Mastitis is one of the most costly diseases in dairy cows worldwide. Increased somatic cell count (SCC) is an indication of mastitis, often subclinical, which implies bacterial infection without clinical signs of inflammation. The aim was to investigate the occurrence of elevated udder SCC (defined as ≥200,000 cells/mL) over the lactation period, and before and after the dry period, for cows of different parity. The aim was also to analyze the association between prevalence and incidence of increased udder SCC and information on cow and herd level, such as breed and milking system type. Data were extracted from the Swedish Official Milk Recording Scheme between January 1, 2008, and December 31, 2011, including all herds with a yearly average of >60 cows. The data include descriptive information on herd and cow level and the results from the systematic test milking. The data included the following: for 2009, 239,182 cows in 1,633 herds; for 2010, 251,852 cows in 1,680 herds; and for 2011, 247,746 cows in 1,596 herds. The results show a peak in elevated udder SCC during the late summer season and that the highest proportion of cases occurs during the first lactation month; the latter was most prominent for primiparous cows. Forty-seven percent of all cows with elevated SCC recovered during dry period (went from high to low SCC), whereas 34% of all cows with low SCC before the dry period had an elevated SCC at first testing after calving. For first lactation cows, 19% had an elevated SCC at first test milking. When the outcomes for the 3 consecutive years were reanalyzed, it was confirmed that the effect of fixed factors such as breed, milk yield, and parity did not change over time, whereas the effect of milking system type did. For the incidence of becoming a new case and the prevalence of cows with elevated udder SCC, automatic milking system (AMS) was associated with reduced SCC in 2009 but associated with increased SCC in 2011. Regarding the proportion of new cases of elevated SCC per cow and year, AMS appeared to be a risk factor for all 3 yr, but the effect decreased over time. The shift for AMS from protective to risk factor regarding incidence of new cases and number of recordings with elevated SCC might reflect a change of the AMS population over these years. The findings indicate the need for appropriate udder health management customized to the system.

One of the most common and reliable ways of monitoring udder health and milk quality in dairy herds is by monthly cow composite somatic cell counts (CMSCC). However, such sampling can be time consuming, and more automated sampling tools entail extra costs. Machine learning methods for prediction have been widely investigated in mastitis detection research, and CMSCC is normally used as a predictor or gold standard in such models. Predicted CMSCC between samplings could supply important information and be used as an input for udder health decision-support tools. To our knowledge, methods to predict CMSCC are lacking. Our aim was to find a method to predict CMSCC by using regularly recorded quarter milk data such as milk flow or conductivity. The milk data were collected at the quarter level for 8 wk when milking 372 Holstein-Friesian cows, resulting in a data set of 30,734 records with information on 87 variables. The cows were milked in an automatic milking rotary and sampled once weekly to obtain CMSCC values. The machine learning methods chosen for evaluation were the generalized additive model (GAM), random forest, and multilayer perceptron (MLP). For each method, 4 models with different predictor variable setups were evaluated: models based on 7-d lagged or 3-d lagged records before the CMSCC sampling and additionally for each setup but removing cow number as a predictor variable (which captures indirect information regarding cows' overall level of CMSCC based on previous samplings). The methods were evaluated by a 5-fold cross validation and predictions on future data using models with the 4 different variable setups. The results indicated that GAM was the superior model, although MLP was equally good when fewer data were used. Information regarding the cows' level of previous CMSCC was shown to be important for prediction, lowering prediction error in both GAM and MLP. We conclude that the use of GAM or MLP for CMSCC prediction is promising.

The aim of this retrospective single-cohort study was to investigate if a rapid change in feeding, management, or housing or an increasing incidence of claw diseases or udder health problems is associated with decreased reproductive performance. Data on individual cows and herds were retrieved from the Swedish official milk recording system and questionnaire data on feeding system was obtained from the regional dairy associations. In total, 63,561 cows in 759 herds were included in the study. The associations between the probability of pregnancy at first insemination and number of inseminations per animal submitted for artificial insemination and potential predictor variables were investigated using a logistic regression model and a Poisson regression model, respectively. The results indicated that cows with severe claw lesions or an increasing somatic cell count after calving had a lower probability of pregnancy at first insemination and had a higher number of inseminations per animal submitted for artificial insemination than healthy cows. Variables representing a change in housing, production system, or milking system within the period from 6 mo before calving until establishment of a new pregnancy were significantly associated with decreased reproductive performance. No differences in fertility were observed between cows milked in an automatic milking system compared with cows milked conventionally. The results indicate that a change of system, rather than the actual type of milking or housing system negatively affects reproductive performance. Special attention should therefore be paid to the fertility of cows when the herd management is changing. It is also important to prevent claw lesions and increasing cell counts after calving to avoid a decrease in reproductive performance.

The aim of the study was to describe large Swedish dairy herds with high and low mortality risk in calves during the first 90 d of life, using herd-level data, and to evaluate if high calf mortality risk is associated with other herd-level management variables that influence cow health. A total of 57 Swedish dairy herds met the inclusion criteria of affiliation to the Swedish official milk recording scheme, herd size of ≥140 and ≥160 cows in 2008/2009 and 2009/2010, and calf mortality risks, classified as high (HM; calf mortality risk at least 3.5% in 2008/2009 and 5.5% in 2009/2010; n=28) or low (LM; calf mortality risk less than <1.5% in 2008/2009 and 2009/2010; n=29), and were thus included in the study. The data used in this study were collected from the Swedish Dairy association during the milking year 2009/2010. For LM herds, the calf mortality risk ranged from 0 to 1.46 (median=0.66) in 2008/2009 and from 0 to 1.48 (median=0.67) in 2009/2010. For HM herds, the calf mortality risk ranged from 3.57 to 11.52 (median=6.15) in 2008/2009 and from 5.88 to 18.23 (median=8.39) in 2009/2010. Median age at death was 28 d for HM and 37 d for LM herds. Associations between type of herd (HM or LM) and the production variables were evaluated using multi-correspondence analysis and logistic regression models covering the areas "mortality and culling," "health," "herd/production variables," and "fertility." Herds with HM risks during d 1 to 90 were associated with higher on-farm mortality rate in cows, lower average milk yield, higher incidence of antibiotic treatment, and a higher proportion of purchased animals. These results indicate that herds with HM risk during d 1 to 90 have coexisting issues concerning cow management and health. Future research is needed to evaluate if identifying HM herds and working with advisory and preventive manners at these herds also can be positive for a reduction of on-farm mortality and antibiotic usage, which are important issues from a global perspective.

No abstract