Empirical Bayesian Research Articles

Detecting geographical clusters of low birth weight and/or preterm birth in Japan

In Japan, mean birth weight has significantly decreased from 3152 g in 1979 to 3018 g in 2010 and the prevalence of preterm birth (PTB) has risen to 5.7% in the last thirty years. However, the presence and magnitude of geographical differences in low birthweight (LBW) and/or PTB in Japan is not well understood. We implemented spatial analysis to identify localized clusters and hot spots of LBW and/or PTB during 2012–2016. The Japan national birth database was used in this study. A total of 5,041,685 (male: 2,587,415, female: 2,454,270) births were used for spatial analysis using empirical Bayes estimates of the incidence rate of LBW and/or PTB and spatial scan tests to detect hot-spot areas with p values calculated from Monte Carlo iterations. The most and second likely clusters were located in two areas: (1) the small islands in south-west Japan (Amami and Okinawa, Relative risk = 1.09–1.67 with p < 0.001) and (2) the cities on the base of Mt. Fuji, stretching over three neighboring prefectures of Yamanashi, Shizuoka and Kanagawa (Relative risk = 1.10–1.55 with p < 0.001), respectively. We need to optimize the medical resource allocations based on the evidence in geographical clustering of LBW and/or PTB at specific locations in Japan.

Pharmacokinetic comparability between two populations using nonlinear mixed effect models: a Monte Carlo study.

'Are two populations the same or are they different' is a question that is often faced in clinical pharmacology trials e.g., a pharmacokinetic trial studying a particular drug in racially different groups. To address this question, concentration-time data were simulated from a reference and test population, where in the latter the clearance, sample size, and sampling design were systematically varied. It was of interest to determine whether the estimates of clearance from the two groups were the same or different. Two approaches were used to estimate the empirical Bayes estimates (EBEs) for clearance. One approach developed a population pharmacokinetic model for the reference population and the EBEs for the reference population were estimated from this model. The parameters of the reference population were fixed to their maximum likelihood estimates. The model was then applied to the test population dataset to estimate the EBEs of the test population using the MAXEVAL = 0 option in NONMEM. A second approach, the theta approach, combined the reference and test datasets into a single dataset and used population as a covariate in the model; the EBEs were estimated from this combined model. The power and type I error rate of each approach were calculated for each treatment combination using a variety of statistical tests to determine whether there was a difference in the distribution of the EBEs in the reference population compared to the test population. Our results suggest that either MAXEVAL or theta approaches can be used with informative sampling designs. In addition to reasonable power and type I error, both approaches gave almost identical results under a dense sampling design. To statistically compare the distribution of EBEs of pharmacokinetic parameters from a reference group to that of a test group, a T-test and DTS eCDF test are equally useful.

A Review of Crash Modification Factor (CMF) Estimation

Background Road authorities and road safety experts are involved with estimating the expected outcomes originating from road safety treatments. Information derived from proposed treatments enables planners to make comparisons between the expected savings from crash reductions and associated treatment costs. Aim The purpose of this review is to provide direction to agencies and practitioners interested in estimating safety effectiveness. Methods Specifically, this study discusses the main methods for developing CMFs, including an overview of each method, data considerations, and their strengths and weaknesses. It also discusses the techniques of estimating combined CMFs resulting from multiple safety treatments. Results The review showed that observational Before–After (BA) studies with the Empirical Bayes (EB) and Full Bayes (FB) approaches provides enhanced consistency and precision for the estimated safety effectiveness. Alternatively, the cross-sectional method can be adopted in cases where observational BA studies are not practical due to data restrictions. Five additional techniques for estimating combined CMFs are also reviewed. Conclusion The study notes that while there has been substantial research in the broad area, very few studies have reported comparative methods of combined CMF estimation. Future research directions and research gaps are also highlighted in this review.

