Individual participant data meta-analyses should not ignore clustering

Neoplastic Progression Risk in Females With Barrett's Esophagus: A Systematic Review and Meta-Analysis of Individual Patient Data

BACKGROUND AND AIMS

Females with Barrett's esophagus (BE) have a lower risk of neoplastic progression than males, but sufficiently powered risk analyses are lacking. This systematic review and meta-analysis of individual patient data (IPD) aimed to provide more robust evidence on neoplastic progression risk in females.

METHODS

We conducted a systematic literature search of 3 electronic databases (Medline, Embase, Google Scholar) from inception until August 2023. Eligible studies (1) reported original data on progression from nondysplastic BE, indefinite for dysplasia, or low-grade dysplasia to high-grade dysplasia or esophageal adenocarcinoma; and (2) included female and male patients. IPD were quality controlled by 2 independent reviewers. The primary outcome was the association between sex and neoplastic progression risk, adjusted for risk factors using multivariable Cox regression analysis. Secondary outcomes were sex differences in time to progression and annual progression rate.

RESULTS

IPD were obtained from 11 of 66 eligible studies, including 2196 (31%) females. Neoplastic progression risk was lower in females (hazard ratio for males vs females, 1.44; 95% confidence interval, 1.13-1.82) after adjusting for age, smoking, medication use, hiatal hernia, BE length, and baseline pathology. The annual progression rate was 0.88% in females vs 1.29% in males. Time to progression was similar in both sexes: 3.7 years (interquartile range, 2.1-7.7 years) in females and 4.2 years (interquartile range, 2.0-8.1 years) in males.

CONCLUSION

Although females had a lower neoplastic progression risk, sex differences were smaller than previously reported, and time to progression was similar for both sexes. Future research should focus on other factors than sex to identify low- and high-risk BE patients.

Prevalence of non-communicable diseases among household contacts of people with tuberculosis: A systematic review and individual participant data meta-analysis

OBJECTIVE

To investigate the prevalence of non-communicable diseases among household contacts of people with tuberculosis.

METHODS

We conducted a systematic review and individual participant data meta-analysis. We searched Medline, Embase and the Global Index Medicus from inception to 16 May 2023. We included studies that assessed for at least one non-communicable disease among household contacts of people with clinical tuberculosis. We estimated the non-communicable disease prevalence through mixed effects logistic regression for studies providing individual participant data, and compared it with estimates from aggregated data meta-analyses. Furthermore, we compared age and sex-standardised non-communicable disease prevalence with national-level estimates standardised for age and sex.

RESULTS

We identified 39 eligible studies, of which 14 provided individual participant data (29,194 contacts). Of the remaining 25 studies, 18 studies reported aggregated data suitable for aggregated data meta-analysis. In individual participant data analysis, the pooled prevalence of diabetes in studies that undertook biochemical testing was 8.8% (95% confidence interval [CI], 5.1%-14.9%, four studies). Age-and sex-standardised prevalence was higher in two studies (10.4% vs. 6.9% and 11.5% vs. 8.4%) than the corresponding national estimates and similar in two studies. Prevalence of diabetes mellitus based on self-report or medical records was 3.4% (95% CI 2.6%-4.6%, 14 studies). Prevalence did not significantly differ compared to estimates from aggregated data meta-analysis. There were limited data for other non-communicable diseases.

CONCLUSION

The prevalence of diabetes mellitus among household contacts was high while that of known diabetes was substantially lower, suggesting the underdiagnosis. tuberculosis household contact investigation offers opportunities to deliver multifaceted interventions to identify tuberculosis infection and disease, screen for non-communicable diseases and address shared risk factors.

Efficacy of Biologics in Patients with Allergic Bronchopulmonary Aspergillosis: A Systematic Review and Meta-Analysis

BACKGROUND

Treatment of allergic bronchopulmonary aspergillosis (ABPA) is challenging. Biological therapies have been reported as adjunctive treatments for ABPA, primarily in case series or case reports. This study aimed to analyze the efficacy of biologics for managing ABPA both qualitatively and quantitatively.

METHODS

All articles on APBA published in October 2023 were searched in PubMed, Web of Science, ClinicalTrials.gov, and Embase databases. The effects of interest were the mean changes from baseline for outcomes, including exacerbation rates, oral corticosteroids usage (OCS), and total immunoglobulin E (IgE) levels. Reported outcomes were quantitatively synthesized by usual or individual patient data (IPD) meta-analyses. PROSPERO registration number: CRD42022373396.

RESULTS

A total of 86 studies were included in the systematic review including 346 patients. Sixteen studies on omalizumab were pooled for the usual meta-analysis. Omalizumab therapy significantly reduced exacerbation rates (- 2.29 [95%CI - 3.32, - 1.26]), OCS dosage (- 10.91 mg [95%CI - 18.98, - 2.85]), and total IgE levels (- 273.07 IU/mL [95%CI - 379.30, - 166.84]), meanwhile improving FEV1% predicted (10.09% [95%CI 6.62, 13.55]). Thirty-one studies on dupilumab, mepolizumab, or benralizumab were pooled to perform an IPD meta-analysis, retrospectively. Both dupilumab and mepolizumab significantly reduced exacerbation rates, OCS, and total IgE levels. Benralizumab showed a similar trend, but it was not statistically significant. Tezepelumab showed weak evidence of its effects on ABPA. All five biologics led to milder clinical symptoms (e.g., cough, wheezing) with serious adverse effects that happened once in omalizumab treatment.

CONCLUSION

These results indicate the clinical benefit of omalizumab, dupilumab, and mepolizumab in patients with ABPA. Further randomized, controlled studies with a larger sample size and longer follow-up are needed to confirm these findings.

Reduced drug use as an alternative valid outcome in individuals with stimulant use disorders: Findings from 13 multisite randomized clinical trials

BACKGROUND AND AIMS

Total abstinence has historically been the goal of treatment for substance use disorders; however, there is a growing recognition of the health benefits associated with reduced use as a harm reduction measure in stimulant use disorders treatment. We aimed to assess the validity of reduced stimulant use as an outcome measure in randomized controlled trials (RCTs) of pharmacological interventions for stimulant use disorder.

DESIGN

We conducted a secondary analysis of a pooled dataset of 13 RCTs.

SETTING AND PARTICIPANTS

Participants were individuals seeking treatment for cocaine or methamphetamine use disorders (N = 2062) in a wide range of treatment facilities in the United States.

MEASUREMENTS

We validated reduced stimulant use against a set of clinical indicators drawn from harmonized measurements, including severity of problems caused by drug use, comorbid depression, global severity of substance use and improvement, severity of drug-seeking behavior, craving and high-risk behaviors, all assessed at the end of the trial, as well as follow-up urine toxicology. A series of mixed effect regression models was conducted to validate reduction in frequency of use against no reduction in use and abstinence.

FINDINGS

More participants reduced frequency of primary drug use than achieved abstinence (18.0% vs. 14.2%, respectively). Reduced use was significantly associated with decreases in craving for the primary drug [60.1%, 95% confidence interval (CI) = 54.3%-64.7%], drug seeking behaviors (41.0%, 95% CI = 36.6%-45.7%), depression severity (39.9%, 95% CI = 30.9%-48.3%), as well as multiple measures of global improvement in psychosocial functioning and severity of drug-related problems, albeit less strongly so than abstinence. Moreover, reduced use was associated with sustained clinical benefit at follow-up, as confirmed by negative urine tests (adjusted odds ratio compared with those with no reduction in use: 0.50, 95% CI = 0.35-0.71).

CONCLUSION

Reduced frequency of stimulant use appears to be associated with meaningful improvement in various clinical indicators of recovery. Assessment of reduced use, in addition to abstinence, could broaden the scope of outcomes measured in randomized controlled trials of stimulant use disorders and facilitate the development of more diverse treatment approaches.

Factors associated with tuberculosis treatment initiation among bacteriologically negative individuals evaluated for tuberculosis: an individual patient data meta-analysis

Background

Globally, over one-third of pulmonary tuberculosis (TB) disease diagnoses are made based on clinical criteria after a negative diagnostic test result. Understanding factors associated with clinicians' decisions to initiate treatment for individuals with negative test results is critical for predicting the potential impact of new diagnostics.

Methods

We performed a systematic review and individual patient data meta-analysis using studies conducted between January/2010 and December/2022 (PROSPERO: CRD42022287613). We included trials or cohort studies that enrolled individuals evaluated for TB in routine settings. In these studies participants were evaluated based on clinical examination and routinely-used diagnostics, and were followed for ≥1 week after the initial test result. We used hierarchical Bayesian logistic regression to identify factors associated with treatment initiation following a negative result on an initial bacteriological test (e.g., sputum smear microscopy, Xpert MTB/RIF).

Findings

Multiple factors were positively associated with treatment initiation: male sex [adjusted Odds Ratio (aOR) 1.61 (1.31-1.95)], history of prior TB [aOR 1.36 (1.06-1.73)], reported cough [aOR 4.62 (3.42-6.27)], reported night sweats [aOR 1.50 (1.21-1.90)], and having HIV infection but not on ART [aOR 1.68 (1.23-2.32)]. Treatment initiation was substantially less likely for individuals testing negative with Xpert [aOR 0.77 (0.62-0.96)] compared to smear microscopy and declined in more recent years.

Interpretation

Multiple factors influenced decisions to initiate TB treatment despite negative test results. Clinicians were substantially less likely to treat in the absence of a positive test result when using more sensitive, PCR-based diagnostics.

