Bayesian predictive probabilities: a good way to monitor clinical trials

BACKGROUND

Bayesian methods, with the predictive probability (PredP), allow multiple interim analyses with interim posterior probability (PostP) computation, without the need to correct for multiple looks at the data. The objective of this paper was to illustrate the use of PredP by simulating a sequential analysis of a clinical trial.

METHODS

We used data from the Laryngobloc trial that planned to include 480 patients to demonstrate the equivalence of success between a laryngoscopy performed with the Laryngobloc® device and a control device. A crossover Bayesian design was used. The success rates of the two laryngoscopy devices were compared. Interim analyses, computed from random numbers of subjects, were simulated.

RESULTS

The PostP of equivalence rapidly reached the predefined bound of 0.95. The PredP computed with an equivalence margin of 10% reached the efficacy bound between 352 and 409 of the 480 included patients. If a frequentist analysis had been made on the basis of 217 out of 480 subjects, the study would have been prematurely stopped for equivalence. The PredP indicated that this result was nonetheless unstable and that the equivalence was, thus far, not guaranteed.

CONCLUSIONS

Based on these interim analyses, we can conclude with a sufficiently high probability that the equivalence would have been met on the primary outcome before the predetermined end of this particular trial. If a Bayesian approach using PredP had been used, it would have allowed an early termination of the trial by reducing the calculated sample size by 15-20%.

Executive Functions Are Associated with Fall Risk but not Balance in Chronic Cerebrovascular Disease

BACKGROUND

Older people's deficits in executive functions (EF) have been shown to lead to higher fall risk, postural sway, and reduced speed. Crucially, EF impairments are even more pronounced in individuals with chronic cerebrovascular disease (CVD), namely vascular cognitive impairment.

METHODS

In this retrospective cross-sectional study, we used a complete neuropsychological battery, including the Trail Making Test (TMT) and physical measures, such as the Morse fall and EQUI scales, to assess 66 individuals with chronic CVD. Linear regressions, Bayesian analyses, and model selection were performed to see the impact of EF, global cognition, and vascular parkinsonism/hemiplegia on physical measures (fall risk and balance).

RESULTS

The TMT part B and BA correlated (r = 0.44 and r = 0.45) with Morse fall scale. Only EF significantly explained fall risk, whereas global cognition and vascular parkinsonism/hemiplegia did not. These findings were confirmed by Bayesian evidence and parsimony model selection. Balance was not significantly correlated with any of the neuropsychological tests.

CONCLUSIONS

This is the first study investigating the relationship between cognitive and physical measures in a sample of older people with chronic CVD. The results are consistent with previous findings that link EF with fall risk in CVD.

Informing water distribution line rehabilitation through quantitative microbial risk assessment

Poor urban water quality has been linked to diminished source water quality, poorly functioning water treatment systems and infiltration into distribution lines after treatment resulting in microbiological contamination. With limited funding to rehabilitate distribution lines, developing nations need tools to identify the areas of greatest concern to human health so as to target cost effective remediation approaches. Herein, a case study of Hyderabad, Pakistan was used to demonstrate the efficacy of combining quantitative microbial risk assessment (QMRA) for multiple pathogens with spatial distribution system modeling to identify areas for pipe rehabilitation. Abundance of Escherichia coli, Enterococcus (enterococci), Salmonella spp., Shigella spp., Giardia intestinalis, Vibrio cholera, norovirus GI and adenovirus 40/41, were determined in 85 locations including the source water, treatment plant effluent and the city distribution lines. Bayesian statistics and Monte Carlo simulations were used in the QMRA to account for left-censored microbial abundance distributions. Bacterial and viral abundances in the distribution system samples decreased as follows: 9400 ± 19,800 norovirus gene copies/100 mL (average ± standard deviation, 100% of samples positive); 340 ± 2200 enterococci CFU/100 mL (94%), 71 ± 97 Shigella sp. CFU/100 mL (97%), 60 ± 360 E. coli CFU/100 mL (89%), 35 ± 79 adenovirus gene copies/100 mL (100%), and 21 ± 46 Salmonella sp. CFU/100 mL (76%). The QMRA revealed unacceptable probabilities of illness (>1 in 10,000 illness level) from the four exposure routes considered (drinking water, or only showering, tooth brushing, and rinsing vegetables consumed raw). Disease severity indices based on the QMRA combined with mapping the distribution system revealed areas for targeted rehabilitation. The combined intensive sampling, risk assessment and mapping can be used in low- and middle-income countries to target distribution system rehabilitation efforts and improve health outcomes.

Bayesian Causality

Although no universally accepted definition of causality exists, in practice one is often faced with the question of statistically assessing causal relationships in different settings. We present a uniform general approach to causality problems derived from the axiomatic foundations of the Bayesian statistical framework. In this approach, causality statements are viewed as hypotheses, or models, about the world and the fundamental object to be computed is the posterior distribution of the causal hypotheses, given the data and the background knowledge. Computation of the posterior, illustrated here in simple examples, may involve complex probabilistic modeling but this is no different than in any other Bayesian modeling situation. The main advantage of the approach is its connection to the axiomatic foundations of the Bayesian framework, and the general uniformity with which it can be applied to a variety of causality settings, ranging from specific to general cases, or from causes of effects to effects of causes.

Uncertainty in Functional Network Representations of Brain Activity of Alcoholic Patients

In spite of the large attention received by brain activity analyses through functional networks, the effects of uncertainty on such representations have mostly been neglected. We here elaborate the hypothesis that such uncertainty is not just a nuisance, but that on the contrary is condition-dependent. We test this hypothesis by analysing a large set of EEG brain recordings corresponding to control subjects and patients suffering from alcoholism, through the reconstruction of the corresponding Maximum Spanning Trees (MSTs), the assessment of their topological differences, and the comparison of two frequentist and Bayesian reconstruction approaches. A machine learning model demonstrates that the Bayesian reconstruction encodes more information than the frequentist one, and that such additional information is related to the uncertainty of the topological structures. We finally show how the Bayesian approach is more effective in the validation of generative models, over and above the frequentist one, by proposing and disproving two models based on additive noise.

Elucidating user behaviours in a digital health surveillance system to correct prevalence estimates

Estimating seasonal influenza prevalence is of undeniable public health importance, but remains challenging with traditional datasets due to cost and timeliness. Digital epidemiology has the potential to address this challenge, but can introduce sampling biases that are distinct to traditional systems. In online participatory health surveillance systems, the voluntary nature of the data generating process must be considered to address potential biases in estimates. Here we examine user behaviours in one such platform, FluTracking, from 2011 to 2017. We build a Bayesian model to estimate probabilities of an individual reporting in each week, given their past reporting behaviour, and to infer the weekly prevalence of influenza-like-illness (ILI) in Australia. We show that a model that corrects for user behaviour can substantially affect ILI estimates. The model examined here elucidates several factors, such as the status of having ILI and consistency of prior reporting, that are strongly associated with the likelihood of participating in online health surveillance systems. This framework could be applied to other digital participatory health systems where participation is inconsistent and sampling bias may be of concern.

Local and global analysis of macromolecular atomic displacement parameters

This paper describes the global and local analysis of atomic displacement parameters (ADPs) of macromolecules in X-ray crystallography. The distribution of ADPs is shown to follow the shifted inverse-gamma distribution or a mixture of these distributions. The mixture parameters are estimated using the expectation-maximization algorithm. In addition, a method for the resolution- and individual ADP-dependent local analysis of neighbouring atoms has been designed. This method facilitates the detection of mismodelled atoms, heavy-metal atoms and disordered and/or incorrectly modelled ligands. Both global and local analyses can be used to detect errors in atomic models, thus helping in the (re)building, refinement and validation of macromolecular structures. This method can also serve as an additional validation tool during PDB deposition.

