Charles R. Curtis

Path model of individual-calf risk factors for calfhood morbidity and mortality in New York Holstein herds☆

Abstract A multivariable path model (based on field data) of the interrelationships between individual-calf management practices and calfhood morbidity and mortality was constructed. Data were from 1171 heifer calves who lived greater than 24 h and were kept as replacements on 26 farms in the vicinity of Cornell University between July 1983 and 30 April 1985. Outcomes of interest were the first occurrence of the following clinical signs as diagnosed by farmers on standardized check-off forms: scours/diarrhea within 14 days of birth and from 15 to 90 days of age; dull, listless, droopy ears or off feed within 90 days of birth; cough, runny nose or eyes or trouble breathing within 90 days of birth; death prior to 90 days of age. Risk-factor data were based on information collected on standardized check-off sheets for each individual calf. Place of birth, calving events, colostrum administration and initial housing, as well as any prophylactic or other treatments, were recorded by the person caring for the calf. Logistic regression was used to estimate the path coefficients (i.e. relative risks: RRs). Time ordering of morbidity and mortality variables inherent in the data were used to construct the hypothesized model. Management appeared to affect directly and indirectly (via scours within 14 days of birth, scours from 15 to 90 days of age and dullness) the risk of respiratory illness. Being born in loose housing increased risk of both scours within 14 days of birth and scours from 15 to 90 days of age. Calves tied versus being housed in group pens were at increased risk of scours within 14 days of birth (RR=1.5). Calves of first-calf heifers were at increased risk of respiratory illness (RR=1.5), and calves whose dams were vaccinated against Escherichia coli were at decreased risk of death (RR=0.2). Calves with scours within 14 days of birth, dullness and scours from 15 to 90 days of age were at increased risk of respiratory illness (RR=2.5, 7.7 and 3.1, respectively). Although respiratory illness had the largest direct RR for death, dullness actually had a greater effect on risk of death because of its additional indirect association through respiratory illness (dullness increased risk of respiratory illness which then increased risk of death).

Descriptive epidemiology of calfhood morbidity and mortality in New York holstein herds

Abstract A prospective cohort study was done of 1171 Holstein heifer calves born from July 1983 to 30 April 1985 and kept to be raised as replacements in 26 commercial dairy herds in the vicinity of Cornell University. The objectives were to examine the interrelationships of management with morbidity and mortality at both the individual and herd level, and the long-term effects of morbidity on survical. The study population, data collection and processing methods, and the descriptive epidemiology (e.g. age and temporal distributions) of calfhood morbidity, mortality and treatments are described. The clinical signs (as diagnosed by farmers on standardized check-off forms) studied were: scours/diarrhea within 14 days of birth and from 15 to 90 days of age; dull, listless, droopy ears, or of feed within 90 days of birth; cough, runny nose or eyes, or trouble breathing within 90 days or birth; death prior to 90 days of age. The crude incidence rates of scours within 14 days of birth, scours from 15 to 90 days of age, dullness, respiratory illness and death were 9.9, 5.2, 7.7, 7.4 and 3.5 per 100, respectively, for the period of study. The only seasonal effect on incidence was for scours within 14 days of birth (which was higher in winter compared to summer across both years of the study). The incidence rates of scours from 15 to 90 days of age, dullness and respiratory illness were higher in the first year of the study compared to the second year; there was no effect of season on these outcomes. The mortality rate within 90 days of birth was not related either to year of study or season. The median ages at occurence of scours within 14 days of birth, scours from 15 to 90 days of age, dullness, respiratory illness, and death were 6, 30, 16, 25 and 28 days, respectively. Risks of clinical signs were highest in the first week of life, while risk of death was highest in the third week. Calves with only scours within 14 days of birth or only scours from 15 to 90 days of age were treated primarily with oral electrolytes, oral antibiotics, and restriction of milk consumption. Calves with only dullness or only respiratory illness were treated primarily with intramuscular antibiotics. The median herd incidence rates of scours within 14 days og birth, scours from 15 to 90 days of age, dullness, respiratory illness and death were 5, 0, 0, 0 and 0 per 100, respectively, during the periof of the study.

