What a Dated Search Still Did Not Find: Wrist-Read Completion and Same-Day Load
Not medical advice. Educational use only. Data from published literature.
Two documentation questions kept showing up as inference rather than citation: one about wrist-worn heart-rate readings used to mark a session “complete,” and one about what to do when school and sport both provoke symptoms on the same day. Labeling them “inferred tension” felt honest. It also felt like an IOU—something that needed a dated look at primary literature, not another re-read of earlier notes.
This Foundation Stack methods piece is narrow on purpose. It asks those two questions of the published literature and reports what a disclosed search found.
The two questions:
- Wrist-read completion. Does a wrist-worn heart-rate reading ever get treated differently, in the published literature, depending on which rule advances a prescription—versus a chest-strap or clinic reading?
- Same-day combined load. Is there a published rule for what to do when a student-athlete has a symptom flare-up on the same day while being tracked on both the return-to-school and return-to-sport ladders at once?
Neither question has a sentence in a trial Methods section or consensus appendix that answers it. This piece goes back to primary literature to check.
Abbreviations used here: BCTT = Buffalo Concussion Treadmill Test; RTL / RTP = return to learn / return to sport (consensus documents often label sport steps RTS); PMID = PubMed identifier (linked below for every source).
1. Problem
The two questions combine facts that are documented separately. What was missing was a stated rule for the combination.
The result, stated up front: a dated, disclosed search of primary literature still did not locate either rule. Both come back checked silence rather than a new resolution. That upgrade matters on its own terms: it changes what a clinician can honestly write on a documentation note today, from an inference about how two facts might interact to an absence someone actually went and checked.
Wrist-read completion, in full. One common way to authorize the next increase in a home aerobic prescription is a session-ramp rule: advance once the next scheduled session (or clinic visit) is marked complete. Micay’s fixed percentage ramp and Kurowski’s duration-titration protocol both use that trigger—session marked complete—even though their dose shapes differ PMID: 30018795 PMID: 27120294. Separately, validation work shows that a wrist-worn optical heart-rate reading often disagrees with a reference reading by more than most prescribed heart-rate bands are wide—too wide to trust for confirming a session stayed inside a narrow band PMID: 28709155 PMID: 29881626. Neither literature strand states whether those two facts combine: does a session-ramp completion trigger depend on the session having been confirmed in-band, and does that confirmation’s reliability change depending on which device did the reading?
Same-day combined load, in full. Return-to-learn and return-to-sport are parallel step-ladders. A practical default used in clinic notes is: if either track generates a symptom escalation of more than 2/10 lasting more than an hour on a given day, defer advancement on both tracks the next day. That specific same-day hold has no published source—it is a proposed clinical-judgment default, not a citation. The high-level consensus gate (full RTL before unrestricted RTP) is published; the micro-rule for same-day combined load is not.
Why these two. Many documentation combinations in adolescent PPCS aerobic care are simply silent—no source addresses them either way. These two stand out because each points to a specific mechanism connecting two separately-documented facts, not just an unaddressed pairing. A silent cell means no one has said anything about a combination; these two meant something more specific—two things that might pull against each other—which is what made them worth a dated search rather than another backlog entry.
The wrist-optical channel’s exclusion from adjudicating a narrow heart-rate band stays untouched as a clinical caution grounded in validation LOA widths. This piece asks a narrower question: whether any primary source conditions advancement on device class, or states a same-day cross-track hold—and reports what a real search found.
2. Methods — search protocol
Two independent search questions were run as their own literature-scout cycle, dated 2026-07-15:
| Field | Value |
|---|---|
| Corpus | GLP research-assistant corpus (concussion-research-rag), queried via its RAG CLI |
| Synthesis | LLM synthesis step over retrieved chunks; near-miss chunks inspected directly |
| Independent check | Direct full-text keyword scan (pypdf) of the Amsterdam and Berlin consensus statements — not RAG retrieval alone |
| Not claimed | An exhaustive PubMed/Embase systematic review; grey literature or offline guideline supplements were not searched |
A RAG (retrieval-augmented generation) search over a working corpus is not equivalent to a formal systematic-review search of Embase, Medline, and Cochrane with a documented multi-string, multi-reviewer audit trail: it retrieves whichever chunks a vector index scores as most relevant to a query, then runs an LLM synthesis step over those chunks. That is fast enough to run twice for a short methods piece, but it stops well short of what a full systematic review would require. For context, the underlying corpus covers several hundred concussion and PPCS aerobic-exercise papers; each query below returned only 8–10 chunks worth inspecting, which reflects how sparse this specific corner of the literature actually is.