Validation of PATHFx 2.0: An open-source tool for estimating survival in patients undergoing pathologic fracture fixation

Treatment decisions in patients with metastatic bone disease rely on accurate survival estimation. We developed the original PATHFx models using expensive, proprietary software and now seek to provide a more cost-effective solution. Using open-source machine learning software to create PATHFx version 2.0, we asked whether PATHFx 2.0 could be created using open-source methods and externally validated in two unique patient populations. The training set of a well-characterized, database records of 189 patients and the bnlearn package within R Version 3.5.1 (R Foundation for Statistical Computing), was used to establish a series of Bayesian belief network models designed to predict survival at 1, 3, 6, 12, 18, and 24 months. Each was externally validated in both a Scandinavian (n = 815 patients) and a Japanese (n = 261 patients) data set. Brier scores and receiver operating characteristic curves to assessed discriminatory ability. Decision curve analysis (DCA) evaluated whether models should be used clinically. DCA showed that the model should be used clinically at all time points in the Scandinavian data set. For the 1-month time point, DCA of the Japanese data set suggested to expect better outcomes assuming all patients will survive greater than 1 month. Brier scores for each curve demonstrate that the models are accurate at each time point. Statement of Clinical Significance: we successfully transitioned to PATHFx 2.0 using open-source software and externally validated it in two unique patient populations, which can be used as a cost-effective option to guide surgical decisions in patients with metastatic bone disease.

Estimating the number of usability problems affecting medical devices: modelling the discovery matrix

Background

Usability testing of medical devices are mandatory for market access. The testings’ goal is to identify usability problems that could cause harm to the user or limit the device’s effectiveness. In practice, human factor engineers study participants under actual conditions of use and list the problems encountered. This results in a binary discovery matrix in which each row corresponds to a participant, and each column corresponds to a usability problem. One of the main challenges in usability testing is estimating the total number of problems, in order to assess the completeness of the discovery process. Today’s margin-based methods fit the column sums to a binomial model of problem detection. However, the discovery matrix actually observed is truncated because of undiscovered problems, which corresponds to fitting the marginal sums without the zeros. Margin-based methods fail to overcome the bias related to truncation of the matrix. The objective of the present study was to develop and test a matrix-based method for estimating the total number of usability problems.

Methods

The matrix-based model was based on the full discovery matrix (including unobserved columns) and not solely on a summary of the data (e.g. the margins). This model also circumvents a drawback of margin-based methods by simultaneously estimating the model’s parameters and the total number of problems. Furthermore, the matrix-based method takes account of a heterogeneous probability of detection, which reflects a real-life setting. As suggested in the usability literature, we assumed that the probability of detection had a logit-normal distribution.

Results

We assessed the matrix-based method’s performance in a range of settings reflecting real-life usability testing and with heterogeneous probabilities of problem detection. In our simulations, the matrix-based method improved the estimation of the number of problems (in terms of bias, consistency, and coverage probability) in a wide range of settings. We also applied our method to five real datasets from usability testing.

Conclusions

Estimation models (and particularly matrix-based models) are of value in estimating and monitoring the detection process during usability testing. Matrix-based models have a solid mathematical grounding and, with a view to facilitating the decision-making process for both regulators and device manufacturers, should be incorporated into current standards.

National population mapping from sparse survey data: A hierarchical Bayesian modeling framework to account for uncertainty

High-resolution population estimates are essential for government planning, development projects, and public health campaigns, but countries where this information is most needed are often where recent national census data are least available. We present a modeling framework that combines recent neighborhood-scale microcensus surveys with national-scale data from satellite images and digital maps to estimate population sizes for every 100-m grid square nationally. We present a case study from Nigeria where population estimates with national coverage were produced using household survey data from 1,141 locations. This work represents a significant step toward achieving high-resolution population estimates with national coverage from sparse population data while providing reliable estimates of uncertainty at any spatial scale.

Population estimates are critical for government services, development projects, and public health campaigns. Such data are typically obtained through a national population and housing census. However, population estimates can quickly become inaccurate in localized areas, particularly where migration or displacement has occurred. Some conflict-affected and resource-poor countries have not conducted a census in over 10 y. We developed a hierarchical Bayesian model to estimate population numbers in small areas based on enumeration data from sample areas and nationwide information about administrative boundaries, building locations, settlement types, and other factors related to population density. We demonstrated this model by estimating population sizes in every 10- m grid cell in Nigeria with national coverage. These gridded population estimates and areal population totals derived from them are accompanied by estimates of uncertainty based on Bayesian posterior probabilities. The model had an overall error rate of 67 people per hectare (mean of absolute residuals) or 43% (using scaled residuals) for predictions in out-of-sample survey areas (approximately 3 ha each), with increased precision expected for aggregated population totals in larger areas. This statistical approach represents a significant step toward estimating populations at high resolution with national coverage in the absence of a complete and recent census, while also providing reliable estimates of uncertainty to support informed decision making.

Racial, Ethnic, and Geographic Disparities in Novel Coronavirus (Severe Acute Respiratory Syndrome Coronavirus 2) Test Positivity in North Carolina

Background

Emerging evidence suggests that black and Hispanic communities in the United States are disproportionately affected by coronavirus disease 2019 (COVID-19). A complex interplay of socioeconomic and healthcare disparities likely contribute to disproportionate COVID-19 risk.

Methods

We conducted a geospatial analysis to determine whether individual- and neighborhood-level attributes predict local odds of testing positive for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). We analyzed 29 138 SARS-CoV-2 tests within the 6-county catchment area for Duke University Health System from March to June 2020. We used generalized additive models to analyze the spatial distribution of SARS-CoV-2 positivity. Adjusted models included individual-level age, gender, and race, as well as neighborhood-level Area Deprivation Index, population density, demographic composition, and household size.

Results

Our dataset included 27 099 negative and 2039 positive unique SARS-CoV-2 tests. The odds of a positive SARS-CoV-2 test were higher for males (odds ratio [OR], 1.43; 95% credible interval [CI], 1.30–1.58), blacks (OR, 1.47; 95% CI, 1.27–1.70), and Hispanics (OR, 4.25; 955 CI, 3.55–5.12). Among neighborhood-level predictors, percentage of black population (OR, 1.14; 95% CI, 1.05–1.25), and percentage Hispanic population (OR, 1.23; 95% CI, 1.07–1.41) also influenced the odds of a positive SARS-CoV-2 test. Population density, average household size, and Area Deprivation Index were not associated with SARS-CoV-2 test results after adjusting for race.

Conclusions

The odds of testing positive for SARS-CoV-2 were higher for both black and Hispanic individuals, as well as within neighborhoods with a higher proportion of black or Hispanic residents—confirming that black and Hispanic communities are disproportionately affected by SARS-CoV-2.

Using informative prior based on expert opinion in Bayesian estimation of the transition probability matrix in Markov modelling-an example from the cost-effectiveness analysis of the treatment of patients with predominantly negative symptoms of schizophrenia with cariprazine

Background

When patient health state transition evidence is missing from clinical literature, analysts are inclined to make simple assumptions to complete the transition matrices within a health economic model. Our aim was to provide a solution for estimating transition matrices by the Bayesian statistical method within a health economic model when empirical evidence is lacking.

Methods

We used a previously published cost-effectiveness analysis of the use of cariprazine compared to that of risperidone in patients with predominantly negative symptoms of schizophrenia. We generated the treatment-specific state transition probability matrices in three different ways: (1) based only on the observed clinical trial data; (2) based on Bayesian estimation where prior transition probabilities came from experts’ opinions; and (3) based on Bayesian estimation with vague prior transition probabilities (i.e., assigning equal prior probabilities to the missing transitions from one state to the others). For the second approach, we elicited Dirichlet prior distributions by three clinical experts. We compared the transition probability matrices and the incremental quality-adjusted life years (QALYs) across the three approaches.

Results

The estimates of the prior transition probabilities from the experts were feasible to obtain and showed considerable consistency with the clinical trial data. As expected, the estimated health benefit of the treatments was different when only the clinical trial data were considered (QALY difference 0.0260), its combination with the experts’ beliefs were used in the economic model (QALY difference 0.0253), and when vague prior distributions were used (QALY difference 0.0243).

