Bayesian Model Averaging: A Systematic Review and Conceptual Classification

Global age-sex-specific mortality, life expectancy, and population estimates in 204 countries and territories and 811 subnational locations, 1950-2021, and the impact of the COVID-19 pandemic: a comprehensive demographic analysis for the Global Burden of Disease Study 2021

BACKGROUND

Estimates of demographic metrics are crucial to assess levels and trends of population health outcomes. The profound impact of the COVID-19 pandemic on populations worldwide has underscored the need for timely estimates to understand this unprecedented event within the context of long-term population health trends. The Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) 2021 provides new demographic estimates for 204 countries and territories and 811 additional subnational locations from 1950 to 2021, with a particular emphasis on changes in mortality and life expectancy that occurred during the 2020-21 COVID-19 pandemic period.

METHODS

22 223 data sources from vital registration, sample registration, surveys, censuses, and other sources were used to estimate mortality, with a subset of these sources used exclusively to estimate excess mortality due to the COVID-19 pandemic. 2026 data sources were used for population estimation. Additional sources were used to estimate migration; the effects of the HIV epidemic; and demographic discontinuities due to conflicts, famines, natural disasters, and pandemics, which are used as inputs for estimating mortality and population. Spatiotemporal Gaussian process regression (ST-GPR) was used to generate under-5 mortality rates, which synthesised 30 763 location-years of vital registration and sample registration data, 1365 surveys and censuses, and 80 other sources. ST-GPR was also used to estimate adult mortality (between ages 15 and 59 years) based on information from 31 642 location-years of vital registration and sample registration data, 355 surveys and censuses, and 24 other sources. Estimates of child and adult mortality rates were then used to generate life tables with a relational model life table system. For countries with large HIV epidemics, life tables were adjusted using independent estimates of HIV-specific mortality generated via an epidemiological analysis of HIV prevalence surveys, antenatal clinic serosurveillance, and other data sources. Excess mortality due to the COVID-19 pandemic in 2020 and 2021 was determined by subtracting observed all-cause mortality (adjusted for late registration and mortality anomalies) from the mortality expected in the absence of the pandemic. Expected mortality was calculated based on historical trends using an ensemble of models. In location-years where all-cause mortality data were unavailable, we estimated excess mortality rates using a regression model with covariates pertaining to the pandemic. Population size was computed using a Bayesian hierarchical cohort component model. Life expectancy was calculated using age-specific mortality rates and standard demographic methods. Uncertainty intervals (UIs) were calculated for every metric using the 25th and 975th ordered values from a 1000-draw posterior distribution.

FINDINGS

Global all-cause mortality followed two distinct patterns over the study period: age-standardised mortality rates declined between 1950 and 2019 (a 62·8% [95% UI 60·5-65·1] decline), and increased during the COVID-19 pandemic period (2020-21; 5·1% [0·9-9·6] increase). In contrast with the overall reverse in mortality trends during the pandemic period, child mortality continued to decline, with 4·66 million (3·98-5·50) global deaths in children younger than 5 years in 2021 compared with 5·21 million (4·50-6·01) in 2019. An estimated 131 million (126-137) people died globally from all causes in 2020 and 2021 combined, of which 15·9 million (14·7-17·2) were due to the COVID-19 pandemic (measured by excess mortality, which includes deaths directly due to SARS-CoV-2 infection and those indirectly due to other social, economic, or behavioural changes associated with the pandemic). Excess mortality rates exceeded 150 deaths per 100 000 population during at least one year of the pandemic in 80 countries and territories, whereas 20 nations had a negative excess mortality rate in 2020 or 2021, indicating that all-cause mortality in these countries was lower during the pandemic than expected based on historical trends. Between 1950 and 2021, global life expectancy at birth increased by 22·7 years (20·8-24·8), from 49·0 years (46·7-51·3) to 71·7 years (70·9-72·5). Global life expectancy at birth declined by 1·6 years (1·0-2·2) between 2019 and 2021, reversing historical trends. An increase in life expectancy was only observed in 32 (15·7%) of 204 countries and territories between 2019 and 2021. The global population reached 7·89 billion (7·67-8·13) people in 2021, by which time 56 of 204 countries and territories had peaked and subsequently populations have declined. The largest proportion of population growth between 2020 and 2021 was in sub-Saharan Africa (39·5% [28·4-52·7]) and south Asia (26·3% [9·0-44·7]). From 2000 to 2021, the ratio of the population aged 65 years and older to the population aged younger than 15 years increased in 188 (92·2%) of 204 nations.

INTERPRETATION

Global adult mortality rates markedly increased during the COVID-19 pandemic in 2020 and 2021, reversing past decreasing trends, while child mortality rates continued to decline, albeit more slowly than in earlier years. Although COVID-19 had a substantial impact on many demographic indicators during the first 2 years of the pandemic, overall global health progress over the 72 years evaluated has been profound, with considerable improvements in mortality and life expectancy. Additionally, we observed a deceleration of global population growth since 2017, despite steady or increasing growth in lower-income countries, combined with a continued global shift of population age structures towards older ages. These demographic changes will likely present future challenges to health systems, economies, and societies. The comprehensive demographic estimates reported here will enable researchers, policy makers, health practitioners, and other key stakeholders to better understand and address the profound changes that have occurred in the global health landscape following the first 2 years of the COVID-19 pandemic, and longer-term trends beyond the pandemic.

FUNDING

Bill & Melinda Gates Foundation.

Footprint of publication selection bias on meta-analyses in medicine, environmental sciences, psychology, and economics

Publication selection bias undermines the systematic accumulation of evidence. To assess the extent of this problem, we survey over 68,000 meta-analyses containing over 700,000 effect size estimates from medicine (67,386/597,699), environmental sciences (199/12,707), psychology (605/23,563), and economics (327/91,421). Our results indicate that meta-analyses in economics are the most severely contaminated by publication selection bias, closely followed by meta-analyses in environmental sciences and psychology, whereas meta-analyses in medicine are contaminated the least. After adjusting for publication selection bias, the median probability of the presence of an effect decreased from 99.9% to 29.7% in economics, from 98.9% to 55.7% in psychology, from 99.8% to 70.7% in environmental sciences, and from 38.0% to 29.7% in medicine. The median absolute effect sizes (in terms of standardized mean differences) decreased from d = 0.20 to d = 0.07 in economics, from d = 0.37 to d = 0.26 in psychology, from d = 0.62 to d = 0.43 in environmental sciences, and from d = 0.24 to d = 0.13 in medicine.

