Hypothesis tests for population heterogeneity in meta-analysis

An Exact Bayesian Model for Meta-Analysis of the Standardized Mean Difference with Its Simultaneous Credible Intervals

While Bayesian methodology is increasingly favored in behavioral research for its clear probabilistic inference and model structure, its widespread acceptance as a standard meta-analysis approach remains limited. Although some conventional Bayesian hierarchical models are frequently used for analysis, their performance has not been thoroughly examined. This study evaluates two commonly used Bayesian models for meta-analysis of standardized mean difference and identifies significant issues with these models. In response, we introduce a new Bayesian model equipped with novel features that address existing model concerns and a broader limitation of the current Bayesian meta-analysis. Furthermore, we introduce a simple computational approach to construct simultaneous credible intervals for the summary effect and between-study heterogeneity, based on their joint posterior samples. This fully captures the joint uncertainty in these parameters, a task that is challenging or impractical with frequentist models. Through simulation studies rooted in a joint Bayesian/frequentist paradigm, we compare our model's performance against existing ones under conditions that mirror realistic research scenarios. The results reveal that our new model outperforms others and shows enhanced statistical properties. We also demonstrate the practicality of our models using real-world examples, highlighting how our approach strengthens the robustness of inferences regarding the summary effect.

Exploring the causal association between serum metabolites and erectile dysfunction: a bidirectional Mendelian randomisation study

Erectile dysfunction is a common sexual disorder in men. Some studies have found a strong association between some serum metabolites and erectile dysfunction. To investigate this association further, we used bidirectional Mendelian randomisation to investigate causality and possible biological mechanisms.Firstly, this study screened the statistics of genome-wide association studies of serum metabolites and erectile dysfunction to obtain instrumental variables. Inverse variance weighting was used as the primary method for causal effect analysis of instrumental variables in forward or reverse Mendelian randomisation, and the results obtained by MR-Egger regression and the weighted median method were used as references. Subsequently, the metabolites causally associated with erectile dysfunction were subjected to replication analyses and meta-analyses, and the results of the meta-analyses were analysed by pathway analyses to find influential pathways. In this process, Mendelian randomisation results need to be assessed for stability and reliability using sensitivity analysis.It was found that a total of six serum metabolites were causally associated with erectile dysfunction in a forward Mendelian randomisation study. 1,3,7-trimethyluraten (0.85 (0.73-0.99), P = 0.0368), ergothioneine (0.65 (0.45-0.94), P = 0.0226) and gamma-glutamylglutamate (0.63 (0.46-0.88), P = 0.0059) were protective against the development of erectile dysfunction, whereas 2-hydroxyhippurate (1.10 (1.02-1.19), P = 0.0152), N2,N2-dimethylguanosine (1.57 (1.02-2.40), P = 0.0395) and octanoylcarnitine (1.38 (1.06-1.82), P = 0.0183) were able to induce the development of erectile dysfunction. In addition, metabolic pathway analysis showed that 1,3,7-trimethylurate was able to influence the development of erectile dysfunction via the caffeine metabolism pathway (P = 0.0454). On the other hand, reverse Mendelian randomisation analysis showed that erectile dysfunction reduced serum homocitrulline levels (0.99 (0.97-1.00), P = 0.0360). Sensitivity analyses, including heterogeneity tests and pleiotropy tests, confirmed the reliability of the results.In conclusion, this study demonstrated a bidirectional causal relationship between serum metabolites and erectile dysfunction using bidirectional Mendelian randomisation analysis and replication meta-analysis. On this basis, this study provides a new direction of thinking and strong evidence for the therapeutic application and adjunctive diagnosis of serum metabolites in erectile dysfunction, and provides a certain reference value for subsequent related studies.

Correcting for measurement error under meta-analysis of z-transformed correlations

This study mainly concerns correction for measurement error using the meta-analysis of Fisher's z-transformed correlations. The disattenuation formula of Spearman (American Journal of Psychology, 15, 1904, 72) is used to correct for individual raw correlations in primary studies. The corrected raw correlations are then used to obtain the corrected z-transformed correlations. What remains little studied, however, is how to best correct for within-study sampling error variances of corrected z-transformed correlations. We focused on three within-study sampling error variance estimators corrected for measurement error that were proposed in earlier studies and is proposed in the current study: (1) the formula given by Hedges (Test validity, Lawrence Erlbaum, 1988) assuming a linear relationship between corrected and uncorrected z-transformed correlations (linear correction), (2) one derived by the first-order delta method based on the average of corrected z-transformed correlations (stabilized first-order correction), and (3) one derived by the second-order delta method based on the average of corrected z-transformed correlations (stabilized second-order correction). Via a simulation study, we compared performance of these estimators and the sampling error variance estimator uncorrected for measurement error in terms of estimation and inference accuracy of the mean correlation as well as the homogeneity test of effect sizes. In obtaining the corrected z-transformed correlations and within-study sampling error variances, coefficient alpha was used as a common reliability coefficient estimate. The results showed that in terms of the estimated mean correlation, sampling error variances with linear correction, the stabilized first-order and second-order corrections, and no correction performed similarly in general. Furthermore, in terms of the homogeneity test, given a relatively large average sample size and normal true scores, the stabilized first-order and second-order corrections had type I error rates that were generally controlled as well as or better than the other estimators. Overall, stabilized first-order and second-order corrections are recommended when true scores are normal, reliabilities are acceptable, the number of items per psychological scale is relatively large, and the average sample size is relatively large.

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.

Early-life participation in cognitively stimulating activities and risk of depression and anxiety in late life

BACKGROUND

Early-life stressful experiences are associated with increased risk of adverse psychological outcomes in later life. However, much less is known about associations between early-life positive experiences, such as participation in cognitively stimulating activities, and late-life mental health. We investigated whether greater engagement in cognitively stimulating activities in early life is associated with lower risk of depression and anxiety in late life.

METHODS

We surveyed former participants of the St. Louis Baby Tooth study, between 22 June 2021 and 25 March 2022 to collect information on participants' current depression/anxiety symptoms and their early-life activities (N = 2187 responded). A composite activity score was created to represent the early-life activity level by averaging the frequency of self-reported participation in common cognitively stimulating activities in participants' early life (age 6, 12, 18), each rated on a 1 (least frequent) to 5 (most frequent) point scale. Depression/anxiety symptoms were measured by Patient Health Questionnaire (PHQ-9) and Generalized Anxiety Disorder Screener (GAD-7). We used logistic regressions to estimate odds ratios (OR) and 95% confidence intervals (CI) of outcome risk associated with frequency of early-life activity.

RESULTS

Each one-point increase in the early-life composite cognitive activity score was associated with an OR of 0.54 (95% CI 0.38-0.77) for late-life depression and an OR of 0.94 (95% CI 0.61-1.43) for late-life anxiety, adjusting for age, sex, race, parental education, childhood family structure, and socioeconomic status.

CONCLUSIONS

More frequent participation in cognitively stimulating activities during early life was associated with reduced risk of late-life depression.

Accuracy and precision of fixed and random effects in meta-analyses of randomized control trials for continuous outcomes

Meta-analyses of treatment effects in randomized control trials are often faced with the problem of missing information required to calculate effect sizes and their sampling variances. Particularly, correlations between pre- and posttest scores are frequently not available. As an ad-hoc solution, researchers impute a constant value for the missing correlation. As an alternative, we propose adopting a multivariate meta-regression approach that models independent group effect sizes and accounts for the dependency structure using robust variance estimation or three-level modeling. A comprehensive simulation study mimicking realistic conditions of meta-analyses in clinical and educational psychology suggested that imputing a fixed correlation 0.8 or adopting a multivariate meta-regression with robust variance estimation work well for estimating the pooled effect but lead to slightly distorted between-study heterogeneity estimates. In contrast, three-level meta-regressions resulted in largely unbiased fixed effects but more inconsistent prediction intervals. Based on these results recommendations for meta-analytic practice and future meta-analytic developments are provided.

