Monitoring recovery after clinical mastitis in dairy cows after 2 different types of treatment

Summary

A prospective observational study, split over two periods, was set up to describe initial symptom patterns and convalescence progression of treated clinical cases of mastitis on 29 Danish dairy farms. In the first study period all cows were treated with antibiotics (group A); in the second study period cows were treated with classical homeopathy (group B1) or antibiotics (group B2). Initial symptom and treatment patterns were examined by factor analyses. The 1st analysis included all clinical observations made on treatment day. The difference between the 3 groups in mean scores for revealed factors was tested by Students t-test. The 2nd and 3rd analyses were performed separately for groups A and B1 and included all variables measured. Differences between groups A and B1 in prevalence of clinical mastitis scores (no, mild, moderate and severe mastitis symptoms), at the end of the follow-up period were compared by 2 test. Changes over time in the prevalence of cows with clinical mastitis (no symptoms vs. any symptoms) was tested for groups A and B1 by logistic analysis with inclusion of effects of time, lactation stage, parity, logarithm of the individual somatic cell count (logSCC), yield (kg milk/day) and acute and chronic mastitis score in the model. The first factor analysis revealed 2 factors, describing 52% of the variation in the data; these were labelled ‘acute symptoms’ and ‘subacute symptoms’. There were significant differences between the 3 groups in mean scores of revealed factors; factor 1 was highest for group A and factor 2 was highest for group B1. The subsequent factor analyses resulted in 5 and 6 factors explaining approx 50% of the variation in the data and related to clinical symptoms, but for group B1 also to exterior and behaviour of the cows. The initial comparison of mastitis severity between groups revealed no significant differences in mastitis status. When corrected for cows withdrawn, this difference became significant (P<0.001). The logistic analyses revealed that there was a significant effect of time after treatment, logSCC and parity (only group B1).