Six Protocols, One Condition: A Clinician’s Guide to Adolescent PPCS Aerobic Exercise Prescription
Not medical advice. Educational use only. Data from published literature.
Somewhere in my second pass through the PPCS aerobic exercise literature, I stopped taking notes on outcomes and started counting something else: how many different ways clinicians were being told to prescribe the same intervention.
The Buffalo group measures a heart-rate threshold on a treadmill and prescribes at 80% of that number. Toronto tells athletes to walk or cycle at a slow to moderate pace and stop before breathlessness. NICE tells primary care clinicians not to provoke moderate symptom increase. Berlin gives six stages and a 24-hour rule. Pittsburgh SE uses graded aerobic rehabilitation under hospital PT for adolescents with prolonged symptoms (weeks, not days, post-injury). A symptom-limited approach skips the device entirely.
Six protocols. One condition. No shared decision framework.
For most adolescents moving through a school athletic trainer, a community PT, or a pediatrician’s office, this fragmentation is invisible—until it isn’t.
1. Why this fragmentation matters
The standard reassurance about PPCS is that most adolescents recover within four weeks. That is true. The 15–30% who do not are the problem this literature is trying to solve PMID: 35481264.
What the literature has not solved is the coordination problem that emerges the moment a slow-to-recover adolescent enters the care system. The athletic trainer clears the athlete for light aerobic work using symptom-limited rules. The pediatrician, trained on Buffalo, wants a formal BCTT before writing a prescription. The physical therapist uses a graded protocol borrowed from Pittsburgh SE. The parents have read something online about the Toronto return-to-sport stages. Everyone is following published guidance. No one is following the same guidance.
This isn’t a knowledge gap in the traditional sense. The six protocols reviewed here largely agree on the underlying theory—that sub-symptom-threshold aerobic exercise is beneficial, that strict rest is ineffective, and that individualization matters. Where they part ways is operational: what test, what formula, what progression rule, what setting.
The cost of that operational disagreement is borne by the adolescent. A 15-year-old with six weeks of post-concussive headache does not need a debate about whether 80% HRt or a 2/10 symptom cap is theoretically superior. She needs the adults in her care team to be working from the same page.
The protocols did not diverge because their authors disagreed about whether aerobic exercise helps PPCS. On that question, the evidence has grown substantially more consistent over the past decade. They diverged because they were solving different problems in different rooms.
Leddy and the Buffalo group were asking: how do we safely titrate exercise for a patient who has been symptomatic for months and whose threshold we need to measure precisely? The answer required a treadmill, a trained observer, and a formal protocol. McCrory and the Berlin consensus were asking: what should every clinician worldwide do immediately after a sport-related concussion? The answer required simplicity and a six-stage framework that needed no equipment. NICE was asking: what guidance can emergency and primary care clinicians act on in a single visit, with no follow-up guaranteed?
Each question produced a good answer. The problem is that a slow-to-recover 15-year-old moving through the care system encounters all of these answers simultaneously, delivered by different clinicians who do not know they are working from different frameworks.
The fragmentation also matters for how the research base is interpreted. If six protocols are in active clinical use, the external validity of any single trial becomes harder to assess. A trial conducted under Buffalo-protocol conditions tells us something important about what happens when those conditions are met. It tells us less about what happens in a school setting where none of those conditions exist, even if the underlying intervention is nominally the same.
Foundation Stack starts here—with the methods—because methods are the lens through which all subsequent evidence must be read. The six approaches are laid out side by side below, not to crown a winner (the evidence doesn’t yet support one) but so the clinician, the parent, and the adolescent can see exactly where they agree, where they diverge, and which protocol fits which setting.
2. Comparison of six published approaches
Table note: All approaches emphasize individualization, sub-symptom thresholds, and symptom monitoring PMID: 30742254. Foundation Stack value: collapse six clinical narratives into two decision trees—measure HRt or track symptoms.
