Abstract
The risk for transmission of highly pathogenic avian influenza A(H5N1) virus from dairy cows to humans is currently low; however, personal protective equipment (PPE) use during work activities on dairy farms has not been well described. PPE use can protect farmworkers when they are working with highly pathogenic avian influenza A(H5N1)–infected cows. The Colorado Department of Public Health and Environment (CDPHE) and the Colorado Department of Agriculture (CDA) offered PPE to all Colorado farms before or during an A(H5N1) outbreak in cows in 2024. CDPHE surveyed 83 dairy workers from three farms with a confirmed bovine A(H5N1) outbreak. Frequently reported farm worker activities included milking cows or working in the milking parlor (51%), cleaning cow manure (49%), and transporting cows (46%). Frequently reported PPE items available to workers before A(H5N1) outbreaks included gloves (88%), eye protection (e.g., safety glasses or goggles) (76%), rubber boots or boot covers (71%), and head covers (69%). N95 respirator use was low among workers who were exposed to ill cows after detection of A(H5N1) virus (26%). PPE use while working with ill cows increased a mean of 28% after detection of A(H5N1) virus on surveyed farms; use of eye protection while milking cows increased the most (40%). Public health PPE distribution, education, and collaboration with CDA might have increased PPE use on dairy farms with A(H5N1) virus–infected cows and mitigated risk for farmworkers acquiring A(H5N1) virus.

Investigation and Results
Background
On April 25, 2024, Colorado detected highly pathogenic avian influenza (HPAI) A(H5N1) virus in cows on a dairy farm. During April–August 2024, an additional 63 dairy farms with A(H5N1) virus–positive cows were identified* in Colorado. Although transmission risk for A(H5N1) viruses from animals to humans is low, transmission has occurred in the United States, including to a dairy worker in Colorado in 2024 and to poultry workers in 2022 and 2024 (1–5). Whereas A(H5N1) virus–infected poultry typically experience rapid and high mortality, cows tend to recover (<2% herd mortality) and require ongoing care such as milking during illness (6). Exposure to ill? cows and raw milk from acutely infected dairy cows poses a transmission risk to workers. CDC recommends use of personal protective equipment (PPE) by persons who are in contact with or near dairy cows, raw milk, or items that might be contaminated with A(H5N1) viruses (5,7,8). Recommended PPE includes fluid-resistant coveralls, an optional waterproof apron, a National Institute for Occupational Safety and Health (NIOSH)–approved particulate respirator (e.g., an N95§ filtering facepiece respirator [FFR]), goggles, a head or hair cover, gloves, boots or boot covers, and an optional face shield.

The Colorado Department of Public Health and Environment (CDPHE) collaborated with the Colorado Department of Agriculture (CDA) to respond to dairy farm A(H5N1) outbreaks? in Colorado. CDPHE and CDA conducted outreach to farm owners, offering site visits to educate and advise them about protecting dairy farmworkers from ill cows. A letter with information on A(H5N1) prevention, PPE best practices, and a PPE order link was sent to dairy facilities and industry partners (7). The link was publicly available on CDA’s website for ordering a free 1-month supply of N95 FFRs, surgical masks, face shields, goggles, and nitrile gloves to all Colorado farms before or during an outbreak.

Dairy farmworkers complete numerous tasks involving close contact with cows, but their specific duties and PPE use are not well described. This report includes survey results describing work activities and PPE use on three dairy farms in Colorado before and after A(H5N1) outbreaks were identified.

