Release assessments
The probability Pre was estimated with 2 conditional probabilities: The probability of ASFV infection in a country o (P1 ) and the probability of selecting an ASFV contaminated meat during the month m before detection of the infection (P2 ). Probability of an ASFV infection in the country o (P1 ) was modelled using a beta distribution (α1, α2), where the α1 was the number of months with at least one undetected ASF outbreak in the country and α2 was the number of months considered for the analysis (OIE, 2019).
The P2 was also estimated with a beta distribution (α3, α4), where α3 was the quantity (kg) of potentially infected pig products in the country o in the month m(Qim ) and α4 was the quantity of pig products produced in countryo in the month m (Nm ). The quantity of infected pig products in the country o (Qim ) was calculated with 3 parameters: (NI ) estimated number of infected pigs in the country which was calculated by multiplying the number of undetected outbreaks (Ou ), average herd size (To ), and intraherd prevalence (Hp ) similar to the assessment of the ASF introduction by live pig imported legally during the HRP, (Pm ) the probability that an ASFV infected pig is transformed into meat and (Mp ) the average weight of the products (kg) obtained per slaughtered pig. The probability Pm was estimated as follows:
Pm = P3 × P4 ×Psm × Pus
In this formula, P3 is the probability that a pig survives ASFV infection; P4 is the probability that the ASFV infected pig survives the transportation;Psm is the probability of a pig going to a slaughterhouse during a specific month based on the pig census and monthly proportion of pigs slaughtered on average (EUROSTAT, 2014) and;Pus is the probability that an ASFV-infected pig is undetected during the clinical checks in the slaughterhouse.