| Approach | Testing | Prescription intensity | Progression rule | Typical setting | Source |
|---|---|---|---|---|---|
| Buffalo | BCTT treadmill: 3.2 mph start (≤5′10″), +1% grade/min | 80% HRt, 20 min/day | Weekly BCTT retest; advance when HRt rises | Outpatient adolescent PPCS; precision titration | 31105634 |
| Symptom-limited | No device; subjective symptom monitoring | Symptom increase ≤2/10, RPE 11–14 | If no worsening over 24 h, add 5–10 min | Home/community; limited equipment | 30742254 |
| Pittsburgh SE | Graded aerobic test + PT supervision | Individual sub-symptom threshold set on site | Increase load as tolerated | Prolonged PPCS (4–16 weeks post-injury); pediatric hospital PT | 27120294 |
| Toronto | Six-stage model; Stage 2 qualitative assessment | Slow to moderate pace; stop before breathlessness | Advance after ≥24 h asymptomatic per stage | Six-stage RTP derived from Berlin 2016 consensus; school return-to-play pathway | 28446457 |
| UK NICE | Red-flag screen + light activity | Do not provoke moderate symptom increase | Step up as symptoms improve | Emergency/primary care triage | NICE NG232 |
| Berlin 2016 | Clinical assessment; no device required | Begin aerobic work without provoking symptoms | Six stages; ≥24 h asymptomatic per stage | Global sport-concussion consensus (McCrory 2017) | 28446457 |
Reading the table
Two patterns are worth noting before moving to the disagreements.
First, the six approaches cluster into two families. Buffalo and Pittsburgh SE are anchored to physiological measurement—they require a test, produce a number, and build the prescription around that number. The remaining four—symptom-limited, Toronto, NICE, Berlin 2016—are anchored to symptom monitoring. They require observation rather than measurement, and they produce a qualitative threshold rather than a heart-rate target. The two families represent different points on a precision-versus-deployability tradeoff, not a better-versus-worse hierarchy.
Second, the “Typical setting” column does more work than it might appear to. It’s the column that should actually drive protocol selection. A clinician working in an emergency department should default to NICE, because it was designed for her setting, rather than Buffalo simply because it has the most published validation data. A school athletic trainer without treadmill access who applies symptom-limited rules isn’t using an inferior protocol—she’s using the one that fits her constraints.
The table is a matching tool. It answers “which approach fits where I am?” rather than “which approach is best?”
3. Three points of disagreement
Disagreement 1: Heart-rate formulae vs symptom caps
The Buffalo approach uses BCTT-measured threshold heart rate (HRt), with exercise prescribed at 80% HRt PMID: 31105634. Pittsburgh work also anchors prescription to BCTT PMID: 27120294. Symptom-limited, NICE, and Toronto protocols rely on symptom caps rather than formulae—typically a ≤2/10-point increase in symptoms PMID: 30742254 PMID: 28446457. Toronto Stage 2 describes only qualitative “slow to moderate” walking or cycling PMID: 28446457 (six-stage staging derived from the Berlin 2016 consensus, McCrory 2017). The split reflects setting: Buffalo targets outpatient PPCS and favors precision; NICE targets emergency and primary care and favors simplicity NICE guideline NG232. Reviews suggest formula-based testing is useful for assessment but less deployable in community settings than symptom-based rules PMID: 30742254.
The split is not “both equally unvalidated.” BCTT-anchored prescription buys precision when you have a treadmill, trained staff, and weekly follow-up. Symptom caps buy deployability when you do not. Reviews support both classes of tool for adolescent concussion—formula testing for assessment, symptom rules for community follow-through PMID: 30742254. What remains genuinely contested is how fast to advance (Disagreement 2), not whether symptom caps are a lesser science. For the school athletic trainer in Section 4—no treadmill, two-week referral wait—I would default to symptom-limited staging first, not because BCTT is wrong, but because waiting for treadmill access often means no structured activity at all.
Disagreement 2: Progression criteria
Buffalo requires weekly BCTT retesting and advancement only after HRt increases PMID: 31105634. Pittsburgh SE (Kurowski 2017) did not mandate weekly BCTT retesting; load was advanced as tolerated under PT supervision. In this pilot RCT of adolescents with prolonged PPCS (4–16 weeks post-injury), the aerobic training group showed significantly faster symptom improvement than a stretching control (group × time interaction, p = 0.044 on the PCSI), even though adherence in the aerobic group was lower than in the stretching group. Notably, the stretching group also improved relative to baseline PMID: 27120294.
That finding narrows what a n≈19 pilot can claim. Both arms improved; aerobic training improved faster on the PCSI interaction. The trial was not powered to separate “aerobic-specific physiology” from “supervised structure plus contact with staff.” It does not collapse this piece’s six-way comparison—Buffalo, Toronto, NICE, and Berlin still disagree on how to set intensity and when to advance—but it does mean Kurowski cannot be read as proof that only aerobic load beats rest. Progression mechanics—weekly BCTT retest versus 24-hour symptom rules—remain the honest axis of disagreement here. Toronto return-to-sport staging and NICE align on ≥24 hours without symptom escalation before advancing PMID: 28446457 NICE guideline NG232. The tension is physiological titration (Buffalo) versus time-plus-symptom rules (community guidelines). Pediatric reviews argue strict rest is ineffective and early activity is beneficial, so most guidance has shifted toward “24 h without worsening” as sufficient to progress PMID: 30742254.
