Perceived Risks, Travel Constraints, and Destination Perception: A Study on Sub-Saharan African Medical Travellers

: In medical travel, previous studies have investigated the factors that inﬂuence medical travellers to receive treatment outside the country. However, most of these studies are limited to travel motivations and perceptions of medical services at destinations. The main objective of this study was to investigate the relationship between medical travellers’ perceived risks, travel constraints, and destination image based on medical and non-medical attributes. This is a quantitative study whereby the data was collected from 306 sub-Saharan African medical travellers, who visited India for the treatment. The study found that physical-health risk has a signiﬁcant negative inﬂuence on destination image based on medical attributes. The service quality risk has a negative e ﬀ ect on destination image based on both medical and non-medical attributes, and destination risk has a negative e ﬀ ect on destination image based on medical attributes. The study also found that travel constraints have a negative inﬂuence on both medical and non-medical destination image.


Introduction
Several countries have benefited from international medical travel in terms of increased GDP, improved healthcare infrastructure, attracting foreign currency, promoting tourism, and, more importantly, checking the brain drain of skilled medical professionals [1,2]. However, the reason behind this phenomenon is the unavailability of certain types of treatments such as bypass surgery, cancer treatment, transplantations, eyecare in a home country, or treatment that is available with low quality or at a very high price. International medical travel is defined as "the citizens of source country travel to foreign destination hospitals with the sole purpose of obtaining necessary medical procedures and treatment". Many researchers disagree with using the term "medical tourism" for a journey performed solely in search of medical treatment, as the term otherwise mainly defines the phenomenon of north-to-south travel where receiving health and medical care was combined with leisurely activities. Researchers argue that the majority of individuals traveling outside the country for treatment are needy patients who have the sole motivation to receive the required treatment [3,4].
International travel for medical services in the sub-Saharan region of Africa has been growing rapidly in the last decade [5]. Although the specific figures are not available on the number of travelers traveling outside the country for medical needs, general figures can provide a broad picture of the size of international medical travel in sub-Saharan Africa. According to Crush and Chikanda [6], more than 2 million patients mainly from sub-Saharan Africa traveled to South Africa between 2003-2008 at an average of 400,000 per annum. In the year 2010, an average of 3000 Nigerians traveled to India for

Literature Review and Development of Research Hypotheses
The theory of planned behaviour (TPB) can predict the behaviour of individuals through attitude and intention. Intention develops through attitude, subjective norms, and perceived behavioural control and leads to actual behaviour. Previous studies applied the TPB model to study medical travel [18,19] and investigate the travel behaviour of travellers.

Perceived Destination Image (Medical and Non-Medical)
Destination image is "the total of ideas, beliefs, and impressions individual possesses about the attributes of destination and or activities available at a destination after processing information from various sources over a time period" [24]. Echtner and Ritchie [25] described the destination image as "an overall imagery picture individual obtains in his mind." The literature gives the impression that the attributes of a medical travel destination can be categorised into medical and non-medical. Hospitals providing services to medical travellers publicise their services through websites and media, and potential travellers predominantly refer to these mediums to select their preferred hospital or clinic to receive treatment [26,27]. The main features medical travellers look for are reliability and quality of services and cost. The importance of the cost factor is high in the context of Asian medical travel destinations as the general perception is that the cost will be significantly lower than their home countries. In terms of medical attributes, medical travel destinations emphasise sophistication, advanced technology, cleanliness and efficiency, variety of procedures, low price, accreditation of hospitals, affiliation of doctors, qualified and smart staff, efficiency of different languages, lavishly furnished wards, and testimonials of the patients [28][29][30][31].
The literature also revealed that support services play an important role in attracting medical travellers. Travellers look for proper accommodation, transportation, cleanliness, and personal safety and security [28]. They also look for the opportunity to tour the destination [32]. Many medical travellers prefer Thailand due to the availability of world-class tourism infrastructure and leisure activities [33]. Services for companions such as accommodation, transport service, and food, etc. play a role in creating a favourable image of a destination [29]. For example, India receives travellers mainly from the low-income countries such as neighbouring countries and Africa, and they often have limited options for accommodation and other services like food and transportation. Prakash, Tyagi [10] found that most medical travellers were not satisfied with the accommodation services and taste of food in India.
In travel and tourism literature, perceived image is considered to be the most important factor in the decision-making of travellers. The results of the previous studies indicate that the destination image was a direct antecedent that influences tourists to visit or revisit a destination [34]. If the image of a destination is positive, it will influence travellers to visit and vice-versa. The process of decision-making of medical travellers is based on evaluating various attributes of the destination, and the image that forms in the mind of the travellers about the destination will influence the choice of destination.

