The interval that matters
Imagine a health department receiving an unusual laboratory result late on a Friday. The result may be important. It may also reflect a reporting error, a change in testing, or an event already investigated elsewhere. Somebody has to establish what it means, find the responsible people, and decide what action is justified. A technically excellent detection system can still leave that person with an unanswerable question.
Our position is straightforward: biological preparedness should be judged across the entire interval between an observation and a useful response. That means measuring the time spent collecting information, interpreting it, assigning responsibility, and obtaining the resources to act. Improving one step while the others remain slow may produce a better demonstration without creating a stronger public-health capability.
WHO's collaborative surveillance framework puts cooperation among institutions and sectors inside the definition of surveillance capacity. It connects disease monitoring, laboratory information, and other evidence to decisions. The implication for researchers is substantial: the interfaces between organisations deserve the same attention as the instruments inside them.
Every signal has a field of view
Wastewater monitoring shows both the promise and the limits of early warning. CDC explains that it can capture community infection trends without depending on everyone developing symptoms, seeking care, or receiving an individual test. It can therefore complement clinical reporting when access to testing is uneven. But municipal wastewater networks do not represent everyone. Rural areas and communities using septic systems may be absent. Measurements also cannot be converted directly into a reliable count of infected people. These qualifications are central to CDC's interpretation guidance.
A useful surveillance interface should make that field of view explicit. The reader needs the collection period, the population represented, known coverage gaps, and an explanation of what the measurement can establish. Otherwise a clean national map may conceal a patchwork of very different local observations.
Consider a hypothetical comparison between two districts. One has regular monitoring; the other has none. Displaying the second district as having low activity would manufacture confidence from missing information. The honest display distinguishes low measured activity from an absence of measurement. That distinction should survive every aggregation, export, and briefing.
New York's polio response shows what connections deliver
In 2022, an unvaccinated adult in Rockland County, New York, developed paralytic polio. The subsequent investigation connected a clinical diagnosis with environmental monitoring and vaccination activity. By August 10, authorities had tested 260 wastewater samples from Rockland and Orange counties; 21 were positive for poliovirus. Twenty positive samples collected during May, June, and July were genetically linked to the patient's virus. Some samples had originally been collected for SARS-CoV-2 surveillance. The response included clinician alerts, enhanced surveillance, assessment of vaccination coverage, and additional vaccination clinics. These findings are documented in CDC's contemporaneous investigation.
The retrospective timing matters. Those samples helped establish earlier circulation; their existence does not mean the system had prospectively issued an early polio warning. A responsible assessment keeps discovery after an event separate from warning before it.
Our reading is that the case demonstrates the value of connected capabilities. Clinical recognition, preserved information, laboratory interpretation, environmental surveillance, and a practical prevention response each contributed something different. The lesson for investment is to fund the connections and the ability to reuse established infrastructure, while being precise about what a particular system actually achieved.
Finding an event is different from finding it early
WHO's Epidemic Intelligence from Open Sources system brings publicly available information into public-health assessment. A news report or other public signal can prompt authorities to verify an event. The important word is verify: an item appearing in a feed is a lead, with a source and a level of uncertainty.
A 2025 evaluation examined 646 events officially notified to WHO by countries in the African region during 2018–2023. Researchers found corresponding EIOS signals for 81% of those events, but signals preceding official notification for 47.4%. The median timing difference was zero days. The study was retrospective and used notification to WHO as its reference; it did not establish when local authorities first knew about each event.
Those distinctions should shape procurement and evaluation. A system can have broad coverage without providing much additional time. A useful trial should specify its comparator, preserve timestamps, and examine whether the earlier information changed a decision. Counting collected articles or matched historical events cannot answer that question. The research result is valuable precisely because it exposes several different measures of performance.
Genomic capacity must extend beyond the sequence
WHO's 2022–2032 genomic surveillance strategy places sequencing inside a broader laboratory and surveillance system. It calls for connected capacity from collection and diagnostics through analysis and data sharing. Its stated goal is timely access to sequencing for pathogens with pandemic or epidemic potential across all 194 Member States by 2032. That is a strategic objective, not evidence that access has already been achieved.
The engineering consequence is easy to overlook. A sequence must be connected to enough context to answer the relevant public-health question, while respecting restrictions on sensitive information. Analysts must also know how the observations were selected. A fast result from a narrow or poorly understood sample can be useful, but its limits have to remain attached.
For a research team, the starting point should be a decision that someone needs to make. Work backwards from that decision to the evidence required, the uncertainty that can be tolerated, and the reporting deadline. This gives technical development a concrete purpose. It also makes failures visible before they become features in a polished dashboard.
Trust determines what a surveillance programme can sustain
Public-health surveillance involves choices about whose information is collected, who can inspect it, and what consequences follow. WHO's ethical guidance addresses privacy, autonomy, equity, and public benefit. It recognises that surveillance can help expose neglected suffering while also creating risks for the people being observed.
Those obligations need an operational form. A programme should be able to explain its purpose, justify the detail it collects, define access responsibilities, and describe how affected communities benefit. The argument for retaining a particular field should be stronger than the convenience of having it. A useful system also needs a route for correcting an error before that error becomes an enduring description of a place or population.
Our judgement is that trust should be treated as a condition of long-term performance. An institution asking communities to cooperate cannot postpone questions about ownership and accountability until deployment. International collaboration makes this harder and more necessary: data may cross a border faster than the agreements governing its interpretation. Those agreements should be built while there is time to discuss them carefully.
Evaluate the handover, then fund the weakness
A practical preparedness exercise can begin with an ordinary hypothetical signal and follow it through the organisation. Who receives it? Who checks it? Who can contact a relevant laboratory? Who has authority to commit resources? Which of those steps still works outside normal hours? Record the delays and the information people repeatedly have to request.
We propose judging a surveillance programme on four concrete outcomes: additional time gained over an existing process, the quality of decisions supported, the burden placed on people reviewing alerts, and the populations it reliably represents. These are proposed evaluation criteria, not universal thresholds. Each setting must establish what useful performance means.
This approach changes the funding conversation. The limiting factor may be an analytical method. It may also be a missing agreement, a vacant role, or a reporting process that cannot accept the evidence being produced. Repairing that weakness can have more value than adding another source of data. Biological early warning succeeds when an institution can turn an uncertain observation into a timely, proportionate decision—and explain why it made it.