Meta-analysis review for pilot and large-scale constructed wetlands: Design parameters, treatment performance, and influencing factors

Despite their longstanding use in environmental remediation, constructed wetlands (CWs) are still topical due to their sustainable and nature-based approach. While research and review publications have grown annually by 7.5 % and 37.6 %, respectively, from 2018 to 2022, a quantitative meta-analysis employing advanced statistics and machine learning to assess CWs has not yet been conducted. Further, traditional statistics of mean ± standard deviation could not convey the extent of confidence or uncertainty in results from CW studies. This study employed a 95 % bootstrap-based confidence interval and out-of-bag Random Forest-based driver analysis on data from 55 studies, totaling 163 cases of pilot and full-scale CWs. The study recommends, with 95 % confidence, median surface hydraulic loading rates (HLR) of 0.14 [0.11, 0.17] m/d for vertical flow-CWs (VF) and 0.13 [0.07, 0.22] m/d for horizontal flow-CWs (HF), and hydraulic retention time (HRT) of 125.14 [48.0, 189.6] h for VF, 72.00 [42.00, 86.28] h for HF, as practical for new CW design. Permutation importance results indicate influent COD impacted primarily on COD removal rate at 21.58 %, followed by HLR (16.03 %), HRT (12.12 %), and substrate height (H) (10.90 %). For TN treatment, influent TN and COD were the most significant contributors at 12.89 % and 10.01 %, respectively, while H (9.76 %), HRT (9.72 %), and HLR (5.87 %) had lower impacts. Surprisingly, while HRT and H had a limited effect on COD removal, they substantially influenced TN. This study sheds light on CWs' performance, design, and control factors, guiding their operation and optimization.

Comparisons of various estimates of the I2 statistic for quantifying between-study heterogeneity in meta-analysis

Assessing heterogeneity between studies is a critical step in determining whether studies can be combined and whether the synthesized results are reliable. The I 2 statistic has been a popular measure for quantifying heterogeneity, but its usage has been challenged from various perspectives in recent years. In particular, it should not be considered an absolute measure of heterogeneity, and it could be subject to large uncertainties. As such, when using I 2 to interpret the extent of heterogeneity, it is essential to account for its interval estimate. Various point and interval estimators exist for I 2 . This article summarizes these estimators. In addition, we performed a simulation study under different scenarios to investigate preferable point and interval estimates of I 2 . We found that the Sidik-Jonkman method gave precise point estimates for I 2 when the between-study variance was large, while in other cases, the DerSimonian-Laird method was suggested to estimate I 2 . When the effect measure was the mean difference or the standardized mean difference, the Q -profile method, the Biggerstaff-Jackson method, or the Jackson method was suggested to calculate the interval estimate for I 2 due to reasonable interval length and more reliable coverage probabilities than various alternatives. For the same reason, the Kulinskaya-Dollinger method was recommended to calculate the interval estimate for I 2 when the effect measure was the log odds ratio.

Meta-analysis of identified genomic regions and candidate genes underlying salinity tolerance in rice (Oryza sativa L.)

Rice output has grown globally, yet abiotic factors are still a key cause for worry. Salinity stress seems to have the more impact on crop production out of all abiotic stresses. Currently one of the most significant challenges in paddy breeding for salinity tolerance with the help of QTLs, is to determine the QTLs having the best chance of improving salinity tolerance with the least amount of background noise from the tolerant parent. Minimizing the size of the QTL confidence interval (CI) is essential in order to primarily include the genes responsible for salinity stress tolerance. By considering that, a genome-wide meta-QTL analysis on 768 QTLs from 35 rice populations published from 2001 to 2022 was conducted to identify consensus regions and the candidate genes underlying those regions responsible for the salinity tolerance, as it reduces the confidence interval (CI) to many folds from the initial QTL studies. In the present investigation, a total of 65 MQTLs were extracted with an average CI reduced from 17.35 to 1.66 cM including the smallest of 0.01 cM. Identification of the MQTLs for individual traits and then classifying the target traits into correlated morphological, physiological and biochemical aspects, resulted in more efficient interpretation of the salinity tolerance, identifying the candidate genes and to understand the salinity tolerance mechanism as a whole. The results of this study have a huge potential to improve the rice genotypes for salinity tolerance with the help of MAS and MABC.

Bias in Laboratory Medicine: The Dark Side of the Moon

Physicians increasingly use laboratory-produced information for disease diagnosis, patient monitoring, treatment planning, and evaluations of treatment effectiveness. Bias is the systematic deviation of laboratory test results from the actual value, which can cause misdiagnosis or misestimation of disease prognosis and increase healthcare costs. Properly estimating and treating bias can help to reduce laboratory errors, improve patient safety, and considerably reduce healthcare costs. A bias that is statistically and medically significant should be eliminated or corrected. In this review, the theoretical aspects of bias based on metrological, statistical, laboratory, and biological variation principles are discussed. These principles are then applied to laboratory and diagnostic medicine for practical use from clinical perspectives.

Methods for two nonstandard problems arising from the Luria-Delbrück experiment

The fluctuation experiment, devised by Luria and Delbrück in 1943, remains the method of choice for measuring microbial mutation rates in the laboratory. While most inference problems commonly encountered in a fluctuation experiment can be tackled by existing standard algorithms, investigators from time to time run into nonstandard problems not amenable to any existing algorithms. A major obstacle to solving these nonstandard problems is the construction of confidence intervals for mutation rates. This note describes methods for two important categories of nonstandard problems, namely, pooling data from separate experiments and analyzing grouped mutant count data, focusing on the construction of likelihood ratio confidence intervals. In addition to illustrative examples using real-world data, simulation results are presented to help assess the proposed methods.

An analytical approach to confidence interval estimation of river microplastic sampling

Microplastics (MPs), plastic particles <5 mm in diameter, are emerging ubiquitous pollutants in natural environments, including freshwater ecosystems. As rivers facilitate efficient transport among landscapes, monitoring is crucial for elucidating the origin, dynamics, and fate of MPs. However, standardized methodologies for in situ sampling in freshwater environments remain undefined to date. Specifically, evaluating the sampling error of MP concentration estimates is crucial for comparing results among studies. This study proposes a novel method for computing confidence intervals (CIs) from a single estimate of numerical concentration (expressed in particles·m-3). MPs are expected to disperse according to purely random processes, such as turbulent diffusion, and to consequently exhibit a random distribution pattern that follows a Poisson point process. Accordingly, the present study introduced a framework based on the Poisson point process to compute CIs, which were validated using MP samples from two urban rivers in Chiba, Japan, obtained using a mesh with an opening size of 335 μm. Random number simulations revealed that the CIs were applicable when ≥10 MPs were present in a sample. Further, when ≥50 MPs were present in a sample, the sampling error (95% CI) was within ±30% of the concentration estimates. The proposed framework allows for the intercomparison of single river MP samples despite the lack of sample replicates. Further, the present study emphasizes that the volume of sampled river water is the only controllable parameter that can reduce the sampling error.

