scholarly journals Testing for funnel plot asymmetry of standardized mean differences

2019 ◽  
Vol 10 (1) ◽  
pp. 57-71 ◽  
Author(s):  
James E. Pustejovsky ◽  
Melissa A. Rodgers
Keyword(s):  
Funnel Plot ◽  
Funnel Plot Asymmetry ◽  
Mean Differences

scholarly journals Recommendations for examining and interpreting funnel plot asymmetry in meta-analyses of randomised controlled trials

BMJ ◽  
2011 ◽  
Vol 343 (jul22 1) ◽  
pp. d4002-d4002 ◽  
Author(s):  
J. A. C. Sterne ◽  
A. J. Sutton ◽  
J. P. A. Ioannidis ◽  
N. Terrin ◽  
D. R. Jones ◽  
...  
Keyword(s):  
Randomised Controlled Trials ◽  
Controlled Trials ◽  
Funnel Plot ◽  
Funnel Plot Asymmetry ◽  
Randomised Controlled ◽  
Meta Analyses

scholarly journals Author response: Standardized mean differences cause funnel plot distortion in publication bias assessments

2017 ◽  
Author(s):  
Peter-Paul Zwetsloot ◽  
Mira Van Der Naald ◽  
Emily S Sena ◽  
David W Howells ◽  
Joanna IntHout ◽  
...  
Keyword(s):  
Publication Bias ◽  
Author Response ◽  
Funnel Plot ◽  
Mean Differences

scholarly journals Does retirement trigger depressive symptoms? A systematic review and meta-analysis

2021 ◽  
Vol 30 ◽  
Author(s):  
A. Odone ◽  
V. Gianfredi ◽  
G. P. Vigezzi ◽  
A. Amerio ◽  
C. Ardito ◽  
...  
Keyword(s):  
Mental Health ◽  
Systematic Review ◽  
Depressive Symptoms ◽  
Protective Effect ◽  
Meta Analysis ◽  
Assessment Tools ◽  
Cochrane Library ◽  
High Quality ◽  
Funnel Plot ◽  
Funnel Plot Asymmetry

Abstract Aims Retirement is a major life transition that may improve or worsen mental health, including depression. Existing studies provide contradictory results. We conducted a systematic review with meta-analysis to quantitatively pool available evidence on the association of retirement and depressive symptoms. Methods We applied PRISMA guidelines to conduct a systematic review and meta-analysis to retrieve, quantitatively pool and critically evaluate the association between retirement and both incident and prevalent depression and to understand better the potential role of individual and contextual-level determinants. Relevant original studies were identified by searching PubMed, Embase, PsycINFO and the Cochrane Library, through 4 March 2021. Subgroup and sensitivity meta-analyses were conducted by gender, study design (longitudinal v. cross-sectional studies), study quality score (QS) and considering studies using validated scales to diagnose depression. Heterogeneity between studies was evaluated with I2 statistics. Results Forty-one original studies met our a priori defined inclusion criteria. Meta-analysis on more than half a million subjects (n = 557 111) from 60 datasets suggested a protective effect of retirement on the risk of depression [effect size (ES) = 0.83, 95% confidence interval (CI) = 0.74–0.93], although with high statistical heterogeneity between risk estimates (χ2 = 895.19, df = 59, I2 = 93.41%, p-value < 0.0001). Funnel plot asymmetry and trim and fill method suggested a minor potential publication bias. Results were consistent, confirm their robustness and suggest stronger protective effects when progressively restricting the included studies based on quality criteria: (i) studies with the highest QS [55 datasets, 407 086 subjects, ES = 0.81, 95% CI = 0.71–0.91], (ii) studies with a high QS and using validated assessment tools to diagnose depression (44 datasets, 239 453 subjects, ES = 0.76, 95% CI = 0.65–0.88) and (iii) studies of high quality, using a validated tool and with a longitudinal design (24 datasets, 162 004 subjects, ES = 0.76, 95% CI = 0.64–0.90). We observed a progressive reduction in funnel plot asymmetry. About gender, no statistically significant difference was found (females ES = 0.79, 95% CI = 0.61–1.02 v. men ES = 0.87, 95% CI = 0.68–1.11). Conclusions Pooled data suggested that retirement reduces by nearly 20% the risk of depression; such estimates got stronger when limiting the analysis to longitudinal and high-quality studies, even if results are affected by high heterogeneity. As retirement seems to have an independent and protective effect on mental health and depressive symptoms, greater flexibility in retirement timing should be granted to older workers to reduce their mental burden and avoid the development of severe depression. Retirement may also be identified as a target moment for preventive interventions, particularly primary and secondary prevention, to promote health and wellbeing in older ages, boosting the observed impact.

