Laboratory surveillance of epizootic animal diseases in Kazakhstan, 2016–2018: testing patterns, geographic concentration, and data gaps

Aggregated national laboratory records can provide insight into the allocation and outputs of veterinary surveillance, although they cannot independently establish population-level disease occurrence. We retrospectively analyzed annual territorial laboratory records to characterize surveillance for epizootic animal diseases in Kazakhstan from 2016 to 2018. The dataset comprised aggregated numbers tested and results recorded as reactive across disease–host categories and territorial reporting units. We evaluated testing volume, recorded positivity, geographic coverage, and concentration using the Herfindahl–Hirschman Index (HHI) and Gini coefficient. The analyzed dataset included 13,657 tests and 1,686 recorded reactive results across 33 tested disease–host categories and 16 territorial reporting units. Annual recorded positivity was 13.10% in 2016, 18.43% in 2017, and 5.43% in 2018, despite relatively stable testing volumes. Theileriosis in cattle accounted for 4,307 tests and 849 reactive results (19.71%; 95% CI, 18.55–20.93), followed in testing volume by dourine in horses (3,072 tests), salmonellosis across animal species (1,429 tests), and surra in camels (1,066 tests). For anthrax and blackleg, the number of recorded reactive results equaled the number tested (128/128 and 108/108, respectively), requiring cautious interpretation in the absence of sampling-purpose metadata. The five most frequently tested disease–host categories accounted for 77.25% of all tests. Surveillance activity was also geographically concentrated: Kyzylorda Region accounted for 47.16% of tests and 70.88% of recorded reactive results. Concentration indices were HHI = 0.267 and Gini = 0.689 for testing volume and HHI = 0.515 and Gini = 0.798 for reactive results. These findings indicate that the observed surveillance outputs were strongly influenced by a limited number of disease–host–territory combinations, while numerous listed diseases were tested infrequently or not tested during particular reporting periods. The absence of consistent metadata on animal species, sampling purpose, diagnostic methods, and underlying animal populations limited epidemiological interpretation. Standardized, sample-level reporting incorporating these variables would improve the comparability, interpretability, and risk-based use of national veterinary surveillance data.