Related Experiment Video
Updated: Nov 22, 2025

11:50
Clinical Practice Protocol of Creative Music Therapy for Preterm Infants and Their Parents in the Neonatal Intensive Care Unit
Published on: January 7, 2020
27.1K
Communication improvement reduces BPSD: a music therapy study based on artificial neural networks
Alfredo Raglio1, Daniele Bellandi2, Luca Manzoni3
1Music Therapy Research Laboratory, Istituti Clinici Scientifici Maugeri IRCCS, Via Maugeri, 27100, Pavia, Italy. alfredo.raglio@icsmaugeri.it.
Summary
Active music therapy effectively reduces behavioral disturbances in dementia patients. Patient characteristics like the Barthel Index and improved communication predict positive outcomes, aiding clinical practice.
Area of Science:
- Gerontology
- Neuroscience
- Psychiatry
Background:
- Music therapy is recognized for improving behavioral disturbances, cognitive functions, and quality of life in dementia patients.
- Relational active music therapy specifically targets reducing behavioral disturbances by enhancing communication, particularly non-verbal cues.
Purpose of the Study:
- To explore baseline characteristics predicting positive outcomes in dementia patients undergoing music therapy.
- To investigate the relationship between behavioral disturbances and improvements in communication within the therapeutic intervention.
Main Methods:
- Utilized linear correlation and semantic connectivity maps to analyze 27 variables from 70 patients with moderate-severe dementia.
- Assessed predictive variables for response and complex connections between clinical factors and the relational aspect of music therapy.
Main Results:
- The Barthel Index was the primary predictor of positive response, followed by Neuropsychiatric Inventory (NPI) sub-items like Disinhibition and Depression.
- Improved communication/relationship was directly linked to being a 'responder,' also associated with younger age, higher Mini Mental State Examination scores, and female sex.
Conclusions:
- Active music therapy is appropriate for reducing behavioral disturbances in dementia.
- Unsupervised artificial neural networks can support clinical practice by identifying predictive factors and understanding intervention outcomes.

