Mass casualty and pandemic disaster management

While chapter is vaguely relevant to Section 2.1.1 of the CICM Second Part General Exam Syllabus (Second Edition), which identifies “pandemic and major incident planning and response” as core knowledge for the intensive care specialist, the only SAQ of the Intelligent Assessment Design Era was blueprinted to the "Leader and Manager" syllabus topic, Section 2.3.1 ("Intensive Care Administration"). That was Question 12 from the first paper of 2024, where the candidates were invited to describe their plan for digging in and defending against a natural disaster, as opposed to evacuating. Prior to this, Question 19 from the second paper of 2013 was the only SAQ to ever ask about an external disaster response, and was mostly interested in the facility-level response to a large scale influenza pandemic - an entirely different set of challenges.

The pandemic question could be easily answered by any person who  was even peripherally aware of the COVID pandemic, and it is not essential to be intimately familiar with the summary statement of the Task Force for Mass Critical Care of the American College of Chest Physicians, as well as the recommendations of the European Society of Intensive Care Medicine Task Force for Intensive Care Unit Triage during an Influenza Epidemic or Mass Disaster. We have all become inadvertently expert at pandemic management. 

The answer suggested in the Discussion section of Question 19  incorporates the recommendations of the abovementioned bodies, as well as the suggestions from the college answer. There is also the LITFL article on this topic which is heavily exam-oriented. The LITFL article in turn takes its inspiration from this 2006 MJA article, as well as published experiences from the recent H1N1 pandemic. Beyond pandemics, other historical disasters we have learned from have included...

• Madrid bombing (de Ceballos et al, 2004)
• SARS epidemic (Hawryluck et al, 2005)z
• Tokyo subway sarin gas attack (Okumura et al, 2005)
• London subway bombing (Shirley and Mandersloot, 2008)
• Almost any day in Israel (Aschkenasy-Steuer et al, 2005)

Non-LITFL resources for this sort of thing include:

• Mass Casualty Incidents, a book by M.Lynn (2016)
• Briggs (2016) - "Disaster Management and Preparedness." 
• Davis et al (2005) - "Hospital bed surge capacity in the event of a mass-casualty incident." 
• Avidan et al (2007) - "Civilian hospital response to a mass casualty event: the role of the intensive care unit." 
• Rosenfeld et al (2005) - a comment on Australian terrorism response preparedness
• Lynn et al (2006)  - "Ten commenadments for hospital planning"

The various writings by Mauricio Lynn are particularly good. Lynn is ex-military, a retired Lieutenant Colonel from the Israeli Military where he served for 18 years (some of that time as Chief of Trauma). His disaster cred comes from having been involved in the Narobi bombing (American embassy) and the 1999 earthquake in Turkey. Moreover, the Israeli experience of mass casualties is formidable. Aschkenasy-Steuer et al (2005) were able to compile a retrospective review spanning four years, during which a nightmarish 33 mass casualty events occurred, totalling 541 victims (twenty of the events involved more than ten wounded).

Challenges of the mass casualty disaster incident

What the hell is a "mass casualty event"? Well, the numerical cut-off is ten simultaneous cases. This is a totally arbitrary number.

In summary, in a mass casualty disaster incident:

• There is not enough hospital beds
• There is not enough ICU beds
• There is not enough ICU-trained staff
• ICU beds and operating theatres will not be immediately available to arriving casualties

Thus:

• Critical care staff will need to provide care outside of the ICU
• Non-critical care staff will need to provide basic ICU-level care
• ICU beds will need to be created
• ICU staff numbers will need to be expanded by the use of "creative rostering" and certain human rights abuses such as recall from leave and the use of retired or seconded staff.

The effectiveness of disaster response is usually limited by bed capacity as the primary problem (you can stretch existing staff and make them work harder, but beds are much less elastic). The ability to adequately expand bed numbers ("surge capacity") depends on several factors:

• Number of critically ill casualties
• Type of disaster (influenza pandemic with its isolation requirements will be very different to a mass trauma scenario with its increase operating theatre demand)
• Duration of the casualty-generating event (i.e. it is easier to deal with a single train wreck than with a month-long Ebola outbreak)
• Infrastructure availability (i.e. is the hospital large enough, or are we talking about erecting MASH-style tents?)
• Quantity and quality of critical care required (i.e. of the casualties, will everybody need blood products like in the case of massive blast trauma, or will everybody end up on ECMO like the swine flu epidemic?)