Adult Organophosphate and Carbamate Insecticide Exposure and Sperm Concentration: A Systematic Review and Meta-Analysis of the Epidemiological Evidence

BACKGROUND

Evidence of the negative impacts of contemporary use insecticides on sperm concentration has increased over the last few decades; however, meta-analyses on this topic are rare.

OBJECTIVES

This investigation assessed the qualitative and quantitative strength of epidemiological evidence regarding adult exposure to two classes of contemporary use insecticides-organophosphates (OPs) and N -methyl carbamates (NMCs)-and sperm concentration using robust and reproducible systematic review and meta-analysis methods.

METHODS

Three scientific databases (PubMed, Scopus, and Web of Science), two U.S. government databases (NIOSHTIC-2 and Science.gov), and five nongovernmental organization websites were searched for relevant primary epidemiological studies published in any language through 11 August 2022. Risk of bias and strength of evidence were evaluated according to Navigation Guide systematic review methodology. Bias-adjusted standardized mean difference effect sizes were calculated and pooled using a three-level, multivariate random-effect meta-analysis model with cluster-robust variance estimation.

RESULTS

Across 20 studies, 21 study populations, 42 effect sizes, and 1,774 adult men, the pooled bias-adjusted standardized mean difference in sperm concentration between adult men more- and less-exposed to OP and NMC insecticides was - 0.30 (95% CI: - 0.49 , - 0.10 ; P Satt < 0.01 ). Sensitivity and subgroup analyses explored statistical heterogeneity and validated the model robustness. Although the pooled effect estimate was modified by risk of bias, insecticide class, exposure setting, and recruitment setting, it remained negative in direction across all meta-analyses. The body of evidence was rated to be of moderate quality, with sufficient evidence of an association between higher adult OP and NMC insecticide exposure and lower sperm concentration.

DISCUSSION

This comprehensive investigation found sufficient evidence of an association between higher OP and NMC insecticide exposure and lower sperm concentration in adults. Although additional cohort studies can be beneficial to fill data gaps, the strength of evidence warrants reducing exposure to OP and NMC insecticides now to prevent continued male reproductive harm. https://doi.org/10.1289/EHP12678.

Estimation of heterogeneity variance based on a generalized Q statistic in meta-analysis of log-odds-ratio

For estimation of heterogeneity varianceτ 2 in meta-analysis of log-odds-ratio, we derive new mean- and median-unbiased point estimators and new interval estimators based on a generalized Q statistic,Q F , in which the weights depend on only the studies' effective sample sizes. We compare them with familiar estimators based on the inverse-variance-weights version of Q ,Q IV . In an extensive simulation, we studied the bias (including median bias) of the point estimators and the coverage (including left and right coverage error) of the confidence intervals. Most estimators add 0.5 to each cell of the 2 × 2 table when one cell contains a zero count; we include a version that always adds 0.5 . The results show that: two of the new point estimators and two of the familiar point estimators are almost unbiased when the total sample size n ≥ 250 and the probability in the Control arm (p iC ) is 0.1, and when n ≥ 100 andp iC is 0.2 or 0.5; for 0.1 ≤ τ 2 ≤ 1 , all estimators have negative bias for small to medium sample sizes, but for larger sample sizes some of the new median-unbiased estimators are almost median-unbiased; choices of interval estimators depend on values of parameters, but one of the new estimators is reasonable whenp iC = 0.1 and another, whenp iC = 0.2 orp iC = 0.5 ; and lack of balance between left and right coverage errors for small n and/orp iC implies that the available approximations for the distributions ofQ IV andQ F are accurate only for larger sample sizes.

On the Q statistic with constant weights in meta-analysis of binary outcomes

BACKGROUND

Cochran's Q statistic is routinely used for testing heterogeneity in meta-analysis. Its expected value (under an incorrect null distribution) is part of several popular estimators of the between-study variance, [Formula: see text]. Those applications generally do not account for use of the studies' estimated variances in the inverse-variance weights that define Q (more explicitly, [Formula: see text]). Importantly, those weights make approximating the distribution of [Formula: see text] rather complicated.

METHODS

As an alternative, we are investigating a Q statistic, [Formula: see text], whose constant weights use only the studies' arm-level sample sizes. For log-odds-ratio (LOR), log-relative-risk (LRR), and risk difference (RD) as the measures of effect, we study, by simulation, approximations to distributions of [Formula: see text] and [Formula: see text], as the basis for tests of heterogeneity.

RESULTS

The results show that: for LOR and LRR, a two-moment gamma approximation to the distribution of [Formula: see text] works well for small sample sizes, and an approximation based on an algorithm of Farebrother is recommended for larger sample sizes. For RD, the Farebrother approximation works very well, even for small sample sizes. For [Formula: see text], the standard chi-square approximation provides levels that are much too low for LOR and LRR and too high for RD. The Kulinskaya et al. (Res Synth Methods 2:254-70, 2011) approximation for RD and the Kulinskaya and Dollinger (BMC Med Res Methodol 15:49, 2015) approximation for LOR work well for [Formula: see text] but have some convergence issues for very small sample sizes combined with small probabilities.

CONCLUSIONS

The performance of the standard [Formula: see text] approximation is inadequate for all three binary effect measures. Instead, we recommend a test of heterogeneity based on [Formula: see text] and provide practical guidelines for choosing an appropriate test at the .05 level for all three effect measures.

Location-scale models for meta-analysis

Heterogeneity is commonplace in meta-analysis. When heterogeneity is found, researchers often aim to identify predictors that account for at least part of such heterogeneity by using mixed-effects meta-regression models. Another potentially relevant goal is to focus on the amount of heterogeneity as a function of one or more predictors, but this cannot be examined with standard random- and mixed-effects models, which assume a constant (i.e., homoscedastic) value for the heterogeneity variance component across studies. In this paper, we describe a location-scale model for meta-analysis as an extension of the standard random- and mixed-effects models that not only allows an examination of whether predictors are related to the size of the outcomes (i.e., their location), but also the amount of heterogeneity (i.e., their scale). We present estimation methods for such a location-scale model through maximum and restricted maximum likelihood approaches, as well as methods for inference and suggestions for visualization. We also provide an implementation via the metafor package for R that makes this model readily available to researchers. Location-scale models can provide a useful tool to researchers interested in heterogeneity in meta-analysis, with the potential to enhance the scope of research questions in the field of evidence synthesis.

On the Q statistic with constant weights for standardized mean difference

Cochran's Q statistic is routinely used for testing heterogeneity in meta-analysis. Its expected value is also used in several popular estimators of the between-study variance, τ 2 . Those applications generally have not considered the implications of its use of estimated variances in the inverse-variance weights. Importantly, those weights make approximating the distribution of Q (more explicitly, Q IV ) rather complicated. As an alternative, we investigate a new Q statistic, Q F , whose constant weights use only the studies' effective sample sizes. For the standardized mean difference as the measure of effect, we study, by simulation, approximations to distributions of Q IV and Q F , as the basis for tests of heterogeneity and for new point and interval estimators of τ 2 . These include new DerSimonian-Kacker-type moment estimators based on the first moment of Q F , and novel median-unbiased estimators. The results show that: an approximation based on an algorithm of Farebrother follows both the null and the alternative distributions of Q F reasonably well, whereas the usual chi-squared approximation for the null distribution of Q IV and the Biggerstaff-Jackson approximation to its alternative distribution are poor; in estimating τ 2 , our moment estimator based on Q F is almost unbiased, the Mandel - Paule estimator has some negative bias in some situations, and the DerSimonian-Laird and restricted maximum likelihood estimators have considerable negative bias; and all 95% interval estimators have coverage that is too high when τ 2 = 0 , but otherwise the Q-profile interval performs very well.

