Posture Retention Devices for Older Adults Undergoing Horseback Riding Therapy

Horseback riding therapy is an activity that affects the user both mentally and physically through engagement with a horse [1]. In recent years, the rapid aging of the population has become a global issue, and several measures are being considered to extend healthy life expectancy. Studies on the effects of horseback riding therapy have reported improvements in trunk stability, muscle strength, balance ability, and muscle tone in patients with cerebral palsy [2-4].

In previous studies on older participants in the field of riding therapy, Diniz et al. [5] reported improvements in dynamic balance and flexibility, and Mello et al. [6] reported that equine therapy promoted a decrease in blood pressure in older participants with hypertension. Various changes occur in humans as they age [7]. One of the most common changes is postural deterioration, generally observed in older patients with kyphosis of the spine. Older adults often develop muscular and spinal deformities due to aging. Because of these postural changes, older adults tend to have poor posture on horseback and experience difficulty maintaining the correct riding position.

Several studies have used motion analysis systems and muscle activity measurement analyses to verify the effectiveness of horseback riding therapy [4,8]. However, few studies have measured and analyzed humans and horses simultaneously on horse tracks. One issue with this is the difficulty of conducting experiments by installing sensors on horses while controlling them in the indoor riding arena. Brandão et al. [8] reported differences in muscle activity in the trunk and lower limbs, depending on the horse-riding equipment used in riding therapy, and emphasized the importance of using appropriate horse-riding equipment. Nakamura et al. [9] also studied devices tailored to patients with physical disabilities who often become unstable on horseback and found that the use of devices promoted appropriate posture in the users. Although there is growing interest in the types of devices and harnesses available for use during horseback riding, few studies have focused on developing equipment to improve the posture of older riders. Therefore, this study aimed to verify the effectiveness of a prototype riding device and identify any areas for improvement by comparing usability survey results and quantitative and qualitative postural analysis results of skilled and unskilled older adults with riding experience.

Participants

The participants included four skilled older adults (two men and two women), with a mean age of 68.75 (±2.17) years, who had experience in horseback riding and were independent in their activities of daily living. The older adults with riding experience had ridden horses for at least 5 years and could control the gait without assistance.

The unskilled group with little riding experience comprised four older adults (two men and two women) with a mean age of 81.5 (±4.63) years who also were independent in their activities of daily living. They had riding experience of fewer than 20 instances within 3 years and could not control the horse’s gait by themselves. As risk management measures, those with systolic blood pressure of 150mmHg and diastolic blood pressure of 100mmHg or higher before riding were excluded; all the participants were required to wear a helmet and gloves when riding. This study was approved by the Ethical Review Committee of the University of Tokyo (Review No.: 20-209). Further, the participants granted informed written consent after being briefed regarding the purpose and methods of the study and the risk of falling from the horse.

Experimental device and horse/harness

The prototype “wedge-shaped horseback riding device”: The material of the wedge-shaped horseback riding device (hereinafter, the “wedge”) was fabricated from rigid polyurethane foam (density: 35kg/m3). The left and right parts of Figure 1 show the wedge-shaped horseback riding device in the top and side views, respectively. The edges were filleted, a coating was applied, and the surface was treated with a special coating that followed the expansion and contraction of the rubber. As shown in Figure 2, the cut side of the wedge was positioned at the front and placed between the saddle and the rider’s buttocks.

Figure 1:The wedge-shaped horseback riding device in the above (left) and side (right) views.

Figure 2:The wedge-shaped horseback riding device installation method.

Horse/harness: One 11-year-old (150cm in height) Criollo gelding, usually used for therapeutic riding and that had been trained and acclimatized, was used as the therapy animal. The horse was equipped with a saddle with a handle, safety stirrups, a head strap, and a mute with a puller; the horse was handled by a staff member familiar with horses (Figure 3).

Figure 3:Horses and harnesses.

Experimental setup: Four video cameras (Panasonic V360MS digital high-definition video camera), each fixed to a tripod, were used for video recording of the therapy sessions. The Dartfish 9.0 video analysis software (Dartfish Japan) was used to analyze the captured video images (sampling frequency: 60Hz).

