Effects of a Resilience-Building Energy Management Program on Fatigue and Other Symptoms in Systemic Sclerosis: A Randomized Controlled Trial

Abstract

Objective.

Supported self-management interventions for individuals with systemic sclerosis (SSc) are needed. We examined effects of a 12-week resilience-building, energy management program (called RENEW) for fatigue and other patient-reported outcomes.

Methods.

Participants, who had physician-diagnosed SSc, moderate to severe fatigue, and were ≥ 18 years old, were randomly assigned to RENEW or waitlist control in a 2:1 ratio. The RENEW intervention included an educational website/app plus 9 virtual peer-led health coaching sessions. The primary outcome was change in the Functional Assessment of Chronic Illness Therapy-Fatigue (FACIT-Fatigue) scale. Secondary outcomes were change in Patient Reported Outcomes Measurement Information System (PROMIS) measures of pain interference and depressive symptoms and Connor-Davidson Resilience Scale. Outcomes were assessed at baseline, 6 weeks, and 12 weeks. Multiple imputation was conducted; linear mixed models were used to assess group differences. A three-way interaction with group, time, and SSc duration was examined in each model. This trial is registered with ClinicalTrials.gov, NCT04908943.

Results.

Among 173 participants (mean age 54.5 years ± 11.7; 93% female, 85% White), 47% had diffuse cutaneous SSc; 57% were ≤ 5 years from diagnosis. At 12 weeks, compared to controls, RENEW participants had a clinically meaningful fatigue improvement (β=−4.7, 95% CI: −6.7 to −2.7; p < 0.001) and improvement in all secondary outcomes. Among RENEW participants, individuals with shorter disease duration had greater improvements in fatigue at 12 weeks.

Conclusion.

A mHealth supported self-management intervention improved fatigue and other outcomes, particularly in newly diagnosed patients. This program may be broadly scalable for SSc symptom management.

Graphical Abstract

Individuals with systemic sclerosis (SSc) rate fatigue as one of their worst symptoms (1,2) and consider fatigue and other psychological needs to be insufficiently addressed in routine clinical care (3). Fatigue is difficult to address because it is multifaceted, co-occurring with pain (4) and sleep disturbance (5) and disease-related issues such as breathlessness (6) and gastrointestinal problems (7). Fatigue may also be influenced by personal non-modifiable factors and external situational stressors, such as financial insecurity and the Covid-19 pandemic (8). Although many contributors to fatigue are potentially modifiable, they may vary among individuals because of heterogeneous SSc symptom experience.

Some fatigue management interventions tested with rheumatologic samples have shown positive fatigue improvements. For instance, in trials including individuals with inflammatory rheumatic disease, fatigue management interventions utilizing cognitive behavioral approaches or individualized exercise programs and delivered by trained health professionals reduced fatigue compared to usual care (9,10). In various chronic conditions, there is strong evidence from systematic reviews that shows theory-based supported self-management approaches using mHealth (website/app platforms) improve fatigue and other symptoms (11–13). However, no trials in SSc have tested supported self-management interventions using mHealth for fatigue. Print and internet-based versions of a general self-management intervention for SSc have been developed (14–16), but a randomized controlled trial (RCT) evaluating the self-guided (uncoached) internet version showed no significant differences in symptoms relative to controls (14). While participants were highly satisfied with intervention content, they did not have access to a health coach, instruction on structured goal-setting, or opportunity to focus on fatigue. Thus, we felt an individually-tailored approach incorporating peer health coaching was needed to effectively address fatigue and other symptoms.

We designed the Resilience-building Energy management to Enhance Wellbeing (RENEW) intervention using a participatory approach with patient partners at University of Michigan. RENEW has theoretical grounding in self-efficacy theory (17) and positive psychology, which teaches people to more optimally respond to stressors to build resiliency (18). The purpose of this study was to examine effects of RENEW on fatigue and secondary outcomes (pain interference, depressive symptoms, and resilience) for individuals with SSc. We hypothesized that participants in the RENEW arm would have a clinically meaningful fatigue reduction following the 12-week program compared to those in the control group. We also hypothesized that pain interference, depressive symptoms, and resilience would improve more for participants in the RENEW arm.

