Adaptive conjoint analysis to determine perceived risk factors of farmers, veterinarians and AI technicians for introduction of BHV1 to dairy farms

A study was carried out to determine the possibility of a more-closed farming system for (Dutch) dairy farms. The objective of the study was to provide effective and economically profitable management advice for improving the animal-health status of farms. Management measures will only be successfully applied if supported by farmers and their advisors (such as veterinarians). Therefore, the perception of farmers and advisors of the importance of various risk factors for the introduction of diseases to a farm was determined by using bovine herpes virus type 1 (BHV1) as an example.

As part of the study, an evening-long workshop was organized and run thrice. In total, 49 farmers, veterinarians and AI technicians participated in these workshops. The computerized questionnaire technique was based on adaptive conjoint analysis (ACA). ACA has the advantage that participants can work with a large number of risk factors in a relatively short period of time. Another advantage of ACA (compared with standard questionnaires) is that the answers from each participant can be checked with regard to consistency with respect to the importance assigned to them. Data from participants with inconsistent responses can be excluded from further analyses. The results of the ACA interview were compared with the risk factors reported in the literature as being associated with BHV1 status (e.g. purchase of cattle, participation in cattle shows) and with farmers' actual management to prevent the introduction of diseases.

The workshop participants were all operating in the dairy sector and they seemed well aware of the risk of direct animal contacts for the introduction of BHV1. Farmers thought visitors to be more risky than did AI technicians and (especially) veterinarians. Farmers who purchased cattle or participated in cattle shows were of the opinion that the risks of direct animal contacts were more important than did farmers who were not involved in those practices. Farmers whose farms were BHV1-positive (and participated in cattle shows more often) thought the risk of participation smaller than did farmers with BHV1-negative farms.     

Relationship between herd-level incidence rate of energy-related postpartum diseases,general risk factors and claw lesions in individual dairy cows recorded at maintenance claw trimming

Background

Laminitis and energy-related postpartum diseases share several risk factors, indicating a common etiology. Thus, a herd’s incidence rate of energy-related postpartum diseases, such as displaced abomasum and clinical ketosis, might reflect the likelihood of cows to suffer from laminitis-related claw lesions. The aim of this study was to investigate the association between herd-level incidence rate of displaced abomasum and clinical ketosis, general risk factors, and claw lesions in individual cows recorded at maintenance claw trimming.

Methods

The dataset consisted of 6773 trimmings, performed between 2004 and 2006 by professional trimmers, from 3607 Swedish Red and Swedish Holstein cows in 26 herds. The herds were classified as having a high, inconsistent-high or low incidence rate of energy-related postpartum diseases, based on the number of recorded cases of veterinary-diagnosed displaced abomasum and clinical ketosis in the Swedish national animal disease recording system during 2002 to 2006, and observations and interviews in connections with herd visits. Generalized linear mixed models were used to investigate the association between herd-level incidence rate of energy-related postpartum diseases and laminitis-related lesions including sole ulcer and sole hemorrhage; and hygiene-related lesions including interdigital dermatitis, digital dermatitis, heel-horn erosion, verrucose dermatitis, and interdigital hyperplasia; and absence of any claw lesion. Systematic effects, including first-order interactions, with P < 0.05 were included in the models. Herd classification was forced into the models, and a random effect of herd was included.

Results

In comparison to herds with a high incidence rate of energy-related postpartum diseases, low-incidence herds showed a lower odds ratio (OR; 0.2) for laminitis-related lesions in cows trimmed during the summer months. Low-incidence herds also showed numerically lower OR estimates for laminitis-related lesions in all parity classes and a numerically lower OR for hygiene-related lesions. In addition, low-incidence herds showed tendencies towards a numerically higher OR for absence of any lesion, irrespective of trimming season or parity.

Conclusions

Only a few statistically significant associations were found, but several tendencies pointed towards better claw health in herds with low as compared with high incidence rate of energy-related postpartum diseases.     