Experimental procedure

The experiment was conducted in an indoor riding arena that was unaffected by wind and rain. After body temperature measurement and hand disinfection, the participants had their vital signs checked for health risk management. Then, they wore helmets and gloves for safety. Landmarks on the participants’ bodies were marked with ping-pong balls to record the joint angles using the video analysis software (Figure 4). The landmarks were located on the helmet just above the external occipital protuberance. Other landmarks included the spinous process of the seventh cervical vertebra, the spinous process of the seventh thoracic vertebra, the spinous process of the fourth lumbar vertebra, greater trochanter, lateral femoral epicondyle, and lateral malleolus. To minimize the risk of injury, participants rode horses on a horse-mounting ramp with the support of a riding therapy instructor.

Figure 4:Location of landmarks on a participant’s body.

As shown in Figure 5, a video camera was used to capture the posture in straight-line rides from the sagittal plane and the posture during curved rides from the forehead plane, and the joint motions were later analyzed using motion analysis software. The order of each joint angle value was randomly determined for each participant in the following four conditions: no wedge (0°), wedge at 3°, wedge at 5°, and wedge at 7°, as shown in Figure 1. The participants dismounted directly on the ground with the support of the riding therapy instructor. Additionally, the joint angle motion in the present study was based on the joint range of motion testing method [10,11].

Figure 5:The angles of the neck, spine, and knee joints by motion analysis in the sagittal and frontal planes.

On a different day after the experiment, the riding therapy instructor was asked to evaluate and score the riding posture using still images from the video that were confirmed to have the maximum value (Appendix 1). One point denoted “bad posture,” two points denoted “undecided,” and three points denoted “good posture.” A good riding posture when visualized from the sagittal plane was defined as follows: neck flexion and extension in midposition, shoulder joint in slight flexion, elbow joint in slight flexion, forearm in rotation, wrist in slight dorsiflexion, spinal column in flexion and extension in mid-position, pelvis in anteroposterior tilt in mid-position, hip joint in flexion, knee joint in flexion, and ankle joint in slight dorsiflexion, with the auricularis aligned to the ground in a straight line passing through the acromion, adductor pollicis, and posterior part of the abductor digitorum. In the frontal plane, the movements of the neck in mid-lateral flexion, shoulder joint in mild internal rotation, elbow joint in mild flexion, forearm in rotation, wrist in mild dorsiflexion, spine in mid-lateral flexion, pelvis in mid-tilted position, and hip flexion external rotation were examined. The rider’s spine was evaluated to determine whether it was perpendicular to the horse’s thoracic spine and whether the acromion and iliac crest were symmetrical with respect to the vertical line from the top of the head to the ground.

Analysis

For both the skilled and unskilled groups, each joint angle value obtained in the experiment was averaged for a total of four conditions: no wedge (0°) and wedge device at 3°, 5°, and 7°. The maximum value was checked twice each, once when the horse’s left forelimb was grounded in the straight-line ride and once when the rider showed the greatest outward swing during a curved ride. Each joint angle was measured and analyzed as follows: the neck angle connecting the external occipital ridge, seventh cervical spinous process, and seventh thoracic spinous process; the spinal column angle connecting the seventh cervical spinous process, seventh thoracic spinous process, and the fourth lumbar spinous process; and the knee angle connecting the greater trochanter, lateral femoral epicondyle, and external capsule. When the joint angles were analyzed from the sagittal plane, the neck angle, spinal column angle, and knee angle were analyzed, and when the joint angles were analyzed during a curve, the neck and spinal column angles were analyzed from the anterior forehead plane. In addition, the postures of both groups were analyzed and scored by the instructor twice (straight line and curve) for the four wedge conditions, and the mean of the scores was calculated.

Analysis of Variance (ANOVA) was employed to identify significant differences in posture scores and joint angles, such as the neck, spinal column, and knee angles, and to examine the main effects and interactions of conditions, including wedge angle (0°, 3°, 5°, and 7°), riding experience (skilled and unskilled group), and line types (straight and curved; excluding the case of knee angle) on these responses. Prior to performing the ANOVA, an aligned rank transform [12,13] was performed on the responses whose distribution did not satisfy the normality. Then, the significance of the main effects and interactions was determined using a posthoc multiple comparison method based on the least squares mean and Tukey’s multiple adjustment [14,15]. Statistical and marginal significance were set at p<0.05 and p<0.10, respectively. The R (version 4.4.0) was used for the statistical analysis.

