ABSTRACT
After more than forty years of intensive research, the epidemiological impact of suicide coverage can now reasonably be considered confirmed. First, Philips’ initial findings have been repeatedly reproduced all around the world, extended to other media (Pirkis & Blood, 2010;
Sisask & Värnik, 2012; Notredame, In Press), and successfully assessed with more sophisticated statistical procedures (e.g., Chen, Chen, & Yip, 2011; Hagihara et al., 2014). Based on an extensive review of the so-called media effect studies, Pirkis and Blood (Pirkis & Blood, 2010) went a step further and argued that the robust correlation that links suicide coverage and suicide rates satisfies enough of the Hill’s criteria (Hill, 1965) to be considered causal. Finally, research has refined and characterized the epidemiological dimension of the WE by disclosing the main factors that modulate its magnitude: qualitative (story’s frame, writing style, images, etc.) (Niederkrotenthaler & Sonneck, 2007; Niederkrotenthaler et al., 2010; Blood et al., 2001) and quantitative (coverage extent) properties of the media items (Etzersdorfer, Voracek, & Sonneck, 2004; Pirkis et al., 2006; Suh, Chang, & Kim, 2015), nature of the covered event, reader/viewer’s predisposition (Chen et al., 2010a), and so on.
