Interobserver Reliability of the Animal Welfare Indicators Welfare Assessment Protocol for Horses

Development of a welfare assessment tool for tourist camp elephants in Asia

Background

Approximately one-third of Asian elephants are managed under human care, participating in educational, cultural, religious, and tourist activities. Management conditions vary considerably among venues, raising questions about whether welfare needs are consistently being met, particularly for Asian tourist camp elephants. To evaluate the well-being of elephants engaged in tourist activities, an evidence-based tool is needed for routine assessments to identify potential welfare risks, aid in the development of better camp standards, and enable caretakers to address specific concerns. While many animal welfare tools exist, none have been designed to consider specific environments and management practices faced by elephants living and working in tourist camps.

Methods

Using direct observations and interviews, the Elephant Welfare Assessment Tool (EWAT) was developed for tourist camp elephants using the Five Domains Model as a framework. Measures were selected based on peer-reviewed literature, existing standards and guidelines, and opinions from animal welfare experts working with zoo and tourist camp elephants. The EWAT differs from existing tools by including criteria on work activities and restraint methods (e.g., chaining and ankus use), factors common in Asia but not often encountered by western zoo elephants. Measures were scored using a 0-2 Likert Scale. The tool was tested in Thailand and determined by calculating a content validity index (CVI) and conducting inter-rater and test-retest reliability tests.

Results

The initial tool included 18 animal-based and 21 resource-based measures across four domains: Nutrition (n = 5), Environment (n = 14), Health (n = 10), and Behavior and Mental State (n = 10). Index scores of content validity (CVI) (Item CVI (0.83), Scale CVI/Average (0.98), and Scale CVI/Universal (0.89)) were high. Measures scoring less than 0.83 were removed: the opportunity to mate, the mahout-elephant relationship, and mahout job satisfaction. The final tool consisted of 42 questions related to 36 measures, including 18 animal-based and 18 resource-based measures within the Nutrition (n = 5), Environment (n = 11), Health (n = 10), and Behavior and Mental State (n = 10) domains. Intraclass correlation coefficients (ICC) for inter-rater reliability (0.78-0.90, p < 0.05) and test-retest (0.77-0.91, p < 0.05) analyses conducted at two camps showed good agreement.

Conclusions

This new assessment tool (EWAT) is a context-specific, holistic method designed to offer a practical means of conducting individual and institutional-level assessments of elephant welfare in tourist camps. It is based on the Five Domains Model using reliable and validated animal- and resource-based measures, data collection through direct observation and interviews, and a numerical scoring system. The tool includes several criteria applicable to tourist rather than zoo venues to make it more relevant to the challenges faced by working elephants in Asia.

A comparison of the welfare of free-ranging native pony herds on common land with those used for conservation grazing in the UK

Free-ranging native Dartmoor and Exmoor ponies have not only held strong cultural and environmental significance for thousands of years within their respective national parks, but their environmental benefits and naturally selected characteristics have also been acknowledged and harnessed for conservation grazing and rewilding programmes. Despite a wealth of literature regarding the welfare of sports, leisure and working horses, there is little information concerning the welfare of free-ranging and extensively grazing ponies. The present study compared the welfare of native Exmoor and Dartmoor ponies grazing on the moors in their respective national parks (n = 47) with those that have been translocated to other areas of the UK for use in conservation grazing and rewilding programmes (n = 29) using a specifically designed observational welfare assessment protocol for free-ranging ponies. The results showed a significant difference between common land and conservation grazing ponies in the scores for Body Condition Score, Water Quality and Availability, Environmental Hazards, Human Disturbance, Skin and Coat Condition and the Human Approach Test. Despite no evidence of significant welfare compromise being identified, this study emphasises the importance of year-round monitoring of welfare and the feasibility of the observational welfare protocol to be used by pony keepers and grazing managers in the future.

