Pandemic preparedness: what clinicians see first

In the same week, a physician at a general hospital admits three adults with pneumonia. None of them has any significant prior illness, all have high fever, and the chest X-ray shows changes greater than the physical exam would have suggested. Tests for the most common causes come back negative. Talking with the team, the physician learns that two of the patients work at the same place and that the third is a relative of one of them. The temptation is to think it is the season and move on. The scene is fictional and was written for teaching. This text uses that moment as a setting for clinical reasoning: what the clinician sees first when a new disease begins to circulate, when to be suspicious, when to notify, and which biases delay that perception.

This text is not a surveillance or infection control protocol. Each country and each service has its own rules for notification, isolation and sample collection. The interest here is reasoning: how to notice that a group of patients forms a pattern that none of them, alone, reveals. If you play the case of the day on NEXO, you will recognize the move: sometimes the clue that changes everything is not in the patient, but in the context.

Key points

• Many new health threats were first noticed by clinicians who spotted something outside the norm.
• An unusual cluster of patients, linked by place, work, travel or contact, is a clue that no single patient brings on their own.
• Asking about travel, work and exposure is part of the history of any unexplained febrile illness.
• The International Health Regulations provide for reporting events that may represent a risk, before the agent is confirmed.
• Normalization and "it's seasonal" are among the biases that most delay detection.

Why pandemic preparedness is a reasoning setting

Pandemic preparedness often sounds like a matter for governments and international agencies. But the first link in the chain is almost always a medical visit. The Independent Panel for Pandemic Preparedness and Response, in its review of the COVID-19 pandemic, highlighted the diligence of clinicians who noticed clusters of unusual pneumonias, sent samples and pushed their concerns forward (Independent Panel, 2021).

A pattern that no single patient brings on their own

Each patient in the scene, in isolation, has a pneumonia that could have many causes. What draws attention is the whole: people without prior illness, with a similar picture, linked to one another, in the same week. This kind of clue requires looking beyond the bedside. It only appears when someone brings together information from several visits and asks whether it really is a coincidence. The post on signs, symptoms and history shows how a patient's history includes the place and the people they live with.

What the scene teaches

In the opening scene, the physician has two possible explanations: a coincidence of common pneumonias or something new. The pressure of the service pushes toward the first. Clinical reasoning asks the physician to keep both open until there is information to choose, and to take the protective measures the second would require in the meantime.

What recent history teaches

Sachs and colleagues, on the Lancet Commission on the COVID-19 pandemic, reported 6.9 million notified deaths and estimated 17.2 million deaths up to May 31, 2022 (Sachs and colleagues, 2022). Among the failures identified by the commission are the lack of timely notification of the initial outbreak and the delay in recognizing airborne transmission.

A lost month

The Independent Panel described February 2020 as a lost month, in which measures could and should have been taken to contain the epidemic (Independent Panel, 2021). The same report defends the precautionary principle: faced with a new respiratory pathogen, presume there is transmission between people unless the data show otherwise.

Early recognition makes a difference

The same report notes that, in some countries, early recognition of the threat and a rapid response kept the epidemic small, and concludes that the alert system did not work with the necessary speed in the face of a fast-spreading respiratory virus (Independent Panel, 2021). For those on the front line, the message is that the time between noticing and communicating counts.

The lesson for the clinician

These analyses talk about systems, but the lesson applies to every consultation. Waiting for certainty before acting costs time, and time is the scarcest resource at the start of an outbreak. The post about diagnostic uncertainty shows how to decide well when information is still incomplete.

Sentinel cases: how new threats were first seen

The history of recent emerging diseases is, to a large extent, the history of clinicians and laboratories that noticed something different. The examples below are told in summary form, focusing on what was noticed, and not on who made a mistake.

SARS, 2003

Peiris and colleagues described 50 patients with severe acute respiratory syndrome and observed that the respiratory symptoms and auscultation findings were disproportionately mild compared with the changes on chest radiography (Peiris and colleagues, 2003). In the same work, the authors isolated a coronavirus and found evidence of infection by it in 45 of the 50 patients. In another country, the identification of the disease in Canada was described by Poutanen and colleagues.

MERS, 2012

The description of a new coronavirus isolated from a man with pneumonia, published by Zaki and colleagues, was the starting point for the recognition of Middle East respiratory syndrome. Assiri and colleagues then described 47 people with confirmed disease, of whom 28 died and 45 had pre-existing conditions (Assiri and colleagues, 2013).