Statistical primer: individual patient data meta-analysis and meta-analytic approaches in case of non-proportional hazards

Individual patient data (IPD) meta-analyses build upon traditional (aggregate data) meta-analyses by collecting IPD from the individual studies rather than using aggregated summary data. Although both traditional and IPD meta-analyses produce a summary effect estimate, IPD meta-analyses allow for the analysis of data to be performed as a single dataset. This allows for standardization of exposure, outcomes, and analytic methods across individual studies. IPD meta-analyses also allow the utilization of statistical methods typically used in cohort studies, such as multivariable regression, survival analysis, propensity score matching, uniform subgroup and sensitivity analyses, better management of missing data, and incorporation of unpublished data. However, they are more time-intensive, costly, and subject to participation bias. A separate issue relates to the meta-analytic challenges when the proportional hazards assumption is violated. In these instances, alternative methods of reporting time-to-event estimates, such as restricted mean survival time should be used. This statistical primer summarizes key concepts in both scenarios and provides pertinent examples.

Trunk postural control during unstable sitting among individuals with and without low back pain: A systematic review with an individual participant data meta-analysis

INTRODUCTION

Sitting on an unstable surface is a common paradigm to investigate trunk postural control among individuals with low back pain (LBP), by minimizing the influence lower extremities on balance control. Outcomes of many small studies are inconsistent (e.g., some find differences between groups while others do not), potentially due to confounding factors such as age, sex, body mass index [BMI], or clinical presentations. We conducted a systematic review with an individual participant data (IPD) meta-analysis to investigate whether trunk postural control differs between those with and without LBP, and whether the difference between groups is impacted by vision and potential confounding factors.

METHODS

We completed this review according to PRISMA-IPD guidelines. The literature was screened (up to 7th September 2023) from five electronic databases: MEDLINE, CINAHL, Embase, Scopus, and Web of Science Core Collection. Outcome measures were extracted that describe unstable seat movements, specifically centre of pressure or seat angle. Our main analyses included: 1) a two-stage IPD meta-analysis to assess the difference between groups and their interaction with age, sex, BMI, and vision on trunk postural control; 2) and a two-stage IPD meta-regression to determine the effects of LBP clinical features (pain intensity, disability, pain catastrophizing, and fear-avoidance beliefs) on trunk postural control.

RESULTS

Forty studies (1,821 participants) were included for the descriptive analysis and 24 studies (1,050 participants) were included for the IPD analysis. IPD meta-analyses revealed three main findings: (a) trunk postural control was worse (higher root mean square displacement [RMSdispl], range, and long-term diffusion; lower mean power frequency) among individuals with than without LBP; (b) trunk postural control deteriorated more (higher RMSdispl, short- and long-term diffusion) among individuals with than without LBP when vision was removed; and (c) older age and higher BMI had greater adverse impacts on trunk postural control (higher short-term diffusion; longer time and distance coordinates of the critical point) among individuals with than without LBP. IPD meta-regressions indicated no associations between the limited LBP clinical features that could be considered and trunk postural control.

CONCLUSION

Trunk postural control appears to be inferior among individuals with LBP, which was indicated by increased seat movements and some evidence of trunk stiffening. These findings are likely explained by delayed or less accurate corrective responses.

SYSTEMATIC REVIEW REGISTRATION

This review has been registered in PROSPERO (registration number: CRD42021124658).

Clinical Associations and Prognostic Value of MRI-Visible Perivascular Spaces in Patients With Ischemic Stroke or TIA: A Pooled Analysis

BACKGROUND AND OBJECTIVES

Visible perivascular spaces are an MRI marker of cerebral small vessel disease and might predict future stroke. However, results from existing studies vary. We aimed to clarify this through a large collaborative multicenter analysis.

METHODS

We pooled individual patient data from a consortium of prospective cohort studies. Participants had recent ischemic stroke or transient ischemic attack (TIA), underwent baseline MRI, and were followed up for ischemic stroke and symptomatic intracranial hemorrhage (ICH). Perivascular spaces in the basal ganglia (BGPVS) and perivascular spaces in the centrum semiovale (CSOPVS) were rated locally using a validated visual scale. We investigated clinical and radiologic associations cross-sectionally using multinomial logistic regression and prospective associations with ischemic stroke and ICH using Cox regression.

RESULTS

We included 7,778 participants (mean age 70.6 years; 42.7% female) from 16 studies, followed up for a median of 1.44 years. Eighty ICH and 424 ischemic strokes occurred. BGPVS were associated with increasing age, hypertension, previous ischemic stroke, previous ICH, lacunes, cerebral microbleeds, and white matter hyperintensities. CSOPVS showed consistently weaker associations. Prospectively, after adjusting for potential confounders including cerebral microbleeds, increasing BGPVS burden was independently associated with future ischemic stroke (versus 0-10 BGPVS, 11-20 BGPVS: HR 1.19, 95% CI 0.93-1.53; 21+ BGPVS: HR 1.50, 95% CI 1.10-2.06; p = 0.040). Higher BGPVS burden was associated with increased ICH risk in univariable analysis, but not in adjusted analyses. CSOPVS were not significantly associated with either outcome.

DISCUSSION

In patients with ischemic stroke or TIA, increasing BGPVS burden is associated with more severe cerebral small vessel disease and higher ischemic stroke risk. Neither BGPVS nor CSOPVS were independently associated with future ICH.

Are units of analysis properly considered in orthodontic meta-analyses?

BACKGROUND

Assessment of unit of analysis (UoA) in meta-analysis with cluster, split-mouth, repeated measures designs, and multiple intervention groups is a fundamental step in the analysis. The objective of this study was to evaluate the UoAs in orthodontic meta-analysis and determine the error of the analysis.

METHODS

An electronic search was conducted in the Cochrane Library and PubMed to identify orthodontic systematic reviews (SRs) with meta-analyses published in Cochrane and in the highest impact orthodontic journals between 1 January 2013 and 31 December 2022. SRs with meta-analysis assessing at least one of the following UoAs; cluster trials, crossover trials, multiple observations, or multiple intervention groups were included. Screening and data extraction were undertaken by two investigators independently. Descriptive statistics for the study characteristics were provided. The associations between avoiding the unit analysis error (yes, no) and the study characteristics were examined using Fisher's exact test and chi-square test. Logistic regression was undertaken for the significant predictors.

RESULTS

Eighty SRs were deemed eligible for inclusion. Only 30 per cent of the included SRs avoided UoA errors. Compared to the split-mouth design, repeated measures designs had higher odds of avoiding UoA error (odds ratio: 9.6, 95% confidence interval: 2.8-32.3, P < 0.001). In contrast, fewer odds of avoiding the UoA error were found in the cluster design (OR: 0.2, 95% CI: 0.4-1.3, P = 0.09). Though multiple intervention groups have higher odds of avoiding UoA error than split-mouth studies, this was not statistically significant (OR: 2.1, 95% CI: 0.5-8, P = 0.28). None of the SRs characteristics have influenced the appropriate handling of the unit analysis except the type of the journal; the odds of avoiding the UoA error were higher in Cochrane reviews than the non-Cochrane reviews (OR: 3.3, 95% CI: 1.2-8.7, P = 0.02), and the number of authors (P < 0.05).

CONCLUSIONS

UoA errors are common in orthodontic meta-analyses and were only partially avoided in split-mouth design, repeated measures design, and multiple intervention groups.

Two-stage or not two-stage? That is the question for IPD meta-analysis projects

Individual participant data meta-analysis (IPDMA) projects obtain, check, harmonise and synthesise raw data from multiple studies. When undertaking the meta-analysis, researchers must decide between a two-stage or a one-stage approach. In a two-stage approach, the IPD are first analysed separately within each study to obtain aggregate data (e.g., treatment effect estimates and standard errors); then, in the second stage, these aggregate data are combined in a standard meta-analysis model (e.g., common-effect or random-effects). In a one-stage approach, the IPD from all studies are analysed in a single step using an appropriate model that accounts for clustering of participants within studies and, potentially, between-study heterogeneity (e.g., a general or generalised linear mixed model). The best approach to take is debated in the literature, and so here we provide clearer guidance for a broad audience. Both approaches are important tools for IPDMA researchers and neither are a panacea. If most studies in the IPDMA are small (few participants or events), a one-stage approach is recommended due to using a more exact likelihood. However, in other situations, researchers can choose either approach, carefully following best practice. Some previous claims recommending to always use a one-stage approach are misleading, and the two-stage approach will often suffice for most researchers. When differences do arise between the two approaches, often it is caused by researchers using different modelling assumptions or estimation methods, rather than using one or two stages per se.

Responsiveness of Critically Ill Adults With Multimorbidity to Rehabilitation Interventions: A Patient-Level Meta-Analysis Using Individual Pooled Data From Four Randomized Trials

OBJECTIVE

To explore if patient characteristics (pre-existing comorbidity, age, sex, and illness severity) modify the effect of physical rehabilitation (intervention vs control) for the coprimary outcomes health-related quality of life (HRQoL) and objective physical performance using pooled individual patient data from randomized controlled trials (RCTs).

DATA SOURCES

Data of individual patients from four critical care physical rehabilitation RCTs.

STUDY SELECTION

Eligible trials were identified from a published systematic review.

DATA EXTRACTION

Data sharing agreements were executed permitting transfer of anonymized data of individual patients from four trials to form one large, combined dataset. The pooled trial data were analyzed with linear mixed models fitted with fixed effects for treatment group, time, and trial.

