Considerations for use of descriptive epidemiology to investigate fetal loss in dairy cows

Various descriptive approaches were taken in a prospective investigation to characterize fetal loss in cows on a California dairy. The methods and observations were offered for consideration by practitioners engaged in dairy herd health medicine. For 4,732 pregnancies followed from 2,163 cows in a 6.5 year period, the respective proportions (percentage) of cows aborting (1--cumulative proportion not aborting by 260 days) and abortion densities (abortions per 10,000 cow-days-at-risk) were 10.63 and 6.29 for all fetal deaths, 9.36 and 5.49 for deaths resulting in fetal expulsion, and 1.39 and 0.80 for deaths resulting in mummification. The greatest risk of fetal death (119 deaths/10,000 fetuses/d) was observed between 98 and 105 days of gestation, and median age at fetal death ranged from 99.0 to 105.5 days. Abortion density for fetal deaths resulting in mummification for cows conceiving during September (1.61/10,000 fetuses/d) and October (1.63/10,000 fetuses/d) was tenfold greater than that for cows conceiving in February (0.16/10,000 fetuses/d) and was twice that of the overall rate (0.84/10,000 fetuses/d). For cohorts of nonculled cows, abortion rate increased after 5 years of age, after 5 pregnancies, or after 4 calvings. For cows with at least one previous abortion, the proportion aborting (14.50%) was higher than that for cows without a previous abortion (12.14%). For a given gravidity, abortion rate was higher among cows that had experienced a previous abortion, compared with those that had not. These methods and observations may help provide a logical foundation on which to base clinical hypotheses regarding causes of abortion, and they may offer insight into pitfalls of bias and confounding to be anticipated.