Perceived Travel Risks
Perceived risk reflects the uncertainty regarding possible negative consequences of using a product or service. In consumer behaviour, perceived risk can be defined as 'the combination of uncertainty plus seriousness of outcome involved and the expectation of loss associated with a purchase that acts as an inhibitor to purchase behaviour' [35]. In travel behaviour, perceived risk is "the perceived probability that action may expose an individual to danger and can influence travel decisions if the perceived danger is deemed to be beyond an acceptable level" [36].
In medical travel, travellers from the low and middle income countries are generally perceived as being more vulnerable to different kind of risks at the destination due to lack of knowledge, fewer resources, and prejudices in the local population about them [37,38]. In this study, perception of risks was confined to risks faced by the medical travellers at the destination as pre and post-procedure risks of medical travellers from the high-income countries [39,40]. After a thorough review of the literature, the study categorised the perceived risks faced by medical travellers during their visit as physical-health, service quality, and destination risk. The three dimensions are used in the study to investigate the perceived travel risks of medical travellers.
Physical risk refers to the possibility of physical danger or injury detrimental to health, whereas health risk refers to the possibility of becoming sick while travelling or while at the destination [36]. In medical travel, individuals can be exposed to contracting epidemic diseases due to spending days or weeks as an in-patient. Medical travellers can face the risk of contracting severe blood-borne infections such as HIV, hepatitis, etc. due to improper screening and storage of blood [40]. The health risk for the medical travellers receiving organ transplant is even higher due to the ignorance of standard protocol in donor selection at many destination countries [41]. Therefore, the following can be hypothesised: H1: Perceived physical-health risks to medical travellers will negatively influence perceived destination image based on medical attributes.

H2:
Perceived physical-health risks to medical travellers will negatively influence perceived destination image based on non-medical attributes.
Service quality risk is always a matter of concern among medical travellers. Medical travellers are concerned about the skills of physicians, nurses, and technicians. Due to this, most hospitals providing services to medical travellers emphasise obtaining accreditation from international bodies such as Joint Commission International (JCI). However, there were many cases when medical travellers faced difficult times due to negligence in the quality of services provided such as unwanted procedures, inflated the cost, and prolonged stay in the hospitals [3,40,42]. Therefore, it can be hypothesised that: H3: Perceived service quality risks to medical travellers will negatively influence perceived destination image based on medical attributes.

H4
: Perceived service quality risks to medical travellers will negatively influence perceived destination image based on non-medical attributes.
In relation to destination risk, medical travellers can be subjected to face criminal incidences such as robbery, physical and sexual assaults, and racism-related incidences at the destination. Medical travellers from the low and lower-middle income countries are subjected to higher vulnerability to these kinds of incidences as in many cases they stay at rented houses for long periods due to the complexity of the medical condition and treatment. Studies in leisure travel have found that the perception of travellers on safety and security, service quality, performance, and health risks have a significant negative influence on destination image perception [22,23,43]. In medical travel, perceived risks were strongly associated with attitude [18] and perceived value [19] of the destination and services. Attitude and value are the perceptions based on attributes and characteristics. In this sense, it can be assumed that perceived risks of medical travellers will create a negative perception about the destination attributes. Therefore, it can be hypothesised that: H5: Perceived destination risks to medical travellers will negatively influence perceived destination image based on medical attributes.