Effectiveness of Roux-en-Y Gastric Bypass vs Sleeve Gastrectomy on Lipid Levels in Type 2 Diabetes: a Meta-analysis

INTRODUCTION

Obesity and its co-morbidities, including type 2 diabetes (T2DM) and dyslipidemia, are accompanied by excess cardiovascular morbi-mortality. Aside from excess low density lipoprotein-cholesterol (LDL-C), atherogenic dyslipidemia (AD), mainly characterized by elevated triglycerides and decreased high density lipoprotein-cholesterol (HDL-C) levels, is often present in T2DM obese patients. Bariatric surgery, such as Roux-en-Y gastric bypass (RYGB) and sleeve gastrectomy (SG), has become a reference treatment in that population. However, the respective effects of RYGB vs SG on lipid metabolism in T2DM patients have been rarely studied.

METHODS

A meta-analysis of randomized controlled trials, comparing the effects of RGYBG vs SG on lipid metabolism 12 months after surgery in T2DM patients, was performed.

RESULTS

Four studies including a total of 298 patients (151 patients in the RYGB and 147 patients in the SG group) were examined. Despite a greater decrease in body mass index and greater improvement in glycemic control in RYGB compared to SG. RYGB vs SG was more effective in reducing total cholesterol, LDL-C, and non-HDL-C levels (mean difference [MD] -26.10 mg/dL, 95 % CI -38.88 to -13.50, p<0.00001; [MD] -20.10 mg/dL, 95 % CI -27.90 to -12.20, p<0.00001 and MD 31.90 mg/dl, 95 % CI -46.90 to -16.80, p<0.00001, respectively).

CONCLUSIONS

The superiority of RYGB vs SG in reducing LDL-C, with an effect comparable to a moderate-intensity statin, suggests RYBG should be favored in hypercholesterolemic T2DM patients in order to further reduce cardiovascular risk.

Comparison of Socioeconomic Disparities in Pump Uptake Among Children With Type 1 Diabetes in 2 Canadian Provinces With Different Payment Models

IMPORTANCE

Insulin pumps improve glycemic control and quality of life in children with type 1 diabetes (T1D). Canada's provinces have implemented universal pediatric programs to improve access. However, these programs provide differing financial coverage, allowing for unique cross-jurisdictional comparisons.

OBJECTIVE

To evaluate possible socioeconomic status (SES) disparities in pump uptake in Québec, where pumps are fully funded, with those in Manitoba, where pumps are partially funded.

DESIGN, SETTING, AND PARTICIPANTS

Using health administrative databases and a clinical registry, parallel, population-based cohort studies of children with diabetes were conducted from April 1, 2011, in Québec, and April 1, 2012, in Manitoba, until March 31, 2017. In analysis conducted from July 1, 2019, to November 30, 2021, multivariable Cox proportional hazards regression models were applied to study the association between pump uptake and SES, defined using validated area-based material and social deprivation indices. Children aged 1 to 17 years with T1D were identified using a validated definition in administrative data (Québec) and a clinical registry (Manitoba). Those using pumps before the initiation of provincial programs were excluded.

EXPOSURES

Socioeconomic status.

MAIN OUTCOMES AND MEASURES

Insulin pump uptake.

RESULTS

A total of 2919 children with T1D were identified in Québec: 1550 male (53.1%), mean (SD) age at diagnosis, 8.3 (4.4) years, and 1067 (36.6%) were using a pump. In Manitoba, 636 children were identified: 364 male (57.2%), mean (SD) age at diagnosis, 8.8 (4.4) years, and 106 (16.7%) were using a pump. In Québec, the mean age at diagnosis of T1D was lower in children using the pump compared with those not using a pump (7.6 [4.1] vs 8.7 [4.5] years); sex distribution was similar (562 [52.7%] vs 988 [53.3%] male). No differences in mean (SD) age at diagnosis (8.8 [4.4] vs 8.8 [4.3] years) or sex (57 [53.8%] vs 307 [57.9%] male) were noted in both groups in Manitoba. Increasing material deprivation was associated with decreased pump uptake in both Québec (adjusted hazard ratio [aHR] 0.89; 95% CI, 0.85-0.93) and Manitoba (aHR, 0.70; 95% CI, 0.60-0.82). Inclusion of ethnic concentration did not change this association. Socioeconomic disparities in pump uptake were greater in Manitoba than Québec (P = .006 by t test; Cochran Q, 8.15; P = .004; I2 = 87.7%; 95% CI, 52.5%-96.8%).

CONCLUSIONS AND RELEVANCE

The results of this study suggest that the program of full coverage for pumps available in Québec partially mitigates observed SES disparities in uptake and may be a model to improve access for all children with T1D.

Incidence and prevalence of pressure injuries in children patients: A systematic review and meta-analysis

BACKGROUND

Pressure injuries presently has been a serious healthcare problem all over the world. Children were recognized as the high-risk population of pressure injuries in the latest prevention and treatment of pressure injuries clinical practice guideline. However, the estimates of incidence, and prevalence of pressure injuries in hospitalized children patients vary considerable in relevant published studies.

OBJECTIVE

To systematically quantify the incidence and prevalence of pressure injuries (PIs) in hospitalized children and the most affected PIs sites.

METHODS

A systematic review and meta-analysis was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-analysis guidelines. Electronic databases searches of the Cochrane Library, Pubmed, Web of Science, Medline, Embase, Cumulative Index to Nursing and Allied Health Literature, China Knowledge Resource Integrated Database (CNKI), Wanfang Database, Chinese Biomedical Database (CBM), and Weipu Database (VIP), and hand-search through references were conducted to find relevant articles. Studies were evaluated independently by two researchers and audited by a third researcher. The data were extracted and presented in tables. The risk of bias was assessed using Hoy's tool. The I² statistic and random-effects model were used to assess the heterogeneity. Meta-regression analysis and subgroup analysis were conducted to examine between-study heterogeneity.

RESULTS

A total of 6, 672 articles were screened, and 30 studies with 251, 501 participants were ultimately included in this review. The pooled incidence of PIs for 3, 205 children was 13.5% (95% CI: 10.5-16.5); and the pooled prevalence of PIs for 4, 639 children was 12.2% (95% CI: 8.0-16.3). The most affected body sites were occiput, ears, and nose. Meta-regression and subgroup analysis showed that the inpatient ward, and region were the sources of heterogeneity.

CONCLUSIONS

The incidence and prevalence of PIs was significantly higher than the adults. Our discoveries recommended that healthcare givers ought to pay more consideration to diminish the happens of PIs. Additionally, more research may be needed to improve our understanding of the characteristics of PIs among children and to identify PIs risk factors to prevent and treat it in children effectively.

Bootstrap-Based Between-Study Heterogeneity Tests in Meta-Analysis

Meta-analysis combines pertinent information from existing studies to provide an overall estimate of population parameters/effect sizes, as well as to quantify and explain the differences between studies. However, testing between-study heterogeneity is one of the most challenging tasks in meta-analysis research. Existing methods for testing heterogeneity, such as the Q test and likelihood ratio (LR) test, have been criticized for their failure to control Type I error rate and/or failure to attain enough statistical power. Although better reference distribution approximations have been proposed in the literature, their application is limited. Additionally, when the interest is to test whether the size of the heterogeneity is larger than a specific level, existing methods are far from mature. To address these issues, we propose new heterogeneity tests. Specifically, we combine bootstrap methods with existing heterogeneity tests (i.e., the maximum LR test, the restricted maximum LR test, and the Q test) to overcome the reference distribution issue and denote them as B-ML-LRT, B-REML-LRT, and B-Q, respectively. Simulation studies were conducted to examine and compare the performance of the proposed methods with the regular LR test, the regular Q test, and the Kulinskaya's improved Q test in both random- and mixed-effects meta-analyses. Based on the results of Type I error rates and statistical power, B-REML-LRT is recommended. Additionally, the improved Q test is also recommended when it is applicable. An R package boot.heterogeneity is provided to facilitate the implementation of the proposed tests.

