Research Accomplishment Reports 2007

Animal Health & Grazing

C.N. Carter, L.R. Harrison, W. Northington
Livestock Disease Diagnostic Center

 

Project Description

Objective 1-- Design and implement adaptive computer-based forms for electronic accessioning of essential medical information related to clinical cases sent to veterinary diagnostic laboratories by practicing veterinarians and farmers. Resultant electronic accessioning will facilitate the development and implementation of a near real-time surveillance and epidemiology system for grazing animals.

Milestones Completed During this project period, a web-based diagnostic laboratory accessioning system was developed and tested at the Livestock Disease Diagnostic Laboratory (LDDC) in Lexington, KY and the Breathitt Veterinary Center (BVC) in Hopkinsville, KY. The interface also has the capability of collecting and storing animal health syndromic events (e.g. abortion, respiratory). This, combined with laboratory events will become data streams that will be analyzed by a near-real-time surveillance engine to detect disease clusters resulting in electronic alerts to appropriate animal health officials (e.g. Office of the State Veterinarian, USDA). This will be implemented to capture all accessions, syndromic data and laboratory test results at both diagnostic laboratories by October, 2008.

Also during this reporting period, the clinical source documents (accession forms) used by each laboratory were redesigned to determine the best means of collecting the required data elements to support management and reporting of the active case and to support near real-time surveillance and epidemiology services. Joint accession forms (LDDC and BVC) have been published for the collection of case information and have been mailed to all current laboratory clients. The new LIMS has been designed with accessioning templates that mirror the accession forms to make data capture as simple as possible. Eventually, the electronic forms will be used for direct data capture, thereby eliminating the paper forms.

Objective 2-- Design and develop/apply software systems technology that will enable near real-time surveillance and epidemiology via electronic accessioning and capture of diagnostic laboratory test results related to health problems in grazing animals.

Milestones Completed During this project period, a web-based diagnostic laboratory result capture system was developed and tested at the LDDC and BVC. All laboratory results are stored in a Microsoft SQL Server data base. This allows for the automated construction of data "views" that are passed to a health event cluster detection system at the close-of-business each day. A process that conducts a statistical analysis of specific health events is automatically spawned that reports out disease clusters (GIS map products, statistical data, cluster report) by 8:00 am on the next day. These products and the statistical output is reviewed by laboratory epidemiologists to decide if further action needs to be taken (e.g. field investigation, awareness bulletins, notification of the Office of the State Veterinarian).

Impact

Objective 3-- Design and develop/apply electronic surveillance and epidemiology reporting and alerting systems related to grazing animal disease events for laboratory clients, the State Veterinarian, USDA, and Kentucky Cabinet for Health and Family Services utilizing standards adopted by the USDA for the NAHLN.

Milestones Completed Veterinary diagnostic laboratories are in a unique position to analyze data from large numbers of clinical cases and to help with the early detection of health problems. During this project period, a system was developed and tested for the early detection of clusters of adverse health events in animal populations using diagnostic laboratory data. The system provides automated, near-real-time cluster alarms/alerts and medical situational awareness. The system could, also be used to detect clusters of animal health events of public health significance (e.g. bioterrorist attacks). During this reporting period, a semi-automated reportable disease system was developed and implemented to report animal diseases and positive test results to the Office of the State Veterinarian. Automated reporting/alerting to a broader audience will be achieved in the next phase of this project.

Objective 4-- Develop a web-based farm-level data collection tool that will capture a minimum data base of information on each premise.

Milestones Completed During this project period, a comprehensive, web-based farm registry system was developed and tested on 13 Kentucky farms. In August, 2007, an equine syndromic surveillance pilot study was conducted. The goal of the study was to demonstrate proof-of-concept that syndromic data can reliably be collected at the farm level. It is hoped that the pilot will lead to further studies involving cattle and other farms raising food animals. A web-based client was developed and tested to allow farms to register their unique farm data (full farm info, population at risk, etc) into a Microsoft SQL data base. The client was also designed to allow farm managers to log in each day and to capture the health events that occurred on the farm that day. During the months leading up to the pilot, thirteen farms were enrolled in the study. Farms reported syndromes for the full month of August, 2007. After the end of the collection period, data analysis commenced on what had been collected. Maps were generated using the reported syndromes to demonstrate to farm managers and veterinarians how they could quickly visualize how clusters of health events (e.g. abortion, respiratory illness) are occurring in areas close to their farm. The enrollment of most farms in a region will enable a robust daily statistical analyses that will identity possible clusters of health events and to notify farms, veterinarians, and health officials that a problem may exist. The output products (e.g. maps, graphs, descriptive statistics) maintains strict confidentiality of the location of event clusters. Event clusters will not identify the farms where they occurring, only a regional boundary such as zip code or county. The web-based farm registry and cluster detection algorithms can easily be applied to any type of farm operation.

Publications

Carter CN, Odoi A, Riley J, Smith J, Stepusin R, Cattoi T, McCollum S: Laboratory-based early animal disease detection utilizing a prospective space-time permutation scan statistic. Proceed of the XIII International Symposium of the WAVLD, 2007, p96.

Carter CN, Odoi A, Riley J, Smith J, Cattoi T, McCollum S: Laboratory-based early animal disease detection utilizing a prospective space-time permutation scan statistic. Proceed 50th Conf. AAVLD, p 69, Oct, 2007.
Carter CN, Odoi A: Diagnostic Laboratory Surveillance and Epidemiology: Serving Agriculture and Public Health, Proceed 143rd AVMA Convention, CD-ROM, July, 2006.