Herd-level risk factors associated with Leptospira Hardjo seroprevalence in Beef/Suckler herds in the Republic of Ireland

Background

The aim of the present study was to investigate risk factors for herd seropositivity to Leptospira Hardjo in Irish suckler herds. Herds were considered eligible for the study if they were unvaccinated and contained ≥ 9 breeding animals of beef breed which were ≥ 12 months of age. The country was divided into six regions using county boundaries. Herd and individual animal prevalence data were available from the results of a concurrent seroprevalence study. Herds were classified as either "Free from Infection" or "Infected" based on a minimum expected 40% within-herd prevalence.Questionnaires were posted to 320 farmers chosen randomly from 6 regions, encompassing 25 counties, of the Republic of Ireland. The questionnaire was designed to obtain information about vaccination; reproductive disease; breeding herd details; the presence of recognized risk factors from previous studies; and husbandry on each farm. Data collected from 128 eligible herds were subjected to statistical analysis.

Results

Following the use of Pearson''s Chi-Square Test, those variables associated with a herd being "infected" with a significance level of P < 0.2 were considered as candidates for multivariable logistic regression modelling. Breeding herd size was found to be a statistically significant risk factor after multivariable logistic regression. The odds of a herd being positive for leptospiral infection were 5.47 times higher (P = 0.032) in herds with 14 to 23 breeding animals compared with herds with ≤ 13 breeding animals, adjusting for Region, and 7.08 times higher (P = 0.033) in herds with 32.6 to 142 breeding animals.

Conclusions

Breeding herd size was identified as a significant risk factor for leptospiral infection in Irish suckler herds, which was similar to findings of previous studies of leptospirosis in dairy herds.     

The role of epidemiology in the prevention, diagnosis, and control of infectious diseases of fish

Epidemiologic methods are essential to understanding infectious diseases in aquaculture. Unfortunately, many of these methods are poorly understood or not utilized by fish-health scientists and aquaculturists — often because of the lack of contact with epidemiologists who are willing to investigate fish diseases. In this paper, we describe direct interactions between epidemiologists and fish-health specialists that have resulted in an improved understanding of the causes and management of infectious diseases in aquaculture. We focus on risk-factor studies, risk analysis and infectious-disease modeling, evaluation of diagnostic tests and experimental studies. We also describe characteristics of confined fish populations that make them ideal for developing and testing epidemiologic models and the theoretical and practical challenges of designing and conducting epidemiologic studies in fish farms. Throughout our presentation, emphasis is given to characteristics, opportunities and problems associated mainly with conducting epidemiologic studies to intensive aquaculture systems. We conclude that the development of increased cooperation among epidemiologists, fish-health scientists and aquaculturists will be mutually beneficial and, therefore, efforts for such collaboration should be initiated from all parties involved.     

Analysis of risk factors associated with bovine leukemia virus seropositivity within dairy and beef breeding farms in Japan: A nationwide survey

This cross-sectional study evaluated risk factors associated with farm-level bovine leukemia virus (BLV) seropositivity in 563 dairy and 490 beef farms throughout Japan. Twenty randomly selected cattle on each farm were serologically tested, and farm epidemiologiocal information was obtained through face-to-face interviews. Due to the large number of zero-prevalence dairy and beef farms, data analysis was performed using a zero-inflated negative binomial model, which revealed that the common risk factors associated with higher within-farm seroprevalence were past detection of clinical leukemia and presence of blood-sucking insects. Loose housing on dairy farms and direct contact between calves and adult cattle on beef farms were also identified as risk factors. With regard to farm-level presence of BLV, the presence of purchased cattle was found to be a risk factor in both sectors. Sending heifers to a common ranch was identified as an additional risk factor for dairy farms.     

Factor analysis of a Johne's disease risk assessment questionnaire with evaluation of factor scores and a subset of original questions as predictors of observed clinical paratuberculosis

Factor analysis was used to examine the interrelationships among 38 variables collected as part of a Johne's disease risk assessment questionnaire completed in 2002 on 815 U.S. dairy operations. Eleven factors were extracted, accounting for two-thirds of the variance encountered in the original variables. Responses to many of the risk assessment questions were closely related. Standardized scores on the 11 factors were calculated for operations providing complete information, and were evaluated as predictors in a model-based logistic regression analysis with the outcome being whether operations had observed one or more cows with clinical signs suggestive of paratuberculosis during the previous year. A logistic regression model was also used to evaluate the predictive ability of a reduced subset of approximately one-third of the original variables that was selected to represent the derived factors. The performance of both sets of predictors was comparable with respect to goodness-of-fit and predictive ability. In conclusion, the length of the current risk assessment instrument could be reduced considerably without a substantial loss of information by removing or combining questions that are strongly correlated.     

