Comparing two methods of meta-analysis in clinical research - individual patient data-based (IPD) and literature-based abstracted data (AD) methods: Analyzing five oncology issues involving more than 10,000 patients in randomized clinical trials (RCTs)

6512 Background: Meta-analyses are one of the highest recommendations levels in Evidence-Based Medicine (EBM). Recently, meta-analyses have increased using either IPD or AD methods. Controversy exists regarding reliability, applicability and feasibility of the different methods to draw conclusions from conflicting RCTs and to estimate magnitude of benefit of different treatments. Methods: As seen in the table , we selected 5 major issues in 3 malignancies subjected to IPD meta-analysis, and then conducted AD meta-analyses from publications of the individual studies, using published methods (Bria, Cancer Treat Rev 2006). We required that >90% of patient numbers for both IPD and AD analyses be available. Event-based relative risk ratios (RRs) with 95% confidence intervals (CI) were derived. Fixed- and random-effect models, and absolute benefits (AB) were calculated. Correlations between IPD Hazard Ratios (HRs) and AD-RRs were estimated using a linear regression model according to Pearson (r) and R2 coefficients (parametric) and Spearman (Rho) coefficient (non- parametric). Results: Results are below. A strong linear correlation exists between IPD-HRs and AD-RRs (r=0.994, R2=0.989; p<0.001; Rho=1.00). Conclusions: The strong correlation supports using high quality meta-analyses with either method to resolve major issues. Differences exist between the methods: IDP is well-suited for sensitivity analyses and for hypothesis generation involving issues not originally anticipated; AD is a practical method allowing EBM to be applied rapidly to major issues in oncology in a timely fashion. No significant financial relationships to disclose. [Table: see text]

Download Full-text

How much is reasonable to expect about overall survival (OS) benefit when designing studies with new drugs for patients affected by castration resistant prostate cancer (CRPC)? Meta-analysis of 23 randomized clinical trials (RCT) including 17,640 patients.

Journal of Clinical Oncology ◽

10.1200/jco.2013.31.15_suppl.e16053 ◽

2013 ◽

Vol 31 (15_suppl) ◽

pp. e16053-e16053

Author(s):

Francesco Massari ◽

Francesca Maines ◽

Sara Pilotto ◽

Camillo Porta ◽

Paolo Carlini ◽

...

Keyword(s):

Treatment Strategy ◽

Significant Interaction ◽

Randomized Clinical Trials ◽

Meta Analysis ◽

Random Effect ◽

Random Effect Model ◽

Phase Iii ◽

Sensitivity Analyses ◽

Significant Heterogeneity ◽

Significant Difference

e16053 Background: Treatment decision making in patients affected by CRPC is difficult because the numerous available therapeutic opportunities can significantly affect OS. To demonstrate that comparing results in absence of head-to-head studies may lead to biased survival estimations, a literature-based meta-analysis was conducted. Methods: Hazard Ratios (HR) with 95% confidence intervals (CI) were extracted and cumulated according to a random-effect model from phase III trials. Sensitivity analyses were performed according to: 1) Treatment Strategy (TS, Chemotherapy versus Hormonal versus Immunotherapy versus Other), 2) Comparison (Chemotherapy versus Placebo versus Other), and 3) Disease setting with regard to treatment with Docetaxel (DOC),. Testing for heterogeneity was performed as well. Results: A significant heterogeneity for the 3 sensitivity analyses was found (p<0.0001). The cumulative HR in favor of (any) experimental arm was 0.91 (95% CI 0.84-0.99, p=0.028). We found a significant interaction according to the chosen TS (p<0.0001), in fact a significant difference in OS was more likely to be detected in RCT evaluating hormonal drugs (HR 0.76, 95% CI 0.64-0.92, p=0.005) versus studies testing immunotherapy (HR 1.16, 95% CI 0.86-1.56, p=0.31). With regard to Comparison, a significant interaction (p<0.0001) was found in favor of RCT having placebo as control (HR 0.86, 95% CI 0.76-0.97, p=0.015), versus studies evaluating chemotherapy (HR 1.00, 95% CI 0.84, 1.19, p=0.99). A significant interaction according to DOC-treatment was also detected (p<0.0001), being the Post-DOC the Setting where a significant OS benefit was more likely to be determined (HR 0.77, 95% CI 0.66-0.90, p=0.001). Conclusions: The cross-trials interpretation in absence of formal direct comparisons may drive biased conclusions with regard to OS estimation. When designing trials to evaluate drugs (or strategies) in CRPC, the expected OS benefit must take into account the comparator, the treatment strategy and the (eventual) pre-treatment with DOC.

