【病毒外文文献】2016 Monitoring the Spread of Swine Enteric Coronavirus Diseases in the United States in the Absence of a Regulatory Fra

March 2016 Volume 3 Article 18 1 Case Repo Rt published 14 March 2016 doi 10 3389 fvets 2016 00018 Frontiers in Veterinary Science www frontiersin org Edited by Beatriz Mart nez L pez University of California Davis USA Reviewed by Lina Mur Kansas State University USA Flavie Vial Epi Connect Sweden Correspondence Andres M Perez aperez umn edu Specialty section This article was submitted to Veterinary Epidemiology and Economics a section of the journal Frontiers in Veterinary Science Received 01 October 2015 Accepted 17 February 2016 Published 14 March 2016 Citation Perez AM Alba A Goede D McCluskey B and Morrison R 2016 Monitoring the Spread of Swine Enteric Coronavirus Diseases in the United States in the Absence of a Regulatory Framework Front Vet Sci 3 18 doi 10 3389 fvets 2016 00018 Monitoring the spread of swine enteric Coronavirus Diseases in the United states in the absence of a Regulatory Framework Andres M Perez 1 Anna Alba 1 Dane Goede 1 Brian McCluskey 2 and Robert Morrison 1 1 Department of Veterinary Population Medicine College of Veterinary Medicine University of Minnesota St Paul MN USA 2 Veterinary Services Animal and Plant Health Inspection Service U S Department of Agriculture Fort Collins CO USA The reporting and monitoring of swine enteric coronavirus diseases SECD including porcine epidemic diarrhea virus and porcine delta coronavirus in the United States have been challenging because of the initial absence of a regulatory framework and the emerging nature of these diseases The National Animal Health Laboratory Network the Emergency Management and Response System and the Swine Health Monitoring Project were used to monitor the disease situation between May 2013 and March 2015 Important differences existed between and among them in terms of nature and extent of reporting Here we assess the implementation of these systems from different perspectives including a description and comparison of collected data disease met rics usefulness simplicity flexibility acceptability representativeness timeliness and stability This assessment demonstrates the limitations that the absence of premises identification imposes on certain animal health surveillance and response databases and the importance of federally regulated frameworks in collecting accurate information in a timely manner This study also demonstrates the value that the voluntary and pro ducer organized systems may have in monitoring emerging diseases The results from all three data sources help to establish the baseline information on SECD epidemiological dynamics after almost 3 years of disease occurrence in the country Keywords swine enteric coronavirus epidemiology monitoring United states INt Ro DUCt Io N Since the first detection of porcine epidemic diarrhea virus PEDV in the United States US in May 2013 followed by the subsequent detection of porcine delta coronavirus PDCoV many efforts have been made to monitor the spread of swine enteric coronavirus diseases SECD in the country 1 7 PEDV and PDCoV reporting to the World Organization for Animal Health OIE are not mandatory although the reporting of SECD infection is encouraged due to their emerging nature and important economic impact on swine industry 8 To investigate the temporal and spatial spread of SECD in the US and to obtain information to support decision making the US Department of Agriculture USDA used three data sources namely the National Animal Health Laboratory Network NAHLN 9 11 the Emergency March 2016 Volume 3 Article 18 2 Perez et al SECD Monitoring in the US Frontiers in Veterinary Science www frontiersin org Management and Response System EMRS 12 and the Swine Health Monitoring Project SHMP 13 These sources differed on a number of qualitative and quantitative attributes such as the target population the temporal extent of the collected data the case definition and the nature of reporting The goal of this report is to report the attributes and conduct a critical assessment of the three systems used to monitor the spread of SECD in the US between May 2013 and March 2015 Their evaluation is approached from different perspectives First we describe features associated with each dataset and their respective attributes Second the databases are compared in terms of shape and size of the epidemic curves generated from the data in each Finally different qualitative aspects usefulness simplicity flexibility quality of data acceptability representative ness and timeliness and stability are described and compared in order to identify strengths and weaknesses and assess their potential application to the surveillance for SECD The report here contributes to understand the nature and extent of the data collected to monitor PEDV and PDCoV in the US and helps to establish the baseline information on the SECD epidemiological dynamics after 2 years of disease spread in the country Mate