Quantifying risk factors of coccidiosis in broilers using on-farm data based on a veterinary practice

A study was done to find and quantify risk factors for coccidiosis. The study population consisted of 4774 broiler flocks kept on 177 farms. Flocks were considered a case when at least one bird in the flock showed microscopic presence of oocysts in intestinal scrapings in a grow-out cycle. Other flocks were defined as controls. This was done for three types of Eimeria: Eimeria acervulina, Eimeria tenella and Eimeria maxima. Logistic regression was used to assess variables that influence the occurrence of Eimeria species. There were 49 variables, based on animal, flock or farm level. There was an enhanced risk of coccidiosis due to environmental and management factors that increase the risk of introducing contamination or that are related to hygienic measures. These include lack of use of overalls by visitors, a farmyard which is difficult to clean, bad hygienic status, personnel who might also be working on other farms, presence of other animals on the farm, and feeding and drinking systems which are more difficult to clean. Also, the presence of other diseases on the farm and Eimeria species found in the previous flock increased the risk: of coccidiosis.     

Canine echinococcosis in northwest Libya: assessment of coproantigen ELISA, and a survey of infection with analysis of risk-factors

In order to determine the prevalence and risk factors for canine echinococcosis in different endemic localities in the Tripoli area of northwest Libya, stray dogs were examined post-mortem, and owned dogs screened for Echinococcus granulosus infection using a standardised genus specific coproantigen ELISA. The prevalence of E. granulosus infection at necropsy in stray-dogs was 25.8% (15/58, 95% CI 15.3–39.0%), and 21.6% (72/334, 95% CI 17.3–26.4%) of owned dogs tested were positive by coproantigen ELISA. Sheepdogs appeared to have a significantly higher copro-positive prevalence (19/19 positive, p = 0.003), compared to 23.6% of other dog classes (e.g. 52/220 guard dogs and household pets). Worm burdens in necropsied dogs ranged from 29 to 2900 (mean 1064) and were positively correlated to coproantigen ELISA OD values (r_s = 0.87, p < 0.001), but negatively correlated with dog age (r_s = −0.69, p = 0.001). Dog age was a significant factor in copro-prevalence as there was an increasing coproantigen-positive tendency in younger dogs (≤5 years, p = 0.04). A total of 45/132 (34%, 95% CI 25.9–42.1%) of farms/homestead had at least one dog that was coproantigen positive. Overall copro-prevalence in dogs by locality varied, with Alkhums (Leptis-Magna) district having the highest copro-prevalence at 38.7% (24/62, 95% CI 26.6–50.8%) (p = 0.001). Coproantigen testing of a cohort of owned dogs before and approximately 15 months after praziquantel treatment showed a significant decrease in the coproantigen positive rate from 21.6% (72/334) to 9% (21/233) post-treatment. The overall E. granulosus coproantigen positive rate (‘re-infection rate’) within the same cohort of dogs was 22 % (10/45) by 15 months post-treatment. Significant risk factors for a copro-positive owned dog were associated with non-restraint of dogs, and owners that did not de-worm their dogs. Home slaughtering of livestock and lack of knowledge about E. granulosus transmission were also significant risk factors for a canine coproantigen positive result.     

Molecular epidemiology of bovine tuberculosis in wild animals in Spain: a first approach to risk factor analysis

In human tuberculosis (Mycobacterium tuberculosis), molecular epidemiology has accurately indicated the risk factors involved in active transmission of the disease, by comparing individuals whose isolates belong to a cluster with patients whose strains are considered unique. Nevertheless, this application has not been used in bovine tuberculosis (Mycobacterium bovis). Our study describes the integration of epidemiological data into molecular classification data on M. bovis isolates. These were isolated from wild ungulates in Extremadura (western Spain) with the objective of detecting the risk factors linked to the association of strains in clades, which are indicators of the active spread of the disease. The molecular markers used were spoligotyping + VNTR typing (loci: VNTR 2165, VNTR 2461, VNTR 0577, VNTR 0580, VNTR 3192 VNTR 2163a and VNTR 2163b) on a population of 59 M. bovis strains isolated from deer (Cervus elaphus), 112 from wild boar (Sus scrofa), six from bovines, 28 from pigs and 2 from goats (n = 207). Epidemiological variables included the animal species from which the strain was isolated, pathological condition of the host (incipient lesion, early and late generalisation), date of sampling (during or after the reproductive period) and hunting season. Bivariant analysis was used to establish the risk factors connected to the association of strains and later, the variables were evaluated by means of logistic regression. Molecular typing grouped a total of 131 strains (64.21%) in 28 clusters and 76 isolates shows unique profiles. The association of strains was connected to the appearance of macroscopic lesions during the reproductive period (O.R. 4.80; 95% CI 1.09–22.99, P < 0.005), showing a possible higher transmission during the courting period. This happened mainly during the last hunting season analysed (2002–2003, O.R. 3.69; 95% CI 1.27–11.9, P < 0.05), clashing with the time of higher prevalence of the disease in wild ungulates. Active spread was not connected to any species in particular, or to any concrete pathological condition.     

