Herd-level risk factors associated with teat congestion, teat-end hyperkeratosis, milk flow dynamics, and take-off...

The objective of this study was to investigate herd-level risk factors for teat congestion, teat-end hyperkeratosis, bimodal milk flow curves (bimodality), and the take-off milk flow threshold of the automatic cluster remover system. In this repeated cross-sectional study, data from 212 milk quality evaluations conducted on 133 dairy farms as part of the Quality Milk Production Services extension program (Cornell University) were analyzed. Data on herd characteristics, milking routine variables, milking machine settings, visit type (initial or follow-up), and teat conditions were collected between October 2016 and February 2026. Average herd size was 1,161 lactating cows (range: 94 to 5,000), with most farms using conventional parlors (90.3%) and a minority using rotary parlors (9.8%); milking frequency was 2-times-per-day (14.9%), 3-times-per-day (82.2%), or 4-times-per-day (2.9%). Generalized linear mixed models were used to evaluate associations between explanatory variables and outcomes. Teat congestion was associated with herd size, strip yields, preparation lag time, teat-end hyperkeratosis, and herd size. A 1,000-cow increase in herd size reduced teat congestion by 3.5% (95% CI: -5.4, -1.5). A 10-percentage-point increase in strip yields between 150 and 250 mL decreased teat congestion by 1.2% (95% CI: -2.4, -0.0), whereas a 10-percentage-point increase in observations with preparation lag time < 90 s and teat-end hyperkeratosis increased teat congestion by 0.6 (95% CI: 0.0, 1.2) and 3.0% (95% CI: 1.7, 4.2), respectively. Teat-end hyperkeratosis was associated with the take-off milk flow threshold, average strip yields, bimodality, and milking liner shape. Teat-end hyperkeratosis decreased by 8.3 percentage points (95% CI: -15.8, -0.8) for each 1-kg increase in the take-off milk flow threshold and by 2.1 percentage points (95% CI: -3.5, -0.8) for each 10-percentage-point increase in bimodality, but increased by 1.7 percentage points (95% CI: 1.0, 2.4) for each 10-mL increase in average strip yields. The LSM (95% CI) were 29.8% (25.2, 34.4), 25.9% (21.4, 30.3), and 18.4% (14.0, 22.9) for round, triangular, and square liners, respectively. Bimodality was associated with milking frequency, preparation lag time, and visit type. The LSM (95% CI) for bimodality were 13.4% (6.1, 20.8), 31.8% (28.6, 35.0), and 20.1% (1.0, 39.2) for herds milked 2, 3, and 4 times daily, respectively. Bimodality decreased by 0.6 percentage points (95% CI: -1.2, -0.0) for every 10-s increase in average preparation lag time. The LSM (95% CI) were 18.0% (10.7, 25.3) at initial visits and 25.5% (18.0, 33.0) at follow-up visits. The take-off milk flow threshold differed by parlor type, milking frequency, and visit type. The LSM (95% CI) for the take-off milk flow threshold were 1.1 (0.9, 1.2) kg/min for conventional parlors and 1.4 (1.2, 1.5) kg/min for rotary parlors. For herds milking 2, 3, and 4 times daily, the LSM (95% CI) were 0.9 (0.8, 1.1), 1.1 (1.0, 1.2), and 1.7 (1.3, 2.1) kg/min at initial visits and 1.1 (0.7, 1.5), 1.3 (1.2, 1.4), and 1.2 (0.9, 1.5) kg/min at follow-up visits, respectively. In summary, herd characteristics, milking management, and machine settings were associated with teat condition and milk flow dynamics, and improvements observed between initial and follow-up visits suggest that targeted extension efforts can enhance milking management and teat condition.