Rethinking 13C-metabolic flux analysis - The Bayesian way of flux inference

Metabolic reaction rates (fluxes) play a crucial role in comprehending cellular phenotypes and are essential in areas such as metabolic engineering, biotechnology, and biomedical research. The state-of-the-art technique for estimating fluxes is metabolic flux analysis using isotopic labelling (13C-MFA), which uses a dataset-model combination to determine the fluxes. Bayesian statistical methods are gaining popularity in the field of life sciences, but the use of 13C-MFA is still dominated by conventional best-fit approaches. The slow take-up of Bayesian approaches is, at least partly, due to the unfamiliarity of Bayesian methods to metabolic engineering researchers. To address this unfamiliarity, we here outline similarities and differences between the two approaches and highlight particular advantages of the Bayesian way of flux analysis. With a real-life example, re-analysing a moderately informative labelling dataset of E. coli, we identify situations in which Bayesian methods are advantageous and more informative, pointing to potential pitfalls of current 13C-MFA evaluation approaches. We propose the use of Bayesian model averaging (BMA) for flux inference as a means of overcoming the problem of model uncertainty through its tendency to assign low probabilities to both, models that are unsupported by data, and models that are overly complex. In this capacity, BMA resembles a tempered Ockham's razor. With the tempered razor as a guide, BMA-based 13C-MFA alleviates the problem of model selection uncertainty and is thereby capable of becoming a game changer for metabolic engineering by uncovering new insights and inspiring novel approaches.

Computational Synaptic Modeling of Pitch and Duration Mismatch Negativity in First-Episode Psychosis Reveals Selective Dysfunction of the N-Methyl-D-Aspartate Receptor

Mismatch negativity (MMN) to pitch (pMMN) and to duration (dMMN) deviant stimuli is significantly more attenuated in long-term psychotic illness compared to first-episode psychosis (FEP). It was recently shown that source-modeling of magnetically recorded MMN increases the detection of left auditory cortex MMN deficits in FEP, and that computational circuit modeling of electrically recorded MMN also reveals left-hemisphere auditory cortex abnormalities. Computational modeling using dynamic causal modeling (DCM) can also be used to infer synaptic activity from EEG-based scalp recordings. We measured pMMN and dMMN with EEG from 26 FEP and 26 matched healthy controls (HCs) and used a DCM conductance-based neural mass model including α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid, N-methyl-D-Aspartate (NMDA), and Gamma-aminobutyric acid receptors to identify any changes in effective connectivity and receptor rate constants in FEP. We modeled MMN sources in bilateral A1, superior temporal gyrus, and inferior frontal gyrus (IFG). No model parameters distinguished groups for pMMN. For dMMN, reduced NMDA receptor activity in right IFG in FEP was detected. This finding is in line with literature of prefrontal NMDA receptor hypofunction in chronic schizophrenia and suggests impaired NMDA-induced synaptic plasticity may be present at psychosis onset where scalp dMMN is only moderately reduced. To the best of our knowledge, this is the first report of impaired NMDA receptor activity in FEP found through computational modeling of dMMN and shows the potential of DCM to non-invasively reveal synaptic-level abnormalities that underly subtle functional auditory processing deficits in early psychosis.

Combining physical-based model and machine learning to forecast chlorophyll-a concentration in freshwater lakes

Increasing algal blooms in freshwater lakes have become a serious challenge facing the world. Short-term forecast of chlorophyll-a concentration (Chla) is essential for providing early warnings and taking action to mitigate the risks of algal blooms in freshwater lakes. At present, a variety of data-driven models and physical-based models have been developed for Chla forecast, yet how to effectively combine multiple models for improving the forecast accuracy remains largely unknown. Here we developed an effective model by combining a physical-based model and machine learning algorithms (long short-term memory, LSTM; random forest, RF; support vector machine, SVM) to forecast the Chla in a freshwater lake, and a Bayesian model averaging (BMA) ensemble forecasting method was further proposed to improve the accuracy and reliability of the forecast results. We found that, with the increase of time steps of advance forecast from 1-day to 7-day, the forecast accuracy as measured by R2 of the machine learning algorithms is decreased from 0.95 to 0.68. The combination of physical-based modeling with LSTM had great capability in short-term forecast of Chla, owing to the fact that the physical-based model can provide high-frequency Chla data and LSTM is skilled at forecasting in the sequence. This is also evidenced by the weights in the BMA method. The proposed BMA short-term ensemble forecasting results had the robust performance when compared to each individual machine learning forecast model for the 7-day advance forecast, with the largest R2 (0.834) and the smallest RMSE (0.267 μg/L). In particular, the uncertainty of a single machine learning model can be effectively reduced by the BMA method.

Geographic uncertainties in external exposome studies: A multi-scale approach to reduce exposure misclassification

BACKGROUND

Many studies on environment-health associations have emphasized that the selected buffer size (i.e., the scale of the geographic context when exposures are assigned at people's address location) may affect estimated effect sizes. However, there is limited methodological progress in addressing these buffer size-related uncertainties.

AIM

We aimed to 1) develop a statistical multi-scale approach to address buffer-related scale effects in cohort studies, and 2) investigate how environment-health associations differ between our multi-scale approach and ad hoc selected buffer sizes.

METHODS

We used lacunarity analyses to determine the largest meaningful buffer size for multiple high-resolution exposure surfaces (i.e., fine particulate matter [PM2.5], noise, and the normalized difference vegetation index [NDVI]). Exposures were linked to 7.7 million Dutch adults at their home addresses. We assigned exposure estimates based on buffers with fine-grained distance increments until the lacunarity-based upper limit was reached. Bayesian Cox model averaging addressed geographic uncertainties in the estimated exposure effect sizes within the exposure-specific upper buffer limits on mortality. Z-tests assessed statistical differences between averaged effect sizes and those obtained through pre-selected 100, 300, 1200, and 1500 m buffers.

RESULTS

The estimated lacunarity curves suggested exposure-specific upper buffer size limits; the largest was for NDVI (960 m), followed by noise (910 m) and PM2.5 (450 m). We recorded 845,229 deaths over eight years of follow-up. Our multi-scale approach indicated that higher values of NDVI were health-protectively associated with mortality risk (hazard ratio [HR]: 0.917, 95 % confidence interval [CI]: 0.886-0.948). Increased noise exposure was associated with an increased risk of mortality (HR: 1.003, 95 % CI: 1.002-1.003), while PM2.5 showed null associations (HR:0.998, 95 % CI: 0.997-1.000). Effect sizes of NDVI and noise differed significantly across the averaged and prespecified buffers (p < 0.05).

CONCLUSIONS

Geographic uncertainties in residential-based exposure assessments may obscure environment-health associations or risk spurious ones. Our multi-scale approach produced more consistent effect estimates and mitigated contextual uncertainties.

Local Bayesian Dirichlet mixing of imperfect models

To improve the predictability of complex computational models in the experimentally-unknown domains, we propose a Bayesian statistical machine learning framework utilizing the Dirichlet distribution that combines results of several imperfect models. This framework can be viewed as an extension of Bayesian stacking. To illustrate the method, we study the ability of Bayesian model averaging and mixing techniques to mine nuclear masses. We show that the global and local mixtures of models reach excellent performance on both prediction accuracy and uncertainty quantification and are preferable to classical Bayesian model averaging. Additionally, our statistical analysis indicates that improving model predictions through mixing rather than mixing of corrected models leads to more robust extrapolations.

