Byeong-Cheol Song
Municipal dog shelters are often grouped as outsourced facilities despite operator differences. We analyzed 33,875 dog admissions and 90 municipality-year observations from 2016 to 2025 across nine Korean municipalities, with three per operator category; operator type was fixed and nested within municipality. Among 32,838 known outcomes, live-release rates were 69.1%, 58.1%, and 40.0%, and euthanasia rates were 14.0%, 22.8%, and 42.5% in corporation-operated, veterinary hospital-operated, and individually operated shelters, respectively. The corporation live-release estimate ranged from 62.6% to 72.0% under alternative classifications of unresolved records. Small-sample cluster-robust comparisons were imprecise: the individually versus corporation-operated euthanasia odds ratio was 4.72 (95% confidence interval, 0.40-56.35), and every operator interval included 1.0. Contextual adjustment yielded odds ratios of 0.97 for both outcomes for veterinary hospital- versus corporation-operated shelters. Among 33,836 admissions with common 180-day follow-up, live-release rates were 59.9%, 53.0%, and 39.8%, and euthanasia rates were 9.1%, 16.7%, and 42.0%; annual odds ratios were 1.05 (0.93-1.19) and 0.90 (0.78-1.05), respectively. Results describe operator-linked municipal systems, not causal effects. Standardized follow-up, complete reporting, capacity monitoring, and larger prospective studies are needed.