An IPD meta-analysis of whether microvessel density is a prognostic factor for death in patients with non-metastatic surgically treated non-small-cell lung carcinoma, as undertaken by Trivella et al.. The forest plot shows the individual study hazard ratio estimates (with confidence intervals), which indicate the association between risk of death and an increase of ten microvessel counts, as assessed by measurement of all vessels. A random-effects meta-analysis was used to combine estimates (I2 = 73.7%), and the overall hazard ratio shown is thus the estimated average of all the underlying hazard ratios across studies.
Abo-Zaid et al. BMC Medical Research Methodology 2012 12:56 doi:10.1186/1471-2288-12-56