Metaan: Random-effects Meta-analysis

Kontopantelis, Evangelos; Reeves, David

doi:10.1177/1536867x1001000307

Cited by 131 publications

(92 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, these methods quite often fail to detect heterogeneity and thus produce biased estimates and conclusions, especially for small meta-analyses. The best method, overall, in particular in terms of detecting heterogeneity when it is present, appeared to be a non-parametric bootstrap of the DerSimonian-Laird, which we have implemented in Stata [34]. Although the method often produces positively biased heterogeneity estimates, especially for small meta-analyses, it performs very well on coverage, error-interval estimation and heterogeneity detection.…”

Section: Discussionmentioning

confidence: 99%

A Re-Analysis of the Cochrane Library Data: The Dangers of Unobserved Heterogeneity in Meta-Analyses

2013

Self Cite

View full text Add to dashboard Cite

BackgroundHeterogeneity has a key role in meta-analysis methods and can greatly affect conclusions. However, true levels of heterogeneity are unknown and often researchers assume homogeneity. We aim to: a) investigate the prevalence of unobserved heterogeneity and the validity of the assumption of homogeneity; b) assess the performance of various meta-analysis methods; c) apply the findings to published meta-analyses.Methods and FindingsWe accessed 57,397 meta-analyses, available in the Cochrane Library in August 2012. Using simulated data we assessed the performance of various meta-analysis methods in different scenarios. The prevalence of a zero heterogeneity estimate in the simulated scenarios was compared with that in the Cochrane data, to estimate the degree of unobserved heterogeneity in the latter. We re-analysed all meta-analyses using all methods and assessed the sensitivity of the statistical conclusions. Levels of unobserved heterogeneity in the Cochrane data appeared to be high, especially for small meta-analyses. A bootstrapped version of the DerSimonian-Laird approach performed best in both detecting heterogeneity and in returning more accurate overall effect estimates. Re-analysing all meta-analyses with this new method we found that in cases where heterogeneity had originally been detected but ignored, 17–20% of the statistical conclusions changed. Rates were much lower where the original analysis did not detect heterogeneity or took it into account, between 1% and 3%.ConclusionsWhen evidence for heterogeneity is lacking, standard practice is to assume homogeneity and apply a simpler fixed-effect meta-analysis. We find that assuming homogeneity often results in a misleading analysis, since heterogeneity is very likely present but undetected. Our new method represents a small improvement but the problem largely remains, especially for very small meta-analyses. One solution is to test the sensitivity of the meta-analysis conclusions to assumed moderate and large degrees of heterogeneity. Equally, whenever heterogeneity is detected, it should not be ignored.

show abstract

Section: Discussionmentioning

confidence: 99%

A Re-Analysis of the Cochrane Library Data: The Dangers of Unobserved Heterogeneity in Meta-Analyses

2013

Self Cite

View full text Add to dashboard Cite

show abstract

“…Fixed-effect models weight studies according to the amount of information they contribute, whereas random-effects models incorporate an estimate of between-study variation (heterogeneity) in the weighting. The fixed-effect assumption is that the true treatment effect is the same in each study, despite any differences in study protocols [35]. We believe a fixed effect model is appropriate as larger studies should be given more weight than smaller ones, and as there are few studies used in our meta-analysis, using a random effects model would provide poor estimates of the distribution of the intervention effects.…”

Section: Methodsmentioning

confidence: 99%

Weighing as a stand-alone intervention does not reduce excessive gestational weight gain compared to routine antenatal care: a systematic review and meta-analysis of randomised controlled trials

Fealy

Taylor

Foureur

et al. 2017

BMC Pregnancy Childbirth

View full text Add to dashboard Cite

BackgroundExcessive gestational weight gain is associated with short and long-term adverse maternal and infant health outcomes, independent of pre-pregnancy body mass index. Weighing pregnant women as a stand-alone intervention during antenatal visits is suggested to reduce pregnancy weight gain. In the absence of effective interventions to reduce excessive gestational gain within the real world setting, this study aims to test if routine weighing as a stand-alone intervention can reduce total pregnancy weight gain and, in particular, excessive gestational weight gain.MethodsA systematic review and meta–analysis of randomised controlled trials (RCTs) was conducted between November 2014 and January 2016, and reported using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses. Seven databases were searched. A priori eligibility criteria were applied to published literature by at least two independent reviewers. Studies considered methodologically rigorous, as per the Academy of Nutrition and Dietetics Quality Criteria Checklist for Primary Research, were included. Meta-analysis was conducted using fixed-effects models.ResultsA total of 5223 (non-duplicated) records were screened, resulting in two RCTs that were pooled for meta-analysis (n = 1068 randomised participants; n = 538 intervention, n = 534 control). No difference in total weight gain per week was observed between intervention and control groups (weighted mean difference (WMD) -0.00 kg/week, 95% confidence interval (CI) -0.03 to 0.02). There was also no reduction in excessive gestational weight gain between intervention and control, according to pre-pregnancy body mass index (BMI). However, total weight gain was lower in underweight women (n = 23, BMI <18.5 kg/m2) in the intervention compared to control group (−0.12 kg/week, 95% CI −0.23 to −0.01). No significant differences were observed for other pregnancy, birth and infant outcomes.ConclusionWeighing as a stand-alone intervention is not worse nor better at reducing excessive gestational weight gain than routine antenatal care.Electronic supplementary materialThe online version of this article (doi:10.1186/s12884-016-1207-2) contains supplementary material, which is available to authorized users.

show abstract

“…I 2 values of 0% to 24.9%, 25% to 49.9%, 50% to 74.9%, and 75% to 100% were considered as having no, mild, moderate, and significant thresholds for statistical heterogeneity. 14,15 A random-effects model using restricted maximum likelihood, 16 which is thought to be better than the conventional DerSimonian-Laird method, 17 was performed to provide more conservative estimates of effect in the presence of known or unknown heterogeneity. Sensitivity analysis was conducted by sequentially omitting each study one at a time in an attempt to identify the potential influence of an individual study.…”

Section: Quantitative Data Synthesismentioning

confidence: 99%

Extended Daily Dialysis Versus Continuous Renal Replacement Therapy for Acute Kidney Injury: A Meta-analysis

Zhang

Yang

Eastwood

et al. 2015

American Journal of Kidney Diseases

125

View full text Add to dashboard Cite

Metaan: Random-effects Meta-analysis

Cited by 131 publications

References 21 publications

A Re-Analysis of the Cochrane Library Data: The Dangers of Unobserved Heterogeneity in Meta-Analyses

A Re-Analysis of the Cochrane Library Data: The Dangers of Unobserved Heterogeneity in Meta-Analyses

Weighing as a stand-alone intervention does not reduce excessive gestational weight gain compared to routine antenatal care: a systematic review and meta-analysis of randomised controlled trials

Extended Daily Dialysis Versus Continuous Renal Replacement Therapy for Acute Kidney Injury: A Meta-analysis

Contact Info

Product

Resources

About