Feasibility and preliminary outcomes of a remotely-delivered exercise training intervention for English-speaking Hispanics/Latinos with multiple sclerosis: Results from the FERLA MS study

Abstract

Background

There is limited evidence on the feasibility of remotely delivered exercise interventions for English-speaking Hispanics/Latinos with MS.

Objectives

To evaluate the feasibility and exploratory efficacy of a remotely-delivered, home-based exercise program for English-speaking Hispanics/Latinos with MS and to explore associations among feasibility metrics, outcomes, and social determinants of health (SDOH).

Results

Of 340 individuals contacted, 112 were eligible (33%), 33 enrolled (9.7%), 30 randomized, and 26 completed (87% retention). Adherence averaged 81% (exercise) and 89% (control). No serious adverse events occurred. Exploratory within-group improvements were observed in the exercise group for physical activity (p = 0.03, d = 0.66). Mobility improved by 10.1 points on the MS Walking Scale–12, exceeding the minimal clinically important difference. Exploratory analyses indicated that higher education correlated with greater program compliance (ρ=0.70, p = 0.005) and lower anxiety (ρ=-0.69, p = 0.006).

Conclusions

Our results support the feasibility, safety, and preliminary benefits of remotely-delivered, home-based exercise training for improvements in physical activity and mobility for English-speaking Hispanics/Latinos with MS. Exploratory results warrant evaluation in larger, adequately powered trials.

Clinical Trial Registration

ClinicalTrials.gov (NCT05998616).

Keywords

Multiple sclerosis exercise training feasibility trial Hispanic/Latino social determinants of health

Introduction

Nearly 1 million adults in the United States live with multiple sclerosis (MS).¹ Age-adjusted prevalence is lower among Hispanics/Latinos than non-Hispanic Whites (160 vs. 374 per 100,000), with intermediate estimates among non-Hispanic Black individuals (298 per 100,000).² Despite lower prevalence, incidence is increasing in Hispanic/Latino populations and is accompanied by earlier onset, faster disability accumulation, and more aggressive disease progression.³ These disparities may reflect social determinants of health (SDOH), including socioeconomic status, health literacy, and access to care, which influence diagnosis, treatment, and long-term management.

Despite this growing disease burden, Hispanics/Latinos with MS remain underrepresented in randomized controlled trials (RCTs), including disease-modifying therapy and rehabilitation studies. Fewer than 10% of participants in MS rehabilitation trials are Hispanic/Latino, and no participants in exercise training trials have been reported (Fakolade et al., 2023; Lai et al., 2018). This mismatch limits generalizability and contributes to gaps in culturally responsive symptom management strategies. Recent consensus statements emphasize the need to improve diversity, equity, and inclusion in MS research and care.⁴

Exercise training offers safe, efficacious benefits for MS function, symptoms, and quality of life.^5,6 Nevertheless, there is limited racial/ethnic diversity of the samples in exercise training trials, and this restricts applicability and generalizability of results to non-White/Caucasian populations. Further, SDOH likely affect participation, adherence, and outcomes, but are underexamined in MS exercise research. We initiated a research line examining physical activity behavior in Hispanics/Latinos with MS. Our research revealed significantly lower physical activity levels than the general population, and levels were associated with worse mobility, fatigue, cognition, and depression.⁷ We further identified social-cognitive theory (SCT) variables as correlates of physical activity behavior,⁸ and our most recent qualitative study detailed exercise preferences among Hispanics/Latinos with MS.⁹

As a next step, we proposed a Stage I, NIH-defined trial that examined the feasibility and efficacy of a remotely-delivered exercise intervention for English-speaking Hispanics/Latinos with MS.¹⁰ The intervention was adapted from the Guidelines for Exercise in Multiple Sclerosis (GEMS) program,¹¹ a home-based exercise training protocol based on the exercise guidelines for MS,¹² and supplemented with SCT behavior-change techniques and one-on-one coaching to support adherence. GEMS has demonstrated strong feasibility and efficacy in Caucasians¹³ and African-Americans¹⁴ with MS. To that end, we tailored GEMS to Hispanics/Latinos preferences⁹ for the Feasibility and Efficacy of Exercise Remotely-Delivered for Hispanics/Latinos with MS (FERLA MS) study.¹⁰ This paper reports FERLA MS feasibility, efficacy, and their SDOH associations.

Materials and methods

Study design

Full methodology is described in the published protocol.¹⁰ Procedures were approved by the University of Illinois Chicago IRB (2023–0665), and participants provided informed consent. This parallel-group RCT feasibility study (April 2024–June 2025) was registered at ClinicalTrials.gov (NCT05998616) and terminated early due to NIH funding revocation.

Participants and recruitment

Participants were recruited from the Chicagoland university, outreach events, and advertisements (March 2024–February 2025). Eligibility included adults aged 18–65 years with physician-confirmed MS, relapse-free ≥30 days, ambulatory with/without assistive devices, and <150 min/week of moderate-to-vigorous physical activity. Exclusion included non-Hispanic/Latino self-identification, limited English proficiency, residence outside Chicagoland, no internet/email access, inability to complete Zoom-based assessments, or failure to clear the Physical Activity Readiness Questionnaire.¹⁵

Randomization and study procedures

Baseline assessments comprised remote tests of physical and cognitive function and online surveys assessing health history, demographics, clinical information, physical activity, and SDOH. Participants were randomly allocated (1:1) to the exercise intervention or flexibility control. The intervention adapted the GEMS protocol,¹¹ and the control used the National MS Society (NMSS) Stretching for People with MS: An Illustrated Manual. Adaptations included both structural and cultural tailoring. Structural adaptations included flexible scheduling and remote delivery to reduce transportation and time-related barriers. Cultural adaptations included SCT-based newsletters with family-oriented narratives and culturally relevant scenarios to enhance relatability. Coaching emphasized culturally responsive communication, rapport-building, and recognition of family, work, and structural factors influencing participation. Both groups completed 16 weeks with scheduled behavioral coach chats. Post-intervention assessments mirrored baseline procedures and participants received two $50 gift cards for completing baseline and final assessments.

