In an exercise trial the intervention is the sessions people actually complete, so attendance, intensity and adherence matter as much as the outcome. Capture records session logs, functional tests, participant diaries and safety events in one study with an audit trail. Free sandbox, no credit card.
Free sandbox · No credit card · 21 CFR Part 11 aligned
Supervised session log (demo)
Participant 003-0017 · Week 6 · Session 2 of 3
Session
Session attended
ATTENDDuration
DURPlanned intensity
PLANDelivered as planned
DELIVPerceived exertion (0-10)
RPEAdverse event this session
AEYNWhat an exercise trial needs from its data capture
The study data
Exercise trials come in many forms: supervised programmes in a clinic or gym, home-based programmes with check-ins, structured walking, resistance training in older adults, prehabilitation before surgery, exercise alongside cancer treatment. What they share is that the intervention is complex and behavioural. Unlike a tablet, the dose depends on what the participant does: how often, how hard, for how long, and with what supervision. A trial that records only baseline and endpoint measurements cannot tell whether a null result reflects a programme that did not work or a programme that was not delivered.
That is why reporting guidance for exercise trials is so detailed. The Consensus on Exercise Reporting Template (CERT) is a 16-item checklist developed through an international expert process and designed to align with CONSORT and TIDieR; its last section asks for planned and actual adherence. Reviews of published trials keep finding that reporting is incomplete, so a trial that captures the information in structured fields from the start has a clear advantage when writing the paper. The CONSORT flow can be generated with the CONSORT flow diagram generator once your numbers are in.
A practical way to think about the dataset is in four layers. The first is eligibility and baseline: history, anthropometrics, and screening for exercise safety. The second is the intervention record: a session log per participant per session, with attendance, duration, content, progression and deviations. The third is outcomes: functional tests, imaging or biomarkers, and patient-reported measures. The fourth is safety: events during or after sessions. Capture builds each layer as eCRF forms or ePRO tasks in one study, so the session log and the outcomes share a participant ID and an audit trail.
Define adherence in the protocol before you build: sessions attended divided by sessions prescribed, minutes completed against minutes prescribed, or proportion of sessions at target intensity. Then give each element its own field so the number can be derived in the export, rather than typed in by hand. If the protocol allows make-up sessions or modifications, add a reason field. The calculation itself is done in your analysis; Capture records the elements and the audit trail shows any later changes.
Participants
40
Sessions prescribed
24 each
Median attended
19
Injury events
2
64 percent
25 percent
11 percent
Protocol to build
| Protocol element | What the data looks like | Where it lives in Capture |
|---|---|---|
| Baseline and safety screening | History, blood pressure, medications, readiness to exercise | Medical history, vital signs, concomitant medications |
| Randomization to arms | Exercise vs control, stratified by age, sex or site | Randomization in the same platform; stratified options |
| Session log | Attended, duration, intensity, delivered as planned, exertion | Repeating eCRF form per session or table section with rows per exercise |
| Functional tests | Walk distance, grip strength, VO2, body composition | 6-minute walk, grip strength and SPPB, VO2max, DEXA |
| Participant-reported outcomes | Pain, fatigue, quality of life, home activity diary | ePRO tasks, e.g. fatigue log and pain diary |
| Wearable activity data | Steps, heart rate, sleep from a connected device | Native OAuth device sync for devices the study enables; ask in a demo which devices are available |
| Blinded assessment | Outcome assessor does not know the arm | Role-based access; blinded roles never receive treatment-arm values |
Licensed questionnaires stay with your study documents; Capture does not supply or verify instrument licences.
Create the session form, run a few practice participants through several weeks and look at the export your statistician will use. Free sandbox, every feature, no credit card.
Design and safety
You cannot blind participants to exercise, but you can blind the people who measure outcomes. Assessor blinding is a common safeguard against bias in functional tests. In Capture, blinded roles never receive treatment-arm values; that is enforced by masked database views and row-level security rather than by hiding a field on the screen. Give the assessors a blinded role and the session leaders an unblinded one, and the same study can hold both the delivery record and the outcome record. See randomization software for the allocation side.
Exercise carries risk as well as benefit, and the protocol should say which events are reportable. Falls, muscle and joint injuries, chest pain, syncope and cardiovascular events can all occur in or soon after a session. Capture records each as an adverse event with onset, severity, seriousness, relationship to the intervention and action, and an event during a session can be linked to the session log by date. The adverse event template is the base form. For frail or older participants, see also EDC for frailty and healthy ageing trials.
Home-based programmes depend on the participant’s own reports. A short diary through the ePRO module opens from a link in the phone browser, with no app, and can ask about sessions done, exertion, pain and sleep. Reminders keep completion up and the compliance view shows who is overdue. Where a wearable is part of the design, device data can sync natively for devices the study enables, which complements the self-reported log; see wearable device studies for the pattern.
Build sequence
Frequency, duration, intensity and progression rules, and what counts as delivered as planned.
Upload the protocol for an AI draft, then review. Start from templates for baseline, tests and safety.
A per-session eCRF form for staff and a short ePRO task for home sessions.
Duration and heart-rate ranges, required reason when a session is modified, date logic between sessions and visits.
Run practice participants for several weeks of sessions in the sandbox, approve the forms and start the paid live phase.
Before first participant
The formula, the denominator and who calculates it.
A CERT or TIDieR style description kept with the study documents.
Criteria for exercise readiness and who reviews borderline cases.
Roles assigned and test staff separate from session leaders.
A wide CSV with data dictionary opened in the statistics package.
Yes. A session log can be a repeating eCRF form or a table section with one row per exercise. Attendance, duration, content, intensity and deviations are typed fields with edit checks.
Capture records the components, such as sessions prescribed and attended, and exports them with a data dictionary. Your protocol defines the adherence measure and the calculation is done in the analysis.
Yes. ePRO questionnaires open from a secure link in the phone browser, with no app, and reminders help completion.
Yes. Blinded roles never receive treatment-arm values, enforced by masked database views and row-level security.
Capture provides 21 CFR Part 11-aligned controls: a field-level audit trail, electronic signatures and role-based access. Compliance is shared with the sponsor’s validated use of the system. This is not legal advice.
Keep exploring
EDC for sports medicine trials
Performance and injury studies.
Clinical trial software for rehabilitation research
Therapy-delivered interventions.
EDC for frailty and healthy ageing
Older participants and function.
EDC for wearable device studies
Activity data from devices.
6-minute walk test template
A common functional endpoint.
Build your exercise trial in the free sandbox. No credit card, and you pay only when real participants enrol.