Patients and Methods

We conducted a 12-week, parallel, single-blind RCT (NCT04908943) to evaluate the effectiveness of RENEW. The researchers and study statistician were blind to participants’ group assignment. Participants were randomly allocated into the RENEW intervention or waitlist control (2:1 ratio). Participants were recruited between August 2021 and March 2023 from the University of Michigan’s Scleroderma Clinic, other SSc specialty centers in the US and abroad, an existing SSc registry (n=~500), and flyers posted on SSc social media pages. Interested individuals were screened by phone for eligibility. People were eligible if they were ≥18 years, had physician diagnosed SSc, had moderate to severe fatigue (average score of ≥4/7 on the Fatigue Severity Scale) (19), had computer access and an internet connection, and were able to speak and read English. People were excluded if they were enrolled in other clinical trials during the study period, were undergoing rehabilitation or psychological treatment, or had other issues that precluded meaningful participation in study procedures (e.g., concurrent medical issues; extended travel). The study was approved by the Medical Institutional Review Board at the University of Michigan (HUM00195121) and conducted in accordance with Good Clinical Practice and Declaration of Helsinki.

Randomization.

Participants were randomly allocated into the RENEW group or waitlist control in a 2:1 ratio of intervention to control, with randomization stratified on whether the person received their clinical care at University of Michigan or not. Unequal allocation is relatively common in fatigue management trials (20) and was chosen for two reasons: to maximize participant feedback on the intervention being tested and to reduce dropout in the control group (21,22). Eligible participants signed an electronic consent form, completed baseline assessments online, and were then randomized. A blinded research staff member used a blocked randomization scheme generated by a statistician with randomly mixed block sizes of 3 and 6. When multiple people were enrolled at the same time, randomization order was determined by the timestamping of their baseline assessment completion.

Procedures.

After allocation, participants in the intervention group received guidance from research staff on how to access the RENEW website and app. They were matched with a trained peer health coach to schedule health coaching meetings via Zoom. Health coaches monitored participant progress through an online portal, providing feedback and maintaining communication during health coaching meetings. Phone calls or emails were used to reschedule sessions when needed. All participants, regardless of group assignment, were asked to complete outcome assessments at baseline, 6 weeks, and 12 weeks.

Intervention.

RENEW is a 12-week, online peer-supported fatigue self-management program for individuals with SSc. During development, some features were particularly important to us to incorporate. First, the positive psychology perspective was felt to be necessary because of its focus on wellness through bolstering self-efficacy, positive experiences, and emotions as opposed to focusing on reducing symptom burden or suffering (23). Such positive activity interventions encourage behavioral activation by inviting patients to engage in pleasant activities (24). This approach also posits that positivity can be learned through practice, and these behaviors can improve affect and resiliency (23,25,26). Second, program materials, including the RENEW website and app, were developed in partnership with individuals with SSc at the forefront to counter the typical provider-patient power dynamic. For example, health coaches and patients in the program supplied photos and audio files for the website and app, suggested educational resources, and co-designed the app. Health coaches were active partners in co-developing procedures for coaching and providing guidance for approaching goal setting given different participant situations and disease activity. Third, RENEW health coaches are trained peers who have SSc. Because SSc is rare and not commonly understood by others, we felt that those who share lived experiences of SSc while also serving as a positive role model would be impactful.