Intervention and control conditions

Detailed procedures are described in the protocol paper¹⁰ and summarized in Table 1. Intervention participants completed 16 weeks of progressive aerobic and resistance training. Aerobic training involved moderate-intensity walking monitored with a pedometer (CW-300 Yamax Digi-Walker™) and resistance training included upper-/lower-body, and core exercises with elastic bands. Materials included a manual, logbook, calendar, six SCT-based newsletters, and eight coaching sessions.

Table 1.

Intervention and control components.

Feature	Exercise intervention	Active control
Prescription	Prescriptive Guidelines for Exercise in MS	NMSS Illustrated Stretching Manual for People with MS
Prescription	Individually-tailored based on level of comfort/trainability	Individually-tailored based on level of comfort/trainability
Frequency	3 times per week	3 times per week
Exercise Session Duration	Aerobic: Gradual increase in minutes over time (10–30 min)	Gradual increase in duration of stretch and range of motion exercise 15–60 s per exercise, per category
Exercise Session Duration	Resistance: Gradual increase in sets and reps (1–3 sets, 10–15 reps)
Exercise Session Intensity	Aerobic: Moderate intensity, 100 steps/min (pedometer), modified RPE between 4–6	Light to Moderate Intensity
Exercise Session Intensity	Resistance: Moderate intensity, can finish 10–15 reps of last set comfortably	Light to Moderate Intensity
Intervention Length	16 weeks	16 weeks
Meeting with Coach
Weeks 1–4	Weekly	Weekly
Weeks 5–8	Bi-weekly	Bi-weekly
Weeks 9–16	Monthly	Monthly
Setting	Home and/or community	Home and/or community
Supervision
Who	Trained personnel	Trained personnel
Mode	Remote, teleconferencing	Remote, teleconferencing
Exercise Modes	Aerobic – Walking	Stretching – dynamic and static
Exercise Modes	Resistance – Bands and body weight	Stretching – dynamic and static
Materials Provided	Program manual	Program manual
	Newsletters	Newsletters
	Logbook	Logbook
	Calendar	Calendar
	Online video demonstrations	Online video demonstrations
Equipment Provided	PedometerBands	Yoga mat
Training Oversight	Zoom/Telephone	Zoom/Telephone
Behavioral Intervention	Zoom/Telephone	Zoom/Telephone
Adaptations for Hispanic/Latino Participants	Flexible scheduling	Flexible scheduling
	Remote delivery	Remote delivery
	Culturally adapted SCT newsletters (family-oriented narratives, culturally recognizable names and community references)	Culturally adapted SCT newsletters (family-oriented narratives, culturally recognizable names and community references)
	Culturally responsive coaching	Culturally responsive coaching
Intervention Safety	NMSS resources	NMSS resources
	Fall prevention instructions	Fall prevention instructions
	Zoom/Telephone oversight and AE/SAE reporting	Zoom/Telephone oversight and AE/SAE reporting

Note: NMSS, National Multiple Sclerosis Society; Rep, repetition; RPE, rate of perceived exertion; AE, adverse event; SAE, serious adverse event.

Control participants completed 16 weeks of progressive stretching using the NMSS manual, with an identical coaching schedule. Materials included a manual, logbook, calendar, newsletters, and a yoga mat.

Outcome measures

Outcomes are detailed in the protocol paper¹⁰ and Table 2. Feasibility metrics included process (recruitment, retention, adherence, compliance), resource (study communication, monetary requirements), management (data collection, analysis, storage), and scientific domains (safety, adverse event handling, participant burden). Adherence was defined as sessions completed, and compliance as proportion performed as prescribed. For the intervention group, a session was considered compliant if ≥75% of prescribed aerobic minutes/steps and ≥75% of prescribed resistance exercises were completed. For the control group, compliance was defined as ≥75% of prescribed flexibility categories per session.

Table 2.

Feasibility metrics, outcomes, and definitions.

Metrics	Outcomes and definitions
Process: assesses participant recruitment and retention	a) Recruitment rate: the number of participants screened divided by the total number of participants contactedb) Retention rate: the number of consenting participants who completed 16-week programs divided by the number of participants who were randomizedc) Adherence rate: the number of sessions completed over the 16-week period divided by the total number of prescribed sessions (i.e., 48) Compliance: the number of sessions completed as prescribed
Resources: assesses communication and monetary requirements of the study	d) Communication time with participants: time required for screening phone calls and Zoom coaching sessionse) Communication needs of participants and staff: whether additional contact approaches were neededf) Monetary costs: total cost in US dollars of training materials (i.e., manuals, logbooks, newsletters, calendar), training equipment (i.e., pedometers, resistance bands, yoga mats), postage and packaging, and participant compensation
Management: assesses data management, analysis, and storage	g) IRB approval procedures and amendments: communications between University IRB and staff, and time from submission to approval notificationh) Participant data collection, analysis and storage: time needed to document, clean, analyze and store data
Scientific: assesses trial safety, handling of adverse and serious adverse events, clinical emergencies, and participant burden	i) Reporting and handling adverse events, serious adverse events, and clinical emergencies such as unexpected problems (e.g., injuries) and health issues (e.g., MS relapse, symptoms exacerbation, illness) reported by the participants over the course of the study periodj) Number of events and emergencies, including frequencies and reasonsk) Participant's experienced burden throughout the program

Efficacy outcomes included physical and cognitive function, physical activity, symptoms, and quality of life. Physical function was assessed using 30-s sit-to-stand (30STS) and MS Walking Scale-12 (MSWS-12); cognitive function using Symbol Digit Modalities Test (SDMT) and California Verbal Learning Test II (CVLT-II). Disability and symptoms were measured with Patient Determined Disease Steps (PDDS), Fatigue Severity Scale (FSS), and Hospital Anxiety and Depression Scale (HADS). Physical activity was self-reported using Godin Leisure-Time Exercise Questionnaire (GLTEQ) and health-related quality of life (HRQOL) via Short-Form 12-Item Health Survey (SF-12).

