The primary objectives were to investigate the efficacy of bovine casein hydrolysate (bCNH) as a dry cow therapy at (1) preventing new intramammary infection (IMI) postpartum of all bacteria and coagulase-negative staphylococci (CNS), and (2) curing existing subclinical infections, mainly of CNS. The secondary objective was to measure the effects of bCNH on milk yield, composition, and somatic cell count (SCC) during the lactation postcalving. The trial was conducted as a randomized, blinded controlled experiment. Israeli Holstein dairy cows (n = 170) in first or higher lactations were recruited from 4 large commercial dairy herds. Cows were enrolled following clinical examination and bacteriological sampling of each quarter, which was the experimental unit. Random allocation was implemented at the cow level. All quarters of 100 cows were treated with 1 dose of bCNH (60 mg diluted in 20 mL of sterile solution) and those of 70 control cows were treated with saline solution. Clinical assessment of each cow's general appearance, teat-end leakage, and teat morphology was performed for 0, 1, 2, 3, 7, and 14 d after treatment, together with follow-up clinical observation and clinical examination of udder quarters. Quarter aseptic milk samples were obtained for bacteriological culture 48 h pretreatment, at time of treatment, and 3 and 5 d postcalving. Multivariable analyses were conducted to study the effects of bCNH on cure and prevention of IMI, adjusting for parity, farm, average of daily milk yield for 305 d, and average of monthly SCC values for 305 d of previous lactation. The odds of preventing IMI in cows treated with bCNH at dry-off were 2.15 times higher [95% confidence interval (CI): 1.15 to 4.00] than in cows treated with saline. Prevention was mostly of CNS. The odds of preventing CNS in cows treated with bCNH at dry-off were 2.20 times higher (95% CI: 1.58 to 3.07) than in control cows. The odds of curing IMI caused by CNS in cows treated with bCNH at dry-off were 4.80 times higher (95% CI: 0.75 to 30.75) than in saline-treated cows. Log SCC, adjusted to that of the previous lactation, was lower in the bCNH group compared with controls for 305 d in milk postcalving. The average milk yield per day for 305 d, adjusted to average daily milk yield of previous lactation, was higher by 2.1 kg in the bCNH group compared with controls (95% CI: 1.21 to 3.20). Clinical assessment of udders and cows posttreatment showed no negative effects of bCNH. Incidence of stillbirth, clinical mastitis, retained placenta, endometritis (5 to 12 d postcalving), ketosis, abortions, and reproduction did not differ between the 2 groups. Results suggest that a single intramammary administration of bCNH at dry-off effectively increases milk yield and lowers SCC, prevents new IMI during the dry period, and may be a beneficial alternative for curing existing IMI at dry-off, mainly by CNS.