The RENEW program consists of modules covering topics including physical activity, pacing activities, relaxation techniques, practicing adaptive (positive) thoughts, taking care of one’s body, healthy diet, and sleep. Participants were paired with one of three peer health coaches (two females - one with diffuse cutaneous and one with limited cutaneous SSc - and one male with diffuse cutaneous SSc; all with disease duration >5 years). Health coaches were selected from the Scleroderma Peer Mentors Program that functions out of University of Michigan. The coaches were selected based on their involvement and leadership in the Peer Mentors Program. They were trained by a study principal investigator (SM) in motivational interviewing (27) and participated in a feasibility study of the RENEW intervention to practice with participants prior to the trial. Health coaches participated in creating a standard manualized protocol, practiced with mock participants, and met weekly with the PI to ensure readiness. All sessions were recorded for intervention fidelity checking. Participants had a total of nine health coaching sessions: weekly meetings via Zoom during the first six weeks followed by biweekly meetings during the second six weeks. Sessions lasted from 15 to-30 minutes. The health coach played an important role in introducing the program, guiding participants on accessing the RENEW website and app, and assisting them in setting and tracking health goals, particularly related to managing fatigue. Health coaches were instructed to refer participants to resources collected by our team if they were asked about specific SSc questions not related to health coaching (e.g., medications/other treatments).

Control group.

Participants in the control group completed the same outcome assessments as those in the intervention group and were asked to continue their usual routines throughout the study. At the end of 12 weeks, RENEW materials (website and app) were provided to participants. They were also invited to a single Zoom session with a health coach to guide them through goal setting and provide encouragement and instruction on adapting health behaviors.

Intervention fidelity.

Using an established fidelity checklist (See Supplementary Materials), two trained fidelity assessors viewed a random sampling of recordings by each health coach for each project quarter. For the first three quarters, 20% of recordings were reviewed. Due to consistently high fidelity across all health coaches, 10% of recordings were reviewed for the last four quarters. A total of 129 recordings were reviewed, yielding 98–99% fidelity across ten key coaching elements. When ratings were less than perfect, they were brought to the team to discuss with health coaches. The 1–2% infidelity was primarily explained by exceeding the allotted time allocated to spend with participants (>30 minutes) or making appropriate adjustments for participant circumstances (e.g., answering more questions during the first session or for newly diagnosed patients, not introducing a new goal for a patient struggling with current goals).

Measures.

Demographics and SSc characteristics.

Participants reported their age, sex, race, education level, employment status, and marital status. Race was ascertained by participant selection from the National Institutes of Health race categories and included responses of “other” or “prefer not to answer.” SSc subtype and SSc duration since date of diagnosis were collected through self-report or from medical records (if participants received care at University of Michigan).

Outcomes.

Participants were asked to provide data at baseline, 6, and 12 weeks via secure online survey. The primary outcome was change in Functional Assessment of Chronic Illness Therapy-Fatigue short form (FACIT-F) (28), a reliable, validated self-reported measure used in rheumatic disease populations and SSc studies (29,30). The FACIT-F is a 13-item questionnaire that assesses fatigue and its impact on daily activities on a 5-point Likert scale (1=not at all to 5=very much) over the previous 7 days. The raw score was converted to a T-score (US population mean ±SD, 50±10) based on Patient Reported Outcomes Measurement Information System (PROMIS) scoring; a higher score indicates higher perceived fatigue; a negative change score means improvement. Secondary outcomes included pain interference, depressive symptoms, and resilience. Pain interference was assessed using the PROMIS pain interference short form (31,32), shown to be reliable and valid in SSc (33–35). It consists of 4 items measuring how pain interfered with participants’ daily activities, work around the home, social activities, and household chores on a scale of 1=not at all to 5=very much. Depressive symptoms were evaluated using the PROMIS depression short form (31,36), which is reliable and validated in SSc populations (33–35), which included 4 items measuring feelings of worthlessness, helplessness, depressed mood, and hopelessness on a 5-point Likert scale (1=never to 5=always). Raw scores for both pain and depressed mood were converted to T-scores, where a higher score indicates a greater degree of the domain assessed and a negative change score means improvement. Resilience was assessed using the reliable, validated 10-item Connor-Davidson Resilience Scale (37,38). Participants rated their agreement with statements reflecting resilience-related traits, such as ability to adapt to change, handle stress, and bounce back from difficult situations on a 5-point Likert scale (0=not true at all to 4=true nearly all the time), with higher scores indicating more resilience. A positive change score means improvement. Finally, we used the Patient Global Impression of Change (PGIC) (39), a scale ranging from 1 (No change or condition has gotten worse) to 7 (A great deal better and a considerable improvement) to capture participants’ perceptions of change in daily activities, symptoms, and overall quality of life from baseline to 12 weeks.

