Physiological Maturity of Beef Cattle Carcasses: Terms of Reference
Introduction
The U.S. Government (USG) and the Government of Japan (GOJ) will undertake an interim trading program in beef and beef products. That program, known as a Beef Export Verification (BEV) Program, has certain requirements which the U.S. Department of Agriculturefs (USDA) Agricultural Marketing Service (AMS) will certify to have been met. One of those requirements is that beef and products only from animals 20 months of age or younger may be sold to Japan.
|
Additional Information:
• What is the Beef Export Verification Program (BEV)? • Determination of Cattle Age (20 Months or Less)
|
The U.S. plans to proceed consistent with the framework as outlined in the Annex of the "Joint Press Statement" relative to the issue of determining age through physiological evaluation of carcass maturity. To this extent, the U.S. will collaborate with the Japanese experts to assure a comprehensive study is conducted, representative of the U.S. cattle industry.
The U.S. Study membership includes:
Mr. Barry L. Carpenter, Deputy Administrator, LS, AMS, USDA
Mr. Martin E. OfConnor, Chief, Standardization Branch (SB), LS, AMS, USDA
Dr. Justin R. Ransom, Agricultural Marketing Specialist, SB, LS, AMS, USDA
Dr. Jimmy W. Wise, Agricultural Marketing Specialist, SB, LS, AMS, USDA
Dr. Michael B. Feil, Chief, Statistics Branch, Science and Technology, AMS, USDA
Dr. Gary C. Smith, Distinguished Professor of Meat Science, Colorado State University
Dr. Keith E. Belk, Professor of Meat Science, Colorado State University
Purpose of Study
The USDA/AMS will conduct a special study of the relationship between cattle age and physiological maturity of the carcass. This study is to assist in finding the "end point" in the USDA grading system which reliably indicates animal age to be 20 months or younger.
Carcasses of steers and heifers of known ages (births identified within a one-month period) will be evaluated for physiological maturity for the specific purpose of determining an end-point maturity (e.g., A60, A70, A80, etc.) to ensure that only beef and beef products from steers or heifers 20 months or younger are shipped to Japan under the special marketing program. The study will be designed and the data analyzed utilizing internationally recognized sampling and statistical methods.
The Sample
The U.S. cattle herd consists of approximately 96 million animals. This herd size results in a harvest each year of some 35 million animals. The vast majority of the animals slaughtered are relatively young, produced specifically for high-quality beef and beef products. It is estimated that some 80% or 27 million head of the younger animals are 12 to 30 months of age. This is the population from which the sample of animals with known birth dates will be drawn for carcass evaluation. (The remaining animals slaughtered each year are primarily older cows and bulls from the beef and dairy herds.)
The sample of animals selected will take account of the variation inherent in the U.S. commercial fed-beef population that could influence carcass characteristics. This variation includes such factors as:
- Age (12 - 30 months);
- Sex (steer and heifer);
- Breed (predominately British, predominately continental, Bos indicus influenced, and Holstein);
- Geographical region; and
- Management system.
Sample Size. The size of the sample to be analyzed will be adequate to ensure that it reflects the major characteristics of the population of interest (12 - 30 months of age). Statisticians will determine the specific number that is required to assure that the sample meets the requirements.
(a) The main emphasis for sample herd selection will be to collect information on as many sources of known age cattle as possible marketed within the available window of opportunity offered through the timeline of this Study.
(b) Dependent upon the number of cattle identified, the sample size will be determined using statistical methods to assure accurate representation of the U.S. cattle population.
(c) Birth information is provided through the supply chain and is traceable back to ranch of birth. Randomized traceback will be conducted on selected cattle to ensure the validity of the reporting process back to records indicating the birth period.
(d) The physiological maturity information will be gathered at the individual locations where the evaluation takes place. This information along with birth information will be sent electronically to the Standardization Branch to be analyzed and reported through recognized protocols between US and Japanese experts.
Study Methodology
The study has five major components that will be conducted sequentially.
- Component 1
- Identifying the population of animals meeting the study requirements and determining the sample size needed.
- Component 2
- Selecting the sample of animals required having a known birth date.
- Component 3
- Grading the carcasses to determine the physiological characteristics and specific scores.
- Component 4
- Analyzing the data collected to determine the statistical correlation between chronological age and physiological maturity.
- Component 5
- Completion of the study report and presentation to USG and GOJ officials.
The maturity groups as outlined in the grading standard are: A, B, C, D, and E, respectively. These maturity groups represent the full spectrum of physiological maturity categories of all cattle produced in the United States.