DATA SYNTHESIS

Four trials contributed data resulting in a combined total of 810 patients (intervention n = 403, control n = 407). After receiving trial rehabilitation interventions, patients with two or more comorbidities had HRQoL scores that were significantly higher and exceeded the minimal important difference at 3 and 6 months compared with the similarly comorbid control group (based on the Physical Component Summary score (Wald test p = 0.041). Patients with one or no comorbidities who received intervention had no HRQoL outcome differences at 3 and 6 months when compared with similarly comorbid control patients. No patient characteristic modified the physical performance outcome in patients who received physical rehabilitation.

CONCLUSIONS

The identification of a target group with two or more comorbidities who derived benefits from the trial interventions is an important finding and provides direction for future investigations into the effect of rehabilitation. The multimorbid post-ICU population may be a select population for future prospective investigations into the effect of physical rehabilitation.

Nonlinear effects and effect modification at the participant-level in IPD meta-analysis part 1: analysis methods are often substandard

OBJECTIVES

To review analysis methods used for linear effect modification (LEM), nonlinear covariate-outcome associations (NL) and nonlinear effect modification (NLEM) at the participant-level in individual participant data meta-analysis (IPDMA).

STUDY DESIGN AND SETTING

We searched Medline, Embase, Web of Science, Scopus, PsycINFO and the Cochrane Library to identify IPDMA of randomized controlled trials (PROSPERO CRD42019126768). We investigated if and how IPDMA examined LEM, NL and NLEM, including whether aggregation bias was addressed and if power was considered.

RESULTS

We screened 6,466 records, randomly sampled 207 and identified 100 IPDMA of LEM, NL or NLEM. Power for LEM was calculated a priori in 3 IPDMA. Of 100 IPDMA, 94 analyzed LEM, 4 NLEM and 8 NL. One-stage models were favoured for all three (56%, 100%, 50%, respectively). Two-stage models were used in 15%, 0% and 25% of IPDMA with unclear descriptions in 30%, 0% and 25%, respectively. Only 12% of one-stage LEM and NLEM IPDMA provided sufficient detail to confirm they had addressed aggregation bias.

CONCLUSION

Investigation of effect modification at the participant-level is common in IPDMA projects, but methods are often open to bias or lack detailed descriptions. Nonlinearity of continuous covariates and power of IPDMA are rarely assessed.

Reporting quality and risk of bias of Cochrane individual participant data meta-analyses: A cross-sectional study

OBJECTIVES

This study aimed to assess the reporting quality and risk of bias of Cochrane individual participant data meta-analyses (IPD-MAs).

METHODS

We searched the Cochrane Library and identified the Cochrane IPD-MAs. We used the Preferred Reporting Items for Systematic Review and Meta-Analyses of individual participant data (PRISMA-IPD) assessed the reporting quality of included Cochrane IPD- MAs, and the Risk Of Bias In Systematic reviews (ROBIS) tool was used to assess the risk of bias. We performed stratified and correlation analyses to explore factors affecting the quality.

RESULTS

Forty-six Cochrane IPD-MAs were included in our study. Twenty-six Cochrane IPD-MAs (56.5%) had statistical or epidemiological authors involved, and 31 (67.4%) contained only IPD data. Thirty-five studies (76.1%) did not report whether they used 1-stage or 2-stage methods, and forty (87.0%) did not report the statistical techniques used for missing participant data. We found that the entire compliance reported PRISMA-IPD items of Cochrane IPD-MAs published after 2015 (n = 18; Mean ± SD: 26.61 ± 2.75) was higher than those studies published in 2015 and before (n = 28; Mean ± SD: 22.61 ± 4.73), the difference was statistically significant (p = 0.002). A strong positive correlation was found between the fully reported PRISMA-IPD items and fully accordance ROBIS items (Spearman's: ρ = 0.653, p < 0.001).

CONCLUSIONS

The quality of Cochrane IPD-MAs is not high, especially in the reporting of statistical methods. There was room for further improvement in IPD retrieval, IPD integrity and statistical analyses.

Adjusting for misclassification of an exposure in an individual participant data meta-analysis

A common problem in the analysis of multiple data sources, including individual participant data meta-analysis (IPD-MA), is the misclassification of binary variables. Misclassification may lead to biased estimators of model parameters, even when the misclassification is entirely random. We aimed to develop statistical methods that facilitate unbiased estimation of adjusted and unadjusted exposure-outcome associations and between-study heterogeneity in IPD-MA, where the extent and nature of exposure misclassification may vary across studies. We present Bayesian methods that allow misclassification of binary exposure variables to depend on study- and participant-level characteristics. In an example of the differential diagnosis of dengue using two variables, where the gold standard measurement for the exposure variable was unavailable for some studies which only measured a surrogate prone to misclassification, our methods yielded more accurate estimates than analyses naive with regard to misclassification or based on gold standard measurements alone. In a simulation study, the evaluated misclassification model yielded valid estimates of the exposure-outcome association, and was more accurate than analyses restricted to gold standard measurements. Our proposed framework can appropriately account for the presence of binary exposure misclassification in IPD-MA. It requires that some studies supply IPD for the surrogate and gold standard exposure, and allows misclassification to follow a random effects distribution across studies conditional on observed covariates (and outcome). The proposed methods are most beneficial when few large studies that measured the gold standard are available, and when misclassification is frequent.

Protocol for a Systematic Review and Individual Participant Data Meta-Analysis of Randomized Trials of Screening for Atrial Fibrillation to Prevent Stroke

INTRODUCTION

 Atrial fibrillation (AF) is a common cause of stroke. Timely diagnosis of AF and treatment with oral anticoagulation (OAC) can prevent up to two-thirds of AF-related strokes. Ambulatory electrocardiographic (ECG) monitoring can identify undiagnosed AF in at-risk individuals, but the impact of population-based ECG screening on stroke is uncertain, as ongoing and published randomized controlled trials (RCTs) have generally been underpowered for stroke.

METHODS AND ANALYSIS

 The AF-SCREEN Collaboration, with support from AFFECT-EU, have begun a systematic review and individual participant data meta-analysis of RCTs evaluating ECG screening for AF. The primary outcome is stroke. Secondary outcomes include AF detection, OAC prescription, hospitalization, mortality, and bleeding.After developing a common data dictionary, anonymized data will be collated from individual trials into a central database. We will assess risk of bias using the Cochrane Collaboration tool, and overall quality of evidence with the Grading of Recommendations Assessment, Development and Evaluation approach.We will pool data using random effects models. Prespecified subgroup and multilevel meta-regression analyses will explore heterogeneity. We will perform prespecified trial sequential meta-analyses of published trials to determine when the optimal information size has been reached, and account for unpublished trials using the SAMURAI approach.

IMPACT AND DISSEMINATION

 Individual participant data meta-analysis will generate adequate power to assess the risks and benefits of AF screening. Meta-regression will permit exploration of the specific patient, screening methodology, and health system factors that influence outcomes.

TRIAL REGISTRATION NUMBER

 PROSPERO CRD42022310308.

Clinical characterization of non-ketotic hyperglycemia-related seizures: A systematic review and individual participant data meta-analysis

New-onset epileptic seizures and status epilepticus (SE) are the most frequent neurological manifestations of non-ketotic hyperglycemia (NKH), an acute complication of diabetes mellitus (DM). Treatment consists of the correction of the underlying metabolic disorder, whereas anti-seizure medications (ASMs) may even worsen seizures. Evidence on NKH-related seizures is currently restricted to case reports and small case-series. We conducted a systematic review of the PubMed, Embase, and Cochrane Library databases to provide a comprehensive description of NKH-related seizures. Statistical analyses were performed to explore possible associations of glycemic and osmolarity levels with clinical variables. We selected 130 publications and 332 patients (186 males, mean age: 61.1 years). DM was newly-diagnosed in 40%. Mean glycemia and osmolarity levels at presentation were 529.7 mg/dL and 309.6 mmol/mol, respectively; 22.6% showed other neurological symptoms besides seizures. Focal motor seizures were the prominent seizure type (49.4%); non-motor focal seizures (23.2%) most commonly manifested as visual symptoms. Reflex seizures occurred in 10.5%. Brain MRI in 48.7% of cases showed focal T2 subcortical hypodensity and/or overlying cortical T2 hyperintensity with DWI restriction. ASMs were administered in 54.2% of cases, achieving seizure control in just 18.3%. Higher osmolarity levels were associated with newly-diagnosed DM (p = 0.002) and other symptoms at presentation (p < 0.001). Glycemic values were higher in patients with focal aware seizures with motor onset compared to those with focal seizures without motor onset (p = 0.0046) or focal seizures with impaired awareness (p = 0.0306). Lower glycemic values were associated with reflex seizures (p = 0.036) and ASM administration (p < 0.001). NKH-related seizures should be suspected in adults with new-onset clustering focal seizures arising from the motor or posterior cortices, even in the absence of a history of DM. Typical focal changes on brain MRI, while not pathognomonic, can drive the clinical diagnosis. Statistical associations suggest a key role of hyperglycemia in the excitability of higher-energy-demanding cortical areas.

