Introduction
Bovine respiratory disease complex is the most common cause of illness and death of feedlot cattle. In 2011, the National Animal Health Monitoring System reported that 16.2% of cattle on feedlots with a capacity of ≥ 1,000 head were affected by respiratory disease after arrival on a feedlot.1 Depending on the pathogens and the severity of illness, death from BRDC can occur between 24 and 36 hours after the onset of clinical signs (eg, discharge from the eyes, mouth, and nose; coughing; lethargy; decreased appetite; and rapid, shallow breathing).2 In addition to the negative impact on cattle health, BRDC also reduces profitability through decreased average daily gain in body weight and carcass weight and increased treatment costs.3,4,5,6,7,8
The causes of BRDC are multifactorial, including environmental stressors, animal immunocompetence, and respiratory pathogens.9 Analyses of the morbidity and mortality rates of BRDC indicate that risk factors for BRDC-related illness or death include lower body weight at feedlot entry,10,11,12 body weight loss during transport,13 sex,10,11 commingling cattle from multiple sources,10,14,15 longer shipping distance,10,15 lack of preconditioning,14,16,17 season of the year,11,12,13 cohort size,11,15 and DOF.11 Although environmental stress is an important risk factor for BRDC, few studies11,18,19,20,21 include an investigation of their association with BRDC. Extreme weather conditions (eg, extreme temperatures) may increase stress and weaken the immune system, leading to a higher incidence of disease.21,22,23 However, most of these studies21,22,23 do not include important cattle and management factors that may confound or modify those associations. For example, Cusack et al21 reported the associations among weather conditions and BRDC incidence, but the analysis was not adjusted for important covariates, such as sex or BWA. Cernicchiaro et al11 included covariates for sex, BWA, cohort size, and designated risk at arrival but did not include information on other important covariates, such as the transport distance and region where cattle were purchased, and limited the analysis to those cattle placed during the fall season. Also, whether the development of BRDC in cattle after their arrival at the backgrounding location or feedlot is associated with weather conditions at the purchase site is unknown. Activities around the time of purchase may represent stressors as cattle are moved and commingled and then transported long distances to a backgrounding location or feedlot.24 Extreme weather conditions during and after transport may exacerbate the effects of shipping stress, including food and water deprivation, overcrowding, unexpected noise, and poor air quality.25 The first week at the feedlot or backgrounding location represents another stressor because cattle may have to adapt to new environmental conditions, dietary changes, exposure to new pathogens, and new social dominance. Additionally, cattle may undergo other stressful processing procedures such as castration, vaccination, and dehorning.25
The objective of the study reported here was to quantify associations between weather conditions on the purchase day at the purchase site or during the first week at a backgrounding location or feedlot and cohort-level (lot-level) BRDC cumulative mortality incidence in calves during the first 60 DOF at multiple sites of a large commercial cattle feeding operation. This study also included an investigation of whether the associations between weather conditions and BRDC death were modified by other factors, such as cattle characteristics, purchase region, commingling, and transport distance from the purchase site.
Materials and Methods
Institutional animal care and use committee approval was not needed for this retrospective study because the data were obtained through an existing database. The checklist for Strengthening The Reporting Of Observational Studies in Epidemiology was used to ensure thoroughness of the reporting.26
Feedlot data
Data were compiled from 2015 to 2018 for a large commercial cattle feeding operation (with approx 150,000 to 250,000 purchased head of cattle/y) based in the Midwestern United States. After purchase, each lot of calves was shipped to 1 of 89 feeding locations (backgrounding locations, n = 87; feedlots, 2) owned by the commercial operation and were managed similarly. The database included information routinely collected on management characteristics, morbidity and mortality incidence, and cattle characteristics. Cleaning and management of data were performed with a commercial software programa after exporting source data from the feedlot management software.
A cohort (lot) was defined as a group of cattle initially housed together after arrival to the feeding location. Cumulative BRDC mortality incidence was defined as the proportion of cattle in each cohort that died because of BRDC within the first 60 DOF. Cohorts were excluded if they had < 60 DOF. This was done to capture the majority of BRDC-related deaths and deaths that were more likely to be affected by weather conditions on the day of purchase (day P) and during the first week after arrival, compared with deaths that occurred later. The case definition for BRDC-related death included the following diagnoses: pneumonia, except atypical interstitial pneumonia; BRDC; chronic lung abscesses; pleuritis; undifferentiated fever; and infectious bovine rhinotracheitis. Cohorts were excluded when data were missing in observations at the calf level, mean BWA was < 136 kg or > 409 kg, or cohorts included yearlings. Cohort-level covariates of interest included the number cattle in each cohort, mean BWA (categorized in quartiles [136.0 to < 254.7 kg, 254.7 to < 275.5 kg, 275.5 to < 299.5 kg, and 299.5 to 409 kg]), transport distance from the purchase site to feedlot or backgrounding location (categorized in quartiles [< 520.1 km, 520.1 to < 767.8 km, 767.8 to < 1,082.4 km, and ≥ 1,082.4 km]), season of purchase (spring [March, April, and May], summer [June, July, and August], fall [September, October, and November], and winter [December, January, and February]), year of purchase, commingling (1, 2, 3, or ≥ 4 purchase groups commingled within a cohort), and weather conditions. All cattle underwent standard processing procedures (eg, metaphylaxis, vaccination, de-worming, and implant protocols) dictated by the commercial cattle operation; however, some variation in processing procedures by cattle type and over time across all locations was likely.
