Animal-Assisted Therapy Combined With Standard Rehabilitation for Motor Recovery in Dementia Patients in a Medical and Rehabilitation Care Unit

1. Background and Rationale Dementia is a progressive syndrome characterized by cognitive decline beyond what is expected from normal aging, affecting memory, reasoning, orientation, comprehension, language, and judgment. According to the World Health Organization (WHO), approximately 50 million people worldwide are affected by dementia, with an estimated 10 million new cases per year. This figure is projected to reach 152 million by 2050, making dementia care a major global public health challenge. Current pharmacological treatments for managing behavioral and psychological symptoms of dementia (BPSD) demonstrate limited efficacy and carry significant adverse effects. French health authorities (Haute Autorité de Santé, HAS) therefore recommend prioritizing non-pharmacological interventions (NPIs) as first-line approaches for symptom management.
2. Rationale for the Intervention Animal-Assisted Therapy (AAT) is a targeted, structured therapeutic intervention involving the introduction of trained animals into clinical care, delivered by qualified health professionals with documented assessment of participant progress. Based on the International Association of Human-Animal Interaction Organizations (IAHAIO) definition, AAT is a goal-directed intervention designed to promote therapeutic benefits in human health by introducing animals into care. It emphasizes improvement of physical, cognitive, behavioral, and/or socio-emotional functioning. While AAT has been used since the 18th century, its scientific evidence base for dementia populations remains limited. A 2019 systematic review (Lai et al., Cochrane Database) identified only 9 eligible studies (6 RCTs, 2 cluster RCTs, 1 crossover RCT), showing that AAT may slightly reduce depressive symptoms in dementia patients, while evidence for effects on other outcomes (quality of life, other BPSD) remains inconclusive. Positive effects on agitation, depression, quality of life, and balance have been observed, but no study has specifically addressed the recovery of motor abilities in demented patients in a rehabilitation setting. Since October 2021, the Institut Médical de Sologne (IMDS) has integrated AAT into the care program of elderly patients with dementia hospitalized in its geriatric medical and rehabilitation care unit (SMR). Clinical observation showed that patients resistant to or disengaged from standard rehabilitation became more accessible and demonstrated previously unseen motor capacities when in contact with trained animals. The therapeutic alliance between the nurse-animal therapist, psychomotor therapist, and nursing assistant appeared to facilitate patient engagement, improve therapeutic compliance, and unlock motor potential. This exploratory pilot study aims to formally and objectively measure the impact of this combined approach on motor recovery.
3. Study Design RÉMÉDI-A is a prospective, single-center, cluster-randomized, open-label pilot study conducted in the geriatric SMR ward of the Institut Médical de Sologne (IMDS), 1 rue Cécile Boucher, 41600 Lamotte-Beuvron, France. Cluster randomization is implemented in clusters of 2 patients, randomized every 4 weeks into either the experimental group (standard rehabilitation + AAT) or the control group (standard rehabilitation alone). Cluster randomization is required for organizational and safety reasons: AAT sessions involve a maximum of 2 participants simultaneously. Randomization is centralized and performed using R software by the LNA Santé Research Support Unit (Cellule d'Appui à la Recherche), with block randomization to ensure balanced group sizes over time. The study is open-label, as blinding is not feasible for behavioral interventions.
4. Intervention Description by Arm Arm A - Experimental: Standard Rehabilitation Combined with Animal-Assisted Therapy

In addition to the standard rehabilitation program, participants in the experimental arm receive structured AAT sessions integrating psychomotor rehabilitation with trained animals. Sessions are conducted by a team comprising: (1) a certified nurse-animal therapist (holding a diploma in animal-mediated therapy), (2) a psychomotor therapist, and (3) a nursing assistant.

Frequency: 2 sessions × 1 hour per week, for 4 weeks (8 sessions total, maximum 2 participants per session).

Before the first session, a joint assessment by the nurse-animal therapist and psychomotor therapist defines individualized therapeutic motor objectives for each participant.

Each session follows a structured format:

1. Welcome, round-table "emotions" check-in, and session presentation
2. Introduction of animals and initial animal contact
3. Motor exercises integrating the animal across 5 domains:

• Global motor function: psychomotor obstacle courses with the animal, targeting general dynamic coordination, transfers, and ambulation; ball-throwing activities
• Balance and posture: standing exercises, oculo-manual and oculo-pedal coordination tasks
• Fine motor skills and praxis: grooming (brushing), feeding preparation (chopping vegetables, serving treats), requiring fine manual gestures and tonic adjustment
• Body schema: sensory awareness activities
• Sensory-motor function: verbal interaction, tactile contact, visual observation, auditory stimulation, olfactory stimulation

1. Closing animal contact and session debrief

Animals used: 1 dog (golden retriever/malinois cross) and/or 2-4 rabbits (dwarf angora, lion-head, and/or rex breeds), selected according to session objectives. The dog is particularly suited for global motor function, balance, and posture work; rabbits are preferred for fine motor skills and sensory stimulation. Working with multiple animals provides animal welfare, session flexibility, and group work compatibility.

Arm B - Control: Standard Rehabilitation Participants in the control arm receive the standard rehabilitation program as delivered at IMDS, including physiotherapy, occupational therapy, adapted physical activity (APA), and group psychomotor rehabilitation sessions. The rehabilitation program is personalized based on each participant's clinical profile.

For psychomotor rehabilitation specifically, participants attend group sessions led by the establishment's psychomotor therapist. Frequency: 2 sessions × 1 hour per week, for 4 weeks (8 sessions total). Sessions cover themes including fine and manual motor skills, static and dynamic coordination, posture, muscle tone, and praxis.