Dairy cow personality: Consistency in a familiar testing environment

Personality affects the behavior of dairy cows in response to various situations on farm, such as novel or social challenges, which can affect the welfare of individual cows. Personality traits are evaluated through behavioral testing, typically in unfamiliar environments, but this can be a time-consuming process. Using a novel object test (NOT) and a forced human approach test (FHAT), we tested 222 individual dairy cows and retested a subset of cows (n = 78) 8 mo afterward. The aim was to see if personality traits can be identified in a familiar environment and if traits were consistent in between test and retest. Principal component analysis was used to identify personality traits. Extracted components (personality traits) were named according to determining variables (loadings <-0.63 or >0.63). Based on identified personality traits we calculated the intraclass correlation (ICC) coefficient and the differences between personality trait in test and retest to assess consistency. We identified 3 personality traits from NOT measurements ("explorative," "boldness," and "sociable") explaining 85% of the variance and one trait from behavioral measurements in the FHAT ("trusting") explaining 74% of the variance. Mean ICC of 0.6 with a standard deviation of 0.18 shows that although traits are consistent at herd level, cows react individually in stressful situations. As both tests were carried out in a familiar environment and results are similar to other studies that used unfamiliar environments, we can conclude that tests can be performed in a familiar environment.

Nerve entry points in the mimic musculature of the horse head

Facial expressions are important in pain recognition in horses, but current observation-based pain scales remain subjective. A promising technique to quantitatively measure subtle changes in expression patterns, including changes invisible to the human eye, is surface electromyography (sEMG). To achieve high-quality and reliable sEMG signals, unilateral placement of bipolar electrodes is required in relation to the motor endplates (MEP). We aimed to localize the nerve entry points (NEPs; where the nerve branch first pierced the muscle belly) and the direction of the terminal nerve endings to estimate MEP locations of the innervating nerves in five equine facial muscles involved in pain expression. Three cadaveric Dutch Warmblood horse heads were dissected to identify the NEPs in the musculi caninus, levator anguli oculi medialis, nasolabialis, masseter and zygomaticus. These points were marked with pins and measured in relation to a reference line between two anatomical landmarks near the origin and insertion of the respective muscle. Relative distances were calculated from the most caudally situated landmark. NEPs were located at 33%-38% (caninus), 69%-86% (levator anguli oculi medialis) and 0%-18% (zygomaticus) from the caudal landmark. The nasolabialis showed two innervations zones. Its NEPs were located at 47%-72% (dorsal muscle branch) and 52%-91% (ventral branch). All terminal nerve endings were found to run in rostral direction. The masseter showed numerous NEPs diffusely spread within the muscle belly. Therefore, calculation of relative positions was not performed. These results could form the basis for feasibility studies and standardization of bipolar electrode positioning in vivo to measure facial muscle activity patterns in horses.

Welfare of extensively managed Swedish Gotland ponies

It has been suggested that grazing horses could be used as a credible tool for landscape conservation which would, at the same time, improve horse welfare as opposed to conventional housing. A study was conducted between May 2014 and April 2015 on 12 one year old Gotland ponies managed extensively without supplementary feed. Monthly animal welfare assessments (n = 13) revealed welfare issues in most of the horses, i.e. low body condition score (BCS < 3/5), recurring poor skin condition in 11/12 horses and ocular discharge in 7/12 horses. At the end of the study, compared to the beginning, chafing and poor skin condition increased while coat condition improved. A correlation was found between a negative reaction (score > 0) in the human approach test and BCS < 3 and ocular discharge. Avoidance Distance test values were correlated with faecal parasite counts (> 350 eggs per gram [EPG]). These results indicate that the horses had acceptable welfare during late spring/summer (May-September) and that some horses required additional feed during winter. The animal welfare protocol proved to be an efficient tool for monitoring welfare. The results showed that factors important for extensive management are: daily monitoring; enclosures that provide sufficient feed; access to recovery enclosure; and habituation of horses to human approach.

"Do not look at me like that": Is the facial expression score reliable and accurate to evaluate pain in large domestic animals? A systematic review

Introduction

Facial expression scoring has proven to be useful for pain evaluation in humans. In the last decade, equivalent scales have been developed for various animal species, including large domestic animals. The research question of this systematic review was as follows: is facial expression scoring (intervention) a valid method to evaluate pain (the outcome) in large domestic animals (population)?