COVID-19, 2020

Huang and colleagues described the first 41 hospitalized patients with the new disease: 27 of them had been exposed to the same market, and 40 had fever at the onset of the illness (Huang and colleagues, 2020). The same study found a family cluster and pneumonia with changes on CT in all patients.

Mpox, 2022

In 2022, a series of patients with mpox infection in 16 countries, described by Thornhill and colleagues, showed a pattern of presentation different from the one described in the areas where the disease was traditionally found. The new presentation required clinicians in many places to recognize a disease they did not expect to see.

What the sentinel cases have in common

In all four examples, the new disease did not arrive with a label. It appeared as pneumonia, fever or skin lesions, conditions that clinicians see every day. What made the difference was someone noticing a detail that did not fit: a radiograph worse than the examination, a group of patients linked to the same place, a presentation different from the one described in textbooks. In all of them, clinical observation came before laboratory confirmation, and it was this that led to ordering the right tests.

The unusual cluster

The unusual cluster is the most characteristic clue at the start of an outbreak. It can appear in various forms:

• Several patients with a similar condition, in a short interval, in the same service.
• Patients linked by a place, a job, an event, a trip or close contact.
• A severe condition in people who do not usually fall ill that way, such as young and healthy adults.
• Health professionals falling ill after treating a patient.
• A condition that does not respond to the usual treatment and whose routine tests come back negative.

Seeing the whole picture

The problem is that the clinician rarely sees the whole picture. Patients arrive on different shifts, are seen by different people and sometimes go to different services. Talking with the team, asking whether anyone has seen something similar and looking at the service records are ways of bringing together clues that are scattered.

Keeping the clue

When routine tests come back negative in a severe case, the leftover sample may be the only chance to identify the cause later. Talking to the laboratory about the possibility of storing material, and recording the suspicion on the request, keeps the investigation possible even if the hypothesis still seems unlikely.

The base rate changes

In an outbreak, the probability of a rare disease changes from one day to the next. The post on base rate in diagnosis explains how the frequency of a disease in the reference population weighs on the interpretation of clues. When a cluster appears, the reference population is no longer the usual one and becomes the patients linked to that source.

The question about travel and exposure

The travel and exposure history is the simplest tool for early detection. It costs a few minutes and can change the entire hypothesis. Liddell and colleagues, when describing patients with Ebola virus disease treated outside the areas where the outbreak was occurring, point to early diagnosis as one of the factors that contributed to the good outcome of survivors (Liddell and colleagues, 2015).

What to ask

When faced with an unexplained febrile or respiratory picture, some questions help:

1. Have you traveled in recent weeks, where to and what did you do there?
2. Have you had contact with anyone who was ill with a similar picture?
3. Do you work in a health service, laboratory, with animals or in a place with many people?
4. Have you been at an event, workplace or institution where other people became ill?
5. Have you had contact with sick or dead animals?

Recording the answer

The answer must be recorded, including when it is negative. A record such as "denies travel and contact with sick people" allows another professional, days later, to notice a connection that was not clear at the first visit. The post on telemedicine and clinical reasoning shows how a well-taken history supports decisions even when the physical examination is limited.

Notifying without confirmation

The International Health Regulations (2005), from the World Health Organization, set rules for countries to detect, assess and report events that may represent a public health emergency of international concern. Article 7 asks the country to share with WHO the available information about an unexpected or unusual public health event, whatever its origin. Article 8 allows keeping WHO informed and consulting it about events for which there is not yet enough information to complete the decision instrument in Annex 2, which asks, among other points, whether the event is unusual or unexpected. Grein and colleagues described the WHO process for verifying outbreak rumors and turning large volumes of information into reliable data for action (Grein and colleagues, 2000).

Why not wait

Reporting an unusual event does not require knowing which agent it is. It exists precisely so that surveillance services can investigate while the diagnosis is still open. Waiting for laboratory confirmation can mean days of delay, and those days are the most valuable for containing transmission.

The fear of being wrong

Many clinicians hesitate to notify for fear of causing alarm for nothing. The verification process described by Grein and colleagues also exists to separate unfounded rumors from real events (Grein and colleagues, 2000). Reporting an event that later turns out to be common is part of the normal operation of the system. The post on clinical decision-making shows how to weigh the cost of each type of error.