Disaster management planning specific to the ICU

Specific issues need to be addressed as a part of the disaster management plan:

• Roles of disaster response staff need to be clearly defined
• Communication is usually a big issue; several needs must be addressed (as discussed below)
• The rationing of ICU equipment needs to be carefully protocolised so that there is no disagreement 
• Transfer agreements with other ICUs need to be in place
• If the level of ICU care must be degraded because demand exceeds capacity,  a plan must exist to ensure that this occurs decrementally with defined priorities (as opposed to haphazardly).
• Backup equipment must be available, and inventory needs to be regularly updated
• Backup consumables stores need to be available, and regularly checked (i.e. did all our excess central lines expire in 2010?)  
• Backup infrastructure needs to be organised (including power and gas supplies).
• The plan needs to provide for a scenario where decontamination or total isolation becomes necessary, eg. Ebola or some sort of chemical or radioactive contaminant. Secondary exposure can rapidly multiply your casualties, eg. as in the classical case of a zombie apocalypse. As another example, during the Tokyo sarin gas attack about 23% of hospital staff (and 37% of ICU staff!) suffered some degree of secondary sarin exposure (Okumura et al, 2005). Interestingly, much of the exposure occurred as the victims were undressed in the ED - the attack occurred during winter, and warm winter clothes trapped the gas - releasing it into the faces of rescuers when the casualties arrived to the ED. Fortunately, the Aum Shinrikyo cult were useless at chemistry and only released a low-concentration gas. Anyway: mass decontamination facilities need to be made available. For victims in the ICU who continue to issue forth clouds of poisonous gas, "measures such as attaching a reservoir to the air outlet and emptying the reservoir by continuous suction should be implemented".

The roles of the ICU disaster director: 

So, you're the director of critical care services during a disaster. And .... you get a disaster. What do you do?
This is a suggested "action card" from Shirley et al (2008). It's so good that I reproduced it with zero modification, as a potential future cut-and-paste answer to a CICM SAQ.

• Initial actions of the ICU disaster manager
  • Make contact with the incident medical director (emergency department consultant).
  • Check details of the incident.
  • Assess the need for ICU beds and the timeframe.
  • Liase with the senior ICU nurse and identify potential increases in capacity.
  • Liase with the senior anaesthetist and senior surgeon.
  • Consider the level of response required and identify staffing.
  • Establish the need for satellite ICU/high-dependency unit beds or other beds.
  • Ensure a sufficient number of runners to relay information on patient movements.
  • Establish an ICU control room with updates.
• Clinical care
  • Identify current patients suitable for transfer.
  • Designate senior ICU medical representation for the resuscitation room.
  • Identify likely ICU patients and their dispersal (theatres, computed tomography, or direct admissions).
  • Ensure that care is not compromised in existing non-incident ICU patients.
  • In the event that resources are overwhelmed, liase with other senior clinicians about the diverting of patients.
• Post-incident
  • Identify pitfalls in planning early and address them at the earliest opportunity.
  • Ensure the ongoing welfare and support of staff, patients, and relatives.

Disaster workforce

In preparation for mass casualty incidents, a rigid hierarchical structure needs to be erected to act as the emergency scaffold for all further intensive care staffing questions.

• Administrative roles for disaster incidents need to be assigned in advance, and the protocol for this needs to be widely circulated so that - when it happens - everybody acknowledges the chain of command.
• Specific roles need to be assigned:
  • Director of operations needs to be assigned; they take control of the scenario.
  • A media liason needs to be identified, so that they can take on the role of public relations controller (you can't have journalists swarming though the hospital). In Toronto, the SARS outbreak received massive media attention. "Daily headlines generated widespread fear and panic", complain Hawryluck et al (2005).  Nominate one person: that is the most sensible and well-spoken member of staff who will provide the media with one consistent picture throughout the process.
  • A hospital liason officer should be nominated, who will do the organisational work of transferring patients to other ICUs to make room.
  • Somebody should be the "documentation monkey". In the London bombings this was a role assigned to one specific ICU medical staff member, who was sat down at a computer terminal to provide a constant running update on the injuries and medical status of the admitted patients. 
• Inter-depeartment cooperation is expected. In London, ICU physicians (three of them) invaded the ED and helped with the initial resuscitation. 
• Some sort of system needs to be in place to ensure the wellbeing of the extra ICU staff who are summoned ot the event. In the London bombing scenario, a local supermarket provided free food to the ICU that afternoon - but one cannot depend on random acts of charity for sustaining the caffeine needs of a beyond-capacity intensive care unit. Some sort of standing arrangement must be organised as a part of the disaster management plan.  