Comparison of commonly used methods in random effects meta-analysis: application to preclinical data in drug discovery research

Background

Meta-analysis of preclinical data is used to evaluate the consistency of findings and to inform the design and conduct of future studies. Unlike clinical meta-analysis, preclinical data often involve many heterogeneous studies reporting outcomes from a small number of animals. Here, we review the methodological challenges in preclinical meta-analysis in estimating and explaining heterogeneity in treatment effects.

Methods

Assuming aggregate-level data, we focus on two topics: (1) estimation of heterogeneity using commonly used methods in preclinical meta-analysis: method of moments (DerSimonian and Laird; DL), maximum likelihood (restricted maximum likelihood; REML) and Bayesian approach; (2) comparison of univariate versus multivariable meta-regression for adjusting estimated treatment effects for heterogeneity. Using data from a systematic review on the efficacy of interleukin-1 receptor antagonist in animals with stroke, we compare these methods, and explore the impact of multiple covariates on the treatment effects.

Results

We observed that the three methods for estimating heterogeneity yielded similar estimates for the overall effect, but different estimates for between-study variability. The proportion of heterogeneity explained by a covariate is estimated larger using REML and the Bayesian method as compared with DL. Multivariable meta-regression explains more heterogeneity than univariate meta-regression.

Conclusions

Our findings highlight the importance of careful selection of the estimation method and the use of multivariable meta-regression to explain heterogeneity. There was no difference between REML and the Bayesian method and both methods are recommended over DL. Multiple meta-regression is worthwhile to explain heterogeneity by more than one variable, reducing more variability than any univariate models and increasing the explained proportion of heterogeneity.

The Response of Plants and Mycorrhizal Fungi to Nutritionally-Heterogeneous Environments Are Regulated by Nutrient Types and Plant Functional Groups

Nutrient type and plant functional group are both important in influencing proliferation of roots or hyphae and their benefit to plant growth in nutritionally heterogeneous environments. However, the studies quantifying relative importance of roots vs. hyphae affecting the plant response to nutrient heterogeneity are lacking. Here, we used meta-analysis based on 879 observations from 66 published studies to evaluate response patterns of seven variables related to growth and morphological traits of plants and mycorrhizal fungi in nutritionally heterogeneous environments. We found that phosphorus [P] and organic fertilizer [OF] supply significantly increased shoot (+18.1 and +25.9%, respectively) and root biomass (+31.1 and +23.0%, respectively) and root foraging precision (+11.8 and +20.4%, respectively). However, there was no significant difference among functional groups of herbs (grasses, forbs, and legumes), between herbs and woody species, and between arbuscular mycorrhizal (AM) and ectomycorrhizal (ECM) tree species in the shoot, root and mycorrhizal fungi responses to nutrient heterogeneity, except for root biomass and root foraging precision among grasses, forbs, and legumes, and mycorrhizal hyphal foraging precision between AM and ECM tree species. Root diameter was uncorrelated with neither root foraging precision nor mycorrhizal hyphal foraging precision, regardless of mycorrhizal type or nutrient type. These results suggest that plant growth and foraging strategies are mainly influenced by nutrient type, among other factors including plant functional type and mycorrhizal type.

The effects of judicial supervision on recidivism of offenders in Australia and New Zealand: a systematic review and meta-analysis

This meta-analysis compares recidivism reduction in problem-solving courts employing judicial supervision in Australia and New Zealand to traditional processes. Using a four-phased search strategy, 16 studies totalling a treatment sample of 6588 individuals and 32,147 comparison participants were identified from 7161 unique records. Meta-analyses indicate that the problem-solving courts significantly reduced both the odds and incidences of recidivism compared with standard justice processes but that the heterogeneity observed within the latter analysis plus reliance on weak methodologies limits the strength of these conclusions. Studies at risk of bias may have had an undue influence on the odds of recidivism analysis. Additionally, the benefits of treatment on the incidence of recidivism are closely linked to the overlap of measurement and treatment periods. The findings suggest a positive impact from judicial supervision but further rigorous research is needed that closely matches experimental samples, strictly measures participants post-intervention and meticulously reports pertinent information.

Meta-analysis of the effect of expression of MYB transcription factor genes on abiotic stress

Background

MYB proteins are a large group of transcription factors. The overexpression of MYB genes has been reported to improve abiotic stress tolerance in plant. However, due to the variety of plant species studied and the types of gene donors/recipients, along with different experimental conditions, it is difficult to interpret the roles of MYB in abiotic stress tolerance from published data.

Methods

Using meta-analysis approach, we investigated the plant characteristics involved in cold, drought, and salt stress in MYB-overexpressing plants and analyzed the degrees of influence on plant performance by experimental variables.

Results

The results show that two of the four measured plant parameters in cold-stressed plants, two of the six in drought-stressed, and four of the 13 in salt-stressed were significantly impacted by MYB overexpression by 22% or more, and the treatment medium, donor/recipient species, and donor type significantly influence the effects of MYB-overexpression on drought stress tolerance. Also, the donor/recipient species, donor type, and stress duration all significantly affected the extent of MYB-mediated salt stress tolerance. In summary, this study compiles and analyzes the data across studies to help us understand the complex interactions that dictate the efficacy of heterologous MYB expression designed for improved abiotic stress tolerance in plants.

Green Tea Catechin Association with Ultraviolet Radiation-Induced Erythema: A Systematic Review and Meta-Analysis

Catechins are a part of the chemical family of flavonoids, a naturally occurring antioxidant, and a secondary metabolite in certain plants. Green tea catechins are well recognized for their essential anti-inflammatory, photo-protective, antioxidant, and chemo-preventive functions. Ultraviolet radiation is a principal cause of damage to the skin. Studies observed that regular intake of green tea catechins increased the minimal dose of radiation required to induce erythema. The objectives of this systematic review and meta-analysis are to determine the effectiveness of green tea catechins in cutaneous erythema and elucidate whether green tea catechin consumption protects against erythema (sunburn) inflammation. A comprehensive literature search was conducted to identify the relevant studies. Two researchers carried out independent screening, data extraction, and quality assessment according to the guidelines of Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA). The pooled effect of green tea catechins on protection against erythema was assessed using approaches fixed-effects or random-effects model to quantify the effectiveness of green tea catechins in the erythema dose-response. Studies not be included in meta-analyses were summarized narratively. Six randomized controlled studies of enrolled studies regularly administrated green tea catechins orally for 6 to 12 weeks involving healthy volunteers comprising a total of 100 participants were included in the analysis. The results revealed green tea catechins have favorable protection against erythema inflammation even at increased minimal erythema dose (MED) of ultraviolet radiation. Meta-analysis results confirm oral supplementation of green tea catechins is highly effective at low-intensity ultraviolet radiation-induced erythema response (MED range; 1.25-1.30) compared to placebo, showing a significant pooling difference (p = 0.002) in erythema index (SMD: -0.35; 95% CI, -0.57 to -0.13; I² = 4%, p = 0.40) in the random-effects model. The pro-inflammatory signaling pathways through oral supplementation with green tea catechins are an attractive strategy for photo-protection in healthy human subjects and could represent a complementary approach to topical sunscreens. Therefore, studies that involved green tea catechin in topical applications to human subjects were also evaluated separately, and their meta-analysis is presented as a reference. The evidence indicates that regular green tea catechin supplementation is associated with protection against UV-induced damage due to erythema inflammation.

