3.4. Discussion
On the basis of previous findings from the literature, we derived 10 hypotheses regarding the influence of various factors on the protective behavior of equestrians, which we subsequently tested by means of a multiple regression analysis. Concerning the influence of gender, we hypothesized that female equestrians demonstrate more pronounced safety behavior compared to male equestrians because they are generally more risk-averse [
34] and also tend to be more at risk [
2,
5,
7,
17,
18,
20]. However, we detected no significant effect of gender. Possible causes for this finding may be that horse riding is not classified as hazardous by female equestrians as it is “just” a sport like any other with which they grew up, so they have gotten used to and thus displace the risk inherent to equestrianism. Also, it may be the case that the argument that protective helmets and other safety equipment do not look good plays a more important role for female equestrians and leads to a higher level of rejection [
26]. Or it might also be possible that the more risk-averse females do not practice horse-riding at all. If only less risk-averse women choose to practice horse riding, this might be a possible reason why there was no relationship found between gender and safety behavior. It might be that there is no or only a small relationship between gender-specific risk perception and horse-related safety behavior. A significant debate is underway as to whether the high share of women among equestrians may bias the findings that female riders represent a high-risk group for injury [
2,
7]. In their literature review on horse-related injuries, Hawson
et al. [
4] also found that the younger age groups were predominately female, while the share of men was higher among the older age groups. However, the smaller share of men among the younger age groups was still more likely to get seriously injured. The authors discussed that for certain age groups, especially younger patients, the age- and gender-related distribution of equestrian accidents seems to correspond to that of the basic riding population. Yet this does not seem to hold true for all age groups. For example, the fact that older males are more likely to suffer more serious injury than females in the same age group cannot be explained by the demographic distribution [
4]. However, as the share of male equestrians in the sample is quite low (about 5%), the high share of female participants may have somewhat distorted the results, although this is not very likely regarding the large sample size. Looking at the data in more detail, we could find at least some small differences between male and female equestrians. Male riders were older on average, were more often earning their money through riding, and less often rode the same horse but rather different horses. We further found male riders to practice eventing more frequently, while the share of female equestrians riding outdoors was higher compared to their male counterparts. Regarding the use of protective equipment, we found that male participants owned and used a riding helmet less frequently, but were more likely to own an airbag vest. Regarding the perception of horse-related risk, male riders perceived the general riding risks as more dangerous, whereas the female equestrians perceived the special riding risks as being more problematical. This shows that the relationship between gender and protective behavior seems to be highly complex and quite manifold, so that it might be difficult to detect a clear relationship here. It follows that there is a need for further research to clarify the underlying relationships between gender-related risk, risk perception, and safety behavior.
Regarding the influence of age, we hypothesized that younger equestrians in general (H2a) and children and teenagers in particular show more pronounced safety behavior (H2b), but the present results only confirmed the second hypothesis. The finding that young riders tend to be a high-risk group [
4,
5,
7,
18,
19] and wear helmets more often seems to be reflected in their safety behavior [
26]. However, whether this behavior is self-imposed and possibly influenced by a high risk perception on the part of the young riders or rather mandated by parents or trainers remains to be clarified. Also, the requirements for junior riders to wear helmets are more closely mandated by associations, which might also play a role in this context [
52]. The finding that older equestrians are less likely to wear protective gear could not be confirmed here [
26]. One reason for this could be that the relationship between age and safety behavior is not completely linear. Looking at the correlation between age and the safety behavior index confirms this assumption as we found no significant relationship between them. Although the safety behavior probably decreases with growing years of experience [
10,
26,
27], it might start to decrease—analogous to growing risk aversion when having a child—due to certain events in life such as the birth of a child, increasing domestic or job responsibilities, or getting older in general [
39]. A study on attitudes and behaviors towards helmet use revealed similar findings, showing that those were affected by perceived social responsibility and care not only for other riders, but also for relatives, families, and friends [
27]. In line with this, the results of the present study confirm that adult equestrians with children show a more pronounced safety behavior compared to those without children, proving Hypothesis 3.