H6:
Perceived destination risks to medical travellers will negatively influence perceived destination image based on non-medical attributes.

Travel Constraints
Travel constraints refer to the "factors that inhibit continued travel, cause inability to begin travel, results in the inability to maintain or increase the frequency of travel, and/or lead to negative effects on the quality of travel" [44]. Travel constraints are the key factors that prevent people from initiating or continuing to travel [45].
Although the constraints factor in international medical travel is largely unexplored, the literature indicates that medical travellers face different types of travel constraints such as anxiety, stress, sickness and stage of life, lack of money, travel documents, and travel companion [10,15,46]. Muraina and Tommy [5] found that most Nigerian medical travellers felt stress when travelling abroad for medical needs. Emotional factors also contribute to the constraints faced by medical travellers as in many cases close family members do not travel with the patient and other relatives accompany them while receiving treatment [15]. The constraints related to support services can also be a factor that inhibits international medical travel. According to Prakash, Tyagi [10], most African travellers perform their journey to India via Middle-Eastern countries due to the lack of direct flights. This broken journey could be painful for medical travellers who have severe medical conditions. Structural factors such as flight connectivity, documentation, and visa were major issues associated with medical travellers.
Medical travellers face different types of constraints while planning to travel to a destination. Travel constraints could be specific to a destination which may not be the same for another destination. In this situation, travel constraints will surely affect the perceived image of a destination. The literature in travel and tourism explained the negative effects of constraints on destination image [21,23] and collectively posit that the travel constraints of medical travellers will have a negative effect on the perceived image destination. Therefore, it can be hypothesised that: H7: Travel constraints of medical travellers will negatively influence destination image based on medical attributes.
H8: Travel constraints of medical travellers will negatively influence destination image based on non-medical attributes. (See Figure 1) Sustainability 2020, 12, x FOR PEER REVIEW 5 of 16 The literature in travel and tourism explained the negative effects of constraints on destination image [21,23] and collectively posit that the travel constraints of medical travellers will have a negative effect on the perceived image destination. Therefore, it can be hypothesised that: H7: Travel constraints of medical travellers will negatively influence destination image based on medical attributes.
H8: Travel constraints of medical travellers will negatively influence destination image based on nonmedical attributes. (See Figure 1)

Data Collection Procedure and Sampling
Data were collected from the sub-Saharan African medical travellers visiting two large private hospitals in Delhi and NCR in the months of February-July 2017. Six medical travel facilitators, officially associated with the hospitals and providing services to medical travellers in the vicinity of the hospitals, were engaged in finding medical travellers to participate in the study survey (see Appendix A). Non-probability convenience sampling was used due to the lack of a sample frame. Convenience sampling is the most frequently used technique in quantitative studies [47]. Pick/drop method was used, and the questionnaires were dropped to the facilitators and collected. In total, 700 questionnaires were distributed, and 318 were returned to the researchers (response rate of 45.4%). All of the returned questionnaires were thoroughly scrutinised, and 306 were found fit for data analysis.
To ensure no common method bias in the questionnaire survey, Harman's single factor test was performed. The results revealed that the first factor accounted for 20.10% of the variance, which was less than the given threshold level of 50% of the total variance explained [48]. Studies identified that the minimum sample size of 100 is adequate for PLS-SEM [49]. Chin [50] explained the rule of thumb "ten times rule" for PLS-SEM analysis. Statistical power is another way to estimate the more restrictive minimum sample size recommended for PLS-SEM analysis [51]. The sample size of 306 satisfied all the conditions to estimate the minimum sample size.