Location and weather data
A geospatial application programming interfacea was used within a software programb to determine the latitude and longitude of the purchasing and feeding sites with the mqgeocode command from the mqtime package.27 The geodist commandc was used to calculate the geometric distance between the locations. An application programming interfaced with the darksky package in a commercial software programe was used to collect weather condition data on day P and on each day during the first week at the feeding location. Collected weather condition data included the minimum and maximum air temperatures, minimum and maximum apparent air temperatures (ie, air temperature that is perceived by people, with apparent temperature affected by humidity and wind speed), accumulated amount of precipitation, wind speed, relative humidity, UV index, and dew point.
Data aggregation
To perform a cohort-level analysis, data that were measured at the individual calf level or the purchasing group level were aggregated up to the cohort level (ie, compositional modeling,28 with this type of analysis having been applied to disease prediction models in dairy cattle29,30). For example, a proportion of the sample population had individual-level entry body weights, so the body weights of the cattle within a cohort were averaged to yield a mean cohort-level body weight. Also, many cohorts were formed from groups of cattle (median, 3; range, 1 to 20) purchased from various nearby locations. Therefore, cattle within the same cohort were transported over various distances and likely experienced various weather conditions before transport. Mean transport distance was the mean calf-transport distance within a cohort. Weather condition data were calculated as the mean values within a cohort on day P, on the day of arrival at the backgrounding location or feedlot, and for each day up to day 6 after arrival. Data were further aggregated on the basis of various times as follows: day P (purchase from auction, before arrival at the feeding location [number of transport days: median, 2 days; range, 0 to 13 days]), day 0 (arrival at the feeding location), and days 1 to 2, 3 to 4, and 5 to 6 (after arrival at the feeding location). The amount of precipitation was calculated as the total precipitation during the time period, and all other weather condition data were calculated as the mean value during the time period. Distribution of the precipitation data was skewed right and included many zeros; therefore, the amount of precipitation was calculated as total precipitation, rather than as mean precipitation, to include extreme weather events. Also, the absolute changes in air temperature and relative humidity over the first week at each feeding location were calculated. The absolute change in air temperature was calculated by subtracting the lowest minimum daily air temperature from the highest maximum daily air temperature, and the absolute change in relative humidity was calculated by subtracting the lowest daily humidity from the highest daily humidity. The overall maximum and minimum air temperatures during the first week after arrival were also assessed.
Statistical analysis
The associations between weather condition data and BRDC mortality incidence were assessed by use of a mixed-effects multivariable count model that included a Poisson regression model or, if overdispersion was present, a negative binomial regression model. Random intercepts for year and feeding location were fitted in the model unless their variance estimates were negligible.
The dependent variable was the number of cattle within a cohort that died of BRDC within 60 DOF, and the independent variables were the number of cattle in each cohort, number of unique purchase groups within a cohort (commingling), weaned status (weaned vs not weaned on day P), mean BWA, purchase region (northern part of the United States and Canada vs southern part of the United States), purchase season, aggregated transport distance, and weather condition data. An offset term was included to adjust for the number of cattle in each cohort. All continuous variables were categorized in quartiles to avoid violating the linearity assumption, except for total precipitation that was categorized as a binary variable (1 = precipitation and 0 = no precipitation) because many days had no precipitation. Variables and their descriptions were summarized (Supplementary Table S1, available at: avmajournals.avma.org/doi/suppl/10.2460/ajvr.82.8.644).
Variables were screened in bivariable analyses before inclusion in the multivariable model. Pairwise comparisons between each variable were performed with the Pearson and Spearman correlation tests. If any 2 variables were strongly correlated (correlation coefficient greater than the absolute value of 0.7), 1 of the variables was removed from the analysis on the basis of biological plausibility and data completeness.31 Variables that were significantly (P < 0.2) associated with the outcome (incidence rate ratio of cumulative BRDC deaths) were retained for inclusion in the multivariable model.
Backward elimination was performed to select the variables for inclusion in the final multivariable model. After selection of the main effects, 2-way inter-actions between each weather condition variable and cattle or management characteristic, between each weather variable and purchase season, and between each pairwise set of weather variables were tested for inclusion into the models. Variables were retained when they were significantly (P ≤ 0.05) associated with the outcome, were part of a significant (P ≤ 0.05) interaction term, or were determined to be confounders in the analysis (ie, changed the regression coefficients of other variables by > 20%).32 Multicollinearity was assessed throughout model development by monitoring SE estimates for inflation and model instability (ie, large changes in regression coefficients and P values). Overdispersion was evaluated with a likelihood ratio test (used to compare the estimates from the Poisson and negative binomial regression models). Also, the observed versus expected counts were compared among the Poisson, negative binomial, zero-inflated Poisson, and zero-inflated negative binomial regression models. Incidence rate ratios and their 95% CIs were reported for each variable in the final model. Values of P ≤ 0.05 were considered significant.
Results
The final sample consisted of 3,339 cohorts that were composed of 545,866 cattle that were purchased from the fall of 2015 to the winter of 2018 from 216 purchase sites. Before arriving at the final number of cohorts, 33 cohorts were excluded because of missing precipitation data. Also, counts were low for the following variable categories: cohorts including females (n = 83), unweaned and weaned calves (10), calves purchased directly from a ranch (81), calves not receiving metaphylaxis (22), and cattle from both Canada and the northern and southern parts of the United States (6); therefore, these cohorts also were excluded because the low counts would make estimates for these categories unreliable to interpret.