A comparison of the methods for detecting dyadic patterns in the actor-partner interdependence model

In the actor-partner interdependence model (APIM), various dyadic patterns between an actor and partner can be examined. One widely used approach is the parameter k method, which tests whether the ratio of the partner effect to the actor effect (p/a) is significantly different from pattern values such as -1 (contrast), 0 (actor-only or partner-only), and 1 (couple). Although using a phantom variable was a useful method for estimating the k ratio, it is no longer necessary due to the availability of statistical packages that allow for a direct estimation of the k ratio without the inclusion of the phantom variable. Moreover, it is possible to examine the patterns by testing new variables defined in different forms from the k or using the χ2 difference test. To date, no previous studies have evaluated and compared the various approaches for detecting the dyadic patterns in APIM. This study aims to assess and compare the performance of four different methods for detecting dyadic patterns: (1) phantom variable approach, (2) direct estimation of the parameter k, (3) new-variable approach, and (4) χ2 difference test. The first two methods frequently included multiple pattern values in there confidence interval. Furthermore, the phantom variable approach was prone to convergence issues. The other two alternatives performed better in detecting the dyadic patterns without convergence problems. Given the findings of the study, we suggest a novel procedure for examining dyadic patterns in APIM.

Empirical methods for the validation of time-to-event mathematical models taking into account uncertainty and variability: application to EGFR + lung adenocarcinoma

BACKGROUND

Over the past several decades, metrics have been defined to assess the quality of various types of models and to compare their performance depending on their capacity to explain the variance found in real-life data. However, available validation methods are mostly designed for statistical regressions rather than for mechanistic models. To our knowledge, in the latter case, there are no consensus standards, for instance for the validation of predictions against real-world data given the variability and uncertainty of the data. In this work, we focus on the prediction of time-to-event curves using as an application example a mechanistic model of non-small cell lung cancer. We designed four empirical methods to assess both model performance and reliability of predictions: two methods based on bootstrapped versions of parametric statistical tests: log-rank and combined weighted log-ranks (MaxCombo); and two methods based on bootstrapped prediction intervals, referred to here as raw coverage and the juncture metric. We also introduced the notion of observation time uncertainty to take into consideration the real life delay between the moment when an event happens, and the moment when it is observed and reported.

RESULTS

We highlight the advantages and disadvantages of these methods according to their application context. We have shown that the context of use of the model has an impact on the model validation process. Thanks to the use of several validation metrics we have highlighted the limit of the model to predict the evolution of the disease in the whole population of mutations at the same time, and that it was more efficient with specific predictions in the target mutation populations. The choice and use of a single metric could have led to an erroneous validation of the model and its context of use.

CONCLUSIONS

With this work, we stress the importance of making judicious choices for a metric, and how using a combination of metrics could be more relevant, with the objective of validating a given model and its predictions within a specific context of use. We also show how the reliability of the results depends both on the metric and on the statistical comparisons, and that the conditions of application and the type of available information need to be taken into account to choose the best validation strategy.

Dynamic Model Selection and Optimal Batch Design for Polyhydroxyalkanoate (PHA) Production by Cupriavidus necator

Mathematical modelling of microbial polyhydroxyalkanoates (PHAs) production is essential to develop optimal bioprocess design. Though the use of mathematical models in PHA production has increased over the years, the selection of kinetics and model identification strategies from experimental data remains largely heuristic. In this study, PHA production from Cupriavidus necator utilizing sucrose and urea was modelled using a parametric discretization approach. Product formation kinetics and relevant parameters were established from urea-free experimental sets, followed by the selection of growth models from a batch containing both sucrose and urea. Logistic growth and Luedeking-Piret model for PHA production was selected based on regression coefficient (R2: 0.941), adjusted R2 (0.930) and AICc values (-42.764). Model fitness was further assessed through cross-validation, confidence interval and sensitivity analysis of the parameters. Model-based optimal batch startup policy, incorporating multi-objective desirability, suggests an accumulation of 2.030 g l-1 of PHA at the end of 120 h. The modelling framework applied in this study can be used not only to avoid over-parameterization and identifiability issues but can also be adopted to design optimal batch startup policies.

Estimation of confidence intervals for quantitation of coeluted peaks in liquid chromatography-Photodiode array detection through a combination of multivariate curve resolution-alternating least-square and Bayesian inference techniques

There is a dramatic increase in drug candidates that exhibit complex structures and do not comply with Lipinski's rule of five. One of the most critical and complex technical challenges in the quality control of such drug candidates is the control of analogous substances contained in active pharmaceutical ingredients and related formulations. Although the development of ultrahigh-performance liquid chromatography and high-performance columns has improved efficiency per unit time, the difficulty of peak separation to quantify impurities with similar structures and physicochemical properties continues to rise, and so does the probability of failure to achieve the necessary separation. Coeluting peaks observed in the case of high-performance liquid chromatography (HPLC) with photodiode array detection can be separated using the multivariate curve resolution-alternating least-square (MCR-ALS) method exploiting differences in analyte UV spectra. However, relatively large quantitation errors have been observed for coeluting analogous substances, and the reliability of the corresponding quantitative data requires improvement. Herein, Bayesian inference is applied to separation by the MCR-ALS method to develop an algorithm assigning a confidence interval to the quantitative data of each analogous substance. The usefulness and limitations of this approach are tested using two analogs of telmisartan as models. For this test, a simulated two-component HPLC-UV dataset with an intensity ratio (relative to the main peak) of 0.1-1.0 and a resolution of 0.5-1.0 is used. The developed algorithm allows the prediction confidence interval, including the true value, to be assigned to the peak area in almost all cases, even when the intensity ratio, resolution, and signal-to-noise ratio are changed. Finally, the developed algorithm is also evaluated on a real HPLC-UV dataset to confirm that reasonable prediction confidence intervals including true values are assigned to peak areas. In addition to allowing the separation and quantitation of substances such as impurities challenging to separate by HPLC in a scientifically valid manner, which is impossible for conventional HPLC-UV detection, our method can assign confidence intervals to quantitative data. Therefore, the adopted approach is expected to resolve the issues associated with assessing impurities in the quality control of pharmaceuticals.

Continuity corrected Wilson interval for the difference of two independent proportions

Confidence interval for the difference of two proportions has been studied for decades. Many methods were developed to improve the approximation of the limiting distribution of test statistics, such as the profile likelihood method, the score method, and the Wilson method. For the Wilson interval developed by Beal (1987), the approximation of the Z test statistic to the standard normal distribution may be further improved by utilizing the continuity correction, in the observation of anti-conservative intervals from the Wilson interval. We theoretically prove that the Wilson interval is nested in the continuity corrected Wilson interval under mild conditions. We compare the continuity corrected Wilson interval with the commonly used methods with regards to coverage probability, interval width, and mean squared error of coverage probability. The proposed interval has good performance in many configurations. An example from a Phase II cancer trial is used to illustrate the application of these methods.