Individual SDOH domains included birthplace, biological sex, gender identity, sexual orientation, race/ethnicity, education, household size/income, and employment status. Structural SDOH domains included health insurance coverage, access to health services, housing, neighborhood, transportation, and food insecurity.

Formative feedback was collected via brief post-intervention Zoom interviews. Participants responded to open-ended questions on program experiences, including perceived benefits, challenges, barriers, and suggestions for enhancing remote delivery and cultural relevance. Interviews were recorded and analyzed using content analysis to identify recurring themes for descriptive synthesis.

Statistical analysis

Analyses were conducted using SPSS (Version 24, IBM Corp.). Descriptive statistics summarized sample characteristics and feasibility metrics. Baseline differences were examined via independent t-tests or chi-square tests. Exploratory between-group comparisons were conducted using analysis of covariance (ANCOVA), with group (exercise vs. control) as the fixed factor and adjustment for baseline outcome values, age, and sex. Adjusted mean differences with 95% confidence intervals and Cohen's d were reported. Within-group pre–post changes were examined as secondary exploratory analyses using paired t-tests. Associations between SDOH variables and outcomes were assessed using Spearman's ρ. All tests were two-tailed (α = .05). Given the feasibility design and small sample size, SDOH analyses were conducted as exploratory, hypothesis-generating correlations. Correlations meeting conventional thresholds for moderate magnitude (|ρ|≥0.30) and p < .05 were highlighted for interpretive purposes. Sensitivity analyses adjusting for education were conducted to assess potential confounding.

Results

Feasibility outcomes

A total of 340 individuals were contacted (Figure 1). Baseline characteristics (Table 3) indicated higher education years in the intervention group than in the control (p = 0.009). Of 340 individuals contacted for eligibility, 112 were eligible (33%), and 33 consented and enrolled (66% of planned N = 50). The study was terminated due to funding withdrawal before full recruitment was achieved. Three participants withdrew pre-randomization, 30 were randomized, and 26 completed follow-up (86.7%).

Figure 1.

CONSORT flow diagram.

Table 3.

Baseline characteristics of participants overall and by conditions.

Variables	Overall (n = 26)	Exercise intervention (n = 14)	Control group (n = 12)	p
Age, years (SD)	38.2 (11.6)	36.2 (7.8)	40.4 (15.2)	0.37
Sex, n (%)				0.49
Female	21 (81%)	12 (86%)	9 (82%)	–
Male	5 (19%)	2 (14%)	3 (27%)	–
Education, years (SD)	14.7 (2)	15.7 (2.2)	13.6 (1.6)	0.009
MS type, n (%)				0.29
Relapsing-remitting	22 (85%)	13 (93%)	9 (75%)	–
Secondary Progressive	1 (4%)	0 (0%)	1 (8%)	–
Primary Progressive	3 (12%)	1 (7%)	2 (17%)	–
Disease duration, years (SD)	9.0 (8.0)	7.4 (6.1)	10.8 (9.7)	0.29
DMT use, n (%)	21 (81%)	10 (71%)	11 (92%)	0.20
DMT no use, n (%)	5 (19%)	4 (29%)	1 (8%)	–
PDDS, median (IQR)	2.0 (3.0)	2.0 (2.0)	3.0 (2.0)	0.23
Normal, n (%)	7 (27%)	5 (36%)	2 (17%)	-
Mild, n (%)	4 (15%)	2 (14%)	2 (17%)	-
Moderate, n (%)	4 (15%)	2 (14%)	2 (17%)	-
Gait, n (%)	7 (27%)	3 (21%)	4 (33%)	-
Early Cane, n (%)	2 (8%)	2 (14%)	0 (0%)	-
Bilateral, n (%)	2 (8%)	0 (0%)	2 (17%)	-
MSWS-12 (0–100), (SD)	33.6 (30.0)	29.5 (30.7)	38.4 (29.6)	0.46

Note: P-values were based on independent samples t-test or chi-square difference tests. MS, multiple sclerosis; DMT, disease-modifying therapies; PDDS, Patient-Determined Disease Steps; MSWS, Multiple Sclerosis Walking Scale; IQR, interquartile range; SD, standard deviation.

Adherence (sessions completed/48 prescribed) averaged 81% in the intervention group and 89% in the control group (Table 4). Mean session-level compliance (≥75% prescription met) was 27% ± 30% in the intervention group and 23% ± 36% in the control group.

Table 4.

Adherence and compliance by condition and training type.