The size of each cohort ranged from 8 to 703 calves (median, 159; median of the absolute deviations from the cohort-size median, 67). Overall, the number of calf deaths attributed to BRDC in the first 60 DOF was 12,505 (range, 0 to 42 calves/cohort [0% to 31.4%/cohort]; median, 2 [mean, 2.4%; SEM, 0.05%]; median of the absolute deviations from the median, 2). Cattle were transported 547.6 km (mean; SEM, 881.3 km). Mean (SEM) BWA was 277.1 kg (1.22 kg). Other characteristics of the cohorts were summarized (Table 1).
Frequency of various characteristics of cohorts (n = 3,339) of beef calves (range, 8 to 703 calves/cohort) that were purchased from the fall of 2015 to the winter of 2018 from 216 locations and then shipped to 89 feeding locations as part of a large commercial cattle feeding operation in the Midwestern United States.
| Characteristic | No. (%) |
|---|---|
| Weaned at purchase | |
| Yes | 2,521 (75.5) |
| No | 818 (24.5) |
| Purchase season | |
| Spring (March, April, May) | 1,131 (33.9) |
| Summer (June, July, August) | 560 (16.8) |
| Fall (September, October, November) | 600 (18.0) |
| Winter (December, January, February) | 1,048 (31.4) |
| Commingled purchase groups | |
| 1 | 841 (25.2) |
| 2 | 809 (24.2) |
| 3 | 620 (18.6) |
| ≥ 4 | 1,069 (32.0) |
| Purchase region | |
| Northern United States and Canada | 2,146 (64.3) |
| Southern United States | 1,193 (35.7) |
The final multivariable model included fixed effects for purchase season, transport distance, degree of commingling, mean BWA, range for absolute air temperature during first week at the feeding location, and, on day P, weaned status (weaned or not weaned), precipitation, mean wind speed, mean maximum air temperature, mean minimum air temperature, and interaction terms shipping distance–mean wind speed and mean BWA–mean maximum air temperature.
Cohorts that included weaned calves on day P had a significantly (P < 0.01) lower cumulative BRDC mortality incidence, compared with cohorts that included unweaned calves (Table 2). Cumulative BRDC mortality incidence increased as the degree of commingling increased, and cohorts that did not include com-mingled purchase groups had a significantly (P < 0.01) lower cumulative BRDC mortality incidence, compared with cohorts that included ≥ 3 commingled purchase groups. Cohorts that included ≥ 4 commingled purchase groups had a significantly (P < 0.01) higher cumulative BRDC mortality incidence, compared with cohorts that included 2 and 3 commingled purchase groups, and had a 40% higher cumulative BRDC mortality incidence, compared with cohorts that did not have commingled purchase groups.
Results obtained from a mixed-effects negative binomial regression model* for cumulative mortality incidence attributed to BRDC during the first 60 DOF for the calves of Table 1.
| Variable | IRR (95% CI) | P value |
|---|---|---|
| Weaned at purchase | ||
| No | REF | — |
| Yes | 0.63 (0.58–0.68) | < 0.01 |
| Shipping distance (km) | < 0.01 | |
| < 520.1 | REF | — |
| 520.1 to < 767.8 | 1.02 (0.82–1.27) | 0.86 |
| 767.8 to < 1,082.4 | 1.21 (0.98–1.50) | 0.08 |
| ≥ 1,082.4 | 1.77 (1.47–2.14) | < 0.01 |
| Commingled purchase groups | < 0.01 | |
| 1 | REF | — |
| 2 | 1.11 (1.01 —1.22) | 0.03 |
| 3 | 1.19 (1.08—1.32) | < 0.01 |
| ≥ 4 | 1.40 (1.27—1.55) | < 0.01 |
| BWA (kg) | 0.13 | |
| 136.0 to < 254.7 | REF | — |
| 254.7 to < 275.5 | 0.98 (0.82—1.16) | 0.80 |
| 275.5 to < 299.5 | 0.99 (0.84—1.18) | 0.92 |
| 299.5 to 409.0 | 0.83 (0.69—1.00) | 0.054 |
| Precipitation on day P | ||
| No | REF | — |
| Yes | 0.86 (0.76—0.97) | 0.02 |
| Mean wind speed on day P (km/h) | 0.30 | |
| < 6.1 | REF | — |
| 6.1 to < 9.2 | 0.96 (0.79—1.18) | 0.72 |
| 9.2 to < 12.9 | 0.85 (0.69—1.03) | 0.10 |
| ≥ 12.9 | 0.90 (0.74—1.09) | 0.27 |
| Mean maximum air temperature on day P (°C) | < 0.01 | |
| < 7.4 | REF | — |
| 7.4 to < 16.4 | 1.34 (1.13—1.60) | < 0.01 |
| 16.4 to < 26.7 | 1.43 (1.20—1.70) | < 0.01 |
| ≥ 26.7 | 1.71 (1.41—2.07) | < 0.01 |
| Absolute air temperature change during first week after arrival (day 0 [arrival] to day 6; °C) | < 0.01 | |
| < 1.9 | REF | — |
| 1.9 to < 5.3 | 1.12 (1.03—1.22) | 0.01 |
| 5.3 to < 9.4 | 1.04 (0.95—1.13) | 0.42 |
| ≥ 9.4 | 0.95 (0.87—1.05) | 0.33 |
| Purchase season | < 0.01 | |
| Spring (March, April, May) | REF | — |
| Summer (June, July, August) | 1.33 (1.17—1.51) | < 0.01 |
| Fall (September, October, November) | 1.46 (1.33—1.60) | < 0.01 |
| Winter (December, January, February) | 1.02 (0.93—1.12) | 0.72 |
| Transport distance times mean wind speed on day P† | < 0.01 | |
| Mean BWA times mean maximum air temperature on day P‡ | < 0.01 | |
| Intercept | 0.02 (0.016—0.03) | < 0.01 |
Model included an offset term (ln [number of calves per cohort]) to adjust for the variable number of calves in each cohort. The estimate for random intercept variance for feeding location was 0.15 (95% CI, 0.10 to 0.23), and the overdispersion parameter was 0.39 (95% CI, 0.13 to 0.43).
