How Does Anticipated Regret Influence Health and Safety Decisions? A Literature Review

Abstract

Making important decisions about health and safety often involves contemplating the regret that may arise if a decision turns out badly. Does anticipated regret decrease risky behaviors and promote precautionary behaviors? The present article reviews the role of anticipated regret in health and safety decisions by outlining findings from the most commonly researched topics. In line with an extended theory of planned behavior, the review concludes that anticipated regret generally predicts behavioral intentions, and intentions in turn influence health and safety behavior. The review discusses methodological recommendations and implications for interventions, and identifies unanswered questions and directions for future research.

ACKNOWLEDGMENTS

I thank Erin Austen, Nancy Frye, James Shepperd, Shawnee Totton, and Angela Weaver for their comments on a previous version of this article.

Notes

¹One prospective study indicated that AR did not uniquely predict intentions to sign a donor card, but unexpectedly, AR moderated the intention–behavior relationship, such that the intention–behavior relationship was weakest among participants with the highest AR (Godin, Bélanger-Gravel, Gagné, & Blondeau, 2008). Participants with relatively low intentions drove this effect. To date, though, no other published study has replicated this complex and unanticipated effect.

²In a scenario study designed to assess speeding intentions (Conner, Smith, & McMillan, 2003), traditional TPB items emerged as significant predictors of speeding intentions, but anticipated affect did not; however, ambiguity regarding the affect measure prevents strong conclusions. The anticipated affect measure contained two items that explicitly mentioned “regret,” and these were combined with a measure of feeling “exhilarated.” Because there is no report of the resulting Cronbach's alpha, one cannot determine whether combining these measures was appropriate.

³The only other known study on AR in the context of road safety examined seat belt use in Turkey (Şimşekoğlu & Lajunen, 2008). Adding AR to TPB variables did not significantly improve the prediction of seat belt intentions. However, the explanation for this null effect may be a methodological one: AR was assessed with a one-item measure (which used “sorry” rather than “regret”), whereas all other predictors were assessed with more than one item.

⁴A recently published study investigated the potential role of AR for concealing texting while driving (Gauld, Lewis, & White, 2014). Although participants responded to questions involving both concealed and “obvious” (i.e., nonconcealed) texting while driving, because the authors were interested in predictors of concealed texting, they did not present results for “obvious” texting. Thus, the possibility that AR for texting may be associated with weak intentions to text (without concealment) while driving still awaits empirical investigation.