Changes in Dairy Cow Behavior with and without Assistance at Calving
Round 1
Reviewer 1 Report
The study evaluated duration and frequency of behaviors 9 hours prior to calving in two groups of cows (with or without calving assistance). This study is relevant as it seeks to provide data that may be used by producers in early determination of cows that may require calving assistance. The experimental design is clear. However, the following concerns may be further explained by the authors.
- The title fails to capture the aim of the study which is to evaluate calving behavior in dairy cows with or without assistance. Please revise for clarity.
- Abstract
L23-25 - This is not part of your original objective. My suggestion is that you add that objective in L14-15 i.e. “to characterize calving behavior of dairy cows and to compare the duration and frequency of behaviors for assisted and unassisted dairy cows at calving”.
L25 – add “calving” assistance
L63 – add approval number if available
L73 – several days? Please be specific or give a range of days or average
L146-147 – what is “key” behavior changes? Revise
L120 - In the final three 120 hours prior to calving “,” the proportion of time for lying and contractions increased and
L124 – Add “dairy cows” (tables and figures should be as explanatory as possible). Check all figures and table titles
L174 – this word “main” would suppose 1 or at most 2 behaviors. Revise
L179 180 – You did not analyze behavior data as an alert for birth events. Remove “however, the monitoring of behavioral patterns could be used as an alert for birth events”
L179-183 – Please add that behaviors were monitored within 9 hours of calving.
Additional questions/suggestions
- Was there a rationale for 9 hours prior to calving as the benchmark? Please include in the draft.
- Please describe at what point or what behavior (s) was used to determine that the subset of cows (n=8) required calving assistance. How did you ensure it was uniformly applied?
- Include number of cows in tables and figures titles.
- Kindly clarify figures 2 and 4.
Author Response
The authors thank the reviewer for their constructive comments, which have each been addressed below.
The study evaluated duration and frequency of behaviors 9 hours prior to calving in two groups of cows (with or without calving assistance). This study is relevant as it seeks to provide data that may be used by producers in early determination of cows that may require calving assistance. The experimental design is clear. However, the following concerns may be further explained by the authors.
The title fails to capture the aim of the study which is to evaluate calving behavior in dairy cows with or without assistance. Please revise for clarity.
AU: Changes in dairy cow behavior with and without assistance prior at calving.
Abstract
L23-25 - This is not part of your original objective. My suggestion is that you add that objective in
AU: This has been added in as below.
L14-15 i.e. “to characterize calving behavior of dairy cows and to compare the duration and frequency of behaviors for assisted and unassisted dairy cows at calving”.
AU: This has been altered at suggested to “The aim of this study was to characterize calving behavior of dairy cows and to compare the duration and frequency of behaviors for assisted and unassisted dairy cows at calving”.
L25 – add “calving” assistance
AU: Added as suggested.
L63 – add approval number if available
AU: Approval number added as suggested.
L73 – several days? Please be specific or give a range of days or average
AU: Clarified as “At three weeks”.
L146-147 – what is “key” behavior changes? Revise
AU: This sentence has been revised to “This study found that when monitoring calving the duration and frequency of lying and contraction bouts increased in the last three hours prior to birth compared to other time periods studied.”
L120 - In the final three 120 hours prior to calving “,” the proportion of time for lying and contractions increased and
AU: Added as suggested.
L124 – Add “dairy cows” (tables and figures should be as explanatory as possible). Check all figures and table titles
AU: Added as suggested and edits to tables/figures.
L174 – this word “main” would suppose 1 or at most 2 behaviors. Revise
AU: Revised sentence to “During the study, changes in behavior were largely associated with…”.
L179 180 – You did not analyze behavior data as an alert for birth events. Remove “however, the monitoring of behavioral patterns could be used as an alert for birth events”
AU: Removed as suggested.
L179-183 – Please add that behaviors were monitored within 9 hours of calving.
AU: Added as suggested.
Additional questions/suggestions
Was there a rationale for 9 hours prior to calving as the benchmark? Please include in the draft.
AU: A sentence has been added to the Material and Methods section 2.2 as “The start of the continuous observation period was determined as 9 hours from when the calf was fully expelled at birth using the video recording, and considered a time when no visual signs of calving behavior are observed.”
Please describe at what point or what behavior (s) was used to determine that the subset of cows (n=8) required calving assistance. How did you ensure it was uniformly applied?
AU: A sentence has been added to section 2.1 as “Need for a birth to be assisted was recorded for each cow and determined by the same experienced farm staff from visual assessment on calving progress. Cows were managed and housed within their normal environment.”
Include number of cows in tables and figures titles.
AU: Added as suggested.
Kindly clarify figures 2 and 4.
AU: Figures 2 and 4 are the three-way interaction between time period × behavior × assistance. These are the predicted means from tables 1 and 2.
Reviewer 2 Report
General comments
This retrospective study evaluated changes in dairy cow behavior (i.e., duration and frequency of several patterns) prior to calving. A comparison was made whether there are differences between the behavior of cows that were assisted and unassisted at calving.
