This paper was presented at the 12th Conference of the International Society of Travel Medicine in Boston, MA, USA, May 8–12, 2011.
Corresponding Author: Rosalie Zimmermann, MD, Swiss Tropical and Public Health Institute, Medical and Diagnostic Services, Socinstrasse 57, CH-4002 Basel, Switzerland. E-mail: email@example.com
This study assessed the risk perception ratings of travelers pre- and post-travel and in comparison to the ratings by travel health experts. While most surveys on travel health knowledge, attitudes, and practices focus on malaria and vaccine-preventable diseases, noninfectious travel risks were included in this study.
Pre- and post-travel perception of nine travel-associated health risks was recorded among 314 travelers to tropical and subtropical destinations. All travelers sought pre-travel health advice at the Travel Clinic of the Swiss Tropical and Public Health Institute in 2008 and 2009. In addition, 18 Swiss travel health experts provided an assessment of the respective risks. A validated visual psychometric measuring instrument was used [pictorial representation of illness and self measure (PRISM)].
Travelers and experts rated most risks similarly, except for accidents and sexually transmitted infections (STIs) which experts rated higher. Compared to other risks, accidents ranked highly in both groups and were the only risk perceived higher after travel. Pre- and post-travel perceptions of all other risks were similar with a tendency to be lower after travel. Travelers perceived mosquitoes to be the highest risk before travel and accidents after travel.
Travelers' risk perception appears to be accurate for most risks stated in this study. However, travel health professionals should be aware that some perception patterns among travelers regarding travel-related health risks may be different from professional risk assessment. Important but insufficiently perceived health risks, such as sexual behavior/STIs and accidents, should be considered to be part of any pre-travel health advice package.
Having reached 980 million in 2011, international tourist arrivals are expected to continue growing. Tourist industries are growing fastest in tropical and subtropical countries, where travelers are exposed to specific health risks such as communicable diseases and dangerous road traffic. Professional pre-travel advice about these risks is based on up-to-date epidemiological data rated by experts. However, many travelers are not fully aware of the health hazards,[4-12] and even well-informed travelers do not always take appropriate safety precautions.[13, 14] One reason for this discrepancy may be different risk perceptions among travel health professionals and travelers. The travelers' point of view often remains unknown, as communication in pre-travel consultation is mainly consultant-directed in order to provide concise information and advice. Only a few studies have examined the subjective perception of a range of risks among travelers[6, 11] (T. Zumbrunn and colleagues, unpublished data). Better knowledge about how travelers perceive travel-associated health risks might improve the acceptance of pre-travel advice and contribute to official recommendations. This study assessed the risk perception ratings of travelers pre- and post-travel and in comparison to the ratings by travel health experts. While most surveys on travel health knowledge, attitudes, and practices (KAP studies) focus on malaria and vaccine-preventable diseases, several noninfectious travel risks with real or potential concern for travelers were included in this study.
Data were collected by convenience sampling among two groups of participants: travelers and experts. The experts (n = 30), all Swiss medical doctors and travel health consultants, were recruited at an annual national seminar on travel medicine in January 2010 (n = 28), and at the Swiss Tropical and Public Health Institute (Swiss TPH) (n = 2, coworkers without any other involvement in the study). The travelers (n = 329) were all walk-in clients of the Swiss TPH Travel Clinic who were available to the research assistant during all regular opening hours from July to September 2008 (n = 270) and from March to July 2009 (n = 59). Refusals were infrequent (9% in 2009, no data for 2008). Inclusion criteria were informed consent, age ≥ 18 years, tropical or subtropical destinations (initial consultation for a specific trip), and comprehension of German (study language). Demographic and travel-related data were collected by an anonymous interviewer-administered questionnaire. The travelers' risk perception was assessed immediately before the consultation and 2 to 4 weeks after their return home. In 2008, an additional assessment for a separate study arm was carried out immediately after the consultation to evaluate the impact of the consultation on the pre-travel risk perception and the possible qualification of the measuring instrument (described below) for routine application before pre-travel consultations (T. Zumbrunn and colleagues, unpublished data) (Figure 1). For reasons of consistency, the experts assessed all risks separately for an average traveler to Africa, Latin America, and Asia/Pacific. The study was approved by the Ethics Committee of Basel.