Download Full-text

A Systematic Review and Meta-Analysis of Seasonal Influenza Vaccination of Health Workers

Vaccines ◽

10.3390/vaccines9101104 ◽

2021 ◽

Vol 9 (10) ◽

pp. 1104

Author(s):

Tingting Li ◽

Xiaoling Qi ◽

Qin Li ◽

Wenge Tang ◽

Kun Su ◽

...

Keyword(s):

Systematic Review ◽

Influenza Vaccination ◽

Randomized Clinical Trials ◽

Health Workers ◽

Meta Analysis ◽

Random Effect ◽

Mean Difference ◽

Sensitivity Analyses ◽

Standardized Mean Difference ◽

Influenza Vaccinations

A systematic review and meta-analysis was conducted to estimate the pooled effect of influenza vaccinations for health workers (HWs). Nine databases were screened to identify randomized clinical trials and comparative observational studies that reported the effect of influenza vaccination among HWs. The risk ratio (RR), standardized mean difference, and 95% confidence interval (CI) were employed to study the effect size using fixed/random-effect models. Subgroup analyses and sensitivity analyses were conducted accordingly. Publication bias was examined. Sixteen studies (involving 7971 HWs from nine countries) were included after a comprehensive literature search. The combined RR regarding the incidence of laboratory-confirmed influenza was 0.36 (95% CI: 0.25 to 0.54), the incidence of influenza-like illness (ILI) was 0.69 (95% CI: 0.45 to 1.06), the absenteeism rate was 0.63 (95% CI: 0.46 to 0.86), and the integrated standardized mean difference of workdays lost was −0.18 (95% CI: −0.28 to −0.07) days/person. The subgroup analysis indicated that vaccination significantly decreases the incidence of laboratory-confirmed influenza in different countries, study populations, and average-age vaccinated groups. Influenza vaccinations could effectively reduce the incidence of laboratory-confirmed influenza, absenteeism rates, and workdays lost among HWs. It is advisable, therefore, to improve the coverage and increase the influenza vaccination count among HWs, which may benefit both workers and medical institutions.

Download Full-text

Adjuvant LH-RH agonists for premenopausal early breast cancer: A meta-analysis of randomized clinical trials (RCTs) exploring the magnitude of the disease-free and overall survival (DFS/OS) benefit

Journal of Clinical Oncology ◽

10.1200/jco.2007.25.18_suppl.592 ◽

2007 ◽

Vol 25 (18_suppl) ◽

pp. 592-592 ◽

Cited By ~ 1

Author(s):

P. Carlini ◽

E. Bria ◽

F. Cuppone ◽

P. Papaldo ◽

C. Nisticò ◽

...

Keyword(s):

Breast Cancer ◽

Early Breast Cancer ◽

Randomized Clinical Trials ◽

Meta Analysis ◽

Random Effect ◽

Breast Cancer Patients ◽

Ovarian Ablation ◽

Inverse Variance ◽

Lh Rh ◽

Event Based

592 Background: LH-RH agonists are considered as a milestone for adjuvant treatment for premenopausal early breast cancer patients (pts). All RCTs in which ovarian ablation (OA) with/without tamoxifen (TAM) and/or chemotherapy (CT) was compared with tamoxifen (TAM) and/or chemotherapy (CT) were pooled to estimate the magnitude of the benefit in both DFS and OS. Methods: A literature-based meta-analysis was accomplished, and event-based relative risk ratios (RRs) with 95% confidence interval (CI) were derived. A fixed- (FEM) and a random-effect (REM) model according to the inverse variance and heterogeneity (Het) test were applied as well. Absolute benefits (AB) and the number of pts needed to treat (NNT) were calculated. A sensitivity analysis to test for effect robustness in 4 sub-populations (OA/OS vs CT; OA/OS + CT vs CT; OA + TAM vs CT; OA + TAM vs observation) was accomplished. Results: Fifteen RCTs were gathered (12,545 pts); one RCT did not report the OS result. Results are depicted in the table . Conclusions: Although differences across RCTs exist in median follow-up time (as demonstrated by heterogeneity), OA seems to significantly improve DFS when combined with CT over CT alone, when combined with TAM versus CT alone and when combined with TAM versus observation. Actually, these DFS benefits do not translate into an OS benefit, with the exception of the last subgroup. [Table: see text] No significant financial relationships to disclose.