See Figure 1 for graphical depiction of data regarding the interaction term of transport distance–mean wind speed on day P.
See Figure 2 for graphical depiction of data regarding the interaction term of mean BWA–mean maximum air temperature on day P.
— = Not applicable. IRR = Incidence rate ratio. REF = Reference group.
Cumulative BRDC mortality incidence was significantly (P < 0.01) higher in summer and fall, compared with spring and winter, and was 33% higher in summer and 46% higher in fall than in spring. Precipitation on day P was significantly (P < 0.05) associated with a lower cumulative BRDC mortality incidence, such that cumulative BRDC mortality incidence in cohorts that were purchased on a day (day P) with precipitation was 14% lower, compared with in cohorts on day P without precipitation. The interaction term mean wind speed on day P–transport distance indicated that wind speed was significantly (P < 0.01) associated with cumulative BRDC mortality incidence only among cattle that were transported the longest distances (≥ 1,082.4 km; Figure 1). Lower mean wind speeds on day P were associated with a higher cumulative BRDC mortality incidence among cattle that were transported the longest distances (≥ 1,082.4 km). Among these cattle that were transported the longest distances, cumulative BRDC mortality incidence was 4.9% for cohorts that experienced the lowest mean wind speeds (< 6.1 km/h) on day P, compared with 2.2% for cohorts that experienced the highest mean wind speeds (≥ 12.9 km/h) on day P. Interactions between mean wind speed and mean maximum air temperature and wind speed and season on day P were not significant.
Categorized by mean wind speed (km/h) on day P, the graphical plot shows the model-adjusted mean cumulative BRDC mortality incidence (%) within the first 60 DOF for cohorts of beef calves (n = 3,339; range, 8 to 703 calves/cohort) that were purchased (on day P) from the fall of 2015 to the winter of 2018 from 216 locations and then transported over various distances (4 categories: < 520.1 km [squares], 520.1 to < 767.8 km [circles], 767.8 to < 1,082.4 km [triangles], and ≥ 1,082.4 km [diamonds]) to 89 feeding locations (backgrounding location, 87; feedlot, 2) as part of a large commercial cattle feeding operation in the Midwestern United States, as determined from a mixed-effects negative binomial regression model. Whiskers denote 95% CI.
Citation: American Journal of Veterinary Research 82, 8; 10.2460/ajvr.82.8.644
Categorized by mean wind speed (km/h) on day P, the graphical plot shows the model-adjusted mean cumulative BRDC mortality incidence (%) within the first 60 DOF for cohorts of beef calves (n = 3,339; range, 8 to 703 calves/cohort) that were purchased (on day P) from the fall of 2015 to the winter of 2018 from 216 locations and then transported over various distances (4 categories: < 520.1 km [squares], 520.1 to < 767.8 km [circles], 767.8 to < 1,082.4 km [triangles], and ≥ 1,082.4 km [diamonds]) to 89 feeding locations (backgrounding location, 87; feedlot, 2) as part of a large commercial cattle feeding operation in the Midwestern United States, as determined from a mixed-effects negative binomial regression model. Whiskers denote 95% CI.
Citation: American Journal of Veterinary Research 82, 8; 10.2460/ajvr.82.8.644
Categorized by mean wind speed (km/h) on day P, the graphical plot shows the model-adjusted mean cumulative BRDC mortality incidence (%) within the first 60 DOF for cohorts of beef calves (n = 3,339; range, 8 to 703 calves/cohort) that were purchased (on day P) from the fall of 2015 to the winter of 2018 from 216 locations and then transported over various distances (4 categories: < 520.1 km [squares], 520.1 to < 767.8 km [circles], 767.8 to < 1,082.4 km [triangles], and ≥ 1,082.4 km [diamonds]) to 89 feeding locations (backgrounding location, 87; feedlot, 2) as part of a large commercial cattle feeding operation in the Midwestern United States, as determined from a mixed-effects negative binomial regression model. Whiskers denote 95% CI.
Citation: American Journal of Veterinary Research 82, 8; 10.2460/ajvr.82.8.644
Absolute air temperature change (absolute difference between maximum and minimum air temperatures) during the first week at the feeding location was significantly (P < 0.01) associated with cumulative BRDC mortality incidence. Cumulative BRDC mortality incidence was significantly (P < 0.01) higher for cohorts that experienced air temperature changes of 1.9°C to < 5.3°C, compared with that for cohorts that experienced air temperature changes of < 1.9°C. However, cohorts that experienced the largest air temperature change (≥ 9.4°C) had a significantly (P = 0.04) lower cumulative BRDC mortality incidence, compared with cohorts that experienced air temperature changes of 5.3°C to < 9.4°C. The interaction term mean maximum air temperature on day P–mean BWA indicated that higher mean maximum air temperature was associated (P < 0.01) with higher cumulative BRDC incidence among cohorts that included calves with low body weights (136.1 to < 254.7 kg and 254.7 to < 275.5 kg), but was not significantly associated with cumulative BRDC mortality incidence in the other body weight classes (275.5 to < 299.5 kg and 299.5 to 409.0 kg; Figure 2).