Power analysis.

Sample size was determined prior to trial initiation. For 80% power, a sample size of 168 participants (112 RENEW/56 control; 2:1 ratio) was needed to detect a clinically important mean difference of 0.4 SD in fatigue (40). Sample size was adjusted to account for an estimated 15% dropout. To ensure diverse representation of males, Indigenous, and people of color, we attempted to oversample later in the project, which yielded an additional five participants for a total sample size of 173. Due to the project period ending, we stopped recruitment for oversampling at that point.

Statistical analysis.

Descriptive statistics were reported for sample baseline demographics and SSc characteristics by treatment group. Analyses were conducted among the intent-to-treat population. We expected each participant had their primary and secondary outcomes completed at baseline, 6 weeks, and 12 weeks (providing a total of 519 records). Most records, 464 (89.4%), had complete data; 3.5% had one measurement missing. We assumed that missing data occurred at random and applied multiple imputation. The imputation model included four outcome measures – fatigue, pain, depression, and resilience, and seven non-missing variables – treatment group, time, age, sex, race, SSc subtype, and SSc duration. Twenty imputed datasets were generated using the multivariate normal distribution method for all numeric variables and using the fully conditional specification method for categorical measures (41). Linear mixed models were fitted to compare change of domain score in RENEW vs. control group over time for primary and secondary outcomes. Each model fitted the change of domain score from baseline as outcome, and included baseline measurement, treatment group, time, treatment group*time, age, sex, race, SSc subtype, and SSc duration as fixed factors, and a random intercept term for participant to account for the correlation between repeated observations within an individual. Given that individuals with SSc commonly experience severe symptoms during the initial five years (42), we examined the moderating role of SSc duration (groups: 0–1 years, 2–5 years, or >5 years) by testing a three-way interaction (group*time*SSc duration) using linear mixed models for all outcomes. Effect sizes, defined as Cohen’s D, were calculated, and considered large when d=0.8 or higher, medium when d=0.5–0.8, and small when d=0.2–0.5 (43). To determine if participants achieved a minimal clinically important difference (MCID) in fatigue, the between-group difference value was compared to previously established thresholds, with a change of 3 points improvement considered a clinically important difference (44). The MCIDs for pain interference and depressive symptoms were defined as 4 points and 3 points improvement, respectively (40). Regarding resilience, although there is no clearly established threshold, studies determined the MCID for resilience to be half of the standard deviation (SD), which is nearly 10% change in score (45). The results of the linear mixed effect model and proportion of MCID from each imputed dataset were combined for the inference using Rubin’s rule (46). Calculation of effect sizes was based on observed data. Adherence was evaluated using participant attendance of health coaching sessions. We established a cutoff of ≥7 (of 9 total) as the criterion for determining adequate adherence (i.e., 78% of sessions attended). Attrition was determined by the number of participants who did not complete the 12-week survey or both the 6 and 12 week surveys. Participants were classified as withdrawn or lost to follow-up if they notified us of their intention to withdraw, were unresponsive to contact attempts, or missed several consecutive health coaching sessions due to various reasons (e.g., prolonged illness). Linear mixed models and MCID analyses were performed using SAS software version 9·4; calculation of Cohen’s D was performed using R Studio version 4.3.0. Other statistical analyses were conducted using SPSS version 27 (Armonk, NY: IBM Corp). The level of significance was set at p<0.05.

