Assessing antimicrobial resistance occurrence in the Portuguese food system : poultry, pigs and derived food, 2014-2018

• Published in: Zoonoses and public health 2022-06, Vol.69 (4), p.312-324
• Main Authors: Costa, Miguel, Cardo, Miguel, D'Anjo, Maria Cara, Leite, Andreia
• Format: Article
• Language: English
• Subjects: Abattoirs; Animal health; Animal populations; Animals; Anti-Bacterial Agents - pharmacology; Anti-Bacterial Agents - therapeutic use; Anti-Infective Agents; Antimicrobial agents; Antimicrobial resistance; Campylobacter; Chickens - microbiology; Confidence intervals; Drug resistance; Drug Resistance, Bacterial; E coli; Escherichia coli; Escherichia coli Infections - veterinary; Farms; Female; Food; Food industry; Food security; Food systems; Foodborne pathogens; Foods; HACCP; Pork; Portugal - epidemiology; Poultry; Quinolones; Risk analysis; Risk factors; Salmonella; Sampling; Statistical analysis; Sulfonamides; Summer; Surveillance; Swine; Swine Diseases - epidemiology; Tetracyclines; Turkeys
• ISSN: 1863-1959; 1863-2378
• DOI: 10.1111/zph.12920

Research Areas: Public, Environmental & Occupational Health ; Infectious Diseases ; Veterinary Sciences Antimicrobial resistance (AMR) spread in the food system is a major threat to public and animal health. We describe AMR trends in zoonotic bacteria and identify risk factors associated with AMR occurrence in animals and derived food in Portugal. Data from the Portuguese AMR surveillance programme on food- producing ani-mals, 2014–2018, were used. AMR frequencies were calculated for Escherichia coli, Campylobacter and Salmonella in broilers, turkeys, pig populations and their derived food products. AMR-associated factors were studied for Salmonella isolates: popula-tion, sampling stage (farm, slaughterhouse and processing plant), sample type (en-vironmental, carcase and food), sampler (HACCP, industry, official and official and industry), sample context (control and eradication programmes and monitoring), year (2014–2018) and season. Logistic regression was applied to estimate crude odds ratio and adjusted odds ratio (aOR) with 95% confidence intervals (95% CI). The final models were obtained using a backward stepwise method. This study in-cluded 2,157 Escherichia coli, 561 Campylobacter and 1,071 Salmonella isolates. The highest prevalence estimates amongst tested antimicrobials for each bacterial spe-cies in 2014–2018 had the following ranges: (i) (fluoro)quinolones: E. coli: 8 4 % – 9 3 % , Campylobacter: 94%– 98%; (ii) tetracyclines: E. coli: 68%–91%, Campylobacter: 87%– 91%, Salmonella: 72%; (iii) penicillins: E. coli: 82%–100%; (iv) sulphonamides: E. coli: 68%– 82%. Compared with the reference categories for host (broiler), year (2014), season (winter) and sampler (HACCP own checks), resistance to at least one anti-microbial in Salmonella was significantly less likely in laying hens (aOR 0.28; 95% CI: 0.18– 0.42), 2016 (aOR 0.56; 95% CI: 0.33–0.93), 2017 (aOR 0.29; 95% CI: 0.17–0.51) and 2018 (aOR 0.35; 95% CI: 0.20–0.61), autumn (aOR 0.63; 95% CI: 0.40–0.97) and more likely to occur in broiler products (aOR 5.14; 95% CI: 2.61–10.54), pork products (aOR 6.84; 95% CI: 3.74–12.98) and official and industry combined sampling (aOR 2.16; 95% CI: 1.06–4.47). This study reveals a high prevalence of Salmonella resistance, especially during the summer and in post-farm stages of the Portuguese food system to nearly all antimicrobials and in the summer in farms to (fluoro)quinolones. Measures to tackle resistance are required