Estimated true prevalence and associated risk factors of bovine paratuberculosis antibodies in dairy herds in Bangladesh.

Bovine paratuberculosis (PTB), caused by Mycobacterium avium subsp. paratuberculosis (MAP), is a chronic infection that undermines cattle productivity and farm profitability. In Bangladesh, no previous studies have estimated PTB prevalence or identified associated risk factors. This cross-sectional study (January 2023 to December 2024) was conducted across 14 districts and 138 dairy farms to estimate true MAP antibody prevalence and identify factors associated antibody detection. Bulk milk samples from all herds (n = 138) and 880 individual milk samples from ELISA-positive herds were tested using a commercial indirect ELISA. True herd- and cow-level antibody prevalence was estimated using Bayesian latent class models. Risk factors were identified using multivariable logistic regression (herd-level) and mixed-effects logistic regression (cow-level). The apparent antibody prevalence was 7.2% at the herd-level and 12.2% at the cow-level, with herd-level prevalence ranging from 0.9% to 37%. After adjusting for test sensitivity and specificity, true antibody prevalence estimates decreased to 4.9% (herd level) and 9.4% (cow level). Herds with more than two retained placenta cases had significantly higher odds of MAP antibody detection (odds ratio [OR] = 9.42) than herds with two or fewer such events. At the cow level, older age (>7.9 years) was associated with higher odds of antibody positivity compared with cows ≤4 years. Cows with anestrus (OR = 3.89) or repeat breeding (OR = 2.63) had significantly higher odds of MAP antibody detection compared with cows without these disorders. Milk antibody levels showed a negative association with milk yield, with each unit increase in the sample-to-positive ratio corresponding to a 0.018 L/day reduction. These findings indicate that apparent antibody prevalence substantially overestimates PTB exposure in the study areas. The associations with retained placenta, anestrus, and repeat breeding suggest potential reproductive impacts that warrant further investigation. Given the purposive sampling strategy and reliance on milk ELISA, the results should be interpreted within the context of the study population. Enhancing diagnostic capacity and integrating PTB considerations into reproductive herd management may help reduce productivity losses, while larger, probability-based studies are needed to assess the national burden.