Measurement
All the items were measured on a seven-point Likert scale that ranged from (1) strongly disagree to (7) strongly agree. For physical-health risk, three items; service quality risk, three items; and

Data Collection Procedure and Sampling
Data were collected from the sub-Saharan African medical travellers visiting two large private hospitals in Delhi and NCR in the months of February-July 2017. Six medical travel facilitators, officially associated with the hospitals and providing services to medical travellers in the vicinity of the hospitals, were engaged in finding medical travellers to participate in the study survey (see Appendix A). Non-probability convenience sampling was used due to the lack of a sample frame. Convenience sampling is the most frequently used technique in quantitative studies [47]. Pick/drop method was used, and the questionnaires were dropped to the facilitators and collected. In total, 700 questionnaires were distributed, and 318 were returned to the researchers (response rate of 45.4%). All of the returned questionnaires were thoroughly scrutinised, and 306 were found fit for data analysis.
To ensure no common method bias in the questionnaire survey, Harman's single factor test was performed. The results revealed that the first factor accounted for 20.10% of the variance, which was less than the given threshold level of 50% of the total variance explained [48]. Studies identified that the minimum sample size of 100 is adequate for PLS-SEM [49]. Chin [50] explained the rule of thumb "ten times rule" for PLS-SEM analysis. Statistical power is another way to estimate the more restrictive minimum sample size recommended for PLS-SEM analysis [51]. The sample size of 306 satisfied all the conditions to estimate the minimum sample size.

Measurement
All the items were measured on a seven-point Likert scale that ranged from (1) strongly disagree to (7) strongly agree. For physical-health risk, three items; service quality risk, three items; and destination risk, six items were adapted from Fuchs and Reichel [52], Fuchs [53], and Sönmez and Graefe [54], respectively. Five items for travel constraints were adapted from Huang and Hsu [55]. For destination image based on medical attributes, nine items; and non-medical attributes, nine items were adapted from Saiprasert [56]. Pre-testing of the survey questionnaire was conducted using the debriefing method of personal interviews suggested by Hunt, Sparkman Jr. [57].

Data Analysis and Findings
Most of the study respondents were male (73.5%), and most of the respondents were married (66.3%). The study divided the age group into five categories; the age of the respondents was evenly distributed in four groups except for 36-45 years which represented 32.4% of the total respondents. Table 1 details the socio-demographic characteristics of the respondents.

Assessment of Measurement Model
The factor loadings of all the items used exceeded the recommended value of 0.708 by Hair, Sarstedt [58] except for nine items. However, these items were retained in the analysis as the average variance extracted (AVE) of the correspondent variables of these items were higher than the recommended value of 0.50. The composite reliability (CR) for the constructs ranged from 0.842-0.959. The values exceeded the cut-off value of 0.70 for CR as recommended by Hair, Sarstedt [58]. Table 2 summarises the loadings, CR, and AVE for the items and constructs.  The discriminant validity of the model was tested by Heterotrait-Monotrait (HTMT) ratio. This study used a conservation level of 0.85 (i.e., HTMT 0.85 ) as suggested by Henseler, Ringle [59]. Table 3 shows that the discriminant validity of the model was established since all the results of the HTMT 0.85 criterion were below the critical value of 0.850.

Assessment of the Structural Model
In PLS, the main evaluation criterion for the goodness of the structural model is that the R 2 measures the coefficient of determination and the level of significance of the path coefficient [60]. The recommended value [61] of R 2 between 0.02 and 0.12 is weak, 0.13 and 0.25 moderate, and 0.26 and above substantial. However, Hair, Ringle [60] qualified these figures and suggested that a high R 2 is dependent on specific research context. The R 2 of destination image-medical attributes is 0.102 and non-medical attributes 0.083 which shows the percentage of total variance explained of these variables by respective exogenous variables. Figure 2 shows the respective exogenous variables of the two endogenous variables.