Conclusions

Imputing zeros to missing transition probabilities in Markov models might be untenable from the clinical perspective and may result in inappropriate estimates. Bayesian statistics provides an appropriate framework for imputing missing values without making overly simple assumptions. Informative priors based on expert opinions might be more appropriate than vague priors.

Intact Context-Dependent Modulation of Conflict Monitoring in Childhood ADHD

Objective: Conflict monitoring is well known to be modulated by context. This is known as the Gratton effect, meaning that the degree of interference is smaller when a stimulus-response conflict had been encountered previously. It is unclear to what extent these processes are changed in ADHD. Method: Children with ADHD (combined subtype) and healthy controls performed a modified version of the sequence flanker task. Results: Patients with ADHD made significantly more errors than healthy controls, indicating general performance deficits. However, there were no differences regarding reaction times, indicating an intact Gratton effect in ADHD. These results were supported by Bayesian statistics. Conclusion: The results suggest that the ability to take contextual information into account during conflict monitoring is preserved in patients with ADHD despite this disorder being associated with changes in executive control functions overall. These findings are discussed in light of different theoretical accounts on contextual modulations of conflict monitoring.

A Restricted Four-Parameter IRT Model: The Dyad Four-Parameter Normal Ogive (Dyad-4PNO) Model

Recently, there has been a renewed interest in the four-parameter item response theory model as a way to capture guessing and slipping behaviors in responses. Research has shown, however, that the nested three-parameter model suffers from issues of unidentifiability (San Martín et al. in Psychometrika 80:450-467, 2015), which places concern on the identifiability of the four-parameter model. Borrowing from recent advances in the identification of cognitive diagnostic models, in particular, the DINA model (Gu and Xu in Stat Sin https://doi.org/10.5705/ss.202018.0420 , 2019), a new model is proposed with restrictions inspired by this new literature to help with the identification issue. Specifically, we show conditions under which the four-parameter model is strictly and generically identified. These conditions inform the presentation of a new exploratory model, which we call the dyad four-parameter normal ogive (Dyad-4PNO) model. This model is developed by placing a hierarchical structure on the DINA model and imposing equality constraints on a priori unknown dyads of items. We present a Bayesian formulation of this model, and show that model parameters can be accurately recovered. Finally, we apply the model to a real dataset.

Spatiotemporal variation and determinants of population's PM2.5 exposure risk in China, 1998-2017: a case study of the Beijing-Tianjin-Hebei region

PM2.5 pollution has emerged as a global human health risk. The best measure of its impact is a population's PM2.5 exposure (PPM2.5E), an index that simultaneously considers PM2.5 concentrations and population spatial density. The spatiotemporal variation of PPM2.5E over the Beijing-Tianjin-Hebei (BTH) region, which is the national capital region of China, was investigated using a Bayesian space-time model, and the influence patterns of the anthropic and geographical factors were identified using the GeoDetector model and Pearson correlation analysis. The spatial pattern of PPM2.5E maintained a stable structure over the BTH region's distinct terrain, which has been described as "high in the northwest, low in the southeast". The spatial difference of PPM2.5E intensified annually. An overall increase of 6.192 (95% CI 6.186, 6.203) ×103 μg/m3 ∙ persons/km2 per year occurred over the BTH region from 1998 to 2017. The evolution of PPM2.5E in the region can be described as "high value, high increase" and "low value, low increase", since human activities related to gross domestic product (GDP) and energy consumption (EC) were the main factors in its occurrence. GDP had the strongest explanatory power of 76% (P < 0.01), followed by EC and elevation (EL), which accounted for 61% (P < 0.01) and 40% (P < 0.01), respectively. There were four factors, proportion of secondary industry (PSI), normalized differential vegetation index (NDVI), relief amplitude (RA), and EL, associated negatively with PPM2.5E and four factors, GDP, EC, annual precipitation (AP), and annual average temperature (AAT), associated positively with PPM2.5E. Remarkably, the interaction of GDP and NDVI, which was 90%, had the greatest explanatory power for PPM2.5E ' s diffusion and impact on the BTH region.

Assembly of Capsids from Hepatitis B Virus Core Protein Progresses through Highly Populated Intermediates in the Presence and Absence of RNA

The genetic material of viruses is protected by protein shells that are assembled from a large number of subunits in a process that is efficient and robust. Many of the mechanistic details underpinning efficient assembly of virus capsids are still unknown. The assembly mechanism of hepatitis B capsids has been intensively researched using a truncated core protein lacking the C-terminal domain responsible for binding genomic RNA. To resolve the assembly intermediates of hepatitis B virus (HBV), we studied the formation of nucleocapsids and empty capsids from full-length hepatitis B core proteins, using time-resolved small-angle X-ray scattering. We developed a detailed structural model of the HBV capsid assembly process using a combination of analysis with multivariate curve resolution, structural modeling, and Bayesian ensemble inference. The detailed structural analysis supports an assembly pathway that proceeds through the formation of two highly populated intermediates, a trimer of dimers and a partially closed shell consisting of around 40 dimers. These intermediates are on-path, transient and efficiently convert into fully formed capsids. In the presence of an RNA oligo that binds specifically to the C-terminal domain the assembly proceeds via a similar mechanism to that in the absence of nucleic acids. Comparisons between truncated and full-length HBV capsid proteins reveal that the unstructured C-terminal domain has a significant impact on the assembly process and is required to obtain a more complete mechanistic understanding of HBV capsid formation. These results also illustrate how combining scattering information from different time-points during time-resolved experiments can be utilized to derive a structural model of protein self-assembly pathways.

A Multilevel Bayesian Approach to Improve Effect Size Estimation in Regression Modeling of Metabolomics Data Utilizing Imputation with Uncertainty

To ensure scientific reproducibility of metabolomics data, alternative statistical methods are needed. A paradigm shift away from the p-value toward an embracement of uncertainty and interval estimation of a metabolite's true effect size may lead to improved study design and greater reproducibility. Multilevel Bayesian models are one approach that offer the added opportunity of incorporating imputed value uncertainty when missing data are present. We designed simulations of metabolomics data to compare multilevel Bayesian models to standard logistic regression with corrections for multiple hypothesis testing. Our simulations altered the sample size and the fraction of significant metabolites truly different between two outcome groups. We then introduced missingness to further assess model performance. Across simulations, the multilevel Bayesian approach more accurately estimated the effect size of metabolites that were significantly different between groups. Bayesian models also had greater power and mitigated the false discovery rate. In the presence of increased missing data, Bayesian models were able to accurately impute the true concentration and incorporating the uncertainty of these estimates improved overall prediction. In summary, our simulations demonstrate that a multilevel Bayesian approach accurately quantifies the estimated effect size of metabolite predictors in regression modeling, particularly in the presence of missing data.

Reporting Bayesian Results

Because of the different philosophy of Bayesian statistics, where parameters are random variables and data are considered fixed, the analysis and presentation of results will differ from that of frequentist statistics. Most importantly, the probabilities that a parameter is in certain regions of the parameter space are crucial quantities in Bayesian statistics that are not calculable (or considered important) in the frequentist approach that is the basis of much of traditional statistics. In this article, I discuss the implications of these differences for presentation of the results of Bayesian analyses. In doing so, I present more detailed guidelines than are usually provided and explain the rationale for my suggestions.

Effect of Volume on Eccentric Overload-Induced Postactivation Potentiation of Jumps

PURPOSE

To investigate the postactivation potentiation (PAP) effects of different eccentric overload (EOL) exercise volumes on countermovement-jump (CMJ) and standing-long-jump (LJ) performance.