scholarly journals THU0475 THE EFFICACY OF ORAL GLUCOCORTICOSTEROIDS FOR PAIN IN RHEUMATOID ARTHRITIS: A PRELIMINARY REPORT OF A META-ANALYSIS

2020 ◽  
Vol 79 (Suppl 1) ◽  
pp. 474.2-475
Author(s):  
D. Mcwilliams ◽  
D. Thankaraj ◽  
R. Morgan ◽  
J. Jones-Diette ◽  
D. Walsh
Keyword(s):  
Meta Analysis ◽  
Bodily Pain ◽  
Significant Heterogeneity ◽  
Research Support ◽  
Funnel Plot ◽  
Specific Effects ◽  
Grant Support ◽  
Subgroup Analyses ◽  
Mean Differences

Background:Glucocorticosteroids (GCs) are used to provide rapid relief of symptoms in people with active RA. Their use is recommended by most RA management guidelines and systematic reviews, although the magnitude of their benefit above placebo is uncertain. Persistent pain remains a problem in RA, even despite optimal immunomodulatory management. Systemic GC use may be associated with important adverse events.Objectives:To quantify the specific effects of oral GCs for RA pain.Methods:A systematic literature review was performed for RCTs using GCs in RA compared to inactive treatment. Trials were included whether or not participants received DMARD treatments, so long as a specific effect could be assigned to GCs. Medline, Embase and Cochrane databases were searched until November 2019 and 2 reviewers independently assessed titles, abstracts and full texts. Data for pain were synthesized in a meta-analysis. This study is part of a wider review (PROSPERO CRD42019111562).For subgroup analyses, follow up time points of 0-3 months, >3 - 6 months and >6 months were selected to address duration of effect. Individual studies could contribute to each of the 3 follow up subgroups.Meta-analysis was performed on standardized mean differences (SMDs, bodily pain data) and mean differences (MDs, 100mm VASpain only) of change from baseline (sd), using the Meta and Metafor packages in R. Heterogeneity was quantified using I2and tau statistics. Bias was assessed with a funnel plot and Eggers test.Results:15983 papers, 470 abstracts and 152 full texts were assessed. Pain data from 12 RCTs were suitable for the meta-analyses. The most common pain metric was the 100mm VASpain (9 trials).Study populations ranged from n=12 to n=350 participants, 50% to 71% were female with mean ages from 43 to 66 years. Baseline scores for VASpain ranged from to 34 to 66 mm. Means were reported for DAS28 (from 4.9 to 5.8), ESR (25 to 60mm) and CRP (5 to 27mg/L).Data synthesis at the reported primary time point/end point showed a statistically significant reduction in bodily pain in participants treated with GCs; SMD = -0.36 (10 studies, 1377 participants, 95% CI, -0.59 to -0.14, p=0.002) with significant heterogeneity (I2= 66%, tau = 0.27, p<0.01). The Funnel plot suggested asymmetry, favouring GCs (Eggers p = 0.007).Subgroup analyses were used to investigate the time course of specific effects on pain. Efficacy displayed time-related decreases after initiation. From 0-3 months SMD= -0.56 (95% CI, -0.76 to -0.36, p<0.001, 9 studies, 936 participants, I2= 43%, Eggers p= 0.002). Efficacy was lower at >3 - 6 months (SMD= -0.32, 95%CI -0.52 to -0.11, p=0.002, 3 studies, 382 participants, I2=0%, Eggers p=0.75) and further reduced at >6 months (SMD= -0.07, 95%CI, -0.23 to 0.08, p=0.357, 4 studies, 665 participants, I2= 7%, Eggers p=0.43).For trial data collected during concomitant oral GC dosage, mean difference (MDs) in 100mm VASpain was -14mm (95% CI, -20mm to -9mm) greater improvement in GC than control in the 0-3 month period (8 studies, 1047 participants, I2= 70%). For later follow ups, MDs at >3 to 6 months were -6mm (95% CI, -11mm to -1mm, 3 studies, 537 participants), and -1mm (95% CI, -6mm to 4mm, 3 studies, 369 participants) at >6 months.Conclusion:Oral GCs have efficacy for pain in RA but the mean effect is of borderline clinical importance, and greatest shortly after steroids are commenced. GCs were usually used alongside other treatments as part of a combination. Future research might determine who might gain most benefit from systemic GCs, and improve other treatments to reduce the burden of pain.Acknowledgments:Dr Douglas Grindley for help devising search strategyDisclosure of Interests: :Daniel McWilliams Grant/research support from: Grant support from Pfizer Ltd, Divya Thankaraj: None declared, Rheinallt Morgan: None declared, Julie Jones-Diette: None declared, David Walsh Grant/research support from: Grant support from Pfizer Ltd and Eli Lilly, Consultant of: Consultancy to Eli Lilly, Pfizer, Abbvie and GSK (all payments made to University of Nottingham). Consultancy to Love Productions(all payments made to the University of Nottingham).

