Heritability and correlations were estimated for the cost of health treatments during first lactations of Holstein cows (n = 2,214) in 8 high-performance dairy herds. A fixed treatment cost (including veterinary fees, pharmaceutical expense, and labor cost) was assigned to 14 types of health treatment cost. Treatments were grouped into 5 categories: mastitis (MAST; including diagnostic tests), reproduction (REPRO; cystic ovary, retained placenta, metritis), lameness (LAME; hoof treatments), metabolic (META; milk fever, displaced abomasum, ketosis, digestive), and miscellaneous (MISC; respiratory, injury, other). Total health cost (THC) was summed for each cow. The fixed effect of herd and the random effect of cow nested within herd were fitted with linear animal models using ASReml. The estimates of heritability were significant for MAST, LAME, META, and THC (Table 1). Genetic correlations between categories of health treatment cost were largest for MAST and REPRO (0.85 ± 0.20); however, phenotypic correlations between all categories were small (r <0.16). The THC had a large genetic correlation with 305-d milk production (0.44 ± 0.18) and somatic cell score (0.93 ± 0.13). The genetic correlation (−0.60 ± 0.16) between THC and udder depth indicated shallower udders were genetically associated with less THC. Genetic selection for reduced health treatment cost should be possible from the use of producer-recorded health treatment records that are supplemented with treatment cost. Table 1. Estimates of heritability (in bold), genetic correlations (above diagonal), and phenotypic correlations (below diagonal) for the treatment costs of 5 health categories and total health treatment cost for first parity

	MAST	REPRO	LAME	META	MISC	THC
MAST	0.13*	0.85*	0.34	0.52	0.66	0.92*
REPRO	0.00	0.04	0.41	0.73*	0.59	0.91*
LAME	0.03	−0.01	0.10*	0.56*	0.21	0.65*
META	0.02	0.14*	0.02	0.12*	0.40	0.85*
MISC	0.04*	0.02	−0.05*	0.16*	0.04	0.72*
THC	0.34*	0.66*	0.27*	0.63*	0.39*	0.27*

*Estimate significantly different from zero based on 95% CI.