Discussion and Implications
The study found that those medical travellers who perceived high physical-health risks possessed negative perceptions of the destination image based on medical attributes. Physical-health risks were measured based on facing health-related issues, getting sick due to contaminated food and drinking water, and contracting infectious diseases. The image of many medical travel destinations is weak in terms of hygienic conditions, pollution, infectious diseases, physical injuries, and safety and security [3,[67][68][69]. The findings of this study support the arguments by researchers on the perception of risks. Many western governments release their travel advisories periodically and warn their citizens about contracting infectious diseases such as malaria, Japanese encephalitis, cholera, typhoid, hepatitis, tuberculosis, and poor air quality, and physical injuries due to accidents while travelling to countries of Asia and Latin America. Most of the well-established medical travel destinations are in this region. Western governments also advise those citizens who have existing medical conditions to take extra care of their health while travelling to emerging countries [70][71][72][73]. The Governments of Canada, America, and the UK have issued special travel advice to medical travellers about health-related issues at medical travel destinations such as blood-borne infections like HIV, hepatitis-B due to infected blood transfusion, acquiring malaria from local blood banks, contact with multi-drug resistant bacteria in hospitals, and illegal organ selling. Although the situation related to the prevalence of infectious diseases, physical injuries, and hygiene is no better in most of the medical travel source countries than the destination countries, a wide range of sources of information such as medical travel blogs, forums, websites, and travel advisories from western governments may create health-related worries in the minds of most medical travellers, especially when they are already suffering serious medical conditions. The medical travel destinations and the hospitals providing services to medical travellers should take initiatives to reduce the physical-health related concerns to increase the confidence of medical travellers. The marketing managers must take The path coefficients were significant for the relationships between travel physical-health risk and destination image-medical attributes (β = −0.139, p < 0.10), service quality risk and destination image-medical and non-medical attributes (β = −0.128, p < 0.05; β = −0.138, P < 0.10 respectively), destination risk and destination image-medical attributes (β = −0.233, p < 0.05), and travel constraints and destination image-medical and non-medical (β = −0.176, p < 0.01; β = −0.186, p < 0.05, respectively). Therefore, H1, H3, H4, H5, H7, and H8 were supported (See Table 4). The predictive sample reuse technique, also known as the Stone-Geisser's Q 2 , can be used as a criterion for predictive relevance in addition to looking at the magnitude of R 2 [62]. The Q 2 value greater than zero indicates that exogenous variables have predictive relevance for endogenous variables [53]. Table 5 details the values of Q 2 , GoF, and power. The study conducted global goodness-of-fit (GoF) for the SEM as suggested by Tenenhaus, Amato [63]. Wetzels, Odekerken-Schröder [64] guidelines were followed to estimate the GoF value which can be used as a cut-off for the global validation of the PLS model. The GoF value for the study model of 0.326 exceeded the cut-off value of 0.25 for a moderate effect size of R 2 given by Akter and Hani [65]. The power analysis for the study model was done by using the Danielsoper.com [66]. Cohen [61] argued that there is a useful role for power analysis in assessing completed research particularly in which non-significant results were obtained. The result of the power analysis indicates that the regression models hold adequate power (> 0.85), as given by Cohen [61]. Table 5. Results of Blindfolding, GoF, and Power Analysis.