What the literature has not addressed directly is the cost of getting the progression rate wrong in either direction. Advancing too quickly risks symptom flare and psychological setback—a return to baseline after apparent progress is one of the most demoralizing experiences for adolescents and families already navigating prolonged recovery. Advancing too slowly risks deconditioning, increasing anxiety about exercise, and potentially extending the recovery timeline through inactivity.
The 24-hour rule offers a practical middle ground, but it was not derived from a controlled comparison of progression rates. It represents expert consensus on what is safe rather than what is optimal. Whether a more aggressive or more conservative progression would produce meaningfully different outcomes in adolescent PPCS remains an open question that current trials are not powered to answer.
Disagreement 3: Environmental factors
Most RCTs were conducted in temperate climates. The Buffalo protocol uses treadmill testing in indoor, room-temperature conditions PMID: 31105634. Toronto and UK guidance mention walking and skating but do not quantify effects of cold, ice, or altitude on HRt PMID: 28446457. Systematic reviews call for individualized prescription yet offer little evidence for environmental correction PMID: 30742254. Clinicians in high-altitude or heavy-snow regions therefore extrapolate published data without validated adjustment tools PMID: 30742254.
The absence of environmental correction data is partly a consequence of how the evidence base was built. RCTs are expensive and logistically complex; researchers conduct them where infrastructure supports them, which typically means university sports medicine clinics in temperate urban settings. The resulting protocols carry implicit assumptions about temperature, altitude, and surface conditions that are rarely stated because they are rarely noticed in the environments where they were developed.
There is a practical workaround worth naming, even without validated correction factors. Clinicians in high-altitude or cold-weather environments can apply a more conservative interpretation of any HRt-based threshold—treating the measured value as an upper bound rather than a target, and defaulting to symptom monitoring as the primary progression signal. The Buffalo protocol doesn’t specify this workaround, but it stays consistent with the underlying principle: stay below the threshold that provokes symptoms.
4. Matching the protocol to your setting
The six protocols were not designed for the same environment. Buffalo was developed and validated in a university sports medicine clinic with treadmill access, trained staff, and weekly follow-up capacity. That infrastructure is not universal.
Consider a more common scenario: a 14-year-old presents to a school athletic trainer three weeks after a soccer collision. Symptoms have not resolved. There is no treadmill in the building. The nearest sports medicine physician is 45 minutes away. The next appointment is in two weeks. The parents work full-time and cannot supervise daily exercise sessions.
In this setting, the BCTT-anchored Buffalo protocol is technically correct and practically inaccessible. The symptom-limited approach—no device, no formula, cap activity when symptoms rise more than 2/10 points—is what the evidence supports for exactly this scenario. Toronto’s qualitative stage 2 description (“slow to moderate pace, stop before breathlessness”) translates directly into language a parent can monitor from a text message.
The environmental variable compounds this. Most RCTs generating the evidence base were conducted in temperate, low-altitude, indoor settings. Pocatello sits at 1,356 meters. Winter temperatures drop below −20°C. Ice and snow surfaces change gait mechanics and perceived exertion in ways the published correction literature has not addressed. A clinician in this environment applying Buffalo-derived HRt values without adjustment is extrapolating beyond the data, whether they know it or not.
Buffalo isn’t wrong here—protocol selection is simply a matching problem: match the protocol to the infrastructure, the supervision capacity, the climate, and the patient. The table in Section 2 is built for that job.
The honest answer for a resource-limited community setting is: start with symptom-limited, document carefully, and escalate to BCTT-anchored prescription if the patient has access to a facility that can support it.
5. One-page SOP: clinician / parent / adolescent checklist
The checklist below is designed for the resource-limited community setting described in Section 4. Rather than a clinical protocol, it’s a shared language document—a way for the clinician, the parent, and the adolescent to work from the same framework even when they aren’t in the same room.
The three-column format is intentional. Most published protocols are written for clinicians. They describe what the clinician should measure, prescribe, and monitor. They say less about what the parent should watch for, and almost nothing about what the adolescent should understand about their own recovery. The gap between clinical prescription and family execution is where real-world adherence breaks down. This checklist attempts to close that gap by giving each party their own column.