Download Full-text

Efficacy and safety of cyclosporine in the management of coronavirus disease 2019: A protocol for systematic review and meta-analysis

10.1101/2021.07.03.21259959 ◽

2021 ◽

Author(s):

Ibtihal Abdallah ◽

Mohamed Aabdien ◽

Mohammed Danjuma

Keyword(s):

Systematic Review ◽

Randomized Clinical Trials ◽

Meta Analysis ◽

Random Effect ◽

Primary Objective ◽

Clinical Severity ◽

World Health ◽

Cochrane Central Register ◽

Meta Analyses

Introduction Cyclosporine may improve the clinical course and outcomes of Coronavirus disease 2019 (COVID-19) due to its antiviral and anti-cytokine effects as shown in vitro. A few ongoing trials are exploring the benefit of adding it to the standard of care (SOC) of COVID-19 patients. Objectives The primary objective is to evaluate the severity of COVID-19, determined by oxygen saturation, intensive care unit (ICU) admission, or the World Health Organization COVID-19 clinical severity scale in patients treated with oral or intravenous cyclosporine added to SOC compared SOC alone or placebo. Secondary objectives include mortality, length of hospitalization, length of ICU stay, and laboratory measurements as well as the safety outcomes of cyclosporine. Methodology A systematic review and meta-analysis of randomized clinical trials and observational studies that compared cyclosporine to placebo or SOC in COVID-19 patients will be conducted. PubMed, EMBASE, Web of Science, Cochrane Central Register of Controlled Trials, Google Scholar, and ClinicalTrials.gov will be explored for studies that satisfy pre-specified inclusion criteria. Quality assessment of all included studies will be performed. Meta-analyses will be done utilizing random effect models to estimate the effect of cyclosporine on the severity of COVID-19. Heterogeneity will be assessed utilizing Q statistics. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines will be followed. Results The result of this synthesis will inform potential changes in the management of COVID-19 patients, especially regarding the role of calcineurin inhibitors. Additionally, it will serve as hypothesis generating for potential future prospective studies.

Download Full-text

The median and the mode as robust meta-analysis methods in the presence of small study effects

10.1101/288050 ◽

2018 ◽

Cited By ~ 1

Author(s):

Fernando Pires Hartwig ◽

George Davey Smith ◽

Amand Floriaan Schmidt ◽

Jack Bowden

Keyword(s):

Meta Analysis ◽

Small Study ◽

Sensitivity Analyses ◽

Literature Reviews ◽

Detailed Simulation ◽

The Mean ◽

Modelling Assumptions ◽

Meta Analyses ◽

Based Medicine

AbstractMeta-analyses based on systematic literature reviews are commonly used to obtain a quantitative summary of the available evidence on a given topic. Despite its attractive simplicity, and its established position at the summit of the evidence-based medicine hierarchy, the reliability of any meta-analysis is largely constrained by the quality of its constituent studies. One major limitation is small study effects, whose presence can often easily be detected, but not so easily adjusted for. Here, robust methods of estimation based on the median and mode are proposed as tools to increase the reliability of findings in a meta-analysis. By re-examining data from published meta-analyses, and by conducting a detailed simulation study, we show that these two simple methods offer notable robustness to a range of plausible bias mechanisms, without making any explicit modelling assumptions. In conclusion, when performing a meta-analysis with suspected small study effects, we recommend reporting the mean, median and modal pooled estimates as a simple but informative sensitivity analyses.

Download Full-text

Metformin prescription and aortic aneurysm: systematic review and meta-analysis

Heart ◽

10.1136/heartjnl-2018-314639 ◽

2019 ◽

Vol 105 (17) ◽

pp. 1351-1357 ◽

Cited By ~ 6

Author(s):

Xinyu Yu ◽

Dingsheng Jiang ◽

Jing Wang ◽

Rui Wang ◽

Taiqiang Chen ◽

...

Keyword(s):