Back to the scene

In the opening scene, the doctor does not know whether this is a coincidence or something new. They do not need to know. It is enough to recognize that three patients linked to each other, with severe illness and negative routine tests, form an unusual event. From there, the path is to talk to the infection control service, notify surveillance according to local rules, ask the laboratory to keep the samples and protect the team while the investigation advances. If everything turns out to be a coincidence, the system will have worked as it should. If not, the days gained can make a difference for many people.

An international agreement

On May 20, 2025, the World Health Assembly adopted the WHO Pandemic Agreement, with the goal of making the world more prepared and fairer in the face of future pandemics. The Lancet commission also advocates prevention based on the One Health approach and early warning systems (Sachs et al., 2022).

Empirical isolation

Isolating a patient before knowing the diagnosis is a decision made under uncertainty. It protects other patients and the team while the investigation advances. The reasoning is similar to that of prophylaxis: the cost of isolating without need is usually lower than the cost of not isolating when it was necessary. The CDC and HICPAC isolation precautions guideline explains that, since the infectious agent is often unknown at admission, transmission-based precautions are used empirically, according to the clinical syndrome and likely agents, and then adjusted when the pathogen is identified or when a transmissible infectious cause is ruled out (Siegel et al., 2007).

The precautionary principle at the bedside

The Independent Panel argues that, in the face of a new pathogen, person-to-person transmission should be presumed until proven otherwise (Independent Panel, 2021). At the bedside, this translates into protective measures proportional to the suspicion, reviewed as test results arrive.

Explaining the measure

Empirical isolation can frighten the patient and family, and can generate resistance in the team when the suspicion seems remote. Explaining that this is a precautionary measure, taken while the investigation advances, helps reduce fear and maintain cooperation. The same conversation serves to ask the patient to recount who has been with them in recent days, which can help surveillance find other exposed people.

Reviewing the decision

Empirical isolation is not a definitive decision. When the diagnosis of a common cause is confirmed, or when the investigation rules out the hypothesis of a new disease, the measure can be suspended. The post on diagnostic pause shows how short review moments protect against decisions that stay on autopilot.

Warning signs: a table for the front line

The table below summarizes, in a didactic way, clues that may suggest an unusual event and what they ask of the clinician. It does not replace local protocols.

Clue	Why it draws attention	What it asks for
Several patients with similar illness at the same time	May indicate a common source	Talk to the team and look for links between patients
Severe illness in a young and healthy person	Deviates from the expected pattern	Broaden the differential and the exposure history
Health professional falling ill after providing care	Suggests person-to-person transmission	Review protective measures and notify the service
Travel or contact with an outbreak area	Changes the pre-test probability	Proportional isolation and contact with surveillance
Routine tests negative in a severe illness	The cause may not be in the usual panel	Keep samples and discuss with the laboratory

The biases that delay detection

The biases that delay the detection of a new threat are the same ones that act in any diagnosis, but here the cost is collective. Croskerry describes anchoring, premature closure and other cognitive errors as frequent sources of diagnostic error (Croskerry, 2003).

The "it's seasonal"

When many patients arrive with respiratory symptoms in winter, the explanation "it's the season" is almost always true. That is exactly why it is dangerous: the rule that is right most of the time keeps you from seeing the exception. The post about examples of heuristics in clinical reasoning shows how useful shortcuts fail in predictable situations.

Normalization

Normalization is the tendency to treat what repeats as normal. If the service receives severely ill patients every day, one more seems part of the routine. The question that protects you is simple: is this patient, or this group, similar to what we usually see, or does it only seem so because we are tired?

Calibrated confidence

Overconfidence in the common explanation and excessive alarm at any coincidence are two sides of the same problem. The post about calibrating confidence in diagnosis explains how to adjust the certainty we feel to the strength of the clues we have.

How NEXO helps train this reasoning

To use the case of the day as training, try three habits. Before each guess, ask whether the patient's context changes the probability of the hypotheses. When a clue does not fit, write it down instead of discarding it. After the result, ask which clue about exposure would have led to the diagnosis sooner.

Three exercises with cases

Look for the context. In each case, note what the history says about travel, work and contacts, and what it leaves unsaid.

Imagine the second patient. Ask what would change in your reasoning if another patient with the same presentation arrived the next day.