Incorporating historical control information in ANCOVA models using the meta-analytic-predictive approach

The meta-analytic-predictive (MAP) approach is a Bayesian meta-analytic method to synthesize and incorporate information from historical controls in the analysis of a new trial. Classically, only a single parameter, typically the intercept or rate, is assumed to vary across studies, which may not be realistic in more complex models. Analysis of covariance (ANCOVA) is often used to analyze trials with a pretest-posttest design, where both the intercept and the baseline effect (coefficient of the outcome at baseline) affect the estimated treatment effect. We extended the MAP approach to ANCOVA, to allow for variation in the intercept and the baseline effect across studies, and possibly also correlation between these parameters. The method was illustrated using data from the Alzheimer's Disease Cooperative Study (ADCS) and assessed with a simulation study. In the ADCS data, the proposed multivariate MAP approach yielded a prior effective sample size of 79 and 58 for the intercept and the baseline effect respectively and reduced the posterior standard deviation of the treatment effect by 12.6%. The result was robust to the choice of prior for the between-study variation. In the simulations, the proposed approach yielded power gains with a good control of the type I error rate. Ignoring the between-study correlation of the parameters or assuming no variation in the baseline effect generally led to less power gain. In conclusion, the MAP approach can be extended to a multivariate version for ANCOVA, which may improve the estimation of the treatment effect.

Interpretation of meta-analyses

Our article provides guidance on how to interpret a meta-analysis and introduces the reader to the basics of the underlying statistical analysis. The multiple steps of a meta-analysis including systematic literature search, risk of bias assessment, data extraction and data aggregation are addressed. The Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach allows to score the quality of the evidence of the results revealed by a meta-analysis. Trial sequential analysis has been suggested in recent years as a method to assess the power of a meta-analysis and the risks of false positive or false negative conclusions. We also provide information on other more complex meta-analytical approaches including network meta-analysis for the comparison of several treatments as well as recent developments such as individual patient data meta-analysis and living meta-analysis.

What is the prevalence of fear of cancer recurrence in cancer survivors and patients? A systematic review and individual participant data meta-analysis

OBJECTIVE

Care for fear of cancer recurrence (FCR) is considered the most common unmet need among cancer survivors. Yet the prevalence of FCR and predisposing factors remain inconclusive. To support targeted care, we provide a comprehensive overview of the prevalence and severity of FCR among cancer survivors and patients, as measured using the short form of the validated Fear of Cancer Recurrence Inventory (FCRI-SF). We also report on associations between FCR and clinical and demographic characteristics.

METHODS

This is a systematic review and individual participant data (IPD) meta-analysis on the prevalence of FCR. In the review, we included all studies that used the FCRI-SF with adult (≥18 years) cancer survivors and patients. Date of search: 7 February 2020. Risk of bias was assessed using the Joanna Briggs Institute critical appraisal tool.

RESULTS

IPD were requested from 87 unique studies and provided for 46 studies comprising 11,226 participants from 13 countries. 9311 respondents were included for the main analyses. On the FCRI-SF (range 0-36), 58.8% of respondents scored ≥13, 45.1% scored ≥16 and 19.2% scored ≥22. FCR decreased with age and women reported more FCR than men. FCR was found across cancer types and continents and for all time periods since cancer diagnosis.

CONCLUSIONS

FCR affects a considerable number of cancer survivors and patients. It is therefore important that healthcare providers discuss this issue with their patients and provide treatment when needed. Further research is needed to investigate how best to prevent and treat FCR and to identify other factors associated with FCR. The protocol was prospectively registered (PROSPERO CRD42020142185).

Postural control of the trunk in individuals with and without low back pain during unstable sitting: A protocol for a systematic review with an individual participant data meta-analysis

Introduction

Postural control of the trunk is critical for performance of everyday activities and the health of spinal tissues. Although some studies report that individuals with low back pain (LBP) have poorer/compromised postural control than pain-free individuals when sitting on an unstable surface, others do not. Analyses commonly lack the statistical power to evaluate the relevance of features that could impact the performance of postural control, such as sex, age, anthropometrics, pain intensity or disability. This paper outlines a protocol for a systematic review with an individual participant data (IPD) meta-analysis that aims to synthesise the evidence and evaluate differences of postural control measures between individuals with and without LBP during unstable sitting.

Methods and analysis

A systematic review with IPD meta-analysis will be conducted according to PRISMA-IPD guidelines. To identify relevant studies, electronic databases and the reference lists of included articles will be screened. Unstable seat movements are derived from centre of pressure (CoP) data using a force plate or angle of the seat using motion systems/sensors. The comprehensiveness of reporting and methodological quality of included studies will be assessed. Analysis will involve a descriptive analysis to synthesise the findings of all included studies and a quantitative synthesis using two-stage IPD meta-analysis of studies that include both individuals with and without LBP for which IPD set can be obtained from authors. Analyses will include consideration of confounding variables.

Ethics

Exemption from ethical approval was obtained for this review (University of Queensland, ID: 2019003026).

Systematic review registration

PROSPERO ID: CRD42021124658.

Surgical outcomes in paediatric lateral condyle non-union: A systematic review and meta-analysis

AIMS

Non-union is a known and much-dreaded complication of paediatric lateral condyle fractures. This systematic review aims to pool together individual studies to find out if the timing of fixation and method of fixation impacts surgical outcomes (postoperative union and elbow ROM) in paediatric lateral condyle non-union.

METHODS

A systematic review and individual patient data meta-analysis was conducted according to PRISMA guidelines. All surgical studies with original data on pediatric lateral humeral condyle non-union were included. Patients who did not undergo surgical fixation were excluded.

RESULTS

A total of 12 studies with 177 patients were included. In total, 159 patients (89.8%) achieved bony union postoperatively while 18 patients (10.2%) did not. Mixed-effects logistic regression showed that percutaneous fixation (p-value=0.020) was associated with lower rates of postoperative union compared to open fixation, whereas the age at surgery did not have a significant impact (p-value=0.401). For elbow ROM, mixed-effects linear regression showed that increased age at surgery (p-value=0.007) and reduction of the fracture fragment (vs. in situ fixation) (p-value=0.041) were associated with reduced postoperative ROM whereas female sex (p-value=0.009) and corrective osteotomy (p-value=0.045) were associated with increased postoperative ROM.

CONCLUSION

While the timing of surgical fixation did not significantly impact postoperative bony union, undergoing fixation at an older age was associated with reduced postoperative elbow ROM. In addition, percutaneous fixation may be associated with poorer postoperative union compared to open fixation while anatomical reduction may be associated with reduced postoperative elbow ROM compared to in situ fixation.

LEVEL OF EVIDENCE

IV.

Systematic Review Reveals Lack of Causal Methodology Applied to Pooled Longitudinal Observational Infectious Disease Studies

OBJECTIVES

Among ID studies seeking to make causal inferences and pooling individual-level longitudinal data from multiple infectious disease cohorts, we sought to assess what methods are being used, how those methods are being reported, and whether these factors have changed over time.

STUDY DESIGN AND SETTING

Systematic review of longitudinal observational infectious disease studies pooling individual-level patient data from 2+ studies published in English in 2009. 2014, or 2019. This systematic review protocol is registered with PROSPERO (CRD42020204104).

RESULTS

Our search yielded 1,462 unique articles. Of these, 16 were included in the final review. Our analysis showed a lack of causal inference methods and of clear reporting on methods and the required assumptions.

CONCLUSION

There are many approaches to causal inference which may help facilitate accurate inference in the presence of unmeasured and time-varying confounding. In observational ID studies leveraging pooled, longitudinal IPD, the absence of these causal inference methods and gaps in the reporting of key methodological considerations suggests there is ample opportunity to enhance the rigor and reporting of research in this field. Interdisciplinary collaborations between substantive and methodological experts would strengthen future work.

Conservative management of occult pneumothorax in mechanically ventilated patients: A systematic review and meta-analysis

BACKGROUND

The aim of this systematic review was to investigate the safety and effectiveness of conservative management versus prophylactic intercostal catheter (ICC) insertion for the management of occult pneumothoraces in mechanically ventilated patients.

METHODS

PubMed, Embase, CINAHL, Web of Science, Cochrane Central, and other trial registries were searched. Eligible studies were critically appraised using standardized instruments. Meta-analysis was performed with mixed-methods logistic regression where appropriate and sensitivity analyses were performed with alternative statistical methods (Stata™ 15 or RevMan 5.3) or summarized in narrative. Randomized controlled trials (RCTs) and cohort studies were analyzed separately.

RESULTS

Twelve studies with a total of 354 participants were included; three RCTs (178 participants) and nine cohort studies (176 participants). The majority of the included studies, particularly the cohort studies, were well conducted. Two of the RCTs were rated as low quality. Statistically significant differences were observed in the RCT analysis: ICC insertion (any reason) (odds ratio, 2.86; 95% confidence interval, 1.26-6.43, 2 RCTs) in favor of prophylactic ICC; ICC complications (odds ratio, 0.12; 95% confidence interval, 0.02-0.62, 2 RCTs) in favor of conservative management. Nonstatistically significant differences were observed for progression of pneumothorax, ICC insertion (progression to simple pneumothorax), and ICC insertion (nonpneumothorax reasons). Results of analyses showed high imprecision (wide confidence limits). Conservative management showed a low rate of tension pneumothorax (2.8%). Complications were higher in the ICC group (19.5% vs. 5.8%).

CONCLUSION

Available evidence suggests that conservative management is safe for the management of occult pneumothoraces in mechanically ventilated patients, especially when undergoing short-term (<4 days) ventilation. We recommend that patients undergoing mechanical ventilation for a procedure alone and patients suspected to be ventilated less than 4 days can be conservatively managed.

LEVEL OF EVIDENCE

Systematic review and meta-analysis, level III.