RIaLs aND Met Ho Ds Background of Data sources National Animal Health Laboratory Network The NAHLN is a coordinated network of federal and state institu tions which includes universities and animal disease diagnostic laboratories that collaborate to provide diagnostic testing to animal health surveillance and give response to important dis ease events 9 11 The NAHLN laboratories provided weekly data files of the PEDV PCR test result records 14 15 including results from each sample tested and if available the associated data on the collection site state and animal age A laboratory accession with one or more PEDV PCR positive samples was considered positive In most of the cases NAHLN data did not include information to identify individual premises or herds The coverage of the NAHLN data was that of the participating laboratory s service areas As of May 2014 24 veterinary diag nostic laboratories voluntarily reported PEDV testing data to the USDA NAHLN program office Figure 1 USDA Emergency Management and Response System The USDA EMRS is an information technology system designed to provide timely and effective response to animal health emer gencies including foreign animal disease FAD investigations and state and national animal disease incidents 12 Federal state and tribal animal health officials use the EMRS to record and view animal disease data manage incident response services and resources and create reports and maps to facilitate disease investigations and associated epidemiological analyses On June 5 2014 according to the Federal Order FO issued by the USDA the EMRS was designated as the official system for the recording of all SECD situation data collection information management and reporting The FO mandated the reporting of SECD cases 16 The EMRS also received testing data provided by the NAHLN laboratories The NAHLN data were electronically transmitted into the EMRS to facilitate the response activities The emergency response officials investigated all the SECD case reports that were initiated by sharing the lab results or other communication channels and determined the SECD status of each reported herd premise On the basis of lab test results and consultations with herd owners and herd veterinarians the premises received a status of confirmed or presumptive positive for PEDV PDCoV or both viruses In accordance with the USDA case definition a confirmed positive herd premise was a site with animals that had at least one positive for PEDV or PDCoV to a PCR test and a history of clinical signs consistent with SECD whereas a pre sumptive herd premise had animals that tested positive for either disease without manifesting clinical signs 17 18 The status of the premises was reported into the EMRS as soon as a federal or state official had confirmed all the information sources required and the status could be updated modified or closed until the new information became available Swine Health Monitoring Project This system was designed by the University of Minnesota in col laboration with the American Association of Swine V eterinarians and the National Pork Board and was aimed at monitoring important diseases that affected the swine industry Initially the SHMP included data of porcine reproductive and respiratory syndrome PRRS cases collected since July 17 2013 When the PEDV epidemic started the infrastructure of the SHMP was used to collect SECD incidence data The project collects data routinely from breeding farms including commercial multiplier nucleus The participation in the project and the contribution to the data base on a weekly basis was voluntary 13 The attributes collected by this system included the premises ID the coordinates the state and county of herd the average inventory capacity the type of breeding herd the air filtration status and the location of nearby pig farms 3 miles The individual data were informed to the participants of the study From the data gathered by each source we described and compared the respective epidemic curves evaluation of the se DC Monitoring systems from Different perspectives Based on the guidelines proposed by Salman et al 19 20 we assessed the following aspects of each system 1 usefulness contribution to the prevention and control of diseases 2 simplicity considering the operability and logistics 3 flexibility ability to adapt to changing information needs or operating conditions 4 quality of data completeness and validity of the data recorded 5 acceptability willingness of people and organization to participate 6 representativeness cases detected by the system that represent the true situation FIGURe 1 spatial coverage of the three information systems used to collect data on enteric coronavirus records in the United states between May 2013 and March 2015 top in green the National animal Health Laboratory Network NaHLN middle in blue swine Health Monitoring and bottom in red emergency Management Reporting system Continued March 2016 Volume 3 Article 18 3 Perez et al SECD Monitoring in the US