Outbreak case counts: what the number on the market actually measures

The number on the CDC page is confirmed cases, and confirmed is a technical word rather than a colloquial one. A case qualifies through laboratory evidence - a positive PCR, a measles-specific IgM result, or a viral genotype - or through an epidemiological link to a case that already has that evidence. A person with a textbook rash, fever and cough who is never tested and never linked to a confirmed case is not in this number, however obvious the illness looked in the clinic.

There is a second tier below it. States report both probable and confirmed cases into the national surveillance system, and probable means the clinical picture fits without the laboratory or the link. Those cases exist in the data and are not in the headline count. A run of probable cases that later gets laboratory confirmation therefore arrives in the total weeks after the illness, which is one of the two reasons the count moves in bursts.

The definition itself is not permanent. Case definitions for nationally notifiable conditions are set through position statements by the Council of State and Territorial Epidemiologists and revised annually. A market that spans a definition change is asking a slightly different question at the end than it was at the start, and nothing on the page announces that.

Its official job is to trigger action. Case counts drive vaccination campaigns, school exclusion orders, contact tracing and the allocation of federal support to a state. Because those consequences are expensive, the count is built to be defensible rather than fast, and every design choice in it trades timeliness for the ability to stand behind each entry.

Its second job is to hold a status. The United States has been formally measles-eliminated since 2000, and elimination is defined as the absence of continuous transmission of a single chain for twelve months or more. A count is therefore not only a tally but evidence in a running argument about whether that designation survives - which is why the outbreak-associated share matters more to public health than the total does.

Its third job, more recent, is to settle contracts. The 2026 board carries a ladder of thresholds on the same series - 2,600 by a fixed date, then 3,000, 3,500, 4,000, 5,000, 7,500, 10,000 and 12,500 for the calendar year. They are one question asked eight times at different distances, which makes the ladder unusually good for reading what the market thinks the rest of the season looks like.

Two federal series carry these numbers and they do not agree at any given moment. The measles page maintained by CDC is the fast one: it is refreshed weekly and reflects only confirmed cases notified to CDC as of noon on Thursday. That cutoff is the single most important fact for anyone holding a position into a deadline - a case confirmed by a state on Thursday afternoon is real, is reported, and is not in this week's number.

The National Notifiable Diseases Surveillance System is the slow one. States notify it on their regular electronic transmission schedule, it publishes weekly provisional tables, and it finalises the year afterwards. Provisional and finalised are different numbers by design, and the gap between them is not an error being corrected but a definition being applied later with more evidence.

Below both sit the states, which is where the number is actually generated. A state health department is required to notify CDC of a measles case promptly - within twenty-four hours, by direct email - and many publish their own dashboards on their own schedules. State totals therefore run ahead of the national page most weeks, and a trader watching four state dashboards can see the next national update coming. That is legitimate research, not an edge in the market's own data.