When carcasses are presented from the sample selected, certified USDA meat graders will evaluate all physiological traits that are set forth in the U.S. Standards for Grades of Carcass Beef. This then results in the determination of three scores: one for skeletal maturity (bone ossification); one for leanness; and a combination of the two that is the overall maturity score: the skeletal, lean, and overall maturity scores are recorded in tenths of a degree of maturity groups (e.g., A50, A60, A70, A80, etc.).
Because this study is concerned only with the commercially fed beef population younger than 30 months, C, D and E physiological maturity groups (older animals) will not be represented.
Study Timeline
Animals included in the study sample will come from feedlots where they are being finished for slaughter. The survey of feedlots with known-age steers and heifers will be completed by October 29, 2004. Some animals meeting the study requirements already have been identified and data obtained on those carcasses. Data collection began the week of October 18th and should be concluded by November 27th. Upon completion of the data collection, an analysis will be conducted and a final report presented by December 4, 2004.
Status of the Study
Upon notification to the GOJ of the U.S.fs intent to perform this study, the U.S. immediately began the process of surveying feedlots in order to identify cattle of known age that are scheduled to be slaughtered within the timeline of this study. This first phase has proven to be crucial to the success of this study, as it will assure the opportunity to collect data on as many cattle within the identified population as possible. The U.S. further concluded that it would be impossible to randomly yet methodically collect the exact percentages of cattle which will represent the U.S. beef population, thus data collection began on all cattle identified with a known-age. Through this, the U.S. will be able to model the data to more accurately reflect the actual cattle supply, by correctly accounting for the percentages of production variables (sex, age, breed, and management background).
Survey: The survey of feedlots is continuing beyond the October 29, 2004 date to allow as many cattle of known ages as possible to be included in the target sample. As more cattle of known ages are identified to USDA, these cattle are identified for inclusion in the target sample to assure a representative sample will be available once the analysis begins. In order to ensure adequate representation of each of the areas identified in the draft Study document, it is critical to include all available cattle that qualify for the parameters of the study to be included. This is especially critical for identifying cattle of greater than 20 months of age because so few exist. Statistical rules of analysis will assure accurate representation of variables to be evaluated once all the data is collected.
Sample: The composition of the meat cattle population in the U.S. for the determination of sample selection and size was based on the peer reviewed and published research: National Beef Quality Audit-2000 (McKenna et al, 2002).
The final sample size will be based upon the actual number of cattle in the sample population that have the sufficient range in age needed to ensure that the requisite criteria of the study are attained. The intent of the U.S. is to obtain as much information as possible from the producers of the known age cattle, and to subsequently collect data on a minimum of 2,000 carcasses. Dependent upon the extent of the information collected, a stratification of the data may need to be performed in order to more nearly represent the U.S. cattle population as identified in the National Beef Quality Audit-2000.
Detailed information on how the sample population will be created cannot be determined at this time as the most appropriate model for analysis has not been selected. At this time the priority is to assure that adequate information is collected on qualifying cattle. A meeting of the U.S. scientific group is scheduled for Wednesday, November 10, 2004 in Washington, DC to develop a proposed methodology and rationale paper for this specific issue.
Analysis: The ultimate goal of the U.S. is to develop interval estimates of the odds ratio which gives the likelihood of a specific result accompanied by a quantitative statement of the magnitude and precision for that specific result. To achieve this, analytic methods under consideration include general linear models, recursive partitioning, logistic regression, and stratified analysis.
General linear models consist of multiple regression analysis, analysis of variance, and analysis of covariance. These are used to investigate the functional relationships among variables. The method selected to develop any models is dependent upon the data types.
Recursive partitioning is useful in producing an accurate classifier or uncovering the predictive structure of a data set. These goals are not mutually exclusive. There is frequently considerable overlap between the goals. Recursive partitioning is appropriate when the data set has high dimensionality, a mixture of data types, a nonstandard data structure, and nonhomogeneity, i.e., there are different relationships between variables in different parts of the data set.
Logistic regression could be used to assess the relationship between carcass maturity and chronological age as a two-level variable after the initial relationship has been modeled. Stratified analysis is preceded by categorical analysis which examines the crude data set in very global terms. It creates homogeneous groups for further analysis.
REFERENCES:
McKenna, D.R., D.L. Roeber, P.K. Bates, T.B. Schmidt, D.S. Hale, D.B. Griffin, J.W. Savell, J.C. Brooks, J.B. Morgan, T.H. Montgomery, K.E. Belk, and G.C. Smith. 2002. National Beef Quality Audit-2000: Survey of targeted cattle and carcass characteristics related to quality, quantity and value of fed steers and heifers. J. Anim Sci. 80:1212-1222.