The overall approach of this study is limited as this observational study uses retrospective classification of behavioral events prior to calving. From a research perspective this seems interesting but inference made from these results must be interpreted with caution as behavioral data need to be used prospectively to help farmers intervene when assistance is required. It has been shown in several studies (e.g., Lange et al., 2017 J Dairy Sci. 2017 Jun;100(6):4847-4856) that calving prediction is difficult using behavioral patterns. Rather than using an endpoint (i.e., delivery of the calf) the authors should re-analyze their data using stage II as the starting point (i.e., presence of the calf’s feet or water bag) and evaluate behavioral patterns depending on calving assistance.
The authors should provide a sample size calculation. This study seems underpowered to detect difference between eutocia and dystcocia.
The results need to be discussed more critically, especially in the context to already published studies having a similar objective as well as clarifying how exactly they serve as a practical support for dairy farmers. Furthermore, it needs a few more details concerning the experimental design, especially when it comes to the definition of terms and the statistical analysis.
Specific comments
L 41-45: Long sentence, please consider rewriting.
L 56-61: Please change the order: The objective of the study should be mentioned before the hypothesis.
L 73: “Several days prior to calving […]”: Which criteria did you use to move cows into the calving pens? Did you consider a specific day/week after successful insemination or prior to the expected calving date? Please specify.
L 90-91: Your definition for eating behavior is a little confusing. Putting a head through a feeding barrier and pulling it back is not necessarily associated with the action of feed intake.
L 95: You chose to observe cows during 9 hours prior to calving and split these hours into three-hour time periods. For which reason did you chose these periods of observation and analysis, respectively? As the video cameras guarantee a real-time monitoring, three hours seem a temporally wide range, especially when it comes to prediction of calving and the need for assistance.
L 96: You determined period one ending “with the birth”. Please specify what you mean by that, e.g. the calf being fully expelled or the appearance of the amniotic sac or feet outside the vulva? This is a very important information that is missing.
L 107: You compare calving events that require assistance or no assistance. You do not mention at any point, which criteria you used to decide for assistance or not. How did the farm protocol look like? When was assistance provided? How did you define eutocia versus dystocia?
L 146: Overall, your discussion-section appears quite short and is mainly focusing on the lying behavior. Consider extending this section by also referring to other behavioral patterns you analyzed in this study and discussing them in context to studies that also evaluated behavior prior to calving using video cameras. You could also highlight, what is different or novel here to other studies focusing on this topic.
L 151-153: Please note as a limitation, that one possible reason for finding no differences in behavioral patterns between assisted and unassisted calvings in your study is a rather small sample size and an even smaller sample of cows (8/35) that needed assistance. This makes it more difficult to make a fair comparison with cows that did not need assistance.
L 175-176: The second part of the sentence seems to be a repetition of lines 153-154.
Figures: Overall, you provide many figures. Please note, that the distances of the y-axis (“Proportion”) are not chosen uniformly, which might give a distorted impression at first glance.
Furthermore, the titles of the figures are not clear. You show the predicted mean proportion of time spent doing different behaviors (Figure 1) and the proportion of observations for different behaviors (Figure 3). However, the temporal reference is missing: During the whole 9 hours of observation or only during the final 3 hours?
Tables: You provide a lot of data in these tables. It remains, however, unclear to the reader how to interpret these data. Results from GENLINMIXED models should be presented in a meaningful way.
Author Response
The authors than the reviewer for their constructive comments, which have each been addressed below.
This retrospective study evaluated changes in dairy cow behavior (i.e., duration and frequency of several patterns) prior to calving. A comparison was made whether there are differences between the behavior of cows that were assisted and unassisted at calving.
The overall approach of this study is limited as this observational study uses retrospective classification of behavioral events prior to calving. From a research perspective this seems interesting but inference made from these results must be interpreted with caution as behavioral data need to be used prospectively to help farmers intervene when assistance is required. It has been shown in several studies (e.g., Lange et al., 2017 J Dairy Sci. 2017 Jun;100(6):4847-4856) that calving prediction is difficult using behavioral patterns. Rather than using an endpoint (i.e., delivery of the calf) the authors should re-analyze their data using stage II as the starting point (i.e., presence of the calf’s feet or water bag) and evaluate behavioral patterns depending on calving assistance.
AU: The 9 hours before calving was considered sufficient to observe the changes in behaviors leading up to calving. Visually observing the calf’s feet or water bag are not as reliable as the expulsion of the whole calf. Observing the feet and water bag are very tricky at night under poor lighting. Therefore, the calf being born was considered reliable and the option to take. Further discussion has been included regarding this point as “The failure to find a difference between assisted and unassisted calvings may have been influenced by observations being conducted until the calf was fully expelled and any assistance being subjectively determined by farm staff.”
The authors should provide a sample size calculation. This study seems underpowered to detect difference between eutocia and dystcocia.
