MGMIMass Gathering Medicine

← State of the Evidence

MCI Posture State of the Evidence v0.1

Mass-Casualty Readiness at Planned Events — State of the Evidence

Bottom line for practitioners

A planned event is the one setting in which a mass-casualty incident can be met by a response that was designed, staffed, and rehearsed before the first patient existed. The corpus supports treating the event medical plan as a pre-positioned MCI response: incident command named in advance, triage and hemorrhage-control doctrine drilled, surge conversion of on-site assets scripted, and hospital notification pathways pre-agreed. The strongest outcome data we hold — London 2005 and Boston 2013 — are stories of preparation meeting the event, not improvisation [1] [2]. Manchester adds the sobering corollary: preparation is obligatory precisely because timing and casualty numbers are unpredictable even for well-prepared units [3]. A 2026 addition from UK national guidance sharpens the doctrine on the response side: evacuation is not the default terminal move — it is one option among several, it can itself be the more dangerous one, and for some event types it may not be available at all [4] [5].

What we know

Prepared systems can hold critical mortality down. After the London bombings of 7 July 2005 — the UK's largest mass-casualty event since WWII — 775 casualties and 56 deaths (53 at scene) yielded 20 critically injured patients with critical mortality of 15%, and that mortality was not attributable to resource shortage. Advanced prehospital teams performing initial scene triage reduced over-triage; the receiving centre absorbed a peak surge of 18 seriously injured patients per hour, reached resuscitation capacity within 15 minutes, and used 264 units of blood products in 15 hours [1].

The Boston lineage is the planned-event proof case. Across Boston's trauma centres after the 2013 Marathon bombing, 281 people were injured, 127 were treated at participating trauma centres, there were three immediate scene fatalities (1%) — and no in-hospital mortality. Two-thirds of admitted patients had lower-extremity injuries; of 31 with exsanguinating hemorrhage, 26 arrived with field tourniquets already in place; 54 of 75 admitted needed urgent surgery and 12 (22%) a lower-extremity amputation [2]. The preparedness framing of that response — the event as an exercise the system had, in effect, already run — is the explicit subject of the companion perspective [6].

Surge has a teachable structure. The CO-S-TR model gives incident command a first-hour checklist: command, control, communications, coordination; staff, stuff, space, special (event-specific) considerations; tracking, triage, treatment, transportation. It exists because facility personnel often lack emergency-management experience and need a simple prioritization tool that complements the incident command system [7].

Manchester: exercising is obligatory, and still not sufficient. The 2017 Manchester Arena bombing killed 23 people (including the attacker) and was the deadliest terrorist attack on UK soil since the 2005 London bombings — yet only one of five mass-casualty terrorist attacks in the UK that year. The after-action lessons are structural: multiagency tabletop exercises, mock hospital exercises, and clinician simulation are obligatory preparation; even the best-prepared units face significant challenge because timing and casualty numbers are unpredictable; and formal local and national reviews must feed lessons back into planning [3].

Over-response is a real failure mode. After a roof collapse during a Norwegian military exercise (76 inside, 25 entrapped or buried; 3 deaths, 7 serious injuries), a camp 125 m away was reorganized as a casualty clearing station, and of 417 available personnel plus abundant vehicles and aircraft, only a fraction were used. Firm incident command deliberately prevented an influx of excess resources — the event stayed a mass casualty rather than becoming a disaster [8].

Evacuation is a decision, not a reflex — guidance-level, 2026. The Purple Guide's Contingency & Emergency Planning chapter, whose working group includes an NHS ambulance service and a fire and rescue service, states that it is generally inappropriate to evacuate an event site on a single unconfirmed alert or for an incident that can be isolated, because evacuations "may, potentially, pose a greater risk than the incident itself"; the preferred posture is containment, moving people to areas of relative safety within the site, in a two-stage model of relative then total safety. Where an incident can be localised, it is "almost always best to keep the core event going" — because stopping the show converts a stable, distributed crowd into a simultaneous directed mass, which is the crush mechanism. The governing test: mitigation measures must not create more risk than they remove, and specifically must not increase crowd-crush risk. The chapter supplies vocabulary US doctrine lacks — invacuation (directed movement to interior refuge, distinct from static shelter-in-place), dynamic lockdown, and self-evacuation — and notes UK crowded places are now required to hold such plans [4].

For camping events, evacuation may be structurally impossible. The Guide's Campsites chapter states that at large events where people arrive in significant numbers by public transport, "it may be impossible to close the event and clear the camping area in an emergency. Facilities will have to be brought to the camping areas rather than the people removed to another place of safety." That inverts the default US assumption: the terminal move for a multi-day camping event is shelter-in-place with inbound logistics — water, shelter, power and medical capability moving toward a population that stays put. The obvious shelter, a large marquee, is itself unstable in exactly the wet-and-windy conditions that create the need [5].

