Applied epidemiology: another tool in dairy herd health programs? K Frankena, Jp Noordhuizen, En Stassen To cite this version: K Frankena, Jp Noordhuizen, En Stassen. Applied epidemiology: another tool in dairy herd health programs?. Veterinary Research, BioMed Central, 1994, 25 (2-3), pp.234-238. <hal-00902201> HAL Id: hal-00902201 https://hal.archives-ouvertes.fr/hal-00902201 Submitted on 1 Jan 1994 HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
!!!-! - Staplylococcus aureus and Streptococcus agalactiae intramammary infections. Prev Vet Med 6, 127-142 Jensen NE (1985) Epidemiological aspects of human/animal interrelationship in GBS. Antibiot Chemother 35, 40-48 Klastrup 0 (1978) Hvor star mastitisbekxmpelsen i dag? (Status of the mastitis control to day). Dansk VetTidsskr 61, 273-281 Martin SW, Meek A, Willeberg P (1987) Veterinary epidemiology: Principles and methods. Iowa State University Press, Ames, 10, USA, 343 p SAS Institute Inc (1988) SASlSTAT TM User s Guide, Release 6.03 Edition, Cary, NC, USA SAS Institute Inc (1990) SASO, Technical Report P-200, SAS/STAT 9 Software, CALIS and LOGISTIC Procedures, Release 6.04, Cary, NC, USA Vet Res (1994) 25, 234-238 Elsevier/INRA Applied epidemiology: another tool in dairy herd health programs? K Frankena JP Noordhuizen EN Stassen 1 Agricultural University, Dept Anim Husbandry, PO Box 338, 6700 AH Wageningen; 2 State Univ Utrecht, Dept Herd Health and Reproduction, PO Box 80 151, 3508 TD Utrecht, The Netherlands Summary― Data bases of herd health programs concern data from individual animals mainly. Several parameters that determine herd performance can be calculated from these programs, and by comparing actual values with standard values, areas for further improvement of health (and production) can be advised. However, such advice is usually not backed up by the proper statistical analyses. Moreover, data concerning the environment of the animals are not present and hence advice concerning multifactorial diseases are based on common knowledge and experience. Veterinary epidemiology offers methods that might improve the value of herd health programs by identification and quantification of factors and conditions contributing to multifactorial disease occurrence. Implementation of these methods within herd health programs will lead to more scientifically sound advice. epidemiology / dairy herd health program Résumé ― L épidémiologie appliquée : un outil dans les suivis de troupeaux laitiers. Les suivis de troupeaux laitiers s appuient sur des bases de données contenant les informations relevées au niveau individuel. Celles-ci sont traitées pour obtenir les valeurs des paramètres de performances et de santé du troupeau qui sont alors comparées à des valeurs-objectifs ou à des références. Cette comparaison n est pas réalisée au moyen d analyses statistiques appropriées. De plus, les données relatives à l environnement des animaux ne sont pas enregistrées. L épidémiologie vétérinaire propose des méthodes qui permettent d améliorer les suivis de troupeaux en réalisant l identification et la quantification des facteurs associés à la fréquence des maladies multifactorielles. La mise en oeuvre de telles méthodes permet de conduire des suivis reposant sur des bases plus scientifiques. épidémiologie / médecine préventive en troupeau laitier Correspondence and reprints.
INTRODUCTION Dairy herd health programs have been implemented in various countries during the last decades. These programs focus on the improvement of farm net profit by increasing productivity and reducing costs (Noordhuizen et al, 1985). The outcome of the herd health program is adequate to define a problem within a herd with regard to the objectives set. Subsequently, it is possible to detect high risk groups by calculation of group specific incidences. The latter can be used to indicate potential areas of attention. Such an approach is by definition retrospective and advice always comes too late to prevent the problem. Moreover, the quality of advice concerning multifactorial diseases is based on both the output of herd health programs, and the experience and knowledge of the advisor. Extending herd health programs by adding information about housing, feeding, hygiene, management, etc, makes it possible to identify and quantify factors and conditions contributing to multifactorial disease occurrence. In a more to cal- prospective way, it is then possible culate the probability in a non-problem herd to get a specific problem given a set of factors or conditions present on the farm. In this paper, prevalences of claw disorders are presented from cross-sectional and longitudinal surveys concerning lameness in cattle. Secondly, results of a risk analysis concerning sole haemorrhages in calves are shown to indicate what can be done with risk analyses and thirdly, a scheme showing how this epidemiological method can be used in combination with herd health programs is discussed. MATERIALS AND METHODS In cooperation with the Department of Herd Health and Reproduction of the Faculty of Veterinary Medicine, Utrecht, the Netherlands, 5 studies have been carried out concerning lameness in cattle (table I). In the Netherlands, cattle are housed from October/November until April/May. Studies 1 to 4 were cross-sectional surveys and study 5 was longitudinal in nature as animals were observed 3 times. In studies 1-3, observations were made during the routine herd claw trimming. In studies 1-5, both claws of each hind leg were examined and the clinical and/or subclinical findings were recorded according to a list of diagnoses based on the description by Espinasse et al (1982). The results given in this paper are restricted to sole haemorrhages (SH) often referred to as signs of subclinical laminitis. Information about environmental data was collected by means of a questionnaire. Data were analyzed using multivariate logistic regression from which odds ratios (relative risk estimators) can be easily calculated (Hosmer and Lemeshow, 1989). Detailed information about the statistical analysis and model building strategy is given by Frankena et al (1992).
