EPPM stands for the Extended Parallel Process Model, a framework from health communication that explains why scary health messages sometimes motivate people to protect themselves and sometimes cause them to tune out entirely. Developed by Kim Witte in 1992, the model centers on a simple insight: when you encounter a frightening health message, your brain makes two quick assessments, and the balance between those assessments determines whether you take action or shut down.
The Two Assessments: Threat and Efficacy
The EPPM breaks your reaction to a fear-based health message into two mental evaluations that happen in sequence.
The first is perceived threat, which has two parts. Perceived susceptibility is how likely you believe you are to be affected by the threat. Perceived severity is how serious you believe the consequences would be. A message about lung cancer, for example, needs you to believe both that you personally could develop it and that lung cancer is genuinely devastating. If neither registers, you ignore the message altogether and move on with your day. The EPPM calls this a “no response” outcome.
The second evaluation is perceived efficacy, also with two parts. Response efficacy is whether you believe the recommended action actually works (quitting smoking reduces cancer risk). Self-efficacy is whether you believe you personally can do it (you feel capable of quitting). This second assessment only kicks in once the threat feels real enough to get your attention.
Danger Control vs. Fear Control
The core prediction of the EPPM is that the relationship between perceived threat and perceived efficacy determines one of two very different outcomes.
When both threat and efficacy are high, you enter what the model calls danger control. You feel genuinely at risk, but you also believe you can do something about it. In this state, you focus on the danger itself and take protective action: you get screened, change a behavior, follow a recommendation. Research on public health emergencies has found that perceived severity and susceptibility generate fear that can positively drive protective actions, while response efficacy and self-efficacy generate hope, which further supports those actions.
When threat is high but efficacy is low, you enter fear control. You feel scared, but you don’t believe you can do anything effective or that the recommended action will work. Instead of managing the danger, you manage the fear. The EPPM identifies several specific fear control responses: defensive avoidance (refusing to think about it), denial (convincing yourself the threat doesn’t apply to you), and reactance, which includes dismissing the message as exaggerated or feeling manipulated by whoever created it. This is the “backfire” that public health professionals worry about when using scare tactics.
What the Research Shows
A major meta-analysis published in Psychological Bulletin examined hundreds of fear appeal studies and found clear support for the EPPM’s central claim. Fear appeals that included efficacy statements, telling people what to do and assuring them they could do it, were significantly more effective than those that didn’t. Messages with efficacy statements had a pooled effect size of 0.43, compared to 0.21 for messages without them. That’s roughly double the impact.
Interestingly, the same analysis found that fear appeals without efficacy statements still worked to some degree. This challenges the strongest version of the EPPM’s prediction, which would suggest that high threat without high efficacy should always backfire. In practice, the relationship is more of a sliding scale: efficacy doesn’t flip a binary switch, but it substantially boosts how well a scary message translates into behavior change. The meta-analysis also found that fear appeals worked better when they depicted high susceptibility and severity, and when the recommended behavior was a one-time action rather than an ongoing habit.
How It Looks in Practice
A smoking prevention program for high school students illustrates how the EPPM translates into real campaign design. The program ran five sessions that deliberately targeted each of the model’s four components in sequence. Early sessions focused on raising threat: teaching students about the physiology of addiction, the health consequences of smoking, and the social and psychological damage it causes. A video showed the physical effects on the body. These sessions aimed to increase perceived severity and susceptibility.
Later sessions shifted to building efficacy. Students practiced identifying high-risk social situations where they might be pressured to smoke. They learned problem-solving skills and, through role-playing exercises, rehearsed how to say no and resist peer pressure. These activities targeted both response efficacy (these strategies work) and self-efficacy (I can actually pull this off). After the intervention, students in the program showed significantly increased perceived severity and response efficacy scores compared to a control group.
The same logic applied during COVID-19 communication efforts. Public health messaging that emphasized both the seriousness of the virus and practical steps people could take (masking, distancing, vaccination) aligned with EPPM principles. Research during the pandemic found that trust in healthcare systems played an important role in how these messages landed. People with higher institutional trust were more responsive to the efficacy components, while distrust amplified fear control responses like information avoidance.
Why Trust Changes the Equation
One of the more nuanced findings from pandemic-era EPPM research is that the model’s predictions don’t play out identically for everyone. The mechanisms connecting threat, efficacy, emotions, and behavior varied significantly depending on how much a person trusted the healthcare system. For people with low trust, even well-designed messages that balanced threat with efficacy could still trigger avoidance. Hope, which the research linked to efficacy perceptions, was positively associated with protective actions and negatively associated with information avoidance, but only when the source of the message was considered credible.
Limitations of the Model
The EPPM is one of the most widely used frameworks in health communication, but it has notable gaps. It doesn’t account for individual coping styles. Some people are naturally inclined toward active problem-solving while others default to avoidance, and these personality-level tendencies exist before any message arrives. The Transactional Model of Stress and Coping, a related framework, does incorporate coping style, but the EPPM treats all audience members as running through the same threat-then-efficacy evaluation.
Measurement has also been a persistent challenge. Researchers have found it difficult to separate the emotional experience of fear from the cognitive assessment of threat. In theory, fear is the feeling that arises after you evaluate a threat as serious and relevant. In practice, people who rate a threat as severe also report higher fear, making it hard to tease apart whether the emotion or the intellectual judgment is actually driving behavior. This blurriness makes it tricky to test the model’s predictions with precision, even though the broad pattern of “threat plus efficacy equals action” holds up consistently across studies.