Discussion and Implications
The study found that those medical travellers who perceived high physical-health risks possessed negative perceptions of the destination image based on medical attributes. Physical-health risks were measured based on facing health-related issues, getting sick due to contaminated food and drinking water, and contracting infectious diseases. The image of many medical travel destinations is weak in terms of hygienic conditions, pollution, infectious diseases, physical injuries, and safety and security [3,[67][68][69]. The findings of this study support the arguments by researchers on the perception of risks. Many western governments release their travel advisories periodically and warn their citizens about contracting infectious diseases such as malaria, Japanese encephalitis, cholera, typhoid, hepatitis, tuberculosis, and poor air quality, and physical injuries due to accidents while travelling to countries of Asia and Latin America. Most of the well-established medical travel destinations are in this region. Western governments also advise those citizens who have existing medical conditions to take extra care of their health while travelling to emerging countries [70][71][72][73]. The Governments of Canada, America, and the UK have issued special travel advice to medical travellers about health-related issues at medical travel destinations such as blood-borne infections like HIV, hepatitis-B due to infected blood transfusion, acquiring malaria from local blood banks, contact with multi-drug resistant bacteria in hospitals, and illegal organ selling. Although the situation related to the prevalence of infectious diseases, physical injuries, and hygiene is no better in most of the medical travel source countries than the destination countries, a wide range of sources of information such as medical travel blogs, forums, websites, and travel advisories from western governments may create health-related worries in the minds of most medical travellers, especially when they are already suffering serious medical conditions. The medical travel destinations and the hospitals providing services to medical travellers should take initiatives to reduce the physical-health related concerns to increase the confidence of medical travellers. The marketing managers must take initiatives to spread positive images about destinations in terms of safe food and drinking water and about the measures taken for the prevention of infectious diseases and physical injuries to reduce the perception of physical-health risk among medical travellers.
The results also found that the perceived service quality risk among medical travellers was negatively related to the perception of the destination on medical and non-medical attributes. Service quality risk was measured based on unsatisfactory medical service that fails to meet expectations and is not value for money. However, descriptive analysis shows that the mean value of the variable service quality risk was 2.65 (SD = 1.31), which means that most of the respondents consider service quality risk related to medical travel destination as being low. The literature on medical travel argued that the high-quality medical service provided by the hospitals at the medical travel destination is the most important factor that pushes the majority of the medical travellers to travel to foreign country [3,5,6,74]. However, a negative perception of non-medical attributes is a concern for medical travel destinations. The findings of this study support the arguments of previous studies that issues such as hidden cost, food quality, accommodation, and travel documents need to be addressed for the continuous growth of medical travel [10]. Medical travel destinations must emphasise maintaining and improving the quality of medical and non-medical services offered to medical travellers and highlight them in the marketing and promotional materials to attract more medical travellers.
The results of the study reveal that the destination risks of the medical travellers were negatively related to the medical attributes of the destination. Destination risk was measured based on crime and safety, political instability, racist attacks, and prejudice among the local population. Destination risks have not been given due consideration in medical travel literature. However, most medical travel destinations are middle or lower-middle income countries and rank low on the social progress index [75]. Medical travellers face different kinds of crime and safety issues at the destinations such as the sexual assault on female foreign patients at hospitals, fraud, and opposition to medical travel by nationalist groups [76][77][78][79]. The most alarming aspect of the finding is that problems and issues not related to medical services are damaging the image of the destination based on medical attributes. Safety and security, crime, and political unrest are issues that have damaged the tourism industry of many countries. Medical travel destinations should also keep in mind that the medical traveller could be more vulnerable to destination related risks as they reside in one place for long periods to complete the treatment. Medical travel destinations must take these issues seriously and find solutions to tackle them for the continuous arrival of medical travellers.
The findings also revealed that those medical travellers who perceived travel constraints also perceived the destination negatively based on medical and non-medical attributes. The study provides interesting findings on the travel constraints of medical travellers. The perceived travel constraints were measured based on stress and anxiety felt, difficulty to find companion, the arrangement of money, the arrangement of travel documents, and flight connectivity from the home country. As far as anxiety and stress are concerned, managers must identify the factors which lead to stress and anxiety among medical travellers. More positive information should be spread among potential medical travellers about the medical and support services that will lead to a reduction in stress Managers also need to identify the issues and problems faced by travellers and work on reducing and correcting those problems so that negative word of mouth can be reduced. For the issue of finding a companion, the destination should develop a supporting mechanism that minimises the need of having a companion. Kangas [15] found that finding a suitable companion to travel abroad was a difficult task for medical travellers. She found that the head of the household who was also the earning member of the family and main decision-maker could not accompany the patient due to work commitments and arrangements of money to finance the trip. Therefore, travellers need a companion to travel with them who can take necessary decisions if needed while the patient receives treatment [80]. Destinations must develop a mechanism that can reduce the need to take crucial decisions by the patient at the time of receiving treatment. Detailed communication before arrival between the hospital and the patients about the procedure, treatment, expenses, and other support services will reduce the need for taking spontaneous decisions at the destination.
Studies of medical travellers have found that the arrangement of medical travel documents and flight connectivity from many source countries is a major issue faced by medical travellers [3,10]. Due to the tedious process which requires a lot of paperwork, most medical travellers visit the destination countries on tourist visas which can sometimes complicate their treatment. A tourist visa allows visitors to stay for a shorter period (in some cases for 15 days only), and sometimes medical procedures force travellers to stay for longer periods. In this case, receiving a further extension of the visa gives can be troublesome, especially when patients are undergoing treatment. For example, the Government of India recently took measures to provide e-visa to health travellers from more than 150 countries [81,82]. However, this relaxation does not include those who want to travel to receive medical treatment. Flight connectivity from many source countries such as the sub-Saharan region is another issue to which governments must take care to attract more medical travellers from this region.
The findings suggest that medical travel destinations should also focus on the risk perceptions and travel constraints of medical travellers instead of just promoting medical services to international patients. It is true that the main criterion for selecting a destination by foreign patients is the quality and cost of medical services. However, at present, many countries are developing their healthcare infrastructure to attract foreign patients by providing quality medical care at low cost and the criterion to choose a destination will include factors such as vulnerability to different kind of risks and ease of travel. Destination marketing managers should develop their marketing strategies that provide solutions to risk perceptions and travel constraints of medical travellers. At the same time, destinations must develop policies and measures that reduce the possibility of getting harmed at the destination for medical travellers.