Batch correction methods for nontarget chemical analysis data: application to a municipal wastewater collection system

Nontarget chemical analysis using high-resolution mass spectrometry has increasingly been used to discern spatial patterns and temporal trends in anthropogenic chemical abundance in natural and engineered systems. A critical experimental design consideration in such applications, especially those monitoring complex matrices over long time periods, is a choice between analyzing samples in multiple batches as they are collected, or in one batch after all samples have been processed. While datasets acquired in multiple analytical batches can include the effects of instrumental variability over time, datasets acquired in a single batch risk compound degradation during sample storage. To assess the influence of batch effects on the analysis and interpretation of nontarget data, this study examined a set of 56 samples collected from a municipal wastewater system over 7 months. Each month’s samples included 6 from sites within the collection system, one combined influent, and one treated effluent sample. Samples were analyzed using liquid chromatography high-resolution mass spectrometry in positive electrospray ionization mode in multiple batches as the samples were collected and in a single batch at the conclusion of the study. Data were aligned and normalized using internal standard scaling and ComBat, an empirical Bayes method developed for estimating and removing batch effects in microarrays. As judged by multiple lines of evidence, including comparing principal variance component analysis between single and multi-batch datasets and through patterns in principal components and hierarchical clustering analyses, ComBat appeared to significantly reduce the influence of batch effects. For this reason, we recommend the use of more, small batches with an appropriate batch correction step rather than acquisition in one large batch.Graphical abstract

Estimating the likelihood of roadway pluvial flood based on crowdsourced traffic data and depression-based DEM analysis

Abstract. Water ponding and pluvial flash flooding (PFF) on roadways can pose a significant risk to drivers. Furthermore, climate change, growing urbanization, increasing imperviousness, and aging stormwater infrastructure have increased the frequency of these events. Using physics-based models to predict pluvial flooding at the road segment scale requires notable terrain simplifications and detailed information that is often not available at fine scales (e.g., blockage of stormwater inlets). This brings uncertainty into the results, especially in highly urbanized areas where micro-topographic features typically govern the actual flow dynamics. This study evaluates the potential for flood observations collected from Waze – a community-based navigation app – to estimate the likelihood of PFF at the road segment scale. We investigated the correlation of the Waze flood reports with well-known flood observations and maps, including the National Flood Hazard Layer (NFHL), high watermarks, and low water crossings data inventories. In addition, highly localized surface depressions and their catchments are derived from a 1 m resolution bare-earth digital elevation model (BE-DEM) to investigate the spatial association of Waze flood reports. This analysis showed that the highest correlation of Waze flood reports exists with local surface depressions rather than river flooding, indicating that they are potentially useful indicators of PFF. Accordingly, two data-driven models, empirical Bayes (EB) and random forest (RF) regression, were developed to predict the frequency of flooding, a proxy for flood susceptibility, for three classes of historical storm events (light, moderate, and severe) in every road segment with surface depressions. Applying the models to Waze data from 150 storms in the city of Dallas showed that depression catchment drainage area and imperviousness are the most important predictive features. The EB model performed with reasonable precision in estimating the number of PFF events out of 92 light, 41 moderate, and 17 severe storms with 0.84, 0.85, and 1.09 mean absolute errors, respectively. This study shows that Waze data provide useful information for highly localized PFF prediction. The superior performance of EB compared to the RF model shows that the historical observations included in the EB approach are important for more accurate PFF prediction.

Longitudinal trajectories of region-level suicide mortality in Tokyo, Japan, 2011 to 2021

Background: Suicide mortality in Japan has declined over a period of more than 10 years, however, differences in longitudinal trajectories at a regional level are not well characterized. Objective was to clarify the longitudinal suicide mortality trajectories at the regional level in Tokyo from 2011 to 2021 by considering spatial smoothing, before and during the COVID-19 pandemic. Methods: This longitudinal cross-sectional analysis used fifty-four regions in Tokyo, Japan. Suicide mortality trends used data from the Cabinet Office of the Japanese government from 2011 to 2021. Regional social and environmental characteristics were used as 10 covariates. Empirical Bayes estimates for the standardized mortality ratio were obtained. A conditional autoregressive (CAR) model was applied to capture the spatial correlation for a crude and adjusted with 10 covariates using OpenBUGS. Spatial clusters were also identified by FlexScan, SaTScan, and Tango’s test. Results: Longitudinal trajectories for both males and females were similar to a decreasing trend in all Japan until 2019. In 2020, the beginning of the COVID-19 pandemic, the age-specific suicide deaths were the highest among those in their 20s. However, those were the highest among males in their 50s in 2021. The results of the CAR models adjusted for 10 covariates detected several regions as having higher suicide rates, but those regions were somewhat varied. Conclusion: During the COVID-19 pandemic, both sexes in their 20s and males in their 50s showed a tendency toward an increase in suicides. The detected regions by spatial epidemiology varied with sex.