METHODS

In total, 13 male university soccer players participated in a crossover design study following a familiarization period. Control (no PAP) CMJ and LJ performances were recorded, and 3 experimental protocols were performed in a randomized order: 1, 2, or 3 sets of 6 repetitions of flywheel EOL half-squats (inertia = 0.029 kg·m2). Performance of CMJ and LJ was measured 3 and 6 minutes after all experimental conditions. The time course and magnitude of the PAP were compared between conditions.

RESULTS

Meaningful positive PAP effects were reported for CMJ after 2 (Bayes factor [BF10] = 3.15, moderate) and 3 (BF10 = 3.25, moderate) sets but not after 1 set (BF10 = 2.10, anecdotal). Meaningful positive PAP effects were reported for LJ after 2 (BF10 = 3.05, moderate) and 3 (BF10 = 3.44, moderate) sets but not after 1 set (BF10 = 0.53, anecdotal). The 2- and 3-set protocols resulted in meaningful positive PAP effects on both CMJ and LJ after 6 minutes but not after 3 minutes.

CONCLUSION

This study reported beneficial effects of multiset EOL exercise over a single set. A minimum of 2 sets of flywheel EOL half-squats are required to induce PAP effects on CMJ and LJ performance of male university soccer players. Rest intervals of around 6 minutes (>3 min) are required to maximize the PAP effects via multiple sets of EOL exercise. However, further research is needed to clarify the optimal EOL protocol configurations for PAP response.

Bayesian-based predictions of COVID-19 evolution in Texas using multispecies mixture-theoretic continuum models

We consider a mixture-theoretic continuum model of the spread of COVID-19 in Texas. The model consists of multiple coupled partial differential reaction-diffusion equations governing the evolution of susceptible, exposed, infectious, recovered, and deceased fractions of the total population in a given region. We consider the problem of model calibration, validation, and prediction following a Bayesian learning approach implemented in OPAL (the Occam Plausibility Algorithm). Our goal is to incorporate COVID-19 data to calibrate the model in real-time and make meaningful predictions and specify the confidence level in the prediction by quantifying the uncertainty in key quantities of interests. Our results show smaller mortality rates in Texas than what is reported in the literature. We predict 7003 deceased cases by September 1, 2020 in Texas with 95 % CI 6802-7204. The model is validated for the total deceased cases, however, is found to be invalid for the total infected cases. We discuss possible improvements of the model.

Early administration of glucocorticoid for thyroid storm: analysis of a national administrative database

Background

Thyroid storm is a life-threatening disease with a mortality rate of over 10%. Although glucocorticoids have been recommended as a treatment option for thyroid storm, supportive evidence based on a large-scale clinical research is lacking. The objective of the current study was to evaluate the beneficial effects of glucocorticoids in the treatment of patients with severe thyroid storm.

Methods

A retrospective nationwide cohort study was conducted using a Japanese national administrative claims database. Patients admitted to intensive care units due to severe thyroid storm between the financial years 2013 and 2017 were included in the study. The primary outcome was in-hospital mortality; secondary outcomes were mortality within 30 days and insulin administration during hospitalization. Generalized linear mixed model (GLMM) with maximum likelihood estimation (MLE) and Bayesian estimation using Markov chain Monte Carlo methods (MCMC), in addition to propensity score matching (PSM), were used for statistical analysis.

Results

A total of 811 patients were included in the study, of which 600 patients were treated with glucocorticoids, and 211 patients were treated without glucocorticoids. The early administration of glucocorticoids was not associated with a significant improvement in the in-hospital mortality of patients with thyroid storm [adjusted odds ratio (95% confidence interval) = 1.77 (0.95–3.34), 1.44 (1.14–1.93), and 1.46 (0.72–3.00) in the GLMM (MLE), GLMM (MCMC), and PSM, respectively]. The results of mortality within 30 days were almost identical to the results of in-hospital mortality. However, insulin use was significantly higher in the glucocorticoid group.

Conclusions

This analysis of a nationwide administrative database indicates that the administration of glucocorticoids does not improve the survival of patients with thyroid storm.

Comparing different spatio-temporal modeling methods in dengue fever data analysis in Colombia during 2012-2015

In this paper, we compare a variety of spatio-temporal conditional autoregressive models to a dengue fever dataset in Colombia, and incorporate an innovative data transformation method in the data analysis. In order to gain a better understanding on the effects of different niche variables in the epidemiological process, we explore Poisson-lognormal and binomial models with different Bayesian spatio-temporal modeling methods in this paper. Our results show that the selected model can well capture the variations of the data. The population density, elevation, daytime and night land surface temperatures are among the contributory variables to identify potential dengue outbreak regions; precipitation and vegetation variables are not significant in the selected spatio-temporal mixed effects model. The generated dengue fever probability maps from the model show a geographic distribution of risk that apparently coincides with the elevation gradient. The results in the paper provide the most benefits for future work in dengue studies.

A Randomized Controlled Trial to Investigate Safety and Variability of Egg Excretion After Repeated Controlled Human Hookworm Infection

BACKGROUND

Controlled human hookworm infections could significantly contribute to the development of a hookworm vaccine. However, current models are hampered by low and unstable egg output, reducing generalizability and increasing sample sizes. This study aims to investigate the safety, tolerability, and egg output of repeated exposure to hookworm larvae.

METHODS

Twenty-four healthy volunteers were randomized, double-blindly, to 1, 2, or 3 doses of 50 Necator americanus L3 larvae at 2-week intervals. Volunteers were monitored weekly and were treated with albendazole at week 20.

RESULTS

There was no association between larval dose and number or severity of adverse events. Geometric mean egg loads stabilized at 697, 1668, and 1914 eggs per gram feces for the 1 × 50L3, 2 × 50L3, and 3 × 50L3 group, respectively. Bayesian statistical modeling showed that egg count variability relative to the mean was reduced with a second infectious dose; however, the third dose did not increase egg load or decrease variability. We therefore suggest 2 × 50L3 as an improved challenge dose. Model-based simulations indicates increased frequency of stool sampling optimizes the power of hypothetical vaccine trials.

CONCLUSIONS

Repeated infection with hookworm larvae increased egg counts to levels comparable to the field and reduced relative variability in egg output without aggravating adverse events.

CLINICAL TRIALS REGISTRATION

NCT03257072.

Assessing cetacean populations using integrated population models: an example with Cook Inlet beluga whales

Effective conservation and management of animal populations requires knowledge of abundance and trends. For many species, these quantities are estimated using systematic visual surveys. Additional individual-level data are available for some species. Integrated population modeling (IPM) offers a mechanism for leveraging these data sets into a single estimation framework. IPMs that incorporate both population- and individual-level data have previously been developed for birds, but have rarely been applied to cetaceans. Here, we explore how IPMs can be used to improve the assessment of cetacean populations. We combined three types of data that are typically available for cetaceans of conservation concern: population-level visual survey data, individual-level capture-recapture data, and data on anthropogenic mortality. We used this IPM to estimate the population dynamics of the Cook Inlet population of beluga whales (CIBW; Delphinapterus leucas) as a case study. Our state-space IPM included a population process model and three observational submodels: (1) a group detection model to describe group size estimates from aerial survey data; (2) a capture-recapture model to describe individual photographic capture-recapture data; and (3) a Poisson regression model to describe historical hunting data. The IPM produces biologically plausible estimates of population trajectories consistent with all three data sets. The estimated population growth rate since 2000 is less than expected for a recovering population. The estimated juvenile/adult survival rate is also low compared to other cetacean populations, indicating that low survival may be impeding recovery. This work demonstrates the value of integrating various data sources to assess cetacean populations and serves as an example of how multiple, imperfect data sets can be combined to improve our understanding of a population of interest. The model framework is applicable to other cetacean populations and to other taxa for which similar data types are available.