Categorized by mean maximum air temperature (°C) on day P, the graphical plot shows the model-adjusted mean cumulative BRDC mortality incidence (%) within the first 60 DOF for the cohorts of calves in Figure 1 based on mean BWA (4 categories: 136.0 to < 254.7 kg [squares], 254.7 to < 275.5 kg [circles], 275.5 to < 299.5 kg [triangles], and 299.5 to 409.0 kg [diamonds]), as determined from a mixed-effects negative binomial regression model. Whiskers denote 95% CI.
Citation: American Journal of Veterinary Research 82, 8; 10.2460/ajvr.82.8.644
Categorized by mean maximum air temperature (°C) on day P, the graphical plot shows the model-adjusted mean cumulative BRDC mortality incidence (%) within the first 60 DOF for the cohorts of calves in Figure 1 based on mean BWA (4 categories: 136.0 to < 254.7 kg [squares], 254.7 to < 275.5 kg [circles], 275.5 to < 299.5 kg [triangles], and 299.5 to 409.0 kg [diamonds]), as determined from a mixed-effects negative binomial regression model. Whiskers denote 95% CI.
Citation: American Journal of Veterinary Research 82, 8; 10.2460/ajvr.82.8.644
Categorized by mean maximum air temperature (°C) on day P, the graphical plot shows the model-adjusted mean cumulative BRDC mortality incidence (%) within the first 60 DOF for the cohorts of calves in Figure 1 based on mean BWA (4 categories: 136.0 to < 254.7 kg [squares], 254.7 to < 275.5 kg [circles], 275.5 to < 299.5 kg [triangles], and 299.5 to 409.0 kg [diamonds]), as determined from a mixed-effects negative binomial regression model. Whiskers denote 95% CI.
Citation: American Journal of Veterinary Research 82, 8; 10.2460/ajvr.82.8.644
Discussion
Few studies11,18,19,20,21 include an investigation of the association between weather conditions and illness and death attributed to BRDC, and none specifically include an investigation of the effect of weather conditions on day P for BRDC morbidity and mortality incidence. The use of weather condition data on day P has advantages over use of weather condition data only on the day of arrival at a feedlot in that operation managers and veterinarians could use the data from day P to determine whether changes in cattle processing and management at a feedlot may be needed to reduce the risk of death attributed to BRDC. For example, if cattle are purchased during weather conditions that are associated with increased risk of BRDC death, operation managers may implement methods that reduce stress, reduce the number of commingled purchase groups within a cohort, and reduce the time to fill a pen with a complete group of cattle.10,33,34 In addition, forecasted weather conditions on day P may influence when and how to purchase, transport, and manage cattle on the basis of potential health risks.
Many studies21,30,31 link transportation to physiologic stress responses in cattle. For example, Chirase et al24 indicated that transportation stress increased concentrations of oxidative stress biomarkers in beef calves, and other authors indicated that increased concentrations can lead to immunosuppression and subsequent increased risk of developing BRDC.35,36 Poor weather conditions on day P may exacerbate the stress of transportation. High air temperatures with little relief from wind or rain may subject cattle to heat stress during transport.37 In the present study, precipitation on day P was associated with lower cumulative BRDC mortality incidence. However, no significant effect of precipitation on BRDC morbidity and mortality incidence has been reported in other studies.11,21 Precipitation may be of benefit on day P because dust and associated irritants and pathogens before, during, and following transport may less likely be airborne (ie, will settle instead). Also, precipitation may be beneficial because of its cooling effect on warm days. However, most days (day P) that had precipitation were in the lowest quartile of maximum daily air temperature (< 7.4°C) in the present study. Future studies should include further investigation of the association between precipitation on day P and BRDC morbidity and mortality incidence.
Higher mean wind speed on day P was associated with decreased cumulative BRDC mortality incidence among cattle that were transported long distances (≥ 1,082.4 km). These results were discordant with those of a previous study11 that indicate higher wind speeds are associated with a high risk of illness attributable to BRDC. Similar to the findings in the present study, another previous study21 reveals a negative but nonsignificant correlation with wind speed and BRDC morbidity. However, both of those studies11,21 report on the association between weather conditions only on the day of feedlot arrival and BRDC morbidity and include cattle only during cooler months (September to December), whereas the present study investigated the weather on day P and included cattle from all months. High wind speeds are often beneficial for transport in warmer months because of their cooling effect but may have adverse consequences in colder months. However, the interaction between wind speed and air temperature or season in the present study was not significant. The protective effect of wind speed on the mortality incidence throughout all seasons may have been attributed to the mean maximum air temperature of > 7.4°C on day P for 75% of the cohorts, which indicated that few cohorts experienced cold temperatures on day P. Also, 25% of the cohorts experienced mean maximum air temperatures > 26.7°C on day P, which indicated that they were likely subjected to heat stress because the ideal maximal air temperature for cattle is < 25°C.38 Also, higher mean maximum air temperature on day P was associated with higher cumulative BRDC mortality incidence among calves with a lower body weight (136 to < 275.5 kg). Syf indicated that calves of lower body weights have higher body temperatures after long-haul transport, compared with calves with higher body weights, thus predisposing them to heat stress and subsequent illness attributable to BRDC. This finding differs from findings from Cusack et al,21 who reported that minimum daily air temperature has a significant negative correlation with BRDC morbidity (ie, higher minimum air temperatures are associated with lower BRDC morbidity). However, that study21 includes only cooler months, so the effects of a higher minimum air temperature were likely protective from deleterious cold temperatures.