Diseases in neonatal foals. Part 2: Potential risk factors for a higher incidence of infectious diseases during the first 30 days post partum

Reason for performing study: The development of clinical illness in foals is usually predetermined by perinatal history, management or stressful environmental conditions. Objectives: To determine potential risk factors for an increased incidence of infectious diseases during the first 30 days post partum. Methods: The population consisted of Thoroughbred foals born on stud farms in the Newmarket (UK) area in 2005 (n = 1031). They were followed for their first 30 days. Factors suspected to influence the incidence of infectious neonatal diseases were examined in a logistic regression approach for each of the 3 outcomes (total infectious diseases, systemic disease with diarrhoea and total infectious diseases excluding diarrhoea). All 28 factors were either foal or mare or stud farm related. Results: Several significant risk factors for a higher disease incidence, such as birth complications, colostrum intake by stomach tube and leucocytosis 12–48 h post partum were identified. The factor ‘boarding stud’ seemed to be protective against disease. Conclusion: Some factors, such as the mare's time at stud before foaling, the mare's rotavirus vaccination schedule and fibrinogen‐values that empirically had been linked to the outcome previously were not confirmed as relevant. This included the reported useful prophylactic treatment with antimicrobial drugs. Potential relevance: Factors to be considered when evaluating newborn foals include: stud management, the birth process, route of colostrum intake, white and red blood cells, and the date of birth. These may help to detect foals at risk to develop an infection so that targeted prophylactic measures can be initiated.     

Potential value of the common vetch (Vicia sativa L.) as an animal feedstuff: a review

The objective of this review was to systematically evaluate common vetch seeds as a potential feedstuff for animals, by summarizing and discussing the available published literature covering their nutritional composition as well as their content of antinutritional factors and potential techniques for their reduction. In addition, animal feeding studies that have investigated the effect of inclusion of common vetch seeds on animal growth and performance were identified and evaluated to stimulate interest in their use as a good source of nutrients for inclusion in animal diets. The collective literature shows that common vetch seeds are a less costly (in comparison with alternatives) and rich source of protein and minerals for farmed animals, are of high digestibility and have a high energy content, and can be used to partially or totally replace soya bean meal and/or to replace a large proportion of cereals in the diet. Furthermore, the literature shows that common vetch seeds contain a range of antinutritional factors which, if they are to be utilized in non‐ruminant diets and to increase their utilizing efficiency, need to be removed or inactivated. This can be achieved via certain pre‐processing methods, the combination of which may deliver better results.     

Risk factors for bovine respiratory disease in Australian feedlot cattle: Use of a causal diagram-informed approach to estimate effects of animal mixing and movements before feedlot entry

A nationwide longitudinal study was conducted to investigate risk factors for bovine respiratory disease (BRD) in cattle in Australian feedlots. After induction (processing), cattle were placed in feedlot pens (cohorts) and monitored for occurrence of BRD over the first 50 days on feed. Data from a national cattle movement database were used to derive variables describing mixing of animals with cattle from other farms, numbers of animals in groups before arrival at the feedlot, exposure of animals to saleyards before arrival at the feedlot, and the timing and duration of the animal's move to the vicinity of the feedlot. Total and direct effects for each risk factor were estimated using a causal diagram-informed process to determine covariates to include in four-level Bayesian logistic regression models. Mixing, group size and timing of the animal's move to the feedlot were important predictors of BRD. Animals not mixed with cattle from other farms prior to 12 days before induction and then exposed to a high level of mixing (≥4 groups of animals mixed) had the highest risk of developing BRD (OR 3.7) compared to animals mixed at least 4 weeks before induction with less than 4 groups forming the cohort. Animals in groups formed at least 13 days before induction comprising 100 or more (OR 0.5) or 50–99 (OR 0.8) were at reduced risk compared to those in groups of less than 50 cattle. Animals moved to the vicinity of the feedlot at least 27 days before induction were at reduced risk (OR 0.4) compared to cattle undergoing short-haul transportation (<6 h) to the feedlot within a day of induction, while those experiencing longer transportation durations (6 h or more) within a day of induction were at slightly increased risk (OR 1.2). Knowledge of these risk factors could potentially be used to inform management decisions to reduce the risk of BRD in feedlot cattle.     