Multilevel Regression and Poststratification Versus Survey Sample Weighting for Estimating Population Quantities in Large Population Health Studies

Multilevel regression and poststratification (MRP) is a model-based approach for estimating a population parameter of interest, generally from large-scale surveys. It has been shown to be effective in highly selected samples, which is particularly relevant to investigators of large-scale population health and epidemiologic surveys facing increasing difficulties in recruiting representative samples of participants. We aimed to further examine the accuracy and precision of MRP in a context where census data provided reasonable proxies for true population quantities of interest. We considered 2 outcomes from the baseline wave of the Ten to Men study (Australia, 2013-2014) and obtained relevant population data from the 2011 Australian Census. MRP was found to achieve generally superior performance relative to conventional survey weighting methods for the population as a whole and for population subsets of varying sizes. MRP resulted in less variability among estimates across population subsets relative to sample weighting, and there was some evidence of small gains in precision when using MRP, particularly for smaller population subsets. These findings offer further support for MRP as a promising analytical approach for addressing participation bias in the estimation of population descriptive quantities from large-scale health surveys and cohort studies.

Probabilistic reconstruction of measles transmission clusters from routinely collected surveillance data

Pockets of susceptibility resulting from spatial or social heterogeneity in vaccine coverage can drive measles outbreaks, as cases imported into such pockets are likely to cause further transmission and lead to large transmission clusters. Characterizing the dynamics of transmission is essential for identifying which individuals and regions might be most at risk. As data from detailed contact-tracing investigations are not available in many settings, we developed an R package called o2geosocial to reconstruct the transmission clusters and the importation status of the cases from their age, location, genotype and onset date. We compared our inferred cluster size distributions to 737 transmission clusters identified through detailed contact-tracing in the USA between 2001 and 2016. We were able to reconstruct the importation status of the cases and found good agreement between the inferred and reference clusters. The results were improved when the contact-tracing investigations were used to set the importation status before running the model. Spatial heterogeneity in vaccine coverage is difficult to measure directly. Our approach was able to highlight areas with potential for local transmission using a minimal number of variables and could be applied to assess the intensity of ongoing transmission in a region.

Size- and stage-dependence in cause-specific mortality of migratory brown trout

Evidence-based management of natural populations under strong human influence frequently requires not only estimates of survival but also knowledge about how much mortality is due to anthropogenic vs. natural causes. This is the case particularly when individuals vary in their vulnerability to different causes of mortality due to traits, life history stages, or locations. Here, we estimated harvest and background (other cause) mortality of landlocked migratory salmonids over half a century. In doing so, we quantified among-individual variation in vulnerability to cause-specific mortality resulting from differences in body size and spawning location relative to a hydropower dam. We constructed a multistate mark-recapture model to estimate harvest and background mortality hazard rates as functions of a discrete state (spawning location) and an individual time-varying covariate (body size). We further accounted for among-year variation in mortality and migratory behaviour and fit the model to a unique 50-year time series of mark-recapture-recovery data on brown trout (Salmo trutta) in Norway. Harvest mortality was highest for intermediate-sized trout, and outweighed background mortality for most of the observed size range. Background mortality decreased with body size for trout spawning above the dam and increased for those spawning below. All vital rates varied substantially over time, but a trend was evident only in estimates of fishers' reporting rate, which decreased from over 50% to less than 10% throughout the study period. We highlight the importance of body size for cause-specific mortality and demonstrate how this can be estimated using a novel hazard rate parameterization for mark-recapture models. Our approach allows estimating effects of individual traits and environment on cause-specific mortality without confounding, and provides an intuitive way to estimate temporal patterns within and correlation among different mortality sources.

Expert Elicitation for Latent Growth Curve Models: The Case of Posttraumatic Stress Symptoms Development in Children With Burn Injuries

Experts provide an alternative source of information to classical data collection methods such as surveys. They can provide additional insight into problems, supplement existing data, or provide insights when classical data collection is troublesome. In this paper, we explore the (dis)similarities between expert judgments and data collected by traditional data collection methods regarding the development of posttraumatic stress symptoms (PTSSs) in children with burn injuries. By means of an elicitation procedure, the experts' domain expertise is formalized and represented in the form of probability distributions. The method is used to obtain beliefs from 14 experts, including nurses and psychologists. Those beliefs are contrasted with questionnaire data collected on the same issue. The individual and aggregated expert judgments are contrasted with the questionnaire data by means of Kullback-Leibler divergences. The aggregated judgments of the group that mainly includes psychologists resemble the questionnaire data more than almost all of the individual expert judgments.

Stakeholder perspectives on adaptive clinical trials: a scoping review

Background

Adaptive clinical trials (ACTs) represent an emerging approach to trial design where accumulating data are used to make decisions about future conduct. Adaptations can include comparisons of multiple dose tiers, response-adaptive randomization, sample size re-estimation, and efficacy/futility stopping rules. The objective of this scoping review is to assess stakeholder attitudes, perspectives, and understanding of adaptive trials.

Methods

We conducted a review of articles examining stakeholders encompassing the broad medical trial community’s perspectives of adaptive designs (ADs). A computerized search was conducted of four electronic databases with relevant search terms. Following screening of articles, the primary findings of each included article were coded for study design, population studied, purpose, and primary implications.

Results

Our team retrieved 167 peer-reviewed titles in total from the database search and 5 additional titles through searching web-based search engines for gray literature. Of those 172 titles, 152 were non-duplicate citations. Of these, 119 were not given full-text reviews, as their titles and abstracts indicated that they did not meet the inclusion criteria. Thirty-three articles were carefully examined for relevance, and of those, 18 were chosen to be part of the analysis; the other 15 were excluded, as they were not relevant upon closer inspection.

Perceived advantages to ADs included limiting ineffective treatments and efficiency in answering the research question; −perceived barriers included insufficient sample size for secondary outcomes, challenges of consent, potential for bias, risk of type 1 error, cost and time to adaptively design trials, unclear rationales for using Ads, and, most importantly, a lack of education regarding ADs among stakeholders within the clinical trial community. Perceptions among different types of stakeholders varied from sector to sector, with patient perspectives being noticeably absent from the literature.

Conclusion

There are diverse perceptions regarding ADs among stakeholders. Further training, guidelines, and toolkits on the proper use of ADs are needed at all levels to overcome many of these perceived barriers. While education for principal investigators is important, it is also crucial to educate other groups in the community, such as patients, as well as clinicians and staff involved in their daily implementation.

Changes in firearm mortality following the implementation of state laws regulating firearm access and use

Many US states have tried to regulate firearm storage and use to reduce the 39,000 firearms-related deaths that occur each year. Looking at three classes of laws that regulate children’s access to firearms, the carrying of a concealed firearm, and the use of a firearm in self-defense, we found that state laws restricting firearm storage and use are associated with a subsequent 11% decrease in the firearms-related death rate. In a hypothetical situation in which there are 39,000 firearms deaths nationally under the permissive combination of these three laws, we expect 4,475 (80% CI, 1,761 to 6,949) more deaths nationally than under the restrictive combination of these laws.

Although 39,000 individuals die annually from gunshots in the US, research examining the effects of laws designed to reduce these deaths has sometimes produced inconclusive or contradictory findings. We evaluated the effects on total firearm-related deaths of three classes of gun laws: child access prevention (CAP), right-to-carry (RTC), and stand your ground (SYG) laws. The analyses exploit changes in these state-level policies from 1970 to 2016, using Bayesian methods and a modeling approach that addresses several methodological limitations of prior gun policy evaluations. CAP laws showed the strongest evidence of an association with firearm-related death rate, with a probability of 0.97 that the death rate declined at 6 y after implementation. In contrast, the probability of being associated with an increase in firearm-related deaths was 0.87 for RTC laws and 0.77 for SYG laws. The joint effects of these laws indicate that the restrictive gun policy regime (having a CAP law without an RTC or SYG law) has a 0.98 probability of being associated with a reduction in firearm-related deaths relative to the permissive policy regime. This estimated effect corresponds to an 11% reduction in firearm-related deaths relative to the permissive legal regime. Our findings suggest that a small but meaningful decrease in firearm-related deaths may be associated with the implementation of more restrictive gun policies.

