This study, titled the EpiFIT Study, was an 8-week, parallel-group, randomized controlled pilot feasibility trial that adhered to CONSORT (Consolidated Standards of Reporting Trials; Multimedia Appendix 1) [31] and the CONSORT pilot and feasibility (Multimedia Appendix 2) [32] guidelines. The overall study design consisted of a screening process, a baseline visit (week 1), an 8-week intervention, and a final visit (week 10). The pilot trial was completed as planned at the conclusion of the prespecified study period. In-person visits and exercise sessions were conducted at a private research facility at Iowa State University. As the primary outcomes were feasibility, acceptability, and appropriateness, the study did not meet the criteria for the ClinicalTrials.gov Applicable Clinical Trial (ACT) Checklist and was not registered.

Participants were recruited between April 2024 and December 2024 using a multifaceted approach, including tabling at community and campus events, distributing flyers around the community and the university, direct engagement at local sporting events, and via social media platforms. Additionally, mass emails were sent to alumni, faculty, staff, and students at multiple time points. Recruitment was further supported through partnerships with local hospitals and community-based epilepsy organizations, which facilitated referrals from patients and community networks.

Inclusion criteria were: adults (aged 18-65 years) with doctor-diagnosed epilepsy who had at least 1 seizure in the past year or were taking medication to control seizures, could provide physician’s consent to participate, were on stable seizure medication for ≥28 days prior to enrollment, were not currently meeting the recommended aerobic guidelines from the Physical Activity Guidelines for Americans [9], and were not participating in a structured exercise program. Exclusion criteria were based on recommendations made by the National Institute of Neurological Disorders and Stroke [33]. Participants were excluded from the study if they were taking >3 concomitant antiepileptic drugs, had status epilepticus within the past 2 years, had a neurostimulation device implanted or activated <1 year prior to enrollment (or a battery life unit that would not extend the duration of the trial), had epilepsy surgery <1 year prior to enrollment, had seizures that were triggered by exercise, drank alcohol excessively (defined as >7 per week for women and >14 per week for men [34]), smoked cigarettes, used cannabis, and were pregnant.

Process feasibility was assessed using study-level indicators of recruitment, retention, adherence, survey completion, protocol compliance, and safety. Recruitment was assessed through screening and recruitment rates. A single-item question on the screening survey asked participants how they heard about the study. The screening rate was calculated as the number of completed screening surveys divided by the total months of recruitment (8 months). Recruitment strategies are summarized as counts and percentages. The recruitment achievement rate was calculated by dividing the number of enrolled participants by the target sample size. Recruitment rate was calculated as the number of enrolled participants divided by the total months of recruitment.

Retention rate was defined as the percentage of enrolled participants who completed all study visits. Adherence was measured by recording attendance at study visits and exercise sessions. Adherence rate was defined as the percentage of completed study visits, including baseline and final assessments for all participants (exercise and control groups), as well as exercise sessions for the exercise group. Weekly survey completion was also monitored as a process feasibility indicator. All participants completed a weekly seizure diary, while exercise group participants additionally completed intervention-related check-ins assessing motivation, habit development, coping plans, and commitment to upcoming exercise sessions. Survey completion was calculated as the percentage of expected weekly surveys completed throughout the intervention.

Protocol compliance was assessed using 2 criteria, with priority given to ensuring safety by focusing primarily on participant completion of the full 50-minute exercise duration for each session. The secondary criterion was adhering to the prescribed RPE targets for each session, with participants encouraged to exercise at a pace they felt comfortable with if they could not reach the target RPE. Safety was monitored throughout the intervention period by recording adverse events, including seizure-related events (eg, seizure occurrence), as well as any participant-reported symptoms, complaints, or medical concerns that occurred during study participation. Seizure occurrence was also monitored using participants’ seizure diaries, which were reviewed weekly throughout the intervention period.

Self-reported intervention feasibility, acceptability, and appropriateness were assessed using 3 implementation outcome measures: the Feasibility of Intervention Measure (FIM, α=0.89), the Acceptability of Intervention Measure (AIM, α=0.85), and the Intervention Appropriateness Measure (IAM, α=0.91). These measures can be administered together or individually to evaluate participants’ perceptions of the intervention’s feasibility, acceptability, and appropriateness [50]. Each measure consists of 4 items rated on a 5-point Likert scale, where participants indicated their level of agreement with each statement (1=Completely Disagree to 5=Completely Agree). A total score for each domain was calculated by averaging the item responses, with a score of >3 indicative of intervention feasibility, acceptability, and appropriateness. Both groups completed all implementation outcome measures during their final visit.

Sample size was determined based on available resources, the study timeline, and published recommendations for pilot studies [30,54-56]. A target enrollment of 36 participants (18 per arm) was selected to estimate feasibility parameters.

All statistical analyses were conducted using R (version 4.1.1 or higher; R Core Team) [57]. Descriptive statistics summarized participant characteristics and study outcomes. Categorical variables were summarized using counts and percentages. Continuous variables were summarized using medians and ranges (minimum-maximum values). Self-reported feasibility, acceptability, and appropriateness measures (FIM, AIM, and IAM) were scored according to published procedures by averaging responses across the 4 items within each measure and were presented as mean (SD) scores to align with recommended scoring procedures. There was no missing data.