Farm Participation
Colorado dairy farms with A(H5N1) virus–infected cows during July–August 2024 were eligible to participate in the retrospective survey. CDPHE contacted management at 43 affected dairy farms by telephone. Farm management gauged worker interest and scheduled CDPHE interview site visits. Individual workers voluntarily participated in interviews during site visits. Three farms located <50 miles apart in the same region of Colorado were included as a convenience sample (approximately 250 total workers). One farm included four smaller farms, which were considered a single farm for the project. One participating farm had preordered PPE before A(H5N1) virus detection, and the other two farms received PPE during their initial public health site visit, after A(H5N1) virus detection. All three farms opted to receive public health site visits upon detection. Thirty-seven (64%) of the 58 affected farms also ordered PPE. This activity was reviewed by CDC, deemed not research, and was conducted consistent with applicable federal law and CDC policy.**

Data Collection and Analysis Methods
CDPHE conducted voluntary worker surveys in English or Spanish during site visits to the three participating farms using a structured interview questionnaire. Workers participated anonymously and received a gift card incentive for completing the survey. Interviews gathered information on work exposure, including contact with ill cows, work duties, and PPE use, both in the weeks before and after A(H5N1) virus detection on the farm. No worker or interview information was shared with employers. Summary statistics described work activities and PPE use, and adjacency matrix heat mapping methods compared the two variable group frequencies to identify clusters and overlapping trends. All analyses were conducted using R statistical software (version 2024.09.0; R Foundation).

Participating Farm Workers
During July–August 2024, CDPHE interviewed 83 (34% of approximately 250 total employees) dairy farmworkers; among these, 72 (87%) were interviewed in Spanish. Interviews were conducted a median of 48 days after infected cows were reported to public health (IQR = 47–49 days). Among interviewed dairy workers, 44 (53%) reported exposure to ill cows starting the week before A(H5N1) virus was first detected in the cows.

Reported Work Duties
The most commonly reported work duties were milking cows or working in the milking parlor (51%), cleaning cow manure (49%), and transporting cows (46%) (Table 1). The least commonly reported duties included repair or maintenance and hospital pen work (20%), breeding and artificial insemination (18%), and milk transport (12%).

Persons working in a milking parlor might be responsible for disinfecting and drying teats and attaching and removing milking equipment. Farms might have one parlor in which all cows are milked or have a separate parlor for only ill animals for isolation and biosecurity. Manure management can involve being near cattle and contact with urine and feces. Transporting cows includes moving cows around the farm, loading cows onto vehicles, and operating vehicles. Although moving cows is mainly carried out from a distance (cows typically retreat from a worker approach), this activity occasionally does involve direct cow contact. The survey also identified a decrease in the median number of work duties performed once A(H5N1) virus was detected (from five to three activities).

Reported PPE Use
When asked about their access to individual elements of PPE before outbreak detection, 88% of workers reported access to gloves, 76% reported access to eye protection such as safety glasses or goggles, 71% reported access to rubber boots or boot covers, and 69% reported access to head covers. Reported use of many individual PPE items was higher among dairy workers who reported exposure to ill cows in the week before or week after the detection of A(H5N1) on the farm, compared with those who did not report exposure to ill cows (Table 2). Dairy workers exposed to ill cows during the week after A(H5N1) virus detection reported higher use of gloves (93%), boots or boot covers (83%), head or hair covers (79%), and eye protection (76%) compared with those who reported exposure to ill cows in the week before detection of HPAI A(H5N1). Reported use of N95 FFRs or other respirators and other types of masks was low (9% and 27%, respectively) among workers exposed to ill cows the week before A(H5N1) outbreaks were detected, with higher usage reported among exposed workers in the week after outbreak detection (26% and 36%, respectively). Use of all CDC-recommended PPE was low among workers both in the weeks before (2%) and after (5%) A(H5N1) virus detection. Workers also reported use of items such as sunglasses and bandanas or gaiters; these items are not recommended PPE.

Adjacency matrix frequencies identified a group of work duties with the highest reported PPE use in the weeks before and after A(H5N1) virus detection (Figure). Workers transporting cows, cleaning cow manure, and milking cows reported the highest frequency of use of gloves, rubber boots or boot covers, head covers, and eye protection. The mean use frequency of these PPE items while performing the same activities increased 28% during the week following detection of A(H5N1) virus. The largest increase in reported use frequency of eye protection from the week before to the week after detection of A(H5N1) occurred in workers who milk cows (40%). The highest correlations of work duty and PPE use (24 workers) were wearing gloves while transporting cows and wearing gloves while cleaning manure or feces (Figure).


