Risk perception, travel intentions and self-protective behavior of chronically ill tourists under the protection motivation perspective

Data analysis

Data analysis and structural equation modeling were performed using SPSS 26.0, AMOS 28.0. First, the demographic characteristics of the participants in this study were determined and descriptive statistics and distributions were calculated. Table 1 shows that the gender ratio of the respondents was comparable, with slightly more males (52.8%); the largest proportion of respondents was over 55 years of age (35.3%), but the next largest proportion of people aged 46–55 years of age was 27.4%, which reflects a gradual tendency for chronic diseases to be younger; the majority of respondents were at the undergraduate level and below (93.6%); in terms of the level of household income, respondents were more concentrated in the range of 8,000–10,000 yuan (42.5%), but chronic diseases occur in all types of income groups; cardiovascular diseases and respiratory diseases account for a larger proportion of the respondents (64.1%), which echoes the reality; more than half of the respondents are in a stable stage of the disease (58.9%), of which a larger proportion of the respondents are married (82.3%); the majority of respondents’ annual traveling frequency was 2 or more times (85.1%).

Control and test for homoscedasticity

The common method bias of the data was procedurally controlled by emphasizing that there is no right or wrong answer to the questionnaire, using anonymous questionnaires, guaranteeing the respondents’ right to know, and the use of reverse questions when the questionnaires were distributed. However, the collection of questionnaire data was done in the same measurement setting; therefore, after the questionnaires were returned, this study used the Harman one-way test for common method bias. Unrotated exploratory factor analysis was conducted on all measurement items, and the results showed that the KMO value was 0.900, and the cumulative variance explained by the eight principal components extracted was 63.525%, of which the variance explained by the first principal component was 30.842%, which was less than 40.000%. Therefore, the negative impact caused by common method bias was better controlled in this study.

Reliability and validity tests

Tables 2 and 3 include summary statistics and reliability tests for all potential variable indicators. Reliability is mainly used to analyze the reliability and accuracy of the scale data, and the alpha value of Cronbach for each variable is greater than 0.7, indicating that the scale has good internal consistency and stability. Validity was mainly tested by convergent validity and discriminant validity. The test of variance was conducted by comparing the standardized loadings with all the indicators specified by the other measurement models, and according to the estimation, the convergent validity is good when the AVE is above 0.5 and the CR is above 0.7. As can be seen in Table 2, the validity of the scale is good. In addition, the standardized loading level of each variable is above 0.5, which also indicates that the convergent validity of the scale is better.

As seen in table 3, the square root of the mean extracted variance of each latent variable is greater than the correlation coefficient between the latent variable and the other latent variables, which indicates that the discriminant validity of the measurement model is good.

Path analysis and hypothesis tests

Structural equation analysis was performed using AMOS 28.0, and the results of the model goodness of fit were: CMIN/DF = 1.426 < 3, GFI = 0. 940, NFI = 0. 930, TLI = 0. 975, CFI = 0. 978, all above 0.8, SRMR = 0. 039 < 0. 05, and RMSEA = 0. 029 < 0. 08, which indicated that the model fit was good. The results of direct path test are shown in Table 4, among the factors related to PMT, perceived vulnerability, perceived severity, and response cost, the positive effect on respondents’ risk perception is significant, self-efficacy has a significant negative effect on risk perception, and H1, H2, H3, and H5 are supported, but response efficacy has a non-significant direct path on risk perception (β = 0. 097, p>0. 05); the direct path of respondents’ risk perception to positive effects of Courageous travel and self-protection are both significant, H6, H7 supported.

The model was further tested for mediated effects by running the model 5000 times in AMOS 28.0 using Bootstrap method, resulting in the level values of Bias – Corrected at 95% confidence level as shown in Table 4, the effect is significant when the confidence interval does not contain 0, the presence of mediated effects is determined by whether the indirect effect is significant or not. The results show that the indirect effect value of risk perception between response efficacy and self-protection is 0.036, and between response efficacy and courageous travel is 0.017; the indirect effect value of risk perception between response cost and self-protection is 0.066, and between response cost and courageous travel is 0.051,; the confidence intervals all contain 0, and the mediation effect is not significant., response efficacy and response cost cannot influence decision-making behavior of chronically ill tourists through risk perception.

Moderated effects test

According to Wen’s suggestion⁵⁸, this study used SPSS grouped regression to conduct regression analysis on the samples of tourists with different psychological types separately to test the effect of risk perception on behavioral intention in each group of samples. In this study, the control variables were included in the regression equation in the first step through the hierarchical regression method, and the main effect test of the independent variable (risk perception) was conducted in the second step. The K-means rapid clustering method was used, and after 10 iterations and classifications (See appendix for cluster analysis results), the results showed that it was more reasonable to classify tourists’ psychological types into three categories, and all the items were significant on the basis of Sig value of 0.000, which was able to distinguish the differences in the psychological types of tourists better. Combined with the psychological type of Plog tourists, chronically ill tourists were divided into three categories: adventurous tourists (N = 102, accounting for 20.2%), intermediate tourists (N = 268, accounting for 53.2%) and dependent tourists (N = 134, accounting for 26.6%). The percentage structure of the three psychological types of tourists is similar to the studies of Weaver⁴⁴ and Wang¹⁵, which suggests that dependent tourists are much less than adventurous and intermediate tourists in contemporary society.