Ordinary versus random-effects logistic regression for analyzing herd-level calf morbidity and mortality data

Abstract The analysis of data from populations organized into groups is frequently complicated by cluster (herd) effects. When present, clustering effects influence the probability of a health-related event in a way that is not readily accounted for with classical fixed-effects models (including unconditional logistic regression). Clustering introduces an additional source of (extra-binomial) variation into a logistic regression model, violating the independence and identical-distribution assumptions, and leading to biased variance estimators and spurious statistical significance. One remedy is to treat herd effects as random effects, so that variation even from unmeasured and unmeasurable (but clustered) sources can be accounted for. The logistic-normal regression model is introduced as one such random-effects model; its application is demonstrated using data from a previously described study of calfhood morbidity and mortality in 25 New York dairy herds.

Effects of calfhood morbility on long-term survival in New York Holstein Herds

Abstract The effect of calfhood morbility on long-term survival (i.e. not dying or being sold) was evluated in 1069 heifer replacement calves in 24 herds in the vicinity of Cornell University. These calves were born between July 1983 and 30 April 1985, and were alive at 90 days of age. The specific calfhood morbility variables of interest were the first occurrence of the following clinical signsn as diagnosed by farmers on standardized check-off forms: (1) scours/diarrhea within 14 days of birth and from 15 to 90 days of age; (2) dull, listless, droopy ears or off feed within 90 days of birth; (3) cough, runny nose or eyes, or trouble breathing within 90 days of birth. Outcomes of interest were (1) death, (2d (2) sale and (3) combined death and sale (exit) after 90 days of age and before 11 February 1986. Age-specific incidence rates and survival functions for the outcomes of interest were calculated using life-table techniques. The effects of the morbility variables on survival were modeled using Coxs proportional hazards model. The crude incidence rates of death and sale after 90 days of age for the study period were 2.5 and 5.1 per 100, respectively. Age-specific mortality rates after 90 days of age indicated a rise in mortality at ∼ 1 year of age. The age-specific incidence rate of sale after 90 days of age increased between 1 and 2 years of age (corresponding to the age at which farmers start reducing heifer numbers to the number needed for replacements). There were no significant effects of season and associated year of birth (birth cohort effect) on the survival distributions of death and sale after 90 days of age, although there was on exit after 90 days of age. The only illness related to survival was dullness within 90 days of birth, which increased the hazards rate for death after 90 days of age 4.3-fold above that for heifers without dullness within 90 days of birth, after adjusting for herd and season and associated year of birth.

Path model of herd-level risk factors for calfhood morbidity and mortality in New York Holstein herds

Abstract A multivariable path model of the interrelationships between herd-management policies and herd-level calfhood morbidity and mortality was constructed. Data were from 25 farms in the vicinity of Cornell University, and comprised records from all 962 heifer calves born between 1 November 1983 and 30 April 1985 who lived longer than 24 h and were kept to be raised as replacements. Outcomes of interest were herd-season-year cumulative incidences of the following clinical signs as diagnosed by farmers and recorded on check-off forms: scours/diarrhea within 14 days of birth and from 15 to 90 days of age; dull, listless, droopy ears or off feed within 90 days of birth; cough, runny nose or eyes, or trouble breathing within 90 days of birth; death prior to 90 days of age. Herd-management risk-factor data were based on information from a personal-interview questionnaire administered by the senior author, and were measures of policy rather than of what individual calves actually experienced. Random-effects logistic regression (logistic-normal regression) was used to model incidence rates and to account for the extra-binomial variation due to heterogeneity of disease risk between herds. Herd-management policies directly and indirectly ‘affected’ the herd cumulative incidences of morbidity and mortality. Farms reporting that calves had periodic, or no, access to water (compared to continuous availability) had less scours from 15 to 90 days of age and less dullness. Damp (versus dry) bedding, tying calves ((by a collar and tether) versus not tying calves)), use of individual-calf hutches, and feeding milk once daily (versus twice daily) all ‘increased’ the chance of scours within 14 days of birth. Use of individual-calf stalls ‘increased’ the chance of scours from 15 to 90 days of age, but ‘decreased’ the chance of mortality. Feeding high-moisture ear corn to calves ‘increased’ the chance of scours from 15 to 90 days of age, suggesting that some of the later scours were of nutritional origin. Scours prior to 14 days of age and dullness (but not later scours) ‘increased’ the chance of respiratory illness. Only respiratory illness directly ‘increased’ the chance of mortality. Scours prior to 14 days of age and dullness were related to mortality indirectly via respiratory illness. Scours from 15 to 90 days of age were not related to mortality.