstrated (Kendall's coefficient of concordance, 0.86; weighted quadratic kappa, 0.67; intraclass coefficient, 0.68).Conclusions The IGA scale for EBS demonstrated strong reliability and excellent agreement in assessing wound severity of target lesions both in-clinic and via patient-submitted photographs.

Background: Missing data is a pervasive problem in epidemiology, with complete records analyses (CRA) or multiple imputation (MI) the most common methods to deal with incomplete data. MI is valid when incomplete variables are independent of response indicators, conditional on complete variables - however, this can be hard to assess with multiple incomplete variables. Previous literature has shown that MI may be valid in subsamples of the data, even if not necessarily valid in the full dataset. Current guidance on how to decide whether MI is appropriate is lacking. Methods: We develop an algorithm that is sufficient to indicate when MI will estimate an exposure-outcome coefficient without bias and show how to implement this using directed acyclic graphs (DAGs). We extend the algorithm to investigate whether MI applied to a subsample of the data, in which some variables and complete and the remaining are imputed, will be unbiased for the same estimand. We demonstrate the algorithm by applying it to several simple examples and a more complex real-life example. Conclusions: Multiple incomplete variables are common in practice. Assessing the plausibility of each of CRA and MI estimating an exposure-outcome association without bias is crucial in analysing and interpreting results. Our algorithm provides researchers with the tools to decide whether (and how) to use MI in practice. Further work could focus on the likely size and direction of biases, and the impact of different missing data patterns.

In small meta-analyses (e.g., up to 20 studies), the best-performing frequentist methods can yield very wide confidence intervals for the meta-analytic mean, as well as biased and imprecise estimates of the heterogeneity. We investigate the frequentist performance of alternative Bayesian methods that use the invariant Jeffreys prior. This prior has the usual Bayesian motivation, but also has a purely frequentist motivation: the resulting posterior modes correspond to the established Firth bias correction of the maximum likelihood estimator. We consider two forms of the Jeffreys prior for random-effects meta-analysis: the previously established “Jeffreys1” prior treats the heterogeneity as a nuisance parameter, whereas the “Jeffreys2” prior treats both the mean and the heterogeneity as estimands of interest. In a large simulation study, we assess the performance of both Jeffreys priors, considering different types of Bayesian estimates and intervals. We assess point and interval estimation for both the mean and the heterogeneity parameters, comparing to the best-performing frequentist methods. For small meta-analyses of binary outcomes, the Jeffreys2 prior may offer advantages over standard frequentist methods for point and interval estimation of the mean parameter. In these cases, Jeffreys2 can substantially improve efficiency while more often showing nominal frequentist coverage. However, for small meta-analyses of continuous outcomes, standard frequentist methods seem to remain the best choices. The best-performing method for estimating the heterogeneity varied according to the heterogeneity itself. Röver & Friede’s R package bayesmeta implements both Jeffreys priors. We also generalize the Jeffreys2 prior to the case of meta-regression.

We respond to Lu et al.'s commentary on our recent paper.

By revisiting imputation from the modern perspective of missing data graphs, we correct common guidance about which auxiliary variables should be included in an imputation model. We propose a generalized definition of missing at random (MAR), termed z-MAR, which distinguishes analysis variables from auxiliary variables included only in the imputation model. We derive a corresponding graphical condition, the m-backdoor criterion, which characterizes when a broad class of MAR-based imputation models yields valid inferences. In a sense we formalize, the m-backdoor criterion is both necessary and sufficient for valid imputation. Whereas common guidance is to include all available auxiliary variables in the imputation model, our results show that some auxiliary variables (e.g., colliders) can exacerbate bias. Thus, the modeled auxiliary variables should be restricted, at minimum, to those that causally affect either the incomplete variables or the missingness indicators. This set will always suffice to ensure valid imputation, if any sufficient set does exist among the candidate auxiliary variables. Applying this result does not require full knowledge of the graph. We demonstrate these results through simulations and an empirical application to the Stockholm Public Health Cohort, where including a presumed collider in the imputation model substantially altered estimates.

