#18,212
The state of Michigan has been rightfully lauded for having perhaps the most aggressive testing and monitoring program for farm workers exposed to HPAI in the country. Through their efforts, two dairy workers were identified as having mild H5N1 infection (see here and here).
Michigan's MDARD also invited the USDA to investigate the spread of the H5 virus in farms (see Epidemiological investigation into the spread of H5N1 in Michigan) which concluded the movement of livestock, farm equipment (including trucks), and personnel contributed to the spread of the H5N1 virus.
Other states have been far less cooperative, although Colorado recently asked the CDC to send investigators after 9 people were infected by H5N1 across two poultry farms, and last week announced the most aggressive dairy cattle testing in the nation.
Grading on the curve, Michigan would appear to get an A+, but - as the following MMWR report illustrates - there were limitations to their surveillance as well.
- Of the 23 known (now 26) infected dairies, only 20 appear to have provided employee lists.
- Illnesses and/or symptoms were mostly self reported, and testing was voluntary.
- Only symptomatic individuals were tested.
It is unknown how many other infected farms there might be in Michigan, and it is also possible that some individuals with minor symptoms - or an aversion to being tested - might not have responded to the daily texts.
There is also often a narrow `window' where testing will reliably detect HPAI H5, particularly if these workers were on already on prophylactic antivirals (not stated, but a common practice).
Still, given that perfect is often the enemy of good, this is far-and-away the most comprehensive testing and surveillance program we've seen since HPAI H5 emerged in American dairy cattle, and it may be the best we can hope for.
The MMWR report follows, after which I'll have a brief postscript.Given the anecdotal reports (see EID Journal: Avian Influenza A(H5N1) Virus among Dairy Cattle, Texas, USA) of sick dairy workers who have refused to be tested, and the (likely large) number of unidentified infected herds across the nation, the known number of human infections is likely a significant undercount.
Notes from the Field: Health Monitoring, Testing, and Case Identification Among Persons Exposed to Influenza A(H5N1) — Michigan, 2024
Weekly / July 25, 2024 / 73(29);656–658
Joseph Coyle, MPH1; Natasha Bagdasarian, MD1; Seth Eckel, MPH1; Jeremy Kuo, MPH1; Mary Grace Stobierski, DVM1; James Barber, MPH1; Megan Weinberg, PhD1; Fatema Mamou, MPH1; Sarah Lyon-Callo, PhD1; Michigan Local Health Departments; Bureau of Laboratories; Bureau of Infectious Disease Prevention Investigation Team (VIEW AUTHOR AFFILIATIONS)View suggested citationSummaryArticle PDF
What is already known about this topic?
Highly pathogenic avian influenza (HPAI) A(H5N1) virus has been detected in wild birds and mammals, poultry, and commercial dairy facilities in the United States. A human case in a Texas dairy worker was reported in April 2024.
What is added by this report?
As of May 23, 2024, Michigan had the largest number of affected dairy and poultry facilities linked to the HPAI A(H5N1) outbreak. Active symptom monitoring and testing of exposed workers led to detection of the second and third known dairy-associated HPAI A(H5N1) cases in 2024.
What are the implications for public health practice?
The current risk to the public from HPAI A(H5N1) viruses is low; however, continued symptom monitoring and testing are critical to characterizing genetic or epidemiological changes that might alter the risk assessment.
Full Issue PDF
On March 25, 2024, a Texas dairy farm detected highly pathogenic avian influenza (HPAI) A(H5N1) virus in cows. The outbreak widely spread after interstate cow movement. During March 25–June 17, animals at a total of 102 dairy farms in 12 states, 24 commercial poultry flocks in five states, and multiple backyard flocks tested positive for HPAI A(H5N1) (1,2). This report describes response activities in Michigan, which led to detection of the second and third human cases related to the 2024 HPAI A(H5N1) outbreak. The activity was reviewed by the Michigan Department of Health and Human Services, deemed not research, and was conducted consistent with applicable federal law, state, and departmental policy.*
Investigation and Outcomes
Infected cows from Texas resulted in introduction of HPAI A(H5N1) virus in a Michigan dairy, detected on March 29. As of May 29, a total of 23 Michigan dairies in 10 counties are known to be affected (1). Michigan’s first affected commercial poultry facility was confirmed on April 2; currently, seven affected poultry facilities in four counties have been identified (2). HPAI A(H5N1) virus has also been detected in a backyard flock, pigeons, foxes, cats, opossums, and a racoon in Michigan. Whole genome sequencing results suggest that, since March 2024, all sequenced isolates have ancestral Texas origins (3).
