Next Article in Journal
Feline Responses to Increasing Inclusion of Natural Olive Extract in Liquid or Dry Palatant Formulations Applied to Kibble Diets
Previous Article in Journal
Beyond the Pet-Effect—Examining Bio-Psychosocial Aspects of Pet Ownership and Introducing the “Aunty Phenomenon”
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Pets
Volume 2
Issue 1
10.3390/pets2010012
pets-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Horses’ Cardiovascular Responses to Equine-Assisted Group Therapy Sessions with Children

by
Lena Kreuzer
1,
Anna Naber
2,
Roswitha Zink
2 and
Lisa Maria Glenk
1,3,*
1
Comparative Medicine, The Interuniversity Messerli Research Institute, University of Veterinary Medicine Vienna, Medical University Vienna, University Vienna, 1210 Vienna, Austria
2
E.motion Lichtblickhof, 1140 Vienna, Austria
3
Karl Landsteiner Research Institute for Neurochemistry, Neuropharmacology, Neurorehabilitation and Pain Treatment Mauer-Amstetten, 3362 Mauer-Amstetten, Austria
*
Author to whom correspondence should be addressed.
Pets 2025, 2(1), 12; https://doi.org/10.3390/pets2010012
Submission received: 25 January 2025 / Revised: 26 February 2025 / Accepted: 28 February 2025 / Published: 3 March 2025
Browse Figure
Versions Notes

Abstract

:
Children with psychosocial, developmental or physical impairments benefit from equine-assisted therapy (EAT) in multiple ways. However, to date, the animal perspective of such interventions has received comparatively less scientific dedication. Thus, heart rate (HR) and heart rate variability (HRV) of seven therapy horses that lived in an open stable environment and participated in therapeutic group sessions with children were monitored within 60 min prior to and within 90 min after EAT. Moreover, cardiovascular activity was compared to a control condition on a day without any EAT sessions. No significant differences in HR or HRV were found, neither before nor after EAT nor under the control condition. These findings do not give rise to any concern regarding horse welfare in the studied setting but cannot be generalized to a broader population of equines serving in EAT. Given the enormous heterogeneity in EAT, more in-depth research is warranted using behavioral and physiological indices of equine health and wellbeing.

1. Introduction

Equine-assisted therapy (EAT) has been shown to be effective in various populations of recipients [1,2,3,4,5]. Especially children with psychosocial, developmental or physical impairments benefit from therapeutic interventions that are facilitated by horses [6,7,8,9]. In recent years, research into the animal perspective of EAT has significantly advanced, however, due to the heterogenic nature of EAT programs (i.e., their therapeutic goal, basic profession of therapists, number and characteristics of recipients, protocol, duration, time course of the sessions and time intervals between sessions), the generalizability of scientific outcomes is still limited. Accordingly, more studies are needed to reflect the diversity and complexity of EAT by addressing not only recipient health-related outcomes but also the impact on the animals involved.
In mammalian species, including equids, exposure to challenge stimulates an immediate activation of the sympathetic–adrenal–medullary (SAM) axis, which results in a discharge of noradrenaline along sympathetic fibers [10]. Since the secretion of noradrenaline and adrenaline directly impacts cardiovascular rhythmicity, SAM axis monitoring via non-invasive measurements is a suitable approach to capture arousal-related adaptations in animals [11]. Heart rate variability (HRV) analysis contains detailed information of the time span between two adjacent heart beats (=RR interval) and thereby mirrors the natural irregularities in heart rate patterns. Beat-to-beat variations originate from the interplay between the parasympathetic and sympathetic branches of the autonomic nervous system (ANS) [12]. Stressful experiences trigger sympathetic fibers, which in turn impact the heart to accelerate and decrease variations in time intervals between individual heartbeats. In contrast, parasympathetic activity decelerates the heart rate (HR) and results in higher variations in beat-to-beat intervals.
The standard deviation of all RR intervals (SDRR) represents the overall variability of the time intervals between single heart beats [13]. A higher SDRR therefore suggests a more efficient interaction of sympathetic and parasympathetic autonomic functioning. The amount of parasympathetic activity is associated with the root mean square of successive differences (RMSSD) of subsequent RR intervals. In spectral HRV analysis, high frequency (HF) is regulated by the breath and vagal tone whereas low frequency (LF) is facilitated by parasympathetic as well as sympathetic function (Baumert et al., as cited in [13]). The frequency domain indicator LF/HF reflects the sympathetic and parasympathetic balance (Eckberg, as cited in [13]), while the standard deviation of the diameter of the scatterplot (SD1) considers the short-term variability of individual heart beats and parallels parasympathetic tone [14]. Despite the benefits of SAM axis activity in enabling cardiovascular adaptions in order to cope with challenges, too frequent and prolonged autonomic activation causes exhaustion [10]. In fact, chronic stress experiences have detrimental effects for health and wellbeing. Research into the animal perspective on EAT should ideally be based on non-invasive scientific methodology that only minimally disrupts the human–animal interaction process [15]. As the ANS integrates physical, mental and environmental stimulation, assessment of cardiovascular functioning has been considered a suitable approach in equine science [16,17].
Previous studies using cardiovascular activity as welfare indicators in therapy horses during human–animal interaction sequences have proposed inconsistent findings. Arrazola and Merkies [18] reported increases in therapy horses’ HRs over the course of a 10-week EAT protocol which was accompanied by a decrease in socio-positive behaviors. In contrast, hippotherapy for disabled individuals did not affect equine HRs [19]. In a study comparing equine HR responses to people with and without a diagnosed post-traumatic stress disorder, higher HRs were found when horses interacted with traumatized individuals [20]. Higher equine HR was measured during the anticipatory and on horseback phase of EAT compared to performing groundwork [21]. Previous data indicated that the spectral HRV indicator LF/HF increased during the active working sequence compared to when the horses were prepared by the recipients [22]. Similarly, when horses were guided by recipients with an intellectual disability through an obstacle course including a slalom, crossing over a pole and balancing on a pedestal, their LF/HF increased compared to activities that were less demanding [23]. In free-ranging horses seasonal influences on HRV were found. The HR and RMSSD were lower during winter. An increase in activity, raining and humidity positively influenced the RMSSD whereas the ambient temperature negatively affected it. In free-ranging horses the parasympathetic activity was higher from 12 a.m. to 6 p.m. [24]. Therapy horses, in contrast, showed elevated values during the night and lower values during the day [25].
Thus, the aim of this study was to monitor the autonomous responsiveness of therapy horses related to their regular participation in EAT by assessing HR and HRV on days with (i.e., experimental condition) and without (i.e., control condition) scheduled group therapy sessions. If the EAT sessions substantially impact ANS functioning, a significant difference in the mean HR and parameters of HRV would be expected from pre- to post-EAT or would emerge between the two conditions.