Multi-model sequential analysis of MRI data for microstructure prediction in heterogeneous tissue

We propose a general method for combining multiple models to predict tissue microstructure, with an exemplar using in vivo diffusion-relaxation MRI data. The proposed method obviates the need to select a single 'optimum' structure model for data analysis in heterogeneous tissues where the best model varies according to local environment. We break signal interpretation into a three-stage sequence: (1) application of multiple semi-phenomenological models to predict the physical properties of tissue water pools contributing to the observed signal; (2) from each Stage-1 semi-phenomenological model, application of a tissue microstructure model to predict the relative volumes of tissue structure components that make up each water pool; and (3) aggregation of the predictions of tissue structure, with weightings based on model likelihood and fractional volumes of the water pools from Stage-1. The multiple model approach is expected to reduce prediction variance in tissue regions where a complex model is overparameterised, and bias where a model is underparameterised. The separation of signal characterisation (Stage-1) from biological assignment (Stage-2) enables alternative biological interpretations of the observed physical properties of the system, by application of different tissue structure models. The proposed method is exemplified with human prostate diffusion-relaxation MRI data, but has potential application to a wide range of analyses where a single model may not be optimal throughout the sampled domain.

Frequentist model averaging for undirected Gaussian graphical models

Advances in information technologies have made network data increasingly frequent in a spectrum of big data applications, which is often explored by probabilistic graphical models. To precisely estimate the precision matrix, we propose an optimal model averaging estimator for Gaussian graphs. We prove that the proposed estimator is asymptotically optimal when candidate models are misspecified. The consistency and the asymptotic distribution of model averaging estimator, and the weight convergence are also studied when at least one correct model is included in the candidate set. Furthermore, numerical simulations and a real data analysis on yeast genetic data are conducted to illustrate that the proposed method is promising.

clusterBMA: Bayesian model averaging for clustering

Various methods have been developed to combine inference across multiple sets of results for unsupervised clustering, within the ensemble clustering literature. The approach of reporting results from one 'best' model out of several candidate clustering models generally ignores the uncertainty that arises from model selection, and results in inferences that are sensitive to the particular model and parameters chosen. Bayesian model averaging (BMA) is a popular approach for combining results across multiple models that offers some attractive benefits in this setting, including probabilistic interpretation of the combined cluster structure and quantification of model-based uncertainty. In this work we introduce clusterBMA, a method that enables weighted model averaging across results from multiple unsupervised clustering algorithms. We use clustering internal validation criteria to develop an approximation of the posterior model probability, used for weighting the results from each model. From a combined posterior similarity matrix representing a weighted average of the clustering solutions across models, we apply symmetric simplex matrix factorisation to calculate final probabilistic cluster allocations. In addition to outperforming other ensemble clustering methods on simulated data, clusterBMA offers unique features including probabilistic allocation to averaged clusters, combining allocation probabilities from 'hard' and 'soft' clustering algorithms, and measuring model-based uncertainty in averaged cluster allocation. This method is implemented in an accompanying R package of the same name. We use simulated datasets to explore the ability of the proposed technique to identify robust integrated clusters with varying levels of separation between subgroups, and with varying numbers of clusters between models. Benchmarking accuracy against four other ensemble methods previously demonstrated to be highly effective in the literature, clusterBMA matches or exceeds the performance of competing approaches under various conditions of dimensionality and cluster separation. clusterBMA substantially outperformed other ensemble methods for high dimensional simulated data with low cluster separation, with 1.16 to 7.12 times better performance as measured by the Adjusted Rand Index. We also explore the performance of this approach through a case study that aims to identify probabilistic clusters of individuals based on electroencephalography (EEG) data. In applied settings for clustering individuals based on health data, the features of probabilistic allocation and measurement of model-based uncertainty in averaged clusters are useful for clinical relevance and statistical communication.

Automating Model Comparison in Factor Graphs

Bayesian state and parameter estimation are automated effectively in a variety of probabilistic programming languages. The process of model comparison on the other hand, which still requires error-prone and time-consuming manual derivations, is often overlooked despite its importance. This paper efficiently automates Bayesian model averaging, selection, and combination by message passing on a Forney-style factor graph with a custom mixture node. Parameter and state inference, and model comparison can then be executed simultaneously using message passing with scale factors. This approach shortens the model design cycle and allows for the straightforward extension to hierarchical and temporal model priors to accommodate for modeling complicated time-varying processes.

Unified tumor growth mechanisms from multimodel inference and dataset integration

Mechanistic models of biological processes can explain observed phenomena and predict responses to a perturbation. A mathematical model is typically constructed using expert knowledge and informal reasoning to generate a mechanistic explanation for a given observation. Although this approach works well for simple systems with abundant data and well-established principles, quantitative biology is often faced with a dearth of both data and knowledge about a process, thus making it challenging to identify and validate all possible mechanistic hypothesis underlying a system behavior. To overcome these limitations, we introduce a Bayesian multimodel inference (Bayes-MMI) methodology, which quantifies how mechanistic hypotheses can explain a given experimental datasets, and concurrently, how each dataset informs a given model hypothesis, thus enabling hypothesis space exploration in the context of available data. We demonstrate this approach to probe standing questions about heterogeneity, lineage plasticity, and cell-cell interactions in tumor growth mechanisms of small cell lung cancer (SCLC). We integrate three datasets that each formulated different explanations for tumor growth mechanisms in SCLC, apply Bayes-MMI and find that the data supports model predictions for tumor evolution promoted by high lineage plasticity, rather than through expanding rare stem-like populations. In addition, the models predict that in the presence of cells associated with the SCLC-N or SCLC-A2 subtypes, the transition from the SCLC-A subtype to the SCLC-Y subtype through an intermediate is decelerated. Together, these predictions provide a testable hypothesis for observed juxtaposed results in SCLC growth and a mechanistic interpretation for tumor treatment resistance.

General-Purpose Methods for Simulating Survival Data for Expected Value of Sample Information Calculations

BACKGROUND

Expected value of sample information (EVSI) quantifies the expected value to a decision maker of reducing uncertainty by collecting additional data. EVSI calculations require simulating plausible data sets, typically achieved by evaluating quantile functions at random uniform numbers using standard inverse transform sampling (ITS). This is straightforward when closed-form expressions for the quantile function are available, such as for standard parametric survival models, but these are often unavailable when assuming treatment effect waning and for flexible survival models. In these circumstances, the standard ITS method could be implemented by numerically evaluating the quantile functions at each iteration in a probabilistic analysis, but this greatly increases the computational burden. Thus, our study aims to develop general-purpose methods that standardize and reduce the computational burden of the EVSI data-simulation step for survival data.

METHODS

We developed a discrete sampling method and an interpolated ITS method for simulating survival data from a probabilistic sample of survival probabilities over discrete time units. We compared the general-purpose and standard ITS methods using an illustrative partitioned survival model with and without adjustment for treatment effect waning.

RESULTS

The discrete sampling and interpolated ITS methods agree closely with the standard ITS method, with the added benefit of a greatly reduced computational cost in the scenario with adjustment for treatment effect waning.

CONCLUSIONS

We present general-purpose methods for simulating survival data from a probabilistic sample of survival probabilities that greatly reduce the computational burden of the EVSI data-simulation step when we assume treatment effect waning or use flexible survival models. The implementation of our data-simulation methods is identical across all possible survival models and can easily be automated from standard probabilistic decision analyses.