As shown in Table 5, there are significant differences in the influence coefficient and adjusted R² of risk perception on courageous travel among different groups of tourists’ psychological types. Specifically, the influence coefficient of the adventurous tourists group is 0.621, the influence coefficient of the intermediate tourists group is 0.488, and the influence coefficient of the dependent tourists group is 0.287; the influence coefficient and adjusted R² of the risk perception on self-protecting behaviors are significantly different. Specifically, the influence coefficient of the dependent tourists group is 0.391, the influence coefficient of the intermediate tourists group is 0.381, and the influence coefficient of the adventurous tourists group is 0.362. The above results indicate that the tourists’ psychological type has a nonlinear moderating role in giving the relationship between perceived risk and behavioral intention, and H10a-b are supported. Among them, risk perception has the strongest influence on adventurous tourists’ courageous travel, followed by intermediate tourists and weakest by dependent tourists; risk perception has the strongest influence on dependent tourists’ self-protective behavior, followed by intermediate tourists and weakest by adventurous tourists.

Discussions and implications

Main conclusions

Vulnerable groups with chronic illnesses constitute an expanding niche market that has gone largely unnoticed by travel scholars. Despite physical or mental impairments, many in this group are able and eager to travel⁵⁹. Research has explored the health risk perceptions and travel intentions of chronically ill travelers to understand the various types of travelers at different stages of their illnesses and to promote more inclusive tourism practices^17,36,42. The main findings of the study are as follows:

The factors related to PMT, vulnerability, severity, and response cost, had a significant positive effect on risk perception, which suggests that an increase in threat factors or predicted costs leads to a stronger risk perception among chronically ill tourists, these findings are in line with those of²². On the other hand, self-efficacy was found to have a significant negative effect on risk perception, which is a new finding from the investigation of this study for the chronically ill group, people with high self-efficacy, who believe in their ability to defend themselves against the threat of disease and weaken the risk factor. However, no effect of response efficacy on risk perception was found in this study, which laterally reflects that chronically ill tourist, although believing in the ability of self-defense against risks, expressed concerns about whether preventive measures are implemented and enforced.

Second, this study confirmed the impact of risk perception on self-protective and actual travel behavior of chronically ill tourists. As expected, risk perception was not only able to influence tourists’ self-protection motivation, but also directly influenced tourists’ courageous travel, which in the cognitive horizons of previous studies often tends to favor travel avoidance or reduced travel intentions^29,36. For chronically ill tourists, risks are not entirely unavoidable, and their health risks are predictable, so behavioral intention not only focuses on the arousal of self-protection motivation, but also may motivate individuals to shift to a positive attitude toward travel, combining the results of the present study with the perspective of self-efficacy theory, the self-efficacy of the individual to cope with the risk is one of the necessary elements to enhance their courage to travel.

The study also confirmed the mediating role of risk perception. The results showed that risk perception was a partial mediator between PMT factors and courageous travel, self-protection. Therefore, the travel behaviors and protective behaviors of chronically ill tourists largely depend on the degree of strength of risk perception, which is mainly influenced by PMT factors, especially threat factors and efficacy factors, and there is a step-by-step process of conduction and reinforcement in influencing tourists’ risk perception as well as the results of behavioral intentions, a finding that is consistent with the viewpoints of existing studies which expands the formation mechanism of risk perception and behavioral intention under the perspective of protection motivation^15,17. The results of the moderating effect further explain the ability of different psychological types of tourists to tolerate perceived risks. Specifically, in the influence path of “risk perception → courageous travel”, the main effect of adventurous tourists is the strongest, which indicates that adventurous tourists have a high sense of self-efficacy, they are self-confident and extroverted, and they are eager to take risks, their strong motivation and autonomy play the most important supportive role for travel behavior, which is consistent with the research results of Xie’s study³⁷. In the influence path of “risk perception →self-protective behavior”, the main effect of dependent tourists is the strongest, which indicates that dependent tourists are risk averse, conservative, and easily affected by threat factors, and play the most supportive role in self-protective behaviors.

In summary, perceived health threats strengthen the risk perception of chronically ill tourists, stimulate their self-protection awareness, and in order to reduce the fear of risk itself, tourists choose to self-protect themselves when making travel decisions; however, tourists with higher levels of risk mastery and confidence in dealing with risk, who have a high degree of self-efficacy and a high level of risk-tolerance, will weaken their own risk perception and choose to travel courageous. The influence of risk perception on the decision-making behavior of adventurous, intermediate and dependent tourists was further explained by the psychological types of Plog tourists.