2. Materials and Methods

2.1. Therapy Horses

Data were collected during regular activities of the EAT center Lichtblickhof, an organization offering animal-assisted therapy to children and adolescents dealing with grief, trauma, palliative diseases or disability. At Lichtblickhof, therapy horses live in a stable herd of 20 individuals in an appropriate environment, consisting of an open stable with unlimited access to paddocks and a riding hall nearby [23]. Access to hay, trees, water and enrichment elements was provided ad libitum. In sum, seven therapy horses of various breeds with a mean age of 17.7 ± 7.54 SD years and 8.8 ± 6.8 SD years of experience in EAT participated in the study (see Table 1).
At the time of the experiments, all animals were in good health, confirmed by veterinary screening (>2 times/year). Prior to the onset of the study, autonomic activity assessments were carried out for each participating horse. Horses underwent regular training sessions and were encouraged to communicate states of joy, arousal or when they require a break during participation in EAT-related activities via audible exhale communication [26]. All individuals were free from medications and displayed normal equine behavior.

2.2. Study Protocol

Each horse was measured once under the experimental condition (E) and under the control conditions (C). One EAT group session of 60 min served as the EAT session under the experimental condition. In the experimental condition, the HR and RR intervals were measured for one hour prior to EAT and one and a half hours after EAT, which took place between 2 p.m. and 6:15 p.m. No measurements were carried out during EAT to account for discretion during the therapeutic process. In the control condition, no EAT took place and horses remained in their herd during that equivalent time span. This schedule was chosen to interfere least with the ongoing EAT sessions at the EAT center. As indicated in Table 2, HR and RR intervals were recorded continuously 60 min before and 90 min after EAT but were divided into 30 min sequences for analysis. EAT only took place under E.

2.3. EAT Sessions

EAT sessions were carried out with small recipient groups of two to four children who had a mean age of 9.4 (±3.3 SD) years and did not show any motor impairment but exhibited impaired cognitive or social functioning. The group (children, therapist and horse) was familiar and met regularly once a week in the EAT center. A low intensity of movement for the horse, mainly standing or slow walking gait, characterized the performed EAT units. That way, the comparability between conditions could be ensured best as these were the most common observable activity patterns when horses roamed freely in the open stable. EAT sessions included preparation of the horses, groundwork and guided rides in the surrounding area of the association, relaxation exercises for the EAT recipient or free interactions (including stroking, talking to the horse and playing) in the riding hall. At least one other therapy horse was present in the riding arena during EAT sessions. On control days, the time of measurement was matched to the respective one under the experimental condition, but the horse remained in the open stable instead of performing EAT. The order of the conditions was randomized and scheduled so that half of the horses started with the experimental and the other half with the control condition. In general, no additional EAT services, riding or training activities took place on the days of data collection.

2.4. Heart Rate Recordings

The HR and RR intervals were measured by a Polar® V800 telemetric device combined with an equine H7 heart rate sensor belt set (Polar® Electro Oy, Kempele, Finland). As the H7 belt set is a wireless device, it was deemed suitable for measurements in the open stable. The measurement device was always attached to the horse in the open stable. The winter coat of the horses demanded careful moistening prior to attachment. The H7 belt can easily be adjusted to the various sizes of horses and the horse’s coat was wetted, which in addition to electrode gel ensured adequate contact for data transmission. An equine chest protector (normally used against grating) was put on top. A small sewed-on pocket at chest height stored the V800 monitor during measurements close to the sensor. The data collection (−60 min) began after a ten-minute acclimatization period prior to the measurement, which was deemed suitable from the procedures of an earlier study protocol [23]. Data were recorded continuously. After each measurement the data were transferred to a PC.

2.5. Activity Protocol: Behavioral Monitoring

All measurements were initiated in the open stable within the herd. Accounting for the fact that HR measurements are prone to artifacts of movement [27], changes in animal posture or gait need to be considered. Accordingly, during data collection the activity of the horse in the open stable was logged on a per second basis in Microsoft® Excel® 2016 MSO to identify periods of movement or activity to complement the heart rate data. Walking (as opposed to standing) was defined as a fluent movement of all legs in the same direction. Additionally, the time a horse spent eating was noted. Potential disturbances (people entering, noises) were recorded as well, irrespective of whether the horse showed any behavioral response to the stimulus. To log the activity pattern, the experimenter remained within sight with the horse but kept a distance of at least two meters and did not interact. Continuous focal sampling was used, where a single experimenter observed one horse at a time.
If the experimenter lost sight of the respective horse, it was logged and the experimenter changed place accordingly. After each 30 min sequence, the horse was approached calmly by the experimenter to check the data transmission of the measuring device. The activity protocol revealed that each horse stood for at least 20 min in each 30 min sequence. To minimize the influence of motion on the recordings, 20 min of standing were chosen for the final analysis of HR and HRV within each predefined sequence of 30 min. Accordingly, the activity protocol of each horse was inspected and, starting from the beginning of each sequence, only periods of standing that lasted at least for 30 s were included to merge the final 20 min sample. If a disturbance was noted in the activity protocol, a time span lasting from 5 s before to 30 s after the disturbance was excluded to eliminate biased data. If a disturbance lasted noticeably longer than 30 s, the frequency and kind of disturbance during the respective 30 min sequence were registered and considered as a confounding factor. The time which a horse spent eating while standing was calculated within each 20 min sample.

