“If history repeats itself, and the unexpected always happens, how incapable must Man be of learning from experience?”
—George Bernard Shaw (1856–1950)

Spotting the signals of new or re-emerging risks is a major challenge for risk managers and risk assessors. Can incidents be avoided before they happen?

Data can be drawn from myriad sources. The data are of varying relevance, quality, timeliness and veracity. How do you “see the wood for the trees,” and how do you ensure that appropriate action is taken and that communications are proportionate and appropriate?

Since its establishment in April 2000, the UK Food Standards Agency (FSA) has managed more than 12,000 food safety incidents of varying scale and complexity. During its incident investigations, the FSA routinely gathers and generates large amounts of food safety data from thousands of sources.

The FSA believes that the early identification or prediction of incidents is fundamental to the implementation of timely, proportionate and decisive actions to protect consumers. It is, therefore, examining how it can use more creatively and effectively the data it generates and collects to help it make predictions about potential future food safety risks. By sharing this information and working better with partners, the agency is also aiming to develop a more innovative and proactive approach to food safety management.

A Few Definitions
    So what do we mean when we talk about emerging risks? Working with stakeholders, the FSA has come up with the following definitions:

•    Emerging risk: a new and unexpected hazard or threat

•    Re-emerging risk: a known hazard or threat that is increasing in frequency of occurrence or severity

The words hazard and threat have been chosen deliberately to encompass both unintentional risks and the deliberate acts of fraudsters or other individuals with malicious intent.

Time frame is also important. The risks we are looking at are those that emerge in the short to medium term.

Often, the terms emerging risks and horizon scanning are used interchangeably, but within the FSA system, we have made a clear distinction between them. Horizon scanning is a specific technique used to help organizations identify risk in the medium to long term to identify evidence gaps. These longer-term risks will ultimately be impacted by what happens now, and therefore, early identification of these longer-term risks will aid risk mitigation.

The FSA is working with a UK university[1] to obtain data about that longer-term view and to inform its emerging risks program, most particularly on global chain analyses, which will be covered later.

Emerging Risks Detection System
Putting together a system to detect emerging risks isn’t rocket science. It needs a few key elements, however. These are the following:

A protocol. That is, a consistent framework within which you operate that is flexible, adaptable and scalable.

An intelligence strategy. The strategy identifies where and how to look for information. The FSA has chosen to base its strategy on the National Intelligence Model (NIM), which provides a structured approach to the assessment of the intelligence that is gathered. NIM is the business model for police enforcement in the UK and is the policy of the Association of Chief Police Officers (ACPO). NIM has a statutory basis under the ACPO (2005) Code of Practice on NIM.[2] This was issued by the UK home secretary under the Police Reform Act of 2002.

All police forces in England and Wales are required to implement NIM, and it provides an intelligence-led proactive approach to enforcement. Essentially, by using intelligence wisely, police forces can identify where they can best deploy their resources to maximum effect.

Figure 1 shows a schematic of the FSA’s emerging risks model. The model cycles information from data collection through evaluation to action but includes a feedback loop from evaluation to data collection so that evaluation can be refined by additional data.

Specialized IT software helps the FSA use data more effectively to predict future risk. This could be by identifying signals pertaining to unusual events that may be the precursors of future food safety issues.

Where data suggest the need for action, the FSA will use its existing response or information-sharing systems to ensure that consumers are protected appropriately.

Reliable Data Sources
The FSA has identified several thousand potential data sources that it could access. These provide the key intelligence but must be chosen with care and their limitations and biases understood. Are they relevant, valid and credible? The FSA has therefore decided to derive the core intelligence it uses to help identify future risks from the statistical analysis of the agency’s historical incident and food fraud data, as well as from reports from the European Commission’s Rapid Alert System for Food and Feed. These data are well characterized, reliable and come from trusted sources.

In addition, the FSA has a large number of other sources of information available for scrutiny, including research and surveillance data, media reports and expert opinions. Outputs from horizon-scanning activities, mentioned previously, provide further intelligence that helps the FSA better understand the key drivers that might affect food safety in the future.

Skilled Human Interventions
This is absolutely vital. No web crawler, algorithm or software gadget can replace know-how, common sense and an inquiring mind. This is key to making sure that information is treated appropriately and proportionately.

The involvement of stakeholders in the detection and assessment of future safety risks is crucial, so the FSA has established a large number of formal and informal networks at local, national and international levels. These networks provide opportunities for intelligence gathering and sharing through collaboration with industry, enforcement authorities and international partners such as the European Food Safety Authority (EFSA), based in Parma, Italy.

Since 2010, EFSA has piloted a program for the identification, assessment and communication of emerging risks.[3] An EFSA scientific colloquium in 2010[4] considered the challenges and opportunities associated with the detection of emerging risks, and the authority’s EMRISK Unit has recently published a report on the lessons learned from its pilot program. The FSA’s emerging risks program is complementary to EFSA’s.

Key Considerations
As mentioned above, emerging risks are assessed in the short to medium term but inform and are informed by the medium to longer term horizon-scanning processes. What this means is that the detection system looks at the following:

What is typical? The FSA has drawn on incident data to provide core intelligence. Working with statisticians, a base line was derived for each of the incident categories with which the FSA deals, allowing trends to be predicted. The data have been summarized and an example is shown in Figure 2.

