The European Commission (EC) has requested the European Food Safety Authority (EFSA) to provide scientific and technical assistance on an assessment of the capacity of the proposed revised bovine spongiform encephalopathy (BSE) monitoring programme in Croatia to allow the detection of BSE, both classical and atypical strains, with a design prevalence of at least 1 case per 100,000 of the adult cattle population of the EU26, i.e. the epidemiological unit made of all European Union (EU) Member States (MS) except Bulgaria and Romania.
Regulation (EC) No 999/2001 allows the revision of individual surveillance programmes if there is a demonstrable improvement of the BSE situation based on epidemiological criteria. Following various past assessments, all MS except Croatia, Bulgaria and Romania (EU25) have been authorised to modify their BSE monitoring programmes.
On 22 January 2015, Croatia submitted to the EC a request to revise its BSE monitoring programme. Under this revision Croatia would stop testing all healthy slaughtered (HS) cattle and would test all ‘at risk’ cattle identified through active surveillance (i.e. emergency slaughter (ES) fallen stock (FS) and animals identified as disease suspect at ante mortem inspection (AM) above 36 months of age) plus all TSE clinically suspected cases (CS) obtained via passive surveillance.
The EC favourably assessed the Croatian implementation of the BSE monitoring system, the bovine identification and traceability system and the feed ban, resolving that the EU26 should be considered as one unique epidemiological unit for this evaluation, consistent with the approach adopted in previous assessments of BSE surveillance in the EU.
Data related to the EU25 and to Croatia have been updated to apply the Cattle Transmissible Spongiform Encephalopathy (TSE) Monitoring Model (C-TSEMM), an in-house model originally developed to evaluate the performance of different BSE monitoring regimes in cattle, and previously applied to similar assessments. The model, that considers as surveillance components CS, HS, FS and ES (including AM), is designed to estimate the sensitivity of surveillance systems in terms of the minimum design prevalence detected for a group of countries that apply the same surveillance regime. It is not possible to directly model the design prevalence in a scenario where the surveillance regime of a MS joining an epidemiological unit is different from that of the unit (EU25), such as the proposed BSE monitoring programme of Croatia.
Following the description of the requirement described in the ToR, the sensitivity of the surveillance system has been described as the design prevalence that surveillance scenarios for the populations of interest are able to detect. This estimated design prevalence is compared to the minimum requirement of 1 case per 100,000. If a result shows a denominator larger than 100,000, the detection level is referred to as a lower design prevalence than the one required, and indicates that the surveillance scenario has a higher sensitivity than the minimum required.
The results of the assessment showed that, using data up to 2014, the former EU25 BSE surveillance regime (HS older than 72 months, ES and FS older than 48 months, all CS) would allow the detection of BSE (both classical and atypical strains) with a design prevalence of at least 1 per 4,684,990 of the adult population (cattle older than 24 months), higher (i.e. less sensitive) than the 1 per 6,354,930 obtained when applying data up to 2011, but lower than the minimum required (1 per 100,000). This reduction in sensitivity is due to the cessation of the compulsory testing of HS animals since 2013.
The results of the assessment showed that, using data up to 2014, the current EU25 BSE surveillance regime (no HS testing, ES and FS older than 48 months, all CS) would allow the detection of BSE (both classical and atypical strains) with a design prevalence of at least 1 per 3,769,555 of the adult population, higher (i.e. less sensitive) than the 1 per 4,021,940 obtained when applying data up to 2011, but lower than the minimum required (1 per 100,000). This reduction in sensitivity is due to the lower absolute number of ‘at risk’ animals of active surveillance and TSE clinically suspected cases tested across the EU25 between 2012 and 2014.
After the addition of Croatia to the EU25 epidemiological unit (EU26) and assuming the current EU25 surveillance regime for all MS including Croatia, the design prevalence of the surveillance in the EU26 would be 1 per 3,789,838 adult population, lower than the minimum required (1 per 100,000).
If Croatia implemented the proposed surveillance regime (no HS testing, ES and FS older than 36 months, all CS) the sensitivity at the EU26 level would be slightly higher (i.e. lower design prevalence) than the one applying the current EU25 surveillance regime.
It is recommended that the C-TSEMM model is run on an annual basis, and with updated data, in order to monitor the ability of the current surveillance regime to detect BSE at both MS and EU26 levels. The outputs could be included in the EU TSE Annual Report. However, the adequacy and assumptions of the model should be revised considering the evolution of the epidemiological situation of BSE in the EU. It is also recommended to monitor the data at MS level in order to evaluate the coverage by testing group in the context of changing standing cattle populations, and to identify any shortcomings affecting the overall sensitivity of the surveillance system.