Show-stop now has documented doctrine. The Crowd Management chapter sets out a complete show-stop procedure: defined as an agreed response to an immediate threat to life; internal triggers explicitly including crowd collapse and excessive crowd density; a named, identifiable set of persons with authority to initiate; advance written agreement with the performer's management; a designated audience communicator with a competent alternative, because some performers cannot address a crowd in an emergency; performers to remain on stage where safe, as the highest-authority calming voice on site; and a formal restart pathway — where primacy was transferred to a statutory service, control returns to the organiser by a signed Return of Authority form. Failure of the stop/restart procedure is flagged as carrying both safety and legal consequences [9]. Command handover is likewise specified: transfer of primacy to the statutory services on declaration of a major emergency, including partial primacy in which the emergency services own the cordon while the organiser retains control — and medical operations — everywhere else [4].

The planned-event advantage

The corpus consistently locates disaster causation in absent preparation: across 156 mass-gathering crowd disasters (1971–2011), the recurring mix was high crowd density, restricted access, poor fire safety, minimal crowd control, and lack of onsite medical care [10]. A planned event inverts every element of that list before gates open. The Toronto "half-a-million strong" concert response shows the ceiling of pre-positioning: a 124-bed on-site medical facility including a field hospital, with tiered assets from bicycle paramedics to physicians, absorbing 1,870 patient contacts (42/10,000) from more than 450,000 attendees [11]. Boston demonstrates the same logic at the system level — short transport times, immediate surgical access, and field hemorrhage control already in hand [2]. Doctrine now institutionalizes the posture: NFPA 3000 exists as a consensus standard for active-shooter/hostile-event response programs [12], Stop the Bleed / Hartford Consensus establishes public hemorrhage-control doctrine [13], FEMA's IS-15 provides special-events contingency planning curriculum [14], and NATA positions establish venue emergency action planning in sport [15] [16].

What's contested or fragile

Triage-method evidence is thin. There is no single validated national mass-casualty triage guideline in the US; variability in processes, tags, and nomenclature risks confusion and miscommunication across jurisdictions. The Model Uniform Core Criteria (general considerations, global sorting, lifesaving interventions, individual assessment) were built by expert consensus to impose interoperability — a standardization project, not an outcomes-validated algorithm [17]. Exercising fidelity is asserted, not measured. Manchester makes exercising obligatory doctrine, but the same source stresses that even well-prepared units are stressed by real events — the corpus contains no measure of how faithfully tabletop performance predicts real performance [3]. Casualty-flow assumptions are fragile. London's 18-serious-patients-per-hour peak, with resuscitation capacity saturated in 15 minutes [1], warns against planning on smooth arrival curves; Romundstad warns from the opposite direction that uncontrolled responder inflow is itself a hazard [8]. The Purple Guide contradicts itself on greenfield evacuability — and the contradiction is instructive. Its Contingency chapter holds that a greenfield event, where smoke and fire disperse, presents lower risks and easier evacuation than an enclosed venue; its Campsites chapter holds that a transport-fed camping event may be impossible to evacuate at all. Both chapters were published on the same day and the conflict is unreconciled anywhere in the publication. The resolution this synthesis adopts: the Contingency claim is about hazard propagation and holds for single-day drive-in greenfield events; the Campsites claim is about egress transport capacity and receiving destinations, and governs multi-day camping events. Evacuation difficulty is set by where the crowd can go, not by how fast the hazard spreads [4] [5].

What we don't know

  • Venue-level surge conversion. No entry in this corpus quantifies how, or how fast, an event medical footprint converts into an MCI treatment area — bed-conversion times, staffing-flip ratios, supply depth. Feldman describes capacity; nothing tests its conversion under attack conditions.
  • Show-stop outcome science. The procedural gap has closed — who stops the show, on what triggers, and how it restarts are now documented guidance [9] — but the outcomes question remains open: no study measures whether, when, or how stopping an event changes casualty counts. Note also a governance gap inside the documented procedure itself: the listed show-stop personnel do not include the event medical lead, although two of the six internal triggers (crowd collapse, excessive crowd density) are conditions the medical team detects first, as clustered presentations from a sector the front-of-stage spotters cannot see. A defined presentation threshold routed to event control is a cheap, testable fix.
  • The extraction backlog, named honestly. Several domain entries — including MIMMS [18], WHO's key considerations [19], and the spontaneous-gathering literature [20] — lack extracted summaries and can be cited here only for their existence. The Purple Guide chapter entries cited above are chapter-level extractions read 2026-08-13/15; portions of each chapter remain unread and are listed in their entry notes.
  • Crowd-crush clinical knowledge remains inadequate. Even for Itaewon, a landmark event, causes of death and injury spectra are poorly identified; traumatic asphyxiation is widely implicated as the primary cause of death, but analysis is hampered by restricted official reports and incomplete documentation [21].