(5) Standard Care Program Both groups receive standard geriatric medical care during hospitalization (medical monitoring, nursing care, nutritional support, medication management). Animal-assisted therapy is an add-on intervention for the experimental group; it does not replace any component of standard care.

(6) Visit Schedule and Procedures

• Pre-inclusion (Day -2 to Day 0) Eligibility verification; provision of information letter; reflection period (up to 2 days)
• Inclusion / Baseline (Day 0, Week 1) Written informed consent; demographic and clinical data; EGP (items 1-10.5); IADL; ADRQL; Tinetti; MMSE; sensory deficit assessment; individual rehabilitation assessment; randomization
• Weeks 2-5 (4-week rehabilitation period) 8 rehabilitation sessions (2/week); daily behavioral grid evaluation (emotional, social, motor dimensions) at each session; adverse event monitoring
• End-of-study assessment (Week 6, ± 7 days) EGP; IADL; ADRQL; Tinetti; number of sessions attended; current medications; adverse event documentation (7) Assessment Tools Geronto-Psychomotor Examination (EGP) - Primary Outcome Measure The EGP is a standardized French geriatric psychomotor assessment tool evaluating items 1 to 10.5, covering the following domains: global motor function, balance and posture, fine motor skills and praxis, body schema, and sensory-motor function. Items are scored on discrete scales and aggregated to a composite psychomotor score. Higher scores indicate better psychomotor functioning. (Detailed scoring scale and item-level descriptions are provided in Annex XIII of the protocol. Score range and MCID: per Annex XIII.)

IADL - Lawton \& Brody Instrumental Activities of Daily Living Scale The IADL scale is an 8-item instrument assessing functional independence in instrumental activities of daily living: telephone use, shopping, food preparation, housekeeping, laundry, transportation, medication management, and financial management. Each item is scored 0 (dependent) or 1 (independent). Score range: 0 to 8; higher score = greater functional independence/autonomy.

ADRQL - Alzheimer's Disease-Related Quality of Life Scale The ADRQL is a 40-item proxy-rated quality of life instrument specifically designed for people with dementia, assessing five domains: social interaction, awareness of self, feelings and mood, enjoyment of activities, and response to surroundings. Items are rated on a binary format by a caregiver or clinician familiar with the participant. Score range: 0 to 100; higher score = better quality of life.

Tinetti Performance-Oriented Mobility Assessment The Tinetti scale is a two-part observational assessment of gait and balance in older adults. Balance subscale: 16 items, scored 0-16. Gait subscale: 12 items, scored 0-12. Total score range: 0 to 28; higher score = better balance and gait performance; lower score = greater fall risk. Risk categories: score <19 = high fall risk; 19-24 = moderate fall risk; ≥25 = low fall risk.

Qualitative Behavioral Grid (Emotional, Social, and Motor Dimensions) A structured behavioral observation grid assessing three dimensions of participant functioning during each rehabilitation session: (1) emotional dimension, (2) social dimension, and (3) motor dimension. Evaluated by the rehabilitation team at each of the 8 sessions during the 4-week rehabilitation period (Weeks 2-5). Score range and direction: per protocol Annex XIII.

(8) Statistical Analysis Plan Statistical analyses will be conducted by the LNA Santé Research Unit using R software (latest version).

Descriptive statistics: Continuous variables presented as mean ± SD (if normality confirmed by Shapiro-Wilk test) or median [Q1-Q3, min-max]. Categorical and discrete variables described as n (%).

Between-group comparisons: Student's t-test (primary) or Mann-Whitney U test (non-normal distributions) for continuous variables; Pearson's Chi-square test (with Monte Carlo simulation if expected cell counts <5) for categorical variables.

Variable classification: EGP item scores and IADL scores treated as discrete variables. ADRQL subscores and total score, Tinetti score, and behavioral dimension scores treated as continuous variables. The Tinetti scale may additionally be used as an ordinal categorical variable (fall risk categories).

Within-group pre/post comparisons: Paired Student's t-test (if difference is normally distributed) or Wilcoxon signed-rank test for continuous variables; McNemar test or omnibus symmetry test for categorical variables.

Significance level: All tests two-sided, global alpha = 5%. p < 0.05 considered statistically significant. 95% confidence intervals reported throughout. Missing data will be documented.

Sample size: 32 participants over 15 months of inclusion. No prior published study on motor outcomes with AAT in demented rehabilitation patients was identified at study design. Sample size was therefore based on the establishment's inclusion capacity and target population representativeness. This pilot study is designed to validate the AAT rehabilitation protocol (feasibility), and to provide effect size estimates for the design of a subsequent multicenter randomized controlled trial.

(9) Benefit/Risk Analysis

Expected benefits:

• Improved global motor function and autonomy
• Facilitated therapeutic interaction with participants
• Improved participant adherence to rehabilitation sessions
• More efficient rehabilitation outcomes

Identified risks (minimal):

• Animal scratches or bites during AAT sessions (minor, managed on site)
• Undiscovered allergic reaction to animal fur or saliva (minimized by eligibility - exclusion criteria)
• Anxiety activation or paradoxical reaction to animal contact (managed by clinical team; premature study withdrawal may be decided by the coordinating investigator)

This study is classified as minimal-risk interventional research (Loi Jardé, Category 2). No ANSM authorization is required. Insurance has been taken out by LNA Santé with Relyens Mutual Insurance.