Effects of Positive-Negative Message Framing on Health-Related Attitudes, Intentions, and Behaviors: A Meta-Analytic Review with particular focus on Health Behavioral Characteristics Moderators and Cultural-level Moderators

Health message framing refers to ways in which information about favorable consequences of adhering to health recommendations and unfavorable consequences without adhering to health recommendations can be presented (Cesario et al., 2013; Rothman et al., 2006; Joyal-Desmarais et al., 2022). Findings have been mixed and contradictory with hundreds of studies being conducted since the previous large-scale meta-analysis on health message framing (Gallagher & Updegraff, 2012). This study included three-level meta-analyses on message framing studies regarding health behavioral attitudes, behavioral intentions, and behaviors, with 1295 effect sizes, 430 studies from 66 countries. Comparing positive-framing and negative-framing, results indicated tentative evidence for very small positive-framed advantage for health promotion and/or illness prevention that involved encouragement of healthy behavior (g = 0.04) and support for very small negative-framed advantage (g = -0.08) for health detection. Furthermore, moderation of behavioral frequency was inconclusive. Comparing gain-framing and loss-framing, loss-framing outperformed gain-framing for people from collectivistic and survival-oriented cultures/samples (gs = -0.26 to -0.29), with tentative evidence for gain-framing advantage for individualistic cultures (g = 0.07). Findings showed inconclusive evidence for moderation of indulgence-restraint, power distance, and dialecticism. We also found support for gain-framing advantage for health recuperation behavioral intentions (g = 0.20). Implications and directions, including culturally-congruent health communication (Betsch et al., 2016), more analyses and research on real-world benefits, costs, and risks of message framing choices, and realistic effect size expectations for sample size determination, were discussed. We called for more multi-regional studies, experiments that tease apart confounds, and studies examining mechanisms. Keywords: message framing, health behavior, behavioral function, cultural moderation, meta-analysis