Associations between the proportion of Salmonella seropositive slaughter pigs and the presence of herd level risk factors for introduction and transmission of Salmonella in 34 Danish organic, outdoor (non-organic) and indoor finishing-pig farms

This paper evaluates the association between herd level risk factors for introduction and transmission of Salmonella in farms with three different production systems: organic, outdoor (non-organic) and indoor finishing-pig farms, and the presence of seropositive animals in the herds. Potential risk factors for Salmonella in the three pig production systems were identified through a literature review, and management information as well as serological data were collected in 34 pig farms: 11 organic farms, 12 outdoor farms, and 11 indoor farms. There were no general differences in the proportion of Salmonella seropositive animals in the organic, outdoor, and indoor pig farms. Correspondence analysis showed that the occurrence of seropositive animals in the herds was mostly associated to the risk of introducing Salmonella in the herds by purchasing and transporting growing pigs. No associations between herd risk factors for transmission and survival of Salmonella and seropositive animals in the herds were found.     

Building a picture: Prioritisation of exotic diseases for the pig industry in Australia using multi-criteria decision analysis

Diseases that are exotic to the pig industry in Australia were prioritised using a multi-criteria decision analysis framework that incorporated weights of importance for a range of criteria important to industry stakeholders. Measurements were collected for each disease for nine criteria that described potential disease impacts. A total score was calculated for each disease using a weighted sum value function that aggregated the nine disease criterion measurements and weights of importance for the criteria that were previously elicited from two groups of industry stakeholders. One stakeholder group placed most value on the impacts of disease on livestock, and one group placed more value on the zoonotic impacts of diseases. Prioritisation lists ordered by disease score were produced for both of these groups. Vesicular diseases were found to have the highest priority for the group valuing disease impacts on livestock, followed by acute forms of African and classical swine fever, then highly pathogenic porcine reproductive and respiratory syndrome. The group who valued zoonotic disease impacts prioritised rabies, followed by Japanese encephalitis, Eastern equine encephalitis and Nipah virus, interspersed with vesicular diseases. The multi-criteria framework used in this study systematically prioritised diseases using a multi-attribute theory based technique that provided transparency and repeatability in the process. Flexibility of the framework was demonstrated by aggregating the criterion weights from more than one stakeholder group with the disease measurements for the criteria. This technique allowed industry stakeholders to be active in resource allocation for their industry without the need to be disease experts. We believe it is the first prioritisation of livestock diseases using values provided by industry stakeholders. The prioritisation lists will be used by industry stakeholders to identify diseases for further risk analysis and disease spread modelling to understand biosecurity risks to this industry.     

Epidemiological studies of pig diseases: 1. Use of French protocols for risk factor assessment to predict the health status of Australian herds

Objective To determine in Australian pig herds the accuracy of French protocols for risk factor assessment.
Procedure Data on health indicators and risk factors were collected for three syndromes, 'pre-weaning diarrhoea', 'post-weaning diarrhoea' and 'respiratory problems', using the French protocols. The protocols were used on 118 occasions in 32 Western Australian pig herds during 3 years (1988 to 1991).
Results There was a wide variation in pre-weaning performance, for example growth rate was 107 to 273 g/day (< 200 g/day in 33% of herds). Respiratory lesions at weaning were associated with poor pre-weaning performance. Post-weaning (21 days after weaning) growth rate was 114 to 408 g/day (< 250 g/day in 54% of herds). In the grower herds, 91% of herds had pneumonia, and growth rate was 439 to 625 g/day (< 550 g/day in 54% of herds). Pleurisy as well as pneumonia was associated with reduced growth rate. The risk factor most closely associated with respiratory health status was air volume per pig.
Conclusion Risk factors were most accurate at predicting the health status in post-weaning problems. A weaning weight of at least 7.9 kg and weaning age of 30 days optimised weaner performance. Stocking densities and shed designs providing at least 3 m³ air volume and 0.6 m² floor space per pig throughout the growing phase should be considered for an improved respiratory health status. Australian pig sheds often do not provide a satisfactory environment for optimum health. The technique of risk factor assessment as an aid to the maintenance of health in pig herds is applicable in Australia, but further research is necessary to determine the most important Australian risk factors.     