scholarly journals Standardized mean differences cause funnel plot distortion in publication bias assessments

eLife ◽  
2017 ◽  
Vol 6 ◽  
Author(s):  
Peter-Paul Zwetsloot ◽  
Mira Van Der Naald ◽  
Emily S Sena ◽  
David W Howells ◽  
Joanna IntHout ◽  
...  
Keyword(s):  
Sample Size ◽  
Publication Bias ◽  
Small Sample Size ◽  
Small Sample ◽  
Mean Difference ◽  
Funnel Plot ◽  
Funnel Plots ◽  
Mean Differences ◽  
Meta Analyses ◽  
Positive Results

Meta-analyses are increasingly used for synthesis of evidence from biomedical research, and often include an assessment of publication bias based on visual or analytical detection of asymmetry in funnel plots. We studied the influence of different normalisation approaches, sample size and intervention effects on funnel plot asymmetry, using empirical datasets and illustrative simulations. We found that funnel plots of the Standardized Mean Difference (SMD) plotted against the standard error (SE) are susceptible to distortion, leading to overestimation of the existence and extent of publication bias. Distortion was more severe when the primary studies had a small sample size and when an intervention effect was present. We show that using the Normalised Mean Difference measure as effect size (when possible), or plotting the SMD against a sample size-based precision estimate, are more reliable alternatives. We conclude that funnel plots using the SMD in combination with the SE are unsuitable for publication bias assessments and can lead to false-positive results.

Analyzing Observed Composite Differences Across Groups

Methodology ◽  
2013 ◽  
Vol 9 (1) ◽  
pp. 1-12 ◽  
Author(s):  
Holger Steinmetz
Keyword(s):  
Monte Carlo ◽  
Structural Equation Modeling ◽  
Measurement Invariance ◽  
Structural Equation ◽  
Monte Carlo Study ◽  
Equation Modeling ◽  
Factor Loadings ◽  
Latent Mean Differences ◽  
Partial Invariance ◽  
Mean Differences

Although the use of structural equation modeling has increased during the last decades, the typical procedure to investigate mean differences across groups is still to create an observed composite score from several indicators and to compare the composite’s mean across the groups. Whereas the structural equation modeling literature has emphasized that a comparison of latent means presupposes equal factor loadings and indicator intercepts for most of the indicators (i.e., partial invariance), it is still unknown if partial invariance is sufficient when relying on observed composites. This Monte-Carlo study investigated whether one or two unequal factor loadings and indicator intercepts in a composite can lead to wrong conclusions regarding latent mean differences. Results show that unequal indicator intercepts substantially affect the composite mean difference and the probability of a significant composite difference. In contrast, unequal factor loadings demonstrate only small effects. It is concluded that analyses of composite differences are only warranted in conditions of full measurement invariance, and the author recommends the analyses of latent mean differences with structural equation modeling instead.

Sign in / Sign up

Export Citation Format

Share Document