Hypothesis 4a and 4b refer to the relationship between riding experience and safety behavior. Regarding the skill level, the results did not confirm that novice equestrians show a more pronounced safety behavior compared to more advanced or professional riders, although the results did confirm that, in terms of years of experience, the more experienced riders show a less pronounced safety behavior (see
Table 6). The finding that experienced riders make less of an effort to protect themselves may have several causes. Firstly, more experienced riders may already have had some probably minor accidents in the past and underestimate the risk for more serious injuries. Also, it could be that as experience increases the control over the horse will improve, which might in turn lead to a false sense of security. In line with the findings by Haigh and Thompson [
27], the perception of being able to control the horse and read equine behavior could favor the rejection of safety equipment. As shown by previous research, another possible reason could lie within the fact that some of the more experienced riders seem to relate the use of protective headgear with being an inexperienced rider [
26,
27] and therefore try to evade the use of protective equipment or other protective measures that might characterize them as newbies. In contrast to more experienced riders, the finding that novice riders do not see the necessity of wearing protective equipment might be influenced by the fact that they might not have experienced or observed a serious fall yet. However, in line with findings indicating that less experienced equestrians tend to—according to the higher risk they are exposed to—protect themselves more often by means of a safety helmet than experienced riders [
10], the supposed negative relationship between increasing experience and decreasing protective behavior could again be confirmed. It has to be noted that the classification of riding skills was based on the participants’ self-assessment, which can lead to inaccuracies. Due to the growing heterogeneity within equestrianism, it is increasingly difficult [
53], especially across different riding disciplines, to have a reliable skill classification. Hasler
et al. [
17] suggest that a comparable educational level-injury risk index that tracks the true improvement in skills could be a more reliable measure of the relationship between experience and safety behavior. Furthermore, it is possible that not all advanced equestrians reject wearing safety equipment and prejudices might be getting slowly removed. Yet, both novice riders—due to the high number of accidents [
4,
7,
10,
15,
17,
29,
40]—and more experienced riders—due to the increased severity of sustained injuries and reduced willingness to exert safety behavior [
7,
10]—represent rather important target groups for sensitization regarding equestrian-related risks. There is a need for more detailed information regarding the risk awareness and attitudes of these target groups to better assess their behavior. One option to sensitize these target groups would be to use an experienced rider giving a testimonial who serves for both novice as well as advanced riders as an inspiring model giving advice on the safe handling of horses through seminars or courses. Sports celebrities are widely used in classical advertising with the aim of improving awareness and recall, driving sales, and influencing behavior. However, it must be noted that the particular celebrity has to comply with certain conditions and has, for example, to match the product or the topic to be effective [
54,
55]. More specifically, previous research results have shown that the use of celebrities within public health campaigns can be able to influence health-related attitudes, beliefs and risk behavior and seems to be a promising possibility within the present context [
56,
57].
Relating to horse accident experience, the high share of each of the 98% of equestrians who stated they had already experienced or witnessed a horse-related accident is consistent with the high number of equestrian injuries found in the literature [
9,
10,
11,
14]. The results of the multiple regression analysis confirmed Hypothesis 5a, which assumed a positive influence of the severity of a personally experienced accident on the individual’s protective behavior. However, we found no significant influence for the severity of an observed accident (H5b). It is plausible that accidents that are experienced firsthand arouse a greater awareness of horse-related risks. If the accident has only been experienced indirectly by observation, the personal distance towards the risk seems to be larger and the corresponding risk might be more easily dismissed. Yet this differs from O’Neil’s [
41] findings that the injury of alpine ski racers did have a psychological impact on the respective teammates. One reason for this could be that individuals might think they are safe and believe that misfortunes only ever happen to others. However, witnessing traumatic injuries can distort such beliefs. To cope with such traumatic events, people engage in different coping strategies [
58]. One of these strategies might consist of mental distancing from the injury. In line with this, a study on helmet use showed that it is being frequently argued by equestrians that one could do without safety equipment as the risk is perceived to be controllable [
27]. Such a false belief can be more easily maintained in the case of a witnessed accident as the riders themselves have not lost control and can talk themselves into believing that they might have reacted differently. The extent to which a rider can maintain this false belief might further depend on the observed severity of the injury. However, this is only speculation and it would be interesting to explore whether there is a certain degree of injury necessary to realize a change in safety behavior and what additional factors might play a role regarding the influence of observed accidents. As research in this field is said to focus too heavily on the effects on the direct victim and has frequently neglected the potential effects of serious injuries on the witness, it would be interesting to take a deeper look at this phenomenon within equestrianism and other kinds of sports [
58].