Conclusion, Limitations, and Future Recommendations
The study contributes to the body of knowledge in two ways. First, the study was performed on actual international medical travellers, as most of the previous quantitative empirical studies [19,20] were performed on potential medical travellers. Therefore, this study provides useful insight into the travel behaviour of actual medical travellers. Second, the study conceptualised and integrated different factors in a single framework that are crucial in the travel decision-making of medical travellers and empirically tested their relationships. Based on the path coefficient values, it was found that physical-health risk (β = −0.139, p < 0.10), service quality risk (β = −0.128, p < 0.05), and destination risk (β = −0.233, p < 0.05) had negative relationship with destination image based on medical attributes, whereas, service quality risk was negatively associated (β = −0.138, p < 0.10) with destination image based on non-medical attributes also. Travel constraints were negatively associated with both destination images based on medical and non-medical attributes. It was found that physical-health risk and destination risk have no relationship with destination image based on non-medical attributes. The empirical evidence of the study suggests that perceived risks and travel constraints of medical travellers negatively affect their perception about the destination on different aspects such as quality of medical services, quality of support services, culture, physicians' skills, medical facilities, etc. and influence their travel behaviour. The destination and hospitals providing services to foreign patients must understand the perceptions of medical travellers. The study suggests that destinations and hospitals should be aware of the types of risks and constraints medical travellers face so that their future behaviour can be understood better.
Although the study has considerable practical implications, the study has several limitations. The study was performed on sub-Saharan African medical travellers visiting India. Therefore, care should be given in generalising the study findings to medical travellers from other regions. The study used non-probability convenience sampling due to the lack of a sample frame which limits the generalisability of the study findings. The study sample was skewed toward male respondents. 73.5% of the study respondents were male which means that the findings of the study could be biased towards male medical travellers. Future studies could be performed on medical travellers travelling from other parts of the world, and the findings can be compared to develop a better understanding of the travel behaviour of medical travellers. As this study could not provide equal representation of female medical travellers, future studies can use a more balanced sample of male and female participants so that the results can be generalised for both genders. Future studies can be performed on the different decision-making stages of international medical travel to examine if information sources and other personal factors play a role in the perception of risks and constraints.  Table A1. Survey Items