Dataset Information

Assessing the consistency assumptions underlying network meta-regression using aggregate data.

ABSTRACT: When numerous treatments exist for a disease (Treatments 1, 2, 3, etc), network meta-regression (NMR) examines whether each relative treatment effect (eg, mean difference for 2 vs 1, 3 vs 1, and 3 vs 2) differs according to a covariate (eg, disease severity). Two consistency assumptions underlie NMR: consistency of the treatment effects at the covariate value 0 and consistency of the regression coefficients for the treatment by covariate interaction. The NMR results may be unreliable when the assumptions do not hold. Furthermore, interactions may exist but are not found because inconsistency of the coefficients is masking them, for example, when the treatment effect increases as the covariate increases using direct evidence but the effect decreases with the increasing covariate using indirect evidence. We outline existing NMR models that incorporate different types of treatment by covariate interaction. We then introduce models that can be used to assess the consistency assumptions underlying NMR for aggregate data. We extend existing node-splitting models, the unrelated mean effects inconsistency model, and the design by treatment inconsistency model to incorporate covariate interactions. We propose models for assessing both consistency assumptions simultaneously and models for assessing each of the assumptions in turn to gain a more thorough understanding of consistency. We apply the methods in a Bayesian framework to trial-level data comparing antimalarial treatments using the covariate average age and to four fabricated data sets to demonstrate key scenarios. We discuss the pros and cons of the methods and important considerations when applying models to aggregated data.

SUBMITTER: Donegan S

PROVIDER: S-EPMC6563470 | biostudies-literature | 2019 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Assessing the consistency assumptions underlying network meta-regression using aggregate data.

Donegan Sarah S Dias Sofia S Welton Nicky J NJ

Research synthesis methods 20181112 2

When numerous treatments exist for a disease (Treatments 1, 2, 3, etc), network meta-regression (NMR) examines whether each relative treatment effect (eg, mean difference for 2 vs 1, 3 vs 1, and 3 vs 2) differs according to a covariate (eg, disease severity). Two consistency assumptions underlie NMR: consistency of the treatment effects at the covariate value 0 and consistency of the regression coefficients for the treatment by covariate interaction. The NMR results may be unreliable when the as ...[more]

PMID: 30367548

Similar Datasets

Project description:BackgroundThe applicability of randomised controlled trials of pharmacological agents to older people with frailty/multimorbidity is often uncertain, due to concerns that trials are not representative. However, assessing trial representativeness is challenging and complex. We explore an approach assessing trial representativeness by comparing rates of trial serious adverse events (SAE) to rates of hospitalisation/death in routine care.MethodsThis was an observational analysis of individual (125 trials, n=122,069) and aggregate-level drug trial data (483 trials, n=636,267) for 21 index conditions compared to population-based routine healthcare data (routine care). Trials were identified from ClinicalTrials.gov . Routine care comparison from linked primary care and hospital data from Wales, UK (n=2.3M). Our outcome of interest was SAEs (routinely reported in trials). In routine care, SAEs were based on hospitalisations and deaths (which are SAEs by definition). We compared trial SAEs in trials to expected SAEs based on age/sex standardised routine care populations with the same index condition. Using IPD, we assessed the relationship between multimorbidity count and SAEs in both trials and routine care and assessed the impact on the observed/expected SAE ratio additionally accounting for multimorbidity.ResultsFor 12/21 index conditions, the pooled observed/expected SAE ratio was <1, indicating fewer SAEs in trial participants than in routine care. A further 6/21 had point estimates <1 but the 95% CI included the null. The median pooled estimate of observed/expected SAE ratio was 0.60 (95% CI 0.55-0.64; COPD) and the interquartile range was 0.44 (0.34-0.55; Parkinson's disease) to 0.87 (0.58-1.29; inflammatory bowel disease). Higher multimorbidity count was associated with SAEs across all index conditions in both routine care and trials. For most trials, the observed/expected SAE ratio moved closer to 1 after additionally accounting for multimorbidity count, but it nonetheless remained below 1 for most.ConclusionsTrial participants experience fewer SAEs than expected based on age/sex/condition hospitalisation and death rates in routine care, confirming the predicted lack of representativeness. This difference is only partially explained by differences in multimorbidity. Assessing observed/expected SAE may help assess the applicability of trial findings to older populations in whom multimorbidity and frailty are common.

Project description:BackgroundIn a star-shaped network, pairwise comparisons link treatments with a reference treatment (often placebo or standard care), but not with each other. Thus, comparisons between non-reference treatments rely on indirect evidence, and are based on the unidentifiable consistency assumption, limiting the reliability of the results. We suggest a method of performing a sensitivity analysis through data imputation to assess the robustness of results with an unknown degree of inconsistency.MethodsThe method involves imputation of data for randomized controlled trials comparing non-reference treatments, to produce a complete network. The imputed data simulate a situation that would allow mixed treatment comparison, with a statistically acceptable extent of inconsistency. By comparing the agreement between the results obtained from the original star-shaped network meta-analysis and the results after incorporating the imputed data, the robustness of the results of the original star-shaped network meta-analysis can be quantified and assessed. To illustrate this method, we applied it to two real datasets and some simulated datasets.ResultsApplying the method to the star-shaped network formed by discarding all comparisons between non-reference treatments from a real complete network, 33% of the results from the analysis incorporating imputed data under acceptable inconsistency indicated that the treatment ranking would be different from the ranking obtained from the star-shaped network. Through a simulation study, we demonstrated the sensitivity of the results after data imputation for a star-shaped network with different levels of within- and between-study variability. An extended usability of the method was also demonstrated by another example where some head-to-head comparisons were incorporated.ConclusionsOur method will serve as a practical technique to assess the reliability of results from a star-shaped network meta-analysis under the unverifiable consistency assumption.

Dataset Information

Assessing the consistency assumptions underlying network meta-regression using aggregate data.

Publications

Assessing the consistency assumptions underlying network meta-regression using aggregate data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets