Insurance works because someone is paid to think about bad days before they arrive. Disease surveillance works the same way, and its underwriters are people: the epidemiologists, laboratory scientists and data specialists who watch illness patterns full-time so that the rest of society does not have to. The recent disruption of a $112 million US epidemiology contract — caught up in broader federal cost-cutting — is therefore best understood not as a budget story but as a workforce story, with lessons our business and insurance audience will recognise from every other risk profession.
Start with what the watchers actually watch. Reports flow continuously from hospitals, clinics and laboratories: diagnoses, test results, unusual clusters. Surveillance teams clean that flow, compare it against seasonal baselines, and escalate anomalies. Most anomalies are nothing. The system’s entire value lies in catching the few that are something — early, while responses are still cheap and effective.
Three professions, one early-warning system
The first profession is field epidemiology: trained investigators, mostly in state and local health departments, who interview patients, trace exposures and decide whether scattered cases share a source. The second is the laboratory workforce, whose genomic sequencing can now link — or clear — cases across state lines with a precision earlier generations could not attempt. Agencies such as the CDC coordinate standards and surge support, but the bench work is distributed across dozens of public laboratories.
The third profession is the newest and the least visible: surveillance data engineering. Feeding thousands of incompatible reporting streams into one trustworthy picture is a full-time technical discipline. Practitioners describe their work as plumbing, and the metaphor is exact — essential, invisible, and noticed only on failure.
Why people, not platforms, are the asset
Vendors can supply software, sequencing machines and dashboards. What they cannot supply quickly is judgement: the experienced investigator who hears the significant detail in a patient’s story, the genomicist who recognises an artefact masquerading as a mutation, the data lead who knows which county’s feed silently broke last spring. That judgement is accumulated over years, held in teams, and lost when teams disperse.
This is the asymmetry at the heart of the current dispute. Contracts can be paused by signature; workforces cannot be paused and resumed the same way. Specialists who leave for other sectors rarely wait by the phone. Rebuilding a surveillance team takes recruitment cycles, training fellowships and supervised field experience — timelines measured in years, set against outbreaks that move in weeks.
The risk-management conclusion
Insurers price the difference between prevention and reaction for a living, and surveillance is prevention in its purest form: modest, steady spending that truncates the tail of catastrophic outcomes. The World Health Organization‘s materials on epidemiology make the same point in public-health language — surveillance is the foundation on which every other response stands.
The comparison that should settle it
Consider the two bills a society can pay. The first is the steady, budgeted cost of surveillance: salaries, sequencers, data systems and the occasional false alarm chased down by professionals. The second is the unbudgeted cost of the outbreak caught late — hospital surges, closed schools and workplaces, emergency procurement at panic prices, and the long tail of illness that early action would have prevented. Actuaries do not find this a difficult comparison, and neither, historically, do finance ministries once they have paid the second bill a single time. The workforce question is therefore not whether the watchers are worth their cost, but whether the cost is being paid to the people who actually do the watching.
Whatever the final settlement of the $112 million contract, the durable question for policymakers is whether the workforce it funded remains intact at the end of the argument. Early-warning systems do not fail loudly; they fail silently, on the day the watcher was no longer watching. Keeping skilled watchers in post is the whole policy — a conclusion as familiar in risk management as it is, apparently, easy to forget. More analysis is available on our homepage.