Results

Participants

Figure 1 shows the CONSORT flowchart. One hundred seventy-three participants met eligibility criteria and were randomized (115 to RENEW and 58 to control). Our participants were from diverse backgrounds, including 18 from seven different countries in addition to US (Canada [n=7], France [n=4], Australia [n=3], Guatemala [n=1], Israel [n=1], New Zealand [n=1], United Kingdom [n=1]). The mean age of participants was 54.5±11.7 (Table 1). Most participants were female (93%), White (85%), and married (67%). Nearly 60% attained a college degree or higher, and 44% were working full-time or part-time. Approximately 47% of the sample had diffuse cutaneous SSc. The mean number of years since diagnosis was 8.1±9.2, with 57% having a shorter disease duration, within 5 years of diagnosis. At baseline, the sample had higher fatigue (63.2±5.7) and pain interference (60.5±8.4) compared to the US norm, with a difference of 1.3 SD and 1.1 SD worse than the US norm, respectively. Depressive symptoms were nearly 0.5 SD worse than the US norm (54.7±7.6). The resilience score (26.2±6.5) was slightly lower compared to a large SSc cohort (n=962) who had a mean score of 27.8 (38). Overall, the sample experienced moderate fatigue and pain.

Figure 1:

Study Flowchart

^aIntent-to-treat refers to all randomized participants.

Table 1:

Baseline characteristics of the study participants (N = 173)

	Total	Control group (n = 58)	RENEW group (n = 115)
Demographics
Age, mean (SD), years	54.5 (11.7)	55.1 (12.5)	54.1 (11.3)
Sex, n (%)
Female	161 (93.1)	54 (93.1)	107 (93.0)
Male	12 (6.9)	4 (6.9)	8 (7.0)
Race, n (%)
White	144 (85.2)	51 (92.7)	93 (81.6)
African American/Asian/others	25 (9.5)	4 (7.3)	21 (18.4)
Education status, n (%)
High school/GED	14 (8.3)	7 (12.7)	7 (6.2)
Some college	54 (32.1)	14 (25.5)	40 (35.4)
College degree	50 (29.8)	19 (34.5)	31 (27.4)
Master’s or advanced degree	50 (29.8)	15 (27.3)	35 (31.0)
Work status, n (%)
Full-time employed	54 (32.0)	14 (25.9)	40 (34.8)
Part-time employed	20 (11.8)	8 (14.8)	12 (10.4)
Homemaker	11 (6.5)	4 (7.4)	7 (6.1)
Retired	40 (23.7)	14 (25.9)	26 (22.6)
On disability	44 (26.0)	14 (25.9)	30 (26.1)
Marital status, n (%)
Not married	56 (33.1)	19 (34.5)	37 (32.5)
Married	113 (66.9)	36 (65.5)	77 (67.5)
SSc characteristics
SSc subtype, n (%)
Diffuse	82 (47.1)	32 (55.2)	50 (43.5)
Limited	61 (35.1)	16 (27.6)	45 (39.1)
Overlap	22 (12.6)	8 (13.8)	14 (12.2)
Unspecified/other (e.g., SINE)	8 (5.2)	2 (3.4)	6 (5.2)
Disease duration, mean (SD), years	8.1 (9.2)	9.6 (10.5)	7.4 (8.4)
Disease duration in category
0 – 1 years, n (%)	40 (23.1)	13 (22.4)	27 (23.5)
2 – 5 years, n (%)	59 (34.1)	17 (29.3)	42 (36.5)
> 5 years, n (%)	74 (42.8)	28 (48.3)	46 (40.0)
PROMIS Anxiety, T-score, mean (SD)	58.7 (8.1)	59.8 (7.8)	58.1 (8.2)
PROMIS Sleep Disturbance, T-score, mean (SD)	56.1 (7.6)	56.4 (7.6)	56.0 (7.7)
PROMIS Social Isolation 8a, T-score, mean (SD)	51.8 (7.4)	51.9 (7.0)	51.8 (7.7)
PANAS Positive Affect	29.8 (7.3)	29.0 (7.3)	30.2 (7.3)
PANAS Negative Affect	22.5 (8.0)	23.3 (7.7)	22.1 (8.1)
Fatigue Severity Scale, score, mean (SD) †	5.6 (0.7)	5.7 (0.8)	5.5 (0.7)
FACIT Fatigue, T-score, mean (SD) ‡	63.2 (5.7)	64.4 (6.3)	62.6 (5.3)
PROMIS Pain interference, T-score, mean (SD)	60.5 (8.4)	61.6 (8.5)	59.9 (8.3)
PROMIS Depressed mood, T-score, mean (SD)	54.7 (7.6)	54.9 (7.6)	54.6 (7.6)
Resilience, score, mean (SD)	26.2 (6.5)	26.1 (5.6)	26.3 (6.9)

