Highly pathogenic avian influenza H5N8 in south-west France 2016-2017: A modeling study of control strategies.

In the winter 2016-2017 the largest epidemic of highly pathogenic avian influenza (HPAI) ever recorded in the European Union spread to all 28 member states. France was hit particularly hard and reported a total of 484 infected premises (IPs) by March 2017. We developed a mathematical model to analyze the spatiotemporal evolution of the epidemic and evaluate the impact of control strategies. We estimated that farms rearing ducks were on average 2.5 times more infectious and 5.0 times more susceptible to HPAI than farms rearing other avian species. The implementation of surveillance zones around IPs reduced transmission by a factor of 1.8 on average. Compared to the strengthening of pre-emptive culling measures enforced by French authorities in February 2017, we found that a faster depopulation of diagnosed IPs would have had a larger impact on the total number of infections. For example, halving the time delay from detection to slaughter of infected animals would have reduced the total number of IPs by 52% and total cull numbers by 50% on average. This study showcases the possible contribution of modeling to inform and optimize control strategies during an outbreak.

Impact of Imperfect Disease Detection on the Identification of Risk Factors in Veterinary Epidemiology.

Risk factors are key epidemiological concepts that are used to explain disease distributions. Identifying disease risk factors is generally done by comparing the characteristics of diseased and non-diseased populations. However, imperfect disease detectability generates disease observations that do not necessarily represent accurately the true disease situation. In this study, we conducted an extensive simulation exercise to emphasize the impact of imperfect disease detection on the outcomes of logistic models when case reports are aggregated at a larger scale (e.g., diseased animals aggregated at farm level). We used a probabilistic framework to simulate both the disease distribution in herds and imperfect detectability of the infected animals in these herds. These simulations show that, under logistic models, true herd-level risk factors are generally correctly identified but their associated odds ratio are heavily underestimated as soon as the sensitivity of the detection is less than one. If the detectability of infected animals is not only imperfect but also heterogeneous between herds, the variables associated with the detection heterogeneity are likely to be incorrectly identified as risk factors. This probability of type I error increases with increasing heterogeneity of the detectability, and with decreasing sensitivity. Finally, the simulations highlighted that, when count data is available (e.g., number of infected animals in herds), they should not be reduced to a presence/absence dataset at the herd level (e.g., presence or not of at least one infected animal) but rather modeled directly using zero-inflated count models which are shown to be much less sensitive to imperfect detectability issues. In light of these simulations, we revisited the analysis of the French bovine abortion surveillance data, which has already been shown to be characterized by imperfect and heterogeneous abortion detectability. As expected, we found substantial differences between the quantitative outputs of the logistic model and those of the zero-inflated Poisson model. We conclude by strongly recommending that efforts should be made to account for, or at the very least discuss, imperfect disease detectability when assessing associations between putative risk factors and observed disease distributions, and advocate the use of zero-inflated count models if count data is available.

Swab cloths as a tool for revealing environmental contamination by Q fever in ruminant farms.

Q fever is a zoonotic abortive disease of ruminants mostly transmitted by inhalation of aerosols contaminated by Coxiella burnetii. Clusters of cases or even epidemics regularly occur in humans but, to date, there is no consensus about the best way to carry out outbreak investigations in order to identify potential farms at risk. Although environmental samples might be useful during such investigations, there are few baseline data on the presence of C. burnetii in the environment of ruminant farms. We thus investigated dust samples from cattle, sheep and goat farm buildings in order to (a) estimate C. burnetii detection frequency and bacterial loads in the environment, and (b) determine whether this environmental contamination is associated with series of abortions attributed to Q fever. We considered 113 herds with a recent abortive episode potentially related (n = 60) or not (n = 53) to C. burnetii. Dust was sampled using a swab cloth and tested by a quantitative PCR method targeting the IS1111 gene. Coxiella burnetii DNA was detected on 9 of 50 cattle farms, 13 of 19 goat farms and 30 of 40 sheep farms. On 16 cloths, bacterial loads were higher than 108 genome equivalents, levels as high as in infectious materials such as placentas and aborted foetuses. Overall, the probability of detecting C. burnetii DNA was higher on small ruminant farms than cattle farms, in herds suspected of Q fever and in large herds. We conclude that swab cloths are a putative indicator of contamination of ruminant farms by C. burnetii.

