Medical Direction & Legal State of the Evidence v0.1

Medical Direction, Regulation & Law — State of the Evidence

Domain: medical-direction-legal | Status: DRAFT — pending physician review (BA) Revised: 2026-08-16 (v2) — supersedes the 2026-08-08 draft. What changed: the 51-jurisdiction survey has landed and the Jaslow-update framing is retired; Connecticut's rule is now primary-verified; the San Francisco thread is resolved (a study, not an ordinance, plus its 2025 successor policy); and the 2026 Purple Guide tranche adds the SAG governance layer, the decision-accountability standard, and the P/S capacity discount.

Bottom line for practitioners

In the United States, the legal architecture of event medicine is mostly empty space — but the size of the space is now measured rather than assumed. MGMI's 51-jurisdiction survey found roughly 22 states with some regulatory instrument touching mass-gathering medical care, most built on bare attendance thresholds with no visible derivation. Almost no state tells you whether your festival needs a physician, what a medical plan must contain, or who holds command when the event becomes an incident. What fills the vacuum is professional consensus — the NAEMSP position statement and its resource document — which is authoritative but not enforceable. The United Kingdom shows the counterfactual twice over: the Green Guide's needs-assessment-driven, checklist-specified regime for sports grounds, and a standing multi-agency review body (the Safety Advisory Group) born from the Hillsborough inquiry. Practically: treat ^[1] as your standard of care, borrow the Green Guide's specificity for your medical plan, and assume that if something goes wrong, you will be judged against consensus documents you were never legally required to read — on a decision standard of what was known at the time, provided you documented it.

The regulatory landscape — measured, at last

The founding empirical fact of this domain was a quarter-century old: Jaslow's 1999 survey found only 6 states (12%) regulating medical care at mass gatherings, one requiring physician oversight ^[2]. That survey has now been updated. MGMI's 51-jurisdiction review found on the order of 22 states with an instrument in force — a real expansion since 1999, and still a minority of states — with the count governed by a published inclusion rule (the District of Columbia's permitting matrix, for instance, is ruled outside the count on stated criteria, with the rule and the matrix published rather than silently applied). The instruments are idiosyncratic rather than systematic, and the dominant design is a headcount trigger: thresholds in force include 500, 1,000, 5,000, 15,000, 30,000 and 50,000 attendees, with no evident derivation for any of them.

Two instruments are now primary-verified rather than catalogued:

Connecticut §19-13-B87 — verified verbatim against the state's own portal — requires one physician per 1,000 attendees "or fraction thereof" (the ratio does not round down), plus drug-custody arrangements, patient record-keeping (name, address, tentative diagnosis — the only US instrument we hold that mandates event medical documentation), and a telephone connection to the medical director ^[3].

San Francisco — the domain's cautionary tale about evidence and policy moving in opposite directions. The city's 2006 rule required standby ambulances at gatherings over 15,500. Meites & Brown then measured what it produced across 47 events: a transport rate of 1 per 59,000 attendees against a community baseline of 1 per 20,000 residents per six hours (RR 0.15, p<0.001), with 46% of mandated standby ambulances unused — direct empirical evidence that the headcount trigger over-provisioned ^[4]. The city's current policy, EMSA 7010 (effective 2025), responded not by relaxing the trigger but by lowering it six-fold to 2,500 within a five-level resource ladder ^[5]. Whether that is a considered rejection of the finding or policy drift that never revisited it cannot be determined from the documents; either way, a jurisdiction whose own published evidence and current regulation point in opposite directions belongs in the manuscript.

The consensus spine

Where law is silent, NAEMSP speaks. The 2021 position statement ^[1] — superseding the 2000 position — defines the modern standard: an event medical director who is a physician knowledgeable in EMS and mass-gathering logistics, with documented leadership in the Medical Plan (ICS Form 206), oversight of the Safety Plan (ICS 208), a communications plan on ICS 205 with a dedicated medical channel and a predesignated MCI channel, clinician credentialing against local scope of practice, and participation in unified-command MCI preplanning. Its goals are explicit: on-site stabilization including expedient critical care; reduced burden on host-jurisdiction EMS and hospitals through physician-overseen alternative disposition; MCI readiness; and care at least commensurate with local standards. Planning doctrine combines predictive models (with awareness of their limits), the event's own historical data, and the PPR/MUR/TTHR metrics. The statement is also candid about definitional drift: the >1,000-person threshold is common, resource-based definitions are preferred, and most published data come from events above 25,000 ^[1].

The companion resource document ^[6] supplies the reasoning: historical data from the same or comparable events is the most reliable predictor of resource needs; standardized PPR/TTHR metrics enable comparison; and an event-dedicated medical director should be engaged across planning, preparation, response, and recovery. (Curation note: its stored DOI is unregistered; cite by PMID 26270473 pending verification.) Lund and colleagues add the strategic frame — event medicine as a rehearsal space for disaster preparedness ^[7] — and the WHO research agenda situates the domain within global health security ^[8].

