Ellen Boyd
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Ellen Boyd has a total of 18 co-authors. Here are the top co-authors by number of shared publications:
- Chelsea Gardner Himsworth
University of British Columbia
2 shared publications Chelsea Himsworth
2 shared publicationsJohn Dick
Greenbelt Veterinary Services Ltd., Chilliwack, BC
2 shared publicationsAnya Fiona Smith
1 shared publicationChristine Millar
Ministry of Agriculture and Food Abbotsford BC Canada gov.bc.ca
1 shared publicationDavid L. Renaud
Department of Population Medicine, University of Guelph, Guelph ON
1 shared publication
Latest Publications
- Open Access
- 11 February 2026
- Published by Elsevier in SSRN Electronic Journal
“No Carrot, No Stick”: Exploring Premise Identification Programs Across Canada
Premises Identification (PID) programs in Canada assign unique identification numbers to locations where animals, such as livestock and poultry, are kept. These programs are provincial and territorial, with a purpose of supporting various animal disease outbreaks, natural disaster emergencies, and animal traceability. While PID programs have been operating for a number of years, there has been no study reporting on PID programs in Canada. This qualitative study explored PID programs across Canada through the perspectives of government-employed stakeholders to (1) describe PID programs in participating Canadian provinces and territories, and (2) identify the strengths and weaknesses amongst different PID programs. Stakeholders from nine provinces and one territory (N = 23) participated in virtual one-on-one interviews or focus groups. Data were analyzed inductively using thematic analysis. Findings suggest that PID programs must build trust with farmers through transparent communication of intended data usage, ideally through familiar social networks. Furthermore, compliance with PID appears to be stronger when there are robust cross-compliance measures in place that have proportional incentives and/or enforcement measures. Programs with better capacity and resources were also able to maintain data quality, which was achieved by narrowing the eligibility criteria for PID around the species of interest. This study provides valuable insights for improving the effectiveness of PID programs in Canada; future research should explore the perspectives of other stakeholders, including farmers.
- 11 February 2026
- Published by Elsevier in SSRN Electronic Journal
Open Access“No Carrot, No Stick”: Exploring Premise Identification Programs Across Canada
Premises Identification (PID) programs in Canada assign unique identification numbers to locations where animals, such as livestock and poultry, are kept. These programs are provincial and territorial, with a purpose of supporting various animal disease outbreaks, natural disaster emergencies, and animal traceability. While PID programs have been operating for a number of years, there has been no study reporting on PID programs in Canada. This qualitative study explored PID programs across Canada through the perspectives of government-employed stakeholders to (1) describe PID programs in participating Canadian provinces and territories, and (2) identify the strengths and weaknesses amongst different PID programs. Stakeholders from nine provinces and one territory (N = 23) participated in virtual one-on-one interviews or focus groups. Data were analyzed inductively using thematic analysis. Findings suggest that PID programs must build trust with farmers through transparent communication of intended data usage, ideally through familiar social networks. Furthermore, compliance with PID appears to be stronger when there are robust cross-compliance measures in place that have proportional incentives and/or enforcement measures. Programs with better capacity and resources were also able to maintain data quality, which was achieved by narrowing the eligibility criteria for PID around the species of interest. This study provides valuable insights for improving the effectiveness of PID programs in Canada; future research should explore the perspectives of other stakeholders, including farmers.
- Open Access
- 1 March 2025
- Published by American Dairy Science Association in Journal of Dairy Science
- Vol. 108 (3) , 2749-2755
Understanding Salmonella Dublin in British Columbia through bulk tank milk surveillance
Increased cases of Salmonella Dublin, a bacterial pathogen that primarily affects dairy cattle, have been noted in British Columbia (BC), Canada since 2015. The objective of this cross-sectional study was to 1) understand the prevalence and distribution of S. Dublin in BC dairy farms based on bulk tank milk (BTM) serology, 2) to investigate the degree of variability within percent positivity (%PP) in negative and positive farms, and 3) investigate risk factors associated with positivity. All BC dairy herds' BTM was sampled 4 times from September 2021 to April 2023 to determine S. Dublin positivity (at least one result ≥35% percent positivity on BTM serology). Of the 461 herds sampled 137 (30%) were positive. A multivariable logistic regression model identified 3 risk factors associated with herd level positivity. Specifically, herds with 200 to 500 lactating cattle (OR = 2.42, 95% CI = 1.14-5.36) and herds with > 500 lactating cattle (OR = 21.78, CI = 5.14-152.20) were associated with increased risk of S. Dublin herd level positivity. Farms that had greater than 5 farms within 4 km radius (OR = 1.73, 95% CI = 1.05-2.90) were at greater risk for S. Dublin compared with those with 5 or fewer. Having satisfactory pest control was protective against S. Dublin positivity (OR = 0.28, 95% CI = 0.09-0.80). Percent positivity variability varied greatly among negative, indeterminant (15-34%PP), and positive farms. Negative farms experienced very little variability in %PP (Median = 5.79%, IQR = 3.86%), with variability increasing among the indeterminant farms (Median = 16.04%, IQR = 11.94%) and highest among the positive farms (Median = 35.84%, IQR = 31.59%). Monitoring variability of BTM %PP could function as an accessible early warning tool for producers and veterinarians to predict if their herds may become positive.
