Contents Introduction 9 Kenyan dairy sector 9 Antibiotics in dairy production 11

Shitandi, A. 2004. Risk factors and control strategies for antibiotic residues in milk at farm level in Kenya. Doctoral dissertation. ISSN 1401-6249. ISBN 91-576-6477-3. Antibiotic residues in milk in Kenya constitute a problem and there is strong interest to enhance milk safety. The research sought to establish whether the prevention of residues in milk was viable at the farm level. It also evaluated implications of non-restricted use of antibiotics in terms of resistance among Staphylococcus aureus isolates from the small and large-scale farms, which are the two main producer categories in the country. The research demonstrated the possible use of a low cost microbiological screening assay (two-tube test) for detection of antibiotic residues in a local dairy. The verified limits of detection for selected antibiotics were below or near established Codex Alimentarius standards. During 2000–2001, 14.9% of the analysed 1109 herd milk samples caused inhibition in the microbial test. Eleven percent (n=118) of the samples had ß-lactam type residues in concentrations ≥ 10μg/kg exceeding established limits two fold and the contamination differed significantly (p<0.001) among the two types of producers. Circumstantial evidence was provided between rising incidence of antibiotics and prevalence of drug resistant S. aureus isolates from mastitis milk. The isolates were tested for susceptibility to five antibiotics from different families. The overall mean prevalence of multidrug resistance from 402 isolates was 34.3% for small and 18.0% for large-scale farms (p < 0.05). Five characteristics were identified based on Hazard Analysis Critical Control Point principles, which predicted residue contamination on farms and were associated with a higher risk (p < 0.05). The characteristics were used to develop a simple risk assessment tool, which was validated in local and remote cohort farms. The tool was then applied in a single blind randomised control study on small-scale farms (42 cases and 42 controls) alongside the two-tube test. A 52% reduction in incidence of antibiotic residue violation was observed in the treatment group compared to the control farms (p < 0.05). The risk assessment tool and the two-tube screening assay were concluded to offer a viable strategy to minimize violative antibiotic residues in farm milk within a control program. The results suggest that this approach could also be implemented on other farms in low-income countries.