The binding contrast — now two UK layers

The instrument layer: the Green Guide. Chapter 18 is what enforceable regulation looks like ^[9]. A Medical Needs Assessment governs — provision is set by an MNA developed with the statutory NHS ambulance trust and CQC-registered providers, and deviation from any minimum is permitted only when substantiated by the MNA and recorded in the Medical Plan. The Medical Plan has a 16-element checklist, from treatment-room sizes and command structure through fatalities procedures and travel times to trauma, stroke, heart-attack, and paediatric centres. Event Doctor provision is thresholded, with specified qualifications and event-day duties. A Medical Advisory Group reports to the Safety Advisory Group. Command transfer is explicit: on a declared major incident, all medical staff come under the senior statutory NHS ambulance officer, with written handover protocols ^[9]. A further capacity-side mechanism with no US analogue at any level of government: holding capacity is discounted by the P factor (physical condition) and the S factor (quality of safety management) — a venue run badly has a lower lawful capacity in identical square metres. That is post-Hillsborough machinery; US occupancy codes compute load from area and egress alone ^[10].

The governance layer: Safety Advisory Groups. SAGs exist substantially because of Lord Justice Taylor's Hillsborough inquiry — a crush disaster produced a standing multi-agency review body, mandated for football and generalised voluntarily beyond it. Core membership includes police, fire, the NHS ambulance service, licensing, health representatives and transport; the event's own medical provider is an invited member only — internally coherent, since the Guide requires members to be independent of the event, but it leaves a medical-competence gap the guidance never says how to fill. Submission lead times are concrete: three months standard, four weeks minimum, six months for major events — implying the medical director must be engaged around nine months out, far earlier than typical US practice. Three honest caveats belong in any comparison that cites the model: SAGs are not a legal requirement outside sports grounds; they are advisory with no bespoke powers (though their advice "would certainly have significant influence on any legal process"); and §55 is the sentence every event medical director should read — SAG approval does not transfer responsibility for safety from the organiser. No safe harbour ^[11]. The referral instrument is also structurally different from US permitting: the Guide's own referral diagram contains no attendance factor at all — ten risk gears (location, traffic, organiser inexperience, criminality, iconicity, profile/behaviour, new event/venue, distance, audience, history) drive a central "levels of risk," and the chapter states directly that risk "may be greater with events that may not reach" any attendance trigger ^[11]. One SAG consideration is absent from every US instrument we surveyed: the cumulative consequences of multiple concurrent events in one area — two 20,000-person events on one weekend are two demand curves onto one regional EMS system, and every US permit assesses its event in isolation.

The accountability standard. The Purple Guide's Contingency chapter states the test for decisions made in extreme situations: they will not always be correct, and they are judged on being proportionate, lawful ("there are no exceptions to the law in incident management"), necessary in the light of what was known at the time, and documented with a rationale ^[12]. That is the correct posture against outcome-anchored retrospective scrutiny, and contemporaneous documentation is the trainable habit it implies. The same chapter summarises the Civil Contingencies Act 2004 duties on Category One Responders — the UK's statutory planning hook, and the closest analogue to the state mass-gathering statutes in our survey.

The resourcing heterodoxy worth arguing about. The Campsites chapter proposes sizing multi-day camping-event medical resources to "the demands that would be placed on a GP practice serving a community of similar size" — a residential primary-care model categorically different from the acute PPR model US practice uses, predicting a flat baseline (medication continuity, minor illness, emergency dentistry, mental health) that presentation-rate models do not see. For multi-day events the honest answer is probably both, additively — and the same chapter requires medical cover for the whole time campers are on site, not the licensed entertainment hours ^[13].

What the evidence adds

Three empirical anchors justify the consensus positions. On-site physician care works: at a California motorsports event, on-site physicians cut ambulance transports by 89% (116 to 13; p<0.001), with paramedics able to disposition 52% of patients, protocol-enabled RNs another 39%, and roughly 9% needing physician-level care ^[14]. The base rate of critical illness is low: across 253 events and 3.3 million attendees, 2,762 encounters (0.08%), 95.3% returned to the event, 4.7% referred to an ED, and none admitted to an ICU ^[15]. The documentation on which oversight depends is undisciplined: rates are reported in a "varied, haphazard and author dependent" manner, motivating a minimum data set for mass-gathering health ^[16]. And Meites & Brown add the fourth anchor from the regulatory side: the one US headcount trigger anyone has formally evaluated over-provisioned by roughly a factor of seven ^[4].

