IntroductionIn South Korea, the companion-dog population is strongly skewed toward a small number of small-breed purebred dogs, and distinguishing breed-type differences in recorded diagnoses from confounding by age, neuter status, and healthcare utilization is challenging in primary-care data.MethodsThis retrospective study used electronic medical records (EMRs) from primary-care veterinary clinics in South Korea, covering January 2016–December 2024, to compare the adjusted odds of the 20 most frequently recorded diagnoses between purebred and mixed-breed dogs. Records were standardized using Veterinary Nomenclature (VENOM) coding. For each condition, a separate multivariable logistic regression model estimated the adjusted odds ratio (aOR) and 95% confidence interval (CI) for purebred status, adjusting for age, sex, neuter status, and visit frequency. Age was assigned as age at first recorded diagnosis for cases and last observed age for non-cases.Results and DiscussionAfter adjustment, purebred dogs showed higher odds of recorded diagnosis for nine conditions, including meningoencephalitis (aOR 2.77; 95% CI 1.17–6.57), intervertebral disc disorder (aOR 1.85; 95% CI 1.49–2.29), myxomatous mitral valve degeneration (aOR 1.57; 95% CI 1.46–1.68), and patellar luxation (aOR 1.54; 95% CI 1.38–1.71). Mammary neoplasm showed higher crude prevalence in mixed-breed dogs, and this association remained statistically significant after adjustment for binary neuter status (aOR 0.80; 95% CI 0.68–0.95; Adj p = 0.0202), likely reflecting unmeasured confounding by the timing of neutering. Ten conditions, including chronic kidney disease and unspecified dermatitis, showed no statistically significant difference between groups. These findings indicate breed-type differences in the odds of recorded diagnoses in a Korean primary-care EMR dataset rather than true disease risk or causal effects. Because the cohort was dominated by small-breed dogs and the analyses could not fully account for body size, follow-up duration, detection bias, or diagnostic-coding heterogeneity, the observed associations may reflect breed composition, morphology, owner behavior, healthcare access, and recording practices in addition to underlying biological differences.