Discussion
The lower mortality rate of cows infected with highly pathogenic avian influenza A(H5N1) viruses compared with that of birds can result in prolonged dairy farm worker exposures to ill cows through poorly understood transmission routes. Reported PPE use among workers at three dairy farms with A(H5N1) influenza outbreaks was high for some items, with the largest increase reported in frequency of use of eye protection once outbreaks were detected. Milking cows, the most frequently reported work duty by interviewed workers, is thought to pose higher risk for cow-to-human A(H5N1) virus transmission because of exposure to raw milk (6). A previous human case of A(H5N1) in a dairy worker identified in Michigan reported milk splashing into the eyes before receiving a positive A(H5N1) test result (9). The reported increase in eye protection among workers milking cows after A(H5N1) detection is important for protecting against A(H5N1) exposure. Dairy farms might isolate ill cows to reduce transmission, but the occurrence of asymptomatic A(H5N1) virus infections in cows means that workers can still be exposed during the milking process, highlighting the need for and importance of PPE during milking for all cows on dairy farms with A(H5N1) virus infections detected (10).

Reported use of N95 FFRs and other types of masks was low during most work activities. Dairy farmworkers’ duties often involve exposure to manure or milk that can contaminate respirators and masks, which might result in lower worker compliance with use as well as possible influenza A(H5N1) exposure despite wearing a mask. Hot weather and humid environments found in milking parlors can also make wearing respirators and masks uncomfortable, potentially reducing the likelihood of their use by workers carrying out farm activities, especially during hot summer months.

In this analysis, other types of masks were used more frequently than were N95 FFRs. Development of messaging by public health agencies that is consistent with CDC PPE recommendations (7) would help to educate farm owners and workers about the risks associated with caring for ill dairy cows on farms with A(H5N1) detected and could encourage recommended respirator use. Additional data are needed to guide recommendations for PPE use to protect worker health in these environments. Engaging dairy industry representatives and producers to recommend practices limiting worker contact with dairy cattle and their milk, along with PPE use recommendations in high-risk scenarios such as milking, might increase PPE use.

Limitations
The findings in this report are subject to at least two limitations. First, some PPE use before A(H5N1) outbreaks might have been in response to work conditions (e.g., wearing gloves to protect hands during work with rough materials or wearing hair or head covering to protect from sun exposure) and unrelated to A(H5N1) exposure. Second, high reported PPE use might be correlated with farm engagement with public health. Dairy farms that cooperate with public health might be more likely to request PPE for workers and participate in public health investigations. All three participating farms received PPE from public health before or immediately after the detection of A(H5N1) virus on their farms.

Implications for Public Health Practice
Public health agencies should continue to conduct outreach to farms and educate farm owners about the importance of workers using PPE during farm duties and exposure to ill cows, as well as understanding and reporting signs of human illness during A(H5N1) herd outbreaks. Collaboration with state agricultural partners can strengthen relationships and public health practice at dairy farms. As the A(H5N1) outbreak in dairy herds evolves, providing PPE to farms before outbreaks occur might help increase PPE use, especially during high-risk activities such as milking, and prevent human cases of A(H5N1).


Acknowledgments
Alexis Burakoff, Nicole Comstock, Kaitlin Harame, Ella Livesay, Shannon Matzinger, Mackenzie Owen, Alexandria Rossheim, Colorado Department of Public Health and Environment (CDPHE); STI/HIV Program, CDPHE; Colorado Department of Agriculture.

Colorado Field Team
Rachel Alade, Epidemic Intelligence Service, CDC; Marlee Barton, Colorado Department of Public Health and Environment; Cindy Camarillo, Colorado Department of Public Health and Environment; Lauren Duval, Colorado Department of Public Health and Environment; Rebecca Hermann, Colorado Department of Public Health and Environment; Frankie Lupercio, Colorado Department of Public Health and Environment; Leovi Madera, Colorado Department of Public Health and Environment; Pamela Pagano, Influenza Division, National Center for Immunization and Respiratory Diseases, CDC; Jeannette Rodriguez, Colorado Department of Public Health and Environment