Frontiers in Veterinary Science www frontiersin orgta BLe 1 Comparative features associated with the National animal Health Laboratory Network NaHL emergency Management and Response system eMRs and swine Health Monitoring project sHMp databases Database National animal Health Laboratory Network NaHLN emergency Management and Response system eMRs swine Health Monitoring project sHMp Purpose Compilation of PEDV PCR testing data for analysis reporting and decision support Occurrence of PEDV in the swine herds premises Situational awareness for the participating systems Tracking of disease response and control actions Start date June 16 2013 June 5 2014 May 13 2013 Unit record Test result Premise herd Breeding herd Coverage National all sites National all sites Participating systems Case definition A sample positive to PEDV PCR Confirmed herd at least one pig positive to PEDV PCR plus pigs with clinical signs Farm in which PEDV was reported by the veterinarian based on clinical signs and diagnostic test results Presumptive herd at least one pig positive to PEDV PCR and no clinical signs observed Frequency of data submissions Weekly until September 2014 After some daily submissions using the HL7 electronic messaging Daily Weekly Participation Voluntary NAHLN labs Mandatory according to the federal ordering Voluntary producers of breeding herds Access to available data USDA staff and lab participants State and Federal animal health officials Public as aggregated data Participants raw and de identified data Reporting Weekly reports of amount of positive and negative accessions by week month and state publicly available Weekly reports summaries of the positive premises by week month and state reports available to the public Weekly reports depicting time series for participating systems March 2016 Volume 3 Article 18 4 Perez et al SECD Monitoring in the US Frontiers in Veterinary Science www frontiersin org 7 timeliness speed between steps in a surveillance system and providing feed back information and 8 stability ability to function without failure and availability when the system is needed Although the evaluation of sensitivity and positive predic tive value were not under the scope of this study since it would require other quantitative approaches this work also provided some evidences from the analysis of outcomes Res ULts Comparison of Basic Features and Description and Comparison of epidemic Curves from Data Gathered by each system There were substantial differences between and among the three information systems used to monitor the SECD spread in the US Table 1 Hence the epidemic curves built from the monthly SECD records reported by the three monitoring systems showed different shapes Figure 2 Shades depict the number of reports on each database and state Stars in the NAHLN figure indicate the location of the participating laboratories including Arkansas Livestock and Poultry Commission Veterinary Diagnostic Laboratory Athens Veterinary Diagnostic Laboratory University of Georgia University of Illinois Veterinary Diagnostic Laboratory Illinois Department of Agriculture Galesburg Animal Disease Laboratory Iowa State University Veterinary Diagnostic Laboratory Kansas State Veterinary Diagnostic Laboratory Michigan State University Diagnostic Center for Population and Animal Health University of Minnesota Veterinary Diagnostic Laboratory Veterinary Medical Diagnostic Laboratory University of Missouri University of Nebraska Veterinary Diagnostic Center USDA National Veterinary Services Laboratories Rollins Diagnostic Laboratory North Carolina Department of Agriculture Veterinary Diagnostic Laboratory North Dakota State University Ohio Animal Disease Diagnostic Laboratory Ohio Department of Agriculture Oregon State University Veterinary Diagnostic Laboratory Indiana Animal Disease Diagnostic Laboratory Purdue University Animal Disease Research and Diagnostic Laboratory South Dakota State University and Texas Veterinary Medical Diagnostic Laboratory Texas A M University FIGURe 1 Continued These differences could be explained at least in part to diverse reasons related to the implementation of each system The SHMP was available from the beginning of the epidemic and essential to assess the initial stages since the reporting of SECD from NAHLN or EMRS was not in place at that moment The SHMP data were spread over an extended period of time suggesting a lower transmission rate compared to what may be estimated using the EMRS data noteworthy NAHLN data would not allow estimating a transmission rate In the NAHLN and EMRS many reports were concentrated at the beginning of the curve likely because they included a number of reports that occurred earlier in the epidemic In contrast in the winter of 2014 2015 when the three databases had been in place for almost 2 years all the epidemic curves pointed toward an increase of PEDV incidence versus the previous summer levels which is particularly evident from data in the EMRS database Although the magnitude