Note: This table synthesizes published Buffalo, Toronto, and NICE pathways for education only; it does not replace individual clinical care PMID: 31105634 PMID: 28446457 NICE guideline NG232. Foundation Stack value: one shared sheet so clinician, parent, and adolescent follow the same decision tree—reducing “clinician says exercise, parent hesitates, adolescent is unsure.”
| Clinician | Parent | Adolescent |
|---|---|---|
| □ Screen red flags: vomiting, GCS <15, focal neuro signs → emergency | □ Watch red flags: thunderclap headache, repeated vomiting, slurred speech → call emergency services | □ Learn 0–10 ratings for headache, dizziness, nausea; report morning, midday, evening |
| □ Start BCTT or symptom assessment at 24–48 h; record HRt | □ Relative rest first 48 h: light walking, chores; no exams or marathon gaming | □ 20 min/day aerobic (brisk walk or bike); able to speak full sentences without gasping |
| □ Prescribe: BCTT → 80% HRt, 20 min/day; symptom-limited → stay ≤+2/10 vs baseline | □ Monitor RPE 11–14; stop if symptoms are +2/10 above session start | □ Hard stop: if symptoms are +2/10 vs start, stop immediately and tell an adult |
| □ Progress: weekly BCTT with rising HRt, or 24 h without symptom escalation | □ Log date, duration, symptom scores; avoid “Can I play yet?” as the only metric | □ Schoolwork in blocks: 15 min study, 5 min break; stop and rest eyes if headache +2 |
| □ Clearance: asymptomatic exam, exercise to RPE 19–20 without flare on ≥3 occasions | □ Return to school in stages: half day → full day; sport as spectator → non-contact → contact | □ Step back one level if symptoms return after return to school—do not push through |
| □ Medical release: non-contact training ≥24 h symptom-free; signed note for school | □ Photograph signed clearance for records; reduces disputes with coaches | □ Do not enter contact sport or competition without written medical clearance |
6. Limitations
Generalizability: Evidence is drawn from adolescent PPCS. Children <6 years, adults >65 years, cervical spine injury, and severe anxiety are not validated here. BCTT requires a treadmill and is not applicable to wheelchair users PMID: 31105634.
Environment: RCTs are largely room-temperature, low-altitude settings. For example, Pocatello (~1356 m) with winter ice near −20 °C lacks correction factors for HRt drift PMID: 30742254.
Device error: Prescription trials that titrate to BCTT thresholds or %HRmax typically use chest-strap monitoring, not wrist photoplethysmography PMID: 30715132. Wrist monitors rely on photoplethysmography and show variable accuracy during aerobic exercise PMID: 28709155; in one graded-intensity validation study, limits of agreement versus reference were −20 to 19 bpm for a wrist optical sensor PMID: 33764310. Worts 2019 discusses physiologically based, BCTT-anchored home exercise but does not quantify wrist-device error PMID: 30742254. Community programs that read a wrist target as if it were trial-grade HR measurement may therefore mis-rank intensity relative to the evidence base.1
Long-term outcomes: Endpoints are often symptom scales at 28–90 days. One-year academic performance and migraine recurrence are rarely reported PMID: 30742254.
Adherence and supervision: Trial adherence often exceeds 80%; real-world parental supervision is often <40%. This SOP does not solve unsupervised home exercise PMID: 27120294.
Scope: Educational synthesis for protocol matching—not individualized medical advice. F-1 trainees must not provide remote clinical care.
What these limitations mean in practice
These limitations do not invalidate the protocols reviewed here. They describe the boundary conditions under which published evidence applies—and beyond which clinical judgment must fill the gap.
The most practically significant limitation for community settings is the gap between trial conditions (supervised visits, structured follow-up) and what families can sustain at home. Rather than a reason to withhold exercise prescription, that gap is a reason to calibrate expectations, build the SOP around what parents can realistically do, and document execution in clinical notes instead of assuming trial conditions apply by default.
2026-07-03 — Limitation #3 (device error). Corrected unsupported ±10 bpm attribution to Worts 2019; anchors moved to device-validation studies cited above.
The comparison above will not age well, and that’s by design. New trials will shift the evidence and protocols will be revised. When they are, this piece will be updated with a changelog note rather than quietly replaced, because the point is not to be permanently right but to be consistently honest about what the evidence supports at a given moment.
Series: Foundation Stack · methods · protocol comparison
Footnotes
Errata (2026-07-03), limitation #3. The originally published sentence attributed a ±10 bpm wrist-photoplethysmography margin to Worts et al. 2019 PMID: 30742254. Full-text search of that review (Foundation Stack / GLP verification, 2026-07-03) did not recover that numeric claim or a PPG-versus-ECG accuracy statement. The clinical concern—that community wrist readouts may not match chest-strap or treadmill reference used in prescription evidence—remains; the evidentiary anchor has been corrected to device-validation studies PMID: 28709155, PMID: 33764310. Worts 2019 is retained only for prescription-framework context (BCTT / home exercise), not for optical error magnitude.↩︎