Disaster communication

• Up-to-date contact details must be available from the staff
• Liasons for the families and media need to be identified in advance
• Back-up communications systems need to be in place. During the London bombing, the Royal London Hospital's internal phone system became inundated with calls and had become inoperable (a sort of analog distributed denial of service attack). At the same time, the authorities had shut down the mobile networks to facilitate emergency service communication. Most of the routine communication within the hospital and ICU was handled by runners and internal email. 
• Inter-department communication needs to be rapid and effective. The haematology department at the RLH was somehow the last to find out about the London bombing, which was not cool. Those guys need to get ready with blood products, and you simply cannot rush the defrosting of frozen plasma, etc.

Diversion and triage

How to conserve precious ICU resources? Ensure the patients being sent there are genuinely ICU-ish.

• Pre-hospital triage services will be helpful in ensuring that the majority of the minor casualties will be sent to lower-level hospitals, so that major trauma centres deal predominantly with major trauma.
• Without this, hospitals closes to the incident are rapidly overwhelmed. For example, during the Madrid bombings de Ceballos et al (2005) comment that there was inefficient overtriage to the nearest hospitals. 
• There is usually a large volume of highly distressed but otherwise only mildly affected patients; for instance in the London bombing 194 casualties from the scene of the incident were seen at The Royal London Hospital, but only 27 were admitted - of whom only 7 ended up in the ICU being ventilated. Similarly, in Madrid only 12% of the total casualty numbers were deemed "critically ill". 

If the capacity of the ICU is exceeded even after non-major cases are turned away, there needs ro be a well-established protocol for diversion of patients to other intensive care units, which might mean moving stable medical ICU patients by road to the next nearest low-level ICU. This pattern of diversion needs to be well-established in advance of the disaster, i.e. there needs to be a mandatory acceptance to streamline this process (you can't spend all your time on the phone looking for an accepting ICU because everybody keeps refusing).

Historically:

• Non-incident-related casualties tend to divert spontaneously (i.e. people with stubbed toes sensibly avoid major hospitals when there is a huge explosion incident). 
• In urban disasters, many of the injured make their own way to hospital, often arriving before the more seriously injured casualties. 
• Diversion was particularly successful in the London bombing where a senior surgeon was in charge of triage at the doors, assigning teams of anaesthetists and surgeons to each casualty which merited urgent surgery. 
• One should also anticipate a second wave of wounded referred from smaller hospitals ("reverse diversion")

How to create more ICU beds in long-lasting disasters

This concerns the diversion of resources of other departments to the ICU during prolonged emergencies, such as the need to ventilate numerous respiratory failure patients during the H1N1 pandemic. In this sort of situation, you need more ventilators and more ICU beds.

• Cancel elective surgery
• Transfer non-influenza patients to private and peripheral ICUs
• Transfer stable patients to high dependency beds
• Urgently open extra ICU beds which are not funded:
  • Annex the recovery wards and CCU
  • Convert HDU beds to ICU
• Ration the use of mechanical ventilation for patients genuinely dependent on it, freeing up equipment
• Appropriate anaesthetic machines from OT (those guys got all their elective cases cancelled, so they aren't going to be using them)
• Borrow transport ventilators from ED and retrieval services
• Contact the vendor companies and negotiate emergency equipment hire.

How to summon more ICU staff

The major resource problem during the H1N1 pandemic in Australia was actually the availability of ICU-trained nurses. How to create greater nursing and medical workforce numbers?

• Cancel medical and nursing staff scheduled leave
• Increase working hours for medical and nursing staff
• Bring seconded staff back from secondment
• Increase proportion of casual nursing staff
• Appeal to locum agencies
• Train non-ICU staff to care for ICU-level patients
• Administer antiviral prophylaxis to decrease absenteeism

On the basis of their massive Israeli experience, Aschkenasy-Steuer et al (2005) also recommend the following:

• An up-to-date list of all staff members, permanently posted in a prominent place, is crucial for efficient personnel recruitment
• Staff are called according to residential distance rather than professional status

Sheltering in the ICU during a natural disaster

This fresh new hell was offered to the candidates as their problem in Question 12 from the first paper of 2024, which suggests that the end stage CICM trainee should prepare to become the emergency disaster coordinator for scenarios where evacuation is impossible, even thought they have never previously held a position of such responsibility. This is in fact a plausible scenario for VMO and locum life, where a mercenary intensivist may find themselves in charge of a unit under these conditions, even though they may have literally graduated 15 minutes ago. The appearance of this in the exam papers makes one wonder whether somebody among the examiners had a personal experience of this nature. It would also be logical for this experience to never see the light of publication, as things like this tend to be politically sensitive and it would do nobody's career any favours to report the incompetence of their local head of department or hospital executive to the scientific community. Instead we must construct this from basic principles.