Method

We searched two databases for relevant articles using the search string: "grimace scale" OR "facial expression" AND animal OR "farm animal" NOT "mouse" NOT "rat" NOT "laboratory animal." The risk of bias was estimated by adapting the Quality Assessment of Diagnostic Accuracy Studies (QUADAS) checklist.

Results

The search strategy extracted 30 articles, with the major share on equids and a considerable number on cows, pigs, and sheep. Most studies evaluated facial action units (FAUs), including the eye region, the orbital region, the cheek or the chewing muscles, the lips, the mouth, and the position of the ears. Interobserver reliability was tested in 21 studies. Overall FAU reliability was substantial, but there were differences for individual FAUs. The position of the ear had almost perfect interobserver reliability (interclass coefficient (ICC): 0.73-0.97). Validity was tested in five studies with the reported accuracy values ranging from 68.2 to 80.0%.

Discussion

This systematic review revealed that facial expression scores provide an easy method for learning and reliable test results to identify whether an animal is in pain or distress. Many studies lack a reference standard and a true control group. Further research is warranted to evaluate the test accuracy of facial expression scoring as a live pen side test.

The behavior of horses stabled in a large group at essential resources (watering point and lying halls)

Since there is a practical need, this study focused on the behavior of 51 horses kept in one large group in order to make management suggestions regarding the animal:watering point ratio and the design of lying halls. Drinking events and interactions between pairs of horses were analyzed for 18 days as well as the occupancy time of the watering point. The time period, had significant effects on drinking events with visible interactions (P < 0.05). The highest least square means (LSM) and standard errors (SE) of drinking events with and without visible interaction were observed between 12 noon-2 p.m. (8.6 ± 1.6 and 13.6 ± 2.2 drinking events, respectively) which was also the time period of main occupancy of the watering point (41.9 ± 2.7 min/2h). About 57% of drinking events were classed as with visible interaction, of which over 90% were mild (i.e. laying back of ears, tightening of face). Horses' behavior (resting while standing/others) and lying positions (sternal/lateral) in different lying halls were analyzed over 36 days by generalized mixed models. The sternal position was preferred (12.4 ± 1.4 - 39.2 ± 3.0%) compared to the lateral one (1.4 ± 0.4% - 8.3 ± 1.6%). Older horses spent significantly less time in the sternal position (13.8 ± 1.6%) and rested more standing (45.4 ± 3.3%) compared to younger horses (26.4 ± 4.5% and 29.1 ± 4.9%, respectively) (P < 0.05). Overall, this study provides first insights into drinking and lying behavior of horses kept in a large group.

Quality of Life within Horse Welfare Assessment Tools: Informing Decisions for Chronically Ill and Geriatric Horses

Equine Quality of Life (QoL) is an important concern in decision making in veterinary medicine and is especially relevant for chronically ill or geriatric horses towards the end of their lives. To our knowledge, there is no currently available QoL assessment tool for chronically ill or geriatric horses that assesses equine QoL defined as the horse's evaluation of their life. However, tools exist to assess equine welfare in different contexts. Hence, the aims of this study were to analyse how equine welfare, QoL, well-being and happiness assessment tools label, define and operationalise the concepts and to discuss the tools' suitability to assess equine QoL in the context of end-of-life decisions for chronically ill or geriatric horses. Fourteen articles were found through a systematic literature search, describing ten equine welfare assessment tools and one approach to integrating equine QoL in veterinary practice that suggests QoL assessment parameters. We discuss that some welfare assessment tools have the potential to support the development of a QoL assessment tool informing decisions towards the end of horses' lives if they are adjusted to focus on the horses' experiences, to provide an integration into an overall QoL grade and are tailored to chronically ill or geriatric horses.

