Network meta-analysis of (individual patient) time to event data alongside (aggregate) count data.

Pedro Saramago,Ling-Hsiang Chuang,Marta O Soares

doi:10.1186/1471-2288-14-105

Pedro Saramago, Ling-Hsiang Chuang + Show 1 more

Open Access

https://doi.org/10.1186/1471-2288-14-105

Copy DOI

Abstract

BackgroundNetwork meta-analysis methods extend the standard pair-wise framework to allow simultaneous comparison of multiple interventions in a single statistical model. Despite published work on network meta-analysis mainly focussing on the synthesis of aggregate data, methods have been developed that allow the use of individual patient-level data specifically when outcomes are dichotomous or continuous. This paper focuses on the synthesis of individual patient-level and summary time to event data, motivated by a real data example looking at the effectiveness of high compression treatments on the healing of venous leg ulcers.MethodsThis paper introduces a novel network meta-analysis modelling approach that allows individual patient-level (time to event with censoring) and summary-level data (event count for a given follow-up time) to be synthesised jointly by assuming an underlying, common, distribution of time to healing. Alternative model assumptions were tested within the motivating example. Model fit and adequacy measures were used to compare and select models.ResultsDue to the availability of individual patient-level data in our example we were able to use a Weibull distribution to describe time to healing; otherwise, we would have been limited to specifying a uniparametric distribution. Absolute effectiveness estimates were more sensitive than relative effectiveness estimates to a range of alternative specifications for the model.ConclusionsThe synthesis of time to event data considering individual patient-level data provides modelling flexibility, and can be particularly important when absolute effectiveness estimates, and not just relative effect estimates, are of interest.

Highlights

Network meta-analysis methods extend the standard pair-wise framework to allow simultaneous comparison of multiple interventions in a single statistical model
Statistical model for the data We describe model A in two interrelated parts: part I describes the modelling of the individual patient-level data (IPD) and part II the modelling of the aggregate data (AD)
This work was motivated by a real data example looking at the effectiveness of high compression treatments on the healing of venous leg ulcers

Summary

Introduction

Network meta-analysis methods extend the standard pair-wise framework to allow simultaneous comparison of multiple interventions in a single statistical model. Despite published work on network meta-analysis mainly focussing on the synthesis of aggregate data, methods have been developed that allow the use of individual patient-level data when outcomes are dichotomous or continuous. A piecemeal approach can be avoided if all RCTs evaluating interventions relevant to the treatment decision are considered collectively, for NMA using individual patient data Published work on NMA mainly focuses on the synthesis of aggregate data (AD) (sometimes called summary data, e.g. group means and standard errors available from study reports) [4,5]; methods have been developed that allow use of individual patient-level data (IPD) in NMA, for binary and continuous data [6,7,8]. IPD is likely to produce more precise estimates of treatment effects, even in the absence of treatment–covariate interactions [10]

Objectives

Methods

Results

Discussion

Conclusion