Spatiotemporal matching between medical resources and population ageing in China from 2008 to 2017

Background

Globally, the increasingly severe population ageing issue has been creating challenges in terms of medical resource allocation and public health policies. The aim of this study is to address the space-time trends of the population-ageing rate (PAR), the number of medical resources per thousand residents (NMRTR) in mainland China in the past 10 years, and to investigate the spatial and temporal matching between the PAR and NMRTR in mainland China.

Methods

The Bayesian space-time hierarchy model was employed to investigate the spatiotemporal variation of PAR and NMRTR in mainland China over the past 10 years. Subsequently, a Bayesian Geo-Detector model was developed to evaluate the spatial and temporal matching levels between PAR and NMRTR at national level. The matching odds ratio (OR) index proposed in this paper was applied to measure the matching levels between the two terms in each provincial area.

Results

The Chinese spatial and temporal matching q-statistic values between the PAR and three vital types of NMRTR were all less than 0.45. Only the spatial matching Bayesian q-statistic values between the PAR and the number of beds in hospital reached 0.42 (95% credible interval: 0.37, 0.48) nationwide. Chongqing and Guizhou located in southwest China had the highest spatial and temporal matching ORs, respectively, between the PAR and the three types of NMRTR. The spatial pattern of the spatial and temporal matching ORs between the PAR and NMRTR in mainland China exhibited distinct geographical features, but the geographical structure of the spatial matching differed from that of the temporal matching between the PAR and NMRTR.

Conclusion

The spatial and temporal matching degrees between the PAR and NMRTR in mainland China were generally very low. The provincial regions with high PAR largely experienced relatively low spatial matching levels between the PAR and NMRTR, and vice versa. The geographical pattern of the temporal matching between the PAR and NMRTR exhibited the feature of north-south differentiation.

HARMONIES: A Hybrid Approach for Microbiome Networks Inference via Exploiting Sparsity

The human microbiome is a collection of microorganisms. They form complex communities and collectively affect host health. Recently, the advances in next-generation sequencing technology enable the high-throughput profiling of the human microbiome. This calls for a statistical model to construct microbial networks from the microbiome sequencing count data. As microbiome count data are high-dimensional and suffer from uneven sampling depth, over-dispersion, and zero-inflation, these characteristics can bias the network estimation and require specialized analytical tools. Here we propose a general framework, HARMONIES, Hybrid Approach foR MicrobiOme Network Inferences via Exploiting Sparsity, to infer a sparse microbiome network. HARMONIES first utilizes a zero-inflated negative binomial (ZINB) distribution to model the skewness and excess zeros in the microbiome data, as well as incorporates a stochastic process prior for sample-wise normalization. This approach infers a sparse and stable network by imposing non-trivial regularizations based on the Gaussian graphical model. In comprehensive simulation studies, HARMONIES outperformed four other commonly used methods. When using published microbiome data from a colorectal cancer study, it discovered a novel community with disease-enriched bacteria. In summary, HARMONIES is a novel and useful statistical framework for microbiome network inference, and it is available at https://github.com/shuangj00/HARMONIES.

Multilevel modelling as a tool to include variability and uncertainty in quantitative microbiology and risk assessment. Thermal inactivation of Listeria monocytogenes as proof of concept

Variability is inherent in biology and also substantial for microbial populations. In the context of food safety risk assessment, it refers to differences in the response of different bacterial strains (between-strain variability) and different cells (within-strain variability) to the same condition (e.g. inactivation treatment). However, its quantification based on empirical observations and its incorporation in predictive models is a challenge for both experimental design and (statistical) analysis. In this article we propose the use of multilevel models to quantify (different levels of) variability and uncertainty and include them in the predictions. As proof of concept, we analyse the microbial inactivation of Listeria monocytogenes to thermal treatments including different levels of variability (between-strain and within-strain) and uncertainty. The relationship between the microbial count and time was expressed using a (non-linear) Weibullian model. Moreover, we defined stochastic hypotheses to describe the different types of variation at the level of the kinetic parameters, as well as in the observations (microbial counts). The model parameters (kinetic parameters and variances) are estimated using Bayesian statistics. The multilevel approach was compared against an analogous, single-level model. The multilevel methodology shrinks extreme parameter estimates towards the mean according to uncertainty, thus mitigating overfitting. In addition, this approach enables to easily incorporate different levels of variation (between-strain and/or within-strain variability and/or uncertainty) in the predictions. On the other hand, multilevel (Bayesian) models are more complex to define, implement, analyse and communicate than single-level models. Nevertheless, their ability to incorporate different sources of variability in predictions make them very suitable for Quantitative Microbial Risk Assessment.

An objective Bayesian analysis of life's early start and our late arrival

Life emerged on Earth within the first quintile of its habitable window, but a technological civilization did not blossom until its last. Efforts to infer the rate of abiogenesis, based on its early emergence, are frustrated by the selection effect that if the evolution of intelligence is a slow process, then life's early start may simply be a prerequisite to our existence, rather than useful evidence for optimism. In this work, we interpret the chronology of these two events in a Bayesian framework, extending upon previous work by considering that the evolutionary timescale is itself an unknown that needs to be jointly inferred, rather than fiducially set. We further adopt an objective Bayesian approach, such that our results would be agreed upon even by those using wildly different priors for the rates of abiogenesis and evolution-common points of contention for this problem. It is then shown that the earliest microfossil evidence for life indicates that the rate of abiogenesis is at least 2.8 times more likely to be a typically rapid process, rather than a slow one. This modest limiting Bayes factor rises to 8.7 if we accept the more disputed evidence of ¹³C-depleted zircon deposits [E. A. Bell, P. Boehnke, T. M. Harrison, W. L. Mao, Proc. Natl. Acad. Sci. U.S.A. 112, 14518-14521 (2015)]. For intelligence evolution, it is found that a rare-intelligence scenario is slightly favored at 3:2 betting odds. Thus, if we reran Earth's clock, one should statistically favor life to frequently reemerge, but intelligence may not be as inevitable.

New Insights Into the Kinetics and Variability of Egg Excretion in Controlled Human Hookworm Infections

Four healthy volunteers were infected with 50 Necator americanus infective larvae (L3) in a controlled human hookworm infection trial and followed for 52 weeks. The kinetics of fecal egg counts in volunteers was assessed with Bayesian multilevel analysis, which revealed an increase between weeks 7 and 13, followed by an egg density plateau of about 1000 eggs/g of feces. Variation in egg counts was minimal between same-day measurements but varied considerably between days, particularly during the plateau phase. These analyses pave the way for the controlled human hookworm model to accelerate drug and vaccine efficacy studies.

Bayesian analysis for social data: A step-by-step protocol and interpretation

The paper proposes Bayesian analysis as an alternative approach for the conventional frequentist approach in analyzing social data. A step-by-step protocol of how to implement Bayesian multilevel model analysis with social data and how to interpret the result is presented. The article used a dataset regarding religious teachings and behaviors of lying and violence as an example. An analysis is performed using R statistical software and a bayesvl R package, which offers a network-structured model construction and visualization power to diagnose and estimate results.•

The paper provides guidance for conducting a Bayesian multilevel analysis in social sciences through constructing directed acyclic graphs (DAGs, or "relationship trees") for different models, basic and more complex ones.

•

The method also illustrates how to visualize Bayesian diagnoses and simulated posterior.

•

The interpretations of visualized diagnoses and simulated posteriors of Bayesian inference are also discussed.

bayes4psy-An Open Source R Package for Bayesian Statistics in Psychology

Research in psychology generates complex data and often requires unique statistical analyses. These tasks are often very specific, so appropriate statistical models and methods cannot be found in accessible Bayesian tools. As a result, the use of Bayesian methods is limited to researchers and students that have the technical and statistical fundamentals that are required for probabilistic programming. Such knowledge is not part of the typical psychology curriculum and is a difficult obstacle for psychology students and researchers to overcome. The goal of the bayes4psy package is to bridge this gap and offer a collection of models and methods to be used for analysing data that arises from psychological experiments and as a teaching tool for Bayesian statistics in psychology. The package contains the Bayesian t-test and bootstrapping along with models for analysing reaction times, success rates, and tasks utilizing colors as a response. It also provides the diagnostic, analytic and visualization tools for the modern Bayesian data analysis workflow.