2.6. Data Management and Statistics

A low battery status in the sensor unit of the measuring device led to abnormal HRs during the three control recording sessions, so these had to be excluded from the data pool. The final sample consisted of seven horses (N = 7) regarding the comparison within the experimental condition and four horses (N = 4) regarding the comparison between the conditions. For HRV analysis, the respective 20 min of standing were chosen in the Kubios HRV program (Kuopio, Finland) and merged into one sample per 30 min. A very low threshold level was chosen to detect artifacts according to van Vollenhoven et al. [28]. The percentage of replaced beats was noted for each 20 min sample. The characteristics of each merged sample (mean duration of the single periods of standing and the percentage of replaced beats by the artifact correction) were noted.
Within the experimental condition, each parameter (HR, SDRR, RMSSD, LF/HF, SD1) at each time point was tested for normal distribution by the Shapiro–Wilk test. The time points before EAT (−60 min, −30 min) were compared to the time points after EAT (+0 min, +30 min, +60 min), as well as once with each other. This was performed by an ANOVA for repeated measures or a Friedman test as a non-parametric alternative. A Bonferroni correction was applied, resulting in an alpha level of a = 0.0125. The comparison of each parameter at one time point (−60 min/−30 min/+0 min/+30 min/+60 min) between conditions was performed by a paired samples t-test or a Wilcoxon signed-rank test as a non-parametric alternative. The alpha level remained at a = 0.05.
Effect size measures were used between conditions (Glass’ ∆) and for repeated measurements within the experimental condition (drm, pooled) [29] to evaluate the strength of EAT on changes in the horses’ activity protocol.

3. Results

3.1. Experimental Condition

Regarding the HR and HRV assessments over the experimental condition (N = 7), one-way repeated measures ANOVA with a Greenhouse–Geisser correction (as Mauchly’s test of sphericity showed that this assumption was not met) revealed no statistically significant main effect of time point on the HR of horses within the experimental condition (F(4, 24) = 1.93, p = 0.177, η2p = 0.249, see Figure 1).
Regarding the HRV indices, there was neither a statistically significant difference of the SDRR (Friedman test: χ2(4) = 4.91, p = 0.296) or RMSSD (Friedman test: χ2(4) = 2.63, p = 0.622) nor of the LF/HF (Friedman test: χ2(4) = 6.74, p = 0.150) or the SD1 (Friedman test: χ2(4) = 2.62, p = 0.622); see Table 3.

3.2. Experimental Versus Control Condition

HR and HRV (Mn ± SD) at each time point were compared between the experimental and the control condition (see Table 4). The pairwise comparison of the HR between conditions by a paired samples t-test revealed no statistically significant difference in the measurements prior to EAT (−60 min, −30 min) or after EAT (+0 min, +30 min, +60 min). Similarly, STDRR, RMSSD and SD1 at each respective time point compared between the experimental and the control condition did not reveal any statistical significance.
The Wilcoxon signed-rank test showed no statistically significant difference in the LF/HF compared between conditions (−60 min: Z = −1.6, p = 0.109; −30 min: Z = −5.4, p = 0.593; +0 min: Z = −7.3, p = 0.465; +30 min: Z = 0.0, p = 1.000, +60 min: Z = −3.7, p = 0.715).

3.3. Behavioral Monitoring

Behavioral monitoring was carried out in order to control for artifacts of movement, activity or disturbances that could bias the HR and HRV data. The descriptive statistics (duration of the behaviors standing and eating) as well as disturbances (frequency) and the amount of replaced beats (%) for each sequence of the protocol are shown in Table 5.
The descriptive results reveal that horses stood longer at a time and spent less time eating directly after the EAT session (+0 min) compared to before (−60 min: drm pooled, standing = 0.62, drm pooled, eating = −0.38; −30 min: drm pooled, standing = 0.54, drm pooled, eating = −0.87) within the experimental condition. The longest mean duration of standing at a time was shown under the control condition (+30 min). Compared to the experimental condition, horses spent more time eating under the control condition at time points −60 min (∆ = 0.81; large effect) and +0 min (∆ = 0.77; intermediate effect). The frequency of disturbances was similar across both conditions. The percentage of replaced beats under the control condition was lowest at −60 min and highest at +30 min and +60 min. The percentage of replaced beats under the experimental condition was lowest at −60 min and highest at +0 min and +30 min.

