A comparison of confidence distribution approaches for rare event meta‐analysis

2021 ◽  
Author(s):  
Brinley N. Zabriskie ◽  
Chris Corcoran ◽  
Pralay Senchaudhuri
Keyword(s):  
Meta Analysis ◽  
Rare Event ◽  
Confidence Distribution

scholarly journals Meta-analysis of rare adverse events in randomized clinical trials: Bayesian and frequentist methods

2020 ◽  
pp. 174077452096913
Author(s):  
Hwanhee Hong ◽  
Chenguang Wang ◽  
Gary L Rosner
Keyword(s):  
Adverse Events ◽  
Coverage Probability ◽  
Randomized Clinical Trials ◽  
Meta Analysis ◽  
Statistical Significance ◽  
Rare Event ◽  
Model Performance ◽  
Study Results ◽  
Wide Range ◽  
Study Heterogeneity

Background/aims: Regulatory approval of a drug or device involves an assessment of not only the benefits but also the risks of adverse events associated with the therapeutic agent. Although randomized controlled trials (RCTs) are the gold standard for evaluating effectiveness, the number of treated patients in a single RCT may not be enough to detect a rare but serious side effect of the treatment. Meta-analysis plays an important role in the evaluation of the safety of medical products and has advantage over analyzing a single RCT when estimating the rate of adverse events. Methods: In this article, we compare 15 widely used meta-analysis models under both Bayesian and frequentist frameworks when outcomes are extremely infrequent or rare. We present extensive simulation study results and then apply these methods to a real meta-analysis that considers RCTs investigating the effect of rosiglitazone on the risks of myocardial infarction and of death from cardiovascular causes. Results: Our simulation studies suggest that the beta hyperprior method modeling treatment group-specific parameters and accounting for heterogeneity performs the best. Most models ignoring between-study heterogeneity give poor coverage probability when such heterogeneity exists. In the data analysis, different methods provide a wide range of log odds ratio estimates between rosiglitazone and control treatments with a mixed conclusion on their statistical significance based on 95% confidence (or credible) intervals. Conclusion: In the rare event setting, treatment effect estimates obtained from traditional meta-analytic methods may be biased and provide poor coverage probability. This trend worsens when the data have large between-study heterogeneity. In general, we recommend methods that first estimate the summaries of treatment-specific risks across studies and then relative treatment effects based on the summaries when appropriate. Furthermore, we recommend fitting various methods, comparing the results and model performance, and investigating any significant discrepancies among them.

scholarly journals Incidence of sudden cardiac arrest and death in young athletes and military members: a systematic review and meta-analysis

2021 ◽  
Author(s):  
Aaron Lear ◽  
Niraj Patel ◽  
Chanda Mullen ◽  
Marian Simonson ◽  
Vince Leone ◽  
...  
Keyword(s):  
Cardiac Arrest ◽  
Incidence Rate ◽  
Meta Analysis ◽  
National Level ◽  
Sudden Cardiac Arrest ◽  
Rare Event ◽  
Risk Of Bias ◽  
Low Risk ◽  
Military Members

Abstract Objectives: The goals of this review are to evaluate the quality of the evidence on the incidence of sudden cardiac arrest and death (SCA/D) in athletes and military members; and to estimate annual incidence of SCA/D. Data Sources: MEDLINE, Embase, Cochrane CENTRAL, Web of Science, BIOSIS, Scopus, SPORT discus, PEDro, and clinicaltrials.gov were searched from inception to dates between 2/21/19–7/29/19. Study Selection: Studies which reported incidence of SCA/D or both in athletes, or military members under age 40 were eligible for inclusion. 40 studies were identified for inclusion Data Extraction: Risk of bias was assessed using a validated, customized tool for prevalence studies in all included studies. 12 were found to be low ROB, with the remaining 28 moderate or high ROB. Data was extracted for narrative review, and meta-analysis. Data Synthesis: Random-effects meta-analysis was performed in studies judged to be low risk of bias in two separate categories: 5 studies on regional or national level data including athletes at all levels, and both sexes included 130 events of SCD, with a total of 11,272,560 athlete years showing a cumulative incidence rate of 0.98 [95%CI: 0.62, 1.53] per 100,000 athlete years, with high heterogeneity with I2 of 78%; 3 Studies on competitive athletes aged 14 to 25 were combined, and included 183 events, and 17,798758 athlete years showing an incidence rate of 1.91[95%CI: 0.71; 5.14] per 100,000 athlete years with high heterogeneity with I2 of 97%. The remaining low risk of bias studies were in military members, and were not synthesized. Conclusion: The worldwide incidence of SCD is a rare event. Low risk of bias studies indicate incidence to be below 2 per 100,000 athlete years. Overall, the quality of the evidence available is low, but there are high quality individual studies to inform the question of incidence levels. PROSPERO Registration: CRD42019125560

scholarly journals Impact of including or excluding both-armed zero-event studies on using standard meta-analysis methods for rare event outcome: a simulation study

