H5N1 in the United States: CDC surveillance expanded rapidly, but exposed workers still fall through the gaps

The emergence of highly pathogenic avian influenza A(H5N1) in US dairy cattle in 2024 created a new and persistent animal–human interface. A new review of the Centers for Disease Control and Prevention response concludes that surveillance, diagnostics and genomic risk assessment have expanded substantially, but uneven testing, limited participation by exposed workers and uncertainty around mild or asymptomatic infections remain important blind spots.

Dairy cattle changed the surveillance landscape

A study identify the detection of clade 2.3.4.4b H5N1 in US dairy cattle in 2024 as a major turning point. Unlike sporadic spillover from wild birds or poultry, infected dairy herds created repeated occupational exposure for farm workers, veterinarians and other people in contact with cattle and raw milk. That required closer coordination between animal health, public health, occupational health, agriculture and food safety.

The CDC response combined monitoring of exposed people with public-health laboratory testing, rapid H5 subtyping, whole-genome sequencing, antigenic characterisation and assessment of antiviral susceptibility.

Human cases were usually mild, but not always

According to the review, most human infections associated with US cattle and poultry outbreaks were mild, frequently presenting with conjunctivitis or upper-respiratory symptoms. However, severe disease and fatalities also occurred. That contrast is important because symptom-based surveillance is more likely to detect severe cases than mild or asymptomatic infections. The review notes that serological studies of exposed workers suggest some infections may have gone undetected by routine symptom monitoring.

Occupational surveillance remains a weak point

The authors highlight several practical limitations. Participation in testing has been incomplete among exposed workers, while access to and consistent use of personal protective equipment can be difficult in real farm settings. Language, employment conditions, concerns about income loss and reluctance to engage with authorities can all complicate surveillance, although the review’s central point is broader: a technically strong surveillance system is only effective if exposed people are willing and able to use it.

CDC guidance has included PPE recommendations, symptom monitoring, testing of exposed workers and antiviral prophylaxis after selected high-risk exposures.

Genomic surveillance is central to pandemic risk assessment

Sequencing plays a major role in determining whether the virus is becoming better adapted to mammals. CDC laboratories monitor viral genomes for mutations associated with mammalian adaptation, changes in receptor binding or polymerase activity, as well as evidence of antiviral resistance.

The review also emphasises antigenic characterisation, assessing how well currently circulating viruses match existing candidate vaccine viruses.

This means surveillance is not simply counting human cases. It is also continuously asking whether the biological properties of the virus are changing in ways that could increase human transmission risk.

The main remaining gaps

The authors identify several persistent uncertainties:

• the true frequency of mild and asymptomatic infections;

• incomplete testing among exposed workers;

• uneven PPE use;

• rapidly changing outbreak data;

• and insufficient integration of animal and human surveillance information.

The review therefore argues for more accessible worker-centred testing, transparent data sharing and tighter integration between veterinary and public-health datasets.

A One Health problem by definition

The cattle outbreaks illustrate why H5N1 cannot be managed through human surveillance alone. Human risk is determined partly by what happens in dairy herds, poultry flocks and wild birds, and the effectiveness of public-health preparedness therefore depends on how quickly animal outbreaks are detected and controlled.

The authors conclude that CDC preparedness has improved substantially, but its effectiveness depends on sustained One Health coordination, continued monitoring of viral evolution, diagnostic performance and antiviral susceptibility, and better alignment between animal and human data systems.

The central lesson is that the US H5N1 response is increasingly sophisticated at the laboratory level, but the next major gains may depend less on new diagnostics than on reaching exposed workers earlier and connecting animal and human surveillance more effectively.