Training type	Metric	Threshold	Meeting threshold (%)	Mean compliance (%) ± SD	Mean # sessions(± SD)	Range
Exercise Intervention (n = 14)
Overall	Adherence	—	81%	—	39 ± 14	0–48
Overall (Aerobic + Resistance, ≥75% both)	Mean Session-level Compliance	≥75% (both components)	—	27% ± 30%	—	—
Aerobic	Adherence	—	84%	—	40 ± 10	24–48
Aerobic (mins)	Compliance	≥75%	30%	—	14 ± 15	0–41
Aerobic (mins)	Compliance	≥90%	28%	—	13 ± 15	0–39
Aerobic (steps)	Compliance	≥75%	23%	—	11 ± 15	0–39
Aerobic (steps)	Compliance	≥90%	21%	—	10 ± 14	0–7
Resistance	Adherence	—	77%	—	37 ± 17	0–48
Resistance (#ex)	Compliance	≥75%	62%	—	30 ± 18	0–48
Resistance (#ex)	Compliance	≥90%	54%	—	26 ± 17	0–48
Control (n = 12)
Control	Adherence	—	89%	—	43 ± 9	22–48
Overall (Flexibility)	Mean Session-level Compliance	—	—	23% ± 36%	—	—
Flexibility (#cat)	Compliance	≥75%	23%	—	11 ± 17	0–47
Flexibility (#cat)	Compliance	≥90%	19%	—	10 ± 16	0–44

Note: #, number; mins, minutes; #ex, number of exercises; #cat, number of categories; SD, standard deviation. Adherence defined as completed sessions divided by the total number (48) prescribed. For the intervention group, session-level compliance required achieving ≥75% of both prescribed aerobic minutes/steps and prescribed resistance exercises within the same session. 90% threshold rows are presented descriptively to characterize higher-fidelity performance.

Screening phone calls averaged 15 ± 6.2 min, coaching sessions averaged 25 ± 10 min, and total coaching time was 5174 min. The pre-intervention Zoom session was longest (mean, 40 ± 11 min), and the self-monitoring session was the shortest (mean, 19 ± 9 min). All participants rescheduled at least once; three required weekly reminders.

Direct costs totaled ∼$11,406, including incentives ($3183), materials/shipping (∼$894), equipment ($2131), printing ($3977), and recruitment ($1221). Personnel costs comprised the largest overall expenditure.

IRB approval required ∼4 months, with eight amendments during the trial. Staff time averaged ∼30 min per assessment and 10 min for interviews; an additional three months were required for data management. Missing data were due to attrition or skipped items and were supplemented using coach notes and self-reports.

Two adverse events (unrelated to study conditions) were reported and resolved, with an average response time of three days. No other significant events or burdens were reported.

Exploratory efficacy outcomes

Post-intervention efficacy outcomes analyzed using ANCOVA (adjusted for baseline values, age, sex, and education) are presented in Table 5. Adjusted between-group differences were not statistically significance for most outcomes. The largest effect was observed for walking mobility (MSWS-12), with lower adjusted scores in the intervention group (adjusted mean difference = –15.4 points, 95% CI: −39.1 to 1.3; p = .06; partial η² = .149). All other comparisons were non-significant (p > .05) with small-to-moderate effect sizes. Within-group pre–post changes are presented in Supplemental Table S1; GLTEQ total scores increased in the intervention group (Δ=10.9, p = .029), whereas all other within-group comparisons were non-significant (p > .05).

Table 5.

Adjusted between-group comparisons of post-intervention outcomes.

Outcome	Adjusted mean (Intervention)	Adjusted mean (Control)	Adjusted mean difference (Int - Con)	95% CI	p-value	Partial η²
MSWS-12 (pts)	19.9 (5.4)	35.3 (5.8)	–15.4	–39.1 to 1.3	0.06	0.149
30STS (reps)	10.9 (0.9)	11.2 (1.0)	–0.3	–3.3 to 2.7	0.84	0.002
SDMT (pts)	43.4 (3.3)	41.9 (3.6)	1.4	–9.6 to 12.5	0.79	0.004
CVLT-II (pts)	44.2 (2.1)	48.4 (2.3)	–4.2	–11.2 to 3.0	0.24	0.069
GLTEQ Total (pts)	27.4 (4.1)	23.5 (4.5)	3.9	–10.1 to 17.9	0.57	0.017
GLTEQ_HCS (pts)	19.3 (3.8)	10.7 (2.1)	8.6	–4.1 to 21.4	0.17	0.091
HADS Depression (pts)	6.1 (1.0)	5.4 (1.0)	0.7	–2.6 to 4.0	0.65	0.011
HADS Anxiety (pts)	6.2 (1.0)	6.5 (1.1)	–0.3	–3.8 to 3.2	0.86	0.002
FSS (pts)	4.4 (0.3)	4.6 (0.3)	–0.2	–1.3 to 0.8	0.28	0.057
SF-12 PCS (pts)	46.9 (1.9)	43.4 (2.1)	3.5	–3.2 to 10.2	0.28	0.060
SF-12 MCS (pts)	46.6 (2.6)	49.4 (2.8)	–2.8	–11.5 to 6.0	0.51	0.022

Note: Data are presented as means ± standard error (SE). Post-intervention outcomes were analyzed using analysis of covariance (ANCOVA), with group (intervention vs. control) as the fixed factor and adjustment for baseline outcome values, age, sex, and education. Values represent adjusted means evaluated at the sample mean of covariates. Two-tailed tests were used (α = .05). Partial η² represents effect size estimates. Analyses are considered exploratory given the feasibility design and small sample size. CI, confidence interval; MSWS-12, Multiple Sclerosis Walking Scale; 30STS, 30-s sit-to-stand; SDMT, Symbol Digit Modalities Test; CVLT-II, California Verbal Learning Test-II; GLTEQ Total, Godin Leisure-Time Exercise Questionnaire total score; GLTEQ_HCS, Godin Leisure-Time Exercise Questionnaire health-contribution score; HADS, Hospital Anxiety and Depression Scale; FSS, Fatigue Severity Scale; SF-12 PCS, 12-Item Short Form Survey Physical Component Score; SF-12 MCS, 12-Item Short Form Survey Mental Component Score; pts, points; reps, repetitions.

Exploratory participant feedback

Interviews averaged 11 ± 4 min. Participants described the program as structured and supportive of consistency amid work and family demands. Reported benefits included increased strength (46%), flexibility (38%), energy (31%), balance (21%), and reduced pain (23%). Commonly valued components included resistance exercises, walking, and stretching. Challenges included time constraints, fatigue, illness, and competing demands (31%), as well as specific exercises perceived as difficult (19%) or insufficiently challenging (15%). Suggestions included digital logs and maintaining the remote format for convenience. Some participants (12%) expressed interest in hybrid or group formats. No cultural barriers were reported.