4. Discussion

This study sought to elucidate the autonomous regeneration of therapy horses subsequent to EAT. Our methodological approach was centered on the request by Koolhaas et al. [30] that, in animal welfare studies, the time span of physiological recovery after confrontation with a potential stressor should be focused on. The present data revealed no statistically significant differences in HR, SDRR, RMSSD, LF/HF and SD1 between measurements before and after a group EAT session. A comparison of each parameter at each measuring time point between the experimental and the control condition did not show any statistically significant differences. The obtained results do not confirm the assumption that the service in EAT was overtaxing the internal autonomic regulation of the horses. In general, all obtained HRs remained within the range of a normal equine resting HR (25–40 bpm) as described by Evans (as cited in [11]) or Ohmura and Jones [31], thereby reflecting good health and performance [32]. Similarly, HRV monitoring revealed values within the physiological ranges described in horses under various circumstances [31,33,34]. The present findings parallel earlier EAT studies [19,25,35] but our study protocol differed from Kreuzer et al. [23], Ayala et al. [21] and Mendonça et al. [22] in that the applied measurements took place before and after EAT but not during sessions.
The absence of any SAM axis response subsequent to an EAT group session excludes a substantial influence of EAT on the ANS of therapy horses. If a stimulus has a strong effect, a time interval is identifiable before the physiological response returns to baseline values [36]. As no such interval was identifiable, the duration of autonomous regeneration, if any, was shorter than the time required to return the horses back to the stable.
In addition, no statistically significant differences were measured between the experimental and the control condition which further confirms the absence of a substantial influence of an EAT session on the ANS. In general, horses are less stressed if they either experience fewer stressors or have the ability to recover more quickly from stressors [37], suggesting that the therapy horses either did not physiologically perceive the EAT session as a stressor at all or quickly recovered from the exposure within a few minutes on the way back to their stable. Koolhaas et al. [30] proposed that, if a physiological response diminishes immediately after the stimulus exposure, the adaptive capacity of an animal is not overtaxed and the stimulus is not systemically perceived as a stressor [30]. Our previous results on therapy horses working under similar conditions are in line with the present findings, suggesting that EAT does not significantly impact autonomic and neuroendocrine cascades in equines [23]. However, the present results apply only for the performance in one group session and cannot be generalized to more frequent or prolonged EAT schedules. Thus, more frequent, longer-lasting sessions or more intense human–horse interactions could provoke relevant changes in the SAM axis response in the short term. It is likely that the frequent pre-exposure to therapeutic work in therapy horses led to a higher predictability and controllability of the situation [38], thereby positively shaping the perception of an EAT session. A recent study has proposed that equine-friendly working modalities (i.e., group outdoor housing, unlimited access to roughage, groundwork-based activities), similar to the EAT protocol of this study, may even positively impact the welfare of the horses and result in more compliance during human–animal interaction [39].
The strength of this study lies in its practical relevance. The horses were monitored during their daily routine and studied in a realistic EAT setting. As studying an animal in most cases affects to some extent what is studied [40] (pp. 125–127), it cannot be ruled out that the attaching of the measurement equipment or observation of their activity pattern did influence the horses and their physiological response. The period of acclimatization to the measuring devices (ten minutes) was deemed appropriate. The data collection at +0 min started as soon as possible to avoid the loss of important data after EAT. As no difference in the parameters at +0 min occurred compared to earlier measurements or across conditions, adjusting the measurement devices did not bias the data. In general, measuring an animal in a familiar environment reduces influences due to an artificial, unknown environment [41] but also comes with some drawbacks. The influence of people entering the open stable, herd dynamics or environmental conditions on the results cannot be exactly determined. Waiblinger et al. [41] raise awareness of the potential influence of herd mates on the behavior of the tested individual. To counteract this, such incidences were noted at the time and excluded from the analysis as much as possible. As the horses could move freely within the open stable, the experimenter had to change place accordingly to remain within range of vision to the horse. By protocolling the activity patterns, it was feasible to conduct the measurements within the open stable and no restriction of the horses, which likely would have negatively impacted their wellbeing, was necessary. Generally, it is challenging to schedule appropriate conditions to study HRV in animals because measurements can be confounded by restriction of movement but also movement itself creates a background noise [42]. To account for these difficulties, the horses were measured in the open stable and free to move and only periods of standing were included in the analysis. The measurement of free-moving horses was already successfully conducted in a study by van Vollenhoven et al. [28], who obtained more reliable HRV measurements on pasture than on stock. As the horses in our study had access to hay or trees ad libitum, eating as a potential influencing factor could not be excluded, but the duration was noted. To account for these factors, the duration of eating, the frequency of potential disturbances and the mean length of standing periods at a time (which were ultimately merged into one 20 min sample) were considered. Our data show that horses stood more at a time and spent a shorter time eating directly after EAT compared to measurements taken before. However, as horses stood longer at a time under the control condition and spent less time eating at other times, the time point rather than the intensity of the observed behaviors differed between conditions. The frequency of disturbances was within a similar range across both conditions. These disturbances consisted of people cleaning the stables and outside area as well as preparation procedures or the conduction of other EAT services in the riding hall. Such incidents occur on a regular basis in the daily routine of the horses and did not provoke an obvious response. Ideally, follow-up studies should focus on video analyses and calculate intra-rater reliability to control for the consistency in behavioral data scoring. For the control condition, future research might consider an equivalent time span where horses are moved to the riding arena and back to the open stable, similarly to the experimental condition, but without any contact with clients. This would enhance the comparability between the conditions.
To reduce the influence of the circadian rhythm, measurements were only scheduled for the second half of the day. Although it was ensured that the horses did not perform EAT or balanced training on the days of data collection, sessions on preceding days were not standardized for the purpose of the study. In this study, RR intervals were corrected by a very low threshold artifact correction. The applied method followed recommendations by van Vollenhoven et al. [28], who advised to not use a very strong correction factor and, more importantly, to report which one was applied. In fact, van Vollenhoven et al. [28] showed that all remaining correction factors led to a similar pattern of HRV. In addition, the applied autoregressive model seems suitable for the analysis of heart beats as it estimates the spectral distribution of intervals between biological signals [43]. It has already been applied to measurements in both humans [43] and horses [11].
In general, the accuracy of the HRV measurement and analysis method influences the outcome. Appropriate artifact correction to parallel physiological variance but at the same time remove background noise and artifacts is an ongoing concern in HRV analysis. Unfortunately, a low battery status of the measuring device led to unreliable recordings with artifacts during three control measurements, which ultimately had to be excluded from further analysis. Loss of data is not uncommon in HRV field studies due to the vulnerability of the method to artifacts [22,25]. For future studies, monitoring of the battery status for each single measurement would be desirable to prevent data loss. Another limitation is that we did not assess the ANS during EAT, however, this aspect was addressed in a previous study [23].
In conclusion, our data suggest no substantial influence of EAT on the autonomic regulation pre- and post-session in therapy horses as no significant differences in HR and HRV were found. The cardiovascular welfare indicators under the control condition, where no EAT took place on that day, were similar to the measurements assessed before and after EAT. In general, EAT group sessions with children in the current study seemed of low intensity (mentally and physically) for the horses involved. Given the heterogeneity in EAT, more in-depth research is warranted using behavioral and physiological indices of equine health and wellbeing.

