Heterogeneity In Meta-analysis Research Articles

Assessment of heterogeneity in meta-analysis

Assessment of heterogeneity in meta-analysis

Comparisons of various estimates of the 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 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 to interpret the extent of heterogeneity, it is essential to account for its interval estimate. Various point and interval estimators exist for . This article summarizes these estimators. In addition, we performed a simulation study under different scenarios to investigate preferable point and interval estimates of . We found that the Sidik-Jonkman method gave precise point estimates for when the between-study variance was large, while in other cases, the DerSimonian-Laird method was suggested to estimate . When the effect measure was the mean difference or the standardized mean difference, the -profile method, the Biggerstaff-Jackson method, or the Jackson method was suggested to calculate the interval estimate for 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 when the effect measure was the log odds ratio.

Orbital CT scan parameters in dysthyroid optic neuropathy: a systematic review and meta-analysis.

Addressing Dysthyroid Optic Neuropathy (DON) is crucial due to its debilitating impact in thyroid eye disease (TED). Prompt treatment can preserve vision. Despite lacking definitive diagnostic criteria, computed tomography's (CT) parameters are commonly used for diagnosis. However, these parameters exist without consensus on their diagnostic performance. Systematic review and meta-analysis. We conducted a meta-analysis of studies assessing orbital CT diagnostic performance for DON in adults with TED. We searched various databases including Medline, PubMed, Scopus, and EMBASE, and others electronic databases, until July 2023. Evaluated CT parameters includes Barrett index (BI), fat prolapse via superior-orbital-fissure (SOF), superior-ophthalmic-vein-dilatation (SOVD), and the Nugent score. Diagnostic Test Accuracy analysis (DTA) was performed using R. A total of 9 articles with documented target parameters, collectively analysed 212 orbits with DON. Nugent score exhibited highest diagnostic ability with a log diagnostic odd ratio (logDOR) of 2.64 (95% CI, 2.02, 3.25). Another significant DON indicator was a BI ≥ 50%, with a logDOR of 1.97 (95% CI, 1.17; 2.77). Conversely, fat prolapse via SOF and SOVD proved less sensitive, with a logDOR of 1.42 and 1.09 respectively. Regarding the SROC curve, Nugent score and the BI have the greatest AUC. Variations in study locale, participant demographics, and measurement methods accounted for heterogeneity in meta-analysis. Nugent score and a BI ≥ 50% prove to be significant diagnostic parameters for DON, distinguishing them from fat prolapse via SOF and SOVD. Prioritizing these parameters can lead to prompt treatment and thus enhanced visual outcomes. CRD42023446376.

Confidence intervals for heterogeneity in meta-analysis of the rare binary events based on empirical likelihood-type methods

In meta-analysis, heterogeneity between independent studies is one of the important reference indicators for comprehensive analyses. It is usually described in terms of the variance between groups of the random-effect model. The existing methods are used to construct their confidence intervals based on the assumption of the normal distribution, but in reality, this assumption is often violated, especially in rare binary events. Recently, the method of the jackknife empirical likelihood was proposed to build confidence intervals; however, it tends to perform not as good as the parametric methods when the number of studies involved is small. Therefore, we propose to use the adjusted jackknife empirical likelihood, transformed jackknife empirical likelihood, and transformed adjusted jackknife empirical likelihood to construct confidence intervals for heterogeneity of the rare binary event. We investigate the performance of the proposed methods through the simulations based on different distributions, especially for skewed distributions and the small sample sizes. Applications to real data with the proposed methods are also provided.

Carbon dioxide laser therapy for the management of genitourinary syndrome of menopause: A meta‑analysis of randomized controlled trials.

