Data collection: animal diseases

Introduction

The European Commission regularly asks EFSA for scientific and technical support in the epidemiological analysis of animal disease outbreaks – such as African swine fever, lumpy skin disease and avian influenza – and to report or assess surveillance data (e.g. Echinococcus multilocularis and avian influenza).

In response to these requests, EFSA has in recent years carried out several data collections and gathered information on outbreaks, surveillance activities and animal populations (poultry, domestic pigs, cattle and wildlife such as wild boar).

The experience gained in these activities led EFSA to reassess the way data is collected on animal populations and animal diseases with a view to optimising the process.

Currently EFSA’s data models are not completely harmonised, its Data Warehouse is not connected to the European Commission’s Animal Disease Notification System (ADNS), and data submission and validation is not automated, which means data collections are labour intensive for both EFSA and data providers in the Member States.

EFSA therefore set up the SIGMA project, with the aim of harmonising data models and revamping processes for data submission, validation, analysis and reporting.

What is the SIGMA project?

The SIGMA project aims to:

  • reduce the amount of data that is submitted manually to EFSA by Member States by setting up automated procedures;
  • avoid double reporting to EFSA and, possibly, to other systems;
  • provide Members States with instruments to automatically produce national reports on animal health and surveillance in a protected environment (secure connections, log-in credentials);
  • increase the quality and comparability of the data received from Member States;
  • shorten the time needed to retrieve up-to-date data that is relevant for risk assessment.

The project is being implemented by a consortium led by the Istituto Zooprofilattico Sperimentale (IZS) Abruzzo e Molise “G. Caporale” in partnership with the Friedrich Loeffler Institut (FLI), the Swedish National Veterinary Institute (SVA), the Bulgarian Food Safety Agency and the Institute of Veterinary Medicine and Animal Sciences, Estonian University of Life Sciences.

SIGMA will be implemented in three phases:

Phase 1

The main goals of the first phase are to:

  • Design a harmonised data model, the SIGMA Animal Disease Data Model (σ-ADM), which includes a section on animal population and another on laboratory results.
  • Provide a comprehensive overview at Member State level (country cards) of the institutions responsible for the collection of data related to animal health and animal populations.

Phase 2

The second phase will be dedicated to the implementation of the framework with the Member States that volunteer to take part in the pilot. This phase will concentrate on data related to avian influenza (AI) and African swine fever (ASF). Main elements:

  • Support the volunteering Member States by designing extract transform load (ETL) processes to select, transform and transmit the relevant national data in line with the σ-ADM.
  • Surveillance reporting on avian influenza: support the transition from the European Commission system to the EFSA data collection framework (DCF).
  • African Swine Fever: integration of the outbreak data collection.
  • Pig and poultry populations: support the setting of data flows within and outside the volunteering Member States.
  • Analytical interactive online tools: preliminary case study on the connection of tools to EFSA’s Data Warehouse for ASF and AI disease outbreak analysis and reporting to be made available to the countries submitting data.

Phase 3

The project will be finalised with the collection of the data and storage in the Data Warehouse. In particular:

  • ETL processes: implementation of the processes (to select, transform and transmit the relevant national data in line with the σ-ADM) in the volunteering Member States.
  • Final report on the quality of the data collected by means of the SIGMA framework.
Data Sources on Animal Diseases in EU Member States
Data Sources on Animal Diseases in EU Member States
 

FAQ

1. What is SIGMA?

The aim of the project is to optimise the submission to EFSA of data about animal population and disease surveillance from EU Member States and neighbouring countries. This will produce a consistent, harmonised and structured database on animal diseases such as African swine fever, avian influenza and lumpy skin disease.
More information here.

2. Who can be partners in the project?

Data providers such as veterinary authorities from Member States and neighbouring countries (e.g. pre-accession countries), academic institutes and central laboratories can participate in the project. As of May 2019, 10 countries – Austria, Spain, Italy, Croatia, Romania, Bulgaria, Estonia, Greece, Sweden and Iceland – are involved.

3. What are the advantages of SIGMA?

  • Minimises the workload of the data providers in data submission, through quicker, automated data extraction and standardisation.
  • Possibility of using the submitted data to produce reports for the EC and for national purposes. An example is the report on E. multilocularis.
  • EFSA is able to produce relevant and timely feedback to risk managers (Member States, other countries and the European Commission).
  • Increases the quality of EFSA outputs.

4. What are the challenges?

The main hurdle to overcome is the different approaches taken in different countries towards: data sharing policy; data dictionaries (definitions of variables) e.g. classification of animal population species and types of animal production and IT infrastructure.

5. Which animal diseases are targeted by SIGMA?

In the initial pilot phase, data will be collected on African swine fever, avian flu and lumpy skin disease. Other diseases and animal populations may be covered in the future.

6. What kind of data are collected?

  • Data on animal populations (e.g. poultry, pigs, bovines): e.g. farm location, farm size, animal species, type of production, etc.
  • Surveillance data on avian influenza, African swine fever and lumpy skin disease, where the main variable fields refer to laboratory data (sample ID, type of sample, date of sampling and testing, diagnostic test used, results, etc.).

7. What resolution of data is required?

Surveillance data can be provided at sample level or as “aggregate-able” data i.e. indicating only the desired level of resolution (e.g. NUTS3 level instead of x-y coordinates). Population data can be provided at farm level (establishment) and/or at single animal level (when opportune) at the desired level of precision (e.g. NUTS3 level instead of x-y coordinates)

8. What type of data are Member States obliged to submit?  

The data that the MSs are obliged to submit are those covered by the relevant EU legislation e.g. data on avian flu surveillance. The additional data requested in the SIGMA project are not compulsory, but essential to perform fit for purpose statistical analysis. Note that all data are protected by data protection legislation.

9. What happens to the data once it has been submitted to EFSA in the SIGMA framework?  

All data submitted to EFSA, once validated by the Member State, are stored in the EFSA Scientific Data Warehouse. The data are then used by EFSA to perform risk assessments. The output of the risk assessment is shared with the data providers to ensure consistency before publication.

10. How protected are the data once submitted to EFSA?

Before the publication of a risk assessment, the data are not accessible to anyone but EFSA. After publication, according to the EFSA policy on transparency, the data are potentially available for consultation upon request. Nonetheless, the data considered confidential by the data providers as well as the ones covered by data protection legislation (e.g. the geo-coordinates of a farm) will not be disclosed.

11. What are the conditions for a country to get involved in SIGMA in terms of staff, IT structures, data management system in place?

No special conditions are required: the SIGMA consortium will first perform an analysis of the data flow at national level.