Meta-analysis of individual patient data versus aggregate data from longitudinal clinical trials

Abstract

Background: In clinical trials following individuals over a period of time, the same assessment may be made at a number of time points during the course of the trial. Our review of current practice for handling longitudinal data in Cochrane systematic reviews shows that the most frequently used approach is to ignore the correlation between repeated observations and to conduct separate meta-analyses at each of a number of time points.

Purpose: The purpose of this paper is to show the link between repeated measurement models used with aggregate data and those used when individual patient data (IPD) are available, and provide guidance on the methods that practitioners might use for aggregate data meta-analyses, depending on the type of data available.

Methods: We discuss models for the meta-analysis of longitudinal continuous outcome data when IPD are available. In these models time is included either as a factor or as a continuous variable, and account is taken of the correlation between repeated observations. The meta-analysis of IPD can be conducted using either a one-step or a two-step approach: the latter involves analysing the IPD separately in each study and then combining the study estimates taking into account their covariance structure. We discuss the link between models for use with aggregate data and the two-step IPD approach, and the problems which arise when only aggregate data are available. The methods are applied to IPD from 5 trials in Alzheimer's disease.

Results: Two major issues for the meta-analysis of aggregate data are the lack of information about correlation coefficients and the effect of missing data at the patient-level. Application to the Alzheimer's disease data set shows that ignoring correlation can lead to different pooled estimates of the treatment difference and their standard errors. Furthermore, the amount of missing data at the patient level can affect these estimates.

Limitations: The models assume fixed treatment effects across studies, and that any missing data is missing at random, both at the patient-level and the study level.

Conclusions: It is preferable to obtain IPD from all studies to correctly account for the correlation between repeated observations. When IPD are not available, the ideal aggregate data are model-based estimates of treatment difference and their variance and covariance estimates. If covariance estimates are not available, sensitivity analyses should be undertaken to investigate the robustness of the results to different amounts of correlation.