Neck angle

Figure 6 & 7 show the mean peak neck angles with Standard Deviations (SD) for each participant in the skilled and unskilled older adults during straight-line and curved rides, respectively. Significant main effects of riding experience (F (1,48)=4.70, p=0.035) and line type (F (1,48)=10.1, p=0.003) were observed. There were no significant main effects of wedge angle (F (1,48)=0.517, p=0.67). A significant interaction between riding experience and line types was also observed (F (1,48)=45.9, p<0.001). Multiple comparisons revealed that the neck angle of the unskilled older adults was significantly greater and smaller than that of the skilled older adults in straight (t (48)=4.86, p<0.001) and curved (t (48)=-4.18, p<0.001) lines, respectively.

Figure 6:Maximum neck angle in a straight line for skilled and unskilled participants.

Figure 7:Maximum neck angle in a curve line for skilled and unskilled participants.

Spine angle

Figure 8 & 9 show the average peak spine angles for the skilled and unskilled older adults during the rides in straight and curved lines, respectively. Similar to the neck angle, riding experiences and line types had significant main effects on the spine angle (riding experience: F (1,48)=7.44, p=0.009, line type: F (1,48)=94.9, p<0.001) and an insignificant main effect of the wedge angle (F (1,48)=0.544, p=0.66). There were no significant interactions among the wedge angle, riding experiences, and line types. Multiple comparisons showed that the spine angle of the skilled participants performed significantly lower than that of the unskilled participants (t (48)=-2.73, p=0.009), and the spine angle in a straight line was significantly smaller than that in a curve line (t (48)=9.74, p<0.001).

Figure 8:Maximum spine angle in a straight line for skilled and unskilled participants.

Figure 9:Maximum spine angle in a curve line for skilled and unskilled participants.

Knee angle

Figure 10 shows the mean and SD of peak knee angles. According to the ANOVA result, riding experiences had a significant main effect on the knee angle (F (1,24)=4.96, p=0.035), but the wedge angle did not (F (1,24)=0.28, p=0.28). There was no significant interaction between the wedge angle and riding experiences. The results of multiple comparisons showed that the unskilled participants had a significantly larger peak knee angle than their skilled counterparts (t (24)=2.23, p=0.036).

Figure 10:Maximum spine angle in a curve line for skilled and unskilled participants.

Soring of riding posture by riding therapy instructor

Figure 11 shows the postural evaluation score for skilled and unskilled participants in the straight and curved lines. The ANOVA results indicated significant main effects of the wedge angle (F (3,48)=6.78, p< 0.001) and riding experiences (F (1,48)=5.39, p=0.025). There was an interaction with marginal significance between the riding experience and the line type (F (1,48)=2.91, p=0.094). Multiple comparisons illustrated that the wedge angles of 3°, 5°, and 7° all had significantly higher postural evaluation scores than the 0° condition (3°: t (48)=2.88, p=0.029; 5°: t (48)=3.81, p=0.002; and 7°: t (48)=3.97, p=0.001). Moreover, skilled participants scored significantly higher than the unskilled ones (t (48)=2.32, p=0.025). Moreover, in the straight-line case, such scores of the skilled participants were significantly larger than those of the unskilled ones (t (48)=2.82, p=0.034).

Figure 11:Postural evaluation score for skilled and unskilled participants in the straight and curved lines.

According to the qualitative postural evaluation by the riding therapy instructor, the neck angles of the skilled participants were in extension with or without a wedge, while those of the unskilled participants were often in flexion. The riding therapy instructor also commented that both skilled and unskilled participants showed more anterior pelvic tilt when a wedge was used. Further, the knees of unskilled participants were extended, and their feet were in a forward position compared to those of the skilled participants. Finally, the riding therapy instructor noted that the skilled participants became closer to a good limb position when using the wedge at ≥7°, while the unskilled ones were closer to a good limb position when using the wedge at ≥5°. Both the skilled and unskilled riders were evaluated as having poor limb position when using the wedge at 0°.