Previous research has also found that equestrians who practice one of the more competitive-oriented English-style riding disciplines wear helmets more frequently [
7,
14,
26,
43]. The present results confirmed that show-jumpers (H6a), eventers (H6a), and dressage riders (H6b) show more pronounced protective behavior. Also, other studies have showed that the less frequent use of helmets, which is characteristic of Western riding [
7,
14,
26] and other more leisure-oriented disciplines, has [
7]—as proposed in hypothesis 6c—a negative effect on the overall protective behavior. In contrast to the riding of gaited horses, which also belongs to the more leisure-oriented riding disciplines, we found no significant relationship. The present results could only partially confirm Hypothesis 6d, which assumes that more leisure-oriented disciplines put less effort in safety behavior. Regarding the relationship between protective behavior and riding outdoors, which several studies consider as one of the most dangerous pastimes on horseback and which often includes a high share of leisure-oriented equestrians of different disciplines, again no significant relationship could be found (H6d) [
6,
14,
19]. It is becoming clear that there are huge variations between the several leisure-oriented riding disciplines and their respective safety behavior. Such information would be helpful for horse sport associations to identify and communicate with more vulnerable groups that have a greater need for safety education. Western riders represent such a vulnerable group, as they often refuse to wear protective helmets as these are not considered appropriate Western-style equipment. In recent years, some producers have tried to develop Western-style protective helmets, but they were not successful on the market. Whether this was due to the look of the helmet, its wearing comfort, or other reasons remains unknown. For the Western riding associations, this implies that it is necessary to work on the development of protective gear that is better accepted by the Western riding culture as well as education about horse-related risk and the advantages and effectiveness of protective gear.
Hypothesis 7 assumed a significant relationship with the breed of the horse. As expected, we could observe a positive relationship between the riding of sport horses such as warmblood or thoroughbred horses and more pronounced safety behavior. It is unclear whether the observed positive relationship between riding of sport horses such as warmblood or thoroughbred horses and more pronounced safety behavior is due to a perception that these breeds are associated with a greater degree of unpredictable behavior and higher risk [
44,
45].
In line with the findings from the literature, the present results indicate that the influence of social groups and peer groups can positively influence helmet use [
26,
27]. Yet, the present study only looked at the influence of one social group, namely other horse owners and riding students from the stable. To keep the number of influencing factors manageable, we looked first at the more general group of other horse owners and riding pupils from the stable. That specific group is likely to have a close and horse-relevant contact with an equestrian and hence also the possibility to exert a strong influence on the latter. In view of the strength of the respective impact, this variable constitutes a quite important influence factor, as it exerts the second highest influence overall and provides a valuable starting point for the promotion of safety behavior. Here, especially trainers, stable managers, and horse sport associations are asked to inform their pupils, members, and clients about safety aspects concerning equestrianism, to reduce safety-related prejudices and to establish a positive security culture among the riders in a stable. As we did not further differentiate the group of horse owners and riding pupils from the stable, it would be useful to ask which individuals and subgroups, such as trainers or equestrian idols and also non-riding friends and family members, influence safety behavior at all and which of these groups exert a particularly high impact. Such knowledge could provide further starting points to enhance preventive behavior.
From all the factors examined within the multiple regression analysis, we identified the attitude towards protective gear as the most influential factor. As expected, a positive attitude exercised a positive influence (H9a) and a negative attitude exercised a negative influence (H9b) on protective behavior. The positive attitude exerted the highest impact overall and the negative attitude had the third-largest influence. The attitude towards safety equipment seems to represent a key aspect when trying to increase the use of safety products. Producers of safety equipment and horse sport associations should continue to try to find out more about the underlying reasons for these attitudes to identify potential enablers and barriers. Deficient design, untraditional appearance, lack of comfort, and doubts about the effectiveness of safety products seem to be major reasons for rejection. Therefore, the research into this area must be continued, especially on safety vests as their effectiveness to reduce horse-related injuries has still to be confirmed and has already been questioned within other sports [
2]. Furthermore, the design and comfort of safety equipment might be other important aspects to look at. In the case of helmets, the design can contribute decisively to the decision to wear safety equipment [
26]. Potentially, if the producers of safety equipment would work on the comfort and look of the product, it might turn into a rather desirable fashion item that riders would more often voluntarily use.