- 1 March 2025
- Published by American Dairy Science Association in Journal of Dairy Science
- Vol. 108 (3) , 2749-2755
Open AccessUnderstanding Salmonella Dublin in British Columbia through bulk tank milk surveillance
Increased cases of Salmonella Dublin, a bacterial pathogen that primarily affects dairy cattle, have been noted in British Columbia (BC), Canada since 2015. The objective of this cross-sectional study was to 1) understand the prevalence and distribution of S. Dublin in BC dairy farms based on bulk tank milk (BTM) serology, 2) to investigate the degree of variability within percent positivity (%PP) in negative and positive farms, and 3) investigate risk factors associated with positivity. All BC dairy herds' BTM was sampled 4 times from September 2021 to April 2023 to determine S. Dublin positivity (at least one result ≥35% percent positivity on BTM serology). Of the 461 herds sampled 137 (30%) were positive. A multivariable logistic regression model identified 3 risk factors associated with herd level positivity. Specifically, herds with 200 to 500 lactating cattle (OR = 2.42, 95% CI = 1.14-5.36) and herds with > 500 lactating cattle (OR = 21.78, CI = 5.14-152.20) were associated with increased risk of S. Dublin herd level positivity. Farms that had greater than 5 farms within 4 km radius (OR = 1.73, 95% CI = 1.05-2.90) were at greater risk for S. Dublin compared with those with 5 or fewer. Having satisfactory pest control was protective against S. Dublin positivity (OR = 0.28, 95% CI = 0.09-0.80). Percent positivity variability varied greatly among negative, indeterminant (15-34%PP), and positive farms. Negative farms experienced very little variability in %PP (Median = 5.79%, IQR = 3.86%), with variability increasing among the indeterminant farms (Median = 16.04%, IQR = 11.94%) and highest among the positive farms (Median = 35.84%, IQR = 31.59%). Monitoring variability of BTM %PP could function as an accessible early warning tool for producers and veterinarians to predict if their herds may become positive.
- Open Access
- 1 January 2024
- Published by Wiley in Transboundary and Emerging Diseases
- Vol. 2024 (1) , 9461144
A Retrospective Analysis of Postmortem Salmonella Dublin Cases in Dairy Cattle in British Columbia
Salmonella Dublin is a bovine‐adapted bacterial pathogen that primarily affects dairy cattle. The incidence of S. Dublin has been increasing across North America, including strains that are multidrug resistant. In British Columbia, the Ministry of Agriculture’s Animal Health Center (AHC) reported an increase in cases since 2015, warranting an investigation into how S. Dublin is spreading within the province. The objectives of this study were to make use of historical data collected from dairy farms across the province to (1) describe S. Dublin cases diagnosed at the AHC between 2007 and 2021, (2) identify risk factors for S. Dublin transmission across British Columbia dairy farms, and (3) identify any potential biases associated with passive laboratory‐based data that may apply to our results. We found that S. Dublin cases diagnosed at the AHC have been increasing over time. Over half of the cases had respiratory symptoms; however, clinical signs tended to be highly variable. The prevalence of respiratory symptoms was mirrored by florfenicol treatment and was suggested to be due to using a first‐line antibiotic for more common causes of pneumonia when presented with an S. Dublin case. Calves were 38 times more likely to have S. Dublin when compared to adults (odds ratio = 38.43, confidence interval = 7.26–203.64), and given the sample population (postmortem cases), it is reasonable to conclude clinical disease is most severe in this age group. Farm premise accounted for a large amount of variability within our model (92% of unexplained variance), suggesting that farm‐level management practices may be the most important risk factor for S. Dublin infection. In total, only 54% of BC dairy farms submitted to the laboratory between 2007 and 2021; however, proximity to the laboratory did not appear to influence submissions as proportionally; farms within the Fraser Valley submitted as frequently as farms from other regions. We strongly suggest that future work explore factors associated with farm management practices, given our findings regarding the clustering by premises.