Author Contributions

Conceptualization, L.K., A.N., R.Z. and L.M.G.; methodology, L.K., R.Z. and L.M.G.; software, L.K.; validation, L.K., A.N., R.Z. and L.M.G.; formal analysis, L.K.; investigation, L.K.; resources, A.N. and R.Z.; data curation, L.K. and L.M.G.; writing—original draft preparation, L.M.G.; writing—review and editing, L.K. and L.M.G.; visualization, L.K.; supervision, L.M.G.; project administration, A.N. and R.Z. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

All methods and procedures used in these studies were discussed and approved by the institutional ethics and animal welfare committee in accordance with GSP guidelines and national legislation (protocol codes ETK-05/01/2018 and ETK-11/09/2018, date of approval: 17 October 2018).

Informed Consent Statement

Informed consent was obtained from the owners of the horses who were involved in the study.

Data Availability Statement

Data will be made available upon reasonable request.

Acknowledgments

We acknowledge any support that we received from the stakeholders of the organization Lichtblickhof to make this research project possible.

Conflicts of Interest

This article is a revised and expanded version of a master thesis by Lena Kreuzer. The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
MDPIMultidisciplinary Digital Publishing Institute
EATEquine-assisted therapy
HRVHeart rate variability
SAMSympathetic–adrenal–medullary
ANSAutonomic nervous system
HRHeart rate
SDRRStandard deviation of all RR intervals
RMSSDRoot mean square of successive differences
HFHigh frequency
LFLow frequency
SD1Standard deviation of the diameter of the scatterplot
EExperimental condition
CControl condition

