Objectives and hypotheses

Using a type 1 hybrid effectiveness-implementation study design [22], this three-arm parallel randomized controlled trial (RCT) will evaluate the effects of 1:1 and group-based supervised PA, compared to a 10-week waitlist control, among post-secondary students across three timepoints: baseline (T1), post-intervention (T2; 6-weeks), and follow-up (T3; 1-month post-intervention). Primary outcomes will include the immediate changes (T1–T2) in mental health indices (i.e., anxiety, depression, psychological distress, and well-being). Secondary outcomes will include changes from baseline to follow-up (T1–T3) and maintenance effects from post-intervention to follow-up (T2–T3) in mental health indices, as well changes in social well-being indices (i.e., social connectedness, social support), and PA behavior. The aims of the study include: (1) examining group differences between 1:1 PA delivery, group-based PA delivery, and the 10-week waitlist control arm on the primary and secondary outcomes; and (2) grounded in process evaluation recommendations [24], to explore implementation outcomes (i.e.,., reach, adherence, acceptability, fidelity) that may be linked to variation in primary and secondary outcomes while offering insight for wider dissemination. Based on evidence supporting various PA modalities for improving mental health [25], both 1:1 and group-based PA are hypothesized to be more effective than the waitlist control on primary and secondary outcomes, with no expected differences between the PA conditions on primary outcomes. Drawing on the social identity approach to health [20], it is hypothesized that group-based delivery will result in greater improvements in secondary outcomes including social well-being outcomes and maintenance effects compared to 1:1 delivery. However, 1:1 delivery is expected to yield more favourable implementation outcomes, due to its potential for stronger individualized support [14,26]. The results will provide novel insight into the effectiveness of different PA program delivery styles on student mental health, social well-being, and PA behaviour, while also offering implementation considerations to support the sustainability and scale-up of PA interventions across post-secondary campuses.

Trial design

A 3-arm parallel RCT assessing the intervention arms (1:1 and group-based PA delivery) compared to a 10-week control arm (waitlist control) will be conducted. A parallel arm design will be implemented, whereby students will be randomized to a study arm, and each study arm will be allocated a different intervention. The protocol adheres to CONSORT guidelines [27] and SPIRIT [28] recommendations for reporting of clinical trial protocols (see S1 Checklist). Ethical approval for the study was obtained November 24, 2023 (protocol # 45228) and the study was retrospectively registered on ClinicalTrials.gov (Study identifier: NCT06350877; Registered: April 2, 2024). The clinical trial was retrospectively registered due to the initial phase of data collection being focused on piloting and refining the data collection methods.

Study setting

The trial will leverage the infrastructure and support provided by Sport and Recreation Services, Health and Wellness Services, and the Mental Health and Physical Activity Research Centre at the University of Toronto’s St. George campus. A collaborative implementation approach involving the knowledge-user groups (i.e., researchers, professionals in on-campus sport and recreation and mental health, students) will be used to gather insights on embedding tailored and structured PA programs for student mental health within existing sport and recreation, and mental health services on post-secondary campuses.

Patients and public involvement

The study is informed by previous research on student perspectives of PA programming for mental health [29] and a mixed methods evaluation of the proposed PA intervention protocol, within a 1:1 delivery setting [14]. Student feedback will be collected throughout the study and used to inform insights for wider program dissemination and scale-up. Using a train-the-trainer model [30], the research team will collaborate with Sport and Recreation Services to provide standardized training in behavior change coaching and PA program delivery for mental health to certified student PA coaches. These trained and certified coaches will be responsible for delivering the PA intervention. Referral pathways from on-campus mental health and accessibility services will be examined. Consistent with type 1 hybrid effectiveness-implementation studies [22], process-related information on coach training and the referral pathways to the program will be collected to inform future testing and broader dissemination of educational tools and implementation guidelines. Input from on-campus mental health and sport and recreation professionals, as well as students, has informed the evaluation targets for the trial. Input emphasized the importance of examining the effectiveness of different PA delivery styles on broader well-being outcomes, such as social well-being. Knowledge-users (i.e., sport and recreation professionals, mental health professionals, students, researchers) will also be consulted during knowledge mobilization and dissemination to enhance the useability and uptake of the findings.