Problems and alternatives of testing significance using null hypothesis and P-value in food research

A testing method to identify statistically significant differences by comparing the significance level and the probability value based on the Null Hypothesis Significance Test (NHST) has been used in food research. However, problems with this testing method have been discussed. Several alternatives to the NHST and the P-value test methods have been proposed including lowering the P-value threshold and using confidence interval (CI), effect size, and Bayesian statistics. The CI estimates the extent of the effect or difference and determines the presence or absence of statistical significance. The effect size index determines the degree of effect difference and allows for the comparison of various statistical results. Bayesian statistics enable predictions to be made even when only a small amount of data is available. In conclusion, CI, effect size, and Bayesian statistics can complement or replace traditional statistical tests in food research by replacing the use of NHST and P-value.

Algorithm for the early prediction of drug stability using bayesian inference and multiple measurements: Application for predicting the stability of silodosin tablets

The stability of active pharmaceutical ingredients (APIs) and formulations has become a major chemistry, manufacturing, and control (CMC) concern in the pharmaceutical industry because it can determine the feasibility of research and development, the development period, and the development costs of a certain formulation. To streamline the research and development of pharmaceutical products and create useful pharmaceutical products at an early stage, a technology that predicts the stability of formulations at an early stage and with a high degree of accuracy is needed. When predicting the stability of a substance, highly reliable data are required; however, the stability data are affected by analytical variations that depend on the experimenter, measurement device, and conditions used. Although these variations greatly affect the prediction accuracy, a stability prediction method that considers these variations has not yet been developed. Here, short-term stability data under accelerated conditions were obtained at three institutions using silodosin tablets as a model sample. By combining Bayesian inference with the temporal change in the amount of the main degradation products obtained and the conventional humidity-corrected Arrhenius equation, we developed a new algorithm that provides a narrow confidence interval, even when using data with variations. By using this algorithm and setting an appropriate number of conditions, we were able to obtain a valid confidence intervals in a short period of time. Here, by performing more measurements than those suggested by the minimum measurement frequency indicated in the guideline specified in the International Council for Harmonisation (ICH) of Technical Requirements for Pharmaceuticals for Human Use, we developed a method that can be used to reasonably predict the long-term stability of the drugs, even if the data measurement interval is short. Our results will help solve various problems in today's pharmaceutical product development scenario and contribute to worldwide health and welfare.

Association Between Biological Lead Concentrations and Autism Spectrum Disorder (ASD) in Children: a Systematic Review and Meta-Analysis

Studies have been conducted in different countries of the world to illustrate a link between autism spectrum disorder (ASD) and lead (Pb) in different specimens such as hair, blood, and urine. Therefore, we carried out a systematic review and meta-analysis to determine the association between Pb concentration in biological samples (blood, urine, and hair) and ASD in children through case-control and cross-sectional studies. In this systematic review, PubMed, Web of Sciences, Scopus, and Google Scholar were searched for relevant studies from January 2000 to February 2022. A random-effects model was used to pool the results. The effect sizes were standardized mean differences (proxied by Hedges' g) followed by a 95% confidence interval. Pooling data under the random effect model from blood and hair studies showed a significant difference between the children in the ASD group and the control group in blood lead level (Hedges' g: 1.21, 95% CI: 0.33-2.09, P = 0.01) and hair level (Hedges' g: 2.20, 95% CI: 0.56-3.85, P = 0.01). For urine studies, pooling data under the random effect model from eight studies indicated no significant difference between the children in the ASD group and control group in urinary lead level (Hedges' g: - 0.34, 95% CI: - 1.14,0.45, P = 0.40). Moreover, the funnel plot and the results of the Egger test for the blood and urine samples showed no publication bias, while, for the hair samples, the funnel plot illustrated the existence of publication bias.

Admission Heart Rate and Mortality in Critically Ill Patients with Acute Aortic Dissection

The association between admission heart rate (HR) and the mortality of critically ill patients with acute aortic dissection (AAD) remains unclear.The data were extracted from the Medical Information Mart for Intensive Care (MIMIC-III) database. Cox regression models and Kaplan-Meier (KM) survival curve were used to explore the association between admission HR and 90-day, 1-year, and 3-year mortality in patients with AAD. Sensitivity analyses were conducted to assess potential bias.A total of 374 eligible AAD patients were included and divided in 4 groups according to admission HR (HR ≤ 70, 71-80, 81-90, and > 90 beats per minute (bpm) ). The patients with AAD in the group with HR > 90 bpm had higher 90-day, 1-year, and 3-year mortality than those in the groups with HR ≤ 70, 71-80, and 81-90 bpm. After adjusting for age, sex, BMI, systolic blood pressure, diastolic blood pressure, SOFA score, SAPSII score, Stanford type, hypertension, coronary artery disease, liver disease, atrial fibrillation, valvular disease, intensive care unit mechanical ventilation, aortic surgery, and thoracic endovascular aortic repair, patients with admission HR > 90 bpm had a higher risk of 90-day, 1-year, and 3-year mortality [adjusted hazard ratio, 95% confidence interval, 5.14 (2.22-11.91) P < 0.001; 4.31 (2.10-8.84) P < 0.001; 3.01 (1.66-5.46) P < 0.001] than those with HR 81-90 bpm. The 90-day, 1-year, and 3-year mortality were similar among the groups with HR ≤ 70, 71-80, and 81-90 bpm.Admission HR > 90 bpm was independently associated with all-cause mortality in critically ill AAD patients, either type A or B aortic dissection.

The bias for statistical significance in sport and exercise medicine

OBJECTIVES

We aimed to examine the bias for statistical significance using published confidence intervals in sport and exercise medicine research.

DESIGN

Observational study.

METHODS

The abstracts of 48,390 articles, published in 18 sports and exercise medicine journals between 2002 and 2022, were searched using a validated text-mining algorithm that identified and extracted ratio confidence intervals (odds, hazard, and risk ratios). The algorithm identified 1744 abstracts that included ratio confidence intervals, from which 4484 intervals were extracted. After excluding ineligible intervals, the analysis used 3819 intervals, reported as 95 % confidence intervals, from 1599 articles. The cumulative distributions of lower and upper confidence limits were plotted to identify any abnormal patterns, particularly around a ratio of 1 (the null hypothesis). The distributions were compared to those from unbiased reference data, which was not subjected to p-hacking or publication bias. A bias for statistical significance was further investigated using a histogram plot of z-values calculated from the extracted 95 % confidence intervals.

RESULTS

There was a marked change in the cumulative distribution of lower and upper bound intervals just over and just under a ratio of 1. The bias for statistical significance was also clear in a stark under-representation of z-values between -1.96 and +1.96, corresponding to p-values above 0.05.

CONCLUSIONS

There was an excess of published research with statistically significant results just below the standard significance threshold of 0.05, which is indicative of publication bias. Transparent research practices, including the use of registered reports, are needed to reduce the bias in published research.