Genitourinary symptoms of menopause (GSM) affect ~50% of women after menopause. Recently, CO2 laser therapy has been used for managing GSM but without high quality evidence. The present review assessed the effectiveness of CO2 laser therapy in the management of GSM. PubMed, Embase, Web of Science, CENTRAL and Scopus databases were searched for randomized controlled trials (RCTs), published up to June 30, 2023, comparing CO2 laser and sham laser treatments for GSM management. The outcomes of interest included Female Sexual Function Index (FSFI), Vaginal Health Index (VHI) and visual analog scale (VAS) for dyspareunia, dryness, burning, itching and dysuria. A total of seven RCTs were included in the review and meta-analysis, with 6/7 studies using three sessions of laser therapy, 4-8 weeks apart. Meta-analysis demonstrated no statistically significant difference in FSFI [mean difference (MD), -1.48; 95% CI, -5.85, 2.89; I2=45%] and VHI scores (MD, -0.18; 95% CI, -1.66, 1.31; I2 =72%) between laser and control groups. Meta-analysis also demonstrated no statistically significant difference in VAS scores for dyspareunia (MD, -1.63; 95% CI; -4.06, 0.80; I2=91%), dryness (MD, -1.30; 95% CI, -3.14, 0.53; I2=75%), burning (MD, -0.76; 95% CI, -2.03; 0.51 I2=56%), itching (MD, -0.28; 95% CI, -0.95, 0.38; I2=0%) and dysuria (MD, 0.15; 95% CI, -0.37, 0.67; I2=23%) between the groups. The included RCTs had low risk of bias. In conclusion, meta-analyses of high-quality sham-controlled RCTs indicated that CO2 may not have any beneficial effect on GSM. Limited data and high heterogeneity in meta-analyses in this area of research are important limitations that need to be addressed by future RCTs.

Combining meta-analysis with multiple imputation for one-step, privacy-protecting estimation of causal treatment effects in multi-site studies.

Missing data complicates statistical analyses in multi-site studies, especially when it is not feasible to centrally pool individual-level data across sites. We combined meta-analysis with within-site multiple imputation for one-step estimation of the average causal effect (ACE) of a target population comprised of all individuals from all data-contributing sites within a multi-site distributed data network, without the need for sharing individual-level data to handle missing data. We considered two orders of combination and three choices of weights for meta-analysis, resulting in six approaches. The first three approaches, denoted as RR + metaF, RR + metaR and RR + std, first combined results from imputed data sets within each site using Rubin's rules and then meta-analyzed the combined results across sites using fixed-effect, random-effects and sample-standardization weights, respectively. The last three approaches, denoted as metaF + RR, metaR + RR and std + RR, first meta-analyzed results across sites separately for each imputation and then combined the meta-analysis results using Rubin's rules. Simulation results confirmed very good performance of RR + std and std + RR under various missing completely at random and missing at random settings. A direct application of the inverse-variance weighted meta-analysis based on site-specific ACEs can lead to biased results for the targeted network-wide ACE in the presence of treatment effect heterogeneity by site, demonstrating the need to clearly specify the target population and estimand and properly account for potential site heterogeneity in meta-analyses seeking to draw causal interpretations. An illustration using a large administrative claims database is presented.

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

BackgroundCochran’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, tau ^2. Those applications generally do not account for use of the studies’ estimated variances in the inverse-variance weights that define Q (more explicitly, Q_{IV}). Importantly, those weights make approximating the distribution of Q_{IV} rather complicated.MethodsAs an alternative, we are investigating a Q statistic, Q_F, 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 Q_{IV} and Q_F, as the basis for tests of heterogeneity.ResultsThe results show that: for LOR and LRR, a two-moment gamma approximation to the distribution of Q_F 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 Q_{IV}, 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 n ge 100 but have some convergence issues for very small sample sizes combined with small probabilities.ConclusionsThe performance of the standard chi ^2 approximation is inadequate for all three binary effect measures. Instead, we recommend a test of heterogeneity based on Q_F and provide practical guidelines for choosing an appropriate test at the .05 level for all three effect measures.

Data Heterogeneity in Meta-Analyses: Statistical Methods, Interpretation, and Guidance.

Data heterogeneity determines whether sufficiently strong conclusions can be derived from synthesizing and aggregating the available literature. Multiple tools are available to calculate data heterogeneity, but each tool has pros and cons. Providing a prediction interval may be the most beneficial since it allows readers to quantify heterogeneity in a clear and clinically relevant form. However, the ultimate decision on which tool to use is left to the discretion of the researcher. This decision should be decided upon during the study inception.

Heterogeneity in meta-analysis: a comprehensive overview.

