Pig and piglet.

2018-19 Rapid Ag: A comprehensive surveillance system to control influenza virus in pigs

June 8, 2017

Principal Leader

Montserrat Torremorell


Department of Veterinary Population Medicine

Funding Awarded

Renewal of 2016-2017 Project

  • 2018 Fiscal Year: $63,449
  • 2019 Fiscal Year: $63,449

The Problem

It is an unprecedented time for influenza infections. In pigs, the 2009 H1N1 pandemic had devastating effects on the swine industry and changed the landscape of circulating viruses in the US herd, with both animal and public health consequences. Therefore the risk that influenza infections pose to animal agriculture and people can hardly be overstated.


The original intent of this proposal was to have a surveillance system in place to guide decisions to control influenza at the farm level. The vision was to have a near real time surveillance system where information on influenza diagnostics would provide information on influenza status and type of influenza strains circulating in farms overtime. This information paired with farm level information such as geospatial coordinates, transport data, management practices and other diseases co-circulating at the farm level is instrumental to investigate which factors are responsible for the introduction, maintenance and dissemination of influenza virus overtime. Influenza is considered the second most costly viral disease in pigs in the U.S with cost estimates of $3.23-$10/pig (Dykhuis et al., 2012, Donovan et al., 2008) with the added risk to public health given its zoonotic nature and the role that pigs play in the emergence of novel strains that can transmit to people (Jhung et al., 2013). Thus there is an urgent need to develop strategies to predict, control and prevent influenza to improve food systems while preventing the risk to public health (Stärk and Häsler, 2015).

As part of this proposal, and working in collaboration with two production systems comprising of 78 farms, researchers piloted a near real time surveillance program capable of assessing incidence, prevalence and type of influenza infections present in herds. This surveillance system has allowed them to identify seasonal patterns of influenza herd prevalence which may guide timing of intervention strategies such as vaccination. The seasonality pattern observed so far indicates that influenza herd prevalence starts to increase in the fall and has two peaks of high prevalence, one in winter and one in spring, and has the lowest prevalence in the summer. This pattern was predictable across the 5 years of data analyzed and was explained in part by environmental factors of temperature and absolute humidity but not relative humidity. Air temperature and absolute humidity were negatively correlated with the levels of influenza over time (r=-0.33 and -0.43 respectively). This association can help explain in part the seasonality of influenza and both absolute humidity and temperature have been shown to affect influenza survivability and transmissibility in other studies (Guan et al., 2016; Lowen et al., 2014).

The surveillance system developed so far is also helping quantify the frequency of new strain introduction at the farm level, and potential spread across systems. The detection of genetically distinct strains was common in the farms studied and data showed a complex dynamic landscape of influenza strains co-circulating, but also emerging and disappearing overtime. We were able to identify 7 hemagglutinin 1 (H1) and 6 H3 genetically different clades of influenza circulating in the participating farms. Additionally, we reported that 21% of farms had at least 3 viruses, 18% had at least 2 and 41% had at least 1 influenza virus.  

As evidenced by the genetic diversity observed within the farms, as well as the high prevalence level detected overt time, a key question is whether vaccination remains a viable option to control influenza infections in pigs. Vaccination is one of the most common strategies to control influenza in swine herds. Influenza vaccines tend to prevent influenza clinical signs and mitigate the negative impact  on performance. However, one of the main imitations of current vaccines is the limited cross-protection that some of the vaccines may confer against circulating strains. Although vaccine use is widespread in pigs (80% in large herds of 500 or more breeding females) (NAHMS, 2012), there has not been a comprehensive effort to assess the effectiveness of vaccines in terms of controlling endemic infections, preventing the establishment of new infections and producing a pig that represents a low risk during the growing period. Despite the common use of vaccination there are many questions that remain unanswered in terms of timing, frequency and protocol, as well as type of vaccine in regards to homology to the circulating strains. In this study we seek to understand many of the factors that drive vaccination success measured not only in terms of improved performance and clinical signs but also prevalence, incidence and persistence of influenza in herds.  