HIGHLIGHTS

Expected value of sample information (EVSI) quantifies the expected value to a decision maker of reducing uncertainty through a given data collection exercise, such as a randomized clinical trial. In this article, we address the problem of computing EVSI when we assume treatment effect waning or use flexible survival models, by developing general-purpose methods that standardize and reduce the computational burden of the EVSI data-generation step for survival data.We developed 2 methods for simulating survival data from a probabilistic sample of survival probabilities over discrete time units, a discrete sampling method and an interpolated inverse transform sampling method, which can be combined with a recently proposed nonparametric EVSI method to accurately estimate EVSI for collecting survival data.Our general-purpose data-simulation methods greatly reduce the computational burden of the EVSI data-simulation step when we assume treatment effect waning or use flexible survival models. The implementation of our data-simulation methods is identical across all possible survival models and can therefore easily be automated from standard probabilistic decision analyses.

Identifying the Best Approximating Model in Bayesian Phylogenetics: Bayes Factors, Cross-Validation or wAIC?

There is still no consensus as to how to select models in Bayesian phylogenetics, and more generally in applied Bayesian statistics. Bayes factors are often presented as the method of choice, yet other approaches have been proposed, such as cross-validation or information criteria. Each of these paradigms raises specific computational challenges, but they also differ in their statistical meaning, being motivated by different objectives: either testing hypotheses or finding the best-approximating model. These alternative goals entail different compromises, and as a result, Bayes factors, cross-validation, and information criteria may be valid for addressing different questions. Here, the question of Bayesian model selection is revisited, with a focus on the problem of finding the best-approximating model. Several model selection approaches were re-implemented, numerically assessed and compared: Bayes factors, cross-validation (CV), in its different forms (k-fold or leave-one-out), and the widely applicable information criterion (wAIC), which is asymptotically equivalent to leave-one-out cross-validation (LOO-CV). Using a combination of analytical results and empirical and simulation analyses, it is shown that Bayes factors are unduly conservative. In contrast, CV represents a more adequate formalism for selecting the model returning the best approximation of the data-generating process and the most accurate estimates of the parameters of interest. Among alternative CV schemes, LOO-CV and its asymptotic equivalent represented by the wAIC, stand out as the best choices, conceptually and computationally, given that both can be simultaneously computed based on standard Markov chain Monte Carlo runs under the posterior distribution. [Bayes factor; cross-validation; marginal likelihood; model comparison; wAIC.].

Analysis of a Binary Outcome Dichotomized from an Underlying Continuous Variable in Clinical Trials

Binary-valued outcome is often seen in many clinical trials across therapeutic areas. It is not uncommon that such binary endpoints are derived from a continuous variable. For example, in diabetes clinical trials, the proportion of patients with HbA1c< 7% is often investigated as one of the key objectives, where HbA1c is a continuous-valued variable reflecting the averaged blood glucose value from the previous three months. Most of the time, if not all, the mean of those binary endpoints were estimated directly through the binary variable defined by the corresponding cutoff. Alternatively, by the nature of the derivation, that quantity could also be estimated by leveraging the density of the underlying continuous variable and computing the area under the density curve up to a threshold. This paper provides a few methods in relation to density estimation. Extensive simulation studies were conducted based on real clinical trial data to compare these estimation approaches against the direct estimation of the proportions. Simulation results showed that the density estimation approaches in general benefited from a smaller mean squared error in early phase studies where the sample size is limited. The density estimation approach is certainly expected to introduce bias, however, a favorable bias-variance trade-off may make these approaches attractive in early phase studies.

Forming intracluster gas in a galaxy protocluster at a redshift of 2.16

Galaxy clusters are the most massive gravitationally bound structures in the Universe, comprising thousands of galaxies and pervaded by a diffuse, hot intracluster medium (ICM) that dominates the baryonic content of these systems. The formation and evolution of the ICM across cosmic time1 is thought to be driven by the continuous accretion of matter from the large-scale filamentary surroundings and energetic merger events with other clusters or groups. Until now, however, direct observations of the intracluster gas have been limited only to mature clusters in the later three-quarters of the history of the Universe, and we have been lacking a direct view of the hot, thermalized cluster atmosphere at the epoch when the first massive clusters formed. Here we report the detection (about 6σ) of the thermal Sunyaev–Zeldovich (SZ) effect2 in the direction of a protocluster. In fact, the SZ signal reveals the ICM thermal energy in a way that is insensitive to cosmological dimming, making it ideal for tracing the thermal history of cosmic structures3. This result indicates the presence of a nascent ICM within the Spiderweb protocluster at redshift z = 2.156, around 10 billion years ago. The amplitude and morphology of the detected signal show that the SZ effect from the protocluster is lower than expected from dynamical considerations and comparable with that of lower-redshift group-scale systems, consistent with expectations for a dynamically active progenitor of a local galaxy cluster.

Predictors of response and remission in patients with treatment-resistant depression: A post hoc pooled analysis of two acute trials of esketamine nasal spray

This exploratory post hoc analysis of two pooled 4-week, phase 3, double-blind, placebo- and active-controlled studies that compared esketamine nasal spray plus a newly initiated oral antidepressant (ESK+AD; n = 310) with a newly initiated oral AD plus placebo nasal spray (AD+PBO; n = 208) in patients with treatment-resistant depression (TRD) examined baseline patient demographic and psychiatric characteristics as potential predictors of response (≥50% reduction from baseline in Montgomery-Åsberg Depression Rating Scale [MADRS] total score) and remission (MADRS total score ≤12) at day 28. Overall, younger age, any employment, fewer failed ADs in the current depressive episode, and reduction in Clinical Global Impression-Severity (CGI-S) score at day 8 were significant positive predictors of response and remission at day 28. Treatment assignment was an important predictor of both response and remission. Patients treated with ESK+AD had 68% and 55% increased odds of achieving response and remission, respectively, versus those treated with AD+PBO. In the ESK+AD group, attainment of response and remission was more likely in patients who were employed, without significant anxiety at baseline, and who experienced a reduction in CGI-S score at day 8. Identification of predictors of response and remission may facilitate identification of those patients with TRD most likely to benefit from ESK+AD. Trial Registration: ClinicalTrials.gov: NCT02417064 (clinicaltrials.gov/ct2/show/NCT02417064) and NCT02418585 (clinicaltrials.gov/ct2/show/NCT02418585).