Finally, the last hypothesis expected both a positive influence of risk perception in general (H10a) and horse-related risk perception in particular (H10b) on protective behavior. However, the present results only confirmed the second hypothesis. Surprisingly, we detected a small but significant negative relationship. We measured risk perception in general as the personal risk perception of basic and high risk situations. The basic situation comprised moderate health risks such as having too much stress or insufficient sleep or exercise. The high-risk situations included activities such as extreme sports and motor sports. The reasons for the observed negative relationship with protective behavior are not clear so far and a more detailed examination of the connection between risk perception in general and horse-related risk perception in particular is necessary. Perhaps the phenomenon of risk suppression is a possible explanation. In this sense, more risk-averse people in general, especially when they have finally decided to participate in a high-risk sport or hobby such as equestrianism, might willingly suppress the associated risk, which might be considered quite harmless compared to other high-risk sports. Certain findings in the recent literature partially confirm this phenomenon [
26,
27], which shows that equestrians generally state that they believe they can control horse-related risk. Future research projects need to scrutinize this assumption. Perception of the level of danger associated with the equestrian activity in general and specific riding situations positively influenced the protective behavior patterns reported by equestrians. This relationship engenders a responsibility of horse sport associations to educate their members about the inherent horse-related risks to produce more safety-oriented behavior in equestrians.
3.5. Limitations
As already discussed in the sample description (see
Section 3.1), it is unclear whether the sample is representative so the generalizability of the present study may be limited. The present study is subject to self-selection bias in that certain types of respondents participated in the survey. Those with a high interest in horse-related safety may be overrepresented. Therefore, the extent to which the transfer of the results of the present study to the German equestrian population may be limited. However, given the high number of participants and the finding that sociodemographic variables did not exert a strong influence, it is likely that these results provide an important first approximation in this area of study.
The study’s methodological limitations include selected method of analysis and the calculation of the index to measure safety behavior. Although multiple regression analysis is a method able to identify the relationship among two or more variables, this does not automatically imply that this relationship is also causal [
59]. However, based on logical considerations, this affects only a small number of variables, such as show-jumping. For instance, it could be that some equestrians might only dare to jump at all because they practice very pronounced safety behavior. Such a relationship could further be linked to the phenomenon of risk compensation, in connection with which it is being discussed whether the wearing of protective gear can also exert the opposite effect on safety behavior, such as the use of protective equipment like helmets or safety vests, giving a false sense of security and promoting risk-prone behavior instead of reducing it [
2,
7,
60]. Another methodological limitation concerns the dependent variable. The safety behavior index mainly covered the use of protective equipment in different riding situations, which might imply that factors concerning protective gear exert a disproportionally high impact. Since several studies showed that not only horse riding itself is dangerous, but also simply handling the horse can lead to serious injuries, such as trampling, being kicked, or being bitten [
5,
8], it would be interesting to look at the specific protective behavior when handling a horse and compare it to the behavior when riding.
Moreover, it has to be noted that it is difficult to judge the quality of protective gear as it might be useful in the case of serious injury but might not avoid dangerous situations overall [
4,
14]. It may even be possible, as already discussed above, that some kind of risk compensation is at work such that the wearing of protective gear can also result in riskier behavior [
2,
7,
60]. Future research regarding the use and effectiveness of protective gear but also the impact of additional measures that can reduce horse-related risk could provide additional useful information. As already proposed in the literature, an important additional safety measure constitutes the improvement of the predictability of horses through better education and understanding of equine learning and behavior patterns, building on recent findings from research on horse ethology and equitation science, as it is a commonly cited cause of human injury [
4]. In this context, improving riders’ competence in physical skills such as fitness, balance, the proper application of aids, and falling techniques should make them more resilient to injury and falls; establishing clear rules and legislation requiring the mandatory wearing of approved safety gear and increasing general awareness of horse-related risk for both individuals and the general public are further possible measures to reduce horse-related risks. The establishment of good practices and a comprehensive safety management within stables will ensure a safe environment [
6,
8,
12,
19].