Humanitarian Emergency Medicine Training

Clinicians preparing for a first humanitarian deployment usually prepare for the wrong thing. They revise tropical medicine, paediatrics and trauma — all reasonable — and then find that the difficulty was never the diagnosis. It was working through an interpreter, deciding how to allocate three ventilators, and functioning in a team that formed four days ago.

None of that means clinical preparation is unnecessary. It means the preparation should be weighted differently from how it feels intuitive to weight it.

If you want to rehearse resource-limited decisions, run a low-resource case in the app.

The clinical adjustments that matter

Four changes to how you practise, rather than four new bodies of knowledge.

History and examination carry the diagnostic load. With limited or no laboratory and imaging, the clinical assessment stops being a screening step and becomes the investigation. In practice this means slowing down at the bedside, examining more thoroughly than hospital habits encourage, and being explicit with yourself about what you are inferring rather than measuring.

You will treat on probability. Empirical treatment based on the clinical picture and local epidemiology, without confirmation, is normal practice rather than a compromise. That requires knowing the local disease pattern — which your organisation will brief and which you should read before arrival — and being comfortable committing without a result.

Your formulary is a list, not a pharmacy. Work from the actual list you will hold, not from the drugs you are used to prescribing. Reading through an essential medicines list and noting what you would do without each familiar agent is a genuinely useful evening's preparation.

Bedside diagnostics become disproportionately valuable. Ultrasound, a glucometer and a means of measuring haemoglobin change management far more than they do in a well-equipped hospital, because they are the only objective data available. If you can extend your point-of-care ultrasound competence before deployment, it has a high return.

The non-clinical skills that dominate

Most people report that these determined how effective they were, and almost none of them are taught in clinical training.

• Working through an interpreter. A specific skill. Speak to the patient rather than the interpreter, use short complete sentences, avoid idiom, check understanding by asking the patient to explain back rather than to confirm. Also worth knowing: your interpreter is frequently from the affected community and is carrying the same event you are.
• Working within a coordination structure. Responses are coordinated, and freelancing — however clinically well-intentioned — damages the wider effort. Understand the structure you are joining, your organisation's place in it, and the reporting you are expected to do.
• Handover and documentation with no continuity. You may never see the patient again, and the next clinician may not speak your language. Notes that are legible, structured and comprehensible to someone with a different training background are a clinical intervention, not administration.
• Teamwork with people you met on Tuesday. Teams form fast, from multiple organisations and nationalities, with different assumptions about hierarchy and about who may question whom. Establishing explicitly how your team will raise concerns is worth doing in the first meeting, not the first crisis.
• Cultural and religious context around illness and death. Consent, examination, family involvement and the handling of a body all vary, and getting them wrong causes real harm to both the family and your team's ability to work. Ask, do not assume, and ask before you need to know.
• Security awareness. Your organisation's security briefing is not a formality. Access, curfews and movement rules are clinical constraints, because they determine who you can reach and when.

The decisions nobody rehearses

These are the parts of humanitarian work that people describe years later. Thinking them through in advance does not make them easy, but it makes them survivable.

Allocating a resource that is genuinely insufficient. Not triage in the familiar sense of ordering treatment, but deciding who receives a limited resource when others will not. The frameworks for this exist and your organisation will have a position; know it before you need it, apply it consistently, and document the reasoning. The thing that damages people afterwards is usually not the decision but the sense of having made it arbitrarily.

Withholding treatment you know how to give. A patient with a condition you could manage well at home, and cannot manage here. This is the specific ethical injury of humanitarian work, and it is not solved by anything except recognising it in advance as an expected feature rather than a personal failure.

Working at the limits of your competence. You will be the most qualified person present for problems you are not trained in. The reasonable approach is the same one that applies anywhere: do what is within your competence, seek remote advice, be honest with the patient and the team about the limits, and document what you decided and why.

Deciding when to stop. Your own capacity is a resource with a finite limit, and a clinician working beyond it becomes a hazard. Deciding in advance what your indicators are — and telling a colleague what they are — is far more reliable than judging it in the moment.

A preparation plan for the months before

Ordered by return on effort rather than by how interesting it is.

1. Read your organisation's clinical protocols and the standards it works to. Not the general literature — the actual documents you will be held to. This is the highest-value thing on the list and the one most often skipped.
2. Learn the epidemiology of where you are going. The common presentations, the seasonal patterns, and what the local health system can and cannot do. Your differential should be re-weighted before you arrive.
3. Practise your existing medicine with the resources removed. Run cases with no imaging, no laboratory, and a restricted drug list. This is the closest available simulation of the actual cognitive task, and it is more useful than acquiring new specialist knowledge.
4. Extend bedside skills that do not depend on infrastructure. Ultrasound, airway management, wound and fracture management, obstetric emergencies. Skills that work without a building behind them.
5. Sort your own affairs and your own health. Vaccinations, prophylaxis, dental review, medication supply, insurance, and the practical arrangements at home. Being distracted by an unresolved problem at home is a common and avoidable performance drag.
6. Set up your offline reference material before you travel. Anything that needs installation, download or account setup should be done while you have reliable internet. Core training and reference tools are designed to work offline after setup, but installation, updates and some account services need a connection — so the setup is a pre-departure task, not an on-arrival one.
7. Agree with someone at home how you will stay in contact. And what they should do if you go quiet for longer than expected.

Preparing for afterwards

The part of preparation that gets least attention and matters over the longest timeframe. Returning is frequently harder than deploying: the work stops abruptly, the people around you have no frame of reference for it, and the contrast between what you have just seen and ordinary life is jarring.

Three things worth arranging in advance. Know what psychological support your organisation offers and how to access it, before you need it rather than after. Arrange some deliberate time before returning to clinical work at home. And identify one person who has done similar work and will talk to you about it, because the most consistently reported thing that helps is being understood by someone who does not need it explained.

None of that is a sign of fragility. It is the same anticipatory planning that the rest of this article recommends, applied to the predictable consequences of the work.