Role of Backyard Flocks in Transmission Dynamics of Highly Pathogenic Avian Influenza A(H5N8) Clade 2.3.4.4, France, 2016-2017.

Highly pathogenic avian influenza A(H5N8) clade 2.3.4.4 spread in France during 2016-2017. We assessed the biosecurity and avian influenza virus infection status of 70 backyard flocks near H5N8-infected commercial farms. One flock was seropositive for clade 2.3.4.4. Backyard flocks linked to commercial farms had elevated risk for H5 infection.

Spatio-temporal patterns of highly pathogenic avian influenza virus subtype H5N8 spread, France, 2016 to 2017.

IntroductionFrance is one of Europe's foremost poultry producers and the world's fifth largest producer of poultry meat. In November 2016, highly pathogenic avian influenza (HPAI) virus subtype H5N8 emerged in poultry in the country. As of 23 March 2017, a total of 484 confirmed outbreaks were reported, with consequences on animal health and socio-economic impacts for producers. Methods: We examined the spatio-temporal distribution of outbreaks that occurred in France between November 2016 and March 2017, using the space-time K-function and space-time permutation model of the scan statistic test. Results: Most outbreaks affected duck flocks in south-west France. A significant space-time interaction of outbreaks was present at the beginning of the epidemic within a window of 8 km and 13 days. This interaction disappeared towards the epidemic end. Five spatio-temporal outbreak clusters were identified in the main poultry producing areas, moving sequentially from east to west. The average spread rate of the epidemic front wave was estimated to be 5.5 km/week. It increased from February 2017 and was negatively associated with the duck holding density. Conclusion: HPAI-H5N8 infections varied over time and space in France. Intense transmission events occurred at the early stages of the epidemic, followed by long-range jumps in the disease spread towards its end. Findings support strict control strategies in poultry production as well as the maintenance of high biosecurity standards for poultry holdings. Factors and mechanisms driving HPAI spread need to be further investigated.

A tool to support the identification of suspect cases of exotic diseases in cattle.

Maintaining vigilance with regard to the introduction of exotic diseases is a challenge, particularly because these diseases are numerous, some are not well known, and they are not immediately suspected by people in day-to-day practice, specifically veterinary practitioners. The objective of this article is to present a tool to support the identification of suspect cases of exotic diseases in cattle, based on a Bayesian approach. A list of 22 exotic diseases in mainland France was selected mainly on the basis of their potential consequences if introduced, and the ability to detect them on a clinical basis. In response of a set of epidemio-clinical criteria observed in the field this tool provides a list of exotic diseases by descending order of likelihood. The tool's performance was assessed by simulation. When simulating epidemio-clinical observations of each of the 22 diseases included in the tool with some uncertainty, the right disease was ranked in the first place between 83.8% and 100% of the times, and always in the five most likely diseases. Even when some noise was introduced in the epidemio-clinical observations simulated by addition of criteria non-characteristic of the simulated diseases, the right disease was always in the five most likely diseases. This tool could be usefully included in a global approach aiming to improve vigilance against exotic diseases.

Iso-population partition: An innovative epidemiological approach to mapping and analyzing spatially aggregated data.

In epidemiology, data are often aggregated using administrative boundaries or regular spatial lattices. Iso-population partitioning methods allow the aggregation of small units for which population data are available into larger units that are contiguous, as compact as possible, and have a similar population size. The objective of this paper was to study the influence of three spatial data aggregation approaches on data visualization and data analysis: iso-populated units (IPUs), administrative units, and iso-geometric units. This study was conducted using results and simulations from the brucellosis clinical surveillance system for dairy cattle in France. Our findings indicate that using spatial partitioning methods for generating IPUs enhances the ability to interpret the spatial distribution of epidemiological indicators under study. In addition, it provides information on population density and improves the consistency of the power of statistical tests across units. By defining the target population size per spatial unit, IPUs can be used to control the statistical power of a study. Finally, by adding criteria based on environmental factors to generate spatial units, they can be used to control the variation of exposure to these factors within the units.

Respective influence of veterinarians and local institutional stakeholders on the event-driven surveillance system for bovine brucellosis in France.