A comparison of hypothesis tests for homogeneity in meta-analysis with focus on rare binary events

Analysis of rare binary events is an important problem for biomedical researchers. Due to the sparsity of events in such problems, meta-analysis that integrates information across multiple studies can be applied to increase the efficiency of statistical inference. Although it is critical to examine whether the effect sizes are homogeneous across all studies, a comprehensive review of homogeneity tests has been lacking, and in particular, no attention has been paid to infrequent dichotomous outcomes. We systematically review statistical methods for homogeneity testing. By conducting an extensive simulation analysis and two case studies, we examine the performance of 30 tests in meta-analysis of rare binary outcomes. When using log-odds ratio as the association measure, our simulation results suggest that there is no uniform winner. However, we recommend the test proposed by Kulinskaya and Dollinger (BMC Med Res Methodol, 2015, 15), which uses a gamma distribution to approximate the null distribution, for its generally good performance; for very rare events coupled with small within-study sample sizes, in addition to the Kulinskaya-Dollinger test, we further recommend the conditional score test based on the random-effects hypergeometric model proposed by Liang and Self (Biometrika, 1985, 72:353-358). One should be cautious about the use of the Wald tests, the Lipsitz tests (Biometrics, 1998, 54:148-160), and tests proposed by Bhaumik et al (J Am Stat Assoc, 2012, 107:555-567).

A Systematic Review and Meta-Analysis of the Diagnostic Accuracy of the Advanced Clinical Solutions Word Choice Test as a Performance Validity Test

Thorough assessment of performance validity has become an established standard of practice in neuropsychological assessment. While there has been a large focus on the development and cross-validation of embedded performance validity tests (PVTs) in recent years, new freestanding PVTs have also been developed, including the Word Choice Test (WCT) as part of the Advanced Clinical Solutions Effort System. And, while the WCT's general utility for identifying invalid performance has been demonstrated in the ensuing decade since its initial publication, optimal cut-scores and associated psychometric properties have varied widely across studies. This study sought to synthesize the existing diagnostic accuracy literature regarding the WCT via a systematic review and to conduct a meta-analysis to determine the performance validity cut-score that best maximizes sensitivity while maintaining acceptable specificity. A systematic search of the literature resulted in 14 studies for synthesis, with eight of those available for meta-analysis. Meta-analytic results revealed an optimal cut-score of ≤ 42 with 54% sensitivity and 93% specificity for identifying invalid neuropsychological test performance. Collectively, the WCT demonstrated adequate diagnostic accuracy as a PVT across a variety of populations. Recommendations for future studies are also provided.

On the finite sample distribution of the likelihood ratio statistic for testing heterogeneity in meta-analysis

In meta-analysis, hypothesis testing is one of the commonly used approaches for assessing whether heterogeneity exists in effects between studies. The literature concluded that the Q-statistic is clearly the best choice and criticized the performance of the likelihood ratio test in terms of the type I error control and power. However, all the criticism for the likelihood ratio test is based on the use of a mixture of two chi-square distributions with 0 and 1 degrees of freedom, which is justified only asymptotically. In this study, we develop a novel method to derive the finite sample distribution of the likelihood ratio test and restricted likelihood ratio test statistics for testing the zero variance component in the random effects model for meta-analysis. We also extend this result to the heterogeneity test when metaregression is applied. A numerical study shows that the proposed statistics have superior performance to the Q-statistic, especially when the number of studies collected for meta-analysis is small to moderate.

The influence of nonnormality from primary studies on the standardized mean difference in meta-analysis

In this study we investigated the influence of data nonnormality in the primary studies on meta-analysis of the standardized mean difference (SMD) for a two-independent-group design. The bias, mean squared error, and confidence interval coverage probability of the mean effect sizes under different types of population distributions were compared. Also, the performance of the Q test was examined. The results showed that oppositely skewed distributions (i.e., distributions skewed in different directions) showed poor performance for point and interval estimates of mean effect sizes in meta-analysis, especially when the tails were pointing toward each other. The previously found adverse impacts due to nonnormality in primary studies do not disappear when primary studies with nonnormal data are meta-analyzed, even when the average sample size and number of studies are large. The results also showed that, when the tails were pointing toward each other, the Type I error rates of the Q test were inflated. We suggest that the impact of violating the assumption of normality should not be ignored in meta-analysis.

Fitness consequences of different migratory strategies in partially migratory populations: A multi-taxa meta-analysis

Partial migration—wherein migratory and non‐migratory individuals exist within the same population—represents a behavioural dimorphism; for it to persist over time, both strategies should yield equal individual fitness. This balance may be maintained through trade‐offs where migrants gain survival benefits by avoiding unfavourable conditions, while residents gain breeding benefits from early access to resources.

There has been little overarching quantitative analysis of the evidence for this fitness balance. As migrants—especially long‐distance migrants—may be particularly vulnerable to environmental change, it is possible that recent anthropogenic impacts could drive shifts in fitness balances within these populations.

We tested these predictions using a multi‐taxa meta‐analysis. Of 2,939 reviewed studies, 23 contained suitable information for meta‐analysis, yielding 129 effect sizes.

Of these, 73% (n = 94) reported higher resident fitness, 22% (n = 28) reported higher migrant fitness, and 5% (n = 7) reported equal fitness. Once weighted for precision, we found balanced fitness benefits across the entire dataset, but a consistently higher fitness of residents over migrants in birds and herpetofauna (the best‐sampled groups). Residency benefits were generally associated with survival, not breeding success, and increased with the number of years of data over which effect sizes were calculated, suggesting deviations from fitness parity are not due to sampling artefacts.

A pervasive survival benefit to residency documented in recent literature could indicate that increased exposure to threats associated with anthropogenic change faced by migrating individuals may be shifting the relative fitness balance between strategies.

Overdispersion models for correlated multinomial data: Applications to blinding assessment

Overdispersion models have been extensively studied for correlated normal and binomial data but much less so for correlated multinomial data. In this work, we describe a multinomial overdispersion model that leads to the specification of the first two moments of the outcome and allows the estimation of the global parameters using generalized estimating equations (GEE). We introduce a Global Blinding Index as a target parameter and illustrate the application of the GEE method to its estimation from (1) a clinical trial with clustering by practitioner and (2) a meta-analysis on psychiatric disorders. We examine the impact of a small number of clusters, high variability in cluster sizes, and the magnitude of the intraclass correlation on the performance of the GEE estimators of the Global Blinding Index using the data simulated from different models. We compare these estimators with the inverse-variance weighted estimators and a maximum-likelihood estimator, derived under the Dirichlet-multinomial model. Our results indicate that the performance of the GEE estimators was satisfactory even in situations with a small number of clusters, whereas the inverse-variance weighted estimators performed poorly, especially for larger values of the intraclass correlation coefficient. Our findings and illustrations may be instrumental for practitioners who analyze clustered multinomial data from clinical trials and/or meta-analysis.

Postmortem imaging findings and cause of death determination compared with autopsy: a systematic review of diagnostic test accuracy and meta-analysis

BACKGROUND

The aim of this study was to evaluate the sensitivity of postmortem computed tomography (PMCT), postmortem magnetic resonance imaging (PMMR) and PMCT angiography (PMCTA) compared with autopsy in cases of adult death investigations.

METHODS

For this systematic review and meta-analysis, Embase, PubMed, Scopus, Web of Science and Medline were searched for eligible studies in October 2016; a follow-up literature search was conducted in March 2018. Studies referring to PMCT, PMCTA and/or PMMR of more than 3 cases with subsequent autopsy were included. Data were extracted from published texts in duplicate. The extracted outcomes were categorized as follows: soft tissue and organ findings, skeletal injuries, haemorrhages, abnormal gas accumulations and causes of death. The summary measure was sensitivity, if 3 or more studies were available. To combine studies, a random effects model was used. Variability and heterogeneity within the meta-analysis was assessed.

