SingFit Data

SingFit Study Pilots & Projects

​

An Equitable ML-Based Music Intervention for At-risk Older Adults

​

Although the adoption of digital health interventions has shown promise in reducing cognitive-related health disparities in rural communities, there remains a lack of consideration regarding rural culture in the design and implementation of such interventions. The purpose of this project is to develop an intelligent recommendation system capable of identifying the optimal therapeutic music components to elicit engagement and resonate with diverse rural-residing older adults at-risk for Alzheimer’s disease (AD). By creating a more comprehensive, individualized platform, SingFit can serve as a more inclusive and culturally relevant tool for at-risk older adults in rural communities who have been traditionally underrepresented in the development of music-based digital health interventions.

 

Specific Aims

1. Develop culturally inclusive user personas – We will identify the unique needs, preferences, and backgrounds of at-risk rural-residing older adults to learn more about their goals, challenges, and perspective regarding music-based digital health intervention.

2. Design an Intelligent Recommendation System – Develop a ML model using collaborative-based filtering that utilizes user data to identify optimal therapeutic music components for enhancing engagement and emotional resonance among rural-residing older adults at-risk for AD.

3. Assess acceptability of personalized therapeutic music sessions – Conduct an online survey to evaluate user feedback of the song/guided prompts generated by the recommendation system among a diverse group of rural-residing older adults at-risk for AD. Collect preliminary data on user acceptability (sentiment,cultural relevance, and perceived benefit).

 

Project Description

The project will leverage machine learning through the combination of data collected from surveys and user testing to inform and improve the app's usability, cultural relevance, and acceptability among rural-residing older adults at-risk for AD. The expected duration of the project is 12 months and will consist of two phases:

 

Phase I – Data collection and preprocessing. Data will be collected from 500 participants using an online survey. Specifically, we will collect feedback on one of SingFit’s session playlists designed to reduce mood symptoms associated with depression, a known modifiable risk factor for AD. Participants will assess 4 randomly selected sets of songs/guided prompts related to the session goal and provide ratings using a 5-point Likert scale (strongly disagree-strongly agree) for questions designed to measure perceived benefit, sentiment, relevance, and preference. Additional data regarding demographic characteristics and musical preferences will

also be collected. The data will be used to develop culturally inclusive user personas (e.g., segment analysis) and preprocessed in preparation for Phase II of the project.

 

Phase II – Collaborative-based filtering implementation and evaluation. Utilizing the Data Core services from MassAITC, a collaborative-based filtering technique will be implemented to build a recommendation system using the data collected from Phase I. This approach will allow the system to make recommendations based on the preferences of similar users in conjunction with our already established expert tailored system to maximize the accuracy and relevance of recommendations. The recommendation accuracy of the collaborative-based filtering algorithm will be assessed using a performance metric (e.g., root mean square), in addition to user survey data from a separate sample of 30 participants. Similar to Phase I, participants will be provided sets of songs/guided prompts generated by the recommendation system based on the target goal (Reduce Depression). Feedback will be used to evaluate sentiment, cultural relevance, and perceived benefit as well as determine areas for improvement.

​

​

An Evaluation of SingFit STUDIO's Utility for Promoting Well-Being and Independence among Older Adults

​

This quality improvement project aims to gather valuable user insights to enhance user experience, product features, accessibility, and overall utility. Participants in the project will include caregivers and their adult loved one or client (55 and older, mild-to moderate cognitive decline, taking at least 1 medication for cognition or mood-related symptoms). The data collected will be used to evaluate the utility of SingFit STUDIO.

 

Objectives

1. Assess the feasibility of SingFit including ease of implementation, acceptability as well as the adaptability to the home environment.

2. Analyze the cost-effectiveness of implementing SingFit as operationalized by appropriate hospitalization days, reduced caregiver and staff burden, and other related healthcare metrics.

3. Evaluate the effectiveness of SingFit STUDIO, focusing on cognitive engagement, behavioral management, and overall quality of life.

 

The quality improvement project is broken down into 2 parts and based on the the Institute for Healthcare Improvement’s Model for Improvement:

 

Part 1 - Rapid Cycle Improvement: Caregivers use the therapeutic music app with client/loved one for 4 sessions. Caregivers provides feedback about the app's usability, functionality, user

experience, and perceived benefit.

 

Part 2 - Plan-Do-Study-Act: 

A randomized quality improvement protocol will be implemented with a different

sample of caregivers who will use the therapeutic music app with client/loved one for approximately 24 sessions (a total of 100 caregiver-client pairs). Caregivers will be randomly assigned to their usual routine with their client/loved one vs. their usual routine augmented with the therapeutic music app. All caregivers will also receive biweekly health education materials. If

randomization is not possible, all caregivers will receive the app to augment their usual routine. At baseline and at the end of the project, the caregivers and their clients/loved ones will complete cognitive, mood, and quality of life assessment measures as well as report information regarding medication use and hospitalization frequency to assess the cost-effectiveness of the app. They will also provide feedback about the app's usability, functionality, user experience, and perceived benefit.