Interferometric SAR Coherence Magnitude Estimation by Machine Learning

Current interferometric wide area ground motion services require the estimation of the coherence magnitude as accurately and computationally effectively as possible. However, a precise and at the same time computationally efficient method is missing. Therefore, the objective of this article is to improve the empirical Bayesian coherence magnitude estimation in terms of accuracy and computational cost. Precisely, this article proposes the interferometric coherence magnitude estimation by Machine Learning (ML). It results in a non-parametric and automated statistical inference. However, applying ML in this estimation context is not straightforward. The number and the domain of possible input processes is infinite and it is not possible to train all possible input signals. It is shown that the expected channel amplitudes and the expected interferometric phase cause redundancies in the input signals allowing to solve this issue. Similar to the empirical Bayesian methods, a single parameter for the maximum underlaying coherence is used to model the prior. However, no prior or any shape of prior probability is easy to implement within the ML framework. The article reports on the bias, the standard deviation and the root mean square error (RMSE) of the developed estimators. It was found that ML estimators improve the coherence estimation RMSE from small samples ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$2 N &lt; 30$</tex-math></inline-formula> ) and for small underlaying coherence compared to the conventional and empirical Bayes estimators. For three interferometric samples ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$N = 3$</tex-math></inline-formula> ) and a zero coherence magnitude, the bias related to the sample estimator improves from 0.53 to 0.39 by 27.8%. Assuming the maximum underlaying coherence is 0.6, the bias is reduced by 33.0% to 0.36 for the less strict and by 45.5% to 0.29 for the strict prior. The developed ML coherence magnitude estimators are suitable and recommended for operational InSAR systems. For the estimation, the ML model is extremely fast evaluated because no iteration, numeric integration or Bootstrapping is needed.

Spatial and Temporal Disparities of Leptospirosis Transmission in Sarawak (Malaysia), 2011-2018

This study is to analyse the spatial distribution of leptospirosis and identify its high and low incidence clusters in Sarawak. The annual incidence rate at the district level was calculated using confirmed report of leptospirosis cases from year 2011 to 2018. Empirical Bayes estimation smoothing of relative risks was used to display the spatial distribution of leptospirosis across the study region. Moran's Global Index and Local Indicators of Spatial Association (LISA) were used to analyse the existence of global and local spatial autocorrelation. Data were analysed using ArcGIS and Geoda software at the district level. The annualised average incidence reported during the study period was 20.83 per 100,000 population, with the highest cases reported in year 2018 (n = 870). The Global Moran's Index revealed spatial clustering of leptospirosis incidence in 2012 (Moran's I: 0.23), 2013 (Moran's I: 0.33), and 2014 (Moran's I: 0.45), while 2011 (Moran's I: -0.01), 2015 and 2016 (Moran's I: 0.09), 2017 (Moran's I: 0.13), and 2018 (Moran's I: 0.04) showed random patterns. High incidence clusters of leptospirosis were primarily congested in the Southeast of Sarawak, involving districts such as Kapit, Belaga, Song, Tatau and Lubok Antu. Spatial and temporal patterns of leptospirosis incidence were heterogeneous across Sarawak. This study facilitates the implementation of targeted interventions and control measures for leptospirosis in Sarawak by identifying spatial cluster and outliers.

Estimating effective connectivity in Alzheimer's disease progression: A dynamic causal modeling study.