Propagating uncertainty about molecular evolution models and prior distributions to phylogenetic trees

Phylogenetic trees constructed from molecular sequence data rely on largely arbitrary assumptions about the substitution model, the distribution of substitution rates across sites, the version of the molecular clock, and, in the case of Bayesian inference, the prior distribution. Those assumptions affect results reported in the form of clade probabilities and error bars on divergence times and substitution rates. Overlooking the uncertainty in the assumptions leads to overly confident conclusions in the form of inflated clade probabilities and short confidence intervals or credible intervals. This paper demonstrates how to propagate that uncertainty by combining the models considered along with all of their assumptions, including their prior distributions. The combined models incorporate much more of the uncertainty than Bayesian model averages since the latter tend to settle on a single model due to the higher-level assumption that one of the models is true. Nucleotide sequence data illustrates the proposed model combination method.

A Statistical Approach for Assessing the Compliance of Integrated Continuous Glucose Monitoring Systems with Food and Drug Administration Accuracy Requirements

To assess the compliance of "integrated" continuous glucose monitoring (CGM) systems with U.S. Food and Drug Administration requirements, the calculation of confidence intervals (CIs) on agreement rates (ARs), that is, the percentage of CGM measurements lying within a certain deviation of a comparator method, is stipulated. However, despite the existence of numerous approaches that could yield different results, a specific procedure for calculating CIs is not described anywhere. This report, therefore, proposes a suitable statistical procedure to allow transparency and comparability between CGM systems. Three existing methods were applied to six data sets from different CGM performance studies. The results indicate that a bootstrap-based method that accounts for the clustered structure of CGM data is reliable and robust. We thus recommend its use for the estimation of CIs of ARs. A software implementation of the proposed method is freely available (https://github.com/IfDTUlm/CGM_Performance_Assessment).

Speeding up interval estimation for R2-based mediation effect of high-dimensional mediators via cross-fitting

Mediation analysis is a useful tool in biomedical research to investigate how molecular phenotypes, such as gene expression, mediate the effect of an exposure on health outcomes. However, commonly used mean-based total mediation effect measures may suffer from cancellation of component-wise mediation effects of opposite directions in the presence of high-dimensional omics mediators. To overcome this limitation, a variance-based R-squared total mediation effect measure has been recently proposed, which, nevertheless, relies on the computationally intensive nonparametric bootstrap for confidence interval estimation. In this work, we formulate a more efficient two-stage cross-fitted estimation procedure for the R-squared measure. To avoid potential bias, we perform iterative Sure Independence Screening (iSIS) in two subsamples to exclude the non-mediators, followed by ordinary least squares (OLS) regressions for the variance estimation. We then construct confidence intervals based on the newly-derived closed-form asymptotic distribution of the R-squared measure. Extensive simulation studies demonstrate that the proposed procedure is hundreds of times more computationally efficient than the resampling-based method with comparable coverage probability. Furthermore, when applied to the Framingham Heart Study, the proposed method replicated the established finding of gene expression mediating age-related variation in systolic blood pressure and discovered the role of gene expression profiles in the relationship between sex and high-density lipoprotein cholesterol. The proposed cross-fitted interval estimation procedure is implemented in R package RsqMed.

Clinical trials evaluating desensitising agents. Some design and analysis issues

INTRODUCTION

The purpose of this short communication is to draw attention to an efficient design for trials to evaluate desensitising agents, and an appropriate statistical analysis.

METHODS

Two recent sensitivity trials conducted by the Bristol Dental School Clinical Trials Group are reviewed.

RESULTS

The methodology used was effective to establish efficacy of the products evaluated.

CONCLUSIONS

This methodology is recommended for wider use.

CLINICAL SIGNIFICANCE

Effective clinical trial methodology enables establishment of efficacy of desensitising products leading to patient benefit.

Double bias correction for high-dimensional sparse additive hazards regression with covariate measurement errors

We propose an inferential procedure for additive hazards regression with high-dimensional survival data, where the covariates are prone to measurement errors. We develop a double bias correction method by first correcting the bias arising from measurement errors in covariates through an estimating function for the regression parameter. By adopting the convex relaxation technique, a regularized estimator for the regression parameter is obtained by elaborately designing a feasible loss based on the estimating function, which is solved via linear programming. Using the Neyman orthogonality, we propose an asymptotically unbiased estimator which further corrects the bias caused by the convex relaxation and regularization. We derive the convergence rate of the proposed estimator and establish the asymptotic normality for the low-dimensional parameter estimator and the linear combination thereof, accompanied with a consistent estimator for the variance. Numerical experiments are carried out on both simulated and real datasets to demonstrate the promising performance of the proposed double bias correction method.

New Confidence Intervals for Relative Risk of Two Correlated Proportions

Biomedical studies, such as clinical trials, often require the comparison of measurements from two correlated tests in which each unit of observation is associated with a binary outcome of interest via relative risk. The associated confidence interval is crucial because it provides an appreciation of the spectrum of possible values, allowing for a more robust interpretation of relative risk. Of the available confidence interval methods for relative risk, the asymptotic score interval is the most widely recommended for practical use. We propose a modified score interval for relative risk and we also extend an existing nonparametric U-statistic-based confidence interval to relative risk. In addition, we theoretically prove that the original asymptotic score interval is equivalent to the constrained maximum likelihood-based interval proposed by Nam and Blackwelder. Two clinically relevant oncology trials are used to demonstrate the real-world performance of our methods. The finite sample properties of the new approaches, the current standard of practice, and other alternatives are studied via extensive simulation studies. We show that, as the strength of correlation increases, when the sample size is not too large the new score-based intervals outperform the existing intervals in terms of coverage probability. Moreover, our results indicate that the new nonparametric interval provides the coverage that most consistently meets or exceeds the nominal coverage probability.

Assessing the relative performance of fast molecular dating methods for phylogenomic data

Advances in genome sequencing techniques produced a significant growth of phylogenomic datasets. This massive amount of data represents a computational challenge for molecular dating with Bayesian approaches. Rapid molecular dating methods have been proposed over the last few decades to overcome these issues. However, a comparative evaluation of their relative performance on empirical data sets is lacking. We analyzed 23 empirical phylogenomic datasets to investigate the performance of two commonly employed fast dating methodologies: penalized likelihood (PL), implemented in treePL, and the relative rate framework (RRF), implemented in RelTime. They were compared to Bayesian analyses using the closest possible substitution models and calibration settings. We found that RRF was computationally faster and generally provided node age estimates statistically equivalent to Bayesian divergence times. PL time estimates consistently exhibited low levels of uncertainty. Overall, to approximate Bayesian approaches, RelTime is an efficient method with significantly lower computational demand, being more than 100 times faster than treePL. Thus, to alleviate the computational burden of Bayesian divergence time inference in the era of massive genomic data, molecular dating can be facilitated using the RRF, allowing evolutionary hypotheses to be tested more quickly and efficiently.