Eligibility criteria

Post-secondary students will be recruited based on the following eligibility criteria: (a) a post-secondary undergraduate or graduate student enrolled either part-time or full-time; (b) fluent in English (e.g., proficiency in reading and verbal expression – written and oral); (c) able to attend in-person PA sessions at the campus athletics and recreation centre; (d) moderately or insufficiently active (≤ 23 units of weekly leisure activity) based on interpretation scores from the Godin Leisure-Time Exercise Questionnaire [GLTEQ; 31]; and (e) experiencing self-reported ‘poor’, ‘fair’ or ‘good’ mental health in the past month. Exclusion criteria will include: (a) physically active (24 units or more of weekly leisure activity); (b) unsuccessful PA clearance using the PA readiness questionnaire [PAR-Q⁺; 32]; and (c) self-reported ‘very good’ or ‘excellent’ mental health.

Schedule of enrolment, interventions, and assessments

After screening for eligibility, students will be contacted to schedule an in-person intake session with a program coordinator for the trial to provide informed consent, complete the baseline assessment (T1), and conduct randomization. In the intervention arms and control arm, study outcomes will be assessed at baseline (T1), 6-weeks post baseline (T2), and at 1-month follow-up (T3). Aligning with SPIRIT [28] and CONSORT [27] guidelines intervention flow will be tracked (see Fig 1) and the schedule of enrolment, interventions, and assessments has been provided (see Fig 2). A 1-month follow-up was selected for the current study given that, to date, only one known study has examined the maintenance effects of PA interventions on student mental health outcomes [16]. The aim is to assess whether the 1-month effects observed in prior research can be replicated in the current context [33]. If such findings are observed, this follow-up period will provide an important foundation for future studies to evaluate the sustainability and longer-term impact of these interventions. To support student retention at T3, a modest financial incentive will be offered in the form of a $25 CAD gift card (student’s choice of Amazon, Indigo, or Starbucks). Students’ reasons for discontinuing the trial will be collected, no further data will be collected following withdrawal from the trial.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Participant flow.

https://doi.org/10.1371/journal.pone.0330851.g001

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Schedule of enrolment, interventions, and assessments according to the Standard Protocol Items: Recommendations for Intervention Trials.

https://doi.org/10.1371/journal.pone.0330851.g002

Allocation and blinding

Simple randomization will be used to allocate eligible students to the intervention arms using Research Randomizer [www.randomizer.org; 34,35]. Each student will be individually randomized, with an equal chance of being assigned to any of the three groups (1:1 PA, group-based PA, or waitlist control). While simple randomization may lead to unequal group sizes, block randomization was deemed unsuitable due to the risk of generating groups not comparable on covariates [34]. Stratified randomization was not possible given the continuous nature of student enrollment [34]. To account for the potential group size imbalance, primary statistical analyses will use linear mixed effects modeling, which are robust to unequal sample sizes and unbalanced data [36].

To minimize selection bias, allocation concealment will be used by having an investigator who will generate and assign random group allocations only after students have completed enrollment and the baseline (T1) assessment. Random allocations will be generated in real-time and not stored in advance. Students will be partially blinded in that students will be unaware of the purpose of group allocation or study hypotheses but will know they were assigned to an intervention arm or the control arm. To prevent care provider bias, PA coaches from Sport and Recreation Services will also not be informed about the purpose of group allocation or the study hypotheses. Investigators will not be blinded due to their roles in program coordination and statistical analyses. Recruiting an external individual to manage randomization, program coordination, and analyses is not feasible given the scope of the study and availability of resources. Nonetheless, several steps will be taken to minimize bias. Allocation concealment will be maintained until after baseline assessment, and analyses will follow the pre-specified plan outlined in this protocol paper. Outcome measures include open-ended and standardized self-report items, completed by students through a secure online data capture platform [REDCap; 37] without direct investigator involvement. The study protocol is also pre-registered to support transparency and reduce the risk of selective reporting.