Alzheimer's disease (AD) affects the whole brain from the cellular level to the entire brain network structure. The causal relationship among brain regions concerning the different AD stages is not yet investigated. This study used Dynamic Causal Modeling (DCM) method to assess effective connectivity (EC) and investigate the changes that accompany AD progression. We included the resting-state fMRI data of 34 AD patients, 31 late mild cognitive impairment (LMCI) patients, 34 early MCI (EMCI) patients, and 31 cognitive normal (CN) subjects selected from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. The parametric Empirical Bayes (PEB) method was used to infer the effective connectivities and the corresponding probabilities. A linear regression analysis was carried out to test if the connection strengths could predict subjects' cognitive scores. The results showed that the connections reduced from full connection in the CN group to no connection in the AD group. Statistical analysis showed the connectivity strengths were lower for later-stage patients. Linear regression analysis showed that the connection strengths were partially predictive of the cognitive scores. Our results demonstrated the dwindling connectivity accompanying AD progression on causal relationships among brain regions and indicated the potential of EC as a loyal biomarker in AD progression.

Socioeconomic Status, the Countries' Socioeconomic Development and Mental Health: Observational Evidence for Persons with Spinal Cord Injury from 22 Countries.

Objectives: Evidence on social inequalities in mental health of persons with physical impairments is limited. We therefore investigate associations of individual-level socioeconomic status (SES) and the country-level socioeconomic development (SED) with mental health in persons with spinal cord injury (SCI). Methods: We analyzed data from 12,588 participants of the International SCI Community Survey from 22 countries. To investigate individual-level inequalities, SES indicators (education, income, financial hardship, subjective status) were regressed on the SF-36 mental health index (MHI-5), stratified by countries. Country-level inequalities were analyzed with empirical Bayes estimates of random intercepts derived from linear mixed-models adjusting for individual-level SES. Results: Financial hardship and subjective status consistently predicted individual-level mental health inequalities. Country-level SED was inconsistently related to mental health when adjusting for individual-level SES. It however appeared that higher SED was associated with better mental health within higher-resourced countries. Conclusion: Reducing impoverishment and marginalization may present valuable strategies to reduce mental health inequalities in SCI populations. Investigations of country-level determinants of mental health in persons with SCI should consider influences beyond country-level SED, such as cultural factors.

Integrated human behavior and tick risk maps to prioritize Lyme disease interventions using a 'One Health' approach

Lyme disease (LD) risk is emerging rapidly in Canada due to range expansion of its tick vectors, accelerated by climate change. The risk of contracting LD varies geographically due to variability in ecological characteristics that determine the hazard (the densities of infected host-seeking ticks) and vulnerability of the human population determined by their knowledge and adoption of preventive behaviors. Risk maps are commonly used to support public health decision-making on Lyme disease, but the ability of the human public to adopt preventive behaviors is rarely taken into account in their development, which represents a critical gap. The objective of this work was to improve LD risk mapping using an integrated social-behavioral and ecological approach to: (i) compute enhanced integrated risk maps for prioritization of interventions and (ii) develop a spatially-explicit assessment tool to examine the relative contribution of different risk factors.The study was carried out in the Estrie region located in southern Québec. The blacklegged tick, Ixodes scapularis, infected with the agent of LD is widespread in Estrie and as a result, regional LD incidence is the highest in the province. LD knowledge and behaviors in the population were measured in a cross-sectional health survey conducted in 2018 reaching 10,790 respondents in Estrie. These data were used to create an index for the social-behavioral component of risk in 2018. Local Empirical Bayes estimator technique were used to better quantify the spatial variance in the levels of adoption of LD preventive activities. For the ecological risk analysis, a tick abundance model was developed by integrating data from ongoing long-term tick surveillance programs from 2007 up to 2018. Social-behavioral and ecological components of the risk measures were combined to create vulnerability index maps and, with the addition of human population densities, prioritization index maps. Map predictions were validated by testing the association of high-risk areas with the current spatial distribution of human cases of LD and reported tick exposure. Our results demonstrated that social-behavioral and ecological components of LD risk have markedly different distributions within Estrie. The occurrence of human LD cases or reported tick exposure in a municipality was positively associated with tick density and the prioritization risk index (p < 0.001).This research is a second step towards a more comprehensive integrated LD risk assessment approach, examining social-behavioral risk factors that interact with ecological risk factors to influence the management of emerging tick-borne diseases, an approach that could be applied more widely to vector-borne and zoonotic diseases.