"Shelter in place" is the sum total of the detail in the hospital disaster preparedness plan presented in Question 12. That somebody scored as highly as 6.5 for this answer is mindboggling, and that trainee should be immediately invited to chair the committee to prepare a better plan. The evacuation of an entire ICU to another ICU across the country is a colossal pain in the arse, and this is not lost on most people who have ever had to deal even with just the OCD-triggering kabelsalat of lines and tubes that occurs during a routine CT transfer. The task becomes insurmountable when the ICU is large, the patients are unstable, or the distance as immense as (for example) Far North Queensland. Moreover it is likely that the rest of the hospital is not getting evacuated, which means that theoretically somebody could get sick during the period of the disaster, and require ICU-level care (so, you will be ICU-ing during the storm, irrespective of whether or not you have managed to move all your patients).  Borrowing from the US experience of Hurricane Sandy, via McGinty et al (2017), the decisions regarding evacuation versus a heroic stay-and-hide approach were mostly motivated by:

• Risk to patient health from evacuation, related to their acuity
• Prior experience
• Cost
• Ability to maintain continuity of operations
  • Planned utility outages
  • Anticipated flooding
  • Integrity of power water and gas supplies
• Timing (i.e. no time to evacuate, the disaster is here)

After all these things are taken into consideration, often the decision to stay and shelter is fairly clear. Evacuation of a hospital or an ICU is rare and is mostly triggered by the sudden loss of primary and backup power or gas supply, which forces one's hand. To shelter in place is otherwise the default setting (Bagaria et al, 2009). From that, one might expect that there should be some kind of clear and widely known strategy for preparing an ICU for this ordeal, but in fact that is not the case, or at least if it exists then it is buried deep in the corporate document pile of your local health district.

Staffing

• Allocate roles for specific aspects of the preparation, eg. "roster person", "family liaison", "SITREP monitor", "pathology/blood bank runner", "electronic medical record downtime record keeper"
• Assess the disaster staffing. Ask staff members about their plans. " Most staff found the decision difficult, balancing the needs of their own families with that of the community", Sexton et al remark about their doctor's and nurses response to Cyclone Yasi, preparing to work double shifts as the Queensland government were telling them to grab their loved ones and run.
• Allow for storage space and personal space for staff who are likely to be cut off from their homes by weather/water
• Source additional meals for trapped staff

Internal communication

• Develop an internal mechanism of distributing information to the staff- ideally not reliant on landline phone services 

Equipment

• Charge removable power sources and stock reserves of bottled water for staff and patients
• Create paper copies of electronic medical records and reference texts (eg. pharmacy manuals) for the eventuality of power failure
• Create a reserve of torches and batteries

Consumables

• Ensure pharmacy is well stocked and estimate several weeks of stock for common drugs, expecting prolonged disruption to supply chains and increased demand
• Share a pool with the other remaining services (ED, ward) to centralise ordering
• Liaise with gas suppliers and acquire excess stock of oxygen and compressed air cylinders

Infrastructure

• If possible, relocate vital systems (pathology department analysers, ICU, backup generators ) out of easily flooded basements and to higher floors
• Move patient beds away from windows

Patient care

• Establish a patient manifest: decide who needs to stay and who can be evacuated
• Send out anyone who can be retrieved safely, to decrease the workload of the remaining staff

Families and visitors

• Establish the identities of all next of kin and secure contact details
• Discorage visiting
• Determine which family plan to evacuate and which plan to remain local

External communication

• Disaster-proof devices (emergency landlines, satellite phones, radio) need to be located checked and distributed to key staff
• Maintain open lines (avoid unnecessary phone calls on critical protected phones)

Facility level support

• Recommend for the closure of non-emergency theatres and cancellation of elective surgery
• Recommend for the retrieval of any patient for whom it would be safe and appropriate
• Listen for communication from central disaster control 

Media and public engagement

• Observe for updates from media sources regarding the progress of the disaster and the prospects of improvement or new challenges (eg. access limitations due to rising water)

Debrief

• Engage with the post-disaster debrief with staff and hospital leadership 
• Ensure counselling services and leave/rest are available for staff who were imposed upon during the period of upheaval.

Preparedness

• Simulated tabletop exercises involving other departments
• Internal position of disaster preparation officer, to maintain documents and revise plans as infrastructure changes