of the peak differed between databases the epidemic curves reflected an upward trend of PEDV from October 2014 until January 2015 Because one single state Iowa accounts for almost 50 of the swine farms in the country the epidemic curves were also explored at state level to assess if the pattern observed in FIGURe 2 Monthly swine enteric coronavirus records reported in the United states by the three monitoring systems between May 2013 and March 2015 March 2016 Volume 3 Article 18 5 Perez et al SECD Monitoring in the US Frontiers in Veterinary Science www frontiersin org Iowa dominated the evidence for the entire country or if a similar pattern was observed at the state level for those states that contribute to most of the swine production in the country Although the disease dynamics seemed to be slightly different between states probably associated with different times of disease introduction demographic and epidemiological condi tions the most swine densely populated states showed similar trends over time suggesting that the pattern observed in the entire country is not just a reflection of what occurred in Iowa Figure 3 evaluation of the se CD Monitoring systems Usefulness The NAHLN database provided information of almost every single SECD diagnosis in the US This information was crucial to determine which SECD viruses were circulating However this system was not sufficient to determine the main meas ures of disease frequency or for decision making since the individual identification of infected premises or herds was not recorded In contrast to NAHLN the EMRS database comprised complete and reliable data at the farm level for epidemiological purposes including all farm types and containing data for posi tive farms When the EMRS was in place the SECD occurrence and spread could be investigated and the system was used to determine prevention and control measures The SHMP served to assess the initial stages of the epidemic when most of the farms were infected The SHMP data referred only to sow farms from some systems and could not be inferred to the entire country but this system helped to create the founda tions for an ongoing monitoring system database for emerging and non reportable diseases in the US Simplicity The USDA NAHLN asked to volunteer labs and universities their SEDC testing data and share this information electronically This system followed by NAHLN was much simpler and cheaper than the EMRS FIGURe 3 Monthly swine enteric coronavirus records in the four Us states with more positive herds reported between May 2013 and March 2015 March 2016 Volume 3 Article 18 6 Perez et al SECD Monitoring in the US Frontiers in Veterinary Science www frontiersin org The EMRS had to collect and verify all the data received at the premise level and these tasks required extensive field work by USDA staff The SHMP asked to volunteer swine producers of breeding herds and only operated in some states Logistics of implementa tion were simpler than both NAHLN and EMRS Flexibility The NAHLN demonstrated the capacity to adapt as a result of the legal requirements of the FO on June 5 2014 First a new Laboratory Messaging System LMS data management system was launched to store and manage all PEDV testing data This new data system was the USDA s repository for laboratory test ing data and was connected to other key data systems including EMRS The LMS provided a system to transmit testing data electronically to USDA by using HL7 electronic result messaging technology As of March 2015 five NAHLN laboratories were electronically messaging their PEDV test results to USDA rather than providing weekly data files Another major change included the requirement of the premises identification data with the March 2016 Volume 3 Article 18 7 Perez et al SECD Monitoring in the US Frontiers in Veterinary Science www frontiersin org laboratory records However even with this requirement the PEDV laboratory test records often did not include the premises identifiers since the individual lab policies or the premises infor mation were not provided by the submitters The inclusion of the national premises identification data with the testing records slowly increased from 40 in June 2014 to nearly 90 in October 2014 In terms of flexibility logistics for implementation of the EMRS was more complex and consequently its adaptation to new situations required more time and effort than the NAHLN or the SHMP Quality of Data The labs that participated in the NAHLN had been previously certified for testing samples to confirm SECD diagnosis The testing data of NAHLN did not record the premises identification and prevented USDA from knowing exactly how many farms or premises were infected The lack of this attribute was a major constraint in using the NAHLN database for decision making The location of herds could be roughly determined by the Collection Site State variable reported by the lab but unfortunately this information was not completely reliable The labs could not validate the locatio