Dynamic predictions from longitudinal CD4 count measures and time to death of HIV/AIDS patients using a Bayesian joint model

A Bayesian joint modeling approach to dynamic prediction of HIV progression and mortality allows individualized predictions to be made for HIV patients, based on monitoring of their CD4 counts. This study aims to provide predictions of patient-specific trajectories of HIV disease progression and survival. Longitudinal data on 254 HIV/AIDS patients who received ART between 2009 and 2014, and who had at least one CD4 count observed, were employed in a Bayesian joint model of disease progression. Different forms of association structure that relate the longitudinal CD4 biomarker and time to death were assessed; and predictions were averaged over the different models using Bayesian model averaging. The individual follow-up times ranged from 1 to 120 months, with a median of 22 months and IQR 7-39 months. The estimates of the association structure parameters from two of the three models considered indicated that the HIV mortality hazard at any time point is associated with the rate of change in the underlying value of the CD4 count. Model averaging the dynamic predictions resulted in only one of the hypothesized association structures having non-zero weight in almost all time points for each individual, with the exception of twelve patients, for whom other association structures were preferred at a few time points. The predictions were found to be different when we averaged them over models than when we derived them from the highest posterior weight model alone. The model with highest posterior weight for almost all time points for each individual gave an estimate of the association parameter of -0.02 implying that for a unit increase in the CD4 count, the hazard of HIV mortality decreases by a factor (hazard ratio) of 0.98. Functional status and alcohol intake are important contributing factors that affect the mean square root of CD4 measurements.

Change plane model averaging for subgroup identification

Central to personalized medicine and tailored therapies is discovering the subpopulations that account for treatment effect heterogeneity and are likely to benefit more from given interventions. In this article, we introduce a change plane model averaging method to identify subgroups characterized by linear combinations of predictive variables and multiple cut-offs. We first fit a sequence of statistical models, each incorporating the thresholding effect of one particular covariate. The estimation of submodels is accomplished through an iterative integration of a change point detection method and numerical optimization algorithms. A frequentist model averaging approach is then employed to linearly combine the submodels with optimal weights. Our approach can deal with high-dimensional settings involving enormous potential grouping variables by adopting the sparsity-inducing penalties. Simulation studies are conducted to investigate the prediction and subgrouping performance of the proposed method, with a comparison to various competing subgroup detection methods. Our method is applied to a dataset from a warfarin pharmacogenetics study, producing some new findings.

Using Social Media for Social Motives Moderates the Relationship between Post-Traumatic Symptoms during a COVID-19-Related Lockdown and Improvement of Distress after Lockdown

Previous research reported inconsistent results on the relationship between social media (SM) use and psychological well-being, suggesting the importance of assessing possible moderators, e.g., motives for SM use. However, no longitudinal study has yet investigated whether, among people who use SM, specific motives for using SM may represent protective/risk factors for the development of psychological distress, especially after a stressful event. Our longitudinal study aimed at assessing the moderating role of motives for using SM (i.e., coping, conformity, enhancement, social motives) in the relationship between COVID-19 pandemic-related post-traumatic stress symptoms during the lockdown and changes in general distress after lockdown. At Time 1 (during the first lockdown in Italy), 660 participants responded to an online survey, reporting their post-traumatic symptoms, motives for using SM, and general distress (i.e., anxiety, depression, and stress symptoms). At Time 2 (three months later, following lockdown), 117 participants volunteered to continue with the follow-up survey assessing general distress symptoms again. Results showed that among those who had experienced more severe post-traumatic symptoms at Time 1, using SM for social motives was associated with more improvement of general distress symptoms. No evidence was found of moderating effects of other motives for SM use. The findings suggest that social connections may have helped to cope with stress during forced confinement, and that SM use may be beneficial for mental health when motivated by maintaining social interactions.

Robust Bayesian meta-analysis: Model-averaging across complementary publication bias adjustment methods

Publication bias is a ubiquitous threat to the validity of meta-analysis and the accumulation of scientific evidence. In order to estimate and counteract the impact of publication bias, multiple methods have been developed; however, recent simulation studies have shown the methods' performance to depend on the true data generating process, and no method consistently outperforms the others across a wide range of conditions. Unfortunately, when different methods lead to contradicting conclusions, researchers can choose those methods that lead to a desired outcome. To avoid the condition-dependent, all-or-none choice between competing methods and conflicting results, we extend robust Bayesian meta-analysis and model-average across two prominent approaches of adjusting for publication bias: (1) selection models of p-values and (2) models adjusting for small-study effects. The resulting model ensemble weights the estimates and the evidence for the absence/presence of the effect from the competing approaches with the support they receive from the data. Applications, simulations, and comparisons to preregistered, multi-lab replications demonstrate the benefits of Bayesian model-averaging of complementary publication bias adjustment methods.

Recovering crossed random effects in mixed-effects models using model averaging

Random effects contain crucial information to understand the variability of the processes under study in mixed-effects models with crossed random effects (MEMs-CR). Given that model selection makes all-or-nothing decisions regarding to the inclusion of model parameters, we evaluated if model averaging could deal with model uncertainty to recover random effects of MEMs-CR. Specifically, we analyzed the bias and the root mean squared error (RMSE) of the estimations of the variances of random effects using model averaging with Akaike weights and Bayesian model averaging with BIC posterior probabilities, comparing them with two alternative analytical strategies as benchmarks: AIC and BIC model selection, and fitting a full random structure. A simulation study was conducted manipulating sample sizes for subjects and items, and the variance of random effects. Results showed that model averaging, especially Akaike weights, can adequately recover random variances, given a minimum sample size in the modeled clusters. Thus, we endorse using model averaging to deal with model uncertainty in MEMs-CR. An empirical illustration is provided to ease the usability of model averaging.

Informed Bayesian survival analysis

Background

We provide an overview of Bayesian estimation, hypothesis testing, and model-averaging and illustrate how they benefit parametric survival analysis. We contrast the Bayesian framework to the currently dominant frequentist approach and highlight advantages, such as seamless incorporation of historical data, continuous monitoring of evidence, and incorporating uncertainty about the true data generating process.

Methods

We illustrate the application of the outlined Bayesian approaches on an example data set, retrospective re-analyzing a colon cancer trial. We assess the performance of Bayesian parametric survival analysis and maximum likelihood survival models with AIC/BIC model selection in fixed-n and sequential designs with a simulation study.

Results

In the retrospective re-analysis of the example data set, the Bayesian framework provided evidence for the absence of a positive treatment effect of adding Cetuximab to FOLFOX6 regimen on disease-free survival in patients with resected stage III colon cancer. Furthermore, the Bayesian sequential analysis would have terminated the trial 10.3 months earlier than the standard frequentist analysis. In a simulation study with sequential designs, the Bayesian framework on average reached a decision in almost half the time required by the frequentist counterparts, while maintaining the same power, and an appropriate false-positive rate. Under model misspecification, the Bayesian framework resulted in higher false-negative rate compared to the frequentist counterparts, which resulted in a higher proportion of undecided trials. In fixed-n designs, the Bayesian framework showed slightly higher power, slightly elevated error rates, and lower bias and RMSE when estimating treatment effects in small samples. We found no noticeable differences for survival predictions. We have made the analytic approach readily available to other researchers in the RoBSA R package.

Conclusions

The outlined Bayesian framework provides several benefits when applied to parametric survival analyses. It uses data more efficiently, is capable of considerably shortening the length of clinical trials, and provides a richer set of inferences.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12874-022-01676-9.