Posture improvement

In straight-line rides, the skilled participants’ neck angle was in extension, regardless of the presence or absence of the wedge (Figure 6). Meanwhile, the unskilled participants’ neck angle was in the flexed position with or without the wedge (Figure 6), and the average peak angle of the unskilled participants was 12-14° greater than that of the skilled participants. Further, in the straightline rides, the average peak neck angle was significantly smaller in the skilled group than in the unskilled group, possibly because of the difference in riding skill. However, we believe that skilled riders checked the horse’s direction of travel, looked forward, and maintained their neck in the mid-position. Riders should always check the front of the horse, observe the surroundings, and pay attention to any danger in terms of the neck angle while riding. Therefore, unskilled riders, or rather unskilled older adults, should keep the neck in the mid-position, and by flexing the neck and looking downward, it is easy to assume a posture with trunk flexion and a neck hump [16].

During the curves, the skilled participants had a significantly larger neck angle than the unskilled participants (Figure 7). The reason for this is thought to be that, compared to unskilled older adults, skilled older adults try to control their posture on curves mainly by neck movements. The spinal column was flexed in both the skilled and unskilled participants, regardless of the presence or absence of the wedge (Figure 8 & 9). The unskilled participants had greater spine flexion than the skilled participants. When the 5° wedge was used, the unskilled participants were closer to the mid-position of flexion and extension of the spine than the skilled participants. The average peak angle of the spine in straight-line rides was 3-6° greater in the skilled group than in the unskilled group, regardless of whether the wedge was used. This result may be because the skilled riders were able to maintain their spinal column in an intermediate flexion-extension position without the use of a wedge, while the unskilled riders were able to achieve extension of the spinal column by correcting the pelvis from a posterior tilt to an intermediate position using the wedge. Particularly, the mean peak spine angle was significantly larger in the curve rides than in the straight-line rides for both skilled and unskilled participants. Compared to the unskilled participants, the skilled ones were not in a slouching posture, but were close to a good posture in which both left and right sitting bones were loaded in a balanced manner. Meanwhile, the unskilled riders had a smaller angle between the neck and spinal column than the skilled riders because they had less riding experience, making it difficult for them to predict the horse’s sway and posture in the direction of travel.

The knee angle during straight-line rides was found to be significantly larger in the unskilled group than in the skilled group (Figure 10). In the skilled participants, the vertical line from the auricle to the ground passed through the acromion, adductor pollicis, and posterior part of the external capsule when viewed from the sagittal plane, and the riding therapy instructor generally evaluated their posture better when the wedge angle was 5°, 7°, and 3° in order than when it was 0°. The joint angles used as the standard in the present study were the joint angles when an experienced horse rider was riding a stationary horse. Therefore, when viewed from the sagittal plane, a cervical angle of 160°, a thoracic angle of 170°, and a knee joint angle of 60° were the standards for good posture. A cervical angle of 180° and a thoracic angle of 180° measured from the frontal plane were considered to be the standards for good posture.

Riding therapy instructor posture evaluation

The riding therapy instructor scored the skilled participants significantly higher than the unskilled participants (Figure 11). In straight-line rides, the scores were significantly higher for the skilled participants, implying that the riding posture of the skilled participants was better than that of the unskilled participants. Furthermore, the scores of the skilled participants were significantly higher using the 3°, 5°, and 7° wedges than using the 0° wedge during curves. These results indicated that even for skilled riders, the posture was improved when the wedge device was used compared to when it was not used. The reason the scores of the unskilled riders were significantly lower than those of the skilled ones was that the unskilled riders had fewer body parts oriented to a good riding posture than the skilled riders, considering pelvic tilt, kyphosis, anterior foot position, and greater lateral bending of the spinal column.

Regarding the effects of horse harnesses, Lage et al. [8] improved the existing stirrups and investigated the effects of the presence or absence of stirrups on muscle activities in the trunk and lower limbs of children with different diseases [8]. They reported that the usage of the harness could affect the muscle activity of the trunk and lower extremities. However, the joint movements of the trunk and lower limbs differ greatly for each muscle group, and there is no mention of maintaining or improving posture with the harness. Our research can be novel in that we developed a prototype of an orthotic device to assist older adults in harnessing and analyzed its effectiveness by focusing on joint motion. Several studies have reported that horseback riding provides sensory and motor activation, establishes anticipatory and compensatory coordination strategies, and improves balance, muscle strength, muscle flexibility, and motor coordination, even in the presence of postural impairments [17-19]. Our achievement should be significant because it extends the scope of improvement in horseback riding therapy, as we have developed a device to assist in improving posture in older adults.