Measurement of Risk Perception
The visual psychometric measuring instrument applied to record the participants' risk perception, pictorial representation of illness and self measure (PRISM), was developed in 1995 and validated for the assessment of the subjective burden of suffering in patients with chronic diseases.[15, 16] It consists of a white board in DINA4 format with a fixed yellow disc, symbolizing a subject's “self” in her/his current life situation, and a movable disc, representing an illness, which is placed on the board by the subject (Figure 2). The distance between the “self” and the illness [self-illness separation (SIS)] is the primary outcome of PRISM and inversely proportional to the perceived importance of the illness. For this study, “life situation” was specified as the planned journey and “illness” replaced by nine health risks. The primary outcome was adjusted to self-risk separation (SRS) as a proxy for the risk perception. According to the severity, frequency of occurrence, or estimated concern for travelers, the following risks were included: general risk (overall danger of a specific journey), mosquitoes, malaria, rabies, epidemic outbreaks, sexually transmitted infections (STIs), accidents, terrorist attacks, and vaccination-associated adverse events (VAEs). In 2008, pre-travel data collection was carried out by a computer application of PRISM (T. Zumbrunn and colleagues, unpublished data). For technical reasons, pre-travel data in 2009, expert data, and all post-travel data were collected using hard copies of the computer application. The forms were scanned and distances measured by means of a computer-aided design (CAD) program. The CAD coordinates were converted to the original scale (cm).
Differences between the median perceptions of travelers and experts with nonoverlapping confidence intervals (CIs) were considered as statistically significant. The CIs were calculated using a bootstrap re-sampling method with 500 replicates. Linear regression was applied to detect differences among traveler subgroups and the SRS log10-transformed prior to analysis. A two-sided p value < 0.05 was considered as statistically significant. No adjustments for multiple testing were made. All analyses were performed using PASW Statistics 18 and R version 10.
Demographics and Travel Characteristics
Of 329 participating travelers, 15 were excluded from data analysis because they either did not complete the pre-travel survey (n = 10), came for a follow-up vaccination (n = 2), did not travel to at least one tropical or subtropical destination (n = 2), or were younger than 18 years (n = 1). The return rate of questionnaires was 70% for travelers after travel (n = 230) and 60% for experts (n = 18). Demographic and travel-related characteristics of the travelers are presented in Table 1. About 50% were women and 40% were older than 40 years. Most traveled for leisure (79%). Asia/Pacific (38%) and Africa (36%) were the most common regions of destinations. More than half of all participants had previously visited the respective region (56%). Nearly half (42%) consulted the Travel Clinic less than 4 weeks prior to departure. The median planned duration of the journey was 3 weeks (interquartile range 16–32 days).
Table 1. Demographic and travel-related characteristics of the participating travelers in absolute numbers and percentages; percentages are rounded to whole numbers. Total and missing data are excluded for a better overview
Multiple answers possible.
Arterial hypertension (24), mental disorders (13), status after stroke (2), immunosuppression (2), HIV (0), pregnancy (2).
Previous travel consultations for other trips at the Swiss TPH Travel Clinic
Figure 3 shows the risk perception of travelers versus experts. According to the experts, the highest risks for travelers are accidents followed by mosquitoes, STIs, malaria, rabies, and epidemic outbreaks. Terrorist attacks and VAEs were ranked lowest. Contrary to the experts' assessment, the travelers perceived accidents and STIs as significantly lower risks [accidents: median SRS 13.3 cm, 95% CI: 12.9–14.3 cm (travelers) vs 7.8 cm, 95% CI: 6.8–8.8 cm (experts); STIs: 23.6 cm, 95% CI: 23.1–24.3 cm (travelers) vs 14.4 cm, 95% CI: 12.6–16.4 cm (experts)]. STIs ranked third for the experts and last for the travelers, while all the other risks ranked similarly in both groups. Compared to the experts' assessment, the travelers' risk perception of VAEs was higher (not statistically significant) (Figure 3). The travelers' pre- and post-travel risk perceptions were similar with a trend toward a lower risk perception after travel for most items. Only accidents were perceived as a higher risk after travel, but still ranked lower than the experts' assessment in absolute figures. Thus, only mosquitoes (rank 1 to 2) and accidents (rank 2 to 1) changed position on the ranking list after travel. With the exception of STIs, the experts showed similar or smaller ranges of distribution than the travelers (Figure 3).