P-value, compatibility, and S-value

Misinterpretations of P-values and 95% confidence intervals are ubiquitous in medical research. Specifically, the terms significance or confidence, extensively used in medical papers, ignore biases and violations of statistical assumptions and hence should be called overconfidence terms. In this paper, we present the compatibility view of P-values and confidence intervals; the P-value is interpreted as an index of compatibility between data and the model, including the test hypothesis and background assumptions, whereas a confidence interval is interpreted as the range of parameter values that are compatible with the data under background assumptions. We also suggest the use of a surprisal measure, often referred to as the S-value, a novel metric that transforms the P-value, for gauging compatibility in terms of an intuitive experiment of coin tossing.

To curb research misreporting, replace significance and confidence by compatibility: A Preventive Medicine Golden Jubilee article

It is well known that the statistical analyses in health-science and medical journals are frequently misleading or even wrong. Despite many decades of reform efforts by hundreds of scientists and statisticians, attempts to fix the problem by avoiding obvious error and encouraging good practice have not altered this basic situation. Statistical teaching and reporting remain mired in damaging yet editorially enforced jargon of "significance", "confidence", and imbalanced focus on null (no-effect or "nil") hypotheses, leading to flawed attempts to simplify descriptions of results in ordinary terms. A positive development amidst all this has been the introduction of interval estimates alongside or in place of significance tests and P-values, but intervals have been beset by similar misinterpretations. Attempts to remedy this situation by calling for replacement of traditional statistics with competitors (such as pure-likelihood or Bayesian methods) have had little impact. Thus, rather than ban or replace P-values or confidence intervals, we propose to replace traditional jargon with more accurate and modest ordinary-language labels that describe these statistics as measures of compatibility between data and hypotheses or models, which have long been in use in the statistical modeling literature. Such descriptions emphasize the full range of possibilities compatible with observations. Additionally, a simple transform of the P-value called the surprisal or S-value provides a sense of how much or how little information the data supply against those possibilities. We illustrate these reforms using some examples from a highly charged topic: trials of ivermectin treatment for Covid-19.

Continuity corrected score confidence interval for the difference in proportions in paired data

For paired binary data, the hybrid method and the score method are often recommended for use to calculate the confidence interval for risk difference. These asymptotic intervals do not control the coverage probability. We propose to develop a new score interval with continuity correction to further improve the performance of the existing intervals. The traditional correction value may be too large which leads to a wide interval. For that reason, we propose three different correction values to identify the optimal correction interval with balanced coverage probability and interval width. From simulation studies, we find that a small correction value for the score interval has good performance. In addition, we derive the non-iterative solutions for the developed continuity correction score intervals.

Revised time estimation of the ancestral human chromosome 2 fusion

Background

The reduction of the chromosome number from 48 in the Great Apes to 46 in modern humans is thought to result from the end-to-end fusion of two ancestral non-human primate chromosomes forming the human chromosome 2 (HSA2). Genomic signatures of this event are the presence of inverted telomeric repeats at the HSA2 fusion site and a block of degenerate satellite sequences that mark the remnants of the ancestral centromere. It has been estimated that this fusion arose up to 4.5 million years ago (Mya).

Results

We have developed an enhanced algorithm for the detection and efficient counting of the locally over-represented weak-to-strong (AT to GC) substitutions. By analyzing the enrichment of these substitutions around the fusion site of HSA2 we estimated its formation time at 0.9 Mya with a 95% confidence interval of 0.4-1.5 Mya. Additionally, based on the statistics derived from our algorithm, we have reconstructed the evolutionary distances among the Great Apes (Hominoidea).

Conclusions

Our results shed light on the HSA2 fusion formation and provide a novel computational alternative for the estimation of the speciation chronology.

Variability and uncertainty: interpretation and usage of pharmacometric simulations and intervals

Variability and estimation uncertainty are important sources of variation in pharmacometric simulations. Different combinations of uncertainty and the variability components lead to a variety types of simulation intervals, and many realized and unrealized confusions exist among pharmacometricians on their interpretation and usage. This commentary aims to clarify some of the important underlying concepts and provide a convenient guideline on pharmacometric simulation conduct and interpretation.

The interobserver reliability of clinical relevance in orthopaedic research

PURPOSE

A ratio of observed difference (OD) over the 95% confidence interval (CI) has been shown to be strongly associated with the perceived clinical relevance (CR) of medical research results. The purpose of this study was to evaluate the association between the OD/CI ratio and perceived CR in orthopaedic research.

METHODS

Sixty-seven orthopaedic surgeons completed a survey with 15 study outcomes (mean difference and CI) and were asked if they perceived the findings as clinically relevant. The interobserver reliability of perceived CR and the association between CR and the OD/CI ratio and p-value were assessed.

RESULTS

The interobserver reliability of CR between respondents was moderate (kappa = 0.46, CI 0.45 to 0.48). P-values did not differ between results with and without CR (median difference (MD) - 0.12, CI - 0.74 to 0.0009, p = 0.07). The OD/CI ratio, however, was greater for results with CR (MD 1.01, CI 0.3 to 3.9, p = 0.004). The area under the curve (AUC) for the p-value and OD/CI ratio receiver operating characteristic (ROC) curves was 0.80 (p = 0.01) and 0.97 (p = 0.0003). The cutoff p -value and OD/CI ratio that maximized the sensitivity (SN) and specificity (SP) for CR were 0.001 (SN 80%, SP 80%) and 0.84 (SN 100%, SP 90%). The SN and SP of a p-value cutoff of 0.05 was 100% and 50%.

CONCLUSION

The interobserver reliability of the perceived CR of orthopaedic research findings was moderate. The OD/CI ratio, in contrast to the p-value, was strongly associated with perceived CR making it a potentially useful measure to evaluate research results.

Evaluation of the Use of P Values in Neurosurgical Literature: from Statistical Significance to Clinical Irrelevance

The application and interpretation of P values have caused debate for several decades, and this debate has become particularly relevant in the past few years. The P value represents the probability of seeing results as extreme or more extreme than those observed in a data analysis, were the null hypothesis and other underlying assumptions to be true. While P values are useful in pointing out where an effect may be present, they have often been misused in an attempt to oversell "statistically significant" findings. As P values rely on the spread and number of measurements, a smaller P value does not necessarily imply a larger effect size, which is better assessed via an effect estimate and confidence interval interpreted in the context of the study. The clinical relevance of a computed P value is context dependent. We investigated the current use of P values in a small sample of recent neurosurgical literature. Only a minority of manuscripts that reported statistical significance described confounder adjustment, or effect sizes. A common, incorrect assumption often observed was that statistical significance equals clinical relevance. To enable correct interpretation of clinical significance, it is crucial that authors describe the clinical implications of their findings.

Performing Meta-analyses with Very Few Studies

This chapter contains a methodological framework for choosing a model for the meta-analysis of very few studies and selecting an estimation method for the chosen model by means of study characteristics and by comparing results yielded by different approaches. When the results are inconclusive between different estimation methods, it might be the best solution to refrain from a quantitative meta-analysis but to summarize the study results by means of a qualitative evidence synthesis.

