JDS
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Schneider, M. d. P.
Right arrow Articles by Monardes, H. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Schneider, M. d. P.
Right arrow Articles by Monardes, H. G.
J. Dairy Sci. 86:4083-4089
© American Dairy Science Association, 2003.

Impact of Type Traits on Functional Herd Life of Quebec Holsteins Assessed by Survival Analysis

M. del P. Schneider*, J. W. Dürr{dagger}, R. I. Cue{dagger} and H. G. Monardes{ddagger}

* Facultad de Ciencias Agrarias, Universidad Nacional del Litoral, P. Kreder 2805, Esperanza 3080, Argentina
{dagger} Universidade de Passo Fundo, Faculdade de Agronomia e Medicina Vetorinaria, Barrio São Jose, Brazil
{ddagger} Department of Animal Science, McGill University, Ste-Anne-de-Bellevue, Quebec H9X 3V9, Canada

Corresponding author: R. I. Cue; e-mail: Roger.Cue{at}McGill.ca.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 ACKNOWLEDGEMENTS
 REFERENCES
 
Survival analysis was used to study the effects of composite and descriptive linear type traits on functional herd life of Quebec Holsteins. Functional herd life was defined as the length of life from first calving to death, culling, or censoring, and was adjusted for 305-d milk production. The dataset contained information from 331,105 cows from Quebec province calving for the first time between 1981 and 1995; 58% of the records had type information. Weibull models were fitted to analyze the data. The hazard function was described as the product of a baseline hazard function and the time-independent effects of age at first calving and type, and the time-dependent effects of year of calving, stage of lactation x lactation number, annual change in herd size, 305-d milk production, and herd-year (random). Analyses were done one at a time for each type trait. The strongest relationships between survival and composite type traits were found for final score, mammary system, and feet and legs. Among the linear type traits, the highest impact on functional herd life was found for traits related to the udder.

Key Words: functional herd life • Holstein • survival analysis • type trait

Abbreviation key: FHL = functional herd life, LPL = length of productive life, RCR = relative culling rate, PATLQ = Programme d’Analyse des Troupeaux Laitiers du Quebec


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 ACKNOWLEDGEMENTS
 REFERENCES
 
Conformation is an important component of breeding and selection decisions in dairy cattle populations. The stayability of a cow in the herd can be conditioned by its morphology (leg problems, udder breakdown, etc.). Many studies have reported moderate genetic correlations between conformation traits and herd life, especially the traits related to the udder and feet and legs (Boldman et al., 1992; Klassen et al., 1992; Dekkers et al., 1994; Dürr, 1997; Vollema and Groen, 1998; Buenger et al., 2001; Larroque and Ducrocq, 2001).

However, longevity is a difficult trait to measure, mainly due to the presence of incomplete records (i.e., cows that are still alive when the study is carried out). Ducrocq (1987) defined functional herd life (FHL) as the ability of a cow to remain sound and healthy after adjustment for production level; it refers to the ability to delay involuntary culling. Correcting the length of productive life of a cow for phenotypic milk production is used as an approximation for FHL.

Survival analysis, which deals with censoring in the analysis of time response, has been applied in the animal breeding context to estimate genetic parameters for longevity. It is the methodology applied for routine genetic evaluation for longevity in dairy cattle in France, The Netherlands, Germany, Denmark, The Czech Republic, and Italy (Van der Linde and de Jong, 2003). Length of productive life (LPL), defined as the number of days from first calving to death, culling, or censoring (Ducrocq et al., 1988), is a suitable measure because it is characterized by the presence of incomplete records.

The objective of this study was to evaluate the effects of conformation traits on FHL by means of survival analysis.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 ACKNOWLEDGEMENTS
 REFERENCES
 