Using Behavioral Insights to Market a Workplace Safety Program: Evidence From a Multi-Armed Experiment.

This paper describes how a multi-armed randomized experiment was used to test multiple variants of a behaviorally informed marketing strategy. In particular, we tested whether specific behavioral messages could be used to increase demand for a safety consultation service offered by the U.S. Occupational Safety and Health Administration. Our experiment used a partial factorial design with 19 study arms and a very large research sample-97,182 establishments-to test the impact of various message, formats, and delivery modes compared with an existing (not behaviorally informed) informational brochure and a no-marketing counterfactual. A secondary research goal was to predict the impact of the most successful marketing strategy (i.e., combination of message, format, and mode) so that OSHA would know what to anticipate if that strategy were implemented at scale. We used two related (but distinct) methods to address these two goals. Both begin with a common mixed (i.e., fixed and random effects) ANOVA model. We addressed the first research goal primarily from the fixed effects; we addressed the second research goal by calculating best linear unbiased predictions (BLUPs) from the full mixed model, where the BLUP involves "shrinkage" as in empirical Bayes (EB) approaches. Marketing via brochures was effective overall, nearly doubling the rate of requests for services. However, the behaviorally informed materials performed no better than OSHA's existing informational brochure. This study also highlights the conditions under which a factorial design can be used to efficiently address questions about which of several program variants are most effective.

A Method for Locational Risk Estimation of Vehicle-Children Accidents Considering Children's Travel Purposes.

The reduction in locational traffic accident risks through appropriate traffic safety management is important to support, maintain, and improve children's safe and independent mobility. This study proposes and verifies a method to evaluate the risk of elementary school students-vehicle accidents (ESSVAs) at individual intersections on residential roads in Toyohashi city, Japan, considering the difference in travel purposes (i.e., school commuting purpose; SCP or non-school commuting purpose: NSCP), based on a statistical regression model and Empirical Bayes (EB) estimation. The results showed that the ESSVA risk of children's travel in SCP is lower than that in NSCP, and not only ESSVAs in SCP but also most ESSVAs in NSCP occurred on or near the designated school routes. Therefore, it would make sense to implement traffic safety management and measures focusing on school routes. It was also found that the locational ESSVA risk structure is different depending on whether the purpose of the children's travels is SCP or NSCP in the statistical model. Finally, it was suggested that evaluation of locational ESSVA risks based on the EB estimation is useful for efficiently extracting locations where traffic safety measures should be implemented compared to that only based on the number of accidents in the past.

Complete chloroplast genome structure of four Ulmus species and Hemiptelea davidii and comparative analysis within Ulmaceae species

In this study, the chloroplast (cp) genomes of Hemiptelea davidii, Ulmus parvifolia, Ulmus lamellosa, Ulmus castaneifolia, and Ulmus pumila ‘zhonghuajinye’ were spliced, assembled and annotated using the Illumina HiSeq PE150 sequencing platform, and then compared to the cp genomes of other Ulmus and Ulmaceae species. The results indicated that the cp genomes of the five sequenced species showed a typical tetrad structure with full lengths ranging from 159,113 to 160,388 bp. The large single copy (LSC), inverted repeat (IR), and small single copy (SSC) lengths were in the range of 87,736–88,466 bp, 26,317–26,622 bp and 18,485–19,024 bp, respectively. A total of 130–131 genes were annotated, including 85–86 protein-coding genes, 37 tRNA genes and eight rRNA genes. The GC contents of the five species were similar, ranging from 35.30 to 35.62%. Besides, the GC content was different in different region and the GC content in IR region was the highest. A total of 64-133 single sequence repeat (SSR) loci were identified among all 21 Ulmaceae species. The (A)n and (T)n types of mononucleotide were highest in number, and the lengths were primarily distributed in 10–12 bp, with a clear AT preference. A branch-site model and a Bayes Empirical Bayes analysis indicated that the rps15 and rbcL had the positive selection sites. Besides, the analysis of mVISTA and sliding windows got a lot of hotspots such as trnH/psbA, rps16/trnQ, trnS/trnG, trnG/trnR and rpl32/trnL, which could be utilized as potential markers for the species identification and phylogeny reconstruction within Ulmus in the further studies. Moreover, the evolutionary tree of Ulmaceae species based on common protein genes, whole cp genome sequences and common genes in IR region of the 23 Ulmaceae species were constructed using the ML method. The results showed that these Ulmaceae species were divided into two branches, one that included Ulmus, Zelkova and Hemiptelea, among which Hemiptelea was the first to differentiate and one that included Celtis, Trema, Pteroceltis, Gironniera and Aphananthe. Besides, these variations found in this study could be used for the classification, identification and phylogenetic study of Ulmus species. Our study provided important genetic information to support further investigations into the phylogenetic development and adaptive evolution of Ulmus and Ulmaceae species.

