Alan H. Meek

Disease, production and culling in Holstein-Friesian cows. I. The data

Abstract A two-and-a-half-year study was undertaken in 32 commercial dairy herds within a 50-mile radius of Guelph, Ontario, in order to provide a comprehensive evaluation of relationships amongst diseases, production parameters and risk of removal in Holstein-Friesian cows. Diseases consisted of common clinical diseases as well as two subclinical diseases (mastitis and ketosis). The data set consisted of a total of 2875 lactation records from 2008 cows. The means and standard deviations of some of the production parameters were: age 4.9 ± 2.5 year, breed class average for milk 140 ± 26 BCM units, milk production per day of life 18.0 ± 3.8 kg and days to conception 118 ± 60 days. Lactational incidence rates of the more common diseases were: reproductive tract infections 0.182, mastitis 0.168, cystic ovaries 0.104, subclinical ketosis 0.096, retained placenta 0.086 and ketosis 0.074. The average lactation mean log somatic cell count was 179 000 cells/ml. The times postpartum to the first diagnosis of all disease conditions are presented.

Associations of clinical findings, bacteriological and histological results of endometrial biopsy with reproductive performance of postpartum dairy cows

Abstract Ninety-seven cows from a commercial Holstein-Friesian dairy herd were studied. Cows were classified as having exhibited good or poor reproductive performance. Unconditional relationships were examined between prostaglandin administration, clinical and palpation findings and histological and bacteriological culture results of endometrial biopsies taken at Days 26 and 40 postpartum and the reproductive performance variable. Variables unconditionally associated with performance were then offered to logistic multiple-regression models to quantify the diagnostic efficacy of these findings. Logistic regression models were calculated for histological, bacteriological, and clinical findings (at Days 26 and 40) first separately and then in a combined model. Cows positive for Actinomyces pyogenes or anaerobic bacteria were more likely to be poor reproductive performers. Variables from Day 26 postpartum were more likely to be conditionally associated with reproductive performance than findings from Day 40. Ovarian findings from Day 26 were associated with subsequent reproductive performance only when the location of the structure relative to the previously gravid horn was considered. In the final combined model, increased inflammation in the stratum compactum and more areas of epithelium increased the likelihood of poor reproductive performance, and increased numbers of lymphocytic foci in the stratum compactum and larger follicles on the gravid-side ovary reduced the risk of poor reproductive performance. This model included information on 55 cows and achieved 84% correct predictions, with a sensitivity (for poor cows) of 92% and a specificity (good cows) of 77%.

Health problems in selected Ontario Holstein cows: frequency of occurrences, time to first diagnosis and associations.

Abstract An observational study of 34 Holstein herds from 7 southwestern Ontario countries was undertaken in order to evaluate common clinical diseases. The data consisted of 2204 lactations, and were collected over 2.5 years (October 1980–June 1983). Lactational incidence risks of seleted diseases and median time post-partum to first clinical diagnosis were: milk fever 0.056 (0 day), udder oedema 0.060 (1 day), foetal membranes retained 48 h or more 0.077 (2 days), ketosis 0.033 (18 days), vaginitis 0.022 (23 days), metritis 0.107 (26 days), mastitis 0.242 (49 days), teat injuries 0.032 (68 days), foot and leg problems 0.036 (89 days), ovarian cysts 0.050 (90.5 days), and abortions 0.012 (296 days). Results of the present study were consistent with the results of other similar studies. Cows younger than 3 years old were at very low risk of experiencing milk fever. Evidence was found that a potential causal association existed between retained placenta and each metritis and ovarian cysts, between teat injuries and mastitis, and between milk fever and ketosis.

Interrelationships among health problems and milk production from consecutive lactations in selected Ontario Holstein cows.

Abstract Clinical health problems and milk production data from Holstein cows from 34 commercial south-western Ontario dairy herds, for which two subsequent lactations were available, were investigated. Associations of health problems and milk production from a previous lactation with the occurrence of health problems in the following lactation were investigated using a stepwise logistic regression procedure. Milk fever, teat injuries and mastitis were the only problems that were associated with previous milk production. Retained placenta, mastitis and ovarian cysts in a lactation were predictors of the same events happening in the subsequent lactation. Long lactations without ovarian problems increased the risk of udder oedema occurring during the following lactation. Reproductive infectious problems showed less capacity to predict health problems in the subsequent lactation than did non-infectious problems. Lactations with low milk production and older animals increased the risk of difficulty at the following calving.

Disease, production and culling in Holstein-Friesian cows VI. Effects of management on disease rates

Effects of management practices on rates of disease in Holstein—Friesian cows in 32 commercial dairy herds located within a 50-mile radius of Guelph, Ontario were investigated in a 212 year study. A personal survey questionnaire was used to collect information on management factors related to housing, nutrition and disease control procedures. The logit transform was applied to the herd disease rates and multiple linear regression used to identify management factors related to disease rates. Large values of the variable “frequency of ration balancing” identified larger more progressive farms which may have had superior disease recording systems. In a second regression analysis, “frequency of ration balancing” was entered into the regression equation prior to step-wise selection of the other variables. The ability of the selected variables to predict herd rates of disease was only moderate (r2 = 0.24 to 0.57). However, of the variables available for selection, those potentially reflecting the farmers attitude toward the dairy herd (e.g., amount of attention paid to individual cows) appeared to be important.

Interrelationships of health problems and age on milk production in selected Ontario Holstein cows.