Weather conditions during the first week at the feeding location may contribute to the stress associated with acclimation to a new environment, new social order, and processing procedures.25 In the present study, however, a clear relationship between absolute air temperature change during the first week at the feeding location and cumulative BRDC mortality incidence was not evident. Large variations in the air temperature are believed to be associated with increased BRDC morbidity and mortality rates, although few studies39 include an investigation of this effect. A few studies18,20 reveal that large air temperature changes are associated with increased BRDC morbidity and mortality rates, but other studies11,19 reveal that those changes are associated with decreased BRDC morbidity. Possible reasons for the negative association between large air temperature changes and BRDC morbidity are poor assessment of illness by feedlot personnel during poor weather conditions11 and decreased viability of BRDC pathogens. Pathogens that may contribute to BRDC such as Mycobacterium bovis and bovine viral diarrhea virus can survive in the environment, but their infectivity and survival depend on the air temperature.40,41,42 Future studies should determine whether air temperature affects the presence and virulence of the most common BRDC pathogens in the environment, including Mannheimia haemolytica, Pasteurella multocida, and Histophilus somni.42 Also, given the unclear relationship between absolute air temperature change and cumulative BRDC mortality incidence in the present study and previous studies,11,18,19,20 further research is needed to determine whether air temperature change after arrival affects the BRDC mortality incidence.
Findings related to other risk factors for BRDC-related mortality incidence in the present study were similar to those from previous studies.10,14,15,20,43,44 Cohorts with weaned calves had lower cumulative BRDC mortality incidence, a finding similar to that reported by Richeson et al.43 Cohorts with more com-mingled cattle had increased cumulative BRDC mortality incidence, which was consistent with findings of other studies.10,14,15,44 Cumulative BRDC mortality incidence was highest in the fall, consistent with findings from other studies,20,45 followed by the summer. Increased BRDC deaths during the fall is speculated to occur because that season is the traditional time for sale (purchase) of beef cattle, which often leads to a greater degree of commingling, crowding, and fatigue of feedlot personnel.39 Delays in loading, unloading, and transporting cattle may be long, which also may increase stress and disease incidence, although these claims have not been adequately researched.
Strengths of the present study were the large sample size and the inclusion of weather condition data on day P. However, the study had limitations. Because of the small number of observations in some categories of the variables, some cohorts had to be excluded. This limited the generalizability of the results to other cohorts, such as those that include female calves, yearlings, and cattle sourced directly from ranches. Also, the present study included calves from only 1 large feeding operation, which limited the generalizability of the results to other feeding operations. Operations in other regions may have different processing and management procedures. In particular, this study only included cohorts that received metaphylaxis, so the results may not be generalizable to cohorts that did not receive metaphylaxis. Plus, other regions may undergo different weather conditions from those in the present study. Another limitation was that physiological stress and immunocompetence status were not directly measured, which are the proposed mechanisms by which weather conditions are associated with disease. Additionally, data on the microenvironment during transport were not available. Other studies46,47,48 reveal that the microcli-mate can vary greatly among compartments within a transport trailer and that a calf's location in a trailer is associated with increased odds of the need for BRDC treatment.
In conclusion, results of the study reported here indicated that weather conditions on day P and during the first week at the feeding location (backgrounding location or feedlot) were significantly associated with cumulative BRDC mortality incidence in the first 60 DOF. These findings may have implications for the health- and economic-risk management of calves. Managers and veterinarians may be able to use available data on weather conditions on day P to intervene, for example, by changing processing procedures at arrival or increasing monitoring of the health of high-risk calves. Additionally, calf procurement and risk management programs may be modified on the basis of observed and predicted weather conditions, such that purchasing or transporting calves with low body weights (< 275.5 kg) or calves that are expected to be transported a long distance (≥ 1,082.4 km) could be planned for days with cooler temperatures or on warmer days that have precipitation or higher wind speeds. Future research is necessary to confirm the findings of the present study and evaluate associations between BRDC morbidity and mortality rates and weather conditions in various feeding operations, as well as investigate whether weather conditions affect markers of physiological stress and immunocompetence and whether the microenvironment during transport affects BRDC morbidity and mortality rates.
Acknowledgments
Funded by USDA National Institute of Food and Agriculture's Agriculture and Food Research Initiative (award No. 2015-67015-23079) and by the USDA National Institute of Food and Agriculture Hatch Act of 1887 (Multistate Research Fund No. 1018845). Funding sources did not have any involvement in the study design, data analysis and interpretation, or writing of the manuscript.
The authors declare no conflicts of interest.
Presented as a poster at the Annual Conference of Research Workers in Animal Diseases, Chicago, November 2019.
The authors thank the participating cattle feeding operation.
Abbreviations
| BRDC | Bovine respiratory disease complex |
| BWA | Body weight on arrival |
| DOF | Days on feed |
Footnotes
MapQuest developer, Geocoding API, Mapquest Inc, Denver, Colo. Available at: developer.mapquest.com/documentation/geocoding-api/. Accessed Aug 1, 2019.
Stata, version 14.2, StataCorp, College Station, Tex.
