Impact of Harp Therapy on Behavioral Disorders in Nursing Home Residents With Neurodegenerative Disease (ARPE)

• Background and Rationale Non-pharmacological interventions (NPIs) are non-invasive, non-pharmacological care techniques designed to prevent, manage, or alleviate certain conditions as complementary approaches to standard treatments. NPIs are central to LNA Santé's SENS care approach, with the objective of integrating them systematically into the management of neurodegenerative disorders associated with behavioral disturbances, in line with French health authority recommendations.

Among these interventions, music therapy is one of the most evidence-supported. Its efficacy has been demonstrated for apathy , anxiety and depression , cognitive and emotional functioning, and pain management. Two approaches exist: active music therapy (participants produce music) and receptive music therapy (participants listen to music), which can be combined.

• Rationale for the Intervention Harp therapy (harpe-thérapie) is a form of receptive music therapy in which a trained therapist plays the harp live during individual or group sessions. Villa Éléonore first introduced harp therapy in late 2023 through a trainee from the first French cohort of harp therapy practitioners. Initial individual sessions with apathetic, bedridden, or palliative residents produced moments of emotional soothing and arousal and were positively received by families. Subsequent group sessions in memory care units in the late afternoon contributed to a general calming effect during and after the session, including a smoother dinner transition and a more serene bedtime routine.

Given these positive preliminary observations and the absence of published literature on harp therapy specifically, Villa Éléonore designed the ARPE study to objectively measure the impact of structured harp therapy sessions on behavioral symptoms in memory care residents. The working hypothesis was that harp therapy would reduce disruptive behavioral symptoms (particularly sundowning-related agitation, verbal and non-verbal aggression, and sleep disturbance), reduce psychotropic medication use, and improve staff well-being.

• Study Design ARPE is a prospective, single-center, single-group observational study. No control group or comparator was included. All eligible residents from both memory care units of Villa Éléonore were offered participation. The study followed a longitudinal repeated-measures design over 6 months, including 12 weeks of weekly harp therapy sessions and a 3-month post-intervention follow-up.

Regulatory classification: This study is classified as Research Not Involving the Human Person (Recherche N'Impliquant pas la Personne Humaine - RNIPH) under Article L1121-1 of the French Public Health Code, as it does not pursue biological or medical knowledge objectives involving experimental procedures. Data collection is governed by French data protection standards under CNIL Reference Methodology MR-004, which ensures pseudonymization.

• Intervention Description Harp Therapy (Receptive Music Therapy) A certified harp therapist delivers one live group harp therapy session per week for 12 consecutive weeks to all participating residents simultaneously in their memory care unit. Sessions take place in the late afternoon (approximately 5:00-6:30 PM), a period associated with increased behavioral disturbances (sundowning syndrome). Residents are seated comfortably and listen to the harp music live. No active participation (playing, singing, or moving) is required. Session duration is not specified in the protocol but is consistent with standard harp therapy practice (approximately 30-45 minutes). Residents who do not wish to attend on a given week may decline.
• Standard of Care All participants continue to receive their usual nursing home care, including ongoing pharmacological treatments and other NPIs (e.g., physiotherapy, occupational therapy). These are documented at each study visit to assess their potential influence on behavioral outcomes.
• Visit Schedule and Procedures Visit Timing Assessments Pre-inclusion (V-S) Day -10 Study presentation to residents and families; information letter distributed; 10-day non-opposition reflection period Inclusion (V0) Day 0 Non-opposition documented; NPI-ES, CMAI; demographics, clinical data, MMSE, GIR, psychotropic medications (past 6 weeks), other NPIs During 12 sessions Weekly (W1-W12) EVIBE (before and after each session); LIKERT staff well-being (twice/week - once post-session, once on a session-free day) Intermediate visit (V6) Week 6 NPI-ES, CMAI; psychotropic medications and other NPIs (past 6 weeks) End-of-session visit (V12) Week 12 NPI-ES, CMAI; psychotropic medications and other NPIs (past 6 weeks) End-of-study assessment 3 months after last session NPI-ES, CMAI; psychotropic medications and other NPIs (past 3 months)
• Assessment Tools NPI-ES (Neuropsychiatric Inventory - Caregiver Version) The NPI-ES is a validated, structured caregiver-administered interview assessing 12 neuropsychiatric symptom domains: delusions, hallucinations, agitation/aggression, depression/dysphoria, anxiety, euphoria/elation, apathy/indifference, disinhibition, irritability/lability, aberrant motor behavior, sleep and nighttime behavior disorders, and appetite/eating abnormalities. Each domain is rated on frequency (1-4) and severity (1-3); domain score = frequency × severity (0-12). Total score ranges from 0 to 144; higher scores indicate greater neuropsychiatric symptom burden.

