The Outbreak Number Is Only the Beginning: Reading November's Health Surveillance Correctly

Surveillance data · preliminaryObservation dates kept

An outbreak headline can look current while describing data collected weeks earlier. A newly released report can describe the previous year. A treatment transfer can appear to create a new case even when the patient was already counted elsewhere. November's public-health reports show why a responsible dashboard needs more than a large number and a recent page timestamp.

WHO released its Global Tuberculosis Report 2026 on November 12. Its central burden estimates concern 2025: approximately 10.6 million people developed TB and 1.1 million died. This is important new reporting about an earlier observation period. It supports renewed attention to access, diagnosis and continuity of treatment, but does not establish a new outbreak in a particular California community. [1]

CDC's November 6 measles update counted 5,182 confirmed US cases as of November 5. That is the latest within-cutoff figure verified for this package and is already reported in the site's November 9 update. It should retain its own date. Updating a webpage on November 12 does not convert those cases into a November 12 measurement. Surveillance counts are preliminary and can change with later reporting or revision. [2]

Respiratory surveillance illustrates another distinction. CDC's November 6 integrated report described low overall national acute respiratory illness with geographic variation, while influenza activity was increasing. Its FluView edition covers the week ending October 31. ECDC's November 9 European summary refers to week44 and different reporting systems. Those summaries can be compared as regional context, but their percentages should not be combined as if they measured one identical population. [3, 4, 5]

For the Bundibugyo Ebola outbreak, WHO's November 6 bulletin uses a November 3 DRC cutoff. Its later dashboard shows 11,114 confirmed cases and 5,433 deaths in DRC as of November 10, while other country rows carry their own dates. The dashboard explains that certain medically evacuated patients are already included in source-country counts. Counting them again at the treatment destination would inflate a total. [6, 7]

A useful site table should therefore include the reporting agency, location, observation date, publication date, case definition and revision note. It should distinguish confirmed, probable and suspected cases, and identify whether a number is cumulative or newly reported during a defined interval. The table need not become intimidating. A short label such as 'confirmed cumulative cases, data through November 10' answers several questions at once.

The clinical question is different from the global counting question. A patient with symptoms needs an assessment based on exposure, timing, individual risk and local advice. A global TB estimate does not diagnose a cough. A distant Ebola outbreak does not mean every traveler has been exposed. Conversely, a low national respiratory level does not make severe symptoms in an individual unimportant. The public-health page should connect people to an appropriate local pathway without attempting individual diagnosis from a dashboard.

For clinics, practical readiness includes a way for patients to disclose relevant symptoms or exposure before entering a crowded waiting room, a current local-health-department contact route, and staff instructions that can be revised when official advice changes. For patients, the helpful message is to seek care for concerning symptoms and tell the receiving service about a relevant exposure or travel history before arrival when possible.

The site's older outbreak snapshots should remain available. Their value is that they show what was known at a given time. A fresh, dated snapshot can come to the top while preserving that history. Readers then see both the current verified picture and how it changed, without mistaking a later publication date, a model estimate or an imported treatment case for a new local event.