What is exceptional? Having derived the expected patterns, unusual or abnormal events that may indicate an emerging issue can be sought. The chart in Figure 2 shows that the expected pattern of incidents is cyclical. Two specific incidents exceeded the expected distribution: a norovirus outbreak in 2010 and a number of consignments of paan (betel) leaves contaminated with Salmonella spp. These unexpected events are reviewed by an internal FSA evaluation board to determine what must be done with the information.

More often than not, there is insufficient information to make an immediate assessment and so further data sources are checked to inform that initial picture.

Global Chain Analysis
Key challenges remain around understanding vulnerabilities in global food chains and their potential impact on future new and re-emerging risks. Work is underway both in the UK and internationally to develop methodologies to aid such understanding.

The FSA is employing three main approaches to analyzing global food chain vulnerabilities and categorizes these vulnerabilities as hazard, threat or value points (Figure 3).

Hazard points are those parts of the supply chain where accidental or unintentional issues arise. These can result from process failures, adventitious contamination or other factors. Such issues are part of any supply chain, and their impact is assessed and mitigation measures implemented using structural analysis techniques such as Hazard Analysis and Critical Control Points.

Threat points are those parts of the chain vulnerable to deliberate or malicious attack. Such vulnerabilities can be assessed using a range of available techniques. In the UK, a Publicly Available Specification, PAS 96 “Defending Food and Drink,” was developed, offering guidance to food businesses as to how to assess and mitigate their vulnerability to malicious attack.

Value points are those parts of the chain vulnerable to economically motivated adulteration. Both regulators and industry are concerned at the apparent increase in frequency of such events. The 2013 European horse meat scandal is one of a long line of products adulterated or subject to substitution. The challenge for industry and regulators is how to implement surveillance and control measures to address what is, in fact, criminal activity.

The FSA is using value chain analysis to help it identify vulnerabilities. Essentially, it consists of a set of simple questions:

•    Where in the supply chain is there opportunity to make money by deviating from best practices?

•    What gives food monetary value and how can this value be enhanced?

How can the profitability of a product be enhanced? Various techniques have been used in the past, including adulteration to improve a food’s perceived quality attributes, substitution of a product or ingredient with a lower-value alternative or extension of product with lower-value alternatives.

The techniques the FSA has developed have helped it identify links in cases involving the fraudulent supply of products and have allowed it to take action to identify and remove those products from sale.

The data are used to study global food supply chains to enable mapping of the potential risks associated with each stage in the chains, from farm to fork, including the individual processes used to manufacture food products. By understanding the features and attributes of each chain and what can occur at each stage, the FSA can look for any unexpected changes that might provide early warning of new or re-emerging risks.

Combining the techniques across supply chains allows structural vulnerabilities to be mapped and assessment made of mitigation measures already in place or that could be introduced by industry or regulators. FSA has begun to map out supply chains and assess those vulnerabilities, but much remains to be done.
Root Cause Analysis

Analysis of food safety incidents has shown that certain types of incident seem to recur regularly. This suggests that the corrective actions taken are not always sustainable or appropriate and are, therefore, not providing long-term preventive solutions. To address this, the FSA has been working to develop methods for identifying the root causes of food safety incidents.

Root cause analysis is based on a very simple premise—asking “Why?” The FSA uses this technique to look at the events leading up to a food safety incident. It can then identify the chain of events and specific step or series of steps within that chain where action could be taken to prevent similar food safety incidents in the future. The FSA believes this technique will also be valuable to food business operators and enforcement colleagues and intends to roll out a training package.

If a number of root causes of similar issues are aggregated and considered together, emerging themes relating to underlying issues might then be uncovered. For example, the analysis of incidents involving undeclared sulfites that could cause problems for those allergic to them has shown that 90 percent of the incidents were caused by poor or insufficient training of food business operators and their staff, resulting in lack of knowledge of either the product or the appropriate labeling requirements.

In Summary
The FSA has developed and is using a range of techniques to identify emerging risks. Figure 4 provides a schematic showing how the various techniques can be combined into global chain analyses. In effect, starting from a map of the supply chain, hazard, threat and value points can be identified. Therefore, the current world of food safety, complete with its flaws, can be characterized and gaps in our knowledge identified.

Horizon scanning and future studies paint a picture of a world not yet realized and identify potential future impacts on the supply chains’ inherent vulnerabilities. Global chain analysis combines the short- and longer-term pictures, highlighting opportunities for mitigating actions or the identification of new risks.

This complementary approach gives the FSA the potential to make more intelligence-led decisions relating to the management of current food safety issues and to enable it to predict and respond earlier to future food safety issues.   

Terry Donohoe is the head of strategy and policy, chemical safety division, at the UK Food Standards Agency.

References
1. Cranfield University’s Centre for Environmental Risks and Futures.
2. whereismydata.files.wordpress.com/2009/01/national-intelligence-model-20051.pdf.
3. www.efsa.europa.eu/en/efsajournal/pub/s1015.htm.
4. www.efsa.europa.eu/en/supporting/pub/114e.htm.