Scleroderma
Conditions
Keywords
Systemic Sclerosis
Brief summary
The purpose of this pilot study is to explore the use of a large language model (LLM) in providing education and behavioral health coaching for individuals with Systemic Sclerosis (SSc). The goal of the LLM is to help user set or modify behavioral goals, provide education, or emotional support as needed by the participant. The primary outcome for this study is to assess the feasibility and acceptability of using an AI-supported health coaching tool over a four-week period.
Detailed description
The study is supported by Dan Barry Trainee Research Program through the Department of Physical Medicine and Rehabilitation at Michigan Medicine.
Interventions
BEHAVIORALRENEW-AI
AI assisted health coaching using the LLM-driven chat functionality.
Sponsors
University of Michigan
Study design
Allocation
NA
Intervention model
SINGLE_GROUP
Primary purpose
SUPPORTIVE_CARE
Masking
NONE
Masking description
The study team members who interact with participants during the enrollment process will be blinded to participant feedback provided during the study to avoid any influence on the participants' experience or outcomes.
Intervention model description
Phase 2 will involve a single-arm, pre-post study design where 20 participants of the original RENEW trial will use the NLP-chat feature integrated into the RENEW app over a 4-week period.
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Participants from our previous clinical trial, who received the original RENEW intervention (n=66), will be invited to participate in this study. * Participants who have a medical diagnosis of systemic sclerosis, any subtype (diffuse, limited, Sine, or overlap) * Participants who report a mean score 4 or more on the Fatigue Severity Scale, indicative of problematic fatigue; potential mean range 1-9 * Participants who have access to a computer and an internet connection * Participants who are able to speak and read English.
Exclusion criteria
* Individuals who did not participate in the original RENEW trial * Individuals who lack access to reliable high-speed internet will be excluded to ensure they can fully engage with the AI-supported health coaching and digital components of the study * Non-English speakers will be excluded because the current version of the AI tool is available only in English * Individuals planning to initiate any new treatments specifically targeting fatigue, pain, or mood symptoms (such as psychological therapy, structured rehabilitation, or new medications) during the 4-week study period will be excluded, to avoid confounding effects on the study outcomes. * Individuals with other significant medical or logistical issues that would impede meaningful participation, such as severe concurrent medical conditions, inability to access the RENEW intervention, or cognitive impairments that prevent them from engaging with study procedures, will also be excluded.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Engagement with the AI chat feature | Up to 4 weeks | Engagement is measured through app analytics (It will be reported as the proportion of participants who use the chat feature.) |
| Satisfaction with AI-generated health coaching | 4 weeks | Satisfaction is self-reported on the mhealth satisfaction survey v 1. It is a 14-item survey with Likert like questions ranging from 1, strongly disagree to 5, strongly agree. Higher scores imply more satisfaction |
| Acceptability of RENEW | 4 weeks | Assessed via focus groups |
Other
| Measure | Time frame | Description |
|---|---|---|
| Change in Fatigue | Baseline, week 4 | Assessed using the Functional Assessment of Chronic Illness Therapy-Fatigue short form (FACIT-F) is a 13 item subscale that allows patient to rate the intensity of their fatigue and its related symptoms on a scale of 0 (not at all) to 4 (very much). The total calculated score varies from 0-52 with a lower score indicating a more severe fatigue level. |
| Change in Resilience | Baseline, week 4 | Assessed using the Connor-Davidson Resilience Scale - 10 item Likert like questions ranging from 0, not true at all to 4, true nearly all the time Higher scores imply more resilience |
| Change in Pain interference | Baseline, week 4 | Assessed using the PROMIS Pain Interference short form - 4 item Likert like questions ranging from 1, not at all to 5, very much. Higher scores imply more pain interference |
| Change in Depressive symptoms | Baseline, week 4 | Assessed using the PROMIS Depression short form - 4 item Likert like questions ranging from 1, never to 5, always Higher scores imply more depression |
Countries
United States
Outcome results
None listed