Systematic Review ◽

Aortic Aneurysm ◽

Weighted Mean Difference ◽

Meta Analysis ◽

Random Effect ◽

Epidemiological Studies ◽

Sensitivity Analyses ◽

Newcastle Ottawa Scale ◽

Randomised Controlled ◽

Meta Analyses

ObjectiveTo assess the association of metformin prescription with the risk of aortic aneurysm, aortic aneurysm events and the enlargement of abdominal aortic aneurysm (AAA).DesignSystematic review and meta-analysis.MethodsWe searched PubMed, Embase and Scopus for epidemiological studies up to November 2018. We included observational studies which evaluated the association of metformin prescription with the risk of aortic aneurysm disease, and we also included studies involving progression and enlargement of AAA. The Newcastle-Ottawa Scale was used to assess the quality of included studies. Random-effect meta-analyses were conducted in line with the between-study heterogeneity. Sensitivity analyses were performed to identify the source of heterogeneity.ResultsEight studies enrolling 29 587 participants met the inclusion criteria and were included in this systematic review. We found that metformin prescription could significantly limit the enlargement of aortic aneurysm (weighted mean difference: −0.83 mm/year, 95% CI −1.38 to −0.28, I2=89.6%) among patients with AAA. Metformin prescription status may be associated with a decreased risk of aortic aneurysm and aortic aneurysm events.ConclusionsAccording to the available epidemiological evidence, metformin prescription could limit the expansion of AAA among patients with this disease, and may be involved with a lower incidence of aortic aneurysm and aortic aneurysm events. Randomised controlled trials are needed to confirm whether metformin could reduce the enlargement of AAA in patients with or without diabetes.

Download Full-text

What is Evidence: Weaknesses of Metaanalyses

European Psychiatry ◽

10.1016/s0924-9338(11)73905-0 ◽

2011 ◽

Vol 26 (S2) ◽

pp. 2202-2202

Author(s):

P. Czobor

Keyword(s):

Multiple Testing ◽

Meta Analysis ◽

Treatment Decisions ◽

Statistical Technique ◽

Evidence Based ◽

Systematic Biases ◽

The Individual ◽

Meta Analyses ◽

Random Fluctuations ◽

Based Medicine

Evidence-based treatment decisions depend on the accumulation of empirical data from individual studies. Individua studies use a variety of methods with variable quality, reflect random fluctuations and systematic biases, and may yield inconsistent findings. Meta-analysis offers a statistical approach to pool relevant studies together, which can reduce the effect of random error and bias, and produce more reliable effect estimates than individual studies. Despite the important contributions that this method provided for evidence-based medicine, it has been criticised as “statistical alchemy for the 21st century” and described as “new bete noir” which should be “stiffled at birth”. Some of this controversy has been driven by poor practice: the arbitrary pooling of dissimilar studies with unrelated outcomes, the application of statistical technique without sufficient expertise, and with insufficient attention to the clinical context. However, controversy also arises from methodological limitations of the meta-analytic approach, which need to be critically examined in order to evaluate the findings. In this presentation, we examine the limitations of meta-analyses both in terms of current procedures (e.g., biased selection, lack of prospective planning, increased likelihood of chance findings due to multiple testing) and methodological shortcomings (e.g., limited ability to handle multivariate outcomes or to incorporate covariates in the meta-analytic model simultaneously at the level of the study, treatment arm and the individual). By bringing together the weaknesses in a systematic way, it is hoped to foster a more reliable and critical appraisal of the empirical evidence both by researchers and clinicians, which will improve treatment decisions.

Download Full-text

The Orchard Plot: Cultivating a Forest Plot for Use in Ecology, Evolution and Beyond

10.32942/osf.io/epqa7 ◽

2019 ◽

Author(s):

Shinichi Nakagawa ◽

Malgorzata Lagisz ◽

Rose E O'Dea ◽

Joanna Rutkowska ◽

Yefeng Yang ◽

...

Keyword(s):

Meta Analysis ◽

R Package ◽

Effect Sizes ◽

Forest Plot ◽

Point Estimates ◽

Aggregate Effect ◽

The Individual ◽

Meta Analyses ◽

Heterogeneous Effect ◽

Intuitive Interpretation

‘Classic’ forest plots show the effect sizes from individual studies and the aggregate effect from a meta-analysis. However, in ecology and evolution meta-analyses routinely contain over 100 effect sizes, making the classic forest plot of limited use. We surveyed 102 meta-analyses in ecology and evolution, finding that only 11% use the classic forest plot. Instead, most used a ‘forest-like plot’, showing point estimates (with 95% confidence intervals; CIs) from a series of subgroups or categories in a meta-regression. We propose a modification of the forest-like plot, which we name the ‘orchard plot’. Orchard plots, in addition to showing overall mean effects and CIs from meta-analyses/regressions, also includes 95% prediction intervals (PIs), and the individual effect sizes scaled by their precision. The PI allows the user and reader to see the range in which an effect size from a future study may be expected to fall. The PI, therefore, provides an intuitive interpretation of any heterogeneity in the data. Supplementing the PI, the inclusion of underlying effect sizes also allows the user to see any influential or outlying effect sizes. We showcase the orchard plot with example datasets from ecology and evolution, using the R package, orchard, including several functions for visualizing meta-analytic data using forest-plot derivatives. We consider the orchard plot as a variant on the classic forest plot, cultivated to the needs of meta-analysts in ecology and evolution. Hopefully, the orchard plot will prove fruitful for visualizing large collections of heterogeneous effect sizes regardless of the field of study.