Exploratory associations between SDOH and study outcomes

The sample varied in education, employment, income, and household size, with 25/26 participants reporting health insurance and a usual source of care (Table 6). Exploratory associations are presented in Table 7. In the intervention group, higher education was associated with greater compliance (ρ = 0.701, p = 0.005), improved verbal learning (ρ = 0.608, p = 0.021), and lower anxiety (ρ = –0.691, p = 0.006). In the control group, employment status was associated with lower adherence (ρ = –0.583, p = 0.047), and household size with improved verbal learning (ρ = 0.597, p = 0.040) and SF-12 MCS (ρ = 0.676, p = 0.016); and higher family income with improved SF-12 PCS (ρ = 0.900, p = 0.037).

Table 6.

Individual and structural social determinants of health.

Domain	Category	n (%, unless indicated otherwise)
Individual SDOH
Birthplace	U.S.-born	21 (80.8)
	Born outside U.S.	5 (19.2)
Sex at birth	Female	21 (80.8)
	Male	5 (19.2)
Gender identity	Woman	21 (80.8)
	Man	5 (19.2)
Sexual orientation	Straight	22 (84.6)
	Bisexual	3 (11.5)
	Other	1 (3.8)
Hispanic/Latino origin	Mexican/Mexican American/Chicano	15 (57.7)
	Puerto Rican	6 (23.1)
	Cuban	1 (3.8)
	Other Hispanic/Latino	4 (15.4)
Race (multiple select)	White	12 (60.0)*
	Black/African American	2 (10.0)*
	American Indian/Alaska Native	4 (20.0)*
	Some other race	9 (45.0)*
Education	High school or less	7 (26.9)
	Some college/Associate degree	12 (46.2)
	Bachelor's or higher	7 (26.9)
Employment status	Working now	14 (50.0)
	Disabled	7 (25.0)
	Other	7 (25.0)
Household size	Median (IQR)	3 (2–4)
Family income (past year)	Median (IQR)	$51,000($25,000-$90,000)
Structural SDOH
Health insurance (any)	Insured	25 (96.2)
	Uninsured	1 (3.8)
Insurance source	Employer	14 (50.0)
	Medicaid	13 (46.4)
	Medicare	3 (11.1)
	Marketplace/private purchase	6 (21.4)
	Other	7 (25.0)
Usual source of care	Regular provider/clinic	26 (92.9)
	No place/Other	2 (7.1)
Last doctor visit	<1 year	26 (92.9)
	≥1 year	2 (7.1)
Housing type	House	12 (48.0)
	Apartment/flat	13 (52.0)
Neighborhood safety	Unsafe at night (agree/strongly agree)	12 (44.4)
	Safe/neutral	15 (55.6)
Transportation	Household vehicles, median (range)	2 (0–5)
Food insecurity (past 12 months)	Often/Sometimes true	7 (28.0)
	Never true	17 (68.0)

Note: *Race allowed multiple selections; percentages are within 20 respondents. SDOH, Social Determinants of Health; U.S., United States; IQR, Interquartile range.

Table 7.

Exploratory spearman correlations between social determinants of health and feasibility and outcome measures per condition.

SDOH variable	Feasibility outcomes		Efficacy outcomes
SDOH variable	Adherence (%)	Compliance ≥75%	ΔCVLT	ΔSF-12 MCS	ΔSF-12 PCS	ΔHADS-Dep	ΔHADS-Anx	ΔFSS	ΔSDMT	ΔSTS
Intervention
Education	–0.011	0.701**	0.608*	0.276	0.307	–0.416	–0.691**	0.136	0.073	0.051
Employment	0.082	0.379	0.408	–0.462	–0.018	0.392	0.315	0.056	0.168	–0.205
Household	0.048	0.193	0.211	–0.560*	0.114	0.223	0.270	–0.185	0.016	0.397
Income	0.257	0.102	–0.450	–0.117	–0.317	–0.264	–0.101	0.594	0.067	0.008
Control
Education	–0.462	0.257	–0.367	–0.065	–0.305	0.226	–0.389	0.246	-0.222	–0.527
Employment	–0.583*	0.576*	–0.098	–0.073	–0.416	0.521	0.400	0.417	-0.147	0.353
Household	–0.158	–0.341	0.597*	0.676*	0.039	–0.044	–0.064	–0.448	0.350	–0.057
Income	0.447	–0.671	0.872	0.100	0.900*	–0.600	–0.308	–0.600	0.500	–0.051

Note. Values shown are Spearman's ρ. Analyses are exploratory and hypothesis-generating. * Indicates p < .05; ** indicates p < .01. Correlations meeting moderate magnitude thresholds (|ρ| ≥ 0.30) and p < .05 are emphasized. SDOH, social determinants of health; Δ, change score; CVLT, California Verbal Learning Test; SF-12 PCS, 12-Item Short Form Survey Physical Component Score; SF-12 MCS, 12-Item Short Form Survey Mental Component Score; HADS, Hospital Anxiety and Depression Scale; Dep, depression; Anx, anxiety; FSS, Fatigue Severity Scale; SDMT, Symbol Digit Modalities Test; STS, 30-s sit-to-stand.

Discussion

The FERLA MS study provides preliminary evidence that a remotely-delivered, home-based exercise intervention is feasible and safe for English-speaking Hispanics/Latinos with MS. These results align with prior work demonstrating the success of tele-supported or hybrid behavioral programs in MS, including those with underrepresented populations.^16,17 These results support the potential for remote exercise programs to reduce barriers related to geography, transportation, and healthcare access, and demonstrate that the GEMS framework can be adapted for this population.