BACKGROUND: The event-driven surveillance system for bovine brucellosis implemented in France aims to ensure the early detection of cases of bovine brucellosis, a disease of which the country has been declared free since 2005. It consists of mandatory notification of bovine abortions by farmers and veterinarians. However, as underlined by a previous qualitative study, several factors influence the decision-making process of actors in the field. This process is particularly influenced by the level of cooperation between institutional stakeholders in their département (a French département being an administrative and territorial unit), veterinarians and farmers. In this context, the objectives of this study were 1) to quantify the respective influence of veterinarians and all local institutional stakeholders on the proportion of notifying farmers and identify which actors have most influence on farmers' decisions; 2) to analyse whether the influence of veterinarians is correlated with that of local institutional stakeholders. RESULTS: In addition to factors relating to the farm itself (production type and herd size), the proportion of notifying farmers was influenced by the number of veterinarians per practice and the veterinary practice's membership of a technical association. This proportion was also influenced by unknown factors relating to the veterinary practice and, to a lesser extent, the département in which the farm was located. There was no correlation between variability in the proportion of notifying farmers among veterinary practices per département and the effect of the département itself. CONCLUSION: To our knowledge, this is the first study to quantify the influence of veterinarians and local institutional stakeholders on the notification process for a mandatory disease. In addition to carrying out regulatory interventions, veterinarians play a major role in encouraging farmers to participate in the surveillance systems. The results of this study, combined with a previous qualitative study, shed light on the need to consolidate the involvement of veterinarians and local stakeholders in the organisation of surveillance by national institutional bodies.

An optimal cut-off point for the calving interval may be used as an indicator of bovine abortions.

The bovine abortion surveillance system in France aims to detect as early as possible any resurgence of bovine brucellosis, a disease of which the country has been declared free since 2005. It relies on the mandatory notification and testing of each aborting cow, but under-reporting is high. This research uses a new and simple approach which considers the calving interval (CI) as a "diagnostic test" to determine optimal cut-off point c and estimate diagnostic performance of the CI to identify aborting cows, and herds with multiple abortions (i.e. three or more aborting cows per calving season). The period between two artificial inseminations (AI) was considered as a "gold standard". During the 2006-2010 calving seasons, the mean optimal CI cut-off point for identifying aborting cows was 691 days for dairy cows and 703 days for beef cows. Depending on the calving season, production type and scale at which c was computed (individual or herd), the average sensitivity of the CI varied from 42.6% to 64.4%; its average specificity from 96.7% to 99.7%; its average positive predictive value from 27.6% to 65.4%; and its average negative predictive value from 98.7% to 99.8%. When applied to the French bovine population as a whole, this indicator identified 2-3% of cows suspected to have aborted, and 10-15% of herds suspected of multiple abortions. The optimal cut-off point and CI performance were consistent over calving seasons. By applying an optimal CI cut-off point to the cattle demographics database, it becomes possible to identify herds with multiple abortions, carry out retrospective investigations to find the cause of these abortions and monitor a posteriori compliance of farmers with their obligation to report abortions for brucellosis surveillance needs. Therefore, the CI could be used as an indicator of abortions to help improve the current mandatory notification surveillance system.

Quantitative and qualitative assessment of the bovine abortion surveillance system in France.

Bovine abortion is the main clinical sign of bovine brucellosis, a disease of which France has been declared officially free since 2005. To ensure the early detection of any brucellosis outbreak, event-driven surveillance relies on the mandatory notification of bovine abortions and the brucellosis testing of aborting cows. However, the under-reporting of abortions appears frequent. Our objectives were to assess the aptitude of the bovine abortion surveillance system to detect each and every bovine abortion and to identify factors influencing the system's effectiveness. We evaluated five attributes defined by the U.S. Centers for Disease Control with a method suited to each attribute: (1) data quality was studied quantitatively and qualitatively, as this factor considerably influences data analysis and results; (2) sensitivity and representativeness were estimated using a unilist capture-recapture approach to quantify the surveillance system's effectiveness; (3) acceptability and simplicity were studied through qualitative interviews of actors in the field, given that the surveillance system relies heavily on abortion notifications by farmers and veterinarians. Our analysis showed that (1) data quality was generally satisfactory even though some errors might be due to actors' lack of awareness of the need to collect accurate data; (2) from 2006 to 2011, the mean annual sensitivity - i.e. the proportion of farmers who reported at least one abortion out of all those who detected such events - was around 34%, but was significantly higher in dairy than beef cattle herds (highlighting a lack of representativeness); (3) overall, the system's low sensitivity was related to its low acceptability and lack of simplicity. This study showed that, in contrast to policy-makers, most farmers and veterinarians perceived the risk of a brucellosis outbreak as negligible. They did not consider sporadic abortions as a suspected case of brucellosis and usually reported abortions only to identify their cause rather than to reject brucellosis. The system proved too complex, especially for beef cattle farmers, as they may fail to detect aborting cows at pasture or have difficulties catching them for sampling. By investigating critical attributes, our evaluation highlighted the surveillance system's strengths and needed improvements. We believe our comprehensive approach can be used to assess other event-driven surveillance systems. In addition, some of our recommendations on increasing the effectiveness of event-driven brucellosis surveillance may be useful in improving the notification rate for suspected cases of other exotic diseases.