Data
Length of productive life was defined as the number of days between date of first calving and date at culling or censoring. The data consisted of 331,105 records of Holstein cows in Quebec herds calving for the first time between March 1, 1981, and March 31, 1995. The raw dataset was provided by the Quebec Herd Analysis Service (PATLQ) and consisted of 2,157,180 lactations. The data contained 67% supervised (official) records and 33% of unsupervised (owner sampler) records. Cows could be enrolled in 2 types of milk recording options: the official option, where milk recording is done by an authorized field supervisor, and the owner sampler option, in which the farmer performs the milk recording himself. Producers enrolled in the official option are entitled to receive official production certificates for each lactation completed by their cows. These certificates are an important marketing tool for selling breeding stock. It has been implicitly accepted that official herds would generally be interested in selling breeding stock and improving production and conformation, whereas owner sampler herds would be primarily interested in improving production (Dürr, 1997). Because data from owner sampler herds are usually not edited as severely as data from official herds, historical files from PATLQ herds include all official herds and only those owner sampler herds with at least 90% of the cows identified and with consistent records for feeding information. This enhances the reliability of the information, but it also creates a selected group of owner sampler herds with possibly better management than the average owner sampler herds. It is important to mention that in the Canadian situation, the distinction between official or owner sampler herds is not synonymous (or equivalent) to registered or grade cows, respectively, because both types of milk recording herds can have registered and grade cows.

The data were merged with 23 conformation traits evaluated by the Holstein Association of Canada; 58% of the records had type information, from which 153,479 (80.3%) corresponded to the official option and 37,688 (19.7%) to the owner sampler option. Therefore, most of the type information corresponds to cows in the official herds. Type information consisted of actual phenotypic type scores of 15 linear descriptive traits (evaluated on a 9-point scale) and 6 composite traits (evaluated on an 18-point scale). Composite traits are calculated directly from the linear score traits (weighting score for each component trait, Holstein Canada, 2003). Front end, body depth, rear attachment width, udder depth, fore teat length, and dairy form linear traits were not included because of limited data. Table 1Go shows the 15 descriptive linear traits included in the analysis, the method of evaluation, a brief description, and the optimal score for each (Holstein Canada, 2003). For all of the composite traits, a high score is desirable (Holstein Canada, 2003). When a cow is evaluated, the classifier’s hand-held computer uses the name, birth date, calving date, lactation number, and a rating for the fullness of the udder (whether the cow has been recently milked or not) to calculate adjustment factors for the mentioned effects when converting the measurements to a descriptive linear code scored from 1 to 9 for on-farm use. The type information corresponded to first classifications. Information about the pedigree status of the cows and about classification (stage of lactation at classification, herd-classifiers, etc.) were not available.


View this table:
[in this window]
[in a new window]
  		Table 1. Description of linear type traits.
 
The censoring criterion was based on the reported disposal codes for each cow. A record was considered to be uncensored if the cow received a termination code indicating that she was either culled or died for any given reason. On the other hand, a record was considered censored if a cow were sold (for dairy purposes), exported, or leased to another herd, if a cow were from herd with discontinuous data collection (stopping milk recording), or if a cow were still alive by March 31, 1995. Forty-three percent of the records were censored.

Model
The following Weibull model was used:


where {lambda}(t) is the hazard function of a cow t days after calving; {lambda}o (t) [{lambda}{rho} ({lambda}t) {rho} - 1] is the Weibull baseline hazard function, with scale parameter {lambda} and shape parameter {rho} (the parameter {rho} was fixed at 1.8, which was estimated in a previous analysis); yi (t') is the time-dependent effect of year of calving, with changes on March 1 of each year; pj ({tau}) is the time-dependent effect of lactation number x stage of lactation (lactation 1, 2, 3, and 4 x 4 stages of lactation), with changes occurring at {tau} = 0, 120, 240, and 305 d of each lactation; zk (t') is the time-dependent effect of annual change in herd size k. Seven classes were defined: a decrease >25%, a decrease of 15 to 25%, a decrease of 5 to 15%, no appreciable change (-5 to +5%), an increase of 5 to 15%, an increase of 15 to 25%, and an increase >25%, with changes at March 1 of each year; am is the time-independent effect of age at first calving (22 to 40 mo of age); du is a dummy variable that indicates the presence or absence of type information. This variable was included to allow all the cows to be considered in the analysis, even the ones without type data, in order to avoid bias. hn (t') is the random time-dependent effect of herd-year, with changes at t' = March 1 of each year (herd-year was assumed to follow a log-gamma distribution, which was algebraically integrated out from the joint posterior density); typeb is the time-independent effect of type trait; and wr ({zeta}) is the time-dependent effect of within herd-year-parity class of milk production at 305 d of lactation (first or later lactation). Five classes were considered: <1.5 SD, between -1.5 and -0.5 SD, between -0.5 and +0.5 SD, between +0.5 and +1.5 SD, and >1.5 SD, with changes at the beginning of a new lactation. The 305-d yields for all cows were standardized to 4% fat and 3.3% protein (PATLQ). The following composite type traits were analyzed: final score, frame-capacity, rump, feet and legs, mammary system, and dairy character. The 15 linear type traits analyzed were stature, size, chest width, loin strength, pin setting, pin width, foot angle, bone quality, rear leg set, median suspensory, udder texture, fore attachment, front teat placement, rear attachment height, and rear teat placement.

