Music Attuned Technology: Care via eHealth (MATCH): A proof‐of‐concept study trialling a music therapy informed mobile application for caregivers of people living with dementia

Abstract Background MATCH (Music Attuned Technology – Care via eHealth) is a music and health application that supports caregivers of people living with dementia to use music strategically to better manage care through virtual training and intuitive music technology. This study trialled a prototype version of the MATCH app with family caregivers and people with dementia residing in the community. Method 16 Dyads trialled the prototype MATCH app. After completing the virtual training in the app, family caregivers created personalised playlists within the app and used strategies learnt during training with the person they care for at least twice per week over 8 weeks. Pre‐post measures on the Neuropsychiatric Inventory Questionnaire (NPI‐Q), Cohen‐Mansfield Agitation Inventory and Acceptability of the mobile app and training were captured to detect change in symptoms and feasibility of the application. Qualitative measures including a participation diary and qualitative interviews captured participant perspectives and usage of the app. Results Significant pre‐post changes in symptom severity (‐4.23 [4.9], 95% CI ‐7.2 to ‐1.3, p = 0.009) and symptom distress (‐5.69 [5.8], 95% CI ‐9.2 to ‐2.2, p = 0.004) on the NPI‐Q were found, but not on the Cohen‐Mansfield Agitation Inventory (‐4.38 [11.6], 95% CI ‐9.9 to 1.34, p = 0.109). Subscales of the NPI‐Q found significant changes to mood symptoms (‐1.23 [1.9], 95% CI ‐2.36 to ‐0.01, p = 0.036) and frontal symptoms (‐1.45 [23], 95% CI ‐2.87 to ‐0.05, p = 0.043), but not agitation/aggression (‐1.1 (1.9], 95% CI ‐2.21 to 0.07, p = 0.062). 85% of carers were satisfied with the MATCH app features, and 92% were satisfied with the music training. Qualitative findings revealed that participants found the strategies to be effective, and that they were able to integrate music‐based strategies into their care routines. Conclusion The MATCH app was found to be feasible and acceptable; family caregivers reported that MATCH training was understandable and useful. There is preliminary evidence that music interventions implemented by trained family caregivers are effective in reducing neuropsychiatric symptoms.

Authors

Institutions

Publication Details

Journal
Alzheimer s & Dementia
Published
2024-12-01
DOI
https://doi.org/10.1002/alz.084857
Citations
1
Primary Topic
Music Therapy and Health
Type
article
Field-Weighted Citation Impact
0.64
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Music Attuned Technology: Care via eHealth (MATCH): A proof‐of‐concept study trialling a music therapy informed mobile application for caregivers of people living with dementia

Dianna Vidas, Zara Thompson, Phoebe A Stretton-Smith, Robyn Woodward‐Kron et al.
1 citations
Alzheimer s & Dementia
Music Therapy and Health
0.64
article

Music Attuned Technology: Care via eHealth (MATCH): A proof‐of‐concept study trialling a music therapy informed mobile application for caregivers of people living with dementia

Dianna Vidas, Zara Thompson, Phoebe A Stretton-Smith, Robyn Woodward‐Kron, Tanara Vieira Sousa, Nicola T. Lautenschlager, Jenny Waycott, Efstratios Skafidas, Amit Lampit, Jeanette Tamplin, Lars Kulik, Felicity A. Baker
article en
1 citations

Abstract

Abstract Background MATCH (Music Attuned Technology – Care via eHealth) is a music and health application that supports caregivers of people living with dementia to use music strategically to better manage care through virtual training and intuitive music technology. This study trialled a prototype version of the MATCH app with family caregivers and people with dementia residing in the community. Method 16 Dyads trialled the prototype MATCH app. After completing the virtual training in the app, family caregivers created personalised playlists within the app and used strategies learnt during training with the person they care for at least twice per week over 8 weeks. Pre‐post measures on the Neuropsychiatric Inventory Questionnaire (NPI‐Q), Cohen‐Mansfield Agitation Inventory and Acceptability of the mobile app and training were captured to detect change in symptoms and feasibility of the application. Qualitative measures including a participation diary and qualitative interviews captured participant perspectives and usage of the app. Results Significant pre‐post changes in symptom severity (‐4.23 [4.9], 95% CI ‐7.2 to ‐1.3, p = 0.009) and symptom distress (‐5.69 [5.8], 95% CI ‐9.2 to ‐2.2, p = 0.004) on the NPI‐Q were found, but not on the Cohen‐Mansfield Agitation Inventory (‐4.38 [11.6], 95% CI ‐9.9 to 1.34, p = 0.109). Subscales of the NPI‐Q found significant changes to mood symptoms (‐1.23 [1.9], 95% CI ‐2.36 to ‐0.01, p = 0.036) and frontal symptoms (‐1.45 [23], 95% CI ‐2.87 to ‐0.05, p = 0.043), but not agitation/aggression (‐1.1 (1.9], 95% CI ‐2.21 to 0.07, p = 0.062). 85% of carers were satisfied with the MATCH app features, and 92% were satisfied with the music training. Qualitative findings revealed that participants found the strategies to be effective, and that they were able to integrate music‐based strategies into their care routines. Conclusion The MATCH app was found to be feasible and acceptable; family caregivers reported that MATCH training was understandable and useful. There is preliminary evidence that music interventions implemented by trained family caregivers are effective in reducing neuropsychiatric symptoms.

Alzheimer s & DementiaVol. 20(S4)
The University of Melbourne (AU)
Openalex Percentile: Top 24%
Music Therapy and Health
0.64
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.