Severity, efficacy, and evidence type as determinants of health message exposure.

This cross-cultural experiment examined the effectiveness of three health message characteristics to foster or inhibit selective exposure to health information. An online magazine was created with eight articles about various health risks. Four articles were manipulated regarding (1) severity of the described health threat (low versus high), (2) suggested efficacy to avoid or minimize negative consequences (low versus high) and (3) type of evidence presented (statistical information versus exemplar information). Respondents from the U.S. and from Germany (n = 301/298) browsed through the magazine while selective exposure was unobtrusively logged. Findings reveal country-specific exposure patterns. A positive main effect of severity was only found for U.S. respondents. Independent of respondents' country, significantly more time was spent with low-severity/high-efficacy messages and high-severity/low-efficacy messages than with articles featuring the often-recommended high-severity/high-efficacy message combination. Respondents generally read more exemplar messages than those with statistical evidence, especially when high efficacy was suggested. Implications of these exposure patterns for the real-life effectiveness of health messages are discussed and an improved theoretical conceptualization of message effectiveness is proposed.