Phenotypic scores were considered as class effects. No specific relationship between type traits and functional herd life was assumed. Analyses were performed for each type trait one at a time; hence, 21 separate analyses were performed on the same dataset according to the model previously described. The Survival Kit V3.0 by Ducrocq and Sölkner (1998) was used. The overall influence of each type trait on FHL was assessed using the likelihood ratio test. Significance was tested by adding each effect in sequential order in the model.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 ACKNOWLEDGEMENTS
 REFERENCES
 
All the effects included in the models were significant at P < 0.001. The most important changes in log likelihood (-2log L) were observed after the inclusion of stage of lactation x lactation number and the effect of 305-d milk yield deviation.

Figures 1Go and 2Go show the contribution of composite and linear type traits in the change of the log likelihood (-2log L), respectively. All of the traits were significant at P < 0.001 level. Among the composite traits, the strongest effects were found for final score and mammary system, followed by feet and legs and rump. The linear traits related to the udder had the strongest effect on the FHL compared with mobility and body traits. Rear attachment height, fore attachment, udder texture, and median suspensory were the udder traits that had the strongest relationship with FHL. Among the mobility traits, bone quality had more impact on FHL than foot angle and rear legs set. Stature had the strongest effect on FHL compared with the other body traits.



View larger version (7K):
[in this window]
[in a new window]
  		Figure 1. Contribution to the likelihood (-2log L) of phenotype for composite traits on functional herd life (fs = final score, dc = dairy character, ms = mammary system, fl = feet and legs, fcap = frame-capacity).

 


View larger version (13K):
[in this window]
[in a new window]
  		Figure 2. Contribution to the likelihood (-2log L) of phenotype for linear traits on functional herd life (black bar = udder traits, slashed bar = mobility traits, dotted bar = body traits; rtp = rear teat placement, rah = rear attachment height, ftp = front teat placement, fa = fore attachment, med = median suspensory, ut = udder texture, rls = rear leg set, bq = bone quality, ft = foot angle, ps = pin setting, pw = pin width, ln = loin strength, chs = chest width, siz = size, sta = stature).

 
Solutions for fixed effects can be expressed as relative culling rate (RCR), defined as the ratio between the estimated risk of being culled under the influence of certain environmental or genetic effects and the mean risk which is normally set to one (reference class, RCR = 1). For example, if the RCR for a given class of a fixed effect is 2, it means that a cow in that class has twice the risk of being culled compared with a cow in the reference class for that effect. Constraints were imposed on the fixed effects to be estimated to obtain a set of meaningfully estimable effects and simplify the interpretation. The class effect that is constrained is used as a reference (RCR = 1). This class (or level) is used for comparison with all the other classes of the same effect. For the type traits, the reference class was imposed to an intermediary class to avoid extreme classes that usually have fewer uncensored failures. For composite traits, it was set to class 10, and for linear traits, it was set to class 5. The differences between the classes and the reference class were significant at P < 0.001 for all the type traits solutions, except for those otherwise mentioned.

Solutions for the environmental effects indicated a higher probability of being culled: for primiparous cows calving at older ages, for cows producing below herd average, for first-lactation cows in the first or last stage of lactation, and for older cows at the end of lactation. The risk of being culled did not change for cows in herds decreasing or increasing in size.

Figure 3Go shows the solutions for composite traits. The relative risks were plotted as functions of the classes of phenotypic score. All the traits followed the same trend: cows classified as poor were more likely to be culled than cows with a good classification score. The decrease in RCR was more marked for final score and mammary system compared with the other traits. For dairy character, there was 2-fold increase in risk for cows with very poor dairy character; for the last classes (14 to 18) there was a slight increase in risk, although the differences in risks were not significantly different from the reference class. For all these traits, a high score is desirable; hence, a high phenotypic score leads to an increased FHL.