All study procedures were approved by Iowa State University’s Institutional Review Board (Approval number #23-375-00) and conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants prior to enrollment. To ensure privacy and confidentiality, all study data were coded with unique identifiers, and identifying information was stored separately from research data on a password-protected computer. Participants did not receive compensation for their participation.

Participant characteristics are presented in Table 1. The median duration of epilepsy diagnosis among participants was 10 years. All participants were taking antiepileptic drugs, with all 4 participants in the exercise group reporting polytherapy (ie, more than one antiepileptic drug) and both participants in the control group reporting monotherapy (ie, one antiepileptic drug). Five of 6 participants reported at least one seizure in the past year. Two participants had a typical seizure pattern of monthly seizures, 1 participant had a typical seizure pattern of yearly seizures, and 2 participants were unsure. Most participants had at least 2 comorbid conditions. None of the participants reported using a dietary approach to manage their seizures.

Figure 1 displays the CONSORT diagram illustrating recruitment efforts and participant flow. The recruitment strategies used yielded 25 completed screening surveys in 8 months, with a screening rate of approximately 3 interested potential participants per month. Most potential participants were recruited via mass emails (20/25, 80%). Other recruitment strategies included word of mouth (2/25, 8%), social media posts (1/25, 4%), flyers in the community (1/25, 4%), and physician referral (1/25, 4%). Of the 25 participants who completed the screening survey, 10 potential participants were screened as potentially eligible based on the screener, and of those, 6 were enrolled in the study, achieving 16.7% of the target sample size (36 participants). Reasons for not enrolling the other 4 potential participants included unable to commute (n=2), not interested in the study (n=1), and inability to be contacted (n=1). The recruitment rate was approximately 0.75 enrolled participants per month.

All participants completed the assessment visits with a 100% retention rate. Out of the 108 study visits (1 baseline visit [both groups], 1 final visit [both groups], and 24 exercise sessions [exercise group]), all 108 were attended, for a study visit adherence rate of 100%. Weekly survey (ie, seizure diary [both groups] and health education check-ins [exercise group]) completion rate was 100% for both groups. Among exercise group participants, the median commitment score for attending their next scheduled exercise session was 100 (range 74-100). Protocol compliance for criterion 1 was met, as 100% of participants completed 24/24 sessions for the full 50-minute duration. Average weekly RPE met the target goal for each week: week 1 (mean 11, SD 0.81) was within the 9-11 range, weeks 2-3 (mean 12.2, SD 1.37 and mean 12.5, SD 0.79) were within the 12-13 range, and weeks 4-8 (mean 14, SD 0.85; mean 14.3, SD 0.42; mean 14.5, SD 0.57; mean 14.8, SD 0.5; and mean 14.8, SD 0.5) were within the 14-15 range. Total RPE compliance for all sessions was 88.54% (85/96 compliant). Deviations from the target RPE were minimal, with 6 sessions 1 unit and 5 sessions 2 units above or below the target goal. One adverse event was recorded during the intervention period. A participant fainted after skipping breakfast, which was attributed to low blood sugar. Following monitoring and physician approval, the participant was able to continue study participation.

No seizures occurred during the supervised exercise sessions. Of the 4 participants, one in the exercise group reported one seizure outside of the exercise sessions and both participants in the control group reported seizures, with a total of 5 seizures reported during the intervention period. When asked about the cause of the seizure, the participant in the exercise group reported missing medication. Other reasons for seizures in the control group included being overtired or having irregular sleep, emotional stress, alcohol or drug use, and fever or overheating.

Participants’ mean scores (out of a total score of 5) for the self-reported feasibility, acceptability, and appropriateness measures were 4.46 (SD 0.60), 4.33 (SD 0.66), and 4.42 (SD 0.80), respectively (Table 2).

Table 3 presents descriptive changes in psychological and physical health outcomes from baseline to the final visit. In the exercise group, median depression and anxiety scores increased from baseline to final. Stigma, general self-efficacy, VO₂ max, weight, BMI, and body fat percentage showed little change. Instigation habit scores increased. Preparation habit scores remained stable. Among QOLIE-31 domains, median seizure worry and cognitive function scores increased. Median medication effects scores decreased. The remaining domains showed little change. Individual participant-level data for all outcomes are provided in Multimedia Appendix 3.

The primary aim of this Phase IIb study was to evaluate whether an 8-week supervised aerobic exercise intervention was feasible, acceptable, and appropriate for adults with epilepsy. Recruitment feasibility was poor; however, intervention implementation among enrolled participants was successful. All enrolled participants completed study visits and exercise sessions with no dropout, and survey completion and protocol compliance were high. No exercise-related seizures occurred during the supervised sessions. Participants also self-reported high perceptions of intervention feasibility, acceptability, and appropriateness. Overall, while recruitment strategies will require further refinement before progressing to a larger trial, the intervention protocol was well received by participants, supporting its continued development and evaluation in future studies.