Propagating clade and model uncertainty to confidence intervals of divergence times and branch lengths

Confidence intervals of divergence times and branch lengths do not reflect uncertainty about their clades or about the prior distributions and other model assumptions on which they are based. Uncertainty about the clade may be propagated to a confidence interval by multiplying its confidence level by the bootstrap proportion of its clade or by another probability that the clade is correct. (If the confidence level is 95% and the bootstrap proportion is 90%, then the uncertainty-adjusted confidence level is (0.95)(0.90) = 86%.) Uncertainty about the model can be propagated to the confidence interval by reporting the union of the confidence intervals from all the plausible models. Unless there is no overlap between the confidence intervals, that results in an uncertainty-adjusted interval that has as its lower and upper limits the most extreme limits of the models. The proposed methods of uncertainty quantification may be used together.

Estimators and confidence intervals of f2 using bootstrap methodology for the comparison of dissolution profiles

BACKGROUND AND OBJECTIVES

The most widely used method to compare dissolution profiles is the similarity factor f₂. When this method is not applicable, the confidence interval of f₂ using bootstrap methodology has been recommended instead. As neither details of the estimator nor the types of confidence intervals are described in the guidelines, the suitability of five estimators and fourteen types of confidence intervals were investigated in this study by simulation.

METHODS

One million individual dissolution profiles were simulated for the reference and test populations with predefined target population f₂ values, where random samples of different sizes were drawn without replacement. From each pair of random samples, five f₂ estimators were calculated, and fourteen types of confidence intervals were obtained using 5000 bootstrap samples. The whole process was repeated 10000 times and the percentage of the similarity conclusions was measured. In addition, the uncertainty associated with the current practice of using f^₂ point estimate alone for the statistical inference was evaluated.

RESULTS

When combined with different types of confidence intervals, the estimated f₂ (f^₂), the bias-corrected f₂ (f^_2,bc), and the variance- and bias-corrected f₂ (f^_2,vcbc) are not suitable estimators due to higher-than-acceptable type I errors. The estimator f^_2,exp, calculated based on the mathematical expectation of f^₂, and f^_2,vcexp, the variance-corrected f^_2,exp, showed acceptable type I errors when combined with any of the ten percentile intervals. However, they have the drawback of low power, which might be addressed by increasing the sample size. To properly control the type I error, samples with at least 12 units should be used.

CONCLUSION

The best combinations of estimator and type of confidence interval are f^_2,exp and f^_2,vcexp combined with any of the ten types of percentile intervals. When the sample f₂ value is close to 50, the use of the confidence interval of f₂ is recommended even when the variability of the dissolution profiles is low and the prerequisites defined in the regulatory guidelines for using the conventional f₂ method are fulfilled in order to control the type I error rate.

The effect of coffee consumption on the non-alcoholic fatty liver disease and liver fibrosis: A meta-analysis of 11 epidemiological studies

INTRODUCTION AND OBJECTIVES

Non-alcoholic fatty liver disease (NAFLD) is a widespread chronic liver disease. It is considered a multifactorial disorder that can progress to liver fibrosis and cause a worldwide public health concern. Coffee consumption may have a protective impact on NAFLD and liver fibrosis. However, the evidence from the previous studies is inconsistent. This meta-analysis summarizes available literature.

MATERIALS AND METHODS

This study comprises two meta-analyses. The first meta-analysis summarizes the effect of coffee consumption on NAFLD in those who did or did not drink coffee. The second analysis compares the risk of liver fibrosis development between NAFLD patients who did or did not drink coffee. Pooled risk ratios (RR) and confidence intervals (CI) of observational studies were estimated.

RESULTS

Of the total collected 321 articles, 11 met our eligibility criteria to be included in the analysis. The risk of NAFLD among those who drank coffee compared to those who did not was significantly lower with a pooled RR value of 0.77 (95% CI 0.60-0.98). Moreover, we also found a significantly reduced risk of liver fibrosis in those who drink coffee than those who did not drink in the NAFLD patients with the relative risk (RR) of 0.68 (95% CI 0.68-0.79).

CONCLUSIONS

Regular coffee consumption is significantly associated with a reduced risk of NAFLD. It is also significantly associated with decreased risk of liver fibrosis development in already diagnosed NAFLD patients. Although coffee consumption may be considered an essential preventive measure for NAFLD, this subject needs further epidemiological studies.

Confidence interval of risk difference by different statistical methods and its impact on the study conclusion in antibiotic non-inferiority trials

Background

Numerous statistical methods can be used to calculate the confidence interval (CI) of risk differences. There is consensus in previous literature that the Wald method should be discouraged. We compared five statistical methods for estimating the CI of risk difference in terms of CI width and study conclusion in antibiotic non-inferiority trials.

Methods

In a secondary analysis of a systematic review, we included non-inferiority trials that compared different antibiotic regimens, reported risk differences for the primary outcome, and described the number of successes and/or failures as well as patients in each arm. For each study, we re-calculated the risk difference CI using the Wald, Agresti-Caffo, Newcombe, Miettinen-Nurminen, and skewness-corrected asymptotic score (SCAS) methods. The CIs by different statistical methods were compared in terms of CI width and conclusion on non-inferiority. A wider CI was considered to be more conservative.

Results

The analysis included 224 comparisons from 213 studies. The statistical method used to calculate CI was not reported in 134 (59.8%) cases. The median (interquartile range IQR) for CI width by Wald, Agresti-Caffo, Newcombe, Miettinen-Nurminen, and SCAS methods was 13.0% (10.8%, 17.4%), 13.3% (10.9%, 18.5%), 13.6% (11.1%, 18.9%), 13.6% (11.1% and 19.0%), and 13.4% (11.1%, 18.9%), respectively. In 216 comparisons that reported a non-inferiority margin, the conclusion on non-inferiority was the same across the five statistical methods in 211 (97.7%) cases. The differences in CI width were more in trials with a sample size of 100 or less in each group and treatment success rate above 90%. Of the 18 trials in this subgroup with a specified non-inferiority margin, non-inferiority was shown in 17 (94.4%), 16 (88.9%), 14 (77.8%), 14 (77.8%), and 15 (83.3%) cases based on CI by Wald, Agresti-Caffo, Newcombe, Miettinen-Nurminen, and SCAS methods, respectively.

Conclusions

The statistical method used to calculate CI was not reported in the majority of antibiotic non-inferiority trials. Different statistical methods for CI resulted in different conclusions on non-inferiority in 2.3% cases. The differences in CI widths were highest in trials with a sample size of 100 or less in each group and a treatment success rate above 90%.

Trial registration

PROSPERO CRD42020165040. April 28, 2020.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13063-021-05686-8.