Trial arms

Eligibility screening and randomization to trial arms will occur on a rolling basis, aligned with designated recruitment periods in the fall, spring, and summer semesters. All PA sessions will be delivered by trained Sport and Recreation PA coaches at the University of Toronto. Following group allocation to an intervention arm (1:1 or group-based PA), students will be matched with a coach and work with the same coach throughout the duration of the 6-week program. All PA sessions for the intervention arm are expected to commence within one week of randomization. Students assigned to 1:1 delivery will be matched immediately with a PA coach to coordinate weekly sessions. Group-based PA sessions will start once 3–8 students have been allocated and will be scheduled at a time that works for most students enrolled. Modifications to the allocated trial arm (e.g., switching from group to 1:1 PA delivery or from the control arm to an intervention arm) will not be permitted within the research protocol. However, students who are unable to follow through with their assigned trial arm will be offered PA sessions outside of the research (1:1 or group-based, depending on student preference) as part of standard service provision. Reasons for not initiating an assigned trial arm will be documented as part of the mixed methods process evaluation, as an indicator of adherence to the trial. To improve adherence to the intervention protocol by students, coaches will be in contact with students weekly to schedule sessions and will send reminder emails to students about upcoming sessions. To improve adherence to the intervention protocol by coaches, a standardized behavior change coaching booklet will be used to guide all PA intervention sessions. Regular bi-weekly check-ins with PA coaches and the research team will be conducted to support consistent delivery and to monitor implementation and student progression through the PA intervention. In addition, fidelity to core intervention processes will be examined based on student-reported experiences (see Implementation Process Evaluation Outcomes). Concomitant mental health care (e.g., therapy, medication use) and engagement in additional on-campus resources will be permitted during the trial and reported on to describe the student sample.

Intervention arms

The intervention arms are grounded in behavior change theory and empirical evidence on PA for mental health [14,25,38,39], with sessions designed to meet students’ PA related needs and preferences. To support student preferences, PA sessions will be delivered in-person in private PA spaces in the Mental Health and Physical Activity Research Center, located in the campus athletics and recreation centre. Public gym spaces will also be used depending on student preferences and interests. While bias related to coaching variability and differences in PA delivery space is plausible, the intervention was intentionally designed using pragmatic principles to reflect real-world conditions [40]. This includes allowing flexibility in implementation, such as permitting coaches with diverse backgrounds to deliver the sessions and enabling students to select preference-based PA delivery space. The purpose is to facilitate enjoyment and key COM-B behavioral processes (capability, opportunity, and motivation), to support ongoing engagement in PA as a mental health management strategy (see Table 1). Indeed, the COM-B model posits that no behavior will occur without sufficient capability, opportunity, and motivation [43]. The mixed methods process evaluation was designed to assess whether core intervention processes are received by students including acceptability (e.g., enjoyment), student-reported experiences of COM-B processes, and the quality of the coach-student relationship, a core component intended to foster a supportive and collaborative environment.

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Intervention core components using the COM-B model and measurement items for implementation fidelity.

https://doi.org/10.1371/journal.pone.0330851.t001

1:1 PA intervention delivery

The 1:1 PA intervention will be a 6-week supervised and individualized program. Participation will involve engaging in a weekly 1-h session. Each 1-h session will include: (1) 30-min of behavior change coaching; and (2) 30-min of individualized and supervised PA training. Intervention material includes a behavior change workbook for facilitating the 30-min of behavior change coaching, which has been informed by evidence-based strategies for promoting mental health through PA engagement [39]. Each week, students will complete the behavior change workbook in session with the program trainer to support the weekly learning objectives and learning experiences (see Table 2). Following completion of the 30-minute behavior change coaching, students will engage in structured and supervised PA. For a detailed description of the PA intervention protocol see [14].