BMJ Open ◽  
2016 ◽  
Vol 6 (8) ◽  
pp. e010983 ◽  
Author(s):  
Ji Cheng ◽  
Eleanor Pullenayegum ◽  
John K Marshall ◽  
Alfonso Iorio ◽  
Lehana Thabane
Keyword(s):  
Simulation Study ◽  
Meta Analysis ◽  
Rare Event ◽  
Event Studies ◽  
Analysis Methods ◽  
Zero Event

VP26 Comparing Statistical Methods For Meta-Analysis Of Rare Event Data

2017 ◽  
Vol 33 (S1) ◽  
pp. 158-159
Author(s):  
Tarang Sharma ◽  
Peter Gøtzsche ◽  
Oliver Kuss
Keyword(s):  
Logistic Regression ◽  
Adverse Events ◽  
Odds Ratio ◽  
Meta Analysis ◽  
Conditional Logistic Regression ◽  
Rare Event ◽  
Sensitivity Analyses ◽  
Analysis Methods ◽  
Binomial Regression ◽  
Events Data

INTRODUCTION:We aimed to identify the validity and robustness of effect estimates for serious rare adverse events in clinical study reports of antidepressant trials, across different meta-analysis methods for rare binary events data (1,2).METHODS:Four serious rare adverse events (all-cause mortality, suicidality, aggressive behaviour and akathisia) were meta-analyzed using different methods (3). The Yusuf-Peto odds ratio (OR), which ignores studies with no events in the treatment arms, was compared with the alternative approaches of generalized linear mixed models (GLMM), conditional logistic regression, a Bayesian approach using Markov Chain Monte Carlo (MCMC) and a beta-binomial regression model.RESULTS:Though the estimates for the four outcomes did not change substantially across the different analysis methods, the Yusuf-Peto method underestimated the treatment harm and overstimated its precision, especially when the estimated odds ratio (OR) deviated greatly from 1. For example the OR for suicidality for children and adolescents was 2.39 (95 percent Confidence Interval, CI 1.32 to 4.33, using the Yusuf-Peto method), but increased to 2.64 (95 percent CI 1.33 to 5.26) using conditional logistic regression, to 2.69 (95 percent CI 1.19 to 6.09) using beta-binomial, to 2.73 (95 percent CI 1.37 to 5.42) using the GLMM and finally to 2.87 (95 percent CI 1.42 to 5.98) using the MCMC approach.CONCLUSIONS:The method used for meta-analysis of rare events data influences the estimates obtained and the exclusion of double zero-event studies can give misleading results. To ensure reduction of bias and erroneous inferences, sensitivity analyses should be performed using different methods and we recommend that the Yusuf-Peto approach should no longer be used. Other methods, in particular the beta-binomial method that was shown to be superior, should be considered instead.

scholarly journals Prevalence of uterine rupture among women with one prior low transverse cesarean and women with unscarred uterus undergoing labor induction with PGE2: A systematic review and meta-analysis

PLoS ONE ◽  
2021 ◽  
Vol 16 (7) ◽  
pp. e0253957
Author(s):  
Giuseppe Chiossi ◽  
Roberto D’Amico ◽  
Anna L. Tramontano ◽  
Veronica Sampogna ◽  
Viola Laghi ◽  
...  
Keyword(s):  
Uterine Rupture ◽  
Meta Analysis ◽  
Rare Event ◽  
Labor Induction ◽  
Cervical Ripening ◽  
Cochrane Library ◽  
Cesarean Birth ◽  
Pooled Prevalence ◽  
Arcsine Transformation ◽  
The Cochrane Library

Background As uterine rupture may affect as many as 11/1000 women with 1 prior cesarean birth and 5/10.000 women with unscarred uterus undergoing labor induction, we intended to estimate the prevalence of such rare outcome when PGE2 is used for cervical ripening and labor induction. Methods We searched MEDLINE, ClinicalTrials.gov and the Cochrane library up to September 1st 2020. Retrospective and prospective cohort studies, as well as randomized controlled trials (RCTs) on singleton viable pregnancies receiving PGE2 for cervical ripening and labor induction were reviewed. Prevalence of uterine rupture was meta-analyzed with Freeman-Tukey double arcsine transformation among women with 1 prior low transverse cesarean section and women with unscarred uterus. Results We reviewed 956 full text articles to include 69 studies. The pooled prevalence rate of uterine rupture is estimated to range between 2 and 9 out of 1000 women with 1 prior low transverse cesarean (5/1000; 95%CI 2-9/1000, 122/9000). The prevalence of uterine rupture among women with unscarred uterus is extremely low, reaching at most 0.7/100.000 (<1/100.000.000; 95%CI <1/100.000.000–0.7/100.000, 8/17.684). Conclusions Uterine rupture is a rare event during cervical ripening and labor induction with PGE2.