Development and Validation of a Test for the Classification of Horses as Broken or Unbroken

Regulation EC 1/2005 has stricter rules for transportation of unbroken (untamed) vs. broken (tamed) horses, but does not provide adequate tools for their identification. This study aimed to develop and validate such a tool. A behavioural test (Broken/Unbroken Test (BUT)) based on approaching, haltering, and leading was applied to 100 horses. Physiological and additional behavioural data were also collected, and the horses' status (broken/unbroken) was assessed by the expert who administered the BUT. Each horse's behaviour during the BUT was scored by four trained observers blinded to the horse's history. The BUT score showed excellent inter-observer, intra-observer, and test-retest reliability (all intraclass correlation coefficients (ICCs) > 0.75). It was also negatively associated with respiratory rate, avoidance distance, and time needed to approach, halter, and lead the horse (p < 0.05 for all). The optimal BUT score cut-off for discrimination between broken and unbroken horses (gold standard: expert judgment) showed 97.8% sensitivity and 97.3% specificity. There was almost perfect agreement between BUT-based and expert classification of horses (ICC = 0.940). These findings confirm the BUT's construct and criterion validity. The BUT could provide officials with a feasible, reliable, and valid tool to identify a horse's broken/unbroken status and, consequently, direct stakeholders towards correct transport procedures.

Evaluation of the Feasibility, Reliability, and Repeatability of Welfare Indicators in Free-Roaming Horses: A Pilot Study

Simple Summary

Animal welfare assessment is an essential tool for maintaining positive animal wellbeing. Validated welfare assessment protocols have been developed for farm, laboratory, zoo, and companion animals, including horses in managed care. However, wild and free-roaming equines have received relatively little attention, despite populations being found worldwide. In the UK, free-roaming ponies inhabit areas of Exmoor, Dartmoor, and New Forest, England, and Snowdonia National Park in Wales, amongst others. Visitors and local members of the public who encounter free-roaming ponies occasionally raise concerns about their welfare, as they are not provided with additional food, water, or shelter. In this study, we evaluated the feasibility, reliability, and repeatability of welfare indicators that can be applied to a population of free-roaming Carneddau Mountain ponies to address such concerns. Our findings indicate that many of the trialed indicators were successfully repeated and had good levels of inter-assessor reliability. Reliable and repeatable welfare indicators for free-roaming and semi free-roaming ponies will enable population managers and conservation grazing schemes to manage the welfare of free-roaming horses and ponies.

Abstract

Validated assessment protocols have been developed to quantify welfare states for intensively managed sport, pleasure, and working horses. There are few protocols for extensively managed or free-roaming populations. Here, we trialed welfare indicators to ascertain their feasibility, reliability, and repeatability using free-roaming Carneddau Mountain ponies as an example population. The project involved (1) the identification of animal and resource-based measures of welfare from both the literature and discussion with an expert group; (2) testing the feasibility and repeatability of a modified body condition score and mobility score on 34 free-roaming and conservation grazing Carneddau Mountain ponies; and (3) testing a prototype welfare assessment template comprising 12 animal-based and 6 resource-based welfare indicators, with a total of 20 questions, on 35 free-roaming Carneddau Mountain ponies to quantify inter-assessor reliability and repeatability. This pilot study revealed that many of the indicators were successfully repeatable and had good levels of inter-assessor reliability. Some of the indicators could not be verified for reliability due to low/absent occurrence. The results indicated that many animal and resource-based indicators commonly used in intensively managed equine settings could be measured in-range with minor modifications. This study is an initial step toward validating a much-needed tool for the welfare assessment of free-roaming and conservation grazing ponies.

Objective Assessment of Chronic Pain in Horses Using the Horse Chronic Pain Scale (HCPS): A Scale-Construction Study

