Animal welfare is one of the most important aspects of animal breeding for the prevention and resistance to diseases. This study investigates the Genome-wide association study (GWAS) for three prevalent health problems in Holstein-Friesian cows: clinical mastitis (MA), clinical metritis (ME), and ovarian cysts (OC). It implements a GWAS approach to analyse genotypic data from 2 112 Polish Holstein-Friesian cows. MA (1 040 sick), ME (1 103 sick), and OC (636 sick) were scored as binary traits. The single nucleotide polymorphism (SNP) effects were estimated using a two-step procedure that involved genomic prediction of breeding values and subsequent estimation of the SNP effect using the back-solve approach. Significant SNPs were genomically annotated using the Variant Effect Predictor and the corresponding genes were tested for enrichment of underlying metabolic pathways. We identified 13 significant SNPs associated with MA, 44 with ME, and 14 with OC. Genes marked by SNPs significant for MA were enriched in functions related to protein export, non-alcoholic fatty liver disease, and autophagy. Pathways enriched in genes significant for ME involved synaptic vesicle cycle and acute myeloid leukaemia. For OC, the insulin-like growth factor 1 receptor gene (IGF1R) was represented in multiple pathways, including ovarian steroidogenesis and PI3K-AKT signalling. These findings provide new information on the genetic architecture and metabolic basis of health traits in Holstein cattle and highlight potential targets for genetic improvement. The limited overlap with previous studies underscores the complexity and polygenic nature of these traits.