Devising an indicator to detect mid-term abortions in dairy cattle: a first step towards syndromic surveillance of abortive diseases.

Bovine abortion surveillance is essential for human and animal health because it plays an important role in the early warning of several diseases. Due to the limited sensitivity of traditional surveillance systems, there is a growing interest for the development of syndromic surveillance. Our objective was to assess whether, routinely collected, artificial insemination (AI) data could be used, as part of a syndromic surveillance system, to devise an indicator of mid-term abortions in dairy cattle herds in France. A mid-term abortion incidence rate (MAIR) was computed as the ratio of the number of mid-term abortions to the number of female-weeks at risk. A mid-term abortion was defined as a return-to-service (i.e., a new AI) taking place 90 to 180 days after the previous AI. Weekly variations in the MAIR in heifers and parous cows were modeled with a time-dependent Poisson model at the département level (French administrative division) during the period of 2004 to 2010. The usefulness of monitoring this indicator to detect a disease-related increase in mid-term abortions was evaluated using data from the 2007-2008 episode of bluetongue serotype 8 (BT8) in France. An increase in the MAIR was identified in heifers and parous cows in 47% (n = 24) and 71% (n = 39) of the departements. On average, the weekly MAIR among heifers increased by 3.8% (min-max: 0.02-57.9%) when the mean number of BT8 cases that occurred in the previous 8 to 13 weeks increased by one. The weekly MAIR among parous cows increased by 1.4% (0.01-8.5%) when the mean number of BT8 cases occurring in the previous 6 to 12 weeks increased by one. These results underline the potential of the MAIR to identify an increase in mid-term abortions and suggest that it is a good candidate for the implementation of a syndromic surveillance system for bovine abortions.

Was the French clinical surveillance system of bovine brucellosis influenced by the occurrence and surveillance of other abortive diseases?

The bovine brucellosis clinical surveillance system implemented in France aims to detect early any case of bovine brucellosis, a disease of which the country has been declared free since 2005. It relies on the mandatory notification of every bovine abortion. Following the spread of the Schmallenberg virus (SBV) in France in 2012 and 2013, and the implementation in 2012 of a clinical surveillance programme of Q fever based on abortion notifications in ten pilot départements, our objective was to study whether these two events influenced the brucellosis clinical surveillance system. The proportion of notifying farmers was analyzed over each semester from June 1, 2009 to June 30, 2013 according to the size and production type of herds, SBV status of départements and the implementation of the Q fever surveillance. Our analysis showed a slight increase in the proportion of notifying farmers as départements became infected by SBV, and after the implementation of Q fever surveillance (during the first semester of 2013). These variations might be explained by an increase in abortion occurrence (congenital deformities in newborns, due to SBV) and/or by an increase in farmers' and veterinarians' awareness (due to the spread of SBV and the implementation of the Q fever surveillance). These results highlight the difficulties in interpreting variations in the proportion of notifying farmers as a consequence of an increase in abortion occurrence. As bovine abortion surveillance can play an important role in the early warning for several diseases, there is a need to explore other ways to monitor abortions in cattle, such as syndromic surveillance using the dates of artificial insemination or calving data.

Why do farmers and veterinarians not report all bovine abortions, as requested by the clinical brucellosis surveillance system in France?