Randomization-based confidence intervals for cluster randomized trials

In a cluster randomized trial (CRT), groups of people are randomly assigned to different interventions. Existing parametric and semiparametric methods for CRTs rely on distributional assumptions or a large number of clusters to maintain nominal confidence interval (CI) coverage. Randomization-based inference is an alternative approach that is distribution-free and does not require a large number of clusters to be valid. Although it is well-known that a CI can be obtained by inverting a randomization test, this requires testing a non-zero null hypothesis, which is challenging with non-continuous and survival outcomes. In this article, we propose a general method for randomization-based CIs using individual-level data from a CRT. This approach accommodates various outcome types, can account for design features such as matching or stratification, and employs a computationally efficient algorithm. We evaluate this method's performance through simulations and apply it to the Botswana Combination Prevention Project, a large HIV prevention trial with an interval-censored time-to-event outcome.

Replicate analyses of OSPAR beach litter data

Replicate surveys of beach litter have seldom been performed in the past. In this study, replicate surveys of beach litter were conducted on the beach north of Hörnum (Sylt, Germany), from 2015 to 2019, applying a slightly modified OSPAR protocol of beach litter monitoring. Descriptive statistics and power analyses were calculated on data resulting from these replicate surveys, to find out whether the scatter of replicate beach litter data decreases and the statistical power increases with increasing numbers of replicate surveys. From 2015 to 2019, mean total abundances, given as numbers of litter items, ranged from 19 to 185 litter items on a 50 m section of beach. With increasing numbers of replicate surveys, the scatter given by the coefficient of variation (CV) significantly increased up to 113%. Statistical power considerably increased with increasing numbers of replicate beach sections, e.g. from 82% (two beach sections) to nearly 100% (five beach sections) for a given reduction of beach litter of 50%. Based on these results from a morphologically straight coastline, the use of replicate surveys would be sensible for the future monitoring of beach litter. However, there is high need for studies, which consider coastlines with varying morphology.

Normative cervical spine kinematics of a circumduction task

Neck pain is a prevalent condition and clinical examination techniques are limited and unable to assess out-of-plane motion. Recent works investigating cervical kinematics during neck circumduction (NC), a dynamic 3D task, has shown the ability to discern those with and without neck pain. The purposes of this study were to establish 1) confidence and prediction intervals of head-to-torso kinematics during NC in a healthy cohort, 2) a baseline summative metric to quantify the duration and magnitude of deviations outside the prediction interval, and 3) the reliability of NC. Thirty-nine participants (25.6 ± 6.3 years, 19F/20M) without neck pain completed left and right NC. A two-way smoothing spline analysis of variance was utilized to determine the mean-fitted values and 90% confidence and prediction intervals for NC. A standardized effect size was calculated and aggregated across all axes (Delta RMSD aggregate), as a summative metric of motion quality. Confidence and prediction intervals were comparable for left and right NC and demonstrated excellent reliability. The average sum of the Delta RMSD aggregate was 2.76 ± 0.55 for left NC and 2.74 ± 0.63 for right NC. The results of this study demonstrate the feasibility of utilizing normative intervals of a NC task to assess head-to-torso kinematics.

Clinical Application of Confidence Interval for Monitoring Changes in Tumor Markers to Determine the Responsiveness to Cancer Treatment

OBJECTIVE

Tumor markers are used to monitor disease progression and determine the responsiveness to cancer treatment. However, there are no standardized criteria for monitoring serial tumor marker measurements. Herein, we have developed a monitoring system for interpreting changes in tumor markers using overlapping 95% confidence intervals (CIs).

METHODS

Two-year data, including 117,289 results for 11 tumor markers in our laboratory, were analyzed. CI ranges for each tumor marker were set based on biological variation, and data were analyzed for each patient assessed at health check-ups and clinics, individually and overall.

RESULTS

The 95^th percentile cut-offs for each tumor marker were much higher in the clinic group than in the health check-up group. In decreasing order, the percentages of results with no overlap in 95% CIs were thyroglobulin antigen, 14.9%; protein induced by vitamin K absence-II (PIVKA), 11.9%; and prostate-specific antigen, 9.8%. After correction using the reference interval, the percentages decreased to less than 5%, except for PIVKA (10.9%).

CONCLUSION

We suggest that our monitoring system can serve as a criterion for the auto-verification of tumor markers. Further studies are required to validate and demonstrate this concept in real clinical situations using actual clinical data reflecting disease progression in cancer patients and responsiveness to cancer treatment.

How to use likelihood ratios to interpret evidence from randomized trials

OBJECTIVE

The likelihood ratio is a method for assessing evidence regarding two simple statistical hypotheses. Its interpretation is simple - for example, a value of 10 means that the first hypothesis is 10 times as strongly supported by the data as the second. A method is shown for deriving likelihood ratios from published trial reports.

STUDY DESIGN

The likelihood ratio compares two hypotheses in light of data: that a new treatment is effective, at a specified level (alternate hypothesis: for instance, the hazard ratio equals 0.7), and that it is not (null hypothesis: the hazard ratio equals 1). The result of the trial is summarised by the test statistic z (ie, the estimated treatment effect divided by its standard error). The expected value of z is 0 under the null hypothesis, and A under the alternate hypothesis. The logarithm of the likelihood ratio is given by z·A - A²/2. The values of A and z can be derived from the alternate hypothesis used for sample size computation, and from the observed treatment effect and its standard error or confidence interval.

RESULTS

Examples are given of trials that yielded strong or moderate evidence in favor of the alternate hypothesis, and of a trial that favored the null hypothesis. The resulting likelihood ratios are applied to initial beliefs about the hypotheses to obtain posterior beliefs.

CONCLUSIONS

The likelihood ratio is a simple and easily understandable method for assessing evidence in data about two competing a priori hypotheses.

Machine learning and statistical approaches for classification of risk of coronary artery disease using plasma cytokines

Background

As per the 2017 WHO fact sheet, Coronary Artery Disease (CAD) is the primary cause of death in the world, and accounts for 31% of total fatalities. The unprecedented 17.6 million deaths caused by CAD in 2016 underscores the urgent need to facilitate proactive and accelerated pre-emptive diagnosis. The innovative and emerging Machine Learning (ML) techniques can be leveraged to facilitate early detection of CAD which is a crucial factor in saving lives. The standard techniques like angiography, that provide reliable evidence are invasive and typically expensive and risky. In contrast, ML model generated diagnosis is non-invasive, fast, accurate and affordable. Therefore, ML algorithms can be used as a supplement or precursor to the conventional methods. This research demonstrates the implementation and comparative analysis of K Nearest Neighbor (k-NN) and Random Forest ML algorithms to achieve a targeted “At Risk” CAD classification using an emerging set of 35 cytokine biomarkers that are strongly indicative predictive variables that can be potential targets for therapy. To ensure better generalizability, mechanisms such as data balancing, repeated k-fold cross validation for hyperparameter tuning, were integrated within the models. To determine the separability efficacy of “At Risk” CAD versus Control achieved by the models, Area under Receiver Operating Characteristic (AUROC) metric is used which discriminates the classes by exhibiting tradeoff between the false positive and true positive rates.