Limitations

The limitations of this study include the small number of participants. Due to the COVID-19 pandemic and the complex, timeconsuming experimental protocol, we were unable to recruit a larger sample. Despite this, we were able to confirm differences between the skilled and unskilled horseback riders and demonstrate that the proposed device improved overall posture in both groups. To better understand detailed effects, such as those associated with continued use and specific posture improvements, future studies should include a larger number of participants. There is also a limitation regarding the generalizability of our findings. Particularly, the manufacturing methods and shapes of horseback riding equipment vary by country and culture, with many established as traditional practices. Therefore, to demonstrate the effectiveness of the wedge device developed in the present study in other countries and cultures, it is essential to consider the relationship between the wedge and the traditional equipment. In summary, the results obtained from the assistive wedge-shaped device for older adults to maintain proper limb positioning suggest that the device could improve the riding posture of older adults. Moreover, we obtained preliminary results that need to be confirmed with several experiments to determine the detailed effectiveness of the device in horseback riding therapy.

In this study, we developed an assistive wedge-shaped device for older adults to maintain proper limb positioning while riding and then evaluated the angles, use, and effectiveness of the device in horseback riding by skilled and unskilled older adults. The results of postural analysis of neck, spine, and knee angles indicated that the use of the device could promote proper limb positioning. Qualitative postural analysis, including riding therapy instructor posture evaluations, also suggested that the device could improve the riding posture of older adults. Our future work includes more trials to explore the detailed effectiveness of the device in horseback riding therapy and the effect of posture improvement through longterm use of the system.

The following is a list of items that the riding therapy instructor considered when scoring the participant’s limb positioning in horseback riding.
a. When the horse is at a standstill, the vertical line from the apex of the ear to the ground passes behind the acromion, adductor pollicis, and external genu when the riding posture is viewed from the sagittal plane. When viewed from the frontal plane, the acromion and iliac crest are symmetrical to the vertical line from the crown of the head to the ground.
b. Even during curves, good limb positioning is essentially the same as in a straight-line ride. A slight lateral flexion and rotation of the neck and spinal column in the direction of travel may occur. The important point is that the rider’s spine is perpendicular to the horse’s spine (thoracic vertebrae) (i.e., the rider’s left and right spinal columns are balanced).

The other detailed evaluation criteria for straight-line and curve horseback riding were as follows.

A. Criteria for straight line-riding:

i. Is the rider’s gaze downward? Is the neck in a flexed position?
ii. Is the rider’s chest open? Is the shoulder joint not internally rotated?
iii. Is the rider’s pelvis tilted backward and the spinal column not kyphotic? Is the pelvis and spinal column in the middle position?
iv. Is the rider relaxed? Is the rider gripping the handles tightly?
v. Is the rider’s ear, shoulder, and heel in a straight-line ride? Is the line connecting the auricle, acromion, and posterior surface of the calcaneus perpendicular to the ground?
vi. Is the rider’s head directly above the spinal column? Not too high up or too low down?
vii. Are the rider’s upper arms perpendicular to the ground?
viii. Are the rider’s feet forward? Are their feet not positioned in front of the girth?
ix. Are the rider’s heels down? Is the ankle in mild dorsiflexion?
x. Are the rider’s feet forward? Are their feet not positioned in front of the girth?
xi. Are the rider’s heels down? Is the ankle in mild dorsiflexion)?

B. Criteria for curve riding:

i. Is the rider’s gaze downward? Is the neck in a flexed position?
ii. Is the rider’s posterior directly above the horse’s spinal column? Particularly, is their posterior not offset to the left or right relative to the saddle?
iii. Are the feet not significantly displaced vertically from side to side? Are there no critical differences in foot position?
iv. Is the rider’s spine and head able to resist centrifugal force?