^*SSc = systemic sclerosis; PROMIS = Patient Reported Outcomes Measurement Information System; FACIT = Functional Assessment of Chronic Illness Therapy; PANAS = Positive and Negative Affect Schedule

^†Fatigue Severity Scale is a 9-item questionnaire that assesses perceived fatigue on a 7-point Likert scale (1 = strongly disagree to 5 = strongly agree). Eligible participants had a score ≥ 4 out of 7 on Fatigue Severity Scale.

^‡For FACIT fatigue, scores are T-scores based on PROMIS scoring of FACIT fatigue.

Among the RENEW group, 98 participants (85.2%) attended ≥7 health coaching sessions. The median number of attended sessions was 9; session duration mean, SD 24±5 minutes. The overall attrition rate was 11.0%, with 10 participants in RENEW and 9 participants in the control group having incomplete assessments (Figure 1). Compared to completers, participants who withdrew or were lost to follow-up were significantly more likely to be a racial minority (African American, Asian, or other; χ²=4.49, p=0.034), have lower educational attainment (high school or GED; χ²=9.13, p=0.028), and have higher pain interference (t [171] =−2.52, p=0.013). No significant differences were found in terms of group assignment, age, sex, marital status, employment status, SSc subtype, or SSc duration. Although differences did not reach statistical significance, participants who withdrew or were lost to follow-up reported higher fatigue (65.0 versus 63.0) and depressive symptoms (56.0 versus 54.5) at baseline when compared to completers. Resilience scores of participants who withdrew or were lost to follow-up and study completers were similar (26.6 versus 26.2). After completing the 12-week outcome assessment, seven participants in the control group (12.1%) opted to attend a health coach meeting.

Primary outcome.

At 12 weeks, participants in the RENEW group, compared to the control group, had significantly more fatigue improvement (β=−4.7, 95% CI: −6.7 to −2.7; p<0.001) (Table 2; Figure 2), which was considered clinically meaningful with a difference of ≥3 points. The effect size was large (d=0.83, CI: 0.48 to 1.19).

Table 2:

Effects of the RENEW intervention on fatigue, pain interference, depressive symptoms, and resilience at 12 weeks in individuals with SSc

	Unadjusted means (SE)	Unadjusted group difference mean with 95% CI			Adjusted† between-group differences estimated least square mean with 95% CI
Outcome	Baseline	6 weeks	12 weeks	Change in RENEW versus Change in Control at 12 weeks	Change in RENEW versus Change in Control at 12 weeks
Fatigue				−3.6 (−5.8 to −1.3)	−4.7 ( −6.7 to −2.7)
RENEW	62.6 (0.49)	58.7 (0.60)	56.7 (0.69)
Control	64.4 (0.82)	62.2 (0.95)	62.1 (0.97)
Pain interference				−1.7 (−4.2 to 0.9)	−2.8 (−5.2 to −0.4)
RENEW	59.9 (0.78)	58.0 (0.80)	57.0 (0.83)
Control	61.6 (1.11)	61.0 (1.20)	60.3 (1.18)
Depressive symptoms				−3.5 (−6.1 to −1.0)	−3.7 (−6.2 to −1.2)
RENEW	54.6 (0.71)	52.3 (0.80)	51.4 (0.74)
Control	54.9 (1.00)	54.8 (1.13)	55.3 (1.18)
Resilience				2.7 (0.7 to 4.8)	2.6 (0.8 to 4.5)
RENEW	26.3 (0.65)	28.3 (0.57)	29.0 (0.59)
Control	26.0 (0.74)	25.3 (1.04)	26.0 (0.93)