Individual Patient Data Meta-Analysis and Network Meta-Analysis

Meta-analyses are often conducted using trial-level summary data. However, when individual patient data (IPD ) is available, there is greater flexibility in the analysis and a wider range of statistical models that can be fitted. There are two approaches to fitting IPD models. The traditional two-stage approach involves analyzing each trial individually in the first stage and then combining trial estimates of treatment effectiveness in the second stage using methods developed for aggregate data meta-analysis. Growing in popularity is the one-stage approach in which trials are analyzed and synthesized within one statistical model whilst the clustering of patients within trials is accounted for. This chapter outlines both fixed effect and random effects one- and two-stage meta-analysis models for continuous, binary, and time-to-event outcomes. The meta-analysis framework is then extended to the scenario where there are more than two treatments and network meta-analysis models are described.The availability of IPD provides greater statistical power for investigating interactions between treatments and covariates. Treatment-covariate interactions contain both within- and across-trial information where the across-trial information may be subject to ecological bias. This chapter presents network meta-analysis models separating out the within- and across-trial information and finishes by considering practical solutions for dealing with missing covariate data, assessing the consistency assumption, combining IPD and aggregate data and specific considerations for time-to-event outcomes.

Behavior Problems Following Childhood TBI: The Role of Sex, Age, and Time Since Injury

OBJECTIVE

To examine the frequency of behavioral problems after childhood traumatic brain injury (TBI) and their associations with injury severity, sex, and social environmental factors.

SETTING

Children's hospitals in the Midwestern/Western United States.

PARTICIPANTS

381 boys and 210 girls with moderate (n = 359) and severe (n = 227) TBI, with an average age at injury of 11.7 years (range 0.3-18) who were injured ≤3 years ago.

DESIGN

Secondary data analysis of a multistudy cohort.

MAIN MEASURES

Child Behavior Checklist (CBCL) administered pretreatment.

RESULTS

Thirty-seven percent had borderline/clinical elevations on the CBCL Total Problem Scale, with comparable rates of Internalizing and Externalizing problems (33% and 31%, respectively). Less parental education was associated with higher rates of internalizing, externalizing, and total problems. Time since injury had a linear association with internalizing symptoms, with greater symptoms at longer postinjury intervals. Younger boys had significantly higher levels of oppositional defiant symptoms than girls, whereas older girls had significantly greater attention-deficit hyperactivity disorder symptoms than boys.

CONCLUSIONS

Pediatric TBI is associated with high rates of behavior problems, with lower socioeconomic status predicting substantially elevated risk. Associations of higher levels of internalizing symptoms with greater time since injury highlight the importance of tracking children over time.

Benefits and harms of Risperidone and Paliperidone for treatment of patients with schizophrenia or bipolar disorder: a meta-analysis involving individual participant data and clinical study reports

BACKGROUND

Schizophrenia and bipolar disorder are severe mental illnesses which are highly prevalent worldwide. Risperidone and Paliperidone are treatments for either illnesses, but their efficacy compared to other antipsychotics and growing reports of hormonal imbalances continue to raise concerns. As existing evidence on both antipsychotics are solely based on aggregate data, we aimed to assess the benefits and harms of Risperidone and Paliperidone in the treatment of patients with schizophrenia or bipolar disorder, using individual participant data (IPD), clinical study reports (CSRs) and publicly available sources (journal publications and trial registries).

METHODS

We searched MEDLINE, Central, EMBASE and PsycINFO until December 2020 for randomised placebo-controlled trials of Risperidone, Paliperidone or Paliperidone palmitate in patients with schizophrenia or bipolar disorder. We obtained IPD and CSRs from the Yale University Open Data Access project. The primary outcome Positive and Negative Syndrome Scale (PANSS) score was analysed using one-stage IPD meta-analysis. Random-effect meta-analysis of harm outcomes involved methods for coping with rare events. Effect-sizes were compared across all available data sources using the ratio of means or relative risk. We registered our review on PROSPERO, CRD42019140556.

RESULTS

Of the 35 studies, IPD meta-analysis involving 22 (63%) studies showed a significant clinical reduction in the PANSS in patients receiving Risperidone (mean difference - 5.83, 95% CI - 10.79 to - 0.87, I2 = 8.5%, n = 4 studies, 1131 participants), Paliperidone (- 6.01, 95% CI - 8.7 to - 3.32, I2 = 4.3%, n = 13, 3821) and Paliperidone palmitate (- 7.89, 95% CI - 12.1 to - 3.69, I2 = 2.9%, n = 5, 2209). CSRs reported nearly two times more adverse events (4434 vs. 2296 publication, relative difference (RD) = 1.93, 95% CI 1.86 to 2.00) and almost 8 times more serious adverse events (650 vs. 82; RD = 7.93, 95% CI 6.32 to 9.95) than the journal publications. Meta-analyses of individual harms from CSRs revealed a significant increased risk among several outcomes including extrapyramidal disorder, tardive dyskinesia and increased weight. But the ratio of relative risk between the different data sources was not significant. Three treatment-related gynecomastia events occurred, and these were considered mild to moderate in severity.

CONCLUSION

IPD meta-analysis conclude that Risperidone and Paliperidone antipsychotics had a small beneficial effect on reducing PANSS score over 9 weeks, which is more conservative than estimates from reviews based on journal publications. CSRs also contained significantly more data on harms that were unavailable in journal publications or trial registries. Sharing of IPD and CSRs are necessary when performing meta-analysis on the efficacy and safety of antipsychotics.

Correlation between lead exposure and cognitive function in 12-year-old children: a systematic review and meta-analysis

Lead exposure as a toxic material especially in children can be recognized as a harmful factor for cognitive function system. This meta-analysis was conducted to estimate the lead exposure effect on cognitive function among 6979 children less than 12 years. The 16 studies were divided into two clusters according to "duration of exposure" and "dose" using k-means partitioning clustering algorithm. Then, subgroup analysis has been performed based on the clustered studies. According to the results of the k-means clustering, dose and duration of exposure were significant factors between all considered variables. A stronger negative significant pooled correlation was observed in higher dose and duration cluster in comparison with another cluster including lower dose and duration (r = - 0.29, P-value < 0.001 vs. r = - 0.08, P-value < 0.001). Overall, a negative significant correlation was observed between lead exposure and cognitive function test score (r = - 0.22, P-value < 0.001). This study confirms the negative effect of lead on cognitive function in children, but needed further investigations to achieve the safe dose and duration of the lead exposure.

Exploring consequences of simulation design for apparent performance of methods of meta-analysis

Contemporary statistical publications rely on simulation to evaluate performance of new methods and compare them with established methods. In the context of random-effects meta-analysis of log-odds-ratios, we investigate how choices in generating data affect such conclusions. The choices we study include the overall log-odds-ratio, the distribution of probabilities in the control arm, and the distribution of study-level sample sizes. We retain the customary normal distribution of study-level effects. To examine the impact of the components of simulations, we assess the performance of the best available inverse-variance-weighted two-stage method, a two-stage method with constant sample-size-based weights, and two generalized linear mixed models. The results show no important differences between fixed and random sample sizes. In contrast, we found differences among data-generation models in estimation of heterogeneity variance and overall log-odds-ratio. This sensitivity to design poses challenges for use of simulation in choosing methods of meta-analysis.

The methodological quality of individual participant data meta-analysis on intervention effects: systematic review

OBJECTIVE

To assess the methodological quality of individual participant data (IPD) meta-analysis and to identify areas for improvement.

DESIGN

Systematic review.

DATA SOURCES

Medline, Embase, and Cochrane Database of Systematic Reviews.

ELIGIBILITY CRITERIA FOR SELECTING STUDIES

Systematic reviews with IPD meta-analyses of randomised controlled trials on intervention effects published in English.

RESULTS

323 IPD meta-analyses covering 21 clinical areas and published between 1991 and 2019 were included: 270 (84%) were non-Cochrane reviews and 269 (84%) were published in journals with a high impact factor (top quarter). The IPD meta-analyses showed low compliance in using a satisfactory technique to assess the risk of bias of the included randomised controlled trials (43%, 95% confidence interval 38% to 48%), accounting for risk of bias when interpreting results (40%, 34% to 45%), providing a list of excluded studies with justifications (32%, 27% to 37%), establishing an a priori protocol (31%, 26% to 36%), prespecifying methods for assessing both the overall effects (44%, 39% to 50%) and the participant-intervention interactions (31%, 26% to 36%), assessing and considering the potential of publication bias (31%, 26% to 36%), and conducting a comprehensive literature search (19%, 15% to 23%). Up to 126 (39%) IPD meta-analyses failed to obtain IPD from 90% or more of eligible participants or trials, among which only 60 (48%) provided reasons and 21 (17%) undertook certain strategies to account for the unavailable IPD.

CONCLUSIONS

The methodological quality of IPD meta-analyses is unsatisfactory. Future IPD meta-analyses need to establish an a priori protocol with prespecified data syntheses plan, comprehensively search the literature, critically appraise included randomised controlled trials with appropriate technique, account for risk of bias during data analyses and interpretation, and account for unavailable IPD.

Hypofractionated radiotherapy in locally advanced bladder cancer: an individual patient data meta-analysis of the BC2001 and BCON trials

BACKGROUND

Two radiotherapy fractionation schedules are used to treat locally advanced bladder cancer: 64 Gy in 32 fractions over 6·5 weeks and a hypofractionated schedule of 55 Gy in 20 fractions over 4 weeks. Long-term outcomes of these schedules in several cohort studies and case series suggest that response, survival, and toxicity are similar, but no direct comparison has been published. The present study aimed to assess the non-inferiority of 55 Gy in 20 fractions to 64 Gy in 32 fractions in terms of invasive locoregional control and late toxicity in patients with locally advanced bladder cancer.