How MGMI operationalizes this

The registry treats these events as the case-seminar corpus — Manchester and Itaewon are registry-case candidates alongside London, Boston, and the Norwegian roof collapse. Module 7 (MCI posture) draws its question bank from the summarized findings above. The MAP composer's surge sections implement CO-S-TR as the structuring skeleton [7] and require a named show-stop authority — a requirement that now has a guidance-level template rather than a bare assertion [9]. Camping-event plans composed for multi-day, transport-fed profiles should carry an explicit shelter-in-place-with-inbound-logistics contingency rather than an evacuation default [5]. Shelf anchors: NFPA 3000 [12], Stop the Bleed [13], FEMA IS-15 [14].

Reading pathway

Start with Biddinger's framing [6], then the two outcome anchors [1] [2]; add Manchester for the exercising mandate [3], Romundstad for command discipline [8], Hick for the surge checklist [7], and Lerner for triage-standardization limits [17]. For response doctrine, the Purple Guide's Contingency chapter for the containment-over-evacuation posture and Crowd Management for the show-stop procedure [4] [9]. Chang closes with what we still cannot see [21].

Citations

  • [1] Aylwin CJ. Reduction in critical mortality in urban mass casualty incidents: analysis of triage, surge, and resource use after the London bombings on July 7, 2005. The Lancet, 2006.
  • [6] Biddinger PD. Be Prepared — The Boston Marathon and Mass-Casualty Events. NEJM, 2013.
  • [2] Gates JD. The Initial Response to the Boston Marathon Bombing. Annals of Surgery, 2014.
  • [3] Craigie RJ, et al. Manchester Arena bombing: lessons learnt from a mass casualty incident. BMJ Military Health, 2018.
  • [21] Chang AH, et al. A Crowd Disaster Study: The Itaewon Seoul Crush. Cureus, 2024.
  • [7] Hick JL. Surge Capacity Concepts for Health Care Facilities: The CO-S-TR Model. Disaster Med Public Health Prep, 2008.
  • [17] Lerner EB. Mass Casualty Triage: An Evaluation of the Science and Refinement of a National Guideline. Disaster Med Public Health Prep, 2011.
  • [8] Romundstad L, et al. Challenges of Major Incident Management When Excess Resources are Allocated. Prehosp Disaster Med, 2004.
  • [10] Soomaroo L, Murray V. Disasters at Mass Gatherings: Lessons from History. PLoS Currents, 2012.
  • [11] Feldman MJ, et al. Half-a-Million Strong: The EMS Response to a Single-Day, Mass-Gathering Event. Prehosp Disaster Med, 2004.
  • [22] Helbing D, Mukerji P. Crowd Disasters as Systemic Failures: Analysis of the Love Parade Disaster. SSRN, 2012.
  • [23] Taylor LJ. The Hillsborough Stadium Disaster: Final Report. Home Office (UK), 1990.
  • [18] Advanced Life Support Group. Major Incident Medical Management and Support (MIMMS), 2011.
  • [19] WHO. Public Health for Mass Gatherings: Key Considerations, 2015.
  • [20] Hawkins ER, Brice JH. Fire Jumpers: Burns and Traumatic Injuries from a Spontaneous Mass Gathering and Celebratory Riot. J Emerg Med, 2010.
  • [12] NFPA 3000 Standard for an Active Shooter/Hostile Event Response (ASHER) Program, 2024.
  • [13] STOP THE BLEED / Hartford Consensus I–IV Compendium, 2016.
  • [14] FEMA. Special Events Contingency Planning (IS-15).
  • [15] Emergency Planning in Athletics (NATA position statement), 2002.
  • [16] Emergency Action Plan Development and Implementation in Sport (NATA position statement), 2024.
  • [4] Events Industry Forum. The Purple Guide — Contingency & Emergency Planning, 2026 (chapter published 2026-01-26; 13 of 27 sections extracted).
  • [5] Events Industry Forum. The Purple Guide — Campsites, 2026 (chapter published 2026-01-26; complete chapter extracted).
  • [9] Events Industry Forum. The Purple Guide — Crowd Management, 2026 (chapter published 2026-01-26; 10 of 30 sections extracted; working group includes G. Keith Still).