The objective assessment of chronic pain is of utmost importance for improving welfare and quality of life in horses. Freedom from disease and pain is one of the 'five freedoms' that are necessary for animal welfare. The aim of this study was to develop a pain scale for the assessment of chronic pain in horses (Horse Chronic Pain Scale; HCPS), which is based on behavioural and facial expressions. The scale was used to assess 53 horses (26 horses diagnosed with chronic painful conditions by means of clinical examination and additional diagnostic procedures (consisting of osteoarthritis, chronic laminitis, chronic back and neck problems, chronic dental disorders) and 27 healthy control animals). Animals were assessed once daily for three consecutive days by two observers that were blinded to the condition of the animals and were unaware of any analgesic treatment regimens. The HCPS consists of two parts, the Horse Chronic Pain Composite Pain Scale (HCP CPS, with behavioural parameters) and the EQUUS-FAP (Equine Utrecht University Scale for Facial Assessment of Pain). The HCP CPS had good inter-observer reliability (intraclass correlation coefficient (ICC) = 0.84, p < 0.001), while the EQUUS-FAP component (with facial expression-based parameters) had poor inter-observer reliability (ICC = 0.45, p < 0.05). The inter-observer reliability of the combined HCPS was good (ICC = 0.78, p < 0.001). The HCPS revealed significant differences between horses with chronic painful conditions and control horses on 2 out of 3 days (p < 0.05). In conclusion, we tested a composite pain scale for the assessment of chronic pain in horses based on behavioural and facial expression-based parameters. Further studies are needed to validate this pain scale before it can be used in practice.

An Ethical Framework for the Use of Horses in Competitive Sport: Theory and Function

Growing ethical concern about equestrian sport is reflected in publications by regulatory authorities, animal charities, and the lay press; and in government debate and social media. However, attempts by regulators and stakeholders to address ethical issues in equine sport have been discipline specific and ad hoc. Ethical frameworks can help stakeholders to make contextual decisions about what should or should not be done in a particular situation. However, when existing animal welfare frameworks and existing sports ethics frameworks are reviewed in this paper, it becomes clear that none provide us with a suitable or sufficient tool for considering ethical issues which can arise in situations where the athlete is a non-human, non-consenting participant. This paper presents the theoretical development of a novel ethical framework, with the aim of providing stakeholders with a tool which they might apply to the consideration of the ethical questions which inevitably arise in relation to (equestrian) sport. The derivation and limitations of the ethical framework are explained. The use of the framework will serve both to underwrite the continuation of the social license to use horses in sport and also to enable those within equestrian sport to critically assess existing and proposed practices and to make welfare-improving adjustments to practice if/where necessary. The theoretical framework as presented here is currently being practically tested and refined in consultation with industry stakeholders, and that research will be submitted for publication in due course.

Evaluation of Inter-Observer Reliability of Animal Welfare Indicators: Which Is the Best Index to Use?

Simple Summary

In order to be effective, on-farm welfare assessment protocols should always rely on reliable, as well as valid and feasible, indicators. Inter-observer reliability refers to the extent to which two or more observers are observing and recording data in the same way. The present study focuses on the problem of assessing inter-observer reliability in the case of dichotomous (e.g., yes/no) welfare indicators and the presence of two observers, in order to decide about the inclusion of indicators in welfare assessment protocols. We compared the performance of the most popular currently available agreement indexes. Some widely used indexes showed their inappropriateness to evaluate the inter-observer reliability when the agreement between observers was high. Other less used indexes, such as Bangdiwala’s B or Gwet’s γ(AC1), were found to perform better and are therefore suggested to assess the inter-observer reliability of dichotomous indicators.

Abstract

This study focuses on the problem of assessing inter-observer reliability (IOR) in the case of dichotomous categorical animal-based welfare indicators and the presence of two observers. Based on observations obtained from Animal Welfare Indicators (AWIN) project surveys conducted on nine dairy goat farms, and using udder asymmetry as an indicator, we compared the performance of the most popular agreement indexes available in the literature: Scott’s π, Cohen’s k, kPABAK, Holsti’s H, Krippendorff’s α, Hubert’s Γ, Janson and Vegelius’ J, Bangdiwala’s B, Andrés and Marzo’s ∆, and Gwet’s γ(AC1). Confidence intervals were calculated using closed formulas of variance estimates for π, k, kPABAK, H, α, Γ, J, ∆, and γ(AC1), while the bootstrap and exact bootstrap methods were used for all the indexes. All the indexes and closed formulas of variance estimates were calculated using Microsoft Excel. The bootstrap method was performed with R software, while the exact bootstrap method was performed with SAS software. k, π, and α exhibited a paradoxical behavior, showing unacceptably low values even in the presence of very high concordance rates. B and γ(AC1) showed values very close to the concordance rate, independently of its value. Both bootstrap and exact bootstrap methods turned out to be simpler compared to the implementation of closed variance formulas and provided effective confidence intervals for all the considered indexes. The best approach for measuring IOR in these cases is the use of B or γ(AC1), with bootstrap or exact bootstrap methods for confidence interval calculation.