The trial demonstrated high feasibility and retention. The 87% retention rate exceeded the 70–85% typically reported in MS rehabilitation trials,^13,18,19 suggesting that remote delivery and behavioral coaching may reduce burden and enhanced accountability. Adherence (81%) modestly exceeded the 50–79% in prior MS exercise trials.^20–22

However, when applying a stringent compliance definition, mean session-level compliance was lower (27% intervention; 23% control). This pattern suggests that although attendance was high, achieving full-session dose fidelity was more challenging. Importantly, compliance was similar across groups, indicating this reflects challenges inherent to remote delivery rather than intervention burden. This strict definition distinguishes attendance from dose fidelity and likely contributed to the lower compliance rate. Such findings align with evidence of fluctuating exercise fidelity in remote interventions.^23,24 Compliance data are seldom reported in MS exercise trials and definitions vary, limiting direct comparisons.²⁵

Trial costs totaled ∼$11,406, driven primarily by participant materials ($3977), incentives ($3183), equipment ($2131), and shipping ($894). These align with our prior African-American MS feasibility trial ($3726; ∼$47 per participant).¹⁴ They further highlight remote program financial feasibility compared to supervised, facility-based interventions that incur substantially higher per-participant costs (e.g., €4619 vs €1820 for home-based).²⁶ This cost profile supports broader implementation of low-cost, home-based programs such as GEMS in MS communities beyond the Hispanics/Latinos.

The intervention proved safe and well-tolerated. Only two non-serious adverse events were reported, and both were unrelated to the intervention or control. This provides further evidence for the safety of home-based, remotely-delivered exercise interventions for diverse MS populations and aligns with prior trials demonstrating low risks in comparable settings.²⁷

Although adjusted between-group differences did not reach statistical significance, several outcomes demonstrated directional improvements favoring the exercise intervention, particularly for walking mobility and physical activity behavior. In exploratory between-group ANCOVA models adjusting for baseline values, age, sex, and education, the largest effect was observed for walking mobility (MSWS-12). The exercise group demonstrated lower adjusted post-intervention MSWS scores compared to controls (adjusted mean difference = −19.2 points), with a moderate effect size (partial η² = .149), although this did not reach conventional statistical significance (p = .06). The 10.1-point within-group improvement in MSWS-12 exceeded the established 4–6 point MCID for persons with MS.²⁸ Similarly, GLTEQ health contribution scores increased from 10.6 to 18.9 in the intervention group, surpassing the 14-point threshold for health-promoting activity,^6,29 whereas controls remained below this benchmark. However, adjusted between-group differences for physical activity and other secondary outcomes did not reach statistical significance. Given the feasibility design and small sample size, these findings should be interpreted as preliminary data rather than definitive evidence of intervention efficacy.

Beyond feasibility and efficacy metrics, this trial incorporated SDOH to explore their influence on engagement and outcomes among Hispanics/Latinos with MS. A novel aspect of FERLA MS was the integration of both individual- and structural-level SDOH into analyses. Structural SDOH (e.g., insurance, healthcare access) demonstrated little variability, limiting associations with outcomes, and likely reflecting the inclusion of English-speaking participants who may have greater healthcare access and resources than non-English-speaking counterparts. In contrast, individual-level factors revealed meaningful patterns: in the intervention group, higher education correlated with better compliance and lower anxiety; in the control group, employment was associated with lower adherence but higher compliance, and higher family income correlated with greater physical HRQOL. These results highlight the complex role of SDOH on exercise participation and suggest it may serve as both facilitator and stressor in behavioral interventions for diverse MS populations.

This pilot trial has several strengths and limitations. It is the first exercise training study targeting Hispanics/Latino adults with MS, an underrepresented group in rehabilitation and behavioral intervention research. The intervention was safe and achieved high retention and adherence. Importantly, SDOH integration into the analytic framework³⁰ added an equity lens often absent in MS exercise trials.^{4,16,17,31,32} However, the intervention was adapted directly from the GEMS program¹¹ with aerobic/resistance retention, but excluded cultural preferences (e.g., dance, bilingual materials). English-only materials preserved GEMS fidelity, but constrained tailoring and accessibility. Structural SDOH demonstrated little variation, likely reflecting English-speaking participants’ greater access to care and resources. Eligibility criteria requiring English proficiency and access to digital technology for remote participation may therefore limit generalizability to the broader Hispanic/Latino MS population, particularly individuals with lower socioeconomic resources or limited access to digital health tools. Associations between SDOH variables and study outcomes were examined using exploratory bivariate correlations and should therefore be interpreted cautiously.

The small sample size limited the ability to conduct comprehensive multivariable modeling without risk of overfitting. Although sensitivity analyses adjusting for education did not change the observed associations, residual confounding cannot be excluded. These findings are hypothesis-generating and intended to inform future adequately powered studies examining socioeconomic influences on intervention engagement and outcomes. Analyses were conducted among participants with complete data, and missing data were not imputed. Given the small sample size and feasibility nature of the trial, we did not apply formal missing data estimation approaches. However, this may introduce bias and limit the generalizability of the findings, as results may reflect a subset of participants who completed the intervention and assessments. Adherence tracking via paper logbooks and coach reports may have underestimated activity levels. The short follow-up also precluded assessing sustained effects. Efficacy and SDOH findings should be interpreted as exploratory. The study was not powered to detect intervention effects and was further limited by early termination and reliance on completers-only analyses. Accordingly, the results should be viewed as preliminary data that inform the feasibility and design of future trials rather than definitive evidence of intervention effectiveness. These limitations are typical of early-phase feasibility trials but highlight directions for future research including inclusion of Spanish-speaking participants, larger trials, and digital adherence tools such as mobile apps or wearables.