METHODS

We did a meta-analysis of individual patient data from patients (age ≥18 years) with locally advanced bladder cancer (T1G3 [high-grade non-muscle invasive] or T2-T4, N0M0) enrolled in two multicentre, randomised, controlled, phase 3 trials done in the UK: BC2001 (NCT00024349; assessing addition of chemotherapy to radiotherapy) and BCON (NCT00033436; assessing hypoxia-modifying therapy combined with radiotherapy). In each trial, the fractionation schedule was chosen according to local standard practice. Co-primary endpoints were invasive locoregional control (non-inferiority margin hazard ratio [HR]=1·25); and late bladder or rectum toxicity, assessed with the Late Effects Normal Tissue Task Force-Subjective, Objective, Management, Analytic tool (non-inferiority margin for absolute risk difference [RD]=10%). If non-inferiority was met for invasive locoregional control, superiority could be considered if the 95% CI for the treatment effect excluded the null effect (HR=1). One-stage individual patient data meta-analysis models for the time-to-event and binary outcomes were used, accounting for trial differences, within-centre correlation, randomised treatment received, baseline variable imbalances, and potential confounding from relevant prognostic factors.

FINDINGS

782 patients with known fractionation schedules (456 from the BC2001 trial and 326 from the BCON trial; 376 (48%) received 64 Gy in 32 fractions and 406 (52%) received 55 Gy in 20 fractions) were included in our meta-analysis. Median follow-up was 120 months (IQR 99-159). Patients who received 55 Gy in 20 fractions had a lower risk of invasive locoregional recurrence than those who received 64 Gy in 32 fractions (adjusted HR 0·71 [95% CI 0·52-0·96]). Both schedules had similar toxicity profiles (adjusted RD -3·37% [95% CI -11·85 to 5·10]).

INTERPRETATION

A hypofractionated schedule of 55 Gy in 20 fractions is non-inferior to 64 Gy in 32 fractions with regard to both invasive locoregional control and toxicity, and is superior with regard to invasive locoregional control. 55 Gy in 20 fractions should be adopted as a standard of care for bladder preservation in patients with locally advanced bladder cancer.

FUNDING

Cancer Research UK.

The Epidemiology, Cause, and Prognosis of Painful Tic Convulsif Syndrome: An Individual Patient Data Analysis of 192 Cases

BACKGROUND

Characterized by the coexistence of trigeminal neuralgia and ipsilateral hemifacial spasm (HFS), painful tic convulsif (PTC) is a rare entity that has not yet been systematically studied.

OBJECTIVE

To systematically explore the epidemiology, cause, prognosis, and prognosis predictors of PTC.

METHODS

We searched PubMed, Web of Science, and the Cochrane Library for relevant studies published between establishment of the library and July 1, 2020. Information on demographics, causes, specific interventions, and intervention outcomes was extracted. We first performed descriptive analysis of demographics, causes, and surgical outcomes of PTC. Univariate and multivariate regression methods were used to explore potential prognosis predictors. Further, a 2-step meta-analysis method was used to validate the identified factors.

RESULTS

Overall, 57 reports including 192 cases with PTC were included in the analysis. The median age of patients with PTC is 54 years (range, 44-62 years), with more patients being female (P < 0.001), initiated as HFS (P = 0.005), and being affected with left side (P = 0.045). The vertebrobasilar artery contributes to >65% of the causes of single vascular compression for PTC. Anterior inferior cerebellar artery/posterior inferior cerebellar artery involvement (odds ratio, 4.050; 95% confidence interval, 1.091-15.031) and older age (P = 0.008) predict freedom from symptoms and recurrence after microvascular decompression, respectively.

CONCLUSIONS

PTC occurs more in middle-aged women between 40 and 60 years old, initiates as HFS, and affects the left side. Vertebrobasilar artery compression is the most common single cause of PTC. Microvascular decompression effectively treated PTC, with a cure rate >80%. Anterior inferior cerebellar artery/posterior inferior cerebellar artery involvement predicts successful surgery and older age predicts recurrence.

Comparison of surgical and endovascular treatments for fusiform intracranial aneurysms: systematic review and individual patient data meta-analysis

Fusiform intracranial aneurysms (FIA) are associated with significant morbidity and mortality. We carried out a systematic review and meta-analysis of individual participant data with propensity score adjustment to compare the functional and angiographic outcomes between surgical and endovascular approaches to FIA. We conducted a systematic review for articles on the treatment of FIA with individual patient-level detailing. Data from patients treated for FIA in our institution from 2010 to 2018 were also collected. The primary studied outcome was morbidity, and secondary outcomes were angiographic results and retreatment. Propensity score-adjusted mixed-effects logistic regression models evaluated treatment options, stratified by anatomical location. Compiling original and published data, there were 312 cases, of which 79 (25.3%) had open surgery, and 233 (74.5%) were treated with endovascular procedures. There were no differences between treatment groups, for neither cavernous ICA (OR 1.04, 95% CI 0.05-23.6) nor supraclinoid aneurysms (OR 7.82, 95% CI 0.65-94.4). Both size (OR 1.11, 95% CI 1.03-1.19) and initial mRS (OR 2.0, 95% CI 1.2-3.3) were risk factors for morbidity, independent of location. Neither age nor rupture status influenced the odds of posterior morbidity. Unfavorable angiographic outcomes were more common in the endovascular group for supraclinoid and vertebrobasilar aneurysms (χ², P < 0.01). There were no differences between morbidity of surgical and endovascular treatments for FIA, regardless of aneurysm location. Size and initial mRS were correlated with functional outcomes, whereas age and rupture status were not. Microsurgery seems to yield better long-term angiographic results compared to endovascular procedures.

An individual patient analysis of the efficacy of using GPi-DBS to treat Huntington's disease

OBJECTIVE

The efficacy of globus pallidus internus-deep brain stimulation (GPi-DBS) for the treatment of Huntington's disease (HD) has not been validated in large-scale studies. We conducted an individual patient analysis to pool outcomes of all of the published HD-GPi-DBS studies.

METHODS

PubMed, Embase and the Cochrane Library were searched for relevant articles. The Unified Huntington's Disease Rating Scale (UHDRS)-motor and UHDRS-chorea improvements were analyzed during different follow-up periods. Secondary outcomes, including UHDRS-motor subitem scores and functional assessment results, were also analyzed. Correlation and regression analyses were conducted to find improvement predictors. This study was registered in PROSPERO (CRD42018105995).

RESULTS

Eighteen studies including 39 patients with 124 visits were analyzed. GPi-DBS significantly improved the UHDRS-motor score in <3 months (p = 0.001), 3-9 months (p < 0.001), and 9-12 months (p < 0.001), but did not continue in later follow-ups. UHDRS-chorea was significantly improved even in the >30-month follow-up (p = 0.003). Functional assessment was not improved 12 months postoperatively (p = 0.196). The Westphal variant of HD (W-HD) gained no motor benefits 6 months postoperatively (p = 0.178). The Westphal variant was the only risk factor for DBS efficacy (p = 0.044). The rate of stimulation-related adverse events was 87.2%.

CONCLUSIONS

GPi-DBS has a stable effect on chorea symptoms in HD patients. Chorea-dominant patients may be the best candidates for surgery, while attention should be paid to postoperative stimulation-related complications. Given that GPi-DBS has limited effects on other motor symptoms, W-HD patients are not surgical candidates.

Impact of PRECISE-DAPT and DAPT Scores on Dual Antiplatelet Therapy Duration After 2nd Generation Drug-Eluting Stent Implantation

PURPOSE

Determining the optimal duration of dual antiplatelet therapy (DAPT) after drug-eluting stent (DES) implantation is an important clinical issue. We evaluated the effects of ischemia (by DAPT score) and bleeding (by PRECISE-DAPT score), as well as the impact of DAPT duration, on clinical outcomes.

METHODS

From pooled analysis of four randomized clinical trials, 5131 patients undergoing second-generation DES implantation were randomized to short-duration (n = 2575; ≤ 6 months) or standard-duration (n = 2556; ≥ 12 months) DAPT groups. This population was further divided into four subgroups according to PRECISE-DAPT (high bleeding risk ≥ 25) and DAPT (high ischemic risk ≥ 2) scores.

RESULTS

Net clinical outcomes (1.3% vs. 1.3%; p = 0.89) and ischemic events (5.0% vs. 4.5%; p = 0.44) did not differ between the two duration groups, although bleeding events were more frequent in patients with standard-duration DAPT (0.4% vs. 0.9%; p = 0.04). Standard-duration DAPT was associated with fewer ischemic events (6.9% vs. 4.0%; p = 0.02) and no increase in bleeding events only among patients at low bleeding risk and high ischemic risk. The other groups show no differences in net clinical outcomes, ischemic events, or bleeding events according to DAPT duration.

CONCLUSION

Compared with short-duration DAPT, standard-duration DAPT was associated with similar net clinical outcomes and ischemic events, but more bleeding events at 12 months after second-generation DES implantation. However, standard-duration DAPT reduced ischemic events without increasing bleeding events among patients at low bleeding and high ischemic risk. When determining DAPT duration, considering both ischemic and bleeding risk can help optimize patient benefits.

CLINICAL TRIAL REGISTRATION

EXCELLENT (NCT00698607), RESET (NCT01145079), IVUS-XPL (NCT01308281), OPTIMA-C (NCT03056118).

