Assessing key assumptions of network meta‐analysis: a review of methods

Donegan, Sarah; Williamson, Paula; D’Alessandro, Umberto; Smith, Catrin Tudur

doi:10.1002/jrsm.1085

Cited by 195 publications

(158 citation statements)

References 114 publications

(284 reference statements)

Supporting

Mentioning

158

Contrasting

Order By: Relevance

“…Homogeneity and consistency assumptions underlie NMA25. Although assessment of heterogeneity in traditional meta-analyses was common, only 4 NMAs (3.92%) assessed the heterogeneity in the entire network by heterogeneity variance parameter (Tau 2 ).…”

Section: Discussionmentioning

confidence: 99%

Epidemiology Characteristics, Methodological Assessment and Reporting of Statistical Analysis of Network Meta-Analyses in the Field of Cancer

Tian

et al. 2016

Sci Rep

View full text Add to dashboard Cite

Because of the methodological complexity of network meta-analyses (NMAs), NMAs may be more vulnerable to methodological risks than conventional pair-wise meta-analysis. Our study aims to investigate epidemiology characteristics, conduction of literature search, methodological quality and reporting of statistical analysis process in the field of cancer based on PRISMA extension statement and modified AMSTAR checklist. We identified and included 102 NMAs in the field of cancer. 61 NMAs were conducted using a Bayesian framework. Of them, more than half of NMAs did not report assessment of convergence (60.66%). Inconsistency was assessed in 27.87% of NMAs. Assessment of heterogeneity in traditional meta-analyses was more common (42.62%) than in NMAs (6.56%). Most of NMAs did not report assessment of similarity (86.89%) and did not used GRADE tool to assess quality of evidence (95.08%). 43 NMAs were adjusted indirect comparisons, the methods used were described in 53.49% NMAs. Only 4.65% NMAs described the details of handling of multi group trials and 6.98% described the methods of similarity assessment. The median total AMSTAR-score was 8.00 (IQR: 6.00–8.25). Methodological quality and reporting of statistical analysis did not substantially differ by selected general characteristics. Overall, the quality of NMAs in the field of cancer was generally acceptable.

show abstract

Section: Discussionmentioning

confidence: 99%

Epidemiology Characteristics, Methodological Assessment and Reporting of Statistical Analysis of Network Meta-Analyses in the Field of Cancer

Tian

et al. 2016

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Note that IF is the logarithm of the ratio of two odds ratios ( RoR ) from direct and indirect evidence in the loop. RoR values truncate at one would also indicate consistency [62]. A forest plot of the estimated summary effects, along with confidence intervals and corresponding predictive intervals (PrI) for all comparisons, summarizes the relative mean effects, the impact of heterogeneity and predictions on each comparison in one plot [63].…”

Section: Methodsmentioning

confidence: 99%

Induction treatments for acute promyelocytic leukemia: a network meta-analysis

et al. 2016

View full text Add to dashboard Cite

Background9 treatments for acute promyelocytic leukemia (APL) have been compared in many randomized controlled trials (RCT). The conclusions have been inconsistent and the purpose of this study is to conduct a network meta-analysis.ResultsRankings of event-free survival are ATRA+RIF (81.2%), ATRA+ATO (69.6%), ATO (50.6%). Rankings of complete remission are ATRA+RIF (79.3%), ATRA+ATO (64.8%), RIF (60.3%), ATO (55.9%). Rankings of avoiding differentiation syndromes are CT (84.3%), ATO (80.3%), RIF (71.6%), ATRA+RIF (49%), ATRA+ATO (40.8%).MethodsA total of 1,666 patients from 12 RCTs were enrolled. The frequentist method was used. Relative risks with 95% confidence intervals were calculated. We produced a network plot, a contribution plot, and a forest plot predictive intervals. The inconsistency factor, the surface under the cumulative ranking curve and the publication bias were evaluated.ConclusionsATRA+ATO is eligible to be the first-line treatment for APL. ATRA+RIF is a prospective alternative to the first-line treatment. RIF or ATO should be reconsidered as another option for de novo APL.

show abstract

“…Next, network meta-analysis was performed by using the GEMTC (V0.6) package and the corresponding results were combined with those obtained from the DerSimonian-Laird random-effects model and Monte Carlo Markov Chain (MCMC). We used R software (V3.2.1) in order to produce the surface under the cumulative ranking curve (SUCRA) and calculated the ranking of different interventions [53]. For each intervention, efficacy and safety outcomes were ranked by the SUCRA: a higher value of SUCRA indicated a higher ranking.…”

Section: Methodsmentioning

confidence: 99%

A mixed analysis comparing nine minimally invasive surgeries for unresectable hepatocellular carcinoma patients

Tao

Ran

et al. 2016

Oncotarget

View full text Add to dashboard Cite

Hepatocellular carcinoma (HCC) is usually managed by the transcatheter arterial chemoembolization (TACE). However, this technique has been challenged since severe complications have been observed in clinical practices. As a result, clinicians have started to seek other minimally invasive surgeries with equivalent efficacy. The corresponding surgeries were assessed by the five outcomes: complete response (CR), partial response (PR), stable disease (SD), progression disease (PD) and objective response rate (ORR). Direct meta-analysis and network meta-analysis were performed and the results were represented by odds ratios (OR), 95% confidence and credential intervals. Furthermore, the value of surface under the cumulative ranking curve (SUCRA)was calculated to provide corresponding rankings.Seventeen studies were incorporated into the network meta-analysis which indicated that TACE + external-beam radiation therapy (EBRT) and drug-eluting beads (DEB) were better than TACE at controllingPD. TACE + EBRT demonstrated their advantages compared to TARE-90Y.However, network meta-analysis comparison showed no significant difference between the corresponding eight treatments with respect to CR, PR, SD and ORR. Moreover, the SUCRA suggested that TACE+EBRT were better than other treatments at treating unresectableHCC.Based on the present results of this network meta-analysis, TACE + EBRT was more effective than the other seven minimally invasive surgeries and therefore it is considered as the optimal treatment for HCC.

show abstract

Assessing key assumptions of network meta‐analysis: a review of methods

Abstract: Presently, we advocate applying existing assessment methods collectively to gain the best understanding possible regarding whether assumptions are reasonable. In our example, consistency was questionable; therefore the NMA results may be unreliable.

Cited by 195 publications

References 114 publications

Epidemiology Characteristics, Methodological Assessment and Reporting of Statistical Analysis of Network Meta-Analyses in the Field of Cancer

Epidemiology Characteristics, Methodological Assessment and Reporting of Statistical Analysis of Network Meta-Analyses in the Field of Cancer

Induction treatments for acute promyelocytic leukemia: a network meta-analysis

A mixed analysis comparing nine minimally invasive surgeries for unresectable hepatocellular carcinoma patients

Contact Info

Product

Resources

About