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Weekly behavior change coaching learning objectives, learning experiences, and associated behavior change techniques.

https://doi.org/10.1371/journal.pone.0330851.t002

Group PA intervention delivery

The group PA intervention will receive the same protocol as described above though delivered in small groups of 3–8 students led by a trained Sport and Recreation PA coach. Drawing on group-dynamic principles [20,52], this number of students is ideal to help foster cohesiveness (e.g., through increased opportunity for interaction, discussion, and feedback/support from the PA coach) while reducing coordination and scheduling barriers. To promote needs satisfaction and preferences in group-based delivery, coaches will promote shared decision making on the types of activities engaged in within the group sessions.

Control arm

Students who are assigned to the 10-week waitlist control arm will be assessed on primary and secondary outcomes at baseline (T1), at 6-weeks (T2), and at 1-month follow-up (T3). At completion of the 1-month follow-up assessment, students in the waitlist control arm will be offered the PA intervention (either group or 1:1 delivery depending on student preference). Following completion of the control arm, behavior will not be monitored or evaluated.

Outcomes

Research outcomes to be assessed include demographic and descriptive data, primary outcomes, secondary outcomes, and implementation process evaluation outcomes.

Demographic and descriptive data

Self-reported data to describe the characteristics of the sample will be collected including age in years, international student identification (i.e., domestic or international student), level of education (i.e., undergraduate, graduate, college program, either full-time or part-time), gender identity (i.e., women, non-binary, two-spirit, man, prefer not to answer), sexual orientation (i.e., heterosexual/straight, gay/lesbian, bisexual, pansexual, asexual, queer, two-spirit, questioning/unsure, I prefer not to answer, other), and ethnicity/race (i.e., Indigenous peoples of Canada, Indigenous outside of Canada, Arab, Black, Chinese, Filipino, Japanese, Korean, Central or South American, South Asian, Southeast Asian, West Asian, White). Students will also be asked to self-report their mental health experiences including history of mental illness using organizational structure from the DSM-5 [53], past month emotional challenges (i.e., loneliness, difficultly coping with stress in a healthy way, difficulty handling emotions, anxiety, social isolation, depression, trouble concentrating, substance use issues, unhealthy social media use), and past year mental health service use (i.e., past-year therapy or counseling, past year medication, therapy or counseling and medication use, not applicable, and other).

Primary outcomes

Depression. The Patient Health Questionnaire [PHQ-9; 54] will be used. The 9-item PHQ measures the presence and severity of depressive symptoms over the past two weeks (e.g., “Feeling tired or having little energy”; “Little interest or pleasure in doing things”) ranging from 0 (not at all) to 3 (nearly every day) and aligns with the DSM-IV criteria to screen for and measure the severity of depression [54]. The total summed score ranging from 0−27 will be used in main analyses and depression symptom severity including minimal (score ranging from 5−9), minor (score ranging from 10−14), moderately severe (score ranging from 15−19) and severe (score > 20) will be calculated for descriptive purposes [54]. The PHQ-9 has high reliability and validity for screening and assessing the severity of depression and is widely used in both clinical practice and research including among post-secondary students [54–56].

Anxiety. The Generalized Anxiety Disorder Questionnaire [GAD-7; 57] will be used. The 7-item GAD assesses the frequency of symptoms associated with anxiety during the past two weeks (e.g., “Feeling nervous anxious or on edge”; “Trouble relaxing”) ranging from 0 (not at all) to 3 (nearly every day). The total summed score ranging from 0−21 will be used in main analyses and anxiety symptom severity including minimal (score ranging from 0−4), mild (score ranging from 5−9), moderate (score ranging from 10−14) and severe (score ranging from 15−21) will be calculated for descriptive purposes [57]. The GAD-7 is a valid and reliable measurement for assessing anxiety symptoms and the psychometric properties among young adults and post-secondary students have been demonstrated [57–59].