Knowledge,attitudes, and practices (KAP) related to brucellosis and factors affecting knowledge sharing on animal diseases: a cross-sectional survey in the dry zone of Sri Lanka

Farmers’ lack of knowledge is assumed to have affected the presence of brucellosis in Sri Lanka for decades. This study, carried out in the Ampara district in the dry zone of Sri Lanka, revealed that there is a significant knowledge gap for brucellosis compared to foot and mouth disease (FMD) (p?<?0.001). Only 8.3% of farmers knew that brucellosis causes cattle abortions. Only 2.6% knew that it is zoonotic. The difference in knowledge of the symptoms and transmission of brucellosis and FMD was significant (p?<?0.001). Farmers’ attitudes and practices related to the spread of the disease were poor. Farmers’ education and spoken language had a negative influence on knowledge. Young people and those with strong social relationships were efficient in knowledge sharing. It can be concluded that brucellosis knowledge, attitudes, and practices are poor; thus, there is a need for more attention in disease control policymaking. Backward farmer groups should be the focus in animal health extension programs.     

Multivariate relationships between epidemiologic risk factors and zoonotic infections among military personnel in the country of Georgia: A non‐linear canonical correlation analysis

Zoonotic diseases are endemic in the country of Georgia. Using the non‐linear canonical correlation (NCC) method, the aim of this study was to examine the relationship between thirteen epidemiological risk factors and seropositivity to five zoonotic infections (anthrax, Q fever, tularemia, leptospirosis, and Crimean‐Congo hemorrhagic fever [CCHF]) among Georgian military recruits during 2014–2016. According to this multivariate statistical technique, which is suitable for the analysis of two or more sets of qualitative variables simultaneously, two canonical variables were identified. These variables accounted for 68% of the variation between the two sets of categorical variables (“risk factors” and “zoonotic infections”). For the first canonical variable, there was a relationship among CCHF (canonical loading, which is interpreted in the same way as the Pearson's correlation coefficient, [cl] = 0.715), tick bites (cl = 0.418) and slaughter of animals (cl = 0.351). As for the second canonical variable, Q fever (cl = ?0.604) and leptospirosis (cl = ?0.486) were related to rodents inside and outside home (cl = ?0.346) and sweeping in or around home (cl = ?0.317). The NCC method allows researchers to obtain additional insights into the complex relationship between epidemiological risk factors and multiple zoonotic infections.     

Anthropogenic factors and the risk of highly pathogenic avian influenza H5N1: prospects from a spatial-based model

Beginning in 2003, highly pathogenic avian influenza (HPAI) H5N1 virus spread across Southeast Asia, causing unprecedented epidemics. Thailand was massively infected in 2004 and 2005 and continues today to experience sporadic outbreaks. While research findings suggest that the spread of HPAI H5N1 is influenced primarily by trade patterns, identifying the anthropogenic risk factors involved remains a challenge. In this study, we investigated which anthropogenic factors played a role in the risk of HPAI in Thailand using outbreak data from the “second wave” of the epidemic (3 July 2004 to 5 May 2005) in the country. We first performed a spatial analysis of the relative risk of HPAI H5N1 at the subdistrict level based on a hierarchical Bayesian model. We observed a strong spatial heterogeneity of the relative risk. We then tested a set of potential risk factors in a multivariable linear model. The results confirmed the role of free-grazing ducks and rice-cropping intensity but showed a weak association with fighting cock density. The results also revealed a set of anthropogenic factors significantly linked with the risk of HPAI. High risk was associated strongly with densely populated areas, short distances to a highway junction, and short distances to large cities. These findings highlight a new explanatory pattern for the risk of HPAI and indicate that, in addition to agro-environmental factors, anthropogenic factors play an important role in the spread of H5N1. To limit the spread of future outbreaks, efforts to control the movement of poultry products must be sustained.