RESULTS

Of 1053 studies, 66 were eligible, encompassing a total of 4213 individuals. For soft tissue and organ findings, there was a high pooled sensitivity with PMCTA (0.91, 95% CI 0.81-0.96), without evidence for between-study variability (Cochrane's Q test p = 0.331, I² = 24.5%). The pooled sensitivity of PMCT+PMMR was very high in skeletal injuries (0.97, CI 0.87-0.99), without evidence for variability (p = 0.857, I² = 0.0%). In detecting haemorrhages, the pooled sensitivity for PMCT+PMMR was the highest (0.88, 95% CI 0.35-0.99), with strong evidence of heterogeneity (p < 0.05, I² > 50%). Pooled sensitivity for the correct cause of death was the highest for PMCTA with 0.79 (95% CI 0.52-0.93), again with evidence of heterogeneity (p = 0.062, I² > 50%).

CONCLUSION

Distinct postmortem imaging modalities can achieve high sensitivities for detecting various findings and causes of death. This knowledge should lead to a reasoned use of each modality. Both forensic evidence and in-hospital medical quality would be enhanced.

Publication bias examined in meta-analyses from psychology and medicine: A meta-meta-analysis

Publication bias is a substantial problem for the credibility of research in general and of meta-analyses in particular, as it yields overestimated effects and may suggest the existence of non-existing effects. Although there is consensus that publication bias exists, how strongly it affects different scientific literatures is currently less well-known. We examined evidence of publication bias in a large-scale data set of primary studies that were included in 83 meta-analyses published in Psychological Bulletin (representing meta-analyses from psychology) and 499 systematic reviews from the Cochrane Database of Systematic Reviews (CDSR; representing meta-analyses from medicine). Publication bias was assessed on all homogeneous subsets (3.8% of all subsets of meta-analyses published in Psychological Bulletin) of primary studies included in meta-analyses, because publication bias methods do not have good statistical properties if the true effect size is heterogeneous. Publication bias tests did not reveal evidence for bias in the homogeneous subsets. Overestimation was minimal but statistically significant, providing evidence of publication bias that appeared to be similar in both fields. However, a Monte-Carlo simulation study revealed that the creation of homogeneous subsets resulted in challenging conditions for publication bias methods since the number of effect sizes in a subset was rather small (median number of effect sizes equaled 6). Our findings are in line with, in its most extreme case, publication bias ranging from no bias until only 5% statistically nonsignificant effect sizes being published. These and other findings, in combination with the small percentages of statistically significant primary effect sizes (28.9% and 18.9% for subsets published in Psychological Bulletin and CDSR), led to the conclusion that evidence for publication bias in the studied homogeneous subsets is weak, but suggestive of mild publication bias in both psychology and medicine.

Risk of heat illness in men and women: A systematic review and meta-analysis

BACKGROUND

Heat illness (HI) is a growing global concern; its incidence has risen dramatically across the world in recent years. The individual factors whereby elevated core temperature produces HI are not well-understood. Given known physiological differences between men and women pertaining to temperature regulation, we hypothesized that women would be at increased risk of HI than men.

OBJECTIVES

We aimed to determine the relative risk of HI in women compared with men through an exhaustive literature review and meta-analysis.

METHODS

We search PubMed and Ovid Medline databases from inception to Apr 2017. Search terms included all permutations of sex and heat illness (including heatstroke and exertional heat illness) with no language restrictions. We included adult or adolescent human data reporting comparable male and female HI rates. One reviewer identified and screened titles and abstracts. Two independent reviewers applied eligibility criteria. Disagreements were resolved with a third reviewer.

RESULTS

Of 5888 articles identified by searches, 36 were included in the systematic review and 22 in the meta-analysis. The mean (standard deviation) quality score was 3.31(1.25)/5. Overall the rate among women was consistently lower than men across the lifespan. The male: female pooled IRR was 2.28 (p < 0.001, 95% CI: 1.66-3.16). There was modest heterogeneity (between-studies variance (τ²) = 0.02). The rates did not differ significantly when corrected for severity or occupation.

DISCUSSION

The rate of HI was significantly increased in men compared with women. Risk for HI might be conferred by psychological and behavioral factors rather than physiological ones. Further research is required to delineate which groups are at greatest risk, leading to the development of mitigation strategies against HI.

OTHER

No funding was received. The authors acknowledge the support of the UK Women in Ground Close Combat Review. The Study was registered with PROSPREO CRD42017064739.

Statistical properties of methods based on the Q-statistic for constructing a confidence interval for the between-study variance in meta-analysis

The effect sizes of studies included in a meta‐analysis do often not share a common true effect size due to differences in for instance the design of the studies. Estimates of this so‐called between‐study variance are usually imprecise. Hence, reporting a confidence interval together with a point estimate of the amount of between‐study variance facilitates interpretation of the meta‐analytic results. Two methods that are recommended to be used for creating such a confidence interval are the Q‐profile and generalized Q‐statistic method that both make use of the Q‐statistic. These methods are exact if the assumptions underlying the random‐effects model hold, but these assumptions are usually violated in practice such that confidence intervals of the methods are approximate rather than exact confidence intervals. We illustrate by means of two Monte‐Carlo simulation studies with odds ratio as effect size measure that coverage probabilities of both methods can be substantially below the nominal coverage rate in situations that are representative for meta‐analyses in practice. We also show that these too low coverage probabilities are caused by violations of the assumptions of the random‐effects model (ie, normal sampling distributions of the effect size measure and known sampling variances) and are especially prevalent if the sample sizes in the primary studies are small.

A flexible approach to identify interaction effects between moderators in meta-analysis

In meta‐analytic studies, there are often multiple moderators available (eg, study characteristics). In such cases, traditional meta‐analysis methods often lack sufficient power to investigate interaction effects between moderators, especially high‐order interactions. To overcome this problem, meta‐CART was proposed: an approach that applies classification and regression trees (CART) to identify interactions, and then subgroup meta‐analysis to test the significance of moderator effects. The aim of this study is to improve meta‐CART upon two aspects: 1) to integrate the two steps of the approach into one and 2) to consistently take into account the fixed‐effect or random‐effects assumption in both the the interaction identification and testing process. For fixed effect meta‐CART, weights are applied, and subgroup analysis is adapted. For random effects meta‐CART, a new algorithm has been developed. The performance of the improved meta‐CART was investigated via an extensive simulation study on different types of moderator variables (ie, dichotomous, nominal, ordinal, and continuous variables). The simulation results revealed that the new method can achieve satisfactory performance (power greater than 0.80 and Type I error less than 0.05) if appropriate pruning rule is applied and the number of studies is large enough. The required minimum number of studies ranges from 40 to 120 depending on the complexity and strength of the interaction effects, the within‐study sample size, the type of moderators, and the residual heterogeneity.

Random or Fixed? An Empirical Examination of Meta-Analysis Model Choices

When conducting meta-analyses, researchers must make decisions about which statistical model is most appropriate for the specific context and aims of the meta-analysis. Although there are several meta-analysis models, most researchers choose between two general models: fixed-effect (FE) and random-effects (RE). Yet, the basis on which these two general models are distinguished and of when it is appropriate to use one or the other varies in the methodological literature. Although model-to-inference inconsistencies have been previously noted, there has been little empirical investigation of whether, and to what extent, the varying conceptualizations of the distinctions between FE and RE models are reflected in published meta-analyses. The present study explores whether conceptualizations of model distinctions among psychological researchers are consistent with those in the methods literature. We also examine model choices and rationales given by psychological researchers in two samples of published meta-analyses in psychology-related journals. We identify four primary categories for distinguishing between FE and RE models, only two of which were predominant in our samples. Although model choice appears to be reported at a moderately high rate, many researchers continue not to provide explicit rationales for their model choices or do not clearly tie model choices to the specific research aims of the meta-analyses. Implications of these findings are discussed.

