Response to Letter
We thank our colleague for his critique. It confirms the need to further explore risk perception of travelers, a topic which has been given consideration by the International Society of Travel Medicine.[1-4] The main goal of our study was to make travel health experts aware of differences in risk perception and to encourage more research. We agree that PRISM, an easily applicable tool, needs to be further validated for risk perception research. A number of methods are available, including risk scales and a variety of questionnaires addressing different aspects of risk perception. As risk perception strongly influences behavior which finally determines the risks, the ideal method to measure people's risk perception, and eventually to validate other methods, should be consistent with their (changed) behavior.
As our priority was to discuss our findings in the context of travel medicine research, integrating concepts of risk perception research would have gone beyond the scope of our study. However, psychological mechanisms influencing risk perception, including both cognitive factors such as the perceived likelihood, severity and susceptibility or the availability heuristic, and emotional factors such as the affect heuristic,[6, 7] are doubtlessly most important to understand risk perception and develop risk conversation strategies. For instance, optimism or optimism bias, an underestimation of likelihood mentioned by our colleague, most likely influenced the travelers' risk perception of STIs and other risks.
Upon cursory comparison, some of our results differ from findings of risk perception research, for example factor-analytic representations, a method of the psychometric paradigm used by our colleague to adjust Figure 3. Factor-analytic representations are three-dimensional frameworks for hazard characteristics. Two axes, the “dread” axis and the “unknown” axis, each represent a set of correlating characteristics while a third axis reflects the number of exposed people. Dread was correlated highest with risk perception. Road traffic accidents, for instance, were characterized as well-known and medium-dreaded[7, 9] or underestimated in terms of personal mortality whereas accidents were perceived as relatively high risk in our study. However, the perception of risks is not static and depends, among other factors, on study population demographics, voluntariness of exposure, media coverage,[6, 7, 11] and on the context. Many studies explore risk perception of specific health hazards in general[6, 7, 9] or in familiar surroundings.[10, 12] Leisure travel is usually voluntary, time-limited, and often involves visiting unfamiliar places. In the context of travel, dreaded or familiar risks might not be the ones our colleague claims them to be. Moreover, one could argue that some findings of our study and their interpretation were, in fact, consistent with concepts of risk perception research,[6-8, 11] such as the underestimation of the risk of STIs, the possibility of media influence on the higher risk perception of terrorist attacks in Asia, the change of perceived risk of accidents after travel, possibly due to a change in availability, and the overall consistency of the experts' risk perception with epidemiological data.
Nevertheless, PRISM or other tools need to be validated, and concepts of psychological and sociological risk perception research need to be integrated in further studies on risk perception in travel medicine. We thus welcome the fact that Dr Zimmer has further encouraged the required discussion of this issue.