Model averaging for sparse seemingly unrelated regression using Bayesian networks among the errors

Multivariate Bayesian linear regression (MBLR) is a popular statistical tool with many applications in a variety of scientific fields. However, a shortcoming is potential model over-complexity, as the model assumes that all responses depend on the same covariates and that all errors are mutually pairwise correlated. The class of Bayesian seemingly unrelated regression (SUR) models generalizes the class of MBLR models by allowing for response-specific covariate sets. In a recent work it has been proposed to employ Gaussian graphical models for learning sparse SUR (SSUR) models with conditional independencies among the errors. The proposed SSUR model infers undirected edges among the errors, and the proposed Reversible Jump Markov Chain Monte Carlo (RJMCMC) inference algorithm relies on approximations of the marginal likelihoods. In this paper, we propose a new refined SSUR model that replaces the undirected graphs (Gaussian graphical models) by directed acyclic graphs (Gaussian Bayesian networks). Unlike the earlier proposed model, our new model is therefore able to learn some directed edges among the errors. And we derive a RJMCMC algorithm that does not require approximations of the marginal likelihoods. In particular, we present an algorithm for sampling covariance matrices that are coherent with a given directed acyclic graph. The proposed RJMCMC algorithm allows for exact Bayesian model averaging across both: the response-specific covariate sets and the directed acyclic graphs.

Effect of meteorological conditions on leisure walking: a time series analysis and the application of outdoor thermal comfort indexes

Leisure walking is affected by meteorological conditions. However, it is still not clear what scales of meteorological conditions and thermal status affect the number of people who choose to leisure walk. Using a time series regression, this study examines the heat-leisure walking relationship by analyzing the effect of the seasons, weather, microclimate, and outdoor thermal comfort on walking count. Eight thermal indexes were selected to estimate the pedestrians' thermal comfort, and their predictive capacities in walking count were evaluated. Particular consideration was given to identifying heat thresholds of walking that determined the tolerance range of pedestrian heat stress. Four years of hourly daytime walking counts and publicly available ASOS meteorological data at Seoul-lo 7017, a pedestrian bridge in Seoul, were used for the analysis. Our findings indicate that walking count is correlated with seasonal climatic variations, with the highest number of pedestrians observed in fall and the lowest in summer. Moreover, air temperature played a vital role, showing that a 5.0 °C rise in temperature was associated with a 1.34% rise in the square root of the walking count. Its impact becomes greater when combined with intense solar radiation and higher absolute humidity. The heat threshold for walking was between 23.8 °C and 26.2 °C. Empirical model indexes showed the highest predictive capacity in walking count at approximately 30.0%, which was followed by rational model indexes at 28.0%.

Correlates of the country differences in the infection and mortality rates during the first wave of the COVID-19 pandemic: evidence from Bayesian model averaging

The COVID-19 pandemic resulted in great discrepancies in both infection and mortality rates between countries. Besides the biological and epidemiological factors, a multitude of social and economic criteria also influenced the extent to which these discrepancies appeared. Consequently, there is an active debate regarding the critical socio-economic and health factors that correlate with the infection and mortality rates outcome of the pandemic. Here, we leverage Bayesian model averaging techniques and country level data to investigate whether 28 variables, which describe a diverse set of health and socio-economic characteristics, correlate with the final number of infections and deaths during the first wave of the coronavirus pandemic. We show that only a few variables are able to robustly correlate with these outcomes. To understand the relationship between the potential correlates in explaining the infection and death rates, we create a Jointness Space. Using this space, we conclude that the extent to which each variable is able to provide a credible explanation for the COVID-19 infections/mortality outcome varies between countries because of their heterogeneous features.

Comparing methods for statistical inference with model uncertainty

Probability models are used for many statistical tasks, notably parameter estimation, interval estimation, inference about model parameters, point prediction, and interval prediction. Thus, choosing a statistical model and accounting for uncertainty about this choice are important parts of the scientific process. Here we focus on one such choice, that of variables to include in a linear regression model. Many methods have been proposed, including Bayesian and penalized likelihood methods, and it is unclear which one to use. We compared 21 of the most popular methods by carrying out an extensive set of simulation studies based closely on real datasets that span a range of situations encountered in practical data analysis. Three adaptive Bayesian model averaging (BMA) methods performed best across all statistical tasks. These used adaptive versions of Zellner’s g-prior for the parameters, where the prior variance parameter g is a function of sample size or is estimated from the data. We found that for BMA methods implemented with Markov chain Monte Carlo, 10,000 iterations were enough. Computationally, we found two of the three best methods (BMA with g=√n and empirical Bayes-local) to be competitive with the least absolute shrinkage and selection operator (LASSO), which is often preferred as a variable selection technique because of its computational efficiency. BMA performed better than Bayesian model selection (in which just one model is selected).

A flexible Bayesian framework for individualized inference via adaptive borrowing

The explosion in high-resolution data capture technologies in health has increased interest in making inferences about individual-level parameters. While technology may provide substantial data on a single individual, how best to use multisource population data to improve individualized inference remains an open research question. One possible approach, the multisource exchangeability model (MEM), is a Bayesian method for integrating data from supplementary sources into the analysis of a primary source. MEM was originally developed to improve inference for a single study by asymmetrically borrowing information from a set of similar previous studies and was further developed to apply a more computationally intensive symmetric borrowing in the context of basket trial; however, even for asymmetric borrowing, its computational burden grows exponentially with the number of supplementary sources, making it unsuitable for applications where hundreds or thousands of supplementary sources (i.e., individuals) could contribute to inference on a given individual. In this article, we propose the data-driven MEM (dMEM), a two-stage approach that includes both source selection and clustering to enable the inclusion of an arbitrary number of sources to contribute to individualized inference in a computationally tractable and data-efficient way. We illustrate the application of dMEM to individual-level human behavior and mental well-being data collected via smartphones, where our approach increases individual-level estimation precision by 84% compared with a standard no-borrowing method and outperforms recently proposed competing methods in 80% of individuals.

The Use of Questionable Research Practices to Survive in Academia Examined With Expert Elicitation, Prior-Data Conflicts, Bayes Factors for Replication Effects, and the Bayes Truth Serum

The popularity and use of Bayesian methods have increased across many research domains. The current article demonstrates how some less familiar Bayesian methods can be used. Specifically, we applied expert elicitation, testing for prior-data conflicts, the Bayesian Truth Serum, and testing for replication effects via Bayes Factors in a series of four studies investigating the use of questionable research practices (QRPs). Scientifically fraudulent or unethical research practices have caused quite a stir in academia and beyond. Improving science starts with educating Ph.D. candidates: the scholars of tomorrow. In four studies concerning 765 Ph.D. candidates, we investigate whether Ph.D. candidates can differentiate between ethical and unethical or even fraudulent research practices. We probed the Ph.D.s' willingness to publish research from such practices and tested whether this is influenced by (un)ethical behavior pressure from supervisors or peers. Furthermore, 36 academic leaders (deans, vice-deans, and heads of research) were interviewed and asked to predict what Ph.D.s would answer for different vignettes. Our study shows, and replicates, that some Ph.D. candidates are willing to publish results deriving from even blatant fraudulent behavior-data fabrication. Additionally, some academic leaders underestimated this behavior, which is alarming. Academic leaders have to keep in mind that Ph.D. candidates can be under more pressure than they realize and might be susceptible to using QRPs. As an inspiring example and to encourage others to make their Bayesian work reproducible, we published data, annotated scripts, and detailed output on the Open Science Framework (OSF).