In recent years, meta-analysis has evolved to a critically important field of Statistics, and has significant applications in Medicine and Health Sciences. In this work we briefly present existing methodologies to conduct meta-analysis along with any discussion and recent developments accompanying them. Undoubtedly, studies brought together in a systematic review will differ in one way or another. This yields a considerable amount of variability, any kind of which may be termed heterogeneity. To this end, reports of meta-analyses commonly present a statistical test of heterogeneity when attempting to establish whether the included studies are indeed similar in terms of the reported output or not. We intend to provide an overview of the topic, discuss the potential sources of heterogeneity commonly met in the literature and provide useful guidelines on how to address this issue and to detect heterogeneity. Moreover, we review the recent developments in the Bayesian approach along with the various graphical tools and statistical software that are currently available to the analyst. In addition, we discuss sensitivity analysis issues and other approaches of understanding the causes of heterogeneity. Finally, we explore heterogeneity in meta-analysis for time to event data in a nutshell, pointing out its unique characteristics.

Corticosteroid in anti-inflammatory treatment of pediatric acute myocarditis: a systematic review and meta-analysis

BackgroundTo evaluate the efficacy of corticosteroids in anti-inflammatory treatment of pediatric acute myocarditis.MethodsWe searched PubMed, Embase and Cochrane library and included studies before October 2022 for clinical trials, observational studies and retrospective studies which reported on children with acute myocarditis treated with corticosteroid anti-inflammatory therapy. The quality of the clinical trials was assessed by Jadad score as an exclusion criterion.ResultsThis systematic review included 6 studies involving 604 pediatric patients with acute myocarditis. Corticosteroid therapy was not associated with reduced risk of mortality due to acute myocarditis (P = 0.53; RR = 0.87; 95% CI = 0.58 to 1.33) compared to anti-failure treatment. There was a significant improvement in pediatric patients’ left ventricular function measured by left ventricular ejection fraction in the group on corticosteroid anti-inflammatory treatment (P = 0.0009; MD = 11.93%; 95% CI = 4.87% to 18.99%). No conclusion can be drawn due to the high heterogeneity in meta-analyses of risk of getting to a clinical endpoint (death or heart transplantation) and changes in left ventricular end-diastolic diameter (LVEDD).ConclusionsCorticosteroid anti-inflammatory therapy in pediatric acute myocarditis patients showed no significant improvement in reducing the risk of mortality, but showed significant improvement in LVEF.

Geographical heterogeneity and dominant polymer types in microplastic contamination of lentic ecosystems: implications for methodological standardization and future research.

This study examines the prevalence and distribution of microplastic polymer types in lentic ecosystems, revealing significant heterogeneity across different geographical regions and ecosystems. The most dominant type of microplastic observed was polyethylene (PE), followed by polypropylene (PP) and polystyrene (PS), which aligns with global production rates. North America, Asia, and Europe were identified as the regions with the highest microplastic contamination, with the United States, China, Italy, and Spain being the most affected countries. The physical characteristics of each ecosystem, such as wind speed, depth, and eutrophication, alongside seasonal variations, and anthropogenic activities, contributed to the observed heterogeneity in microplastics concentrations. The study highlights the need for further research on microplastics in lentic ecosystems, considering their unique physical characteristics and anthropogenic influences. A significant lack of methodological standardization in microplastics research was identified, leading to underestimation of microplastics prevalence and high heterogeneity in meta-analyses.

Disentangling effect size heterogeneity in meta-analysis: A latent mixture approach.

An important task of meta-analysis is to observe, quantify, and explain the heterogeneity across the reported effect sizes of primary studies. A primary issue that challenges this task is the myriad of subtle factors that could have contributed to the observed heterogeneity. We leveraged the recent advances in theoretical machine learning to develop a novel latent mixture-based method for disentangling effect-size heterogeneity in meta-analysis. Mathematical analysis and simulation studies were carried out to demonstrate that, when the observed heterogeneity stems from more than 1 factor, our method can attain a substantially higher statistical power than the traditional methods for moderator analysis without requiring researchers to make judgment calls on which factors to consider or correct for in analyzing the observed heterogeneity. We also conducted a case study with real-world data to show how our method may be used to address long-standing inconsistencies in the literature. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

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.

Heterogeneity estimates in a biased world.