What we don't know

- Whether any threshold works. The survey now tells us how many states regulate and how; no study except Meites & Brown evaluates whether any instrument changes outcomes — and that one found over-provision. - Refusal-of-care medicolegal exposure. RMA/AMA review is a NAEMSP quality-management expectation ^[1], but the medicolegal literature on event refusals is absent from this domain entirely. - The ACEP lineage. ACEP's 1996 crowd-care guideline, 2012 EMS medical-direction policy, and 2024 intervener-physician statement [acep-1996-provision; acep-2012-medical-direction-ems; acep-2024-intervener-physician, unextracted] remain unread, as do NFPA 101, FEMA IS-15, the DHS SEAR framework, WHO guidance, FIFA/UEFA medical articles, and the UKHSA 2024 scoping review [unextracted]. The Purple Guide is now extracted at chapter level for five chapters plus Venue & Site Design; unread remainders are listed in each entry's notes. - Schwartz full text. The resource document's staffing tables and oversight specifics are known only at abstract level ^[6].

How MGMI operationalizes

The composer applies jurisdiction gates: plans generated for a venue inherit the applicable instruments (e.g., NY Part 18, KY 202 KAR, CT §19-13-B87 where triggered) alongside the NAEMSP consensus baseline. Platform positions R1 (physician medical director), R3 (ICS-documented medical plan), and R10 (major-incident command transfer) trace their lineage to ^[1] and ^[9]; the transfer-of-primacy and Return-of-Authority mechanics now have guidance-level templates ^[12] ^[17]. The documentation habit implied by the §138 standard — contemporaneous, rationale-carrying — is a plan line, not advice.

Reading pathway

1. ^[1] — the standard; read first. 2. ^[9] — Chapter 18; the specificity to emulate. 3. ^[11] — the governance model, with its three caveats. 4. ^[4] — the only evaluated US threshold, and what it found. 5. ^[3] — the most demanding US instrument, primary-verified. 6. ^[2] — the historical baseline the survey updates. 7. ^[14], ^[15], ^[16] — the supporting evidence base.

Citations

^[2] Jaslow D, Drake M, Lewis J. Characteristics of state legislation governing medical care at mass gatherings. Prehosp Emerg Care. 1999;3(4):316–320.
^[1] Margolis AM, et al. Position Statement: Mass Gathering Medical Care. Prehosp Emerg Care. 2021. DOI 10.1080/10903127.2021.1903632.
^[6] Schwartz B, Nafziger S, Milsten A, Luk J, Yancey A. Mass Gathering Medical Care: Resource Document for the NAEMSP Position Statement. Prehosp Emerg Care. 2015;19(4):559–568. PMID 26270473.
^[9] Guide to Safety at Sports Grounds ("Green Guide"), 6th ed. Sports Grounds Safety Authority, 2018. Ch. 18, pp. 265–280 (page-verified).
^[3] Connecticut Public Health Code §19-13-B87 (verified verbatim from the CT eRegulations portal, 2026-08-10).
^[4] Meites E, Brown JF. Ambulance Need at Mass Gatherings. Prehosp Disaster Med. 2010;25(6):511–514. DOI 10.1017/s1049023x00008682. PMID 21181684. (Row formerly slugged sf-2006-standby-ambulance; title corrected 2026-08-15.)
^[5] San Francisco EMS Agency Policy 7010, Emergency Medical Services at Special Events, eff. 2025-04-01. (Single-source read; tier table pending visual verification.)
^[14] Grange JT, Baumann GW, Vaezazizi R. On-site physicians reduce ambulance transports at mass gatherings. Prehosp Emerg Care. 2003. DOI 10.1080/10903120390936518.
^[15] Varon J, et al. Critical illness at mass gatherings is uncommon. J Emerg Med. 2003. DOI 10.1016/j.jemermed.2003.03.001.
^[16] Ranse J, Hutton A. Minimum Data Set for Mass-Gathering Health Research and Evaluation. Prehosp Disaster Med. 2012. DOI 10.1017/s1049023x12001288.
^[7] Lund A, Gutman SJ, Turris SA. Mass gathering medicine: a practical means of enhancing disaster preparedness in Canada. CJEM. 2011. DOI 10.2310/8000.2011.110305.
^[8] Tam JS, et al. Research agenda for mass gatherings: a call to action. Lancet Infect Dis. 2012. DOI 10.1016/s1473-3099(11)70353-x.
^[11] · ^[12] · ^[13] · ^[17] · ^[10] Events Industry Forum, The Purple Guide, chapters published 2026-01-26 (chapter-level extractions; coverage in entry notes).
Unextracted shelf (existence/kind only): ^[18], ^[19], ^[20], ^[21], ^[22], ^[23], ^[24], ^[25], ^[26], ^[27], ^[28], ^[29], ^[30], ^[31], ^[32], ^[33], ^[34], ^[35], ^[36], ^[37], ^[38], ^[39], ^[40], ^[41], ^[42], ^[43].

Note at ingest: confirm the survey count phrasing against the published methodology page before publish — this draft states "on the order of 22 states" with DC governed by the published inclusion rule; the page is authoritative on the exact number.