CMAI (Cohen-Mansfield Agitation Inventory) The CMAI is a validated 29-item tool assessing the frequency of agitated behaviors in nursing home residents. Each item is rated on a 7-point scale (1 = never, 7 = several times per hour). Total score ranges from 29 to 203; higher scores indicate greater agitation.

EVIBE (Évaluation Immédiate du Bien-Être) The EVIBE is a brief, standardized immediate well-being assessment tool used in French geriatric and psychogeriatric settings. It is administered by a trained clinician (psychologist, occupational therapist, or psychomotor therapist) before and after each harp therapy session to capture immediate changes in residents' perceived well-being. It uses a visual or ordinal scale to quantify well-being. A higher score indicates better well-being.

LIKERT Staff Well-Being Scale A Likert-type scale completed by voluntary care staff members to rate their own immediate well-being and their perceived ease of managing residents' behavioral difficulties. Administered twice per week: once after a harp therapy session, and once on a day without a session, to capture potential carry-over effects on staff experience. Score ranges from 1 to 5 (or 1 to 7 depending on the version used); higher scores indicate better well-being / greater ease of managing behavior.

• Statistical Analysis Plan Statistical analyses were conducted by the LNA Santé research support unit using R (version 3.6.1). Given the exploratory nature of the study and the absence of a comparator arm, no formal sample size calculation was performed; enrollment was driven by the estimated inclusion capacity of Villa Éléonore (25-30 residents). No interim analysis was planned.

Descriptive statistics: Quantitative variables are described by number of observations, mean, standard deviation, minimum, maximum, 25th/50th/75th percentiles, and 95% bilateral confidence intervals. Qualitative variables are described by frequency counts and relative/absolute percentages.

Inferential statistics: Comparisons across the four time points are performed using one-way analysis of variance (ANOVA) for quantitative variables and chi-squared tests for qualitative variables. All tests are bilateral with an overall alpha risk of 5%. A p-value < 0.05 is considered statistically significant. All confidence intervals are 95% bilateral.

• Benefit-Risk Assessment Expected benefits: Reduction in neuropsychiatric symptoms (particularly sundowning-related agitation, verbal/non-verbal aggression, anxiety, sleep disturbances, aberrant motor behavior); reduction in psychotropic medication use; improvement in residents' and care staff's well-being.

Identified risks: Paradoxical anxiety reactions; emotional activation (reactivation of distress linked to past trauma or memories); auditory discomfort. All risks are considered minimal. Any adverse reaction may lead to withdrawal from the study at the investigator's discretion.X

• Regulatory and Ethical Framework This study is classified as RNIPH under French law (Article L1121-1 of the Public Health Code) and does not require review by a Committee for the Protection of Persons (CPP). Data collection and processing comply with CNIL Reference Methodology MR-004, pursuant to the French Data Protection Act (Law No. 78-17 of 6 January 1978, as amended). A 10-day non-opposition period was offered to all residents and/or their legal representatives before inclusion. The sponsor (LNA Santé) is the data controller. Data are pseudonymized; the correspondence list is held exclusively by the coordinating investigator. Results will be compiled into an internal report for LNA Santé establishments. No external publication is planned.

Inclusion criteria

• Living permanently in a memory care unit (unité protégée) of Villa Éléonore nursing home for at least 2 months
• Diagnosed with a moderate-to-severe neurodegenerative disease (such as Alzheimer's disease or a related condition), AND/OR Mini-Mental State Examination (MMSE) score of 15 or below (or MMSE not evaluable), assessed within the last 12 months
• Presenting moderate-to-severe disruptive behavioral difficulties, such as anxiety, agitation, nighttime wandering, or difficulty falling asleep
• Participant and/or their legal representative has been informed about the study and has not objected to the use of their pseudonymized data

Exclusion criteria

• Unstabilized psychiatric condition
• Uncorrected hearing impairment causing sensory isolation (i.e., without appropriate hearing aids)
• Documented end-of-life status in the medical record