Separate the common from the new. When the common explanation seems sufficient, list the clues it does not explain.

The post about case-based learning in clinical practice shows how cases help fix patterns that rarely appear in daily practice.

The limits on using the game are described on the NEXO Economy page. If you research the teaching of clinical reasoning, the page for Researchers explains how to get in touch. Professionals who write or review cases find information on Curator and on Creators, and institutions can learn about the proposal on the partnerships page.

Frequently asked questions

What is a sentinel case?

It is a patient, or a small group of patients, whose presentation draws attention to a threat that has not yet been recognized. Many emerging diseases were identified from observations like this.

What is an unusual cluster of cases?

It is a set of patients with a similar presentation, linked by place, time, work, travel or contact, in a number or severity above what is expected. It is a clue that no single patient reveals.

Do you need a confirmed diagnosis to notify?

For unusual events, notification can and should be made on suspicion. It allows surveillance to investigate while the diagnosis is still open, according to each country's rules.

What is the International Health Regulations?

It is the set of World Health Organization rules, adopted in 2005, that guides countries to detect, assess and report events that may constitute a public health emergency of international concern.

What is empirical isolation?

It is the isolation of a patient with suspected transmissible disease before the diagnosis is confirmed. The measure protects other people and is reviewed when test results arrive.

How can clinicians prepare for the next pandemic?

By keeping the habit of asking about travel, work and contacts, paying attention to unusual clusters and knowing the path to notify within their own service.

Take-home messages

• I asked about travel, work, contacts and exposure to animals.
• I considered whether this patient is linked to others with a similar presentation.
• I became suspicious when the presentation was severe in someone who does not usually get sick that way.
• I took protective measures proportional to the suspicion, without waiting for confirmation.
• I reported the unusual event according to the rules of my service.
• I reviewed the decision when the test results and the evolution arrived.

Want to practice reasoning with progressive clues? Play the case of the day.

References

1. Assiri A, Al-Tawfiq JA, Al-Rabeeah AA, et al. Epidemiological, demographic, and clinical characteristics of 47 cases of Middle East respiratory syndrome coronavirus disease from Saudi Arabia: a descriptive study. Lancet Infect Dis. 2013. PMID 23891402.
2. Croskerry P. The importance of cognitive errors in diagnosis and strategies to minimize them. Acad Med. 2003. PMID 12915363.
3. Grein TW, Kamara KB, Rodier G, et al. Rumors of disease in the global village: outbreak verification. Emerg Infect Dis. 2000. PMID 10756142.
4. Huang C, Wang Y, Li X, et al. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. Lancet. 2020. PMID 31986264.
5. Independent Panel for Pandemic Preparedness and Response. COVID-19: make it the last pandemic. 2021. Report page.
6. Liddell AM, Davey RT Jr, Mehta AK, et al. Characteristics and clinical management of a cluster of 3 patients with Ebola virus disease, including the first domestically acquired cases in the United States. Ann Intern Med. 2015. PMID 25961438.
7. Peiris JS, Lai ST, Poon LL, et al. Coronavirus as a possible cause of severe acute respiratory syndrome. Lancet. 2003. PMID 12711465.
8. Poutanen SM, Low DE, Henry B, et al. Identification of severe acute respiratory syndrome in Canada. N Engl J Med. 2003. PMID 12671061.
9. Sachs JD, Karim SSA, Aknin L, et al. The Lancet Commission on lessons for the future from the COVID-19 pandemic. Lancet. 2022. PMID 36115368.
10. Siegel JD, Rhinehart E, Jackson M, Chiarello L, Healthcare Infection Control Practices Advisory Committee. 2007 guideline for isolation precautions: preventing transmission of infectious agents in healthcare settings. 2007. Guideline page.
11. Thornhill JP, Barkati S, Walmsley S, et al. Monkeypox virus infection in humans across 16 countries, April-June 2022. N Engl J Med. 2022. PMID 35866746.
12. World Health Organization. International Health Regulations (2005). Publication page.
13. World Health Organization. World Health Assembly adopts historic Pandemic Agreement to make the world more equitable and safer from future pandemics. 2025. News page.
14. Zaki AM, van Boheemen S, Bestebroer TM, et al. Isolation of a novel coronavirus from a man with pneumonia in Saudi Arabia. N Engl J Med. 2012. PMID 23075143.