Human fecal contamination of water, soil, and surfaces in households sharing poor-quality sanitation facilities in Maputo, Mozambique

Identifying the origin of fecal contamination can support more effective interventions to interrupt enteric pathogen transmission. Microbial source tracking (MST) assays may help to identify environmental routes of pathogen transmission although these assays have performed poorly in highly contaminated domestic settings, highlighting the importance of both diagnostic validation and understanding the context-specific ecological, physical, and sociodemographic factors driving the spread of fecal contamination. We assessed fecal contamination of compounds (clusters of 2-10 households that share sanitation facilities) in low-income neighborhoods of urban Maputo, Mozambique, using a set of MST assays that were validated with animal stool and latrine sludge from study compounds. We sampled five environmental compartments involved in fecal microbe transmission and exposure: compound water source, household stored water and food preparation surfaces, and soil from the entrance to the compound latrine and the entrances to each household. Each sample was analyzed by culture for the general fecal indicator Escherichia coli (cEC) and by real-time PCR for the E. coli molecular marker EC23S857, human-associated markers HF183/BacR287 and Mnif, and GFD, an avian-associated marker. We collected 366 samples from 94 households in 58 compounds. At least one microbial target (indicator organism or marker gene) was detected in 96% of samples (353/366), with both E. coli targets present in the majority of samples (78%). Human targets were frequently detected in soils (59%) and occasionally in stored water (17%) but seldom in source water or on food surfaces. The avian target GFD was rarely detected in any sample type but was most common in soils (4%). To identify risk factors of fecal contamination, we estimated associations with sociodemographic, meteorological, and physical sample characteristics for each microbial target and sample type combination using Bayesian censored regression for target concentration responses and Bayesian logistic regression for target detection status. Associations with risk factors were generally weak and often differed in direction between different targets and sample types, though relationships were somewhat more consistent for physical sample characteristics. Wet soils were associated with elevated concentrations of cEC and EC23S857 and odds of detecting HF183. Water storage container characteristics that expose the contents to potential contact with hands and other objects were weakly associated with human target detection. Our results describe a setting impacted by pervasive domestic fecal contamination, including from human sources, that was largely disconnected from the observed variation in socioeconomic and sanitary conditions. This pattern suggests that in such highly contaminated settings, transformational changes to the community environment may be required before meaningful impacts on fecal contamination can be realized.

Phenome-Wide Analysis of Short- and Long-Run Disease Incidence Following Recurrent Pregnancy Loss Using Data From a 39-Year Period

Background It is unclear how recurrent pregnancy loss (RPL) impacts disease risk and whether there is a difference in risk between women with or without a live birth before RPL (primary versus secondary RPL). We investigated the disease risk following RPL, and whether there was a difference between primary and secondary RPL. Methods and Results Using population-wide healthcare registries from Denmark, we identified a cohort of 1 370 896 ever-pregnant women aged 12 to 40 years between 1977 and 2016. Of this cohort, 10 691 (0.77%) fulfilled the criteria for RPL (50.0% primary RPL). Average follow-up was 15.8 years. Incidence rate ratios were calculated in a phenome-wide manner. Diagnoses related to assessment and diagnosis of RPL and those appearing later in life were separated using a mixture model. Primary RPL increased the risk of subsequent cardiovascular disorders, including atherosclerosis, cerebral infarction, heart failure, and pulmonary embolism, as well as systemic lupus erythematosus, chronic obstructive pulmonary disease, anxiety, and obsessive-compulsive disorder. Women with secondary RPL had no increased risk of cardiovascular disorders. However, we observed an increased risk of gastrointestinal disorders such as irritable bowel syndrome and intestinal malabsorption, as well as mental disorders and obstetric complications. Conclusions RPL is a risk factor for a spectrum of disorders, which is different for primary and secondary RPL. Screening following RPL explains some associations, but the remaining findings suggest that RPL influences or shares cause with cardiovascular disorders, autoimmune disorders, and mental disorders. Research into the pathophysiology of RPL and later diseases merits further investigation.

Weyl Prior and Bayesian Statistics

When using Bayesian inference, one needs to choose a prior distribution for parameters. The well-known Jeffreys prior is based on the Riemann metric tensor on a statistical manifold. Takeuchi and Amari defined the α -parallel prior, which generalized the Jeffreys prior by exploiting a higher-order geometric object, known as a Chentsov-Amari tensor. In this paper, we propose a new prior based on the Weyl structure on a statistical manifold. It turns out that our prior is a special case of the α -parallel prior with the parameter α equaling - n , where n is the dimension of the underlying statistical manifold and the minus sign is a result of conventions used in the definition of α -connections. This makes the choice for the parameter α more canonical. We calculated the Weyl prior for univariate Gaussian and multivariate Gaussian distribution. The Weyl prior of the univariate Gaussian turns out to be the uniform prior.

A Bayesian Null Interval Hypothesis Test Controls False Discovery Rates and Improves Sensitivity in Label-Free Quantitative Proteomics

Statistical significance tests are a common feature in quantitative proteomics workflows. The Student's t-test is widely used to compute the statistical significance of a protein's change between two groups of samples. However, the t-test's null hypothesis asserts that the difference in means between two groups is exactly zero, often marking small but uninteresting fold-changes as statistically significant. Compensations to address this issue are widely used in quantitative proteomics, but we suggest that a replacement of the t-test with a Bayesian approach offers a better path forward. In this article, we describe a Bayesian hypothesis test in which the null hypothesis is an interval rather than a single point at zero; the width of the interval is estimated from population statistics. The improved sensitivity of the method substantially increases the number of truly changing proteins detected in two benchmark data sets (ProteomeXchange identifiers PXD005590 and PXD016470). The method has been implemented within FlashLFQ, an open-source software program that quantifies bottom-up proteomics search results obtained from any search tool. FlashLFQ is rapid, sensitive, and accurate and is available both as an easy-to-use graphical user interface (Windows) and as a command-line tool (Windows/Linux/OSX).

SMAUG: Analyzing single-molecule tracks with nonparametric Bayesian statistics

Single-molecule fluorescence microscopy probes nanoscale, subcellular biology in real time. Existing methods for analyzing single-particle tracking data provide dynamical information, but can suffer from supervisory biases and high uncertainties. Here, we develop a method for the case of multiple interconverting species undergoing free diffusion and introduce a new approach to analyzing single-molecule trajectories: the Single-Molecule Analysis by Unsupervised Gibbs sampling (SMAUG) algorithm, which uses nonparametric Bayesian statistics to uncover the whole range of information contained within a single-particle trajectory dataset. Even in complex systems where multiple biological states lead to a number of observed mobility states, SMAUG provides the number of mobility states, the average diffusion coefficient of single molecules in that state, the fraction of single molecules in that state, the localization noise, and the probability of transitioning between two different states. In this paper, we provide the theoretical background for the SMAUG analysis and then we validate the method using realistic simulations of single-particle trajectory datasets as well as experiments on a controlled in vitro system. Finally, we demonstrate SMAUG on real experimental systems in both prokaryotes and eukaryotes to measure the motions of the regulatory protein TcpP in Vibrio cholerae and the dynamics of the B-cell receptor antigen response pathway in lymphocytes. Overall, SMAUG provides a mathematically rigorous approach to measuring the real-time dynamics of molecular interactions in living cells.

Designing and evaluating dose-escalation studies made easy: The MoDEsT web app

BACKGROUND/AIMS

Dose-escalation studies are essential in the early stages of developing novel treatments, when the aim is to find a safe dose for administration in humans. Despite their great importance, many dose-escalation studies use study designs based on heuristic algorithms with well-documented drawbacks. Bayesian decision procedures provide a design alternative that is conceptually simple and methodologically sound, but very rarely used in practice, at least in part due to their perceived statistical complexity. There are currently very few easily accessible software implementations that would facilitate their application.