Download Full-text

Laser Influence on Dental Sensitivity Compared to Other Light Sources Used During In-office Dental Bleaching: Systematic Review and Meta-analysis

Operative Dentistry ◽

10.2341/19-064-l ◽

2020 ◽

Vol 45 (6) ◽

pp. 589-597

Author(s):

BGS Casado ◽

EP Pellizzer ◽

JR Souto Maior ◽

CAA Lemos ◽

BCE Vasconcelos ◽

...

Keyword(s):

Randomized Clinical Trials ◽

Color Change ◽

Meta Analysis ◽

Present Review ◽

Cochrane Library ◽

Light Sources ◽

Eligibility Criteria ◽

Tooth Color ◽

Significant Difference ◽

Meta Analyses

Clinical Relevance The use of laser light during bleaching will not reduce the incidence or severity of sensitivity and will not increase the degree of color change compared with nonlaser light sources. SUMMARY Objective: To evaluate whether the use of laser during in-office bleaching promotes a reduction in dental sensitivity after bleaching compared with other light sources. Methods: The present review was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) and is registered with PROSPERO (CDR42018096591). Searches were conducted in the PubMed/Medline, Web of Science, and Cochrane Library databases for relevant articles published up to August 2018. Only randomized clinical trials among adults that compared the use of laser during in-office whitening and other light sources were considered eligible. Results: After analysis of the texts retrieved during the database search, six articles met the eligibility criteria and were selected for the present review. For the outcome dental sensitivity, no significant difference was found favoring any type of light either for intensity (mean difference [MD]: −1.60; confidence interval [CI]: −3.42 to 0.22; p=0.09) or incidence (MD: 1.00; CI: 0.755 to 1.33; p=1.00). Regarding change in tooth color, no significant differences were found between the use of the laser and other light sources (MD: −2.22; CI: −6.36 to 1.93; p=0.29). Conclusions: Within the limitations of the present study, laser exerts no influence on tooth sensitivity compared with other light sources when used during in-office bleaching. The included studies demonstrated that laser use during in-office bleaching may have no influence on tooth color change.

Download Full-text

Systematic review and meta-analysis of the prevalence of depressive symptoms, dysthymia and major depressive disorders among homeless people

BMJ Open ◽

10.1136/bmjopen-2020-040061 ◽

2021 ◽

Vol 11 (2) ◽

pp. e040061

Author(s):

Getinet Ayano ◽

Asmare Belete ◽

Bereket Duko ◽

Light Tsegay ◽

Berihun Assefa Dachew

Keyword(s):

Systematic Review ◽

Depressive Symptoms ◽

Depressive Disorders ◽

Meta Analysis ◽

Random Effect ◽

Homeless People ◽

Sensitivity Analyses ◽

Prevalence Rates ◽

Major Depressive ◽

Major Depressive Disorders

ObjectivesTo assess the global prevalence estimates of depressive symptoms, dysthymia and major depressive disorders (MDDs) among homeless people.DesignSystematic review and meta-analysis.Data sourcesDatabases including PubMed, Scopus and Web of Science were systematically searched up to February 2020 to identify relevant studies that have reported data on the prevalence of depressive symptoms, dysthymia and MDDs among homeless people.Eligibility criteriaOriginal epidemiological studies written in English that addressed the prevalence of depressive problems among homeless people.Data extraction and synthesisA random-effect meta-analysis was performed to pool the prevalence estimated from individual studies. Subgroup and sensitivity analyses were employed to compare the prevalence across the groups as well as to identify the source of heterogeneities. The Joanna Briggs Institute’s quality assessment checklist was used to measure the study quality. Cochran’s Q and the I2 test were used to assess heterogeneity between the studies.ResultsForty publications, including 17 215 participants, were included in the final analysis. This meta-analysis demonstrated considerably higher prevalence rates of depressive symptoms 46.72% (95% CI 37.77% to 55.90%), dysthymia 8.25% (95% CI 4.79% to 11.86%), as well as MDDs 26.24% (95% CI 21.02% to 32.22%) among homeless people. Our subgroup analysis showed that the prevalence of depressive symptoms was high among younger homeless people (<25 years of age), whereas the prevalence of MDD was high among older homeless people (>50 years of age) when compared with adults (25–50 years).ConclusionThis review showed that nearly half, one-fourth and one-tenth of homeless people are suffering from depressive symptoms, dysthymia and MDDs, respectively, which are notably higher than the reported prevalence rates in the general population. The findings suggest the need for appropriate mental health prevention and treatment strategies for this population group.

Download Full-text