Results

A total of 2 classifiers were developed, both built using 35 cytokine predictive features. The best AUROC score of .99 with a 95% Confidence Interval (CI) (.982,.999) was achieved by the Random Forest classifier using 35 cytokine biomarkers. The second-best AUROC score of .954 with a 95% Confidence Interval (.929,.979) was achieved by the k-NN model using 35 cytokines. A p-value of less than 7.481e-10 obtained by an independent t-test validated that Random Forest classifier was significantly better than the k-NN classifier with regards to the AUROC score.

Presently, as large-scale efforts are gaining momentum to enable early, fast, reliable, affordable, and accessible detection of individuals at risk for CAD, the application of powerful ML algorithms can be leveraged as a supplement to conventional methods such as angiography. Early detection can be further improved by incorporating 65 novel and sensitive cytokine biomarkers. Investigation of the emerging role of cytokines in CAD can materially enhance the detection of risk and the discovery of mechanisms of disease that can lead to new therapeutic modalities.

Identification of Images of COVID-19 from Chest X-rays Using Deep Learning: Comparing COGNEX VisionPro Deep Learning 1.0™ Software with Open Source Convolutional Neural Networks

The novel Coronavirus, COVID-19, pandemic is being considered the most crucial health calamity of the century. Many organizations have come together during this crisis and created various Deep Learning models for the effective diagnosis of COVID-19 from chest radiography images. For example, The University of Waterloo, along with Darwin AI—a start-up spin-off of this department, has designed the Deep Learning model ‘COVID-Net’ and created a dataset called ‘COVIDx’ consisting of 13,975 images across 13,870 patient cases. In this study, COGNEX’s Deep Learning Software, VisionPro Deep Learning™,  is used to classify these Chest X-rays from the COVIDx dataset. The results are compared with the results of COVID-Net and various other state-of-the-art Deep Learning models from the open-source community. Deep Learning tools are often referred to as black boxes because humans cannot interpret how or why a model is classifying an image into a particular class. This problem is addressed by testing VisionPro Deep Learning with two settings, first, by selecting the entire image as the Region of Interest (ROI), and second, by segmenting the lungs in the first step, and then doing the classification step on the segmented lungs only, instead of using the entire image. VisionPro Deep Learning results: on the entire image as the ROI it achieves an overall F score of 94.0%, and on the segmented lungs, it gets an F score of 95.3%, which is better than COVID-Net and other state-of-the-art open-source Deep Learning models.

A unified method for improved inference in random effects meta-analysis

Random effects meta-analyses have been widely applied in evidence synthesis for various types of medical studies. However, standard inference methods (e.g. restricted maximum likelihood estimation) usually underestimate statistical errors and possibly provide highly overconfident results under realistic situations; for instance, coverage probabilities of confidence intervals can be substantially below the nominal level. The main reason is that these inference methods rely on large sample approximations even though the number of synthesized studies is usually small or moderate in practice. In this article, we solve this problem using a unified inference method based on Monte Carlo conditioning for broad application to random effects meta-analysis. The developed method provides improved confidence intervals with coverage probabilities that are closer to the nominal level than standard methods. As specific applications, we provide new inference procedures for three types of meta-analysis: conventional univariate meta-analysis for pairwise treatment comparisons, meta-analysis of diagnostic test accuracy, and multiple treatment comparisons via network meta-analysis. We also illustrate the practical effectiveness of these methods via real data applications and simulation studies.

Common methodological issues and suggested solutions in bone research

Bone research is a dynamic area of scientific investigation that usually encompasses multidisciplines. Virtually all basic cellular research, clinical research and epidemiologic research rely on statistical concepts and methodology for inference. This paper discusses common issues and suggested solutions concerning the application of statistical thinking in bone research, particularly in clinical and epidemiological investigations. The issues are sample size estimation, biases and confounders, analysis of longitudinal data, categorization of continuous data, selection of significant variables, over-fitting, P-values, false positive finding, confidence interval, and Bayesian inference. It is hoped that by adopting the suggested measures the scientific quality of bone research can improve.

Recommended confidence intervals for the conditional odds ratio in matched-pairs designs

There has been limited research on the confidence intervals of the conditional odds ratio in matched-pairs design. This article investigates the interval estimation of the conditional odds ratio. We described several confidence intervals, which are available in some situations, and they can produce different results. We tried to determine which method(s) should be recommended for different situations. We derived four confidence intervals from the delta test, the score test, the inferential model test, and the fiducial test, and employed four exact calculation studies to compare the performances of the four methods, in order to make recommendations for small and moderate-to-large sample sizes. All of the methods are illustrated using a real example. And we offered the recommendations for different situations.

Left-censored dementia incidences in estimating cohort effects

We estimate the dementia incidence hazard in Germany for the birth cohorts 1900 until 1954 from a simple sample of Germany's largest health insurance company. Followed from 2004 to 2012, 36,000 uncensored dementia incidences are observed and further 200,000 right-censored insurants included. From a multiplicative hazard model we find a positive and linear trend in the dementia hazard over the cohorts. The main focus of the study is on 11,000 left-censored persons who have already suffered from the disease in 2004. After including the left-censored observations, the slope of the trend declines markedly due to Simpson's paradox, left-censored persons are imbalanced between the cohorts. When including left-censoring, the dementia hazard increases differently for different ages, we consider omitted covariates to be the reason. For the standard errors from large sample theory, left-censoring requires an adjustment to the conditional information matrix equality.

Calibrated Bayesian Credible Intervals for Binomial Proportions

Drawbacks of traditional approximate (Wald test-based) and exact (Clopper-Pearson) confidence intervals for a binomial proportion are well-recognized. Alternatives include an interval based on inverting the score test, adaptations of exact testing, and Bayesian credible intervals derived from uniform or Jeffreys beta priors. We recommend a new interval intermediate between the Clopper-Pearson and Jeffreys in terms of both width and coverage. Our strategy selects a value κ between 0 and 0.5 based on stipulated coverage criteria over a grid of regions comprising the parameter space, and bases lower and upper limits of a credible interval on Beta(κ, 1- κ) and Beta(1- κ, κ) priors, respectively. The result tends toward the Jeffreys interval if the criterion is to ensure an average overall coverage rate (1-α) across a single region of width 1, and toward the Clopper-Pearson if the goal is to constrain both lower and upper lack of coverage rates at α/2 with region widths approaching zero. We suggest an intermediate target that ensures all average lower and upper lack of coverage rates over a specified set of regions are ≤ α/2. Interval width subject to these criteria is readily optimized computationally, and we demonstrate particular benefits in terms of coverage balance.

Statistical Guideline #6. Indicate magnitude and precision in your estimation and use "new statistics"

This is one in a series of statistical guidelines designed to highlight common statistical considerations in behavioral medicine research. The goal is to briefly discuss appropriate ways to analyze and present data in the International Journal of Behavioral Medicine (IJBM). Collectively, the series will culminate in a set of basic statistical guidelines to be adopted by IJBM and integrated into the journal's official Instructions for Authors, and also to serve as an independent resource. If you have ideas for a future topic, please email the Statistical Editor, Suzanne Segerstrom at segerstrom@uky.edu.