Wrist-read completion query (verbatim): Does any adolescent PPCS or sport-related concussion aerobic exercise trial or consensus statement condition session-completion or progression-advancement decisions on the monitoring device class used to confirm in-band heart rate (wrist optical vs. chest strap)? Explicitly look for any paper that links wearable/device type to progression logic. Searched deliberately beyond the device-validation papers already used to establish the wrist-reliability argument.
Same-day combined-load query (verbatim): Does any concussion consensus statement or aerobic exercise protocol give an explicit same-day rule for combined return-to-learn and return-to-sport symptom load — for example holding advancement on both tracks the next day if either shows symptom escalation? Beyond the high-level rule that full return-to-learn must precede unrestricted return-to-sport. The Amsterdam and Berlin consensus documents were checked directly by keyword scan for this specific question, since both already supply the return-to-learn/return-to-sport sequencing gate but had not been searched for a same-day combined-load rule specifically.
A second, broader query was run to catch near-misses the first, narrowly-phrased question might miss: wrist optical heart rate monitor progression OR advancement OR session completion concussion OR PPCS OR BCTT chest strap Polar. This is what surfaced the Leddy/Buffalo Polar H7 detail and the Mercier, Del Rossi, Henke, and Chizuk near-misses listed in Table A below.
Every PMID this search touched was independently re-verified rather than relayed from a prior receipt. A concrete example of that re-verification in practice: the wrist-read completion search’s closest near-miss initially returned two candidate Mercier papers sharing the same last name — PMID 32024486, a 2020 trial protocol, and PMID 39427780, a 2024 completed-trial report from the same research group. Only the 2020 protocol’s abstract matched the specific 70–80%-of-symptom-exacerbation-heart-rate detail this search was checking; the 2024 paper is a later, different publication and was excluded. Table A cites only the verified match. No other PMID issues surfaced this pass.
Prior same-day scout note. An earlier literature note in this series already flagged the same-day RTL+RTP micro-rule as unpublished (Yang 2024 and Wingerson 2024 discuss related loads, not the hold-both-tracks rule). Those papers are not re-listed in Table A below — this piece re-checks the consensus primaries with an explicit same-day search string rather than repeating that earlier partial scout as table rows.
3. Comparison — near-miss audit
Every source below was located, and none settles either question. Nine candidates were examined — six for wrist-read completion, three for same-day combined load — and every Settles? cell is “No.”
| Question | Source (PMID) | Claim tested | What the source actually says | Classification | Settles? |
|---|---|---|---|---|---|
| Wrist-read completion | Leddy/Buffalo RCT (30715132) | Device class named for home monitoring | “a provided Polar H7 Bluetooth Heart Rate Sensor” — a chest strap is specified in Methods | Near-miss — names a device class, but never conditions the advancement trigger on it | No |
| Wrist-read completion | Howell/Wilson et al. 2022 (35489100) | Device class named for home monitoring | Pilot RCT: home sessions tracked with a chest-strap HR monitor (Garmin HRM-Dual); prescription at 80% of exercise-test HR, 20 min/day, 5 days/week | Near-miss — names a device class, same pattern as Leddy; never conditions the advancement trigger on it | No |
| Wrist-read completion | Mercier et al. 2020 (32024486) | Device-conditioned progression | Adult PPCS aerobic-exercise protocol paper: prescription at 70–80% of symptom-exacerbation HR; target updated every 3 weeks via a repeat treadmill test | Near-miss — adult population outside the adolescent query frame; a retest cadence, not a device-class rule | No |
| Wrist-read completion | Del Rossi et al. 2020 (32503036) | Device-conditioned progression | Narrative review of early aerobic exercise after pediatric concussion; discusses symptom-gated dosing, not monitoring hardware | Near-miss — no device-class discussion at all | No |
| Wrist-read completion | Henke et al. 2020 (31629334) | Device-conditioned progression | Evidence review on early low-intensity exercise timing; recovery gated on symptom exacerbation | Near-miss — symptom gate, not a device-class rule | No |
| Wrist-read completion | Chizuk et al. 2023 (36728783) | Device-conditioned progression | Practical-management program for clinics without graded exercise testing; stop rule is symptom exacerbation (≥2-point rise), weekly re-evaluation | Near-miss — an explicit no-device pathway, still not a rule conditioning advancement on device class | No |
| Same-day combined load | Amsterdam 2022 consensus (37316210) | Same-day combined-load rule | “RTL and RTS strategies can occur in parallel… complete full RTL before unrestricted RTS” — a macro sequencing gate | False positive — a sequencing rule, not a same-day symptom-load rule | No |
| Same-day combined load | Berlin 2017 consensus (28446457) | “same-day” keyword hit | “no same-day return to play” — this is the day-of-injury return-to-play ban, an unrelated rule | False positive — different question entirely | No |
| Same-day combined load | Berlin 2017 consensus (28446457) | “concurrent” keyword hit | “published concurrently” (referring to companion systematic reviews) / “concurrent injury to the cervical spine” | False positive — both unrelated uses of the word | No |
4. Disagreements
Both questions land on the same outcome: a dated, disclosed search found nothing that states either combination, in either direction. That upgrades both from inferred tension (an audit-only guess) to a checked silence (a searched, dated, and still-empty result). It does not turn either question into a published rule for or against the combination—absence after search is not the same as a source that forbids the practice.