METHODS

We have created MoDEsT, a free and easy-to-use web application for designing and conducting single-agent dose-escalation studies with a binary toxicity endpoint, where the objective is to estimate the maximum tolerated dose. MoDEsT uses a well-established Bayesian decision procedure based on logistic regression. The software has a user-friendly point-and-click interface, makes changes visible in real time, and automatically generates a range of graphs, tables, and reports. It is aimed at clinicians as well as statisticians with limited expertise in model-based dose-escalation designs, and does not require any statistical programming skills to evaluate the operating characteristics of, or implement, the Bayesian dose-escalation design.

RESULTS

MoDEsT comes in two parts: a 'Design' module to explore design options and simulate their operating characteristics, and a 'Conduct' module to guide the dose-finding process throughout the study. We illustrate the practical use of both modules with data from a real phase I study in terminal cancer.

CONCLUSION

Enabling both methodologists and clinicians to understand and apply model-based study designs with ease is a key factor towards their routine use in early-phase studies. We hope that MoDEsT will enable incorporation of Bayesian decision procedures for dose escalation at the earliest stage of clinical trial design, thus increasing their use in early-phase trials.

Rethinking phosphorus-chlorophyll relationships in lakes

The empirical relationship between total phosphorus and chlorophyll has guided lake management decisions for decades, but imprecision in this relationship in individual lakes limits the utility of these models. Many environmental factors that potentially affect the total phosphorus-chlorophyll relationship have been studied, but here we hypothesize that imprecision can be reduced by considering differences in the proportions of phosphorus bound to three different "compartments" in the water column: phosphorus bound in phytoplankton, phosphorus bound to suspended sediment that is not associated with phytoplankton, and dissolved phosphorus. We specify a hierarchical Bayesian network model that estimates phosphorus associated with each compartment using field measurements of chlorophyll, total suspended solids, and total phosphorus collected from reservoirs in Missouri, USA. We then demonstrate that accounting for these different compartments yields accurate predictions of total phosphorus in individual lakes. Results from this model also yield insights into the mechanisms by which lake morphometric and watershed characteristics affect observed relationships between total phosphorus and chlorophyll.

The influence of environmental reliability in the marshmallow task: An extension study

This study is an extension of an experiment where the reliability of children's environment was manipulated before children completed the Marshmallow Task (Cognition, 2013, Vol. 126, pp. 109-114). In that experiment, Kidd, Palmeri, and Aslin found a significant difference in waiting time between two conditions in which the experimenter demonstrated reliability (by returning with promised reward) or unreliability (by not returning with rewardP). Children who had an unreliable experimenter did not wait as long during the Marshmallow Task, suggesting that delay gratification performance may be, in part, based on a rational decision. Due to the important theoretical and practical implications of this finding, we repeated the procedure of this experiment with 60 3- to 5-year-old children (twice as many as in the original study), but in a more familiar context (e.g., children's school instead of a lab). Using Bayesian analyses, we found an effect (albeit smaller than in the original study) of experimenter reliability as well as a significant gender by condition interaction effect.

Modelling the impact of MAUP on environmental drivers for Schistosoma japonicum prevalence

BACKGROUND

The modifiable areal unit problem (MAUP) arises when the support size of a spatial variable affects the relationship between prevalence and environmental risk factors. Its effect on schistosomiasis modelling studies could lead to unreliable parameter estimates. The present research aims to quantify MAUP effects on environmental drivers of Schistosoma japonicum infection by (i) bringing all covariates to the same spatial support, (ii) estimating individual-level regression parameters at 30 m, 90 m, 250 m, 500 m and 1 km spatial supports, and (iii) quantifying the differences between parameter estimates using five models.

METHODS

We modelled the prevalence of Schistosoma japonicum using sub-provinces health outcome data and pixel-level environmental data. We estimated and compared regression coefficients from convolution models using Bayesian statistics.

RESULTS

Increasing the spatial support to 500 m gradually increased the parameter estimates and their associated uncertainties. Abrupt changes in the parameter estimates occur at 1 km spatial support, resulting in loss of significance of almost all the covariates. No significant differences were found between the predicted values and their uncertainties from the five models. We provide suggestions to define an appropriate spatial data structure for modelling that gives more reliable parameter estimates and a clear relationship between risk factors and the disease.

CONCLUSIONS

Inclusion of quantified MAUP effects was important in this study on schistosomiasis. This will support helminth control programmes by providing reliable parameter estimates at the same spatial support and suggesting the use of an adequate spatial data structure, to generate reliable maps that could guide efficient mass drug administration campaigns.

Testing the magnocellular-pathway advantage in facial expressions processing for consistency over time

The ability to identify facial expressions rapidly and accurately is central to human evolution. Previous studies have demonstrated that this ability relies to a large extent on the magnocellular, rather than parvocellular, visual pathway, which is biased toward processing low spatial frequencies. Despite the generally consistent finding, no study to date has investigated the reliability of this effect over time. In the present study, 40 participants completed a facial emotion identification task (fearful, happy, or neutral faces) using facial images presented at three different spatial frequencies (low, high, or broad spatial frequency), at two time points separated by one year. Bayesian statistics revealed an advantage for the magnocellular pathway in processing facial expressions; however, no effect for time was found. Furthermore, participants' RT patterns of results were highly stable over time. Our replication, together with the consistency of our measurements within subjects, underscores the robustness of this effect. This capacity, therefore, may be considered in a trait-like manner, suggesting that individuals may possess various ability levels for processing facial expressions that can be captured in behavioral measurements.

Effects of influenza vaccination on seasonal influenza symptoms: A prospective observational study in elementary schoolchildren in Japan

Although influenza vaccine has been shown to prevent influenza symptom onset, its further beneficial effects after vaccinated individuals become symptomatic remain undetermined. This epidemiological survey compared influenza symptoms in subjects diagnosed with influenza who were and were not vaccinated. A prospective survey was performed among the 13,217 schoolchildren who attended all 29 public elementary schools in Matsumoto City, Nagano Prefecture, Japan, during the 2014/2015 influenza season. Information about symptoms and background demographic and clinical factors were obtained from a questionnaire. Of these schoolchildren, 2,548 were diagnosed with influenza and 1,122 were previously vaccinated and 1,426 were unvaccinated. Fever duration and frequency of symptoms and hospitalization were compared in vaccinated and unvaccinated children. The hospitalization rate was lower in vaccinated children, whereas symptom frequency and fever duration were similar in the two groups. This study showed that hospitalization was less in vaccinated children. Vaccination may attenuate symptom intensity after symptom onset.

Comparison of Statistical Tests and Power Analysis for Phosphoproteomics Data

Advances in protein tagging and mass spectrometry have enabled generation of large quantitative proteome and phosphoproteome data sets, for identifying differentially expressed targets in case-control studies. The power study of statistical tests is critical for designing strategies for effective target identification and control of experimental cost. Here, we develop a simulation framework to generate realistic phospho-peptide data with known changes between cases and controls. Using this framework, we quantify the performance of traditional t-tests, Bayesian tests, and the ranking-by-fold-change test. Bayesian tests, which share variance information among peptides, outperform the traditional t-tests. Although ranking-by-fold-change has similar power as the Bayesian tests, its type I error rate cannot be properly controlled without proper permutation analysis; therefore, simply relying on the ranking likely brings false positives. Two-sample Bayesian tests considering dependencies between intensity and variance are superior for data sets with complex variance. While increasing the sample size enhances the statistical tests' performance, balanced controls and cases are recommended over a one-side weighted group. Further, higher peptide standard deviations require higher fold changes to achieve the same statistical power. Together, these results highlight the importance of model-informed experimental design and principled statistical analyses when working with large-scale proteomics and phosphoproteomics data.