An empirical Bayes approach to improving population-specific genetic association estimation by leveraging cross-population data.

Populations of non-European ancestry are substantially underrepresented in genome-wide association studies (GWAS). As genetic effects can differ between ancestries due to possibly different causal variants or linkage disequilibrium patterns, a meta-analysis that includes GWAS of all populations yields biased estimation in each of the populations and the bias disproportionately impacts non-European ancestry populations. This is because meta-analysis combines study-specific estimates with inverse variance as the weights, which causes biases towards studies with the largest sample size, typical of the European ancestry population. In this paper, we propose two empirical Bayes (EB) estimators to borrow the strength of information across populations although accounting for between-population heterogeneity. Extensive simulation studies show that the proposed EB estimators are largely unbiased and improve efficiency compared to the population-specific estimator. In contrast, even though the meta-analysis estimator has a much smaller variance, it yields significant bias when the genetic effect is heterogeneous across populations. We apply the proposed EB estimators to a large-scale trans-ancestry GWAS of stroke and demonstrate that the EB estimators reduce the variance of the population-specific estimator substantially, with the effect estimates close to the population-specific estimates.

Assessment of the Impact of the Implementation of a Pre-Hospital Ambulance System on Acute Myocardial Infarction Mortality in a Developing Country.

The effective management of patients with acute myocardial infarction (AMI) is time-dependent. To assess the impacts of the implementation of prehospital care on admission rates and mortality associated with AMI. Retrospective, ecological study, which assessed data from the Brazilian Universal Health System, from all 853 municipalities of Minas Gerais, from 2008 to 2016. Excessive skewness of general and in-hospital mortality rates was smoothed using the empirical Bayes method. This study assessed the relationship between Mobile Emergency Care Service (SAMU) in each municipality and the following 3 outcomes: mortality rate due to AMI, AMI in-hospital mortality, and AMI hospitalization rate, using the Poisson hierarchical model. Rates were corrected by age structure and detrended by seasonality and temporal influences. A confidence interval of 95% was adopted. AMI mortality rates decreased throughout the study, on average 2% per year, with seasonal variation. AMI in-hospital mortality also showed a decreasing trend, from 13.81% in 2008 to 11.43% in 2016. SAMU implementation was associated with decreased AMI mortality (odds ratio [OR] = 0.967, 95% confidence interval [CI] 0.936 to 0.998) and AMI in-hospital mortality (OR = 0.914, 95% CI 0.845 to 0.986), with no relation with hospitalizations (OR = 1.003, 95% CI 0.927 to 1.083). SAMU implementation was associated with a modest but significant decrease in AMI in-hospital mortality. This finding reinforces the key role of prehospital care in AMI care and the need for investments on this service to improve clinical outcomes in low- and middle-income countries.

Variable selection in linear-circular regression models

ABSTRACT Applications of circular regression models are ubiquitous in many disciplines, particularly in meteorology, biology and geology. In circular regression models, variable selection problem continues to be a remarkable open question. In this paper, we address variable selection in linear-circular regression models where uni-variate linear dependent and a mixed set of circular and linear independent variables constitute the data set. We consider Bayesian lasso which is a popular choice for variable selection in classical linear regression models. We show that Bayesian lasso in linear-circular regression models is not able to produce robust inference as the coefficient estimates are sensitive to the choice of hyper-prior setting for the tuning parameter. To eradicate the problem, we propose a robustified Bayesian lasso that is based on an empirical Bayes (EB) type methodology to construct a hyper-prior for the tuning parameter while using Gibbs Sampling. This hyper-prior construction is computationally more feasible than the hyper-priors that are based on correlation measures. We show in a comprehensive simulation study that Bayesian lasso with EB-GS hyper-prior leads to a more robust inference. Overall, the method offers an efficient Bayesian lasso for variable selection in linear-circular regression while reducing model complexity.