scholarly journals Prediction intervals for random-effects meta-analysis: A confidence distribution approach

2018 ◽  
Vol 28 (6) ◽  
pp. 1689-1702 ◽  
Author(s):  
Kengo Nagashima ◽  
Hisashi Noma ◽  
Toshi A Furukawa
Keyword(s):  
Random Effects ◽  
Prediction Interval ◽  
Meta Analysis ◽  
Prediction Intervals ◽  
Point Estimate ◽  
Random Effects Models ◽  
Confidence Distribution ◽  
Large Sample Approximation ◽  
Sample Approximation ◽  
Meta Analyses

Prediction intervals are commonly used in meta-analysis with random-effects models. One widely used method, the Higgins–Thompson–Spiegelhalter prediction interval, replaces the heterogeneity parameter with its point estimate, but its validity strongly depends on a large sample approximation. This is a weakness in meta-analyses with few studies. We propose an alternative based on bootstrap and show by simulations that its coverage is close to the nominal level, unlike the Higgins–Thompson–Spiegelhalter method and its extensions. The proposed method was applied in three meta-analyses.

Impact of Inherited Thrombophilia on Symptomatic Venous Thrombo-Embolism (VTE) in Children: A Systematic Review & Meta-Analysis of 37 Studies Including 2470 Pediatric Patients.

Blood ◽  
2007 ◽  
Vol 110 (11) ◽  
pp. 3189-3189 ◽  
Author(s):  
Guy Young ◽  
Frauke Friedrichs ◽  
Anthony Chan ◽  
Gili Kenet ◽  
Paolo Simioni ◽  
...  
Keyword(s):  
Risk Factors ◽  
Risk Factor ◽  
Fixed Effects ◽  
Meta Analysis ◽  
Rare Event ◽  
Cochrane Library ◽  
Inclusion Criteria ◽  
Inherited Thrombophilia ◽  
Mesh Terms ◽  
The Impact

Abstract Background: Inherited thrombophilia (IT) has been described as a risk factor for venous thromboembolism (VTE) in children. So far the majority of studies performed in the field were either retrospective or prospective on small numbers of patients. Thus, the results are contradictory or inconclusive mainly due to lack of statistical power. The aim of this study was to better estimate the impact of IT on early VTE onset and recurrence in children as a prerequisite to develop primary and secondary treatment options. Methods: A systematic search of publications listed in the electronic databases (Pubmed, Medline, EMBASE, Web of Science, The Cochrane Library) up to August 2007 using key words in combination both as MeSH terms and text words, was conducted. Citations were screened by two independent group members and those meeting the inclusion criteria were retained. Articles were included if published after 1990, when pediatric VTE was started to be systematically investigated. Findings: Twenty case-control and 17 cohort studies from 13 countries met the inclusion criteria. In these studies > 70% of patients had at least one clinical risk factor. The summary odds ratios (OR) and 95% confidence intervals (CI) of included studies under a fixed-effects and random-effects model showed statistically significant associations between the IT traits investigated and VTE onset (table). For the rare event of VTE recurrence, 1227 patients (eight studies) were evaluated: at the present state due to high heterogeneity, a trend towards association with recurrent VTE was found for ≥2 IT traits in the fixed-effects model (0R/CI: 2.8/1.6–4.8). Interpretation: The present meta-analysis gives evidence that the detection of inherited thrombophilia is clincially meaningful in children with VTE and underlines the importance of a pediatric thrombophilia screening program. Summary of Data Risk Factors OR/CI:fixed model OR/CI:random model patients/controls 2470/4119 N/A FV G1691A 3.5/2.9–4.2 3.2/2.3–4.4 FII G20210A 2.2/1.5–3.3 2.2/1.5–3.4 Protein C defiiciency 9.8/5.9–16 9.9/6.1–16.1 Protein S deficiency 7.1/3.9–13.2 6.8/3.7–12.7 Antithrombin deficiency 7.9/3.8–16.6 7.3/3.4–15.3 Lipoprotein(a) 4.4/3,2–5.9 4/2.4–6.6 ≥ 2 risk factors 12.6/7.3–21.8 11.6/6.2–20.2

Sign in / Sign up

Export Citation Format

Share Document