Welfare of horses from Mexico and the United States of America transported for slaughter in Mexico: Fitness profiles for transport and pre-slaughter logistics

Every year thousands of horses from Mexico and the United States of America (USA) are transported to slaughter in Mexico, but little is known about their welfare or pre-slaughter logistics. In this study, we recorded the origin, sex, age and condition of horses (121 journeys, 2648 animals) upon arrival to an abattoir in northern Mexico, including transport details. Horse welfare was measured indirectly via individual scores for body condition, coat quality, lameness, ocular and nasal discharge, as well as reactivity to a chute restraint test, all performed shortly after unloading. The average journey duration was 9.69 (±7.6) hours for horses from Mexico and 16.77 (±4.51) hours for horses from the USA (77 % of all journeys). The prevalence of ocular discharge, nasal discharge, skin wounds, lameness and diarrhoea, were 23 %, 12 %, 11 %, 9 %, 1 % (respectively) of all the horses observed, with no significant differences between Mexican and American horses (P ≥ 0.05). During the chute test the American horses were calmer than the Mexican ones (P < 0.001), who were more restless and aggressive (P = 0.001). Likewise, vocalizations in their three variants during the restraint, neigh/whinny (P = 0.018), nicker (P < 0.001), and snort (P = 0.018), were more common in horses from Mexico. In order to help characterize fitness for transport, a two-step cluster analysis was applied using the welfare indicators, suggesting the existence of four clusters (C) evaluated on arrival at the abattoir (from good to very poor fitness): good (profile C4, n = 769, 29.1 %), average (profile C1, n = 799 horses, 30.2 %), poor (profile C3, n = 586, 22.1 %) and very poor (profile C2, n = 494, 18.6 %). In fact, the C4 best welfare group had 0% lame, 0% nasal discharge, 16.4 % ocular discharge, 7.9 % skin wounds. Instead, the C2 poorest welfare group had 45.8 % lame, 61.1 % nasal discharge, 42.8 % ocular discharge, and 19.9 % skin wounds. Results show potential for using nasal discharge, lameness and ocular discharge as key indicators of horse fitness and welfare on abattoir. The study provides detailed scientific data to help establish strategies regarding optimal days of recovery post-transport and fattening for homogenization of weights between animals of different origins, logistic planning, and optimization of logistic resources to minimize the biological cost of long-distance transport.

On-Farm Welfare Assessment of Horses: The Risks of Putting the Cart before the Horse

Although the question of animal welfare has been an important source of concern in the scientific community for several decades, many aspects are still under debate. On-farm assessments have to be rapid, acceptable to farmers and safe for both the assessors and animals. They are thus very demanding, with multiple decisions to make, such as the choice of appropriate indicators, sampling methods and scoring. Research has moved from resource-based to animal-based criteria, which reflects the subjective welfare state of an animal rather than relying upon external indices. In the present review, we describe two major (i.e., the most frequently/recently tested or disseminated) protocols: one in low-/middle-income countries, and the other in high-income countries, for on-farm assessments of horses, using animal-based resources; we evaluate their strengths and limitations, and then we compare their results with those obtained by various other studies. We propose lines of improvement, particularly in view of public dissemination, and offer suggestions for further refinement or new protocols. We emphasize the high risks of putting the cart before the horse, i.e., proposing protocols that rely upon indicators and sampling methods that need to be refined, as this could lead to under-evaluation (or less likely over-evaluation) of current welfare problems. Because welfare is a subjective experience, the true representation of an individual's actual welfare status has to be evaluated by using objective assessment tools (that are validated and have a scientific basis) used by well-trained observers.