Meta-analyses typically quantify heterogeneity of results, thus providing information about the consistency of the investigated effect across studies. Numerous heterogeneity estimators have been devised. Past evaluations of their performance typically presumed lack of bias in the set of studies being meta-analysed, which is often unrealistic. The present study used computer simulations to evaluate five heterogeneity estimators under a range of research conditions broadly representative of meta-analyses in psychology, with the aim to assess the impact of biases in sets of primary studies on estimates of both mean effect size and heterogeneity in meta-analyses of continuous outcome measures. To this end, six orthogonal design factors were manipulated: Strength of publication bias; 1-tailed vs. 2-tailed publication bias; prevalence of p-hacking; true heterogeneity of the effect studied; true average size of the studied effect; and number of studies per meta-analysis. Our results showed that biases in sets of primary studies caused much greater problems for the estimation of effect size than for the estimation of heterogeneity. For the latter, estimation bias remained small or moderate under most circumstances. Effect size estimations remained virtually unaffected by the choice of heterogeneity estimator. For heterogeneity estimates, however, relevant differences emerged. For unbiased primary studies, the REML estimator and (to a lesser extent) the Paule-Mandel performed well in terms of bias and variance. In biased sets of primary studies however, the Paule-Mandel estimator performed poorly, whereas the DerSimonian-Laird estimator and (to a slightly lesser extent) the REML estimator performed well. The complexity of results notwithstanding, we suggest that the REML estimator remains a good choice for meta-analyses of continuous outcome measures across varied circumstances.

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, . 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, ) rather complicated. As an alternative, we investigate a new Q statistic, , 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 and , as the basis for tests of heterogeneity and for new point and interval estimators of . These include new DerSimonian-Kacker-type moment estimators based on the first moment of , 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 reasonably well, whereas the usual chi-squared approximation for the null distribution of and the Biggerstaff-Jackson approximation to its alternative distribution are poor; in estimating , our moment estimator based on 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 , but otherwise the Q-profile interval performs very well.

Overview of methodologies and statistical strategies in observational studies and meta-analyses on the risk of hepatocellular carcinoma in patients with chronic hepatitis B on entecavir or tenofovir therapy

Entecavir (ETV) and tenofovir disoproxil fumarate (TDF) are first-line antiviral therapies for patients with chronic hepatitis B (CHB) and reduce the risk of disease progression and liver-related complications, as well as improve survival by effectively suppressing viral replication. Nevertheless, since the first publication in 2019 on a lower risk of hepatocellular carcinoma (HCC) in Korean patients receiving TDF than those receiving ETV, the topic has remained a hot and unsettled debate. Multiple studies and meta-analyses have yielded conflicting results. As HCC takes time to develop, studies are mainly observational to benefit from a larger sample size and longer follow-up that provides a higher statistical power to compare the two treatments. However, TDF was available to CHB patients a few years later than ETV in most countries, thus leading to a difference in follow-up duration. Moreover, despite studying the same topic, the difference in data sources and available parameters, inclusion and exclusion criteria, and use of statistical methods complicated the interpretation and comparison of the findings and contributed to between-study heterogeneity in meta-analyses. This review describes some caveats in interpreting and comparing the results from these observational studies and meta-analyses. Future studies should explore better designed observational studies with high-quality data sources, and aggregation of patient data in meta-analysis to tackle between-study heterogeneity.

Why do men have worse COVID-19-related outcomes? A systematic review and meta-analysis with sex adjusted for age

We aimed to study the mechanism behind worse coronavirus disease-19 (COVID-19) outcomes in men and whether the differences between sexes regarding mortality as well as disease severity are influenced by sex hormones. To do so, we used age as a covariate in the meta-regression and subgroup analyses. This was a systematic search and meta-analysis of observational cohorts reporting COVID-19 outcomes. The PubMed (Medline) and Cochrane Library databases were searched. The primary outcome was COVID-19-associated mortality and the secondary outcome was COVID-19 severity. The study was registered at PROSPERO: 42020182924. For mortality, men had a relative risk of 1.36 (95%CI: 1.17 to 1.59; I2 63%, P for heterogeneity <0.01) compared to women. Age was not a significant covariate in meta-analysis heterogeneity (P=0.393) or subgroup analysis. For disease severity, being male was associated with a relative risk of 1.29 (95%CI: 1.19 to 1.40; I2 48%, P for heterogeneity <0.01) compared to the relative risk of women. Again, age did not influence the outcomes of the meta-regression (P=0.914) or subgroup analysis. Men had a higher risk of COVID-19 mortality and severity regardless of age, decreasing the odds of hormonal influences in the described outcomes.

