**5. Discussion**

*Sugar Intake - Risks and Benefits and the Global Diabetes Epidemic*

In addition, processing fluency mediated both the effect of ad framing (gain vs. neutral) and the effect of ad framing (loss vs. neutral) on people's behavioral intention: Processing fluency significantly affected people's behavioral intention, β = .21, *p* < .001. The indirect effect of ad framing (gain vs. neutral) on behavioral intention through processing fluency was significant (γ = −.03, *p* < .01). The indirect effect of ad framing (loss vs. neutral) on behavioral intention through processing fluency

*Final model (ad framing: Loss vs. neither gain nor loss). Note: 1.* χ*2 (18) = 32.872, p < .05; CFI = .995; TLI = .991; RMSEA = .027. 2. The model was evaluated by using loss framing as the reference group. 3. \* indicates p < .05, \*\* indicates p < .01, \*\*\* indicates p < .001. 4. Dotted line indicates the effect is not statistically significant at 95% level of confidence. 5. The indirect effect of loss vs. neither gain nor loss framing* 

*Final model (ad framing: Gain vs. neither gain nor loss). Note: 1.* χ*2 (18) = 32.872, p < .05; CFI = .995; TLI = .991; RMSEA = .027. 2. The model was evaluated by using gain framing as the reference group. 3. \* indicates p < .05, \*\* indicates p < .01, \*\*\* indicates p < .001. 4. Dotted line indicates the effect is not statistically significant at 95% level of confidence. 5. The indirect effect of gain vs. neither gain nor loss framing on behavioral intention through processing fluency is* −*.03, p < .01. 6. The indirect effect of regulatory focus* 

*on behavioral intention through processing fluency is* −*.04, p < .01.*

Based on the results, the statistical diagrams of the final model were presented

**88**

**Figure 4.**

**Figure 3.**

was also significant (γ = −.08, *p* < .001).

*on behavioral intention through processing fluency is* −*.08, p < .001.*

as follows (See **Figures 2**–**4**).

The purpose of this study was to investigate the effectiveness of advertising on people's intentions to control sugar intake. Specifically, three types of ad framing were examined: gain vs. loss vs. neither gain nor loss framing. Moreover, the moderator role of individuals' regulatory focus (promotion focus vs. prevention focus) on the effects of ad framing was explored. In addition, processing fluency was tested as a mediator.

By considering the influences of all tested variables in a whole SEM model, it was found that gain framing was more effective than loss framing in leading people to have greater intentions to limit sugar intake. The positive reaction toward gain-framed ads may be because people do not want to be told not to eat sugar: Many people may find it pleasant to consume sweets and foods with sugar, and limiting sugar intake is a prevention behavior that asks people to give up some kind of pleasure in order to pursue other desirable outcomes. Therefore, using a positive blueprint to persuade them to pursue desirable outcomes may be more effective than using negative illustrations to scare them into giving up their current pleasure.

The findings also showed that both gain and loss framing were more persuasive than the neutral framing in changing peoples' intentions to limit sugar intake. These findings suggest that ad framing (gain vs. loss vs. neither gain nor loss) matters in leading people to have greater intentions to adopt the recommended behavior of limiting sugar intake. Among the three types of ad framing, the neutral framing without emphasizing gain or loss cannot persuade people effectively to reduce sugar intake, while gain framing is the most effective framing to increase people's intention to eat less sugar.

Moreover, there was a significant interaction effect between ad framing (gain vs. loss) and regulatory focus on people's behavioral intentions. Gain framing was more effective in leading promotion-focused individuals to have greater intentions to limit sugar intake than loss framing, while loss framing was more effective in leading prevention-focused individuals to have greater behavioral intentions than gain framing. Based on regulatory focus theory [12], this result may indicate that gain-framed and loss-framed ads separately match people's promotion regulatory focus and prevention focus, which in turn create regulatory fit and lead people to feel right about the message. Therefore, regulatory focus is a moderator in the effects of ad framing (gain vs. loss) on behavioral intention.

Additionally, it was found that processing fluency mediated both the effects of ad framing (gain vs. neutral) and ad framing (loss vs. neutral) on people's behavioral intention to control sugar intake. That is, compared to the neutral-framed ads, both gain- and loss-framed ads were easier for participants to process, and then the greater processing fluency further led to greater advertising persuasiveness. This may be because people already have some knowledge or common sense about the negative consequences of high sugar intake or positive outcomes of controlling sugar intake; thus, compared to the neutral framing, they may process gain and loss framing more fluently with their existing knowledge, and then were better persuaded by the message.

This research provided several implications. The findings insinuate that to persuade the general population (i.e., without knowing their regulatory orientation) to control sugar intake, gain-framed advertising would be the best choice. Instead of always stressing the bad consequences of high sugar consumption to scare people, health professionals should design some positive-framed messages that stress the benefits of lowering sugar intake to stimulate people's stronger desires to control their sugar consumption. Moreover, neutral framing of neither gain nor loss is not

a good choice to persuade people to lower their sugar intake. However, if possible, it should be encouraged to find out people's regulatory orientation in order to better persuade them. For example, hospitals or other health organizations can ask obese patients or the patients with high blood sugar to fill out a questionnaire to know their regulatory focus; and then the health professionals can use different strategies tailored to different patients to help them control sugar intake. Specifically, gainframed messages could be used more often for promotion-focused people, while loss-framed messages should get the priority to be selected for prevention-focused people. In addition, since the findings showed that processing fluency can increase message persuasiveness, making the messages easy to process should be a way to better persuade people to lower sugar intake.
