scholarly journals The influence of study characteristics on coordinate-based fMRI meta-analyses

2017 ◽  
Author(s):  
Han Bossier ◽  
Ruth Seurinck ◽  
Simone Kühn ◽  
Tobias Banaschewski ◽  
Gareth J. Barker ◽  
...  
Keyword(s):  
Random Effects ◽  
Fixed Effects ◽  
Meta Analysis ◽  
Group Analysis ◽  
Likelihood Estimation ◽  
Ordinary Least Squares ◽  
Mixed Effects ◽  
Mixed Effects Models ◽  
Activation Likelihood Estimation ◽  
Meta Analyses

AbstractGiven the increasing amount of neuroimaging studies, there is a growing need to summarize published results. Coordinate-based meta-analyses use the locations of statistically significant local maxima with possibly the associated effect sizes to aggregate studies. In this paper, we investigate the influence of key characteristics of a coordinate-based meta-analysis on (1) the balance between false and true positives and (2) the reliability of the outcome from a coordinate-based meta-analysis. More particularly, we consider the influence of the chosen group level model at the study level (fixed effects, ordinary least squares or mixed effects models), the type of coordinate-based meta-analysis (Activation Likelihood Estimation, fixed effects and random effects meta-analysis) and the amount of studies included in the analysis (10, 20 or 35). To do this, we apply a resampling scheme on a large dataset (N = 1400) to create a test condition and compare this with an independent evaluation condition. The test condition corresponds to subsampling participants into studies and combine these using meta-analyses. The evaluation condition corresponds to a high-powered group analysis. We observe the best performance when using mixed effects models in individual studies combined with a random effects meta-analysis. This effect increases with the number of studies included in the meta-analysis. We also show that the popular Activation Likelihood Estimation procedure is a valid alternative, though the results depend on the chosen threshold for significance. Furthermore, this method requires at least 20 to 35 studies. Finally, we discuss the differences, interpretations and limitations of our results.

Meta-Analysis of Meta-Analyses in Communication: Comparing Fixed Effects and Random Effects Analysis Models

2010 ◽  
Vol 58 (3) ◽  
pp. 257-278 ◽  
Author(s):  
Ashley Anker ◽  
Amber Marie Reinhart ◽  
Thomas Hugh Feeley
Keyword(s):  
Random Effects ◽  
Fixed Effects ◽  
Meta Analysis ◽  
Meta Analyses ◽  
Analysis Models

scholarly journals Prognostic Role of Matrix Metalloproteinases in Cervical Cancer: A Meta-Analysis

2021 ◽  
Vol 28 ◽  
pp. 107327482110337
Author(s):  
Weiwei Chen ◽  
Shenjiao Huang ◽  
Kun Shi ◽  
Lisha Yi ◽  
Yaqiong Liu ◽  
...  
Keyword(s):  
Cervical Cancer ◽  
Matrix Metalloproteinases ◽  
Random Effects ◽  
Fixed Effects ◽  
Prognostic Value ◽  
Meta Analysis ◽  
Free Survival ◽  
Random Effects Models ◽  
Registration Number ◽  
Meta Analyses

Objective Studies have published the association between the expression of matrix metalloproteinases (MMPs) and the outcome of cervical cancer. However, the prognostic value in cervical cancer remains controversial. This meta-analysis was conducted to evaluate the prognostic functions of MMP expression in cervical cancer. Methods A comprehensive search of PubMed, Embase, and Web of Science databases was conducted to identify the eligible studies according to defined selection and excluding criteria and analyzed according to Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines. Fixed and random effects models were evaluated through the hazard ratios (HRs) and 95% confidence intervals (CIs) to estimate the overall survival (OS), recurrence-free survival (RFS), and progress-free survival (PFS). Results A total of 18 eligible studies including 1967 patients were analyzed for prognostic value. Totally 16 selected studies including 21 tests were relevant to the cervical cancer OS, 4 studies focused on RFS, and 1 study on PFS. The combined pooled HRs and 95% CIs of OS were calculated with random-effects models (HR = 1.64, 95% CI = 1.01–2.65, P = .000). In the subgroup analysis for OS, there was no heterogeneity in MMP-2 (I2 = .0%, P = .880), MMP-1 (I2 = .0%, P = .587), and MMP-14 (I2 = 28.3%, P = .248). In MMP-7 and MMP-9, the heterogeneities were obvious (I2 = 99.2% ( P = .000) and I2 = 77.9% ( P = .000), respectively). The pooled HRs and 95% CIs of RFS were calculated with fixed-effects models (HR = 2.22, 95% CI = 1.38–3.58, P = .001) and PFS (HR = 2.29, 95% CI = 1.14–4.58, P = .035). Conclusions The results indicated that MMP overexpression was associated with shorter OS and RFS in cervical cancer patients. It suggested that MMP overexpression might be a poor prognostic marker in cervical cancer. Research Registry Registration Number: reviewregistry 1159.