References

  1. Bizub, A.L.; Joy, A.; Davidson, L. “It’s like Being in Another World”: Demonstrating the Benefits of Therapeutic Horseback Riding for Individuals with Psychiatric Disability. Psychiatr. Rehabil. J. 2003, 26, 377–384. [Google Scholar] [CrossRef] [PubMed]
  2. Corring, D.; Lundberg, E.; Rudnick, A. Therapeutic Horseback Riding for ACT Patients with Schizophrenia. Community Ment. Health J. 2013, 49, 121–126. [Google Scholar] [CrossRef] [PubMed]
  3. Earles, J.L.; Vernon, L.L.; Yetz, J.P. Equine-Assisted Therapy for Anxiety and Posttraumatic Stress Symptoms. J. Trauma. Stress 2015, 28, 149–152. [Google Scholar] [CrossRef]
  4. Romaniuk, M.; Evans, J.; Kidd, C. Evaluation of an Equine-Assisted Therapy Program for Veterans Who Identify as ‘Wounded, Injured or Ill’ and Their Partners. PLoS ONE 2018, 13, e0203943. [Google Scholar] [CrossRef]
  5. Machová, K.; Juríčková, V.; Kasparová, A.; Petrová, K.; Sládková, B.; Svobodová, I. An Evaluation of the Effect of Equine-Facilitated Psychotherapy on Patients with Substance Use Disorders. PLoS ONE 2023, 18, e0286867. [Google Scholar] [CrossRef]
  6. Bass, M.M.; Duchowny, C.A.; Llabre, M.M. The Effect of Therapeutic Horseback Riding on Social Functioning in Children with Autism. J. Autism Dev. Disord. 2009, 39, 1261–1267. [Google Scholar] [CrossRef] [PubMed]
  7. Zadnikar, M.; Kastrin, A. Effects of Hippotherapy and Therapeutic Horseback Riding on Postural Control or Balance in Children with Cerebral Palsy: A Meta-Analysis: Review. Dev. Med. Child Neurol. 2011, 53, 684–691. [Google Scholar] [CrossRef]
  8. Gilboa, Y.; Helmer, A. Self-Management Intervention for Attention and Executive Functions Using Equine-Assisted Occupational Therapy Among Children Aged 6–14 Diagnosed with Attention Deficit/Hyperactivity Disorder. J. Altern. Complement. Med. 2020, 26, 239–246. [Google Scholar] [CrossRef] [PubMed]
  9. Merino-Andrés, J.; García De Mateos-López, A.; Damiano, D.L.; Sánchez-Sierra, A. Effect of Muscle Strength Training in Children and Adolescents with Spastic Cerebral Palsy: A Systematic Review and Meta-Analysis. Clin. Rehabil. 2022, 36, 4–14. [Google Scholar] [CrossRef]
  10. Glenk, L.M.; Kothgassner, O.D. Life Out of Balance: Stress-Related Disorders in Animals and Humans. In Comparative Medicine; Jensen-Jarolim, E., Ed.; Springer International Publishing: Cham, Switzerland, 2017; pp. 97–107. ISBN 978-3-319-47005-4. [Google Scholar]
  11. Rietmann, T.R.; Stuart, A.E.A.; Bernasconi, P.; Stauffacher, M.; Auer, J.A.; Weishaupt, M.A. Assessment of Mental Stress in Warmblood Horses: Heart Rate Variability in Comparison to Heart Rate and Selected Behavioural Parameters. Appl. Anim. Behav. Sci. 2004, 88, 121–136. [Google Scholar] [CrossRef]
  12. McCraty, R.; Atkinson, M.; Lipsenthal, L.; Arguelles, L. New Hope for Correctional Officers: An Innovative Program for Reducing Stress and Health Risks. Appl. Psychophysiol. Biofeedback 2009, 34, 251–272. [Google Scholar] [CrossRef]
  13. Lohninger, A. Herzratenvariabiltät: Das HRV-Praxis-Lehrbuch, 1st ed.; Maudrich: Vienna, Austria, 2017; ISBN 978-3-7089-1495-4. [Google Scholar]
  14. Mäkikallio, T.H.; Perkiömäki, J.S.; Huikuri, H.V. Nonlinear Dynamics of RR Intervals. In Dynamic Electrocardiography; Malik, M., Camm, A.J., Eds.; Wiley: Hoboken, NJ, USA, 2004; pp. 22–30. ISBN 978-1-4051-1960-3. [Google Scholar]
  15. Glenk, L.-M. Trends in Animal Welfare Research in Animal-Assisted Interventions. In Handbook on Animal-Assisted Therapy; Elsevier: Amsterdam, The Netherlands, 2025; pp. 235–248. ISBN 978-0-443-22346-4. [Google Scholar]
  16. De Santis, M.; Contalbrigo, L.; Borgi, M.; Cirulli, F.; Luzi, F.; Redaelli, V.; Stefani, A.; Toson, M.; Odore, R.; Vercelli, C.; et al. Equine Assisted Interventions (EAIs): Methodological Considerations for Stress Assessment in Horses. Vet. Sci. 2017, 4, 44. [Google Scholar] [CrossRef] [PubMed]
  17. Drinkhouse, M.; Birmingham, S.S.; Fillman, R.; Jedlicka, H. Correlation of Human and Horse Heart Rates During Equine-Assisted Therapy Sessions with At-Risk Youths: A Pilot Study. J. Stud. Res. 2012, 1, 22–25. [Google Scholar] [CrossRef]
  18. Arrazola, A.; Merkies, K. Effect of Human Attachment Style on Horse Behaviour and Physiology during Equine-Assisted Activities–A Pilot Study. Animals 2020, 10, 1156. [Google Scholar] [CrossRef]
  19. Potier, J.F.; Louzier, V. Evaluation of Stress Markers in Horses during Hippotherapy Sessions in Comparison to Being Ridden by Beginners. Anim. Welf. 2023, 32, e10. [Google Scholar] [CrossRef]
  20. Merkies, K.; McKechnie, M.J.; Zakrajsek, E. Behavioural and Physiological Responses of Therapy Horses to Mentally Traumatized Humans. Appl. Anim. Behav. Sci. 2018, 205, 61–67. [Google Scholar] [CrossRef]
  21. Ayala, M.D.; Carrillo, A.; Iniesta, P.; Ferrer, P. Pilot Study of the Influence of Equine Assisted Therapy on Physiological and Behavioral Parameters Related to Welfare of Horses and Patients. Animals 2021, 11, 3527. [Google Scholar] [CrossRef] [PubMed]
  22. Mendonça, T.; Bienboire-Frosini, C.; Menuge, F.; Leclercq, J.; Lafont-Lecuelle, C.; Arroub, S.; Pageat, P. The Impact of Equine-Assisted Therapy on Equine Behavioral and Physiological Responses. Animals 2019, 9, 409. [Google Scholar] [CrossRef]