When analyzing a selected sample from a general population, selection bias can arise relative to the causal average treatment effect (ATE) for the general population, and also relative to the ATE for the selected sample itself. In this paper, we provide simple graphical rules that indicate (1) whether a selected-sample analysis will be unbiased for each ATE and (2) whether adjusting for certain covariates could eliminate selection bias. The rules can easily be checked in a standard single-world intervention graph. When the treatment could affect selection, a third estimand of potential scientific interest is the "net treatment difference"-namely the net change in outcomes that would occur for the selected sample if all members of the general population were treated versus not treated, including any effects of the treatment on which individuals are in the selected sample. We provide graphical rules for this estimand as well. We decompose bias in a selected-sample analysis relative to the general-population ATE into (1) "internal bias" relative to the net treatment difference and (2) "net-external bias," a discrepancy between the net treatment difference and the general-population ATE. Each bias can be assessed unambiguously via a distinct graphical rule, providing new conceptual insight into the mechanisms by which certain causal structures produce selection bias.

Observational studies are critical tools in clinical research and public health response, but challenges arise in ensuring the data produced by these studies are scientifically robust and socially valuable. Resolving these challenges requires careful attention to prioritising the most valuable research questions, ensuring robust study design, strong data management practices, expansive community engagement, and access and benefit sharing of results and research materials. This paper opens with a discussion of how well-designed observational studies contribute to biomedical evidence and provides examples from across the clinical literature of how these methods generate hypotheses for future research and uncover otherwise unattainable insights by providing examples from across the clinical literature. Then, we present obstacles that remain in ensuring observational studies are optimally designed, conducted and communicated.

Attention checks are often used to identify and exclude participants who may be responding carelessly. There has been little statistical guidance on the analysis of such studies and on when it is indeed valid to simply exclude inattentive participants. To address this, I first formalize attention checks as measures intended to identify participants whose responses are free of measurement error. Measurement error could arise not only because of careless responding but also if some participants fail to receive the experimental manipulation in its intended form because they did not attend to its contents. I discuss the statistical assumptions under which it is valid to simply exclude inattentive participants. In randomized experiments, this standard analysis may lead to bias if (a) the dependent variable affects attentiveness or (b) there are variables that affect both attentiveness and the dependent variable. The latter assumption is stringent and is likely to be violated in many studies. I suggest a straightforward modification to the standard approach, that is, controlling for variables that affect both attentiveness and the dependent variable. This covariate-adjusted approach requires considerably less stringent assumptions. In two worked examples, I reanalyze previously published experiments on (a) a documentary intended to reduce consumption of meat and animal products and (b) flag-priming effects on political conservatism.

We respond to Madley-Dowd et al's recent article in American Journal of Epidemiology. We show that standard imputation algorithms can fail for simple graphs (such as those used in Madley-Dowd et al's simulation study) even when the full data distribution is identified and an appropriate imputation estimator would be straightforward to design.

Understanding the causal genetic architecture of complex phenotypes is essential for future research into disease mechanisms and potential therapies. Here, we present a novel framework for genome-wide detection of sets of variants that carry non-redundant information on the phenotypes and are therefore more likely to be causal in a biological sense. Crucially, our framework requires only summary statistics obtained from standard genome-wide marginal association testing. The described approach, implemented in open-source software, is also computationally efficient, requiring less than 15 minutes on a single CPU to perform genome-wide analysis. Through extensive genome-wide simulation studies, we show that the method can substantially outperform usual two-stage marginal association testing and fine-mapping procedures in precision and recall. In applications to a meta-analysis of ten large-scale genetic studies of Alzheimer s disease (AD), we identified 82 loci associated with AD, including 37 additional loci missed by conventional GWAS pipeline. The identified putative causal variants achieve state-of-the-art agreement with massively parallel reporter assays and CRISPR-Cas9 experiments. Additionally, we applied the method to a retrospective analysis of 67 large-scale GWAS summary statistics since 2013 for a variety of phenotypes. Results reveal the method's capacity to robustly discover additional loci for polygenic traits and pinpoint potential causal variants underpinning each locus beyond conventional GWAS pipeline, contributing to a deeper understanding of complex genetic architectures in post-GWAS analyses.