BACKGROUND: Since 2005, France has been officially free of brucellosis, an infectious disease that causes abortion in cattle and can be transmitted from cattle to humans. Recent animal and human cases have drawn attention to the need to prevent infection of humans and animals from any primary outbreaks. In order to detect any new outbreaks as soon as possible, a clinical surveillance system requires farmers and veterinarians to report each abortion and to test the aborting cow for brucellosis. However, under-reporting limits the sensitivity of this system. Our objective was to identify the barriers and motivations influencing field actors in their decision to report or not to report bovine abortions. We used a qualitative approach with semi-structured interviews of 12 cattle farmers and their eight veterinarians. RESULTS: Our analysis showed that four main themes influence the decision-making process of farmers and veterinarians: 1) the perceived risk of brucellosis and other abortive diseases; 2) the definition of a suspected case of brucellosis and other abortive diseases adopted by field actors, which is less sensitive than the mandatory definition; 3) the cost-benefit analysis conducted by actors, taking into account regulatory and health aspects, economic and financial losses, technical and practical factors; 4) the level of cooperation within the socio-technical network. We discussed how early detection may be improved by revising the definition of abortion, extending the time frame for notification and generalising the differential diagnosis of the causes of abortion. CONCLUSIONS: In contrast to quantitative approaches, qualitative studies can identify the factors (including unknown factors) influencing the decision-making process of field actors and reveal why they take those factors into consideration. Our qualitative study sheds light on the factors underlying the poor sensitivity of clinical brucellosis surveillance system for cattle in France, and suggests that early detection may be improved by considering actors' perceptions. We believe our findings may provide further insight into ways of improving other clinical surveillance systems and thus reduce the risk of disease.

Inventory of veterinary syndromic surveillance initiatives in Europe (Triple-S project): current situation and perspectives.

Within the current context that favours the emergence of new diseases, syndromic surveillance (SyS) appears increasingly more relevant tool for the early detection of unexpected health events. The Triple-S project (Syndromic Surveillance Systems in Europe), co-financed by the European Commission, was launched in September 2010 for a three year period to promote both human and animal health SyS in European countries. Objectives of the project included performing an inventory of current and planned European animal health SyS systems and promoting knowledge transfer between SyS experts. This study presents and discusses the results of the Triple-S inventory of European veterinary SyS initiatives. European SyS systems were identified through an active process based on a questionnaire sent to animal health experts involved in SyS in Europe. Results were analyzed through a descriptive analysis and a multiple factor analysis (MFA) in order to establish a typology of the European SyS initiatives. Twenty seven European SyS systems were identified from twelve countries, at different levels of development, from project phase to active systems. Results of this inventory showed a real interest of European countries for SyS but also highlighted the novelty of this field. This survey highlighted the diversity of SyS systems in Europe in terms of objectives, population targeted, data providers, indicators monitored. For most SyS initiatives, statistical analysis of surveillance results was identified as a limitation in using the data. MFA results distinguished two types of systems. The first one belonged to the private sector, focused on companion animals and had reached a higher degree of achievement. The second one was based on mandatory collected data, targeted livestock species and is still in an early project phase. The exchange of knowledge between human and animal health sectors was considered useful to enhance SyS. In the same way that SyS is complementary to traditional surveillance, synergies between human and animal health SyS could be an added value, most notably to enhance timeliness, sensitivity and help interpreting non-specific signals.

Assessing the mandatory bovine abortion notification system in France using unilist capture-recapture approach.

The mandatory bovine abortion notification system in France aims to detect as soon as possible any resurgence of bovine brucellosis. However, under-reporting seems to be a major limitation of this system. We used a unilist capture-recapture approach to assess the sensitivity, i.e. the proportion of farmers who reported at least one abortion among those who detected such events, and representativeness of the system during 2006-2011. We implemented a zero-inflated Poisson model to estimate the proportion of farmers who detected at least one abortion, and among them, the proportion of farmers not reporting. We also applied a hurdle model to evaluate the effect of factors influencing the notification process. We found that the overall surveillance sensitivity was about 34%, and was higher in beef than dairy cattle farms. The observed increase in the proportion of notifying farmers from 2007 to 2009 resulted from an increase in the surveillance sensitivity in 2007/2008 and an increase in the proportion of farmers who detected at least one abortion in 2008/2009. These patterns suggest a raise in farmers' awareness in 2007/2008 when the Bluetongue Virus (BTV) was detected in France, followed by an increase in the number of abortions in 2008/2009 as BTV spread across the country. Our study indicated a lack of sensitivity of the mandatory bovine abortion notification system, raising concerns about the ability to detect brucellosis outbreaks early. With the increasing need to survey the zoonotic Rift Valley Fever and Q fever diseases that may also cause bovine abortions, our approach is of primary interest for animal health stakeholders to develop information programs to increase abortion notifications. Our framework combining hurdle and ZIP models may also be applied to estimate the completeness of other clinical surveillance systems.