GEODIST: stata module to compute geographical distances, Picard R, Statistical Software Components, Economics Department, Morrissey College of Arts and Sciences, Boston College, Boston, Mass. Available at: econpapers.repec.org/software/bocbocode/s457147.htm. Accessed Jul 7, 2020.
Darksky, version 1.3.0, Apple Inc, Cupertino, Calif.
R: A language and environment for statistical computing, version 3.6.0, R Foundation for Statistical Computing, Vienna, Austria.
Sy T. Evaluation of long-haul shipping stress for beef calves transported from Hawaii to Washington or California and their ability to recover. PhD dissertation, Animal Science, University of Hawaii at Manoa, Honolulu, Hawaii, 2015.
References
- 1. ↑
National Animal Health Monitoring System. Feedlot 2011: part IV: health and health management on US feedlots with a capacity of 1,000 or more head. Fort Collins, Colo: USDA, 2013.
- 2. ↑
Cockcroft P. Bovine respiratory disease (BRD): diagnosis, prevention and control. In: Cockcroft P, ed. Bovine medicine. Chichester, England: John Wiley & Sons Ltd, 2015;525– 530.
- 3. ↑
Pinchak WE, Tolleson DR, McCloy M, et al. Morbidity effects on productivity and profitability of stocker cattle grazing in the Southern plains. J Anim Sci 2004;82:2773–2779.
- 4. ↑
Snowder GD, Van Vleck LD, Cundiff LV, et al. Bovine respiratory disease in feedlot cattle: environmental, genetic, and economic factors. J Anim Sci 2006;84:1999–2008.
- 5. ↑
Cernicchiaro N, White BJ, Renter DG, et al. Evaluation of economic and performance outcomes associated with the number of treatments after an initial diagnosis of bovine respiratory disease in commercial feeder cattle. Am J Vet Res 2013;74:300–309.
- 6. ↑
Blakebrough-Hall C, McMeniman JP, González LA. An evaluation of the economic effects of bovine respiratory disease on animal performance, carcass traits, and economic outcomes in feedlot cattle defined using four BRDC diagnosis methods. J Anim Sci 2020;98:skaa005.
- 7. ↑
Wang M, Schneider LG, Hubbard KJ, et al. Cost of bovine respiratory disease in preweaned calves on US beef cow-calf operations (2011–2015). J Am Vet Med Assoc 2018;253:624– 631.
- 8. ↑
Johnson KK, Pendell DL. Market impacts of reducing the prevalence of bovine respiratory disease in United States beef cattle feedlots. Front Vet Sci 2017;4:189.
- 9. ↑
Cusack PMV, McMeniman N, Lean IJ. The medicine and epidemiology of bovine respiratory disease in feedlots. Aust Vet J 2003;81:480–487.
- 10. ↑
Sanderson MW, Dargatz DA, Wagner BA. Risk factors for initial respiratory disease in United States' feedlots based on producer-collected daily morbidity counts. Can Vet J 2008;49:373–378.
- 11. ↑
Cernicchiaro N, Renter DJ, White BJ, et al. Associations between weather conditions during the first 45 days after feedlot arrival and daily respiratory disease risks in autumn-placed feeder cattle in the United States. J Anim Sci 2012;90:1328–1337.
- 12. ↑
Hay KE, Morton ML, Mahony TJ, et al. Associations between animal characteristic environmental risk factors and bovine respiratory disease in Australian feedlot cattle. Prev Vet Med 2016;125:66–74.
- 13. ↑
Cernicchiaro N, White BJ, Renter DJ, et al. Effects of body weight loss during transit from sale barns to commercial feedlots on health and performance in feeder cattle cohorts arriving to feedlots from 2000 to 2008. J Anim Sci 2012;90:1940–1947.
- 14. ↑
Step DL, Krehbiel CR, DePra HA, et al. Effects of commingling beef calves from different sources and weaning protocols during a forty-two-day receiving period on performance and bovine respiratory disease. J Anim Sci 2008;86:3146– 3158.
- 15. ↑
Hay KE, Barnes TS, Morton JM, et al. Risk factors for bovine respiratory disease in Australian feedlot cattle: use of a causal diagram-informed approach to estimate effects of animal mixing and movements before feedlot entry. Prev Vet Med 2014;117:160–169.
- 16. ↑
Macartney JE, Bateman KG, Ribble CS. Health performance of feeder calves sold at conventional auctions versus special auctions of vaccinated or condition calves in Ontario. J Am Vet Med Assoc 2003;223:677–683.
- 17. ↑
Hay KE, Morton JM, Schibrowski ML, et al. Associations between prior management of cattle and risk of bovine respiratory disease in feedlot cattle. Prev Vet Med 2016;127:37–43.
- 18. ↑
MacVean DW, Franzen DK, Keefe TJ, et al. Airborne particle concentration and meteorologic conditions associated with pneumonia incidence in feedlot cattle. Am J Vet Res 1986;47:2676–2682.
- 19. ↑
Alexander BH, MacVean DW, Salman MD. Risk factors for lower respiratory tract disease in a cohort of feedlot cattle. J Am Vet Med Assoc 1989;195:207–211.
- 20. ↑
Ribble CS, Meek AH, Janzen ED, et al. Effect of time of year, weather, and the pattern of auction market sales on fatal fibrinous pneumonia (shipping fever) in calves in a large feed-lot in Alberta (1965–1988). Can J Vet Res 1995;59:167–172.