^*SE = standard error; CI = confidence interval; imputed data was used for this analysis

^†Adjusted for age, sex, race, SSc subtype, SSc duration, and baseline values of each outcome.

Figure 2:

Least square mean changes over the study period for fatigue by groups

*Graphs present mean ± 95% confidence interval. For fatigue change scores reported here are T-scores based on Patient Reported Outcomes Measurement Information System scoring of Functional Assessment of Chronic Illness Therapy-Fatigue. Imputed data was used for this analysis

Secondary outcomes.

At 12 weeks, participants in the RENEW group, compared to the control group, also had significantly improved pain interference (β=−2.8, 95% CI: −5.2 to −0.4; p=0.023), depressive symptoms (β=−3.7, 95% CI: −6.2 to −1.2; p=0.004), and resilience (β=2.6, 95% CI: 0.8 to 4.5; p=0.006). Effect sizes were moderate for depressive symptoms (d=0.57, CI: 0.22 to 0.92) and resilience (d=−0.54, CI: −0.90 to −0.18) and small for pain interference (d=0.43, CI: 0.08 to 0.77).

Moderating role of SSc duration

Within the RENEW group, change in fatigue at 12 weeks was found to be significantly moderated by SSc duration. Participants with 0–1 year SSc duration had significantly greater fatigue improvement at 12 weeks compared to those with >5 years SSc duration (β=−3.2, 95% CI: −6.1 to −0.4; p=0.027) (Figure 3). No other three-way interaction effects were found.

Figure 3:

The moderation effect of SSc duration on fatigue in RENEW intervention group

*Graphs present mean ± 95% confidence interval. Imputed data was used for this analysis

Minimum clinically important difference

More participants in the RENEW group achieved a MCID for all outcomes at 12 weeks compared to the control group, and these differences were statistically significant. In the RENEW arm, 63.4% participants achieved MCID for fatigue compared to 41.9% in the control (p=0.008). As for pain interference, 41.9% participants in the RENEW arm achieved MCID compared to 25.9% in the control group (p=0.042). Regarding depressive symptoms, 52.2% participants in the RENEW arm achieved MCID compared to 32.1% in the control group (p=0.013). The MCID for resilience was determined as a 3-point improvement, which was calculated as half of the SD for the entire sample at baseline. Consequently, 48.3% participants in the RENEW arm achieved MCID compared to 23.9% in the control group (p=0.001).

Patient Global Impression of Change (PGIC).

PGIC differed in favor of the RENEW group relative to the control group at 12 weeks. Seventy-nine participants (76.0%) in the RENEW group versus three participants (6.3%) in the control group reported moderately improved symptoms with a score of at least 5 (Moderately better, and a slight but noticeable change).

DISCUSSION

This is the first adequately powered RCT in SSc to demonstrate a significant, clinically meaningful fatigue reduction using a supported self-management intervention delivered via mHealth. In a sample of 173 participants, we found that, compared to a waitlist control group, participants of the RENEW intervention not only had clinically significant improvements in fatigue but also improvements in pain interference, depressive symptoms, and resilience at 12 weeks. Over 75% of RENEW participants perceived an impactful change to their overall symptoms, function, or quality of life throughout the study compared to 6% in the control group. While RENEW participants improved steadily over 12 weeks, control group participants maintained the same or worse symptom levels.

