IPM Take
Epidemiological curves have politics hidden inside them. A line on a surveillance dashboard looks objective, but what gets counted depends partly on the case definition sitting underneath it.
A new US pertussis analysis provides a useful lesson in how to communicate resurgence without distorting it: the increase was real, but the 2020 surveillance definition substantially amplified the magnitude seen in reported national data.
Executive Summary
Researchers evaluated how the US 2020 Council of State and Territorial Epidemiologists pertussis case-definition change affected reported case numbers during the major 2024 resurgence.
Among Enhanced Pertussis Surveillance cases, the proportion newly reportable under the 2020 definition increased from 8.6% in 2021 to 38.0% in 2024. Applying that estimate to preliminary national 2024 surveillance data suggested that 16,462 of 43,321 reported cases would not have met the previous 2014 definition. Reported pertussis increased 133% between 2019 and preliminary 2024 data; the investigators projected that the increase would have been approximately 44% if the earlier definition had remained in use.
Why it matters
- Public authorities: Changes in case definitions should be made visible when communicating long-term disease trends.
- Policymakers: A larger surveillance count does not automatically mean the additional cases are false; broader definitions may deliberately improve sensitivity.
- Researchers and media: Comparing headline totals across years without accounting for measurement changes can misrepresent the magnitude of epidemiological change.
When infectious-disease numbers rise sharply, the public question is usually straightforward: why are more people getting sick? Surveillance systems have to ask an additional question before answering: are we counting disease in exactly the same way we used to?
Pertussis provides an unusually clear example. The United States changed its surveillance case definition in 2020, allowing an acute cough illness of any duration to be classified as confirmed when Bordetella pertussis is detected by PCR or culture. The change reflected increasing confidence in molecular testing and reduced the need for public-health personnel to wait for prolonged cough duration before a laboratory-positive infection could meet the national confirmed-case definition.
That change was designed to improve surveillance sensitivity, but it also complicates comparisons with earlier years. The newly published analysis examined Enhanced Pertussis Surveillance data and then estimated how the definition affected national reporting as case numbers surged again in 2024.
The result is substantial. In 2021, 8.6% of cases in the enhanced surveillance dataset were newly reportable under the 2020 definition. By 2024 the proportion had risen to 38%. Applied to preliminary national data, the investigators estimated that 16,462 of 43,321 reported 2024 cases would not have met the older 2014 criteria.
That does not mean 16,462 cases were invented, misdiagnosed or epidemiologically meaningless. It means they met the newer surveillance criteria but would not have been counted under the old ones. Confusing those two statements would turn a legitimate methodological finding into misinformation.
Nor does the study erase the pertussis resurgence. The investigators estimate that reported cases rose 133% between 2019 and preliminary 2024 data. Had the 2014 definition remained in place, they project an increase of roughly 44%. A 44% rise is still a significant rise; the new definition changed its measured scale.
This is precisely why surveillance governance matters. Case definitions evolve for good reasons as diagnostics improve and clinical understanding changes. Holding definitions permanently static simply to preserve comparability would produce another form of bad data. The responsibility is therefore not to avoid methodological change but to document its effect clearly enough that policymakers, journalists and the public do not interpret two differently constructed numbers as if they were perfectly equivalent.
The issue extends far beyond pertussis. New laboratory tests, revised disease classifications, changing screening programmes and improved reporting can all make disease incidence appear to shift. Sometimes the pathogen changes. Sometimes detection changes. Often both are happening at once.
Public-health communication tends to struggle with that complexity because a single percentage is easier to headline than an explanation of denominators and case definitions. But precision public health requires precision in interpretation as much as precision in testing.
The correct conclusion from the new pertussis analysis is therefore neither that the 2024 resurgence was an artefact nor that every additional reported case represents an equivalent increase in disease burden. The stronger conclusion is that America had a real resurgence measured through a surveillance system that had also become more inclusive.
Both facts belong on the same graph.