Spatial Statistical Models: An Overview under the Bayesian Approach

Spatial documentation is exponentially increasing given the availability of Big Data in the Internet of Things, enabled by device miniaturization and data storage capacity. Bayesian spatial statistics is a useful statistical tool to determine the dependence structure and hidden patterns in space through prior knowledge and data likelihood. However, this class of modeling is not yet well explored when compared to adopting classification and regression in machine-learning models, in which the assumption of the spatiotemporal independence of the data is often made, that is an inexistent or very weak dependence. Thus, this systematic review aims to address the main models presented in the literature over the past 20 years, identifying the gaps and research opportunities. Elements such as random fields, spatial domains, prior specification, the covariance function, and numerical approximations are discussed. This work explores the two subclasses of spatial smoothing: global and local.

Improving astronaut cancer risk assessment from space radiation with an ensemble model framework

A new approach to NASA space radiation risk modeling has successfully extended the current NASA probabilistic cancer risk model to an ensemble framework able to consider sub-model parameter uncertainty as well as model-form uncertainty associated with differing theoretical or empirical formalisms. Ensemble methodologies are already widely used in weather prediction, modeling of infectious disease outbreaks, and certain terrestrial radiation protection applications to better understand how uncertainty may influence risk decision-making. Applying ensemble methodologies to space radiation risk projections offers the potential to efficiently incorporate emerging research results, allow for the incorporation of future models, improve uncertainty quantification, and reduce the impact of subjective bias. Moreover, risk forecasting across an ensemble of multiple predictive models can provide stakeholders additional information on risk acceptance if current health/medical standards cannot be met for future space exploration missions, such as human missions to Mars. In this work, ensemble risk projections implementing multiple sub-models of radiation quality, dose and dose-rate effectiveness factors, excess risk, and latency are presented. Initial consensus methods for ensemble model weights and correlations to account for individual model bias are discussed. In these analyses, the ensemble forecast compares well to results from NASA's current operational cancer risk projection model used to assess permissible mission durations for astronauts. However, a large range of projected risk values are obtained at the upper 95th confidence level where models must extrapolate beyond available biological data sets. Closer agreement is seen at the median ± one sigma due to the inherent similarities in available models. Identification of potential new models, epidemiological data, and methods for statistical correlation between predictive ensemble members are discussed. Alternate ways of communicating risk and acceptable uncertainty with respect to NASA's current permissible exposure limits are explored.

A survival mediation model with Bayesian model averaging

Determining the extent to which a patient is benefiting from cancer therapy is challenging. Criteria for quantifying the extent of "tumor response" observed within a few cycles of treatment have been established for various types of solids as well as hematologic malignancies. These measures comprise the primary endpoints of phase II trials. Regulatory approvals of new cancer therapies, however, are usually contingent upon the demonstration of superior overall survival with randomized evidence acquired with a phase III trial comparing the novel therapy to an appropriate standard of care treatment. With nearly two-thirds of phase III oncology trials failing to achieve statistically significant results, researchers continue to refine and propose new surrogate endpoints. This article presents a Bayesian framework for studying relationships among treatment, patient subgroups, tumor response, and survival. Combining classical components of a mediation analysis with Bayesian model averaging, the methodology is robust to model misspecification among various possible relationships among the observable entities. A posterior inference is demonstrated via an application to a randomized controlled phase III trial in metastatic colorectal cancer. Moreover, the article details posterior predictive distributions of survival and statistical metrics for quantifying the extent of direct and indirect, or tumor response mediated treatment effects.

The Backup Planning Scale (BUPS): A Brief, Self-Reported Measure of a Person's Tendency to Develop, Reserve, and Use Backup Plans

To mitigate uncertainty in their goal pursuits, people use backup plans, i.e., alternative means that are developed to potentially replace "Plan A." Several studies have demonstrated that backup plans can introduce unexpected costs into goal pursuits that decrease a person's motivation to continue using their "Plan A," and reduce their chances for achieving their goal. These existing studies used time-intensive experimental and/or observational approaches to assess the effects of backup planning. The present research examines the newly-developed Backup Planning Scale (BUPS) for its measurement invariance, reliability, validity, and other psychometric characteristics across three independent samples with more than 1,500 participants. Consistent with prior theorizing, we found support for a nine-item, three factor structure for the BUPS, indexing latent factors for a person's tendency to develop, reserve, and replace with (or use) backup plans. Furthermore, a novel "IRTree" based statistical technique provided evidence for the validity of the measure, as participants' responses to the BUPS were associated with their actual developing, reserving, and replacing backup planning behaviors in a logic task. We conclude that the freely-available BUPS is a simple, brief, reliable, and valid self-reported instrument for assessing backup planning behaviors across adulthood.

Quantifying Critical Success Factors (CSFs) in Management of Investment-Construction Projects: Insights from Bayesian Model Averaging

The problem with evaluating investment projects is that there are many factors that determine the degree of their successful conclusion. Consequently, there has been an active debate for years as to which critical success factors (CSFs) contribute most to the performance of construction projects. This is because the practice of empirical research is based on two steps: first, researchers choose a particular model from the space of all possible models, and second, they act as if the chosen model is the only one that fits the data and describes the phenomenon under study. Hence, there are many CSF lists that can be found in the literature, owing to the uncertainty at the model selection stage, which is usually ignored. Alternatively, model averaging accounts for this model uncertainty. In this study, the Bayesian model averaging and data from a survey of Polish construction managers were used to investigate the potential of 28 factors describing a diverse set of characteristics in explaining the performance of construction projects in Poland. Determinants of successful completion of investment projects are categorized by their level of evidential strength, which is derived from posterior inclusion probabilities (PIPs), i.e., providing strong, medium and weak evidence.

Phenomenological Constraints on the Transport Properties of QCD Matter with Data-Driven Model Averaging

Using combined data from the Relativistic Heavy Ion and Large Hadron Colliders, we constrain the shear and bulk viscosities of quark-gluon plasma (QGP) at temperatures of ∼150-350  MeV. We use Bayesian inference to translate experimental and theoretical uncertainties into probabilistic constraints for the viscosities. With Bayesian model averaging we propagate an estimate of the model uncertainty generated by the transition from hydrodynamics to hadron transport in the plasma's final evolution stage, providing the most reliable phenomenological constraints to date on the QGP viscosities.

