Despite no known infections of H5N1 bird flu among dairy cows, Missouri recently discovered a case in an individual who had no apparent exposure to potentially infected animals or related products (i.e., raw milk). A close contact and two health care workers who cared for the person all developed respiratory symptoms but were never tested. There has not yet been a broader increase in other potential cases in the same community indicating efficient human-to-human transmission – the evolution of which is a prerequisite for a possible human epidemic – but this fear underlines the potential danger posed by the continued spread of H5N1.
In recent weeks, California, home to the largest number of dairy cows in the country, became 14the is on report infected herds. It’s possible that H5N1 is even more widespread, including in states with no reported infections among dairy cows, but a lack of testing makes it difficult to pinpoint where the virus is circulating. This bottleneck could be solved by wastewater sampling as close as possible to dairy farms and using genomic sequencing to confirm the presence of H5N1. Sequencing could also assess each detected virus for possible mutations conducive to human transmission and enable phylogenetic analyzes that can help determine which species it may have come from. The more H5N1 is allowed to circulate, especially among dairy cows that are clustered closely together in large numbers and with close human contact, the greater the chance that the virus can evolve for efficient human spread.
Several months after this outbreak, testing on farms has remained limited. Farm owners remain reluctant to allow bulk milk and animal testing for fear of financial loss. Farm workers, many of whom are undocumented, have done so too been hesitant to get tested because of fear of job loss, immigration concerns and, in some cases, because they are unaware there is an outbreak. After nine poultry workers were infected in July, Colorado mandatory routine tests of bulk milk on all dairy farms. This policy soon led to the discovery of 11 infected herds. Colorado required these herds to remain isolated until subsequent testing confirms that viral circulation is no longer present. Massachusetts is the only other state to test all of its dairy herds. Other states have not pursued similar guidelines, likely due to opposition from farm owners and the dairy industry.
Colorado’s approach could be mimicked without mandates by sampling wastewater as close to farms as possible and sequencing it for H5N1. Wastewater does is already being tested of approximately 150 sites for the influenza virus H5 hemagglutinin gene, although this approach does not distinguish between the neuraminidase (“N”) subtype. Sequencing could be used to delineate and monitor the N-subtype different mutations that virologists have identified as potentially necessary for efficient human-to-human transmission. The presence of these mutations could lead to more intensive measures in affected areas, such as testing, vaccination and ensuring workers have personal protective equipment.
The value of a sequence-based approach was demonstrated by a team from the Texas Epidemic Public Health Institute (TEPHI). In one recent letter in the New England Journal of Medicine they report on a two-year proof-of-concept in ten cities for a technique they first described in 2023 (and who won the 2024 STAT Madness competition). Their approach searches for and enriches almost 100 viruses “concern for public health” in wastewater samples where sequencing is performed on detected viral fragments. The researchers started using this approach in May 2022 and found no trace of H5N1 for almost two years. Then, in March 2024, they started seeing it in ten cities in Texas. Sequencing showed that the virus originated from the clade circulating among dairy cows, but did not detect a mutation adaptive for human transmission, consistent with animals as the source.
This type of approach, if decentralized closer to farms and scaled up to urban centers, could provide nationwide insight into the geographic spread of H5N1 and its genomic evolution and help get a handle on transmission. Whenever H5N1 is found in a location, area farms may be required or requested to allow on-farm testing, with no livestock allowed to be moved until cleared from ongoing circulation. Sequencing could also be done in other areas where H5N1 may be circulating, such as Missouri, or where wastewater is found positive for H5 flu without clear links to animal sources, such as in San Francisco Bay.
Four things are needed to implement this strategy. First, a sequencing protocol must be established or developed for scale-up. The TEPHI approach is a proven option. At first glance, agnostic sequencing of nearly 100 pathogens may be overkill for detecting H5N1. It is conceivable that this technique could be adapted to a less comprehensive approach that augments current H5 wastewater monitoring and finds only sequenced samples positive. Efforts to create such a frugal protocol could continue, but to our knowledge have not occurred at any scale and would need to be validated before it can be used more widely, which could take months and require substantial funding. Given the urgency of the current outbreak, it may be more efficient to license and scale up the TEPHI protocol. This protocol has already been validated in peer-reviewed journals and has been widely tested for two years.
Second, state and county officials should identify locations where wastewater should be sampled based on the distribution and location of dairy farms and communities with suspected H5N1. This selection should aim to provide adequate coverage yet be sufficiently disaggregated to allow for sufficient local follow-up and monitoring efforts.
Third, this technique would probably costs considerably more than conventional wastewater testing, including initial capital investments for sequencing equipment and bioinformatics and recurring costs for reagents, software licenses and trained personnel. Required expenditures can be quantified and funds sought from federal and state budgets. Even while awaiting allocations, currently available resources can be immediately focused on the most strategic locations, with others being added as more resources become available.
Fourth, the limitations of wastewater sequencing need to be better understood to inform how its results can best be used to guide public health responses. For example, it is unclear how much virus needs to enter the wastewater before this technology can reliably detect it. If low-grade transmission is not readily established, continued viral circulation could escape control efforts, only to surge again later. It is also unclear how consistently dairy cow waste is flushed into regular wastewater instead of being disposed of separately. These parameters need to be clarified so that strategies can be planned appropriately. In the absence of mandates for on-farm testing, even with imperfect sensitivity, analyzing wastewater may be the best – and only – way to detect and suppress the dairy outbreak.
Missouri officials are conducting serology tests on health care workers and conducting close contact tests, which could indicate whether human-to-human transmission of H5N1 has already occurred. Regardless of the outcome, the lack of a broader footprint of potential cases suggests it may not yet be efficient enough to cause an epidemic. However, the longer the dairy outbreak is allowed to continue, the more likely such a human-transmitted strain will occur. This current fear should wake us up to finally act more aggressively and creatively in addressing this threat.
Ranu S. Dhillon is an instructor in the Division of Global Health Equity at Brigham and Women’s Hospital and Harvard Medical School. He previously served as special advisor to the government of Guinea during the 2013-2016 West African Ebola epidemic. Abraar Karan is an infectious disease physician and epidemiologist at Stanford University; he has worked on the Covid-19 and mpox outbreaks at the provincial, state and federal levels. Devabhaktuni Srikrishna is the founder of Patient knowledgewhich provides practical information to the public about N95 respirators and indoor air filtration.