When describing the severity of illnesses, terms such as “mild”, “severe”, or “critical” are commonly used in medical and public health communication. During the COVID-19 pandemic, these terms served as labels to classify potential health risks and communicate them to the general public. However, severity labels might not always be accurately interpreted. Individuals encountering the term “mild” could, e.g., underestimate the potential severity of the disease, reducing the likelihood of adopting protective measures. Drawing inspiration from meteorological risk communication, we employ an "impact-based" approach. This method assumes that providing information about which symptoms are classified, e.g., as “mild” or “severe”, could lead to a more accurate assessment of the (potential) threat. In an online study, we examine how people generally interpret disease severity labels. Participants match a variety of symptoms typically associated with respiratory diseases to what they believe is the appropriate severity category. Moreover, in a 2x2 experimental design, we investigate how providing a list of possible symptoms in addition to the severity label vs. presenting the label only (factor information) influences perceived risk of a disease and the likelihood to take up protective measures against it. Furthermore, we explore how previous knowledge and susceptibility beliefs about diseases influence the effects by comparing a condition in which we describe the disease as a “Respiratory Disease” to a neutral condition with a fictious disease called „Avirconie“ (factor disease).