AU: Behavioral observations for 35 cows is a considerable amount of observations and work. While more assisted births would be beneficial, there were 23% of births that were assisted. The mixed model approach used is suitable to unbalanced data. The authors recognise that a larger dataset would always be beneficial. The results appear consistent with other studies. The lack of a difference between assisted and unassisted births would probably be more influenced by subjective nature of determining the need to intervene.
The results need to be discussed more critically, especially in the context to already published studies having a similar objective as well as clarifying how exactly they serve as a practical support for dairy farmers. Furthermore, it needs a few more details concerning the experimental design, especially when it comes to the definition of terms and the statistical analysis.
AU: Further discussion and detail has been added throughout.
Specific comments
L 41-45: Long sentence, please consider rewriting.
AU: Sentence has been revised to “To assist a stockperson at calving, and given the importance of a successful birth and potential need for intervention, a number of sensor technologies have been developed. These technologies have largely been based on accelerometers and movement detection [3,4], or an alternative is computer vision [5,6], which have been developed to support farm management and improve animal health and wellbeing, and ultimately productivity.”
L 56-61: Please change the order: The objective of the study should be mentioned before the hypothesis.
AU: Ordered changed as suggested.
L 73: “Several days prior to calving […]”: Which criteria did you use to move cows into the calving pens? Did you consider a specific day/week after successful insemination or prior to the expected calving date? Please specify.
AU: Detail has been added as “At three weeks before expected calving, each cow was moved into one of the three calving pens so that the entire calving process could be closely monitored”.
L 90-91: Your definition for eating behavior is a little confusing. Putting a head through a feeding barrier and pulling it back is not necessarily associated with the action of feed intake.
AU: Actual consumption of feed could not be determined, but feed face activity could be observed and assumed as feeding behaviour, which was recorded as described.
L 95: You chose to observe cows during 9 hours prior to calving and split these hours into three-hour time periods. For which reason did you chose these periods of observation and analysis, respectively? As the video cameras guarantee a real-time monitoring, three hours seem a temporally wide range, especially when it comes to prediction of calving and the need for assistance.
AU: Cows can spend long periods doing the same behaviour, such as lying. The three-hour periods studied allowed comparison between several behaviors carried out during each time period for all cows rather than spending 100% of an hour doing a single behavior. A sentence has been added at the end of the Material and Methods.
L 96: You determined period one ending “with the birth”. Please specify what you mean by that, e.g. the calf being fully expelled or the appearance of the amniotic sac or feet outside the vulva? This is a very important information that is missing.
AU: This detail has been added to section 2.2 as “The start of the continuous observation period was determined as 9 hours from when the calf was fully expelled at birth using the video recording, and considered a time when no visual signs of calving behavior are observed.”
L 107: You compare calving events that require assistance or no assistance. You do not mention at any point, which criteria you used to decide for assistance or not. How did the farm protocol look like? When was assistance provided? How did you define eutocia versus dystocia?
AU: The need for assistance was determined by experienced farm staff. This was visually determined. A sentence has been added to section 2.1 as “Need for a birth to be assisted was recorded for each cow and determined by the same experienced farm staff from visual assessment on calving progress. Cows were man-aged and housed within their normal environment.”
L 146: Overall, your discussion-section appears quite short and is mainly focusing on the lying behavior. Consider extending this section by also referring to other behavioral patterns you analyzed in this study and discussing them in context to studies that also evaluated behavior prior to calving using video cameras. You could also highlight, what is different or novel here to other studies focusing on this topic.
AU: Further discussion has been added as requested.
L 151-153: Please note as a limitation, that one possible reason for finding no differences in behavioral patterns between assisted and unassisted calvings in your study is a rather small sample size and an even smaller sample of cows (8/35) that needed assistance. This makes it more difficult to make a fair comparison with cows that did not need assistance.
AU: We acknowledge that more cows that needed assistance would add and a sentence has been added to the discussion as “Also, only 23% (8 out of the 35 cows) in the current study needed assistance when calving, and therefore further observations of assisted births would add to the study.” Behavioral observations for 35 cows is very time consuming. More assisted births would be better if possible.
L 175-176: The second part of the sentence seems to be a repetition of lines 153-154.
AU: End of sentence removed as suggested.
Figures: Overall, you provide many figures. Please note, that the distances of the y-axis (“Proportion”) are not chosen uniformly, which might give a distorted impression at first glance.
AU: The different y-axis is preferable to allow data points to be seen for each behaviour. The figures have been enlarged.
Furthermore, the titles of the figures are not clear. You show the predicted mean proportion of time spent doing different behaviors (Figure 1) and the proportion of observations for different behaviors (Figure 3). However, the temporal reference is missing: During the whole 9 hours of observation or only during the final 3 hours?
AU: “during the 9 hours prior to calving” has been added to the titles for Figures 1 and 3.
Tables: You provide a lot of data in these tables. It remains, however, unclear to the reader how to interpret these data. Results from GENLINMIXED models should be presented in a meaningful way.
AU: Footnotes have been added to explain the variables shown. Both tables present proportion values, with predicted mean values for behaviors and the three-way interaction of time period × behavior × assistance presented in figures. The mixed model is presented in Section 2.3 to help the reader.