Metformin in Pregnancy Study (MiPS): protocol for a systematic review with individual patient data meta-analysis

INTRODUCTION

Gestational diabetes mellitus (GDM) is a common disorder of pregnancy and contributes to adverse pregnancy outcomes. Metformin is often used for the prevention and management of GDM; however, its use in pregnancy continues to be debated. The Metformin in Pregnancy Study aims to use individual patient data (IPD) meta-analysis to clarify the efficacy and safety of metformin use in pregnancy and to identify relevant knowledge gaps.

METHODS AND ANALYSIS

MEDLINE, EMBASE and all Evidence-Based Medicine will be systematically searched for randomised controlled trials (RCT) testing the efficacy of metformin compared with placebo, usual care or other interventions in pregnant women. Two independent reviewers will assess eligibility using prespecified criteria and will conduct data extraction and quality appraisal of eligible studies. Authors of included trials will be contacted and asked to contribute IPD. Primary outcomes include maternal glycaemic parameters and GDM, as well as neonatal hypoglycaemia, anthropometry and gestational age at delivery. Other adverse maternal, birth and neonatal outcomes will be assessed as secondary outcomes. IPD from these RCTs will be harmonised and a two-step meta-analytic approach will be used to determine the efficacy and safety of metformin in pregnancy, with a priori adjustment for covariates and subgroups to examine effect moderators of treatment outcomes. Sensitivity analyses will assess heterogeneity, risk of bias and the impact of trials which have not provided IPD.

ETHICS AND DISSEMINATION

All IPD will be deidentified and studies contributing IPD will have ethical approval from their respective local ethics committees. This study will provide robust evidence regarding the efficacy and safety of metformin use in pregnancy, and may identify subgroups of patients who may benefit most from this treatment modality. Findings will be published in peer-reviewed journals and disseminated at scientific meetings, providing much needed evidence to inform clinical and public health actions in this area.

One-stage individual participant data meta-analysis models for continuous and binary outcomes: Comparison of treatment coding options and estimation methods

A one-stage individual participant data (IPD) meta-analysis synthesizes IPD from multiple studies using a general or generalized linear mixed model. This produces summary results (eg, about treatment effect) in a single step, whilst accounting for clustering of participants within studies (via a stratified study intercept, or random study intercepts) and between-study heterogeneity (via random treatment effects). We use simulation to evaluate the performance of restricted maximum likelihood (REML) and maximum likelihood (ML) estimation of one-stage IPD meta-analysis models for synthesizing randomized trials with continuous or binary outcomes. Three key findings are identified. First, for ML or REML estimation of stratified intercept or random intercepts models, a t-distribution based approach generally improves coverage of confidence intervals for the summary treatment effect, compared with a z-based approach. Second, when using ML estimation of a one-stage model with a stratified intercept, the treatment variable should be coded using "study-specific centering" (ie, 1/0 minus the study-specific proportion of participants in the treatment group), as this reduces the bias in the between-study variance estimate (compared with 1/0 and other coding options). Third, REML estimation reduces downward bias in between-study variance estimates compared with ML estimation, and does not depend on the treatment variable coding; for binary outcomes, this requires REML estimation of the pseudo-likelihood, although this may not be stable in some situations (eg, when data are sparse). Two applied examples are used to illustrate the findings.

Response to physical rehabilitation and recovery trajectories following critical illness: individual participant data meta-analysis protocol

INTRODUCTION

The number of inconclusive physical rehabilitation randomised controlled trials for patients with critical illness is increasing. Evidence suggests critical illness patient subgroups may exist that benefit from targeted physical rehabilitation interventions that could improve their recovery trajectory. We aim to identify critical illness patient subgroups that respond to physical rehabilitation and map recovery trajectories according to physical function and quality of life outcomes. Additionally, the utilisation of healthcare resources will be examined for subgroups identified.

METHODS AND ANALYSIS

This is an individual participant data meta-analysis protocol. A systematic literature review was conducted for randomised controlled trials that delivered additional physical rehabilitation for patients with critical illness during their acute hospital stay, assessed chronic disease burden, with a minimum follow-up period of 3 months measuring performance-based physical function and health-related quality of life outcomes. From 2178 records retrieved in the systematic literature review, four eligible trials were identified by two independent reviewers. Principal investigators of eligible trials were invited to contribute their data to this individual participant data meta-analysis. Risk of bias will be assessed (Cochrane risk of bias tool for randomised trials). Participant and trial characteristics, interventions and outcomes data of included studies will be summarised. Meta-analyses will entail a one-stage model, which will account for the heterogeneity across and the clustering between studies. Multiple imputation using chained equations will be used to account for the missing data.

ETHICS AND DISSEMINATION

This individual participant data meta-analysis does not require ethical review as anonymised participant data will be used and no new data collected. Additionally, eligible trials were granted approval by institutional review boards or research ethics committees and informed consent was provided for participants. Data sharing agreements are in place permitting contribution of data. The study findings will be disseminated at conferences and through peer-reviewed publications.

PROSPERO REGISTRATION NUMBER

CRD42019152526.

Real-world Independent Testing of e-ASPECTS Software (RITeS): statistical analysis plan

Background: Artificial intelligence-based software may automatically detect ischaemic stroke lesions and provide an Alberta Stroke Program Early CT score (ASPECTS) on CT, and identify arterial occlusion and provide a collateral score on CTA. Large-scale independent testing will inform clinical use, but is lacking. We aim to test e-ASPECTS and e-CTA (Brainomix, Oxford UK) using CT scans obtained from a range of clinical studies. Methods: Using prospectively collected baseline CT and CTA scans from 10 national/international clinical stroke trials or registries (total >6600 patients), we will select a large clinically representative sample for testing e-ASPECTS and e-CTA compared to previously acquired independent expert human interpretation (reference standard). Our primary aims are to test agreement between software-derived and masked human expert ASPECTS, and the diagnostic accuracy of e-ASPECTS for identifying all causes of stroke symptoms using follow-up imaging and final clinical opinion as diagnostic ground truth. Our secondary aims are to test when and why e-ASPECTS is more or less accurate, or succeeds/fails to produce results, agreement between e-CTA and human expert CTA interpretation, and repeatability of e-ASPECTS/e-CTA results. All testing will be conducted on an intention-to-analyse basis. We will assess agreement between software and expert-human ratings and test the diagnostic accuracy of software.  Conclusions: RITeS will provide comprehensive, robust and representative testing of e-ASPECTS and e-CTA against the current gold-standard, expert-human interpretation.

Individual participant data meta-analysis of intervention studies with time-to-event outcomes: A review of the methodology and an applied example

Many randomized trials evaluate an intervention effect on time-to-event outcomes. Individual participant data (IPD) from such trials can be obtained and combined in a so-called IPD meta-analysis (IPD-MA), to summarize the overall intervention effect. We performed a narrative literature review to provide an overview of methods for conducting an IPD-MA of randomized intervention studies with a time-to-event outcome. We focused on identifying good methodological practice for modeling frailty of trial participants across trials, modeling heterogeneity of intervention effects, choosing appropriate association measures, dealing with (trial differences in) censoring and follow-up times, and addressing time-varying intervention effects and effect modification (interactions).We discuss how to achieve this using parametric and semi-parametric methods, and describe how to implement these in a one-stage or two-stage IPD-MA framework. We recommend exploring heterogeneity of the effect(s) through interaction and non-linear effects. Random effects should be applied to account for residual heterogeneity of the intervention effect. We provide further recommendations, many of which specific to IPD-MA of time-to-event data from randomized trials examining an intervention effect.We illustrate several key methods in a real IPD-MA, where IPD of 1225 participants from 5 randomized clinical trials were combined to compare the effects of Carbamazepine and Valproate on the incidence of epileptic seizures.

Statistical approaches to identify subgroups in meta-analysis of individual participant data: a simulation study

Background

Individual participant data meta-analysis (IPD-MA) is considered the gold standard for investigating subgroup effects. Frequently used regression-based approaches to detect subgroups in IPD-MA are: meta-regression, per-subgroup meta-analysis (PS-MA), meta-analysis of interaction terms (MA-IT), naive one-stage IPD-MA (ignoring potential study-level confounding), and centred one-stage IPD-MA (accounting for potential study-level confounding). Clear guidance on the analyses is lacking and clinical researchers may use approaches with suboptimal efficiency to investigate subgroup effects in an IPD setting. Therefore, our aim is to overview and compare the aforementioned methods, and provide recommendations over which should be preferred.

Methods

We conducted a simulation study where we generated IPD of randomised trials and varied the magnitude of subgroup effect (0, 25, 50% relative reduction), between-study treatment effect heterogeneity (none, medium, large), ecological bias (none, quantitative, qualitative), sample size (50,100,200), and number of trials (5,10) for binary, continuous and time-to-event outcomes. For each scenario, we assessed the power, false positive rate (FPR) and bias of aforementioned five approaches.

Results

Naive and centred IPD-MA yielded the highest power, whilst preserving acceptable FPR around the nominal 5% in all scenarios. Centred IPD-MA showed slightly less biased estimates than naïve IPD-MA. Similar results were obtained for MA-IT, except when analysing binary outcomes (where it yielded less power and FPR < 5%). PS-MA showed similar power as MA-IT in non-heterogeneous scenarios, but power collapsed as heterogeneity increased, and decreased even more in the presence of ecological bias. PS-MA suffered from too high FPRs in non-heterogeneous settings and showed biased estimates in all scenarios. Meta-regression showed poor power (< 20%) in all scenarios and completely biased results in settings with qualitative ecological bias.

Conclusions

Our results indicate that subgroup detection in IPD-MA requires careful modelling. Naive and centred IPD-MA performed equally well, but due to less bias of the estimates in the presence of ecological bias, we recommend the latter.