β3-Adrenoceptor Agonist for the Treatment of Bladder Dysfunction in Children: A Systematic Review and Meta-Analysis.

ß3-adrenergic receptor agonists (ß3 agonists) have been used in treatment of overactive bladder (OAB) and neurogenic detrusor overactivity (NDO) in adults. However, their use in children has only recently been approved by the U.S. Food and Drug Administration for patients with NDO. As in adults, the role of ß3 agonists in children may include conditions such as OAB. This systematic review and meta-analysis aims to understand the intended use, efficacy and safety of ß3 agonists in the pediatric population. A literature search was performed in February 2021 across MEDLINE®, Embase®, Scopus®, the Cochrane Library and ClinicalTrials.gov. No language restrictions were placed. All records describing the clinical use of ß3 agonists in pediatric patients (<18 years of age) were included, regardless of the methodological design or outcomes assessed. The identified records were screened by 2 independent authors. The reporting was compliant with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement. Data extraction was performed by 2 independent reviewers, blinded to each other's extractions. The data were pooled using the fixed effects model. Of 367 records identified, 8 studies were included in the review (3 prospective and 5 retrospective). ß3 agonists led to improvements in both urodynamics parameters and self-reported outcomes such as incontinence. Commonly reported side effects were headaches (3%‒5.9%), constipation (3.5%‒5.7%), rhinitis/nasopharyngitis (1.7%‒5.8%) and blurred vision (1.7%‒2.9%). Clinically meaningful changes in safety outcomes (blood pressure, heart rate, electrocardiogram-related changes, liver function) were rare. Before and after ß3 agonist use, pooled effect estimates for maximum cystometric capacity for 171 patients were mean difference of +98.84 ml (95% CI 74.72, 122.96); for complete dryness, assessment of 235 patients showed a Peto odds ratio of 8.68 (95% CI 5.22, 14.45). ß3 agonists appear to be a promising, effective and safe alternative/adjunctive therapy in management of pediatric NDO or OAB, with improvements in both objective urodynamics parameters and subjective patient-reported outcomes following their use.

On ratio measures of heterogeneity for meta-analyses.

As a measure of heterogeneity in meta-analysis, the coefficient of variation (CV) has been recently considered, providing researchers with a complement to the very popular measure. While measures the proportion of total variance that is due to variance of the random effects, the CV is the ratio of the standard deviation of the random effects to the effect of interest. Consequently, the CV provides a different measure of the extent of heterogeneity in a meta-analysis relative to the effect being measured. However, very large CV values can occur when the effect is small making interpretation difficult. The purpose of this article is two-fold. First, we consider variants of the CV that exist in the interval which may be preferable for some researchers. Second, we provide interval estimators for the CV and its variants with excellent coverage properties. We perform simulation studies based on simulated and real data sets and draw comparisons between the methods. For both the CV and its transformations, we recommend confidence intervals using the propagating imprecision method or, as a simpler alternative but at the expense of slightly worse performance in terms of coverage, combining reduced-coverage confidence intervals for the two parameters. These interval estimators typically have better coverage properties for the CV measure than those previously considered.

Investigating heterogeneity in meta-analysis of studies with rare events

In many meta-analyses, the variable of interest is frequently a count outcome reported in an intervention and a control group. Single- or double-zero studies are often observed in this type of data. Given this setting, the well-known Cochran’s Q statistic for testing homogeneity becomes undefined. In this paper, we propose two statistics for testing homogeneity of the risk ratio, particularly for application in the case of rare events in meta-analysis. The first one is a chi-square type statistic. It is constructed based on information of the conditional probability of the number of events in the treatment group given the total number of events. The second one is a likelihood ratio statistic, derived from the logistic regression models allowing fixed and random effects for the risk ratio. Both proposed statistics are well defined even in the situation of single-zero studies. In a simulation study, the proposed tests show a performance better than the traditional test in terms of type I error and power of the test under common and rare event situations. However, as the performance of the two newly proposed tests is still unsatisfactory in the very rare events setting, we suggest a bootstrap approach that does not rely on asymptotic distributional theory and it is shown that the bootstrap approach performs well in terms of type I error. Furthermore, a number of empirical meta-analyses are used to illustrate the methods.