This search deliberately extended beyond the device-validation papers already used for the wrist-reliability argument — the question here is different (does any source condition the advancement trigger on device class), not a re-run of the LOA meta-picture. The clinical caution that a wrist reading cannot adjudicate a narrow heart-rate band stands untouched; this piece only checked whether any source separately ties that reliability question to which advancement rule is running, and found nothing that does either way.
A “confirmed-absent” result is easy to misread. Neither finding means “confirmed unsafe” or “confirmed incompatible.” It means no trial was designed to test whether a wrist-read completion trigger behaves differently under a session-ramp rule, and no consensus statement specifies a same-day cross-track symptom-load rule. Reading either checked silence as an implicit warning would invent a finding this search did not make.
5. SOP — what to write now
Sections 2–4 stay short because there is no new citation to build a case around. Practice changes are here. Both worked examples below are meant to be copied onto an actual note — the exact wording is part of what makes the caveat legible to whoever reads the chart next.
1. Session-ramp sessions. Record which monitoring channel supplied the heart-rate reading for each session, alongside the advancement rule in force. A wrist-optical reading still cannot serve as evidence that a session was completed in-band — published LOA widths for consumer wrist devices under exercise are typically wider than a narrow prescription band, and this search found no exception clause anywhere in primary literature for any progression logic. Nothing changes about what to do; what changes is that “no exception exists in the literature” is now a checked claim, not a guess.
What this looks like on a note: Wrist optical (ancillary) · session-ramp · session 4/8 complete · in-band status: not verifiable from wrist HR — do not use to confirm zone completion. Writing the device class and the caveat on the same line, rather than checking a generic “session complete” box, is what lets a covering AT or a parent reading the binder act on it without re-reading this piece.
2. Same-day RTL+RTP hold. Keep using the proposed rule — if either track generates a symptom escalation of more than 2/10 lasting more than an hour on a given day, defer advancement on both tracks the next day. The label on the documentation note must say adopted clinical-judgment convention, not a citation. Add one line pointing to this piece’s Methods search date (2026-07-15) — a search-trail reference, not a citation the rule doesn’t actually have.
What this looks like on a note: [Date] RTL step 3, RTP step 2 — symptom escalation >2/10 for 90 min on the sport track (meets the >1 h threshold) → hold advancement on both tracks per adopted convention (search trail: this article, Methods 2026-07-15). Reassess [date]. The parenthetical points to a search trail rather than a PMID on purpose — the difference that matters is between a rule nobody has published and a rule nobody bothered to look for.
3. Do not invent a new checkbox. A confirmed-absent result is not grounds to add a new “forbidden combination” field to any documentation form. Silence stays silence; checking it more carefully does not turn it into a rule.
6. Limitations
Confirmed-absent is corpus-bound and dated, not proof of universal non-existence. This search covered the GLP corpus plus a direct scan of the Amsterdam and Berlin consensus documents. It did not search grey literature, conference abstracts, or guideline supplements held offline. A future search could still find something this one did not. A search that wanted to close this gap more definitively would need at least three things this pass did not do: a formal multi-database search (Embase and Medline, not just this corpus) with a documented string and date; a hand search of the CISG consensus statements’ companion systematic-review papers — Berlin 2017’s own text states each consensus question was “the subject of a specific formal systematic review, which is published concurrently with this summary statement,” and this pass did not separately search those companion reviews; and a check of clinic-level quality-improvement or registry data, where a same-day combined-load convention might be documented operationally without ever appearing in a peer-reviewed source. None of those were in scope for this piece.
Scope stays narrow. This piece checks exactly the two questions above — wrist-read completion and same-day combined load — and no others. It does not reopen other documentation combinations, and it does not address citation-blocker edge cases outside these two questions.
No new trial comparisons. This piece only checks whether the published literature has ever stated a rule for either question — and reports that it has not. It does not test whether device class affects outcomes, or whether a same-day combined-load hold changes recovery time.
Series: Foundation Stack · methods · wrist-read completion and same-day load