Reality orientation therapy benefits cognition in older people with dementia: A meta-analysis

BACKGROUND/AIMS

To investigate the effect of reality orientation therapy on cognition, behavioral problems, and depressive symptoms in older people with dementia.

DESIGN

Meta-analysis of randomized controlled trials (RCTs).

DATA SOURCE

Various electronic databases were searched to select eligible articles, which included CINAHL, Embase, Pubmed, ProQuest, PsycINFO, the Cochrane Central Register of Controlled Trials, Google Scholar and Chinese electronic databases comprising Chinese Electronic Periodical Services and China Knowledge Resource Integrated.

REVIEW METHODS

Two reviewers independently extracted data and evaluated the methodological quality of the included articles. All analyses were performed using a random-effects model. Moderator analyses were conducted to identify the superior subgroup of intervention details that benefited from reality orientation therapy.

RESULTS

There was an overall random effect of reality orientation therapy of improving cognitive functions (n = 11, g = 0.39, 95% confidence interval [CI] = 0.25, 0.53); however, reality orientation therapy did not ameliorate behavioral problems (n = 5, g= -0.18, 95% CI= -0.36, 0.00) and depressive symptoms (n = 5, g= -0.17, 95% CI= -0.39, 0.05). A subgroup analysis revealed that studies of reality orientation therapy alone had a greater improvement in cognitive functions compared to those studies adopting reality orientation therapy combined with other types of techniques, but the difference in effect sizes between the two approaches was not significant (g = 0.51 vs. 0.35, p = 0.333).

CONCLUSIONS

Reality orientation therapy can be considered routine care for the maintenance and improvement of cognitive functions in older populations with dementia. Health professionals and caregivers can be further educated and trained to regularly perform reality orientation therapy to improve the cognition of older people with dementia.

Prevalence, Risk Factors, Outcomes, and Treatment of Obstructive Sleep Apnea in Patients with Cerebrovascular Disease: A Systematic Review

BACKGROUND

Obstructive sleep apnea (OSA) is known to increase the risk of cerebrovascular disease (CVD), and patients with CVD have high incidence of OSA. The study aimed to systematically evaluate the prevalence of OSA in patients with CVD.

MATERIALS AND METHODS

Medline, Embase, Science Citation Index, Wanfang, CNKI, and Wiley Online Library were thoroughly searched to identify relevant studies. Random-effects models were used to calculate the pooled rate estimates. Meta-regression and subgroup analysis were performed to explore potential sources of heterogeneity.

RESULTS

Thirty-seven studies with 3242 patients were analyzed. The prevalence of OSA (apnea hypopnea index [AHI] >10) ranged from 34.5% to 92.3%, the random-effects pooled prevalence was 61.9%. Furthermore, the prevalence of sleep disordered breathing (SDB) with AHI  greater than  5 was 70.4%, with AHI greater than 20 was 39.5%, and with AHI greater than 30 was 30.1%. Only 8.3% of the SDB was primarily central apnea. Seventeen studies reported risk factors for OSA, 6 of which used multivariate analyses to extract risk factors. In univariate meta-regression analysis, male had higher prevalence than female (P = .041). OSA was associated with increased length of hospitalization in 2 studies, and 1 long-term study reported severe sleep apnea was associated with poor functional outcome. Among the 5 studies on treatment, 3 indicated that early treatment with CPAP was effective; the remaining studies did not find benefit from CPAP treatment and reported the CPAP acceptance was poor.

CONCLUSIONS

There is high prevalence of OSA in patients with CVD (61.9%). Therefore, accurate diagnosis and treatment to OSA is very important so as to prevent CVD.

On the Multilevel Nature of Meta-Analysis: A Tutorial, Comparison of Software Programs, and Discussion of Analytic Choices

The term "multilevel meta-analysis" is encountered not only in applied research studies, but in multilevel resources comparing traditional meta-analysis to multilevel meta-analysis. In this tutorial, we argue that the term "multilevel meta-analysis" is redundant since all meta-analysis can be formulated as a special kind of multilevel model. To clarify the multilevel nature of meta-analysis the four standard meta-analytic models are presented using multilevel equations and fit to an example data set using four software programs: two specific to meta-analysis (metafor in R and SPSS macros) and two specific to multilevel modeling (PROC MIXED in SAS and HLM). The same parameter estimates are obtained across programs underscoring that all meta-analyses are multilevel in nature. Despite the equivalent results, not all software programs are alike and differences are noted in the output provided and estimators available. This tutorial also recasts distinctions made in the literature between traditional and multilevel meta-analysis as differences between meta-analytic choices, not between meta-analytic models, and provides guidance to inform choices in estimators, significance tests, moderator analyses, and modeling sequence. The extent to which the software programs allow flexibility with respect to these decisions is noted, with metafor emerging as the most favorable program reviewed.

Meta-CART: A tool to identify interactions between moderators in meta-analysis

In the framework of meta-analysis, moderator analysis is usually performed only univariately. When several study characteristics are available that may account for treatment effect, standard meta-regression has difficulties in identifying interactions between them. To overcome this problem, meta-CART has been proposed: an approach that applies classification and regression trees (CART) to identify interactions, and then subgroup meta-analysis to test the significance of moderator effects. The previous version of meta-CART has its shortcomings: when applying CART, the sample sizes of studies are not taken into account, and the effect sizes are dichotomized around the median value. Therefore, this article proposes new meta-CART extensions, weighting study effect sizes by their accuracy, and using a regression tree to avoid dichotomization. In addition, new pruning rules are proposed. The performance of all versions of meta-CART was evaluated via a Monte Carlo simulation study. The simulation results revealed that meta-regression trees with random-effects weights and a 0.5-standard-error pruning rule perform best. The required sample size for meta-CART to achieve satisfactory performance depends on the number of study characteristics, the magnitude of the interactions, and the residual heterogeneity.

Meta-analysis reveals that seed-applied neonicotinoids and pyrethroids have similar negative effects on abundance of arthropod natural enemies

Background

Seed-applied neonicotinoids are widely used in agriculture, yet their effects on non-target species remain incompletely understood. One important group of non-target species is arthropod natural enemies (predators and parasitoids), which contribute considerably to suppression of crop pests. We hypothesized that seed-applied neonicotinoids reduce natural-enemy abundance, but not as strongly as alternative insecticide options such as soil- and foliar-applied pyrethroids. Furthermore we hypothesized that seed-applied neonicotinoids affect natural enemies through a combination of toxin exposure and prey scarcity.

Methods

To test our hypotheses, we compiled datasets comprising observations from randomized field studies in North America and Europe that compared natural-enemy abundance in plots that were planted with seed-applied neonicotinoids to control plots that were either (1) managed without insecticides (20 studies, 56 site-years, 607 observations) or (2) managed with pyrethroid insecticides (eight studies, 15 site-years, 384 observations). Using the effect size Hedge’s d as the response variable, we used meta-regression to estimate the overall effect of seed-applied neonicotinoids on natural-enemy abundance and to test the influence of potential moderating factors.

