Multivariate meta‐analysis of prognostic factor studies with multiple cut‐points and/or methods of measurement

Abstract

A prognostic factor is any measure that is associated with the risk of future health outcomes in those with existing disease. Often, the prognostic ability of a factor is evaluated in multiple studies. However, meta‐analysis is difficult because primary studies often use different methods of measurement and/or different cut‐points to dichotomise continuous factors into ‘high’ and ‘low’ groups; selective reporting is also common. We illustrate how multivariate random effects meta‐analysis models can accommodate multiple prognostic effect estimates from the same study, relating to multiple cut‐points and/or methods of measurement. The models account for within‐study and between‐study correlations, which utilises more information and reduces the impact of unreported cut‐points and/or measurement methods in some studies. The applicability of the approach is improved with individual participant data and by assuming a functional relationship between prognostic effect and cut‐point to reduce the number of unknown parameters. The models provide important inferential results for each cut‐point and method of measurement, including the summary prognostic effect, the between‐study variance and a 95% prediction interval for the prognostic effect in new populations. Two applications are presented. The first reveals that, in a multivariate meta‐analysis using published results, the Apgar score is prognostic of neonatal mortality but effect sizes are smaller at most cut‐points than previously thought. In the second, a multivariate meta‐analysis of two methods of measurement provides weak evidence that microvessel density is prognostic of mortality in lung cancer, even when individual participant data are available so that a continuous prognostic trend is examined (rather than cut‐points). © 2015 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd.

A guide to systematic review and meta-analysis of prognostic factor studies

“…Advanced multivariate meta-analysis methods are also available to handle multiple cutpoints,59 multiple methods of measurement,59 or different adjustment factors in prognostic factor studies 60. An introduction to multivariate meta-analysis has been published in The BMJ 61…”

Prognostic Value of microRNA-133a Expression and Its Clinicopathologic Significance in Non-Small Cell Lung Cancer: A Comprehensive Study Based on Meta-Analysis and the TCGA Database

“…A HR of less than 1 with the 95% CI not overlapping 1 indicates a better prognosis for the miR-133a high-expression group. Considering the many sources of heterogeneity among the studies and consequently among their individual HRs, we calculated an overall HR with a random effects model on all the survival data [29]. Odds ratios (ORs) with 95% CI were used to estimate clinicopathologic parameter data, and heterogeneity was tested using Q-statistic and I 2 -statistic.…”

Time-dependent summary receiver operating characteristics for meta-analysis of prognostic studies

“…However, it is difficult to obtain IPD for all the studies and then it may lead to publication bias and availability bias . For this reason, a literature‐based meta‐analysis—that is, a meta‐analysis based on reported summary statistics—is also important for prognostic studies . However, statistical methods for meta‐analysis of prognostic studies have been less developed .…”