Psychological distress. The 10-item Kessler Psychological Distress Scale [K10; 60] will be used. Students will be asked to indicate how often over the last 30 days they experienced symptoms of psychological distress (e.g., “How often did you feel hopeless?”; “How often did you feel worthless?”) ranging from 1 (none of the time) to 5 (all of the time). The total summed score ranging from 10–50 will be used in analyses, with higher scores reflecting more psychological distress. Evidence of reliability and validity of the K10 has been reported among community and clinical samples [61,62] and is frequently used to monitor psychological distress among post-secondary students [63,64].

Well-being. The Mental Health Inventory-38 [MHI-38; 65] will be used to measure the 14-item psychological well-being subscale. Students will be asked to report how often during the past month they experienced symptoms of psychological well-being (e.g., “How much of the time, during the past month, did you feel relaxed and free from tension”; “During the past month, how much of the time have you generally enjoyed the things you do?”) on a six-point Likert scale ranging from 1 (all of the time) to 6 (none of the time). A total summed score will be used in main analyses ranging from 14–84, where higher scores represent more positive experiences of well-being. The MHI-38 has been widely used to measure well-being, and the psychometric properties have been demonstrated [66,67].

Secondary outcomes

Social support. The Social Provision Scale [SPS-5; 68] will be used to measure social support. The scale consists of 5-items (e.g., “I have relationships where my competence and skill are recognized”; “I feel part of a group who share my attitudes and beliefs”), on a 4-point Likert scale ranging from 1 (strongly disagree) to 4 (strongly agree). A total summed score ranging from 5–20 will be used in main analyses, where higher scores indicate more favorable perceptions of social support. The scales validation has been shown in the general population and the post-secondary student population [68].

Social connectedness. The 8-item Social Connectedness Scale will be used to measure social connectedness [69]. The items portray a general emotional distance between the self and others (e.g., “I feel disconnected from the world around me”; “I don’t feel I participate with anyone or any group”) and reflect behavior, feelings, or both associated with a lack of connectedness on a 6-point Likert scale ranging from 1 (agree) to 6 (disagree). Higher scores will reflect a more reported sense of social connectedness with a potential range of 8–48. The Social Connectedness Scale displays strong psychometric properties among post-secondary students [69].

PA behavior. Self-reported PA behavior will be measured using the GLTEQ [70]. Students will be asked to indicate how many times on average they engaged in mild (e.g., yoga, golf), moderate (e.g., fast walking, baseball), and vigorous (e.g., running, soccer) PA for more than 15 minutes in a typical week. A common modification of assessing the average duration (in hours and minutes) per session of each intensity category of PA will be included [71–73]. Students will also be asked to report the frequency and average duration of resistance exercise (e.g., free weight, bodyweight training). Total scores for mild, moderate, and vigorous activity and resistance exercise will be calculated by multiplying self-reported times per week by average duration of sessions for each type. The scale has been used among post-secondary students [73] and has been validated for use in adult samples [70,73].

Implementation process evaluation outcomes

Informed by previous research [14,29], the mixed methods implementation evaluation plan (see Table 3) will assess intervention fidelity, acceptability, reach, and adherence. Intervention fidelity will assess the extent to which students perceive their coach adhered to core intervention targets guided by the COM-B model, as well as the extent to which the collaborative coach–student relationship supported task, goal, and bond components of working therapeutic alliance [74]. Program acceptability will be assessed based on students’ enjoyment, and their perceptions of the program’s effectiveness as a mental health intervention. In addition, motives for engagement, perceptions of the intervention content (i.e., what worked well and opportunities for improvement), and delivery format preferences (i.e., no preference, 1:1, or group-based) will be assessed to help inform future program refinement to improve acceptability. Reach will assess referral pathways, as well as program inclusion and exclusion considerations. Adherence will be assessed based on the number of students who initiated the intervention, the number of sessions completed, and overall student completion rates.