  23. Kreuzer, L.; Naber, A.; Zink, R.; Millesi, E.; Palme, R.; Hediger, K.; Glenk, L.M. Horses’ Cardiovascular and Glucocorticoid Responses to Equine-Assisted Therapy with Women with Intellectual Disability: An Exploratory Study. J. Appl. Anim. Welf. Sci. 2024, 1–16. [Google Scholar] [CrossRef] [PubMed]
  24. Pohlin, F.; Brabender, K.; Fluch, G.; Stalder, G.; Petit, T.; Walzer, C. Seasonal Variations in Heart Rate Variability as an Indicator of Stress in Free-Ranging Pregnant Przewalski’s Horses (E. ferus Przewalskii) within the Hortobágy National Park in Hungary. Front. Physiol. 2017, 8, 664. [Google Scholar] [CrossRef]
  25. Gehrke, E.K.; Baldwin, A.; Schiltz, P.M. Heart Rate Variability in Horses Engaged in Equine-Assisted Activities. J. Equine Vet. Sci. 2011, 31, 78–84. [Google Scholar] [CrossRef]
  26. Naber, A.; Völk, M.; Hediger, K.; Zink, R. Can Equine Communication via Audible Exhales Improve the Welfare of Therapy Horses?—A Pilot Practice Project. Hum. Anim. Interact. 2023, 2023, 231–243. [Google Scholar] [CrossRef]
  27. Królak, A.; Wiktorski, T.; Bjørkavoll-Bergseth, M.F.; Ørn, S. Artifact Correction in Short-Term HRV during Strenuous Physical Exercise. Sensors 2020, 20, 6372. [Google Scholar] [CrossRef] [PubMed]
  28. Van Vollenhoven, E.; Grant, C.C.; Fletcher, L.; Ganswindt, A.; Page, P.C. Repeatability and Reliability of Heart Rate Variability in Healthy, Adult Pony Mares. J. Equine Vet. Sci. 2016, 46, 73–81. [Google Scholar] [CrossRef]
  29. Lenhard, W.; Lenhard, A. Calculation of Effect Sizes; Psychometrica: Dettelbach, Germany, 2016; Available online: https://www.psychometrica.de/effect_size.html (accessed on 13 February 2025).
  30. Koolhaas, J.M.; Bartolomucci, A.; Buwalda, B.; De Boer, S.F.; Flügge, G.; Korte, S.M.; Meerlo, P.; Murison, R.; Olivier, B.; Palanza, P.; et al. Stress Revisited: A Critical Evaluation of the Stress Concept. Neurosci. Biobehav. Rev. 2011, 35, 1291–1301. [Google Scholar] [CrossRef] [PubMed]
  31. Ohmura, H.; Jones, J.H. Changes in Heart Rate and Heart Rate Variability as a Function of Age in Thoroughbred Horses. J. Equine Sci. 2017, 28, 99–103. [Google Scholar] [CrossRef]
  32. Boudoulas, K.D.; Borer, J.S.; Boudoulas, H. Heart Rate, Life Expectancy and the Cardiovascular System: Therapeutic Considerations. Cardiology 2015, 132, 199–212. [Google Scholar] [CrossRef]
  33. Huangsaksri, O.; Wonghanchao, T.; Sanigavatee, K.; Poochipakorn, C.; Chanda, M. Heart Rate and Heart Rate Variability in Horses Undergoing Hot and Cold Shoeing. PLoS ONE 2024, 19, e0305031. [Google Scholar] [CrossRef]
  34. Kapteijn, C.M.; Frippiat, T.; Van Beckhoven, C.; Van Lith, H.A.; Endenburg, N.; Vermetten, E.; Rodenburg, T.B. Measuring Heart Rate Variability Using a Heart Rate Monitor in Horses (Equus caballus) during Groundwork. Front. Vet. Sci. 2022, 9, 939534. [Google Scholar] [CrossRef]
  35. Fazio, E.; Medica, P.; Cravana, C.; Ferlazzo, A. Cortisol Response to Road Transport Stress in Calm and Nervous Stallions. J. Vet. Behav. 2013, 8, 231–237. [Google Scholar] [CrossRef]
  36. Heart Rate Variability: Standards of Measurement, Physiological Interpretation and Clinical Use. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Circulation 1996, 93, 1043–1065. [Google Scholar]
  37. Norton, T.; Piette, D.; Exadaktylos, V.; Berckmans, D. Automated Real-Time Stress Monitoring of Police Horses Using Wearable Technology. Appl. Anim. Behav. Sci. 2018, 198, 67–74. [Google Scholar] [CrossRef]
  38. Keeling, L.; Jensen, P. Behavioural Disturbances, Stress and Welfare. In The Ethology of Domestic Animals: An Introductory Text; Jensen, P., Ed.; CABI Publishing: Wallingford, UK, 2002; pp. 79–98. ISBN 978-0-85199-752-0. [Google Scholar]
  39. Grandgeorge, M.; Lerch, N.; Delarue, A.; Hausberger, M. From Human Perception of Good Practices to Horse (Equus caballus) Welfare: Example of Equine-Assisted Activities. Animals 2024, 14, 2548. [Google Scholar] [CrossRef] [PubMed]
  40. Moberg, G.P.; Mench, J.A. The Biology of Animal Stress: Basic Principles and Implications for Animal Welfare; CABI Publishing: Wallingford, UK, 2000; ISBN 978-0-85199-359-1. [Google Scholar]
  41. Waiblinger, S.; Boivin, X.; Pedersen, V.; Tosi, M.-V.; Janczak, A.M.; Visser, E.K.; Jones, R.B. Assessing the Human–Animal Relationship in Farmed Species: A Critical Review. Appl. Anim. Behav. Sci. 2006, 101, 185–242. [Google Scholar] [CrossRef]
  42. Vitale, V.; Balocchi, R.; Varanini, M.; Sgorbini, M.; Macerata, A.; Sighieri, C.; Baragli, P. The Effects of Restriction of Movement on the Reliability of Heart Rate Variability Measurements in the Horse (Equus caballus). J. Vet. Behav. 2013, 8, 400–403. [Google Scholar] [CrossRef]
  43. Bernasconi, P.; Messmer, E.; Bernasconi, A.; Thölen, A. Assessment of the Sympatho-Vagal Interaction in Central Serous Chorioretinopathy Measured by Power Spectral Analysis of Heart Rate Variability. Graefe’s Arch. Clin. Exp. Ophthalmol. 1998, 236, 571–576. [Google Scholar] [CrossRef]
Pets 02 00012 g001
Figure 1. HR (±SD) of therapy horses (N = 7) at different time points (x-axis) under the experimental condition. The box indicates the time when an EAT group session took place.
Figure 1. HR (±SD) of therapy horses (N = 7) at different time points (x-axis) under the experimental condition. The box indicates the time when an EAT group session took place.