Electronic supplementary material

The online version of this article (10.1186/s12874-019-0817-6) contains supplementary material, which is available to authorized users.

Gestational weight gain outside the Institute of Medicine recommendations and adverse pregnancy outcomes: analysis using individual participant data from randomised trials

BACKGROUND

High Body Mass Index (BMI) and gestational weight gain (GWG) affect an increasing number of pregnancies. The Institute of Medicine (IOM) has issued recommendations on the optimal GWG for women according to their pre-pregnancy BMI (healthy, overweight or obese). It has been shown that pregnant women rarely met the recommendations; however, it is unclear by how much. Previous studies also adjusted the analyses for various women's characteristics making their comparison challenging.

METHODS

We analysed individual participant data (IPD) of healthy women with a singleton pregnancy and a BMI of 18.5 kg/m2 or more from the control arms of 36 randomised trials (16 countries). Adjusted odds ratios (aOR) and 95% confidence intervals (CI) were used to describe the association between GWG outside (above or below) the IOM recommendations (2009) and risks of caesarean section, preterm birth, and large or small for gestational age (LGA or SGA) infants. The association was examined overall, within the BMI categories and by quartile of GWG departure from the IOM recommendations. We obtained aOR using mixed-effects logistic regression, accounting for the within-study clustering and a priori identified characteristics.

RESULTS

Out of 4429 women (from 33 trials) meeting the inclusion criteria, two thirds gained weight outside the IOM recommendations (1646 above; 1291 below). The median GWG outside the IOM recommendations was 3.1 kg above and 2.7 kg below. In comparison to GWG within the IOM recommendations, GWG above was associated with increased odds of caesarean section (aOR 1.50; 95%CI 1.25, 1.80), LGA (2.00; 1.58, 2.54), and reduced odds of SGA (0.66; 0.50, 0.87); no significant effect on preterm birth was detected. The relationship between GWG below the IOM recommendation and caesarean section or LGA was inconclusive; however, the odds of preterm birth (1.94; 1.31, 2.28) and SGA (1.52; 1.18, 1.96) were increased.

CONCLUSIONS

Consistently with previous findings, adherence to the IOM recommendations seem to help achieve better pregnancy outcomes. Nevertheless, even in the context of clinical trials, women find it difficult to adhere to them. Further research should focus on identifying ways of achieving a healthier GWG as defined by the IOM recommendations.

Evidence synthesis in prognosis research

Over the past few years, evidence synthesis has become essential to investigate and improve the generalizability of medical research findings. This strategy often involves a meta-analysis to formally summarize quantities of interest, such as relative treatment effect estimates. The use of meta-analysis methods is, however, less straightforward in prognosis research because substantial variation exists in research objectives, analysis methods and the level of reported evidence. We present a gentle overview of statistical methods that can be used to summarize data of prognostic factor and prognostic model studies. We discuss how aggregate data, individual participant data, or a combination thereof can be combined through meta-analysis methods. Recent examples are provided throughout to illustrate the various methods.

The biomarker features of miR-145-3p determined via meta-analysis validated by qRT-PCR in metastatic cancer cell lines

MicroRNAs (miRNAs) play important roles in the cancer biology such as proliferation, differentiation, and apoptosis. The pivotal roles that miRNA expression plays, make them ideal candidates for detection of cancer progression as well as cancer metastasis. Especially for breast, lung and prostate cancer which are originated from soft tissues and prone to metastasis. Thus, the aim of this study is to evaluate the expression level of miR-145-3p which is a shared potential biomarker identified by meta-analysis of breast, prostate and lung cancer data sets. Six different data sets representative of three different cancer types were analyzed. These data sets are pooled together to have a master metamiRNA list while getting rid of the platform differentiations between them. As a result, 24 common differentially expressed miRNAs are determined in which miR-145-3p has the topmost rank. To mimic in vivo cancer microenvironment, hypoxia and serum deprivation were used to induce metastasis in breast (MCF-7, MDA-MB-231, MDA-MB-453), prostate (PC3, LNCaP, DU145), lung (A549, NCIH82,) cancer cell lines and noncancerous cell lines of the coresponding tissues (MCF10A, RWPE-1, MRC-5). miR-145-3p expression levels were determined by qRT-PCR. It has been shown that it is down regulated by the induction of metastasis in cancer cell lines while it is up regulated in normal cell lines to suppress the tumor formation. As a conclusion, as representing the same results in three different cancer cell types, miR-145-3p will be a promising biomarker to follow up its expression to detect cancer metastasis.

Investigating the stratified efficacy and safety of pharmacological blood pressure-lowering: an overall protocol for individual patient-level data meta-analyses of over 300 000 randomised participants in the new phase of the Blood Pressure Lowering Treatment Trialists' Collaboration (BPLTTC)

INTRODUCTION

Previous research from the Blood Pressure Lowering Treatment Trialists' Collaboration (BPLTTC) and others has shown that pharmacological blood pressure (BP)- lowering substantially reduces the risk of major cardiovascular events, including ischaemic heart disease, heart failure and stroke. In this new phase, the aim is to conduct individual patient-level data (IPD) meta-analyses involving eligible BP-lowering randomised controlled trials (RCTs) to address uncertainties relating to efficacy and safety of BP-lowering treatment.

METHODS AND ANALYSIS

RCTs investigating the effect of pharmacological BP-lowering, with a minimum of 1000 patient-years of follow-up in each trial arm, are eligible. Our systematic review identified 100 potentially eligible trials. We requested their investigators/sponsors to contribute baseline, follow-up and outcomes data. As of June 2018, the collaboration has obtained data from 49 trials (n=315 046 participants), with additional data currently in the process of being transferred from four RCTs (n=34 642 participants). In addition, data harmonisation has commenced. Scientific activities of the collaboration are overseen by the Steering Committee with input from all collaborators. Detailed protocols for individual meta-analyses will be developed and registered on public platforms.

ETHICS AND DISSEMINATION

Ethics approval has been obtained for this new and extended phase of the BPLTTC, the largest collaboration of de-identified IPD from RCTs. It offers an efficient and ethical manner of re-purposing existing data to answer clinically important questions relating to BP treatment as well as methodological questions relating to IPD meta-analyses. Among the immediate impacts will include reliable quantification of effects of treatment modifiers, such as baseline BP, age and prior disease, on both vascular and non-vascular outcomes. Analyses will further assess the impact of BP-lowering on important, but less well understood, outcomes, such as new-onset diabetes and renal disease. Findings will be published in peer-reviewed medical journals on behalf of the collaboration.

International incidence of psychotic disorders, 2002-17: a systematic review and meta-analysis

BACKGROUND

The last comprehensive systematic review of the incidence of psychotic disorders was published in 2004. New epidemiological data from different settings now permit a broader understanding of global variation. We examined the variation in psychosis by demographic characteristics and study method.

METHODS

For this systematic review and meta-analysis, we searched PubMed, Embase, Web of Science, PsycINFO, and bibliographies, and directly contacted first authors. We sought to obtain citations of original research published between Jan 1, 2002, and Dec 31, 2017, on incidence of non-organic adult-onset psychotic disorder. We included papers that were published or in grey literature and had no language restrictions. Data were extracted from published reports, where possible, by sex, age, and ethnic group. Quality of yield was assessed. Data were assessed using univariable random-effects meta-analysis and meta-regression. We registered our systematic review on PROSPERO, number CRD42018086800.

FINDINGS

From 56 721 records identified, 177 met inclusion criteria. The pooled incidence of all psychotic disorders was 26·6 per 100 000 person-years (95% CI 22·0-31·7). Heterogeneity was high (I2≥98·5%). Men were at higher risk of all psychotic disorders (incidence rate ratio 1·44 [1·27-1·62]) and non-affective disorders (1·60 [1·44-1·77]) than women, but not affective psychotic disorders (0·87 [0·75-1·00]). Ethnic minorities were also at excess risk of all psychotic disorders (1·75 [1·53-2·00]), including non-affective disorders (1·71 [1·40-2·09]). Meta-regression revealed that population registers reported higher rates of non-affective disorders (9·64 [2·72-31·82]), schizophrenia (2·51 [1·24-5·21]), and bipolar disorder (4·53 [2·41-8·51]) than first contact study designs.

INTERPRETATION

We found marked variation in incidence of psychotic disorders by personal characteristics and place. Some geographical variation could be partially explained by differences in case ascertainment methods.

FUNDING

None.

Conducting gene set tests in meta-analyses of transcriptome expression data

Research synthesis, eg, by meta-analysis, is more and more considered in the area of high-dimensional data from molecular research such as gene and protein expression data, especially because most studies and experiments are performed with very small sample sizes. In contrast to most clinical and epidemiological trials, raw data are often available for high-dimensional expression data. Therefore, direct data merging followed by a joint analysis of selected studies can be an alternative to meta-analysis by P value or effect-size merging or, more generally spoken, the merging of results. While several methods for meta-analysis of differential expression studies have been proposed, meta-analysis of gene set tests has very rarely been considered, although gene set tests are standard in the analysis of individual gene expression studies. We compare in this work the different strategies of research synthesis of gene set tests, in particularly the "early merging" of data cleaned from batch effects versus the "late merging" of individual results. In simulation studies and in examples of manipulated real-world data, we found that in most scenarios, the early merging has a higher sensitivity of detecting a gene set enrichment than the late merging. However, in scenarios with few studies, large batch effect, moderate and large sample sizes of late merging are more sensitive than early merging.