View larger version (14K):
[in this window]
[in a new window]
  		Figure 3. Solutions for the composite type traits expressed as relative culling rate (RCR) ({blacksquare} = dairy character; {square} = feet and legs; {blacktriangleup} = rump; {triangleup} = frame-capacity; • = mammary system; {circ} = final score). Reference class: class 10 (RCR = 1).

 
To facilitate the interpretation, solutions for linear traits were grouped in udder (rah, fa, med, ut, ftp, rtp), body (sta, siz, ln, chs, pw, ps), and mobility traits (bq, ft, rls) (Figures 4Go, 5Go, and 6Go respectively). Rear attachment height, fore attachment, and median suspensory had a similar trend; cows with strong attached udders had higher chances of surviving than cows with weak ligaments. Cows with soft udder tissue had more chances of surviving than cows with an intermediate phenotype, whereas cows with coarse and fleshy udders had a risk of being culled increased by 3-fold compared with the reference cow. Front teat placement and rear teat placement showed a different trend. These traits have an optimal intermediate score of 6 and 5, respectively (Holstein Canada, 2003). Cows with a phenotype for centered and inside teats had almost no difference in risk, whereas cows with extremely outside teats were more likely to be culled than cows with centered teats.



View larger version (12K):
[in this window]
[in a new window]
  		Figure 4. Solutions for the linear udder traits expressed as relative culling rate (RCR) ({blacksquare} = fore attachment; {square} = front teat placement; • = rear attachment height; {circ} = rear teat placement; {blacktriangleup} = median suspensory; {triangleup} = udder texture). Reference class: class 5 (RCR = 1).

 


View larger version (10K):
[in this window]
[in a new window]
  		Figure 5. Solutions for the linear body traits expressed as relative culling rate (RCR) ({blacksquare} = stature; {square} = size; • = chest width; {circ} = loin strength; {blacktriangleup} = pin setting; {triangleup} = pin width). Reference class: class 5 (RCR = 1).

 


View larger version (9K):
[in this window]
[in a new window]
  		Figure 6. Solutions for the mobility body traits expressed as relative culling rate (RCR) ({blacksquare} = foot angle; • = bone quality; {blacktriangleup} = rear legs set). Reference class: class 5 (RCR = 1).

 
Stature, size, chest width, and loin strength (Figure 5Go) showed a similar trend; cows classified with a high score had more chances of surviving than a cow in the reference class. Therefore, tall and large cows had more chances of surviving than cows with an intermediate phenotype. Having extremely narrow hips increased the risk of a cow of being culled, whereas a wider hip decreased it. Pin setting showed a different relationship. This trait has an intermediate ideal score of 5 (Holstein Canada, 2003). Cows with extremely high or low hips were at higher risk of being culled than cows with an intermediate phenotype; therefore it showed an intermediate optimum.

Figure 6Go shows the solutions for mobility traits. Cows with coarse bones had a higher risk of being culled than a cow in the reference class; instead, the risk is reduced slightly for cows with flat and clean bones. Solutions for foot angle showed that cows with a low foot had more risk of being culled than cows in the reference class; on the other hand, cows with a steep foot had no difference in risk compared with cows with an intermediate phenotype. For this trait, the ideal score is seven (Holstein Canada, 2003). Rear leg set has an optimal score of 5 (Holstein Canada, 2003). Cows with extremely curved legs had an 80% higher risk of being culled than cows with an intermediate phenotype; on the other hand, cows with straight legs had a 34% higher risk of being culled than reference cows. This trait also displayed an intermediate optimum; however, the risk of being culled is higher for cows with extreme curved legs than for cows with extreme straight legs.

Quantification of the Effect of Mammary System on FHL
Survivor curves for cows with different phenotypes for mammary system were computed to evaluate the impact of different phenotypes on longevity (Figure 7Go). For this estimation, we assumed a reference cow with average environmental conditions and a calving interval of 400 d. Three cows with different phenotypic scores were assumed: one with a good classification (class 14, favorable or desirable), one with an intermediate phenotypic score (class 10), and the last one with a poor classification (class 4, unfavorable or undesirable). Differences in survival rate for cows with different phenotypes were found. About 67% of the cows classified with a low phenotypic score, 83% of cows with an intermediate phenotype, and 90% of cows with a high phenotypic score were expected to be alive at the beginning of the second lactation (400 d after calving). Cows with a good mammary system-strongly attached, well-defined, fine-textured udders-are expected to have a longer FHL.