- 1 January 2024
- Published by Wiley in Transboundary and Emerging Diseases
- Vol. 2024 (1) , 9461144
Open AccessA Retrospective Analysis of Postmortem Salmonella Dublin Cases in Dairy Cattle in British Columbia
Salmonella Dublin is a bovine‐adapted bacterial pathogen that primarily affects dairy cattle. The incidence of S. Dublin has been increasing across North America, including strains that are multidrug resistant. In British Columbia, the Ministry of Agriculture’s Animal Health Center (AHC) reported an increase in cases since 2015, warranting an investigation into how S. Dublin is spreading within the province. The objectives of this study were to make use of historical data collected from dairy farms across the province to (1) describe S. Dublin cases diagnosed at the AHC between 2007 and 2021, (2) identify risk factors for S. Dublin transmission across British Columbia dairy farms, and (3) identify any potential biases associated with passive laboratory‐based data that may apply to our results. We found that S. Dublin cases diagnosed at the AHC have been increasing over time. Over half of the cases had respiratory symptoms; however, clinical signs tended to be highly variable. The prevalence of respiratory symptoms was mirrored by florfenicol treatment and was suggested to be due to using a first‐line antibiotic for more common causes of pneumonia when presented with an S. Dublin case. Calves were 38 times more likely to have S. Dublin when compared to adults (odds ratio = 38.43, confidence interval = 7.26–203.64), and given the sample population (postmortem cases), it is reasonable to conclude clinical disease is most severe in this age group. Farm premise accounted for a large amount of variability within our model (92% of unexplained variance), suggesting that farm‐level management practices may be the most important risk factor for S. Dublin infection. In total, only 54% of BC dairy farms submitted to the laboratory between 2007 and 2021; however, proximity to the laboratory did not appear to influence submissions as proportionally; farms within the Fraser Valley submitted as frequently as farms from other regions. We strongly suggest that future work explore factors associated with farm management practices, given our findings regarding the clustering by premises.
- Open Access
- 10 January 2023
- Published by MDPI AG in International Journal of Environmental Research and Public Health
- Vol. 20 (2) , 1248
Hands off the Mink! Using Environmental Sampling for SARS-CoV-2 Surveillance in American Mink
Throughout the COVID-19 pandemic, numerous non-human species were shown to be susceptible to natural infection by SARS-CoV-2, including farmed American mink. Once infected, American mink can transfer the virus from mink to human and mink to mink, resulting in a high rate of viral mutation. Therefore, outbreak surveillance on American mink farms is imperative for both mink and human health. Historically, disease surveillance on mink farms has consisted of a combination of mortality and live animal sampling; however, these methodologies have significant limitations. This study compared PCR testing of both deceased and live animal samples to environmental samples on an active outbreak premise, to determine the utility of environmental sampling. Environmental sampling mirrored trends in both deceased and live animal sampling in terms of percent positivity and appeared more sensitive in some low-prevalence instances. PCR CT values of environmental samples were significantly different from live animal samples’ CT values and were consistently high (mean CT = 36.2), likely indicating a low amount of viral RNA in the samples. There is compelling evidence in favour of environmental sampling for the purpose of disease surveillance, specifically as an early warning tool for SARS-CoV-2; however, further work is needed to ultimately determine whether environmental samples are viable sources for molecular epidemiology investigations.
- 10 January 2023
- Published by MDPI AG in International Journal of Environmental Research and Public Health
- Vol. 20 (2) , 1248
Open AccessHands off the Mink! Using Environmental Sampling for SARS-CoV-2 Surveillance in American Mink
Throughout the COVID-19 pandemic, numerous non-human species were shown to be susceptible to natural infection by SARS-CoV-2, including farmed American mink. Once infected, American mink can transfer the virus from mink to human and mink to mink, resulting in a high rate of viral mutation. Therefore, outbreak surveillance on American mink farms is imperative for both mink and human health. Historically, disease surveillance on mink farms has consisted of a combination of mortality and live animal sampling; however, these methodologies have significant limitations. This study compared PCR testing of both deceased and live animal samples to environmental samples on an active outbreak premise, to determine the utility of environmental sampling. Environmental sampling mirrored trends in both deceased and live animal sampling in terms of percent positivity and appeared more sensitive in some low-prevalence instances. PCR CT values of environmental samples were significantly different from live animal samples’ CT values and were consistently high (mean CT = 36.2), likely indicating a low amount of viral RNA in the samples. There is compelling evidence in favour of environmental sampling for the purpose of disease surveillance, specifically as an early warning tool for SARS-CoV-2; however, further work is needed to ultimately determine whether environmental samples are viable sources for molecular epidemiology investigations.