- 21. ↑
Cusack PM, McMeniman NR, Lean IJ. Feedlot entry characteristics and climate: their relationship with cattle growth rate, bovine respiratory disease and mortality. Aust Vet J 2007;85:311–316.
- 22. ↑
Yun CH, Wynn P, Ha JK. Stress, acute phase proteins and immune modulation in calves. Anim Prod Sci 2014;54:1561– 1568.
- 23. ↑
Hulbert LE, Moisá SJ. Stress, immunity, and the management of calves. J Dairy Sci 2016;99:3199–3216.
- 24. ↑
Chirase NK, Greene LW, Purdy CW, et al. Effect of transport stress on respiratory disease, serum antioxidant status, and serum concentrations of lipid peroxidation biomarkers in beef cattle. Am J Vet Res 2004;65:860–864.
- 25. ↑
Loerch SC, Fluharty FL. Physiological changes and digestive capabilities of newly received feedlot cattle. J Anim Sci 1999;77:1113–1119.
- 26. ↑
von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol 2008;61:344–349.
- 27. ↑
Voorheis J. mqtime: a Stata tool for calculating travel time and distance using MapQuest web services. Stata J 2015;15:845– 853.
- 28. ↑
Chan D. Functional relations among constructs in the same content domain at different levels of analysis: a typology of composition models. J Appl Psychol 1998;83:234–246.
- 29. ↑
Wisnieski LW, Norby B, Pierce SJ, et al. Cohort level disease prediction by extrapolation of individual-level predictions in transition dairy cattle. Prev Vet Med 2019;169:104692.
- 30. ↑
Wisnieski L, Norby B, Pierce SJ, et al. Cohort-level disease prediction using aggregate biomarker data measured at dryoff in transition dairy cattle: a proof-of-concept study. Prev Vet Med 2019;169:104701.
- 31. ↑
Dormann CF, Elith J, Bacher S, et al. Collinearity: a review of methods to deal with it and a simulation study evaluating their performances. Ecography 2013;26:27–46.
- 32. ↑
Hosmer DW, Lemeshow S, Sturdivant RX. Applied logistic regression. 3rd ed. Hoboken, NJ: John Wiley & Sons, 2013.
- 33. ↑
Barnes T, Hay K, Morton J, et al. Final report: epidemiology and management of bovine respiratory disease in feedlot cattle. North Sydney, NSW, Australia: Meat & Livestock Australia Ltd, 2014.
- 34. ↑
Noffsinger T, Lukasiewicz K, Hyder L. Feedlot processing and arrival cattle management. Vet Clin North Am Food Anim Pract 2015;31:323–340.
- 35. ↑
Mormede P, Soissons J, Bluthe RM, et al. Effect of transportation on blood serum composition, disease incidence, and production traits in young calves: influence of the journey duration. Ann Rech Vet 1982;13:369–384.
- 36. ↑
Mackenzie AM, Drennan M, Rowan TG, et al. Effect of transportation and weaning on humoral immune responses of calves. Res Vet Sci 1997;63:227–230.
- 37. ↑
Schwartzkopf-Genswein KS, Faucitano L, Dadgar S, et al. Road transport of cattle, swine and poultry in North America and its impact on animal welfare, carcass and meat quality: a review. Meat Sci 2012;92:227–243.
- 38. ↑
Hahn GL. Dynamic responses of cattle to thermal heat loads. J Anim Sci 1999;77(suppl 2):10–20.
- 39. ↑
Taylor JD, Fulton RW, Lehenbauer TW, et al. The epidemiology of bovine respiratory disease: what is the evidence for predisposing factors? Can Vet J 2010;51:1095–1102.
- 40. ↑
Depner K, Bauer T, Liess B. Thermal and pH stability of pestiviruses. Rev Sci Tech 1992;11:885–893.
- 41. ↑
Fine AE, Bolin CA, Gardiner JC, et al. A study on the persistence of Mycobacterium bovis in the environment under natural weather conditions in Michigan, USA. Vet Med Int 2011;2011:765430.
- 42. ↑
Stevens ET, Thomson DU, Wileman BW, et al. The survival of bovine viral diarrhea virus on materials associated with livestock production. Bov Pract 2011;45:118–123.
- 43. ↑
Richeson JT, Kegley EB, Powell JG, et al. Weaning management of newly received beef calves with or without continuous exposure to a persistently infected bovine viral diarrhea virus pen mate: effects on health, performance, bovine viral diarrhea virus titers, and peripheral blood leukocytes. J Anim Sci 2012;90:1972–1985.
- 44. ↑
Hay KE, Morton JM, Clements ACA, et al. Population-level effects of risk factors for bovine respiratory disease in Australian feedlot cattle. Prev Vet Med 2017;140:78–86.
- 45. ↑
Loneragan GH, Dargatz DA, Morley PS, et al. Trends in mortality ratios among cattle in US feedlots. J Am Vet Med Assoc 2001;219:1122–1127.
- 46. ↑
White BJ, Blasi D, Vogel LC, et al. Associations of beef calf wellness and body weight gain with internal location in a truck during transportation. J Anim Sci 2009;87:4143–4150.
- 47. ↑
Goldhawk C, Janzen E, González LA, et al. Trailer microcli-mate and calf welfare during fall-run transportation of beef calves in Alberta. J Anim Sci 2014;92:5142–5154.
- 48. ↑
Stanford K, Bryan M, Peters J, et al. Effects of long- or short-haul transportation of slaughter heifers and cattle liner micro-climate on hide contamination with Escherichia coli O157. J Food Prot 2011;74:1605–1610.
