RESUMO
AIMS: The present study employed a network analysis approach to scrutinize a pig supply chain in a repeated outbreak province for human streptococcosis in Thailand and identified important actors that should be focused on for tailoring appropriate interventions. METHODS AND RESULTS: Nakhon Sawan province was chosen as the study site as the cases of human streptococcosis have been consecutively reported since 2014, with the number of cases ranging from 21 to 63. A questionnaire survey was used to collect data from actors along the pig supply chain, including pig farms, slaughterhouses, pork sellers, restaurants and customers. A one-mode-directed network was then constructed. Degree and betweenness centrality values were measured. We found that the supply chain of pork products comprised 314 nodes and 296 directed ties. A retailer got the highest overall degree, out-degree and betweenness centrality values at 35, 34, and 65.3, respectively. For in-degree centrality, the highest was identified in a customer at 9. Interestingly, this customer bought pork products from nine different mobile groceries. CONCLUSIONS: Both public health and veterinary authorities should extend their surveillance activities to cover all actors in the supply chain to strengthen overall disease prevention and control for streptococcosis.
RESUMO
A cross-sectional study evaluated the risk of zoonotic Streptococcus suis (S. suis) illness from consuming raw pork and swine blood in Nakhon Sawan Province. A four-step risk assessment recommended by the Codex Alimentarius Commission was used to evaluate the risk along the pork supply chain. A total of 480 pork and swine blood samples were collected from the abattoir (n = 120) and retail (n = 360) during December 2020 and January 2021. Streptococcus suis in samples was enumerated using a culture-based technique and then confirmed by the biochemical and molecular technique. Streptococcus suis was serotyped by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Two positive swine blood samples were contaminated with non-zoonotic S. suis serotype 23 at retail. In the case of all negative samples, the deterministic prevalence becomes zero and then the risk could not be estimated. Otherwise, the beta probability distribution was used to describe the probabilistic prevalence, while the maximum likelihood estimator was applied to estimate the upper limit of a probability distribution of concentration. The district averages of probabilistic prevalences of zoonotic S. suis in pork products at abattoir and retail were 9.9% and 4.1%, respectively. The district averages of concentrations of zoonotic S. suis in pork and blood samples from abattoir were 6.8 × 10-3 cfu/g and 6.83 cfu/ml and in pork and blood samples from retail were 2.3 × 10-3 cfu/g and 2.30 cfu/ml, respectively. The overall annual risk estimate per 100,000 population in pork and swine blood from abattoir and retail were 9.8 × 10-11 , 2.2 × 10-6 , 5.4 × 10-13 , and 8.3 × 10-8 . These risk estimates were negligible (<10-6 ) except for the annual risk estimate in swine blood from the abattoir. The results from this cross-sectional risk assessment should prompt the food safety regulator to cautiously sample by taking into account the duration of sampling and sample size.