Another key aspect of having a system that integrates diagnostic surveillance information and geospatial coordinates is the ability to conduct spatiotemporal analysis to further understand the transmission and dissemination of infections. Spatiotemporal analyses of porcine reproductive and respiratory syndrome virus (PRRSV) infections have resulted in the detection of disease clusters in time and space (Tousignant et al., 2015). These analyses have been central to understand PRRSV disease epidemiology and to better assess risks of local spread and effectiveness of regional control programs. However, information on the spatiotemporal spread of swine influenza is limited. Pig movement has also been associated with the introduction of influenza diversity into regions (Nelson et al., 2011) and is also associated with the global migration of influenza genetic diversity from North America and Europe to countries in Asia (Nelson et al., 2015). Because influenza is endemic in swine, analysis requires incorporation of sequencing information that allows the tracking of strains by type. As part of this proposal, we intend to study the spatiotemporal spread of influenza virus in the Midwest states of the US using data obtained from the influenza swine health monitoring program (SHMP) participants. A main goal is to evaluate dissemination of endemic influenza strains versus newly introduced strains, in particular ones of human-origin such as the newly detected H3N2 (Rajao et al., 2015). We will use methods from spatial epidemiology with Bayesian phylodynamic modeling (BEAST) to compare whether the spatiotemporal spread of influenza strains varies based on strain type. This information will be critical to understand and predict patterns of influenza spread within regions.    


The ultimate goal of this proposal is to develop and provide recommendations to producers and veterinarians to control influenza in swine herds. More specifically the objectives are: 

  1. To determine the impact of vaccination practices and protocols on status (prevalence) of influenza at weaning
  2. To analyze and monitor spatiotemporal patterns of the spread of influenza at multiple levels


Alkamis, M, Perez, A., and Morrison, R. 2015. Frontiers in Veterinary Science.
Bailey, T. C., Gatrell, A.T, 1995. Interactive spatial data analysis. Longman Scientific & Technical.Essex, England.
Drummond, A. J.; Rambaut, A., BEAST: Bayesian evolutionary analysis by sampling trees. BMC evolutionary biology 2007, 7, 214.
Donovan, T. (2008). Influenza isolate selection methodology for timely autogenous vaccine use (pp. 551-561). Proceedings. Amer Assoc Swine Vet Conf
Dykhuis, C., Painter, T., Fangman, T., Holtkamp, D. (2012). Assessing production parameters andeconomic impact of swine influenza, PRRS and Mycoplasma hyopneumoniae on finishing pigs ina large production system. in: Veterinarians, A.A.o.S.(Ed.), 2012 AASV Annual Meeting: Integrating Science, Welfare, and Economics in Practice. American Asociation of Swine Veterinarians, Denver, Colorado, pp. 75-76.
Guan, J., M. Chan, and A. VanderZaag. 2016. 'Inactivation of Avian Influenza Viruses on Porous and Non-porous Surfaces is Enhanced by Elevating Absolute Humidity', Transbound Emerg Dis: 1-8.
Jacquez, G. M., L.A. Waller., 1999. Theory and Applications for GIS and Remote Sensing. H. T. Mowrer and R. G. Congalton, eds.
Jhung MA, Epperson S, Biggerstaff M, Allen D, Balish A, Barnes N, et al. Outbreak of Variant Influenza A(H3N2) Virus in the United States. Clinical Infectious Diseases. 2013 Dec;57(12):1703-12.
Kulldorff M. A spatial scan statistic. Commun Stat Theory Methods 1997;26:1481–1496.
Kulldorff M., et al., 1998. SaTScan v2.1: Software for the spatial and space-time scan statistics. Bethesda, MD.
Lowen, A. C., and J. Steel. 2014. 'Roles of humidity and temperature in shaping influenza seasonality', J Virol, 88: 7692-5.
Nelson MI, Wentworth DE, Culhane MR, Vincent AL, Viboud C, LaPointe MP, et al. Introductions and Evolution of Human-Origin Seasonal Influenza A Viruses in Multinational Swine Populations. Journal of Virology. 2014 Sep;88(17):10110-9.
Nelson MI, et al. 2015. Nat Commun. 2015 Mar 27;6:6696.
Nelson, M. I., Lemey, P., Tan, Y., Vincent, A., Lam, T. T. Y., Detmer, S., et al. (2011). Spatial Dynamics of Human-Origin H1 Influenza A Virus in North American Swine. Plos Pathogens, 7(6).
Rajão DS, et al. 2015. J Virol. 2015 Nov;89(22):11213-22.Slomka, M.J., Densham, A.L.E., Coward, V.J., Essen, S., Brookes, S.M., Irvine, R.M., Spackman, E., Ridgeon, J., Gardner, R., Hanna, A., Suarez, D.L., Brown, I.H., 2010. Real time reverse transcription (RRT)-polymerase chain reaction (PCR) methods for detection of pandemic (H1N1) 2009 influenza virus and European swine influenza A virus infections in pigs. Influenza and Other Respiratory Viruses 4, 277–293.
Stärk,K.D., Häsler,B. 2015. Preventive Veterinary Medicine, Available online 14 May 2015, ISSN 0167-5877.
Tousignant, S.J.P., Perez, A.M., Morrison, R.B. et al. 2015 Am. J. Vet. Res. 76, 70-76.