An Evaluation of Gene-Diet Interaction Statistical Methods and Discovery of rs7175421-Whole Grain Interaction in Lung Cancer

Dietary factors show different effects on genetically diverse populations. Scientific research uses gene-environment interaction models to study the effects of dietary factors on genetically diverse populations for lung cancer risk. However, previous study designs have not investigated the degree of type I error inflation and, in some instances, have not corrected for multiple testing. Using a motivating investigation of diet-gene interaction and lung cancer risk, we propose a training and testing strategy and perform real-world simulations to select the appropriate statistical methods to reduce false-positive discoveries. The simulation results show that the unconstrained maximum likelihood (UML) method controls the type I error better than the constrained maximum likelihood (CML). The empirical Bayesian (EB) method can compete with the UML method in achieving statistical power and controlling type I error. We observed a significant interaction between SNP rs7175421 with dietary whole grain in lung cancer prevention, with an effect size (standard error) of −0.312 (0.112) for EB estimate. SNP rs7175421 may interact with dietary whole grains in modulating lung cancer risk. Evaluating statistical methods for gene-diet interaction analysis can help balance the statistical power and type I error.

Safety Evaluation of Changing Speed Limit from 55 mph to 60 mph on Two-Lane, Two-Way Road Segments

This paper describes the efforts to evaluate the safety impacts of increasing the speed limit from 55 mph to 60 mph on selected two-lane, two-way state highway road segments in Minnesota, U.S. An empirical Bayes (EB) before–after analysis was used to estimate crash modification factors (CMFs) for both segments (1,909.11 mi) and intersections (1,722 3-leg and 1,191 4-leg). Aggregate analysis conducted using all the segment and intersection data showed a 2.9% increase in total crashes, a 2.5% increase in injury (KABC) crashes, and a 0.05% reduction in the injury (KAB) crashes. These results—along with before-and-after operating speed data from another study by Minnesota Department of Transportation (MnDOT) (2019) showing that the 85th percentile operating speed remained the same and that the mean operating speeds increased by 1 mph following the speed limit increase—can lead to a conclusion that the speed limit increase from 55 mph to 60 mph had a minor effect on combined segment and intersection crashes or operating speeds. It is important to note that these results are specific to the corridors that were selected by MnDOT for the increase in speed limit; caution must be exercised when extending these to systemwide increases in speed limits in Minnesota or in other states, and when estimating long-term effects of speed limit increases as operating speeds can change over a longer period of time.

A geographical study on amyotrophic lateral sclerosis in Rio Grande Do Norte, Brazil, from 2005 to 2018

Objective: amyotrophic lateral sclerosis (ALS) is a rare and fatal neurodegenerative disorder with variable incidence and prevalence worldwide. However, clinical-epidemiological studies on ALS are scarce in Brazil. Thus, this study investigated whether ALS incidence had uniform spatial distribution in population-based cluster analysis in Rio Grande do Norte state (Brazil), from 2005 to 2018. Methods: new cases of ALS were identified in a database of the ALS multidisciplinary care center of the Onofre Lopes University Hospital in Natal (Rio Grande do Norte, Brazil). Approaches were based on incidence (empirical Bayes estimator and Moran's I analysis) and cluster analyses (Moran scatter plot and spatial correlogram). Results: a total of 177 patients (59% males) participated in the study; the mean age of ALS onset was 57 years. Mean annual incidence of ALS was 0.3769 per 100,000 inhabitants (95% confidence interval of 0.0889), higher in males than in females (0.4516 per 100,000 vs. 0.3044 per 100,000). According to spatial statistics, patients were homogeneously distributed throughout the studied area. Conclusion: a low estimate was observed compared with other populations. Results did not indicate areas of increased risk or significant spatial geographic dependence, suggesting a random ALS incidence in Rio Grande do Norte.