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Mixed methods implementation evaluation.

https://doi.org/10.1371/journal.pone.0330851.t003

Sample size

A simulation-based power analysis using the simr package in R was conducted [75]. This analysis estimated the likelihood of detecting a group-by-time interaction in a linear mixed model using restricted maximum likelihood. The simulated design included three groups (1:1 PA intervention, group PA intervention, and waitlist control), with primary and secondary outcomes assessed at three time points: pre-intervention (T1), post-intervention (T2), and follow-up (T3). The model included fixed effects for group, time, and their interaction, as well as a random intercept for students to account for within-subject correlations. A moderate treatment effect (d = .50) was applied at the post-intervention time point, and a smaller sustained effect (d = .20) was applied at follow-up. Simulated data included realistic between- and within-person variability, and the outcome variable was centered to aid model convergence. To reflect anticipated study conditions aligned with typical attrition in PA and mental health intervention research [76,77], the analysis included missing data at random in approximately 20% of observations, with a slightly higher dropout rate (30%) at the follow-up time point. All available data from each student will be used, aligning with an intention-to-treat analysis, and no ad hoc imputation will be performed, as researchers have suggested that linear mixed models without such imputation are more powerful than ad hoc strategies [78].

Based on 1,000 simulations, a total sample size of 114 students (~38 per group) yielded an estimated power of 98.10% (95% CI: 97.05, 98.85) to detect the group-by-time interactions using a Type II F-test with Satterthwaite degrees of freedom at α = .05. This high level of power reflects the modeled effect structure and supports the adequacy of the proposed sample size.

Recruitment

Student recruitment and data collection began November 24, 2023, and is estimated to be completed by October 24, 2025, with results expected by January 24, 2026. Purposive and snowball sampling procedures will be used to recruit post-secondary students who are physically inactive and experiencing poor mental health. Post-secondary students will be recruited and referred to the intervention through the team’s research and professional networks (e.g., health and wellness and student support services; student life listservs; campus mental health listservs). Digital recruitment materials, including email templates and poster advertisements, will provide details about the purpose of the intervention, the procedures involved, eligibility criteria, and a link to the screening questionnaire. The purpose of this broad recruitment strategy is to enhance outreach and more effectively engage the diverse student population. Program reach will be examined to assess who is engaging with the intervention and how. These findings will offer valuable insights into potential disparities in uptake and help inform future strategies to improve accessibility, equity, and representation in program delivery. The screening questionnaire will be administered through REDCap [37] and will allow students to “sign up” up for the intervention by providing their email address and answering several questions to confirm eligibility. The program coordinator will contact eligible students to confirm their involvement in the study and to arrange an intake meeting.

Statistical analyses

Prior to conducting the main analyses, descriptive statistics will be calculated for baseline demographic characteristics (e.g., gender, age, sexual orientation, ethnicity/race) and main study variables at each assessment point (T1, T2, and T3) across the trial arms. To assess baseline equivalence between trial arms, differences in demographic characteristics and main study variables at baseline will be examined using chi-square tests and one-way analysis of variance. Differences in dropout will also be examined based on demographic characteristics and baseline study variables using chi-square tests and independent samples t-tests. The absence of systematic differences in dropout would support the plausibility of the Missing At Random (MAR) assumption, which underlies the use of linear mixed effects modeling in handling missing data [36,78,79].

For the main analyses, linear mixed effects modeling will be used to examine intervention effects on the primary and secondary outcomes over time and model assumptions will be tested [36,79]. Model specifications are provided above for a priori sample size calculation purposes. Models will be estimated using restricted maximum likelihood with the lme4 and lmerTest packages in R Version 2025.05.0 + 496. The intra-class correlation coefficient (ICC) will be reported to quantify the proportion of total variance in the outcome attributable to between-person differences, relative to within-person changes across timepoints. Significant interactions will be probed using within-group and between-group pairwise comparisons conducted with the emmeans package. P value and 95% CI adjustments using Tukey method for comparing a family of 3 estimates will be used to adjust for multiple comparisons.

