A GRAPHICAL DISPLAY USEFUL FOR METAANALYSIS

Graphical methods are frequently used in meta-analysis to summarize th eir results and to explore potential sources of heterogeneity across s tudies, In this paper, we illustrate a graphical method for meta-analy sis of studies with dichotomous exposures and outcomes that complement s other graphical and analytical approaches to meta-analysis, In prosp ective studies, the proportion of cases among the unexposed is plotted on the horizontal axis versus the proportion of cases among the expos ed on the vertical axis. Contour lines for equal values of relative ri sk, odds ratio or risk difference and for the combined estimate of eft :ect and its confidence interval are then superimposed on the graph, I n case-control studies, the proportion of exposed controls is plotted on the horizontal axis versus the proportion of exposed cases on the v ertical axis, although only the contour lines of equal odds ratios yie ld direct epidemiological interpretation, In these graphs, the distrib ution of the individual estimates of effect with respect to the contou r lines offers a clue as to the adequacy of the scale of measurement u sed (additive or multiplicative). This graphical method also permits d irect inspection of the range of disease frequency in follow-up studie s and of the range of exposure in case-control studies, Its use is ill ustrated with the aid of 3 examples derived from the literature.