View larger version (17K):
[in this window]
[in a new window]
  		Figure 7. Expected length of life as function of the class of mammary system phenotype ({blacktriangleup} = high score, {blacksquare} = intermediate score, • = low score).

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 ACKNOWLEDGEMENTS
 REFERENCES
 
Results indicated that udder and feet and legs traits had a strong relationship with FHL. Vollema and Groen (1998) found a significant effect (P < 0.01) of total score on FHL for Dutch dairy cows. Reasonable agreement exists with another study using a different methodology for Canadian Holsteins (Boettcher et al., 1997). They reported that overall conformation, mammary system, fore udder, and feet and legs were the composite traits with the largest linear regression coefficients for number of lactations on sire ETA for type traits; they also found positive but small effects for rump and frame-capacity on herd life.

In the present study, dairy character showed a positive relationship with FHL. However, other studies reported a different result. Boettcher et al. (1997) found a negative genetic correlation between dairy character and herd life when it was corrected for milk production. Also, Rogers et al. (1999) found that sires that rate highest for dairy form also score low for health. It seems that refined Holstein cows with a lot of dairy character tend to be higher producers, but are more vulnerable to disease, which leads to a shorter herd life.

Among the linear traits, the strongest effects were found for udder traits. Using survival analysis, Vollema and Groen (1998), Buenger et al. (2001), Larroque and Ducrocq (2001), and Chirinos et al. (2003) reported very similar results for the Dutch, German, French, and Spanish populations, respectively. Theses studies agreed that udder traits influence the length of productive life. Boettcher et al. (1997) reported significant associations (P < 0.001) between all linear type traits and herd life for Canadian Holsteins using a linear model. Therefore, cows with well-attached udders and strong ligament are able to remain in the herd longer.

Stature and size showed a positive relationship with longevity, taller and bigger cows have better chances of surviving. Boettcher et al. (1997) reported a moderate positive relationship between body traits and herd life in Canadian Holsteins. In European populations, these traits do not contribute much to longevity (Vollema and Groen, 1998; Buenger et al., 2001; Larroque and Ducrocq, 2001). For pin setting, agreement was found with the results of Buenger et al. (2001); they reported a decrease in FHL for cows with very ascending rump. However they reported that extremely wide rump influenced negatively FHL, which is different from the results in the present study where wider rump decreased the risk of a cow of being culled.

Reasonable agreement exists between the present study and the results reported by Buenger et al. (2001) for foot angle and rear legs set; extremely low foot angles and extremely curved and straight legs led to a decreased longevity. Similar results were found by Burke and Funk (1993) for US Holsteins and Boettcher et al. (1997) for Canadian Holsteins.

As shown, there are some differences if we compare the results from Quebec Holsteins with European populations, especially for dairy character and some body traits. In our study, most of the data analyzed correspond to official herds, for which it can implicitly be accepted that official herds are interested in improving both production and conformation. These results might indicate some voluntary culling for type. It is known that in the Canadian breeding goal, conformation is an important component.

Unfortunately, we were not able to distinguish between registered and grade cows because we did not have the necessary information. For the French situation, Larroque and Ducrocq (2001) reported different relationships between conformation traits and herd life for registered and unregistered herds, especially for capacity traits. They assumed that capacity traits were subject to some voluntary culling in registered herds. Further work should look at differences between grade and registered herds to confirm if voluntary culling based on conformation exists.

Results from an earlier study using the same dataset, but including the 15 linear type traits simultaneously in the model (Schneider, 1998Schneider, 1999), showed different results. In the aforementioned study, the traits with the strongest effect on survival were rear attachment height, fore attachment, bone quality, stature, and front teat placement. Comparing the 2 studies, it is interesting to note the case of stature and size. In the present study, both traits had a similar effect and relationship with FHL, but when they were considered together in the model, size had almost no relationship with longevity. Size and stature are highly positively correlated at the genetic (0.88) and phenotypic (0.97) level (Klassen et al., 1992). The same situation was found for udder traits. When median suspensory ligament was included with the other traits, it had almost no relationship with FHL and the RCR was almost flat. Front teat placement and rear teat placement showed the same tendency. Median suspensory had high positive phenotypic (0.85, 0.96) and genetic (0.60, 0.66) correlations with front teat placement and rear teat placement, respectively. The phenotypic and genetic correlations between front teat placement and rear teat placement were (0.63) and (0.85), respectively (Klassen et al., 1992).

