Two reputable sources can publish different plague totals without either one making a simple arithmetic error. They may use different case definitions, reporting dates, geographic boundaries or stages of verification.
The evidence ladder
A suspected case meets a clinical and epidemiological definition that signals the need for investigation. A probable case has stronger supporting evidence but may still lack complete laboratory confirmation. A confirmed case meets the laboratory criteria specified by the relevant surveillance system.
These categories should be displayed separately. Adding them together without a label makes a provisional signal look like a confirmed total.
Zero and unknown are different values
- Zero means the source explicitly established that none were recorded.
- Not reported means the source did not provide that field.
- Not established means the available evidence cannot support a reliable number.
- Not applicable means the measure cannot sensibly be calculated.
Plague Map stores the last three as null and displays them as N/A. Replacing them with zero produces false certainty and distorts totals.
Dates matter as much as numbers
An early situation report may count patients under investigation. A later report can reclassify them after testing. Deaths and cases may also refer to different cut-off dates. A case-fatality calculation is meaningful only when its numerator and denominator use compatible definitions and periods.
Our event records therefore keep a source date, data-through date and editorial-review date. The number should always travel with those fields.