We provide evidence from the FERLA MS study that a remotely-delivered, home-based exercise intervention adapted from the GEMS program is feasible, safe, and provides potential benefits for English-speaking Hispanic/Latino adults with MS. The study achieved high retention and adherence, with preliminary evidence of improvements in physical activity and mobility. These results provide a foundation for future, adequately-powered multi-site trials to confirm efficacy and sustainability, and highlight the potential for culturally-tailored, technology-supported exercise programs to broaden access to evidence-based rehabilitation for diverse MS populations.

Supplemental Material

sj-pdf-1-mso-10.1177_20552173261448001 - Supplemental material for Feasibility and preliminary outcomes of a remotely-delivered exercise training intervention for English-speaking Hispanics/Latinos with multiple sclerosis: Results from the FERLA MS study

Supplemental material, sj-pdf-1-mso-10.1177_20552173261448001 for Feasibility and preliminary outcomes of a remotely-delivered exercise training intervention for English-speaking Hispanics/Latinos with multiple sclerosis: Results from the FERLA MS study by Victoria A Flores, Stephanie L Silveira, David X Marquez, Dominique Kinnett-Hopkins, Augusto Miravalle, Fabian Sierra-Morales and Robert W Motl in Multiple Sclerosis Journal – Experimental, Translational and Clinical

Supplemental Material

sj-docx-2-mso-10.1177_20552173261448001 - Supplemental material for Feasibility and preliminary outcomes of a remotely-delivered exercise training intervention for English-speaking Hispanics/Latinos with multiple sclerosis: Results from the FERLA MS study

Supplemental material, sj-docx-2-mso-10.1177_20552173261448001 for Feasibility and preliminary outcomes of a remotely-delivered exercise training intervention for English-speaking Hispanics/Latinos with multiple sclerosis: Results from the FERLA MS study by Victoria A Flores, Stephanie L Silveira, David X Marquez, Dominique Kinnett-Hopkins, Augusto Miravalle, Fabian Sierra-Morales and Robert W Motl in Multiple Sclerosis Journal – Experimental, Translational and Clinical

Footnotes

Acknowledgements

The authors thank all participants for their time and commitment to the FERLA MS study and acknowledge the support of the Exercise Neuroscience Research Laboratory, as well as Ariel Kidwell and Lexi Huynh, for their assistance with data collection and recruitment.

Consent to participate

All participants provided informed consent electronically through REDCap prior to enrollment in accordance with University of Illinois Chicago IRB procedures.

Consent for publication

Not applicable. This manuscript does not include any individual person's data, images, or videos requiring consent for publication.

Data availability

Data supporting the findings of this study are available from the corresponding author upon reasonable request.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical considerations

This study was approved by the University of Illinois Chicago Institutional Review Board (Protocol #2023-0665). All study procedures complied with institutional and ethical standards for research involving human participants.

Funding

This research was supported by a pilot award from the Chicago Chronic Condition Equity Network (C3EN), funded by the National Institutes of Health. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. Chicago Chronic Conditions Engagement Network, (grant number 5P50MD017349).

ORCID iDs

Victoria A Flores

Dominique Kinnett-Hopkins

Fabian Sierra-Morales

Robert W Motl

Supplemental material

Supplemental material for this article is available online.

References

Wallin

Culpepper

Campbell

, et al. The prevalence of MS in the United States: a population-based estimate using health claims data. Neurology 2019; 92: e1029–e1040. PubMed PMID: 30770430; PubMed Central PMCID: PMC6442006.

Hittle

Culpepper

Langer-Gould

, et al. Population-Based estimates for the prevalence of multiple sclerosis in the United States by race, ethnicity, age, sex, and geographic region. JAMA Neurol 2023; 80: 693–701. PubMed PMID: 37184850; PubMed Central PMCID: PMC10186207.

Kister

Bacon

Cutter

. How multiple sclerosis symptoms vary by age, sex, and race/ethnicity. Neurol Clin Pract 2021; 11: 335–341. PubMed PMID: 34476125; PubMed Central PMCID: PMC8382423.

Hersh

Morrow

Williams

, et al. Diversity, equity, and inclusion in the multiple sclerosis community: a call to action. Int J MS Care 2023; 25: 199–205. PubMed PMID: 37720263; PubMed Central PMCID: PMC10503818.

Motl

Hubbard

Bollaert

, et al. Randomized controlled trial of an e-learning designed behavioral intervention for increasing physical activity behavior in multiple sclerosis. Mult Scler J - Exp Transl Clin 2017; 3: 2055217317734886. PubMed PMID: 29051831; PubMed Central PMCID: PMC5637983.

Motl

Pilutti

. The benefits of exercise training in multiple sclerosis. Nat Rev Neurol 2012; 8: 487–497. PubMed PMID: 22825702.

Silveira

Motl

Marquez

, et al. Rates and patterns of physical activity among hispanics with multiple sclerosis in the NARCOMS registry. Mult Scler Relat Disord 2022; 65: 104018. PubMed PMID: 35810721.

Silveira

Motl

Salter

, et al. Social cognitive theory variables as correlates of physical activity behavior among hispanics with multiple sclerosis. Arch Phys Med Rehabil 2024; 105: 1083–1088. PubMed PMID: 38281577.

Silveira

Motl

Ontiveros

, et al. A Qualitative Inquiry of Exercise Preferences among Hispanic Adults with Multiple Sclerosis. Am J Lifestyle Med 2025; 15598276251334360. doi:10.1177/15598276251334360 PubMed PMID: 40255267; PubMed Central PMCID: PMC12003335.

10.

