The aim of the study is to systematically evaluate the impact of message framing on persuasive communication in an applicative rather than hypothetical setting. The existing literature shows that the framing effect depends on different factors, such as level of involvement and valence of the consequences produced by adopting or not adopting a risky behaviours. The 129 participants had to judge several characteristics of messages describing the negative consequences of smoking or excessive drinking and the positive consequences of non-smoking or moderate drinking. The positive consequences were framed either as gains or avoided losses and the negative consequences were framed either as losses or as foregone gains. Results show that loss messages, followed by gain messages, are evaluated better and processed more easily than no loss and no gain messages. Low consumption participants prefer messages describing negative consequences, while high consumption participants - who are very likely more involved - do not evaluate in a significantly different way messages underlining the negative consequences of persisting with the risky behaviour compared to messages underlining the positive consequences of stopping the behaviour.