Gender, age, destination, and region-related travel experience had different impacts on the travelers' risk perception (Figure 4). The following differences were detected before travel: general risk and mosquitoes were considered as lower risks in Asia/Pacific than in Africa (log10-transformed coefficient 0.07, 95% CI: 0.02–0.12; 0.08, 95% CI: 0.02–0.14), and malaria was perceived as a lower risk in Asia/Pacific and Latin America than in Africa (0.15, 95% CI: 0.09–0.21; 0.19, 95% CI: 0.12–0.26). Men perceived mosquitoes, malaria, and rabies as higher risks than women (−0.09, 95% CI: −0.14 to −0.04; −0.09, 95% CI: −0.15 to −0.04; −0.05, 95% CI: −0.09 to −0.01). Compared to younger participants, travelers aged >40 years considered terrorist attacks as a higher risk and STIs as a lower risk (−0.04, 95% CI: −0.07 to −0.0004; 0.04, 95% CI: 0.002–0.08). Epidemic outbreaks and VAEs were perceived similarly by all subgroups before and after travel. Region-related experience did not influence pre-travel but influenced post-travel risk perception: travelers without experience in the region considered malaria and mosquitoes as lower risks than region-experienced travelers (0.05, 95% CI: 0.004–0.1; 0.06, 95% CI: 0.0006–0.12). Retrospectively, terrorist attacks were perceived as a higher risk in Asia/Pacific than in Africa (−0.05, 95% CI: −0.09 to −0.003), while malaria and general risk (not mosquitoes) were still considered as lower risks in Asia/Pacific than in Africa (0.06, 95% CI: 0.001–0.11; 0.05, 95% CI: 0.003–0.1). Post-travel risk perception was not different among gender, age groups, and travelers to Latin America versus Africa.
The travelers' overall perception of travel-associated health risks was mostly in accordance with the experts' assessment and appears to be accurate for most risks, with the exception of accidents and STIs. Remarkably, all risks were perceived similarly or slightly lower after travel than before, except for accidents. Mosquitoes, the number one perceived risk among travelers (before travel) and malaria, both “classic” pre-travel health topics, ranked highly among experts and travelers and were only outranked by accidents. However, the tendency of having a lower post-travel risk perception was most distinct for malaria and mosquitoes (Figure 3). The interpretation of this finding remains ambiguous, as the associations with the term “mosquitoes” are unknown and might range from “nuisance” and local bite reactions to mosquito-borne diseases. This fact also applies to epidemic outbreaks which were rated as relatively low risk throughout. In general, destination-related differences in risk perception were small with the exception of malaria (Figures 3 and 4). In accordance with the prevalence of Plasmodium falciparum, malaria was perceived as a lower risk in Asia/Pacific and Latin America than in Africa by both experts and travelers, confirming existing knowledge about the disease. The general risk of travel was also considered lower in Asia/Pacific than in Africa. The popularity of travel to Asia/Pacific might lead to this region appearing less risky than other continents. However, terrorist attacks were perceived as a higher risk in Asia/Pacific than in Africa which might have been influenced by the relatively high incidence of terrorist acts and political disturbances in Asia at the time of the study[20, 21] and their media coverage in Switzerland. This was estimated by the number of hits for the keywords “terror* asia*” compared to “terror* africa*”, “terror* south america*” and “terror* latin america*” between 1 January 2008 and 31 August 2009 on an archive portal for Swiss print media articles. Regardless of their destinations, the travelers' perception of VAEs was relatively high which is in accordance with European KAP studies describing negative attitudes toward vaccines and their potential adverse effects.[9, 23] Interestingly, pre-travel risk perception did not differ comparing region-experienced with region-inexperienced travelers who may have utilized other sources of information before visiting the Travel Clinic.