Single-nucleotide polymorphism detecting of some candidate genes related to lipid metabolism in Booroola Merino-Afshari sheep by Bayesian model averaging

This study was designed to identify single-nucleotide polymorphisms (SNPs) of some candidate genes related to lipid metabolism and their association with carcass fat in male crossbred lambs. Hence, 96 of almost 11-month-old Booroola Merino-Afshari crossbred lambs (first-generation backcross) were used by considering their phenotypic carcass traits. Then, DNA was extracted and DNA targets were amplified using designed specific primers by PCR procedure. Identification of potential SNPs was done by a direct sequencing method for LEP, FABP4, DGAT1, GH, and TRIB3 genes using the sequencing-RFLP procedure. Then, the most probable statistical models based on additive and genotypic effects of identified SNPs in each trait were obtained by the Bayesian model averaging (BMA) approach of R software (Ver. 3.3.1) to assess the association of SNPs with traits. Detected SNPs in this study included two SNPs in exon 3 of LEP, one SNP in exon 2 of TRIB3, one SNP in intron 2 of FABP4, one SNP in 5' UTR of DGAT1, and two SNPs in 3' UTR of GH genes. For carcass weight trait, one of the identified SNP genotypes in the LEP (c.587G > A) had a higher probability in the model. Carcass weight of lambs with GA genotype was 2.46 kg heavier than GG genotype. Also, two genes of TRIB3 and GH2 had the highest probability in the models of fat tail and waste weight, respectively. Based on the results, these polymorphisms can be used in the marker-assisted selection of breeding programs and designing DNA chips for genomic selection.

Recovery Capital and Symptom Improvement in Gambling Disorder: Correlations with Spirituality and Stressful Life Events in Younger but Not Older Adults

Although age-related differences have been reported in gambling disorder, prior studies have not examined how age may influence recovery in gambling disorder. Recovery may be influenced by positive factors (e.g., spirituality and recovery capital) and negative factors (e.g., depression, anxiety, and stressful life events). The current study examined associations between these positive and negative factors and gambling disorder DSM-5 symptom improvement in younger and older adults. Younger (less than 55 years of age; n = 86) and older (55 years or older; n = 54) adults, with lifetime gambling disorder treated currently or within the past 5 years in five treatment centers in Israel were assessed using structured scales on past-year and lifetime DSM-5 gambling disorder, intrinsic spirituality, recovery capital, anxiety, depression and stressful life-events. Among younger adults, recovery capital and intrinsic spirituality were associated with gambling disorder symptom improvement. Among older adults, only recovery capital was associated with gambling disorder symptom improvement. Correlations between recovery capital and spirituality (z = 2.34, p = 0.02) and recovery capital and stressful life events (z = 2.29, p = 0.02) were stronger in younger than in older adults. Recovery capital is an important resource that should be considered across older and younger adults with gambling disorder. Spirituality and stressful life events may operate differently across age groups in gambling disorder. Future studies should investigate whether the findings may extend to other groups and the extent to which promoting recovery capital should be integrated into treatments for gambling disorder.

Assessing the appropriateness of antimicrobial therapy in patients with sepsis at a Vietnamese national hospital

Objectives

To assess the appropriateness of empirical antimicrobial therapy for sepsis and septic shock and determine factors associated with patient treatment outcomes at a Vietnamese national hospital.

Methods

A cross-sectional study was conducted on 134 patients diagnosed with sepsis and/or septic shock at Thong-Nhat Hospital, Ho Chi Minh City, Vietnam, from January 2018 to June 2018. Appropriateness of antimicrobial therapy was defined as physician adherence to antimicrobial guidelines using the Sanford Guide to Antimicrobial Therapy and the Vietnam national guidelines. Bayesian model averaging technique was used to identify the related factors associated with patient treatment outcomes.

Results

The median age of patients was 70 years. Organisms were identified in 54.5% of cases and predominated by Escherichia coli and staphylococci. Appropriate empirical antimicrobial agents were initiated in 56.6% (n = 73) of all cases. Of these patients, 31 cases (42.5%) and 61 cases (83.6%) received the antimicrobials in accordance with recommendations related to dosage and route of administration, respectively, bringing the overall rate of appropriate empirical antimicrobial therapy down to 23.3%. Patients who progressed to septic shock, received inappropriate antimicrobial therapy and required ICU admission were more likely to suffer treatment failure.

Conclusions

The study findings suggest that clinicians should appropriately adhere to antimicrobial guidelines, especially in patients with septic shock and those who require ICU care, to improve treatment outcomes.

Could Road Safety Education (RSE) Help Parents Protect Children? Examining Their Driving Crashes with Children on Board

Recent evidence suggests that driving behavior and traffic safety outcomes of parents may be influenced by the extent to which they receive information and education on road safety, as well as the fact of driving with their children on board, which may increase their risk perception. However, there are no studies specifically addressing the case of crashes suffered while driving with children. Hence, this study aimed to describe the relationship between road safety education-related variables and parents' traffic safety outcomes while driving with children on board. For this cross-sectional study, data was retrieved from a sample composed of 165 Spanish parents-all of them licensed drivers-with a mean age of 45.3 years. Through binary logistic regression (logit) analysis, it was found that factors such as gender, having received road safety education (RSE), and having been sanctioned for the performance of risky driving behavior contribute to modulating the likelihood of suffering crashes while driving with children on board. Gender differences showed a riskier status for male parents. In this study, a set of risk factors explaining the involvement in traffic crashes when driving with children as passengers was identified among parents: gender, traffic sanctions, valuation, and exposure to road safety campaigns. Also, substantial limitations in the self-reported degree of received RSE were found, especially considering that risky driving behavior and traffic crash rates with children on board still have a high prevalence among parents.

Bayesian approaches to variable selection: a comparative study from practical perspectives

In many clinical studies, researchers are interested in parsimonious models that simultaneously achieve consistent variable selection and optimal prediction. The resulting parsimonious models will facilitate meaningful biological interpretation and scientific findings. Variable selection via Bayesian inference has been receiving significant advancement in recent years. Despite its increasing popularity, there is limited practical guidance for implementing these Bayesian approaches and evaluating their comparative performance in clinical datasets. In this paper, we review several commonly used Bayesian approaches to variable selection, with emphasis on application and implementation through R software. These approaches can be roughly categorized into four classes: namely the Bayesian model selection, spike-and-slab priors, shrinkage priors, and the hybrid of both. To evaluate their variable selection performance under various scenarios, we compare these four classes of approaches using real and simulated datasets. These results provide practical guidance to researchers who are interested in applying Bayesian approaches for the purpose of variable selection.

Bayesian hierarchical evaluation of dose-response for peanut allergy in clinical trial screening

Risk-based labeling based on the minimal eliciting doses (EDs) in sensitized populations is a potential replacement for precautionary allergen labeling of food allergens. We estimated the dose-response distribution for peanut allergen using data from double-blind placebo-controlled food challenges (DBPCFCs) conducted in the US at multiple sites, testing a population believed to be similar to the general U.S. food allergic population. Our final (placebo-adjusted) dataset included 548 challenges of 481 subjects. Bayesian hierarchical analysis facilitated model fitting, and accounted for variability associated with various levels of data organization. The data are best described using a complex hierarchical structure that accounts for inter-individual variability and variability across study locations or substudies. Bayesian model averaging could simultaneously consider the fit of multiple models, but the Weibull model dominated so strongly that model averaging was not needed. The ED01 and ED05 (and 95% credible intervals) are 0.052 (0.021, 0.13) and 0.49 (0.22, 0.97) mg peanut protein, respectively. Accounting for challenges with severe reactions at the LOAEL, by using the dose prior to the LOAEL as the new LOAEL, the ED01 drops to 0.029 (0.014, 0.074) mg peanut protein. Our results could aid in establishing improved food labeling guidelines in the management of food allergies.