While we included patients with moderate to severe fatigue who had potential for fatigue reduction, the intervention may have been effective because the intervention was intentionally constructed with components shown to increase treatment effects, such as being grounded in behavioral theory, providing guidance by a health coach, and incorporating structured goal-setting (10–13). Our team also worked in tandem with patient partners and used a participatory process in intervention development and refinement, which may have increased relevance and utility of the RENEW intervention. In the context of rare disease, a focus on positive role models and approaches to health behaviors may have contributed to positive effects on depressive symptoms and resilience. Given that adaptive coping and strategies for enhancing positive emotions and behaviors are teachable, this approach has potential to greatly improve engagement in positive health behaviors (23,25,26), and the positive psychology resilience-building focus may be important to integrate into chronic condition self-management approaches.

The inclusion of a peer health coaching component appears to have played a significant role in observed improvements in fatigue and other symptoms in RENEW participants relative to existing SSc self-management interventions (14–16). Peer coaching health interventions have demonstrated effectiveness in improving health behaviors and health outcomes (47,48). In terms of fatigue improvement, peer health coaching may assist with emotional aspects of having SSc. One health coach reported, “I have learned that most people are lonely in this disease.” As social isolation is common in SSc (49), the relationship building and social support that comes with having a peer health coach may be an integral part of this intervention.

Interestingly, among RENEW participants, newly diagnosed individuals experienced the most fatigue improvement. Notably, there were no significant differences in baseline fatigue by SSc duration. Although newly diagnosed patients often begin SSc treatments that could potentially contribute to lower fatigue, we would expect to see similar improvements in newly diagnosed individuals in the control group. The RENEW intervention may have a disproportionate positive impact on newly diagnosed individuals because it helps them with commonly reported issues of not knowing what the disease course will be like, how to advocate for themselves, or symptom unpredictability (49). The SSc-focused materials and peer health coaching may be valuable strategies to help individuals early in the disease. Of note, effects of this intervention did not differ by SSc disease subtype, which we expected as fatigue does not differ by disease subtype (50). Nonetheless, this finding suggests that even if there are unique contributors to fatigue, the fatigue management approach may not need to differ by SSc disease subtype.

Future Directions

Considering the positive results, future research is planned to evaluate effectiveness of the RENEW intervention in real-world settings where it could be delivered and sustained over time. It will be important to examine if the peer health coaching provided in the intervention is the driver of positive effects. If this peer health coaching, performed in real-time, is integral to effectiveness, as opposed to asynchronous support, infrastructure would be needed to train and oversee peer health coaches to ensure integrity of RENEW administration. In this study, the technology of using the website, app, and virtual health sessions offered the opportunity to recruit English-speaking participants worldwide. The possibility of broad dissemination exists, but electronic devices and internet access, both with the capability for videoconferencing, will still be required.

Limitations

We have limited generalizability of study findings to males with SSc. We had a low number of males in the sample (n=12, 7%), despite targeting them in recruitment and including a male peer health coach on the team. Further qualitative studies are needed with male participants to examine perceptions of RENEW and to determine if modifications need to be made for this group. Additionally, our sample was well-educated, so findings may not be generalizable to those with lower educational attainment. As with any behavioral intervention with a waitlist control group, it is possible that attention bias contributed to differences in outcomes. Studies including an inert education group may be warranted in the future to fully examine improvements in patient-reported outcomes.

Conclusion

This study provides evidence that a supported self-management approach delivered via mHealth improved fatigue in SSc and had effects on other symptoms and psychological outcomes. This approach may be a useful adjunct to existing clinical care to provide holistic SSc management, particularly for newly diagnosed individuals.

Significance and Innovations.

• A supported self-management program using peer health coaching, in which we trained individuals with systemic sclerosis to lead coaching sessions, was feasible and highly valued by participants.

• This is the first adequately powered clinical trial in systemic sclerosis that addressed energy management and had positive results on fatigue and other outcomes.

• The website and app developed in partnership with individuals with systemic sclerosis may be important tools for symptom management.