Risk and the food safety chain: animal health, public health and the environment.

Food safety related to the consumption of animal-derived protein encompasses a wide variety of production and processing procedures which begins with the farm and inputs to the animals on the farm (e.g. feed and water) and includes the environment in which animals are reared. Hazards may be physical, artificial or naturally-occurring chemicals, organisms which cannot reproduce outside a specified life-cycle (e.g., parasites such as tapeworm in pigs) or viruses. Other microbes reproduce in the gastrointestinal tract of food animals as well as on the surface of food and in the environment. Methods of risk assessment for physical and chemical hazards have been used for many years. However, with microbial pathogens which can survive and grow on meat, in soil, water or other media, risk assessment methods are at the early stages of development. Due to the broad habitat range, the role of microbial pathogens in the food safety of meat, poultry, fruit and vegetables is important. The use of antibiotics in livestock species may accelerate the development of antibiotic-resistant strains of microbial pathogens, potentially complicating treatment for both animals and humans. The authors discuss the food chain, risk analysis and hazard analysis and critical control points in relation to foodborne pathogens, and introduce general strategies for improving pathogen control on the farm.

The application of probabilistic scenario analysis for risk assessment of animal health in international trade.

Probabilistic scenario analysis (PSA) is a method of risk assessment that has had wide usage in many fields, including engineering, nuclear safety, and financial analysis. It is fast becoming the "gold standard" for risk assessment in many fields, including human health and the environment. It has been successfully applied in animal and plant health issues as well. PSA begins with the identification of a hazard and the development of a step-by-step scenario from some initiating event to the end point at which the hazard occurs. In PSA, the pathway leading to the end point is outlined by a model, called an event tree. Each step leading to the occurrence of the hazard is carefully outlined, called the node on the event tree. At each of these nodes, the probability of the event leading to the end point (hazard occurrence) is evaluated. The evaluation of the probability at each node may be qualitative or quantitative. The evidence used may come from standard epidemiological studies. The model can also accommodate expert opinion, anecdotal evidence, or any other information that can be verified that is pertinent to the event leading to hazard occurrence. The PSA is also quite flexible for it can be quickly revised when new data become available. With careful statement of the evidence and linkage back to bibliographic or other sources, it can provide a transparent, flexible, well-documented approach to risk assessment for animal health. Risk assessment along with regionalization is the key to healthy national herds and free international trade.

Risk analysis systems for veterinary biologicals: a regulator's tool box.

Recent advances in biology and technology have significantly improved our ability to produce veterinary biologicals of high purity, efficacy and safety, virtually anywhere in the world. At the same time, increasing trade and comprehensive trade agreements, such as the Uruguay Round of the General Agreement on Tariffs and Trade (GATT: now the World Trade Organisation [WTO]), have put pressure on governments to use scientific principles in the regulation of trade for a wide range of products, including veterinary biologicals. In many cases, however, nations have been reluctant to allow the movement of veterinary biologicals, due to the perceived threat of importing an exotic disease. This paper discusses the history of risk analysis as a decision support tool and provides examples of how this tool may be used in a science-based regulatory system for veterinary biologicals. A wide variety of tools are described, including qualitative, semi-quantitative and quantitative methods, most with a long history of use in engineering and the health and environmental sciences.

Standardization of nomenclature for animal health risk analysis.

The authors propose standard definitions for terms and concepts commonly used in agricultural risk analysis. The differences between risk analysis and risk assessment are explained, and the relationship of these two terms with the more familiar terms "epidemiology" and "biostatistics" is discussed. The authors stress the importance of differentiating words and phrases which share similar sounds but are subtly different, and the necessity of developing a common nomenclature in agricultural risk analysis. An opportunity is provided for readers to comment on the definitions presented.

Developing a quantitative risk assessment process.

The international animal health community is advocating a more widespread use of quantitative risk assessment in making international trade decisions. In this article, the authors explain why quantitative risk assessment is a valuable tool, outline a process with key steps for developing a quantitative risk assessment, and describe a method for quantifying the uncertainty associated with the results of a risk assessment.

Serological evidence for the presence of Brucella antibodies in sheep and goats on Saint Croix, U.S. Virgin Islands.

A serological survey for Brucella antibodies in sheep and goats was completed on Saint Croix, United States Virgin Islands (USVI). Seroprevalence (at suspect or reactor titer levels) for B. melitensis antibodies was 11.3% for sheep and 2.5% for goats. This is the first report, of which we are aware, of B. melitensis antibodies in sheep or goats in the Caribbean islands.

Use of nonsteroidal anti-inflammatory drugs in food animal practice.

We attempted to determine the extent to which nonsteroidal anti-inflammatory drugs (NSAID) are used in the treatment of food animals, and whether withdrawal times for milk and slaughter are recommended to clients. A survey questionnaire was mailed to a stratified random sample of 2,000 veterinarians whose practices were at least half food animals. A cross-sectional study was used to examine the responses to determine whether differences existed on the basis of a respondent's geographic location, number of years since graduation from veterinary college, and percentage of practice devoted to beef and dairy cattle. The response rate was 71% (1,424/2,000). Of those practitioners responding, 93% (1,325/1,424) reported using NSAID, with approximately 57 (751/1,322), 24 (327/1,322), and 18% (244/1,322) of respondents reporting use more than once a week, once a week, and 1 to 2 times per month, respectively. Dairy practitioners reported more frequent use than did beef practitioners. Use of flunixin meglumine was reported more frequently than the use of aspirin, phenylbutazone, or dipyrone. Approximately 88% (1,146/1,306) of respondents that used NSAID did so in combination with antibiotics. Withdrawal times for milk and meat were made on the basis of guidelines for the antibiotic. When using NSAID alone, recommendations for withdrawal times for milk and meat varied extensively. Overall, practitioners indicated that NSAID were useful and necessary for the treatment of food-producing animals.