For unknown reasons, malaria, mosquitoes and rabies, three vector-borne or vector-associated health problems were perceived as higher risks by men than women before travel (Figure 4). Experts and travelers perceived the rabies risk similarly before and after travel (Figure 3), whereas the separate study arm reported a higher perception of rabies after pre-travel health consultation than before [T. Zumbrunn and colleagues, unpublished data]. Subject to coincidence, the perception might have decreased owing to lack of close encounters with mammals. Nevertheless, as rabies is a rare but always deadly disease in humans with a worldwide distribution, information about rabies needs to be part of pre-travel advice, especially as it is a neglected topic in travel health,[24, 25] and knowledge about rabies is known to be limited among travelers.[6, 9, 26]
Another relatively underrepresented health risk in pre-travel advice is STIs.[27, 28] STIs were perceived as lowest of all risks by the travelers, in significant contrast to the experts, who ranked STIs third, yet with a wide range of distribution (Figure 3). While data about the incidence of STIs among travelers is scarce,[29-31] studies about the sexual behavior of travelers indicate that STIs are not unusual souvenirs, especially among the average 20% of travelers having casual sex abroad, nearly half of which is unprotected (without condoms). However, a low pre-travel risk perception is not surprising as casual sex abroad is often not anticipated or planned and is associated with other potential risky behaviors which are more frequent among travelers than nontravelers[32, 33] such as the consumption of alcohol[13, 27, 28, 32, 33] and/or illicit drugs.[27, 30, 34] A socio-anthropological approach to understanding risk-taking behavior abroad is the concept of “antistructure” applied to tourism. “Antistructure” is the counterpart to the “structure” of everyday life, characterized by a temporary change of norms, values, and social relations while being away from home. Nevertheless, post-travel risk perception of STIs was not higher after travel than before (Figures 3 and 4). Whether some travelers had unprotected casual sex abroad is unknown. There were no gender-related differences in perception although travelers aged >40 years did perceive STIs as a lower risk than younger travelers but, interestingly, only before departure (Figure 4). Studies evaluating demographic or travel-related characteristics associated most with sexual risk-taking behavior show controversial results,[13, 14, 30, 31, 36, 37] and assumptions about the sexual activity according to gender, age, or travel mode should be made with caution. As receiving information about STIs has been associated with safer sex among travelers, STIs and their prevention should be considered as a general part of pre-travel advice, not only for the travelers' protection but also for epidemiological reasons. After all, travelers play a central role in the global spread of STIs, especially travelers to tropical and subtropical regions with the highest worldwide prevalence of STIs including HIV.[38, 39]
Accidents were the only risk perceived higher after travel but significantly lower by travelers than by experts (Figure 3). Injuries, particularly road traffic accidents, are the second most common cause of death abroad after cardiovascular disease[40-43] and the leading cause of death of those aged 15 to 29 years worldwide. Over 90% of road traffic fatalities occur in low- and middle-income countries, including many tourist destinations in the tropics and subtropics. Higher mortality rates due to vehicle accidents have been found among travelers than among the local population. Travelers are often not familiar with poor road conditions and different, partly insufficient or insufficiently enforced road traffic laws, and they might engage in high-risk behavior during vacation. Despite their potential for disability and other complications, little is known about the incidence, type, and severity of nonfatal accidents among travelers. Injuries were reported by 6 to 16% of travelers in three different studies,[14, 46, 47] most of them due to road traffic accidents. The most vulnerable groups on the road are pedestrians, (motor) cyclists, and users of unsafe or overcrowded public transport. Some studies suggest that (young) men are most likely to be involved in (fatal) vehicle crashes[43, 48] and engage in more risk-taking activities than women.[14, 49] However, there were no gender- or age-related differences in the perception of accidents in this study (Figure 4). The post-travel increase in perception is most likely due to observed danger abroad. In other studies, accidents were also rated as a more important health problem during or after a stay abroad than before.[10, 50] In order to raise awareness of this potentially life-threatening risk before departure, information about accidents abroad including practical preventive measures needs to be an integral part of pre-travel health advice.
PRISM has only been validated for the assessment of the subjective burden of a present illness, not for the perception of health risks in the near future and past. Nevertheless, a fast, nonverbal visual tool may take into account the emotional quality of (risk) perception which is subjective among both travelers and experts. Statistical correlation of the perception of risks with their incidence was not an option as up-to-date, comparable data were not available or collected. However, the experts' risk assessment, used as a reference point, proved to be consistent with current literature. Generalization of the results is limited owing to the single location of the study. Additionally, clients of a travel clinic are probably pre-sensitized regarding health risks. Convenience sampling, different periods of data collection, and different associations with unspecified risks may have caused bias to an unknown extent. Travelers known to be more exposed or susceptible to certain risks, for example, persons visiting friends and relatives, persons with chronic illnesses, pregnant women, or business travelers, are interesting target groups for the assessment of risk perception, but underrepresented for analysis in this study (Table 1).
Travelers' risk perception appears to be accurate for most risks stated in this study. However, travel health professionals should be aware that some perception patterns among travelers regarding travel-related health risks may be different from professional risk assessment. We suggest that important but insufficiently perceived health risks, such as sexual behavior/STIs and accidents, should be included in any pre-travel health advice package, whether given in person, printed, or online.
The authors would like to thank Stefanie Zumbrunn-Jegge for contributing the baseline information of this follow-up study and for supporting the team with most valuable inputs. We thank the travelers and experts for participating in the study, and the Travel Clinic team for their help and support.
Declaration of Interests
The authors state that they have no conflicts of interest.