Abstract The interrelationships of common clinical health problems with milk production in Holstein cows from 34 commercial south-western Ontario dairy herds were investigated. The data used were from 1588 lactations. Analyses were conducted on a data file containing all lactations and on a data file containing only lactations having 305 days in milk (305-d) standardized production. Stepwise multiple linear regressions (controlling for herd) were used to estimate the effect of health problems and age on milk production. The effects were investigated with respect to milk production of concurrent and subsequent lactations. Age was associated with a decrease in milk quality and an increase in risk of health problems and total milk produced. Milk fever, retained placenta, abortion and udder problems were detrimental to milk production. Metritis and ovarian cysts had a positive effect on concurrent milk production through an increase in days in milk and ovarian cysts also had a direct effect. The positive effect of ovarian problems on total milk yield in the concurrent lactation was followed by a detrimental effect on milk production in the subsequent lactation. Using only standardized milk production measures introduced a bias in the results, because the effect of short lactations was ignored.

Veterinary epidemiology: challenges and opportunities in research

Abstract It is essential that all groups periodically review their mandate, achievements and direction in light of the changing needs of society. It is suggested that greater use of the naturalistic paradigm in epidemiological research is warranted, as the scientific paradigm, while serving the ‘hard’ sciences well may have shortcomings when it comes to understanding multifactorial issues, particularly if one of these is human behaviour. The challenge then is to ensure on an ongoing basis that epidemiological research remains relevant to societal concerns, and to seek and apply methods that lead to knowledge that will have a positive impact. Epidemiology can contribute to societal need both through research and through the resultant practical application of its findings. The former necessitates that the scientific discipline of epidemiology be strong, and that its methods continue to advance. The latter is more pragmatic and requires integration with other perspectives, for example, economics and management science. There is danger that the latter harmonization of perspectives be considered synonymous with the discipline of epidemiology. Preference here is to use the term population medicine for this. Without this distinction we are in danger of becoming increasingly unfocused. With it we can evolve the discipline of epidemiology and plan our research programs so that they remain relevant to societys changing needs.

Associations between lipoarabinomannan enzyme-immuno-assay test results for paratuberculosis and farm-management factors

Abstract Data from an epidemiological study of paratuberculosis in dairy cattle in Ontario, Canada, were used to identify farm-management factors that were associated with lipoarabinomannan antigen enzyme-immuno-assay (LAM-ELISA) test results for paratuberculosis. The study was conducted in two phases. Phase 1 herd selection was cross-sectional and involved 304 herds. Herd average LAM-ELISA optical density was regressed on management-practice data using least-squares regression, and was associated negatively with teat dipping and the number of years the herd had been established. Phase 2 herd selection was carried out in a case-control manner and involved 120 of the herds that had participated in Phase 1. The 60 highest-risk and 60 lowest-risk herds for paratuberculosis, as defined by Phase 1 LAM-ELISA test results, were visited for sample and data collection during Phase 2. In general, there were no striking differences in management practices observed in a simple comparison of high- to low-risk herds. When herd status was regressed (using logistic regression) on management practices, herd status was associated with calf housing, calf feeding, use of teat dip, adult nutrition, water sources and contact between animals. However, the direction and detail of some of the observed associations were opposite to those that were expected. The results of this analytic observational study were inconclusive. Nevertheless, they provide material to help guide future investigations.

Tail-tip necrosis in Ontario feedlot cattle: two case-control studies.

In 1987–1988, two case-control studies were designed to identify possible determinants and to develop hypotheses regarding causes of tail-tip necrosis in feedlot cattle. In Study A, 13 solid- and 25 slatted-floor farms were compared. Eleven and 14 of the 25 slatted-floor farms were characterized as low- and high-incidence tail-tip necrosis farms, respectively, and were compared in Study B. Data regarding barn structure, management of the herd, health and behavior of the cattle were collected. There was no case of tail-tip necrosis diagnosed in solid-floor barns in 1987. Cattle housed in solid-floor barns were more likely to have access to free-choice minerals, were inspected more often per week, spent more days in the barn and were kept at a lower stocking density. No animal housed in a solid-floor barn was observed to get up or lie down abnormally. In the slatted-floor barns, abnormal locomotor patterns were observed 20% and 25% of the times animals got up and lay down, respectively. Study B showed that space allotment per animal was higher in the low- than in the high-incidence, slatted-floor barns (1.90 and 1.74 m2 per head respectively). Also, cattle housed in slatted-floor barns with a high incidence of tail-tip necrosis, got up abnormally more often (P<0.05). Results suggest that there might be a connection between space allotment, abnormal rising and tail-tip necrosis.

Hatchery-level predictive values for infectious pancreatic necrosis virus and Aeromonas salmonicida in Ontario, Canada.

The probability and uncertainty of correctly classifying the IPNV and Aeromonas salmonicida status of fish-rearing and natural sites in Ontario were estimated through Monte Carlo simulations. Propagating several uncertain inputs showed the extent to which natural variability and our present lack of knowledge affect the probability of site misclassification. For the scenarios investigated, the site-level negative predictive values (SNPVs) were high and fairly constant. The site-level positive predictive values (SPPVs) - given a test specificity ranging between 0.999 and 1.0 - were much lower, more variable, and highly affected by cut-off point and sample size. Substantial uncertainty resides in classifying the pathogen status of test-positive sites, whereas much less uncertainty resides in classifying pathogen status of test-negative sites.