An imputation approach for fitting two-part mixed effects models for longitudinal semi-continuous data

2020 ◽  
Vol 29 (11) ◽  
pp. 3351-3361
Author(s):  
Hyoyoung Choo-Wosoba ◽  
Debamita Kundu ◽  
Paul S Albert
Keyword(s):  
Random Effects ◽  
Fixed Effects ◽  
Clinical Trial Data ◽  
Mixed Effects ◽  
Mixed Effects Models ◽  
Unbiased Estimation ◽  
Continuous Components ◽  
Wide Range ◽  
Parameter Values ◽  
Imputation Approach

Two-part mixed effects models are often used for analyzing longitudinal data with many zeros. Typically, these models are formulated with binary and continuous components separately with random effects that are correlated between the two components. Researchers have developed maximum-likelihood and Bayesian approaches for fitting these models that often require using particular software packages or very specialized software. We propose an imputation approach that will allow practitioners to separately use standard linear and generalized linear mixed models to estimate the fixed effects for two-part mixed effects models with complex random effects structures. An approximation to the conditional distribution of positive measurements given an individual’s pattern of non-zero measurements is proposed that can be easily estimated and then imputed from. We show that for a wide range of parameter values, the imputation approach results in nearly unbiased estimation and can be implemented with standard software. We illustrate the proposed imputation approach for the analysis of longitudinal clinical trial data with many zeros.

scholarly journals Smoking prevalence among hospitalized COVID-19 patients and its association with disease severity and mortality: an expanded re-analysis of a recent publication

2021 ◽  
Vol 18 (1) ◽  
Author(s):  
Konstantinos Farsalinos ◽  
Pantelis G. Bagos ◽  
Theodoros Giannouchos ◽  
Raymond Niaura ◽  
Anastasia Barbouni ◽  
...  
Keyword(s):  
Disease Severity ◽  
Random Effects ◽  
Fixed Effects ◽  
Smoking Prevalence ◽  
Meta Analysis ◽  
Original Study ◽  
Hospitalized Patients ◽  
Prevalence Odds Ratio ◽  
Meta Analyses ◽  
Low Prevalence

Abstract Background There is a lot of debate about the effects of smoking on COVID-19. A recent fixed-effects meta-analysis found smoking to be associated with disease severity among hospitalized patients, but other studies report an unusually low prevalence of smoking among hospitalized patients. The purpose of this study was to expand the analysis by calculating the prevalence odds ratio (POR) of smoking among hospitalized COVID-19 patients, while the association between smoking and disease severity and mortality was examined by random-effects meta-analyses considering the highly heterogeneous study populations. Methods The same studies as examined in the previous meta-analysis were analyzed (N = 22, 20 studies from China and 2 from USA). The POR relative to the expected smoking prevalence was calculated using gender and age-adjusted population smoking rates. Random-effects meta-analyses were used for all other associations. Results A total of 7162 patients were included, with 482 being smokers. The POR was 0.24 (95%CI 0.19–0.30). Unlike the original study, the association between smoking and disease severity was not statistically significant using random-effects meta-analysis (OR 1.40, 95%CI 0.98–1.98). In agreement with the original study, no statistically significant association was found between smoking and mortality (OR 1.86, 95%CI 0.88–3.94). Conclusion An unusually low prevalence of smoking, approximately 1/4th the expected prevalence, was observed among hospitalized COVID-19 patients. Any association between smoking and COVID-19 severity cannot be generalized but should refer to the seemingly low proportion of smokers who develop severe COVID-19 that requires hospitalization. Smokers should be advised to quit due to long-term health risks, but pharmaceutical nicotine or other nicotinic cholinergic agonists should be explored as potential therapeutic options, based on a recently presented hypothesis.