​

​

THE FRONT PORCH CENTER FOR TECHNOLOGY & INNOVATION

The Front Porch Center for Innovation and Wellbeing is a highly respected nonprofit that brings together the residents and staff of senior living communities with research partners and technology companies. The Front Porch Center undertook an eight-week study of the SingFit PRIME™ training, technology and protocols at their Claremont Manor Community in Claremont, California. 

 

Report Summary: “Over the course of 8 weeks, SingFit PRIME™ sessions were very effective with Memory Care and Assisted Living residents. Using a modified facial grimace scale to measure the impact of SingFit, we found that moods improved, memories were revived, and residents were more socially open with the people around them...the staff noted increased levels of engagement among assisted-living residents, and more so with those in memory care. Staff members also indicated that through these SingFit sessions, they were able to better get to know their residents on a deeper and [more] personal level.” 

 

Front Porch Grimace Scale Results: From the Front Porch Report: “Summer House (memory care) and Assisted Living residents did not fill out a survey about their feedback and expectations on SingFit. Instead, facilitators filled out a modified mood scale using the Wong-Baker Facial Grimace Scale (see figure below) to see how SingFit sessions impacted the residents, whether it would be after one session or over time. Across the three populations, it is clear that moods improved between 1-2 points after each SingFit session.” A lower number indicates an elevated mood. 

 

Qualitative Data on Benefits Over Time: From a staff report: “Another important finding worth noting was the residents’ ability to maintain their cognitive abilities. From observation, in the first week of the SingFit program at Summer House (memory care) it was very difficult to get the residents to sing or answer the trivia questions. Towards the end of the pilot, there were improvements such as being able to answer trivia questions, remembering song titles and/or lyrics, and being able to recall a personal memory. For example, in the beginning of the program, one resident could not speak clearly when answering trivia questions. Towards the end of the pilot, she was able to complete parts of a song lyric, moved more to the songs, and spoke more coherently.” 

​

SINGFIT PRIME™ PILOT AT 11 COMMUNITIES

One of the top five Senior Living Communities collected data over 12 weeks in 11 of their locations. The data was from 150 seniors with Dementia singing 3 times per week. SingFit PRIME dramatically increased the mood of residents . Here’s what they observed:

​

• 43% on average mood elevation in Memory Care

• 40% on average mood elevation in Assisted Living

 

The results were even more dramatic with residents who are consistently very sad at the beginning of sessions:

​

• 330% elevation in mood (from very sad to content)

• On average, residents sing, move, talk and smile often during a SingFit PRIME Session

• 89% “sang often” during a SingFit PRIME Session

 

Feedback from facilitators on the effect of SingFit PRIME on residents was positive and constructive on ease of implementation and acceptance of the program.

​

COMMUNITY STATS

• SingFit is in over 437 long term care communities

• SingFit has a retention rate of 95% in 2020 and 94% in 2021

• In a customer survey of 58 facilities, SingFit PRIME facilitators used SingFit approximately 2-3 times a week:

  • 98.3% saw a positive impact on their residents

  • 84.5% reported SingFit made their job easier

  • 91.4% felt SingFit made their job more fulfilling

​

SINGFIT PRIME™ CUSTOMER REPORTS ON OUTCOMES & BENEFITS

In Assisted Living & Community Care: SingFit PRIME™ customers represent the most innovative and respected dementia care service providers in the country. Below are reports from front line staff on how the SingFit PRIME™ program, powered by the SingFit iOS mobile app, creates positive outcomes for their residents. 

 

“Some of the people who hardly talk sing along to songs with SingFit. I have found that if a resident is agitated, SingFit can calm them. SingFit can also wake people up and aids sleep irregularities (for people who tend to sleep through activities all day). It helps residents who pace a lot to focus and participate even in the activities that come later in the day.”  
— Chris Saario, Memory Care Life Enrichment Director at Aegis Living of Marymoor 

 

“I read about SingFit in the AARP Bulletin that we receive monthly. I downloaded it on my iPad and played with it myself and brought it over to my mom at her skilled nursing facility. It was a huge hit. Now that my mother has passed, I cherish the recordings we made together with the SingFit app. Thank you.”   
— Diana Spearman, 57 

 

“Singing with guided lyrics makes singing songs we are not familiar with a lot easier. The musical part of the song helps to trigger the lyrics in the residents’ heads before the words start to be sung. Great program all around providing a boost in energy during the day.”
— Brittney Carson, Country Side Living.