1. Kayoko Yamamoto, Mitsuharu Inaki, Masao Yamane (2005) Research trends on horseback riding therapy (horseback riding for people with disabilities) in Japan. Seinan Jogakuin University Bulletin 9: 66-71.
2. Haehl Victoria, Giuliani Carol, Lewis Cynthia (1999) Influence of hippotherapy on the kinematics and functional performance of two children with cerebral palsy. Pediatric Physical Therapy 11(2): 89-101.
3. Bertoti DB (1988) Effect of therapeutic horseback riding on posture in children with cerebral palsy. Physical Therapy 68(10): 1505-1512.
4. Tim L Shurtleff, John W Standeven, Jack R Engsberg (2009) Changes in dynamic trunk/head stability and functional reach after hippotherapy. Archives of Physical Medicine and Rehabilitation 90(7): 1185-1195.
5. Diniz LH, Corrêa de Mello E, Ribeiro MF, Brandão Lage J, Bevilacqua Júnior DE, et al. (2020) Impact of hippotherapy for balance improvement and flexibility in elderly people. Journal of Bodywork & Movement Therapies 24(2): 92-97.
6. Edneia Correa Mello, Lunna Honorato Diniz, Janaine Brandão Lage, Valéria Silva Monteiro, Fernandes Ribeiro M, et al. (2021) Impact of equine assisted therapy on the cardiovascular parameters of the elderly. ABCS Health Sciences 46: 1-8.
7. Takeshima N, Hayashi R (2020) Age-associated decline of balance ability and the effects of balance exercise in older adults. Japanese Journal of Sports and Health Science 42: 1-15.
8. Brandao Lage J, Fernandes Ribeiro M, Vicente de Paula AT, César Rosa R, Abadio Ferreira A, et al. (2020) Effect of horse riding equipment in activity of trunk and lower limb muscles in equine-assisted therapy. Acta Scientiarum Health Sciences 42(1).
9. Yuji Nakamura, Sasaki Ayaka, Nakajima Sonomi, Sengoku Yasuhito (2021) Kinematic effects of assistive devices in horseback riding therapy for children and people with movement disorders. Journal of Animal-Assisted Education and Therapy 12 (1&2): 1-8.
10. Nancy Berryman Reese, William D Bandy (2016) Joint range of motion and muscle length testing-E-book, (4^th edn), Elsevier Health Sciences, Amsterdam, Netherlands.
11. M Clarkson, Hurabielle Jacques (2005) Joint motion and function assessment: A research-based practical guide. (1^st edn), Lippincott Williams & Wilkins, Pennsylvania, USA, pp. 1-366.
12. Jacob O Wobbrock, Leah Findlater, Darren Gergle, James J Higgins (2011) The aligned rank transform for nonparametric factorial analyses using only anova procedures. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, Vancouver BC, Canada, pp. 143-146.
13. Lisa A Elkin, Matthew Kay, James J Higgins, Jacob O Wobbrock (2021) An aligned rank transform procedure for multifactor contrast tests. Proceedings of the 34^th Annual ACM Symposium on User Interface Software and Technology, Virtual Event USA, pp. 754-768.
14. Lenth RV (2016) Least-squares means: The R package lsmeans. Journal of Statistical Software 69: 1-33.
15. Russell V Lenth, Ben Bolker, Paul Buerkner, Iago Giné-Vázquez, Maxime Herve, et al. (2024) Package ‘emmeans’ R topics documented.
16. Kazunari Onuki, Hiroshi Watanabe, Masao Wada, Tadazhi Ogino, Makoto Maie, et al. (1996) Equestrian. (4^th edn), Japan Central Horse Racing Association Horse Racing School, Japan, p. 226.
17. Miroslav Janura, Christian Peham, Tereza Dvorakova, Milan Elfmark (2009) An assessment of the pressure distribution exerted by a rider on the back of a horse during hippotherapy. Human Movement Science 28(3): 387-393.
18. Deirdra Murphy, Linda Kahn-D'Angelo, James Gleason (2008) The effect of hippotherapy on functional outcomes for children with disabilities: A pilot study. Pediatric Physical Therapy 20(3): 264-270.
19. Ann Hammer, Ylva Nilsagård, Anette Forsberg, Helena Pepa, Elisabeth Skargren, et al. (2005) Evaluation of therapeutic riding (Sweden)/hippotherapy ( United States ). A single-subject experimental design study replicated in eleven patients with multiple sclerosis. Physiotherapy Theory and Practice 21(1): 51-77.