Monitoring of Dairy Workers
Among the 23 affected dairies, 306 persons exposed to affected cows were identified. Lists of exposed persons were obtained by public health officials from 20 (87%) affected dairies. Workers at 12 (60%) of those dairies were enrolled in text-based daily symptom monitoring,† and workers at eight (40%) farms were monitored through a farm point of contact. Because it could be unclear when workers’ exposures to cows ended, some workers were monitored for >50 days.
Twenty (6.5%) exposed workers reported symptoms and were tested for influenza A(H5) virus infection. Among persons who received real-time reverse transcription–polymerase chain reaction testing,§ one received a positive test result from a conjunctival swab, similar to the case of HPAI A(H5N1) reported from a dairy worker in Texas (4).
Before the onset of mild unilateral conjunctivitis, the patient reported direct ocular exposure to raw, unpasteurized milk from an affected cow. A second worker from a different dairy farm experienced respiratory symptoms after close contact with sick cows and received a positive A(H5) virus test result from a nasopharyngeal swab.
In both instances, public health officials rapidly collected patient specimens, which tested positive for HPAI A(H5N1). Neither worker was severely ill, neither required hospitalization, and no household or work contacts reported being ill. Both workers wore some personal protective equipment (PPE), but neither wore a mask or respirator.
Monitoring of Poultry Workers
Among seven affected commercial poultry facilities, 857 persons exposed to affected birds were identified. Lists of exposed persons were obtained from all facilities. Workers from four facilities were directly enrolled in text-based daily symptom monitoring, and workers from three facilities were monitored through a farm point of contact who reported results to public health officials. Eighteen (2.1%) symptomatic persons were identified and tested; all test results were negative for influenza A(H5).
Monitoring of Other Exposed Persons
Federal and state employees who responded to affected farms were also observed for symptoms, as were persons with exposure to HPAI A(H5N1) virus–infected animals (domestic or wild) or humans. Overall, 125 such persons were monitored, and 15 (12%) reported symptoms, 14 of whom received negative influenza A(H5) test results.
Preliminary Conclusions and Actions
Among 1,288 Michigan residents who were monitored for signs and symptoms after potential HPAI A(H5N1) virus exposure, 53 (4.1%) reported signs and symptoms, 52 of whom received testing for influenza A(H5). Two dairy workers received positive test results (3.8% of all persons tested, <1% of all monitored dairy workers).
Although the risk for HPAI A(H5N1) virus to the public remains low, novel influenza A viruses such as A(H5N1) have pandemic potential. Therefore, it is critical to notify persons with exposure to infected animals, provide education and access to PPE,¶ monitor signs and symptoms, test specimens collected from any exposed person with signs and symptoms, and make antivirals available to symptomatic persons as soon as possible.**
Although the percentage of workers who regularly used PPE is not known, the human cases associated with dairy farms in Texas and Michigan demonstrate the potential value of PPE, including eye and respiratory protection, especially on affected farms (4,5). The cases identified to date have resulted in mild illness, which might not have been detected without the collaboration of state officials and the engagement of farms and workers. Streamlined, nonintrusive approaches to monitoring, such as the text-message monitoring used in Michigan, might encourage participation and subsequent testing. A One Health†† approach including collaboration with agriculture departments, farms, and workers is crucial to successful public health response.
Two weeks ago the CDC held a COCA Call for clinicians to help them identify, treat, and report suspected novel flu infections (see CDC COCA Call On H5N1 for Clinicians & Healthcare Centers Now Online).
This presentation also included a 48-slide PDF file.
As we've seen previously from the UKHSA (see TTD (Time to Detect): Revisited), identifying community spread of a novel virus could take weeks, and only come after dozens or even hundreds of people had been infected.
Our ability to detect H5 cases in the community depends not only on a bit luck, but also on how hard we are actually looking.
And right now, in far too many places, we barely appear to be looking at all.