Leptospira serology in small ruminants on St. Croix, U.S. Virgin Islands.

A serological survey of 16 serovars of Leptospira interrogans, previously reported in tropical small ruminants, was undertaken to determine the serovars involved and the prevalence of these antibodies in sheep and goats on St. Croix, U. S. Virgin Islands (USVI). Seven of eight goat herds (108 animals) had at least two seropositive animals in each herd with an individual animal seroprevalence of 26%. The 53 sheep tested (one flock only) showed a 32% seroprevalence. Antibodies against seven serovars were detected in goats (autumnalis, ballum, bataviae, bratislava, canicola, icterohemorrhagiae, and pyrogenes). In addition, hardjo antibodies were detected in sheep. Serovar autumnalis accounted for about 30% of seropositive animals in each species. Many animals showed titers against more than one serovar. The number of seropositive animals suggests Leptospira may be a factor in the health of small ruminants on St. Croix.

Cost of mastitis and its prevention in four dairy cattle herds on St Croix, US Virgin Islands.

Nonclinical mastitis in the dairy herds on St Croix, US Virgin Islands, represents a prevalent (97%) and severe problem (44% of quarters had California Mastitis Test results of 2 or 3). Streptococcus agalactiae (26%) and Staphylococcus aureus (22%) were the most prevalent pathogens. More than half (54%) of all organisms isolated and tested for resistance were resistant to ampicillin and benzyl penicillin; 29% were resistant to trimethoprim/sulfadiazine. Estimates of the loss of milk production suggested that there was a loss of 4.40 to 13.33 lb/cow/day. The yearly economic cost of prevention was estimated for each cow ($22.07); yearly income loss attributable to mastitis was estimated to be from $339 to $1,022/cow/yr. Suggestions for the prevention and control of mastitis are included.

Experimental uses of flunixin meglumine and phenylbutazone in food-producing animals.

Presently, in the United States, there are no nonsteroidal anti-inflammatory drugs, except aspirin, that are approved for use in animals intended for food production. Use of phenylbutazone, flunixin meglumine, and dipyrone for treatment of food animals may be considered in special circumstances. Such use requires strict adherence to FDA guidelines for extra-label use of drugs. Flunixin meglumine and phenylbutazone have been shown to have a favorable influence on the course and outcome of certain diseases. This report reviews information concerning the pharmacology, pharmacokinetics, and therapeutics of phenylbutazone and flunixin as they have been used on an experimental basis in food animals.

Compliance with recommended drug withdrawal requirements for dairy cows sent to market in Michigan.

By examination of computerized individual cow records from January 1981 through June 1985, compliance with premarketing drug withdrawal times was investigated in 23 dairy herds. During this period, 877 cows were culled from 23 herds; 33 cows from 12 herds were in violation of premarketing drug withdrawal times. Six herds had 1 violation, 2 herds had 2 violations, 2 herds had 3 and 4 violations, and 2 herds had 7 and 9 violations. Fifty drugs had been used; 15 drugs were used in violation of drug withdrawal times before cows were sent to market. The most commonly used drugs in violation were oxytetracycline, gentamicin, chloramphenicol, and a combination of lincomycin/spectinomycin. Significant differences between cows in violation and cows not in violation were not observed for age, lactation number, milk production, herd size, or month of culling. However, the 12 herds in violation of premarket drug withdrawal times had significantly (P less than 0.05) more cows culled within 30 days for retained placenta and milk fever, and more cows culled within 30 and 100 days for mastitis, metritis, and displaced abomasum. Although not significant, the 2 herds with 7 and 9 violations had more hired vs family labor than did the other herds with violations.

Drug residues in dairy cattle industry: epidemiological evaluation of factors influencing their occurrence.

A random study of 3000 dairy farmers was designed to determine 1) management factors that may be associated with the occurrence of drug residues; 2) the dairy farmers' attitudes and knowledge about residues; 3) how these variables influence the occurrence of residues in dairy cattle. Management factors perceived as having the greatest influence on drug residues in milk were insufficient knowledge about withdrawal periods, errors due to hired help, insufficient identification and record of animals treated for mastitis, metritis treatment, dry cow treatment for mastitis, and not reading the label. Forty-one percent of farms with residue problems used medicated feeds compared with 38.9% of control farms. Factors significantly associated with the occurrence of residues were herd size, increased number of hired persons, increased frequency of use of medicated feeds, category of medicated feed, and producer's attitude toward the public health significance of residues. Farmers needed more information in five areas: likelihood of persistent residues from various types of chemicals (26.8%), consequences of chemicals occurring in livestock products (22.4%), withholding times for specific chemicals (20.4%), disposal of surplus farm chemicals (17%), preparation of products for use in livestock (13.4%).

The role of vibrissae in behavior: a status review.

Vibrissae or tactile hairs are an important part of the tactile sensory apparatus of many mammals. A wide range of suggested functions found in the literature include food acquisition, prey attack, aggression and attack behavior, facial expression in intraspecies communications, dispersion of pheromones, maintaining head position in swimming, and a wide range of environmental monitoring (e.g., current detection in water, wind direction on land). There is little work done specifically on domestic animals or their feral relatives. Work on the tactile senses in general and vibrissae in particular is an open field of study. A set of general questions for study of vibrissa function in domestic animals is presented.