Pets 02 00012 g001
Table 1. Therapy horse details: Age and experience in EAT are expressed in years. ✓ indicates successful measurements in the respective condition—experimental (E) or control (C). x indicates loss of data.
Table 1. Therapy horse details: Age and experience in EAT are expressed in years. ✓ indicates successful measurements in the respective condition—experimental (E) or control (C). x indicates loss of data.
HorseSexAgeBreedEAT
Experience		(E)
Condition		(C)
Condition
1f9Criollo3✓x
2f23Shetland pony mix17✓✓
3f29Shetland pony18✓✓
4m25Icelandic horse15✓✓
5m8Criollo1✓x
6m8Warmblood1.5✓✓
7f11Criollo5✓x
Table 2. Measurement schedule of the experimental condition (E) and the control condition (C). ✓ indicates EAT performance, X indicates no EAT performance.
Table 2. Measurement schedule of the experimental condition (E) and the control condition (C). ✓ indicates EAT performance, X indicates no EAT performance.
−60 min−30 minEAT+0 min+30 min+60 min
E30 min30 min✓30 min30 min30 min
C30 min30 minX30 min30 min30 min
Table 3. The HRV indicators SDRR, RMSSD, LF/HF and SD1 (Mn ± SD) were assessed before (−60 min; −30 min) and after (+0 min; +30 min; +60 min) EAT. The vertical lines indicate the time point of the EAT group session.
Table 3. The HRV indicators SDRR, RMSSD, LF/HF and SD1 (Mn ± SD) were assessed before (−60 min; −30 min) and after (+0 min; +30 min; +60 min) EAT. The vertical lines indicate the time point of the EAT group session.
N = 7−60 min−30 min+0 min+30 min+60 min
HR [bpm]36.45 ± 6.3535.97 ± 5.9735.22 ± 6.4033.94 ± 5.8634.27 ± 5.46
SDRR [ms]97.42 ± 39.96111.72 ± 40.58111.31 ± 61.04110.06 ± 53.00109.16 ± 46.38
RMSSD [ms]115.46 ± 58.26137.53 ± 66.10134.25 ± 82.99135.12 ± 70.11130.76 ± 69.01
LF/HF2.83 ± 0.962.19 ± 0.772.95 ± 1.332.57 ± 0.963.37 ± 1.59
SD1 [ms]81.71 ± 41.2497.32 ± 46.7895.01 ± 58.7495.62 ± 49.6292.56 ± 48.86
Table 4. The results of the paired samples t-test are shown. The intensity of the HR, SDRR, RMSSD and SD1 at different time points (−60 min; −30 min; +0 min; +30 min; +60 min) was compared between experimental condition (E) and control condition (C).
Table 4. The results of the paired samples t-test are shown. The intensity of the HR, SDRR, RMSSD and SD1 at different time points (−60 min; −30 min; +0 min; +30 min; +60 min) was compared between experimental condition (E) and control condition (C).
Tdfp
HR [bpm]−60 min0.4930.658
−30 min1.0330.379
+0 min1.2030.317
+30 min2.3030.105
+60 min0.9930.40
SDRR [ms]−60 min1.6130.205
−30 min−0.0830.942
+0 min1.3130.281
+30 min0.0830.950
+60 min−0.8930.441
RMSSD [ms]−60 min1.6230.204
−30 min−2.6430.078
+0 min0.3030.784
+30 min−0.2030.854
+60 min−0.7830.495
SD1 [ms]−60 min1.6230.204
−30 min−2.6330.078
+0 min0.3030.79
+30 min−0.2030.854
+60 min−0.7830.495
Table 5. Potentially confounding characteristics of each 20 min sample at each time point (−60 min to +60 min) are listed: The mean duration of standing at a time and eating, the mean frequency of disturbances and the mean % of replaced beats by artifact correction. The vertical lines indicate the time of either an EAT session under the experimental condition (E) or an equally long time span under the control condition (C). Seven horses were included regarding the analysis within E. The comparison between conditions consisted of fewer horses (N = 4).
Table 5. Potentially confounding characteristics of each 20 min sample at each time point (−60 min to +60 min) are listed: The mean duration of standing at a time and eating, the mean frequency of disturbances and the mean % of replaced beats by artifact correction. The vertical lines indicate the time of either an EAT session under the experimental condition (E) or an equally long time span under the control condition (C). Seven horses were included regarding the analysis within E. The comparison between conditions consisted of fewer horses (N = 4).
N = 7 x−60 min−30 min+0 min+30 min+60 min
Estanding[mm:ss]04:1203:3110:0607:4305:10
Eeating[mm:ss]09:0212:0005:2504:0505:14
Edisturbancesfrequency0.430.430.140.430.29
Ereplaced beats%5.29.711.68.28.4
N = 4 x
Estanding[mm:ss]03:2403:3508:5507:2304:35
C 07:0003:3103:1611:2804:24
Eeating[mm:ss]06:3007:1104:4902:0904:39
C 12:4207:3711:3202:3502:42
Edisturbancesfrequency0.50.750.250.750
C 0.250.750.500.25
Ereplaced beats%7.611.511.410.18.9
C 3.36.37.612.113.1
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Kreuzer, L.; Naber, A.; Zink, R.; Glenk, L.M. Horses’ Cardiovascular Responses to Equine-Assisted Group Therapy Sessions with Children. Pets 2025, 2, 12. https://doi.org/10.3390/pets2010012

AMA Style

Kreuzer L, Naber A, Zink R, Glenk LM. Horses’ Cardiovascular Responses to Equine-Assisted Group Therapy Sessions with Children. Pets. 2025; 2(1):12. https://doi.org/10.3390/pets2010012

Chicago/Turabian Style

Kreuzer, Lena, Anna Naber, Roswitha Zink, and Lisa Maria Glenk. 2025. "Horses’ Cardiovascular Responses to Equine-Assisted Group Therapy Sessions with Children" Pets 2, no. 1: 12. https://doi.org/10.3390/pets2010012

APA Style

Kreuzer, L., Naber, A., Zink, R., & Glenk, L. M. (2025). Horses’ Cardiovascular Responses to Equine-Assisted Group Therapy Sessions with Children. Pets, 2(1), 12. https://doi.org/10.3390/pets2010012

Article Metrics

Back to TopTop