Results

Seed-applied neonicotinoids reduced the abundance of arthropod natural enemies compared to untreated controls (d = −0.30 ± 0.10 [95% confidence interval]), and as predicted under toxin exposure this effect was stronger for insect than for non-insect taxa (Q_M = 8.70, df = 1, P = 0.003). Moreover, seed-applied neonicotinoids affected the abundance of arthropod natural enemies similarly to soil- or foliar-applied pyrethroids (d = 0.16 ± 0.42 or −0.02 ± 0.12; with or without one outlying study). Effect sizes were surprisingly consistent across both datasets (I₂ = 2.7% for no-insecticide controls; I₂ = 0% for pyrethroid controls), suggesting little moderating influence of crop species, neonicotinoid active ingredients, or methodological choices.

Discussion

Our meta-analysis of nearly 1,000 observations from North American and European field studies revealed that seed-applied neonicotinoids reduced the abundance of arthropod natural enemies similarly to broadcast applications of pyrethroid insecticides. These findings suggest that substituting pyrethroids for seed-applied neonicotinoids, or vice versa, will have little net affect on natural enemy abundance. Consistent with previous lab work, our results also suggest that seed-applied neonicotinoids are less toxic to spiders and mites, which can contribute substantially to biological control in many agricultural systems. Finally, our ability to interpret the negative effect of neonicotinoids on natural enemies is constrained by difficulty relating natural-enemy abundance to biological control function; this is an important area for future study.

Physiological effects of epigallocatechin-3-gallate (EGCG) on energy expenditure for prospective fat oxidation in humans: A systematic review and meta-analysis

Green tea catechins (GTCs) are known to improve fat oxidation (FOX) during fasted, rested and exercise conditions wherein epigallocatechin-3-gallate (EGCG) is thought to be the most pharmacologically active and has been studied extensively. From the available data of randomized controlled trials (RCTs) on EGCG, we carried out a systematic review and meta-analysis to elucidate whether EGCG consumption indeed increase energy expenditure (EE) and promote FOX. A systematic review of the literature was conducted using electronic databases (PubMed, Embase, Cochrane Library, CINAHL, JICST, JSTPLUS, and JMEDPLUS and others) and eight RCTs were included. RCTs were reviewed using Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and methodological quality was assessed. After data extraction, results were aggregated using fixed- and random-effect approaches and expressed to quantify the relationship between the dose of EGCG for respiratory quotient (RQ), EE and rate of FOX to compare the EGCG and placebo treatments. The meta-analysis results of verities of studies in terms of dose and length of duration revealed that EGCG supplementation provided significant mean difference (MD) when compared with placebo for RQ [MD: -0.02; 95% confidence intervals (95% CI), -0.04 to 0.00; I²=67%; P=.01] and EE [MD: 158.05 kJ/day; 95% CI, 4.72 to 311.38; I²=0%; P=.04] in fixed-effect approach. Changes in FOX did not reach the level of statistical significance. Meta-analyses of EGCG influence on the body mass index, waist circumference and total body fat mass (TBFM) were also examined and their impact on the promotion of FOX is reported. Effect of EGCG doses was also systematically reviewed. Finding showed that EGCG intake moderately accelerates EE and reduces RQ. The analyses revealed that the EGCG resulted in difference in RQ and EE but the effect on the other measures of energy metabolism was relatively mild. Possibly, EGCG alone has the potential to increase metabolic rate at 300 mg dose. Collectively, the outcome supports the findings that EGCG has an effect on metabolic parameters. However, the large prospective trials are needed to confirm the findings.

Pervasive and strong effects of plants on soil chemistry: a meta-analysis of individual plant 'Zinke' effects

Plant species leave a chemical signature in the soils below them, generating fine-scale spatial variation that drives ecological processes. Since the publication of a seminal paper on plant-mediated soil heterogeneity by Paul Zinke in 1962, a robust literature has developed examining effects of individual plants on their local environments (individual plant effects). Here, we synthesize this work using meta-analysis to show that plant effects are strong and pervasive across ecosystems on six continents. Overall, soil properties beneath individual plants differ from those of neighbours by an average of 41%. Although the magnitudes of individual plant effects exhibit weak relationships with climate and latitude, they are significantly stronger in deserts and tundra than forests, and weaker in intensively managed ecosystems. The ubiquitous effects of plant individuals and species on local soil properties imply that individual plant effects have a role in plant-soil feedbacks, linking individual plants with biogeochemical processes at the ecosystem scale.

On the moments of Cochran's Q statistic under the null hypothesis, with application to the meta-analysis of risk difference

W. G. Cochran's Q statistic was introduced in 1937 to test for equality of means under heteroscedasticity. Today, the use of Q is widespread in tests for homogeneity of effects in meta-analysis, but often these effects (such as risk differences and odds ratios) are not normally distributed. It is common to assume that Q follows a chi-square distribution, but it has long been known that this asymptotic distribution for Q is not accurate for moderate sample sizes. In this paper, the effect and weight for an individual study may depend on two parameters: the effect and a nuisance parameter. We present expansions for the first two moments of Q without any normality assumptions. Our expansions will have wide applicability in testing for homogeneity in meta-analysis. As an important example, we present a homogeneity test when the effects are the differences of risks between treatment and control arms of the several studies-a test which is substantially more accurate than that currently used. In this situation, we approximate the distribution of Q with a gamma distribution. We provide the results of simulations to verify the accuracy of our proposal and an example of a meta-analysis of medical data. Copyright © 2012 John Wiley & Sons, Ltd.

Random-effects meta-analysis: the number of studies matters

This paper investigates the impact of the number of studies on meta-analysis and meta-regression within the random-effects model framework. It is frequently neglected that inference in random-effects models requires a substantial number of studies included in meta-analysis to guarantee reliable conclusions. Several authors warn about the risk of inaccurate results of the traditional DerSimonian and Laird approach especially in the common case of meta-analysis involving a limited number of studies. This paper presents a selection of likelihood and non-likelihood methods for inference in meta-analysis proposed to overcome the limitations of the DerSimonian and Laird procedure, with a focus on the effect of the number of studies. The applicability and the performance of the methods are investigated in terms of Type I error rates and empirical power to detect effects, according to scenarios of practical interest. Simulation studies and applications to real meta-analyses highlight that it is not possible to identify an approach uniformly superior to alternatives. The overall recommendation is to avoid the DerSimonian and Laird method when the number of meta-analysis studies is modest and prefer a more comprehensive procedure that compares alternative inferential approaches. R code for meta-analysis according to all of the inferential methods examined in the paper is provided.

An empirical investigation of partial effect sizes in meta-analysis of correlational data

The partial correlation and the semi-partial correlation can be seen as measures of partial effect sizes for the correlational family. Thus, both indices have been used in the meta-analysis literature to represent the relationship between an outcome and a predictor of interest, controlling for the effect of other variables in the model. This article evaluates the accuracy of synthesizing these two indices under different situations. Both partial correlation and the semi-partial correlation appear to behave as expected with respect to bias and root mean squared error (RMSE). However, the partial correlation seems to outperform the semi-partial correlation regarding Type I error of the homogeneity test (Q statistic). Although further investigation is needed to fully understand the impact of meta-analyzing partial effect sizes, the current study demonstrates the accuracy of both indices.

Metaanalyse–praktische Schritte und Entscheidungen im Umsetzungsprozess

In diesem Beitrag wird der Durchführungsprozess metaanalytischer Techniken nach Hunter und Schmidt (2004) Schritt für Schritt beschrieben. In Form eines Tutoriums geben wir evidenzbasierte Empfehlungen, verweisen auf relevante Quellen und Hilfsmittel und bewerten alternative Vorgehensweisen in Bezug auf ihre Güte und Akzeptanz. Wir wenden uns an Forschende, die eine Metaanalyse mit der Besonderheit der Artefaktkorrekturen nach Hunter und Schmidt durchführen wollen und geben Anregungen für die Verbreitung metaanalytischer Ergebnisse sowohl im Wissenschaftskontext als auch im Austausch mit Praktikern.