Censoring in survival analysis: a simulation study of the effect of milk yield on conception

Survival-analysis methods often are used to analyze data from dairy herds w here the outcome of interest is the interval from calving to conception. Th e purpose of this study was to determine whether an association between mil k yield and culling biases the estimation of the effect of milk yield on co nception. This was done by simulating four different scenarios modeling dai ry-cattle milk yield and reproductive performance with known relationships among study factors. Cox's proportional-hazards model was used to analyze t he effect of milk yield on days open under the following four scenarios: (l ) no association between milk yield and culling or between milk yield and c onception; (2) association between milk yield and culling only; (3) associa tion between milk yield and conception only; (4) associations between milk yield and both culling and conception. The analyses also were repeated for data sets with an association between milk yield and culling, but with prob abilities of culling ranging from 0.01 to 0.4. An effect of milk production on culling appeared to cause a small increase in the parameter estimates f or the association of milk yield and days open - particularly when the prob ability of culling was high. The effect of high milk production on median d ays open (as estimated by survival functions) changed by 2 to 4 days when a n association between milli yield and culling was programmed in the simulat ed data sets. (C) 2001 Elsevier Science B.V. All rights reserved.