The results showed that the relationship between phenotypic scores and FHL is not always linear, which is in agreement with the assumptions and results reported by Larroque and Ducrocq (2001) and Buenger et al. (2001). In the present study, some traits showed an intermediate optimum, such as pin setting and rear legs set. Rear teat placement has an optimal score of 5 (Holstein Canada, 2003), but the results did not show a clear intermediate optimum, as in the case of pin setting. Buenger et al. (2001) reported that body depth, rear leg side view, and teat length exhibited an intermediate optimum. Larroque and Ducrocq (2001) did not find traits with an intermediate optimum for the French population.

Sire effects were not included in the model. It has been suggested that the effect of the type trait’s phenotype on culling risk could be biased because part of it is accounted for by the sire effect (Larroque and Ducrocq, 2001; Buenger et al., 2001). However, in comparing the results from the present study with a previous run (results not shown), where sire effects were included in the model, we did not find any differences in the value of the estimates nor in the significance level.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
ACKNOWLEDGEMENTS
 REFERENCES
 
Survival analysis was used to investigate the effect of conformation traits on functional herd life in Quebec Holsteins. Traits describing udder and feet and legs had a strong relationship with survival. It was assumed that culling for low production is voluntary, but it would be interesting to determine to what extent culling for type is voluntary. Also it has been demonstrated that there is a nonlinear relationship between type traits and longevity.


   ACKNOWLEDGEMENTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 ACKNOWLEDGEMENTS
REFERENCES
 
The authors wish to thank V. Ducrocq for providing the Survival Kit V3.0 and valuable assistance throughout the study, and PATLQ for allowing access to data files. The authors further acknowledge the financial support of the Cattle Breeding Research Council (CBRC) of Canada.

Received for publication July 17, 2003. Accepted for publication September 14, 2003.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 ACKNOWLEDGEMENTS
 REFERENCES
 


Boettcher, P. J., L. K. Jairath, K. R. Koots, and J. C. M. Dekkers. 1997. Effects of interactions between type and milk production on survival traits of Canadian Holsteins. J. Dairy Sci. 80:2984–2995.[Abstract]

Boldman, K. G., A. E. Freeman, and B. L. Harris. 1992. Prediction of sire transmitting abilities for herd life from transmitting abilities for linear type traits. J. Dairy Sci. 75:552–563.[Abstract]

Buenger, A., V. Ducrocq, and H. H. Swalve. 2001. Analysis of survival in dairy cows with supplementary data on type scores and housing systems from a region of northwest Germany. J Dairy Sci. 84:1531–1541.[Abstract]

Burke, B. P., and D. A. Funk. 1993. Relationship of linear type traits and herd life under different management systems. J. Dairy Sci. 76:2773–2782.[Abstract]

Chirinos, Z., D. Hernandez, and M. J. Carabaño. 2003. Relación de caracteres de conformación con longevidad funcional en ganado Frisón utilizando técnicas de análisis de supervivencia. Rev. Asoc. Interprof. Desarrollo Agrario. ITEA.II(Suppl.):516–518.

Dekkers, J. C. M., L. K. Jairath, and B. H. Lawrence. 1994. Relationships between sire genetic evaluations for conformation and functional herd life of daughters. J. Dairy Sci. 77:844–854.[Abstract]

Ducrocq, V. P. 1987. An analysis of length of productive life in dairy cattle. Ph.D. Diss., Cornell Univ., Ithaca, NY.

Ducrocq, V. P., R. L. Quaas, E. J. Pollak, and G. Casella. 1988. Length of productive life of dairy cows. 2. Variance component estimation and sire evaluation. J. Dairy Sci. 71:3071–3079.[Abstract/Free Full Text]

Ducrocq, V. and J. Sölkner. 1998. The Survival Kit - a Fortran package for the analysis of survival data. Proc. 6th World Cong. Genet. Appl. Livest. Prod., Armidale, Australia 27:447–448.