For the mixed methods process evaluation, quantitative closed-ended questions will be analyzed using descriptive statistics (means and standard deviations or frequency counts). Independent-samples t-tests and chi-square tests will be used to examine group differences between 1:1 and group delivery formats on process-related outcomes. Open-ended qualitative responses will be analyzed using thematic analysis [80]. An iterative coding process will be used, combining both an inductive (data-driven) and deductive (theory-informed) approach [81]. Findings from both quantitative and qualitative analyses will be integrated during interpretation in the discussion to provide a comprehensive understanding of student experiences and implementation processes.

Ethical considersations

Any important protocol modifications (e.g., eligibility criteria, outcomes, analyses) to the protocol will be submitted to the REB for approval, updated in trial registries, and communicated to students (if applicable). All changes will be clearly documented and reported in trial publications. Student recruitment and data collection is ongoing (see student recruitment for study timeline) with plans for future implementation research and toolkit development, following completion of the current protocol. Written informed consent (See S1 Appendix) will be obtained from all students, and they will be provided with a thorough explanation of the study objectives, the voluntary nature of their participation, their right to withdraw, and the risks and benefits of the study. While there are minimal risks associated with participating in this research, there are potential emotional risks related to group vulnerabilities, particularly due to self-reported poor mental health, as well as the inherent physical risks associated with engaging in PA. To mitigate emotional risks, students will receive a resource sheet outlining accessible and free mental health services available on-campus and in the community after the intake meeting (see S2 Appendix). To mitigate physical risks, the PA sessions will be delivered by certified Sport and Recreation coaches who have received standard training in behavior change coaching and PA program delivery, and students will receive clearance for PA engagement using the PAR-Q⁺ [32]. In addition, bi-weekly meetings will be held with the research team and PA coaches throughout the intervention to ensure safety, address concerns, and support student well-being. All data will be collected, managed, and stored through the secure data capture platform REDCap [37]. Any identifiable information (i.e., email addresses, names) will be password protected and will only be used for program coordination purposes. For further information on data storage and confidentiality see S1 Appendix.

Adverse events will be identified through both spontaneous reports from students and formal solicited check-ins conducted by PA coaches during weekly sessions. Coaches will routinely ask about students’ physical and mental well-being during weekly session, with prompts provided in the behavior change coaching workbook. Coaches will report any concerns to the research team during regular bi-weekly coach check-ins. All adverse events will be reviewed for severity and monitored by the principal investigator. No formal provisions for ancillary or post-trial care are planned, as the intervention involves supervised PA and poses no greater risk than typical supervised PA offered in the community. Any serious adverse events will be reported to the University of Toronto REB in accordance with institutional policy.

Knowledge dissemination

A knowledge mobilization plan to optimize dissemination to our intended audiences (i.e., sport and recreation professionals, mental health professionals, students, academic audiences) has been developed. Practically, the findings from the research trial will serve to inform prevention and intervention efforts for mental health on post-secondary campuses. The research team will collaborate with our knowledge-users in on-campus sport and recreation and student mental health to disseminate the findings through dedicated platforms including Best Practices in Canadian Higher Education and NIRSA Leaders in College Recreation. Moreover, guided by the AGREE II instrument [82], the research findings will inform the development of a toolkit, which will include educational resources for coach training, as well as program delivery considerations to guide future implementation research and practice. A project summary infographic will be distributed to all study participants and shared through university media and mental health outreach outlets. Lastly, to target academic audiences, research findings and subsequent studies will be published in open-access, peer reviewed journals. The findings will also be presented at international and national conferences in the discipline of exercise psychology (e.g., North American Society for the Psychology of Sport and Physical Activity), youth mental health (e.g., International Association for Youth Mental Health) and public health (e.g., International Society for Physical Activity and Health). Due to ethical considerations for human research participant data, data and statistical code will only be available upon project completion (October 24, 2025) by request from the corresponding author, Dr. Catherine M. Sabiston (email: catherine.sabiston@utoronto.ca). The behaviour change coaching workbook used to guide the PA sessions will also be available by request from the corresponding author.