Average treatment effects (ATEs) may be subject to selection bias when they are estimated among only a non-representative subset of the target population. Selection bias can sometimes be eliminated by conditioning on a “sufficient adjustment set” of covariates, even for some forms of missingness not at random (MNAR). Without requiring full specification of the causal structure, we consider sufficient adjustment sets to allow nonparametric identification of conditional ATEs in the target population. Covariates in the sufficient set may be collected among only the selected sample. We establish that if a sufficient set exists, then the set consisting of common causes of the outcome and selection, excluding the exposure and its descendants, also suffices. We establish simple graphical criteria for when a sufficient set will not exist, which could help indicate whether this is plausible for a given study. Simulations considering selection due to missing data indicated that sufficiently-adjusted complete-case analysis (CCA) can considerably outperform multiple imputation under MNAR and, if the sample size is not large, sometimes even under missingness at random. Analogous to the common-cause principle for confounding, these sufficiency results clarify when and how selection bias can be eliminated through covariate adjustment.

Psychologists are often interested in the effect of an internal state, such as ego depletion, that cannot be directly assigned in an experiment. Instead, they assign participants to a manipulation intended to produce this state and use manipulation checks to assess the manipulation’s effectiveness. In this article, I discuss statistical analyses for experiments in which researchers are primarily interested in the average treatment effect (ATE) of the target internal state rather than that of the manipulation. Often, researchers estimate the association of the manipulation itself with the dependent variable, but this intention-to-treat (ITT) estimator is typically biased for the ATE of the target state, and the bias could be either toward the null (conservative) or away from the null. I discuss the fairly stringent assumptions under which this estimator is conservative. Given this, I argue against the status-quo practice of interpreting the ITT estimate as the effect of the target state without any explicit discussion of whether these assumptions hold. Under a somewhat weaker version of the same assumptions, one can alternatively use instrumental-variables (IVs) analysis to directly estimate the effect of the target state. IVs analysis complements ITT analysis by directly addressing the central question of interest. As a running example, I consider a multisite replication study on the ego-depletion effect, in which the manipulation’s partial effectiveness led to criticism and several reanalyses that arrived at varying conclusions. I use IVs analysis to directly account for the manipulation’s partial effectiveness; this corroborated the replication authors’ reported null results.

Bryant et al.'s longitudinal study investigated causes of meat and animal product reduction. While their study design had important strengths, the analysis had significant problems that preclude interpreting the estimates as causal effects or even as meaningful associations. Our reanalysis of this study resolves the statistical issues by using standard causal inference methods for longitudinal studies. Our findings sometimes corroborated those of Bryant et al., but at other times diverged. In contrast to Bryant et al.'s findings, our analyses indicate that increased motivation to reduce meat consumption is associated with lower actual consumption of animal products. This result diverges from Bryant et al.'s surprising finding that motivation to decrease consumption was associated with increased rather than decreased animal product consumption. Additionally, our results suggest that consuming plant-based alternatives and perhaps also handling raw meat were associated with changes in plant-based dietary behavior and ideation. Several other findings corroborated those in the original analysis. We commend Bryant et al.'s study design and appreciate their exceptional support for our performing this reanalysis.

Here we investigate the impacts of media advocating plant-based diets. Search volume for popular films explains the majority of variance in searches for plant-based food, but is not associated with consumption. For three documentaries, we estimated that a standard deviation increase in searches for each film increases searches for plant-based food by up to 43% in the following week. Our findings can inform approaches for raising awareness of sustainable diets.