RESULTS Prevalences regarding dermatitis interdigitalis (DI), dermatitis digitalis (DD) and sole haemorrhages (SH) are shown in table II. Clinical lameness was sporadically encountered. It appeared that large between-herd variation existed for most of the claw disorders (fig 1 for SH in study 4). Subsequently, an explanation of this variation by factors extracted from the questionnaire was attempted. Results of a risk analysis concerning sole haemorrhages (study 4) are shown in table III. Identical analyses can be done for other claw disorders in all studies. It shows that both animal and herd level data are related to SH. DISCUSSION The main purpose of this paper was to indicate some possibilities of applying epidemiological methods within herd health programs. In herd health programs, only clinical disorders are registered. The relation between occurrence of clinical disorders and environmental factors could not be evaluated in this paper because the information about the latter was not available, and collecting that information implied visiting quite a number of herds (questionnaire) which is rather time-consuming. Alternatively, it was chosen to illustrate how epidemiological methods can be applied in herd health programs with already existing data from a cross-sectional study concerning subclinical sole haemorrhages in calves. However, this type of study is not
the best for assessing causality. Causality should be evaluated in a more scientific approach using experimental and cohort type studies. But even now, relevant riskincreasing or risk-decreasing factors might be identified, and more or less quantified, and areas of primary attention and/or priorities with regard to further research can be indicated. Prevalence figures showed that subclinical claw disorders are present at high levels. The fact that clinical lameness is hardly encountered in cross sectional surveys might be due to the routine herd claw trimming that is performed regularly on these farms and the immediate action farmers take in case lameness occurs clinically. However, large between-herd variation existed at the subclinical level, not only in case of sole haemorrhages, but also for other disorders (Frankena et al, 1994). This implies a management influence. Subsequently, the risk analysis showed that the between-herd variation is, at least partly, explained by managerial and environmental conditions and those factors might be used
In a next step, an economic module might be added and financial consequences of the actions can be calculated. The interpretation of the statistical results requires skills and knowledge from the herd health advisor; for example, table III presents at first sight some questionable results. For a more comprehensive discussion about this aspect see Frankena et al (1992). Therefore, some basic conditions should be satisfied, one of them being that continuing education programs are developed or adapted in order to improve that knowledge. Moreover, supporting service units should be available to gain optimal benefit of the proposed system. REFERENCES as priorities for increasing the health standard. How to make such a risk assessment possible within herd health programs is illustrated using the activity pattern of an implemented system (fig 2, Noordhuizen et al, 1985). The block of farm inspection data should be extended with more data concerning management and environment. The analysis part should include appropriate statistical modules from which easily interpretable outcomes (like relative risks or odds ratios) can be extracted. If these odds ratios are available, then the benefit of manipulation of risk factors, in terms of reduction in disease occurrence, can be calculated. Espinasse J, Savey M, Thorley CM, Toussaint Raven E, Weaver AD (1982) ColourAtlas on Disorders of Cattle and Sheep Digit, International Terminology. Editions du Point Vétérinaire, Maisons-Alfort, France, 48 pp Frankena K, Stassen EN, Noordhuizen JP et al (1991) Prevalence of lameness and risk indicators for dermatitis digitalis (Mortellaro disease) during pasturing and housing of dairy cattle. Proc Ann Meet Soc Vet Epidemiol Prev Med, Royal Veterinary College, University of London, April 17-19, 107-118 Frankena K, Keulen KAS van, Noordhuizen JP et al (1992) A cross-sectional study into prevalence and risk indicators of digital haemorraghes in female dairy calves. Prev Vet Med 14, 1-12 Frankena K, Keulen KAS van, Noordhuizen JP et al (1994) A cross-sectional study of prevalence and risk factors of dermatitis interdigitalis in female dairy calves in the Netherlands. Prev VetMed 17, 137-144 Hosmer DW and Lemeshow S (1989) Applied Logistic Regression, John Wiley & Sons, New York, 1 st ed, 307 Noordhuizen JP, Wilbrink HJ, Buurman J (1985) Index lists provided by the VAMPP computer programme for dairy herd health and production control. Vet Q 7, 3-30 Smits MCJ, Frankena K, Metz JHM, Noordhuizen JP (1992) Prevalence of digital disorders in zero-grazing dairy cows. Livest Prod Sci 32, 231-244