Abstract T 6 A data exchange format and

Зарегистрируйтесь, чтобы просмотреть полный документ!

Abstract T 6 A data exchange format and national database for producer-recorded health event data from on-farm management software J. B. Cole, D. J. Null*, and L. R. Bacheller Animal Improvement Programs Laboratory, Agricultural Research Service, USDA, Beltsville, MD 20705 -2350 2010 Introduction There is great interest in producing dairy cows that are healthy and remain in the herd longer. Direct and indirect costs associated with disease represent a significant expense to producers, and selection for improved health may reduce these costs significantly. Genetic response to selection for improved health based upon breeding values from genetic evaluations of field-recorded traits has been well-documented. That genetic variation is not currently being directly utilized for genetic improvement. Standardized health event codes and a data exchange format (Format 6) for the collection of data from onfarm record-keeping systems are necessary for the creation of a national database for health data. Standard codes for 19 health and 3 management events were developed based on the frequency of events in the DRMS data and consultation with veterinary experts. The most frequent health events were mastitis (19%), lameness (5%), metritis (5%), cystic ovary (3%), and retained placenta (2%). Format 6 includes management traits for ease of collection, and because they may be associated with longevity and profitability. Of the event records, 34% were categorized as health events and 59% as health maintenance or management events (e. g. vaccinations, hoof trims not associated with lameness, dry treatments, etc. ). Only health disorders were retained. Health Events Cystic Ovary Metritis Diarrhea/Scours Fever/Hypocalcemia Digestive Problem/Off Feed Nervous System Problem Displaced Abomasum Other Reproductive Problem Downer Cow Respiratory Problem Dystocia Retained Placenta Johne’s Disease (clinical) Stillbirth/Perinatal Survival Ketosis/Acetonemia Teat Injury Lameness Udder Edema Mastitis (clinical) Management Events Body Condition Score Milking Speed Temperament Discussion The database and editing systems were tested by processing actual health event records provided by two dairy records processing centers, DRMS and Ag. Source Cooperative Services (Verona, WI). Ag. Source provided 1, 285 lactation records and included 1, 585 distinct health events. The DRMS data included 63, 423 health events from 23, 332 lactations. The most frequent reported events differed by center and did not overlap. Ag. Source herds most frequently reported diarrhea (23%), ketosis (11%), lameness (11%), and teat injury (21%). The DRMS herds reported mastitis (38%), metritis (16%), and other reproductive problems (15%) most often. These differences may be related to software: Ag. Source herds typically use Dairy. COMP 305™ (Valley Agricultural Software, Tulare, CA), while DRMS herds usually use PCDART™. Standard Health Event Codes Dairy Records Management Systems (Raleigh, NC) provided about 3. 7 million producer-recorded health events for 1, 834 herds from 1997 through 2003. Records included cow ID, date, health event code (4 -character), and a comment field (up to 16 characters of supplemental information). A herd-based dataset included codes used by each farm with a 12 -character code definition. The herd and event datasets were merged, and standardized codes were assigned to the events. Testing Record and Database Structure A Format 6 record includes detailed cow identification, a health event code, an event date, and an optional detail field (Figure 1). This format can provide the necessary data for research into, and implementation of, genetic evaluations for economically-important health traits. Format 6 was designed to be easily extensible, as demonstrated by the addition of a locomotion score event to the specification in July, 2008. The existing table of lactation data, which is used to store lactation information and test day yields, was modified to include health event data transmitted on Format 6. Records are indexed using a composite key constructed from a unique internal animal ID, the calving date initiating a lactation, and the ID for the herd in which the record was initiated. Each record includes two columns associated with health data, an integral event counter and a variable-length string (VARCHAR) that contains the health event segments. Test day (Format 4) and reproductive (Format 5) data are stored in the national dairy database using the same strategy. Incoming Format 6 records are processed nightly using a sophisticated system of edits designed to insure data quality and consistency (Wiggans and Thornton, 2008). When problems are detected in the incoming data detailed error messages are generated for use by the dairy records processing centers. Ag. Source Health Event Records % Cystic Ovary 147 9 Diarrhea 355 23 Digestive Problem — — Displaced Abomasum 124 8 Downer Cow — — Dystocia — — Johne’s Disease — — Ketosis 176 11 Lameness 167 11 Mastitis — — Metritis — — Milk Fever 58 4 Reproduction 115 7 Respiratory Problem 74 5 Retained Placenta 48 3 Teat Injury 326 21 Udder Edema — — Total 1, 545 100 DRMS Records % 2, 325 4 99 <1 1, 867 3 798 1 46 <1 1, 515 2 519 1 2, 373 4 2, 666 4 23, 927 38 10, 141 16 553 1 9, 616 15 1, 839 3 4, 867 8 156 <1 114 <1 63, 423 100 Eighteen Ag. Source events had errors, which were event dates later than the processing date (16%), records with no production data (68%), and differing herd IDs for health and yield data (16%). A total of 3, 920 DRMS events had errors, the most common of which were calving dates that did not match event dates for dystocia (26%) and calving dates with no matching test day data (72%). The potential for improvement of dairy cattle health by means of genetic selection has been well-documented (Lyons et al. , 1991; Zwald et al. , 2004). However, health traits generally have low heritabilities, which means that large numbers of high -quality data are needed to produce accurate breeding value estimates. Format 6 provides a way for processing centers to transmit health event data to AIPL for research, but considerable work remains to convince producers that it is worthwhile to contribute their data. An even greater challenge may be that of encouraging software vendors to use the Format 6 codes in their on-farm applications. In this study, data were carefully inspected and records were assigned to standardized categories. This process must be automated for routine use, but the fact that producers can define their own codes in the on-farm software presents a significant challenge. The dataset of 3. 7 million records from DRMS included over 2, 600 different event codes, including 42 different codes related to diagnosis and treatment of lameness. These results, as well as the experiences of Zwald et al. (2004), underscore the desirability and difficulty of assigning standardized codes to current health event data. The results of this study show that it is feasible to collect health event data from on-farm management systems and combine them with other cow data in a national database. Further research is needed to determine how best to use those data to improve the health of dairy cows. References Animal Improvement Programs Laboratory. 2006. Format 6. http: //aipl. arsusda. gov/formats/fmt 6. html. Accessed 26 June, 2008. Lyons, D. T. , A. E. Freeman, and A. L. Kuck. 1991. Genetics of health traits in Holstein cattle. J. Dairy Sci. 74: 1092— 1100. Wiggans, G. R. , and Thornton, L. L. M. 2008. Processing of data discrepancies for U. S. dairy cattle and effect on genetic evaluation. Proc. Intl. Committee Anim. Recording, Niagara Falls, NY, June 16, pp. 19— 24. Zwald, N. R. , K. A. Weigel, Y. M. Chang, R. D. Welper, J. S. Clay. 2004. Genetic selection for health traits using producer-recorded data. I. Incidence rates, heritability estimates, and sire breeding values. J. Dairy Sci. 87: 4287 — 4294.

Скачать презентацию Abstract T 6 A data exchange format and

c21714da983d18037f5aa150f7276578.ppt

Количество слайдов: 1