Fitting Growth Model Using Nonlinear Regression with Random Parameters

2011 ◽  
Vol 480-481 ◽  
pp. 1308-1312
Author(s):  
Yao Xiang Li ◽  
Li Chun Jiang
Keyword(s):  
Random Effects ◽  
Repeated Measures ◽  
Fixed Effects ◽  
Model Building ◽  
Mixed Effects ◽  
Mixed Effects Models ◽  
Ratio Test ◽  
Nonlinear Mixed Effects ◽  
Nonlinear Mixed Effects Models ◽  
Mixed Effect

Mixed Effect models are flexible models to analyze grouped data including longitudinal data, repeated measures data, and multivariate multilevel data. One of the most common applications is nonlinear growth data. The Chapman-Richards model was fitted using nonlinear mixed-effects modeling approach. Nonlinear mixed-effects models involve both fixed effects and random effects. The process of model building for nonlinear mixed-effects models is to determine which parameters should be random effects and which should be purely fixed effects, as well as procedures for determining random effects variance-covariance matrices (e.g. diagonal matrices) to reduce the number of the parameters in the model. Information criterion statistics (AIC, BIC and Likelihood ratio test) are used for comparing different structures of the random effects components. These methods are illustrated using the nonlinear mixed-effects methods in S-Plus software.

Meta-analysis

2010 ◽  
pp. 447-475
Author(s):  
Janet L. Peacock ◽  
Philip J. Peacock
Keyword(s):  
Publication Bias ◽  
Random Effects ◽  
Fixed Effects ◽  
Meta Analysis ◽  
Patient Data ◽  
Meta Analyses ◽  
Combining Estimates

Meta-analysis: introduction 448 Searching for studies 450 Combining estimates in meta-analyses 452 Heterogeneity 454 Overcoming heterogeneity 456 Fixed effects estimates 458 Random effects estimates 460 Presenting meta-analyses 462 Publication bias 464 Detecting publication bias 466 Adjusting for publication bias 468 Independent patient data meta-analysis 472...

scholarly journals Benefits of Bayesian Model Selection and Averaging for Mixed-Effects Modeling

2021 ◽  
Author(s):  
Daniel W. Heck ◽  
Florence Bockting
Keyword(s):  
Model Selection ◽  
Random Effects ◽  
Mixed Models ◽  
Bayesian Model ◽  
Fixed Effects ◽  
Bayes Factor ◽  
Mixed Effects ◽  
Bayes Factors ◽  
Mixed Effects Models ◽  
Within Subjects

Bayes factors allow researchers to test the effects of experimental manipulations in within-subjects designs using mixed-effects models. van Doorn et al. (2021) showed that such hypothesis tests can be performed by comparing different pairs of models which vary in the specification of the fixed- and random-effect structure for the within-subjects factor. To discuss the question of which of these model comparisons is most appropriate, van Doorn et al. used a case study to compare the corresponding Bayes factors. We argue that researchers should not only focus on pairwise comparisons of two nested models but rather use the Bayes factor for performing model selection among a larger set of mixed models that represent different auxiliary assumptions. In a standard one-factorial, repeated-measures design, the comparison should include four mixed-effects models: fixed-effects H0, fixed-effects H1, random-effects H0, and random-effects H1. Thereby, the Bayes factor enables testing both the average effect of condition and the heterogeneity of effect sizes across individuals. Bayesian model averaging provides an inclusion Bayes factor which quantifies the evidence for or against the presence of an effect of condition while taking model-selection uncertainty about the heterogeneity of individual effects into account. We present a simulation study showing that model selection among a larger set of mixed models performs well in recovering the true, data-generating model.

An Exact Test for Meta-analysis with Binary Endpoints

2007 ◽  
Vol 46 (06) ◽  
pp. 662-668 ◽  
Author(s):  
C. Gromann ◽  
O. Kuss
Keyword(s):  
Random Effects ◽  
Fixed Effects ◽  
Meta Analysis ◽  
Artery Bypass ◽  
Sparse Data ◽  
Random Effects Model ◽  
Exact Test ◽  
Standard Methods ◽  
Meta Analyses ◽  
Empirical Size

Summary Objectives : We reintroduce an exact Mantel-Haenszel (MH) procedure for meta-analysis with binary endpoints which is expected to workespeciallywell i sparse data, e.g., in meta-analyses of safety or adverse events. Methods : The performance of the exact MH procedure in terms of empirical size and power is compared to the asymptotic MH and to the two standard procedures (fixed effects and random effects model) in a simulation study. We illustrate the methods with a metaanalysis of postoperative stroke occurrence after offpump or on-pump surgery in coronary artery bypass grafting. Results : We find that in almost all situations the asymptotic MH procedure outperforms its competitors; especially the standard methods yield poor results in terms of power and size. Conclusions : There is no need to use the reintroduced exact MH procedure; the asymptotic MH procedure will be sufficient in most practical situations. The standard methods (fixed effects and random effects model) should not be used in the sparse data situation.

Sign in / Sign up

Export Citation Format

Share Document