Flores

Silveira

Marquez

, et al. Rationale for a 4-month, parallel-group, randomized controlled trial to assess the feasibility and efficacy of a remotely delivered exercise training intervention for hispanics/latinos with multiple sclerosis (FERLA MS). Pilot Feasibility Stud 2025; 11: 62. PubMed PMID: 40340698; PubMed Central PMCID: PMC12060554.

11.

Adamson

Learmonth

Kinnett-Hopkins

, et al. Feasibility study design and methods for project GEMS: guidelines for exercise in multiple sclerosis. Contemp Clin Trials 2016 Mar; 47: 32–39. PubMed PMID: 26655434.

12.

Latimer-Cheung

Martin Ginis

Hicks

, et al. Development of evidence-informed physical activity guidelines for adults with multiple sclerosis. Arch Phys Med Rehabil 2013; 94: 1829–1836.e7. PubMed PMID: 23770262.

13.

Learmonth

Paul

Miller

, et al. The effects of a 12-week leisure centre-based, group exercise intervention for people moderately affected with multiple sclerosis: a randomized controlled pilot study. Clin Rehabil 2012; 26: 579–593. PubMed PMID: 21984532.

14.

Kinnett-Hopkins

Motl

. Results of a feasibility study of a patient informed, racially tailored home-based exercise program for black persons with multiple sclerosis. Contemp Clin Trials 2018; 75: –8. PubMed PMID: 30342254.

15.

Shephard

. PAR-Q, Canadian home fitness test and exercise screening alternatives. Sports Med 1988; 5: 185–195.

16.

Fakolade

Akbar

Mehelay

, et al. Mapping two decades of multiple sclerosis rehabilitation trials: a systematic scoping review and call to action to advance the study of race and ethnicity in rehabilitation research. Mult Scler Relat Disord 2023; 72: 104606. PubMed PMID: 36917889.

17.

Marrie

Chataway

Bierer

, et al. Enhancing diversity of clinical trial populations in multiple sclerosis. Mult Scler Houndmills Basingstoke Engl 2023; 29: 1174–1185. PubMed PMID: 37555490; PubMed Central PMCID: PMC10413791.

18.

Huynh

TLT

Barstow

Motl

. Feasibility of a theory-based physical activity intervention for persons newly diagnosed with multiple sclerosis. Mult Scler Relat Disord 2024; 83: 105446. PubMed PMID: 38262331.

19.

Sandroff

Wender

CLA

Weber

, et al. Feasibility of remotely delivered and supported aerobic walking exercise training for cognitive processing speed impairment in fully-ambulatory persons with multiple sclerosis. Mult Scler Relat Disord 2023; 74: 104709.

20.

Almarwani

Alosaimi

. Exercise self-efficacy and fatigue as predictors of adherence to home-based exercise among patients with multiple sclerosis. Patient Prefer Adherence 2023; 17: 1441–1449. PubMed PMID: 37342492; PubMed Central PMCID: PMC10278867.

21.

Donkers

Nickel

Paul

, et al. Adherence to physiotherapy-guided web-based exercise for persons with moderate-to-severe multiple sclerosis: a randomized controlled pilot study. Int J MS Care 2020; 22: 208–214. PubMed PMID: 33177956; PubMed Central PMCID: PMC7643843.

22.

Garg

Rutherfurd

Labrum

, et al. Outcomes, and perceptions of a virtual group exercise program in multiple sclerosis. J Neurol Phys Ther JNPT 2024; 48: 54–63. PubMed PMID: 37563754.

23.

Adnan

Roikjaer

Samadzadeh

, et al. At home e-based physical exercise programs in patients with multiple sclerosis: a scoping review. Front Neurol 2024; 15: 1449519. PubMed PMID: 39474367; PubMed Central PMCID: PMC11518785.

24.

Knox

Nickel

Donkers

, et al. Physiotherapist and participant perspectives from a randomized-controlled trial of physiotherapist-supported online vs. Paper-based exercise programs for people with moderate to severe multiple sclerosis. Disabil Rehabil 2023; 45: 1147–1153.

25.

Motl

Pilutti

. Advancements and challenges in exercise training for multiple sclerosis: comprehensive review and future directions for randomized controlled trials. Neurol Ther 2024; 13: 1559–1569.

26.

Tosh

Dixon

Carter

, et al. Cost effectiveness of a pragmatic exercise intervention (EXIMS) for people with multiple sclerosis: economic evaluation of a randomised controlled trial. Mult Scler Houndmills Basingstoke Engl 2014; 20: 1123–1130. PubMed PMID: 24421304.

27.

Learmonth

Herring

Russell

, et al. Safety of exercise training in multiple sclerosis: an updated systematic review and meta-analysis. Mult Scler Houndmills Basingstoke Engl 2023; 29: 1604–1631. PubMed PMID: 37880997; PubMed Central PMCID: PMC10637110.

28.

Hobart

. Prolonged-release fampridine for multiple sclerosis: was the effect on walking ability clinically significant. Mult Scler 2010; 16: P509.

29.

Godin

Shephard

. A simple method to assess exercise behavior in the community. Can J Appl Sport Sci J Can Sci Appl Au Sport 1985; 10: 141–146. PubMed PMID: 4053261.

30.

Cox

Hwang

Haines

, et al. Using the PhenX toolkit to select standard measurement protocols for your research study. Curr Protoc 2021; 1: e149. PubMed PMID: 34038028; PubMed Central PMCID: PMC8251725.

31.

Lai

Cederberg

Vanderbom

, et al. Characteristics of adults with neurologic disability recruited for exercise trials: a secondary analysis. Adapt Phys Act Q APAQ 2018; 35: 476–497. PubMed PMID: 30382753.

32.

Williams

Orlando

Akisanya

, et al. Multiple sclerosis in black and hispanic populations: serving the underserved. Neurol Clin 2024; 42: 295–317. PubMed PMID: 37980120.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.68 MB

0.00 MB

0.02 MB