The goal of this pilot randomized clinical trial is to learn if a music therapy treatment, called AMUSED, can improve engagement and reduce behavioral symptoms in older adults with severe dementia who live in care facilities. The main questions it aims to answer are:

• Is it feasible to conduct a full-scale trial of AMUSED?
• Can investigators identify the best outcome measures to assess impact on behavioral symptoms of dementia?
• Does speech offer a useful indicator of treatment effectiveness? Researchers will compare a group-based music therapy treatment to a reading activity to learn if music therapy leads to greater improvements in behavioral symptoms and speech patterns.

Participants will:

• Participate in either music therapy (includes live music, singing, and rhythmic instrument playing) or a reading group with stories about life and nature and talk about memories.
• Attend small group sessions twice a week for 12 weeks, with each session lasting 40 minutes between lunch and dinner.
• Be observed and assessed for behavioral symptoms, cognition, and speech several times during treatment and at a 4-week follow-up.

Meaningful activity has personal importance or usefulness, fosters a sense of fulfillment, and contributes to well-being, whether through cherished pastimes or goal-oriented endeavors. Persons with dementia (PWD) experience significant deficits in memory, communication, and functional abilities, which poses unique obstacles to engage in meaningful activities. When PWD lack sufficient activity, or if activities are poorly tailored to ability, PWD can demonstrate increased behavioral symptoms. Such symptoms impact the PWD and contribute to caregiver stress. Nonetheless, PWD across the severity continuum often respond favorably to music. Thoughtfully designed music-based interventions (MBIs) may offer PWD opportunities to engage in meaningful activity and socialize. However, empirical evidence to support the use of MBIs with this population is very limited. This pilot study advances a line of research to test and refine a randomized clinical trial design and gather preliminary data on intervention effectiveness beyond a single session.

The overall objectives of this quantitative pilot study are to (i) establish a well-refined clinical trial design, procedures, and outcome measures and (ii) obtain preliminary data of enduring effectiveness of a multimodal music therapy intervention for engaging persons with severe dementia (AMUSED) with regards to behavioral symptoms and speech. AMUSED uses live participant-preferred music and progressively layers singing, touch, and rhythmic instrument playing concurrent with participant behavioral responses. This MBI is grounded in the person-centered Clinical Practice Model for PWD, which provides guidelines for adjusting the degree of support and challenge offered to an individual. This Model is also used to enhance treatment fidelity.

This pilot study builds on a feasibility study that field tested logistics and demonstrated feasibility and acceptability of the AMUSED intervention protocol through a randomized cross-over trial design. The researchers refined recruitment, data collection, protocol training, and intervention processes. As in past similar work, the researchers demonstrated positive engagement during music therapy sessions. However, obtaining longer-term measures of treatment effects (i.e., neuropsychiatric inventory, Fitbit actigraphy, pulse oximetry, functional measures) proved impractical and unreliable in the care milieu. The researchers' recent qualitative work with long-term care staff revealed possible solutions to more feasibly measure such outcomes in a long-term care setting. The current pilot study builds on this past feasibility and qualitative work.

In this quantitative pilot study, the investigators will use a randomized clinical trial design to test methods and procedures that will be used in a future larger clinical trial. This will help enhance the rigor and reproducibility of this research. The aims of this pilot study are to: 1) determine whether the clinical trial protocol is feasible for a full-scale trial, 2) select the most appropriate primary outcome measure(s) of behavioral symptoms of dementia, and 3) determine if spontaneous speech offers a useful indicator of music-based intervention effectiveness in persons with severe dementia. Participants will be older adults with severe dementia who live in a long-term care facility. To isolate the effects of the music and obtain an estimate of cluster effects, care facilities will be randomly assigned to either receive music therapy or reading. Both conditions use the same structure, strategies, and dosage, with music being the key difference.

PARTICIPANT Inclusion Criteria:

• 65 years or older
• late-onset dementia diagnosis from a physician
• stable at facility long enough to establish residency and routine at the facility (about 2 months prior to start of study)
• English is primary language (for pilot study practicality and ensure straightforward evaluation of our aims)
• Dementia is severe: As in the feasibility study, residents' charted score on the Brief Inventory of Mental Status (BIMS) < 7 will be used and no independent function in community affairs, hobbies, chores, or personal care. The BIMS is a component of the Minimum Data Set (MDS), a federally mandated clinical assessment for all residents in US Medicare and Medicaid certified care facilities. It is completed periodically and interpreted by a licensed health care professional employed by the facility who has completed requisite training on the assessment.

PARTICIPANT Exclusion criteria:

• music therapy recipient
• co-occurring movement disorder
• preexisting cognitive disability
• Milder dementia (BIMS 7+); Persons with more significant cognitive decline could arguably respond differently to the intervention than those with milder dementia, a notion that is supported by past music therapy research with this population.

FACILITIES - Inclusion Criteria:

• Any facility in Kentucky that provides residential care for persons with severe dementia (up to 8 facilities).
• At least 5 residents must enroll to make the best use of our resources.
• At least 1 facility staff member available to help with recruitment and periodic observable data collection
• Can accommodate consistent session days and times between lunch and dinner, 2 times per week for 40 minutes, for 12 weeks.
• Willing to communicate questions, concerns, and changes in safety protocols to the research team.
• Can provide a space appropriate for a small group activity (3-5 residents plus a group leader).
• Can ensure staff support is available for safely transitioning participants to/from sessions at the agreed upon session days and time.
• Can assert that music therapy from a credentialed music therapist (MT-BC) is not offered by the facility, to prevent "double dosing" (music entertainment is ok).

FACILITIES - Exclusion Criteria:

-Does not meet one or more inclusion criteria.

This is a small pilot study. Pilot data for preliminary effect size estimating will not be publicly shared. Researchers may request de-identified data and study materials from the principal investigator.

6 months after publication. The principal investigator will consider sharing data and study materials prior to that time frame in consultation with the study team.

The principal investigator will share data, qualitative and quantitative analysis plans, and study materials with other researchers who provide a reasonable rationale for intended use, by email. Requests will be reviewed by the principal investigator Reschke-Hernandez, all co-investigators, and all study team members.

Participants and care providers (music therapists, reading group leaders) will not be masked; both will know what is being received (music therapy or reading). Principal investigator Reschke-Hernandez and Co-Is Schmitt and Abner will remain masked to allocation. Care staff who collect behavior data will not be masked; care staff will know if the facility is receiving music therapy or reading. However, for all other outcomes, research assistants who assess participants will be masked to group assignment without knowledge of the intervention assigned to individual participants. Allocation and assignment information will be kept in a separate secure document that is not accessible to the PI or Co-Is.