Theoretical contributions

This study explores the travel decisions of chronically ill tourists in risky situations using protection motivation theory (PMT). The main theoretical contributions of this study are that it introduces the risk perception and behavioral intention of chronically ill tourists into the PMT framework, focuses on the “population orientation”, and pays attention to the intrinsic characteristics and dynamics of health risks of the chronically ill group, which fills in the gaps of the existing studies; Under the guidance of PMT theory, the re-segmentation of chronic disease tourists’ intention to travel, which is based on the perception of risk and uncertainty, including courageous travel intention with higher risk tolerance and self-protection intention with lower risk tolerance, helps to explain the mediating mechanism of risk information that influences behavior, and comprehensively examines the differential characteristics of tourists’ behavioral intentions, expanding the theory of protective motivation’s applicable scenarios^37,45, validating the research on the validity of the PMT theoretical framework in the field of tourism. The study confirms that risk perception is a multidimensional and complex concept, which is influenced by external risk factors and tourists’ self-competence, and that the strengthening or weakening of an individual’s risk perception tends to change their perception of the controllability of risk, mainly due to threat factors and self-efficacy, and the control of competence.

The study constructed and validated the mediating mechanism of risk perception and the moderating effect of Plog tourists’ psychological types in the process of risk perception influencing tourists’ behavioral intentions, such as resilience, impulsivity, and travel experience, have been identified as moderators and boundary conditions of risk information that influence tourists’ behavioral decisions³⁶. The mediating mechanism of tourists’ risk perception is a research direction that has received much academic attention^34,60, but few studies have explored the mechanism of Plog tourists’ psychological types as moderating variables in the process of risk perception and individual behavioral responses. The study reveals the differences in tolerance to perceived risk and decision-making behavioral differences among different psychological categories of tourists, and distinguishes the cognitive-behavioral responses of different types of tourists driven by individual factors.

Practical implications

The physical and mental health problems of chronically ill groups are widespread, tourism as a basic human right, is a means for socially marginalized groups to achieve social interaction, access to welfare and improve the quality of life. More participatory and inclusive research can promote the de-marginalization of groups such as special groups, and ultimately promote the sustainable development of society. The desire of chronically ill groups for tourism activities has become stronger and stronger, and tourism satisfies their immediate pursuit of pleasurable experiences. The study focuses on the health risk perception and travel intention of chronically ill groups, pays attention to their awareness of risk during the travel, and then improves their social functioning and helps them to establish healthy behaviors to improve their quality of life.

In the process of disseminating information to tourists, tourism enterprises and relevant management departments should take into full consideration the psychological characteristics of chronically ill tourists, and put forward targeted dissemination strategies for tourists of different psychological types, so as to promote tourists to take the initiative to strengthen their own health protection in the process of tourism activities. For example, through the tourism APP for tourists to conduct psychological type test to distinguish the psychological type of tourists, and at the same time for different psychological types of tourists to push the personalized knowledge of educational content, in order to protect the quality of tourists’ experience in the tourism process.

Specifically, adventurous travelers usually have a full understanding of their own disease conditions and coping measures, especially whether they are suitable for high-intensity or risky activities, in addition to knowing the medical facilities of the destination in advance, carry the necessary medication or medical equipment to cope with possible emergencies; Intermediate tourists should choose moderate activities, avoid excessive exertion or adventure, keep the itinerary flexible, adjust the plan at any time according to the physical condition, and learn about the climate, diet and medical conditions of the destination in advance, so as to avoid extra burden on the body as far as possible; dependent tourists are advised to travel with family members, friends or professional medical escorts to get necessary help, give priority to travel destinations with good medical conditions, convenient transportation and comfortable environment, pay attention to their psychological feelings, and communicate with fellow travelers at the right time, so as to avoid affecting the travel experience due to discomfort or anxiety. However, no matter which psychological trait chronic tourists belong to, they should take physical health as a prerequisite, plan their trips reasonably, pay attention to health, and have a certain degree of psychological adjustment ability.

Research limitations

First, this study focused on quantitative analysis methods to explore the causal relationship between variables, given the complexity of the theoretical model, future research can apply mixed methods or qualitative analysis to enhance the interpretation of the framework, such as adding qualitative interviews with patients’ subjective feelings, with the help of experimental paradigms and professional scales in the medical field, to be more targeted and rigorous. In addition, for the application of the model, this study was mainly based on the past tourism experience of the chronically ill group, and future research should consider the mechanism of influencing tourists’ behavior in different tourism stages.

Secondly, although we tried to adopt quotas as much as possible during data collection, respondents mainly cooperated with the survey in the hospitals they visited due to time, resources and social distance, which may limit the generalizability of the findings, and future studies should adopt a wider range of surveys as much as possible.