Recruitment challenges are consistent with prior exercise interventions in adults with epilepsy, where enrollment has similarly fallen short of target samples [16] or exercise interventions have shown limited appeal in this population [58]. Possible contributors to recruitment challenges in this study include the lack of an on-site medical center at the university, which limited direct access to potential participants. Although efforts were made to collaborate with nearby hospitals, ongoing reminders were needed to sustain recruitment momentum. Additionally, the decision to make the intervention supervised was driven by a commitment to ensuring participant safety and proper execution of exercises. However, this decision excluded several potential participants who were interested in participating but were unable to commute, a commonly reported barrier to exercise intervention participation in this population [59]. Yet, a recent theory-based, remotely delivered, 12-week physical activity intervention for adults with epilepsy recruited through a Level 4 Epilepsy Center also fell short of its enrollment target (26/30) [18]. Together, these findings suggest that recruitment challenges in exercise research among adults with epilepsy may represent a broader field-level barrier rather than a problem specific to this study or intervention format. Addressing this barrier will likely require a more comprehensive approach to recruitment that combines stronger clinical partnerships, potentially across multiple sites, with accessible intervention delivery and a better understanding of how to engage adults with epilepsy in exercise research.

Although recruitment was poor, intervention implementation among participants was successful. Similar patterns have been reported across exercise interventions for adults with epilepsy, where recruitment rates were low (9% to 60%), but completion and adherence rates were high, ranging from 75% to 100% [29]. This pattern was also observed in the recent remote trial, which reported high retention, adherence, and acceptability despite falling short of its enrollment target [18]. In this study, participants demonstrated strong engagement throughout the intervention, with high retention, adherence, survey completion, and protocol compliance. Participants also self-reported that the intervention was feasible, acceptable, and appropriate, providing further support for the intervention among those enrolled. The use of BCTs, including goal setting, action planning, commitment contracts, feedback and monitoring, and habit formation strategies, may have supported participant engagement and adherence. Future efforts to move the field forward should focus on improving recruitment while preserving the intervention strategies that support engagement among those who enroll.

The exploratory outcomes should be interpreted cautiously given the small sample size, differences from findings reported in prior exercise intervention studies in adults with epilepsy [20,21,25,26], and contextual factors that may have influenced participant responses. For example, 3 of the 4 participants in the exercise group experienced challenging personal circumstances unrelated to the study during the week of their final assessment, which influenced their responses to the psychological and sleep measures. Additionally, baseline levels of depression and anxiety were higher in the control group than in the exercise group, providing greater opportunity for improvement among control participants. These factors, combined with the small sample size, make it difficult to draw meaningful conclusions about changes in exploratory outcomes.

This study had several limitations. The most significant was the small sample size and difficulties with recruiting participants. Recruitment was limited to a convenience sample of individuals in the local rural area, who were primarily college-aged and non-Hispanic White. Additionally, epilepsy diagnosis was based on physician diagnosis rather than confirmation by an epileptologist or neurologist. Therefore, it is possible that individuals with seizure-like activity or other conditions that mimic epilepsy may have been included. Additionally, epilepsy type was also determined through self-report, which may have resulted in misclassification of epilepsy characteristics. Future studies should incorporate clinical verification of epilepsy diagnosis and subtype when feasible.

The study design also included several methodological limitations. Due to the nature of the intervention, participant and study personnel blinding was not feasible, which may have influenced participant responses and outcome reporting. Additionally, randomization resulted in baseline differences between groups, particularly in the psychological outcomes, which limited meaningful comparison of the exploratory outcomes. Further, exercise intensity was primarily prescribed and monitored using the Borg RPE scale. Although RPE is an established and commonly used proxy indicator for prescribing aerobic exercise intensity [45], it is a subjective measure and may be influenced by individual perception or prior exercise experience [60]. Additionally, participants were permitted to use either personal wearable devices or a Polar H10 heart rate monitor to monitor heart rate during exercise sessions. Because these devices use different methods to estimate heart rate, measurement accuracy may have varied across participants.

Despite these limitations, the study also had several strengths that can inform future exercise interventions in adults with epilepsy. The intervention incorporated a standardized supervised exercise protocol alongside multiple evidence-based behavior change techniques designed to support adherence. Feasibility was evaluated across multiple domains, including process feasibility study-level indicators (ie, recruitment, retention, adherence, weekly survey completion, protocol compliance, and safety) and self-reported feasibility, acceptability, and appropriateness measures. Collectively, these findings provide practical guidance for refining future exercise interventions in adults with epilepsy.

The results of this Phase IIb pilot study suggest that a supervised aerobic exercise intervention for adults with epilepsy is feasible, acceptable, and appropriate to implement among participants who enroll. Using supervised exercise sessions, weekly preparatory habits and attendance-commitment checks, and flexibility within scheduling may be useful strategies to promote adherence among adults with epilepsy. Recruitment barriers remain the primary challenge to progressing this line of research and may represent a broader field-level barrier. Future trials should therefore focus on improving how adults with epilepsy are reached and engaged in exercise research, while preserving the intervention strategies that support strong engagement among those who enroll.