Dürr, J. W. 1997. Genetic and phenotypic studies on culling in Quebec Holstein cows. Ph.D. Diss., McGill Univ., Montreal, Canada.

Holstein Canada. 2003. Subject: Canadian Type Classification Program. Available: http://www.holstein.ca/english/tc/program.htm. Accessed April 2003.

Klassen, D. J., H. G. Monardes, L. Jairath, R. I. Cue, and J. Hayes. 1992. Genetic correlations between lifetime production and linearized type in Canadian Holsteins. J. Dairy Sci. 75:2272–2282.[Abstract]

Larroque, H., and V. Ducrocq. 2001. Relationship between type and longevity in the Holstein breed. Genet. Sel. Evol. 33:39–59.[Medline]

Rogers, G. W., G. Banos, and U. Sander-Nielsen. 1999. Genetic correlations among protein yield, productive life, and type traits from the United States and diseases other than mastitis from Denmark and Sweden. J. Dairy Sci. 82:1331–1338.[Abstract]

Schneider, M. del P. 1998. Effect of type traits on herd life in Holstein cows. M.Sc. Diss., McGill Univ., Montreal, Canada.

Schneider, M. Del P., H. G. Monardes, and R. I. Cue. Effects of type traits on functional herd life in Holstein cows. Workshop on Genetic Improvement of Function Traits in Cattle, Jouy en Josas. France. Interbull Bull. 21:111–116.

Van der Linde, C., and G. de Jong. 2003. MACE for longevity traits. Interbull Bull. 30: 3–9.

Vollema, A. R., and A. F Groen. 1998. Conformation traits in survival analysis of longevity in dairy cattle. 6th World Congr. Genet. Appl. Livest. Prod., Armidale, Australia. 23:371.


This article has been cited by other articles:


Home page
 J DAIRY SCIHome page
A. Sewalem, F. Miglior, G. J. Kistemaker, P. Sullivan, and B. J. Van Doormaal
Relationship Between Reproduction Traits and Functional Longevity in Canadian Dairy Cattle
J Dairy Sci, April 1, 2008; 91(4): 1660 - 1668.
[Abstract] [Full Text] [PDF]

Home page
 J DAIRY SCIHome page
F. Miglior, A. Sewalem, J. Jamrozik, D. M. Lefebvre, and R. K. Moore
Analysis of Milk Urea Nitrogen and Lactose and Their Effect on Longevity in Canadian Dairy Cattle
J Dairy Sci, December 1, 2006; 89(12): 4886 - 4894.
[Abstract] [Full Text] [PDF]

Home page
 J DAIRY SCIHome page
A. Sewalem, F. Miglior, G. J. Kistemaker, and B. J. Van Doormaal
Analysis of the relationship between somatic cell score and functional longevity in canadian dairy cattle.
J Dairy Sci, September 1, 2006; 89(9): 3609 - 3614.
[Abstract] [Full Text] [PDF]

Home page
 J DAIRY SCIHome page
D. P. Berry, B. L. Harris, A. M. Winkelman, and W. Montgomerie
Phenotypic Associations Between Traits Other than Production and Longevity in New Zealand Dairy Cattle
J Dairy Sci, August 1, 2005; 88(8): 2962 - 2974.
[Abstract] [Full Text] [PDF]

Home page
 J DAIRY SCIHome page
A. Sewalem, G. J. Kistemaker, and B. J. Van Doormaal
Relationship Between Type Traits and Longevity in Canadian Jerseys and Ayrshires Using a Weibull Proportional Hazards Model
J Dairy Sci, April 1, 2005; 88(4): 1552 - 1560.
[Abstract] [Full Text] [PDF]

Home page
 J DAIRY SCIHome page
A. Sewalem, G. J. Kistemaker, F. Miglior, and B. J. Van Doormaal
Analysis of the Relationship Between Type Traits and Functional Survival in Canadian Holsteins Using a Weibull Proportional Hazards Model
J Dairy Sci, November 1, 2004; 87(11): 3938 - 3946.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Schneider, M. d. P.
Right arrow Articles by Monardes, H. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Schneider, M. d. P.
Right arrow Articles by Monardes, H. G.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS