U. S. Food and Drug Administration
Center for Food Safety and Applied Nutrition
March 2002

Initiation and Conduct of All 'Major' Risk Assessments within a Risk Analysis Framework

Table of Contents

Part II. A Decision-Based Approach to Identify and Select Risk Assessments

Activity: Identify and Select All 'Major' Risk Assessments

• Concept generation
  
• Problem identification
  
• Data feasibility determination (evaluation and recommendation)
  
• Disposition (selection)

Phase 1: Concept Generation		Phase 2: Problem Identification		Phase 3: Data Feasibility (Evaluation and Recommendation)		Phase 4: Disposition (Selection)
Collect ideas and maintain list of potential risk management questions for which a risk assessment would assist with policy decisions. Develop justification for candidate risk assessment: purpose of assessment scope of problem importance to the Center how results will be used by the Center		The candidate risk assessment and supporting information (justification) are reviewed to determine whether the assessment meets the Center's regulatory needs. Recommended actions: conduct data feasibility study not required for regulatory decision more information needed to make decision		Information is collected and reviewed to determine availability of data needed to answer risk assessment question(s). Recommended actions: conduct quantitative risk assessment conduct a qualitative risk assessment more research needed modify question and conduct alternative assessment		Using the results of the data feasibility determination as an aid, risk assessment(s) to be conducted are selected. Decision based on: technical merit resource availability the Center's priority needs other legitimate factors

When is a Risk Assessment Needed?

A 'major' risk assessment requires a substantial commitment of CFSAN resources. Thus, risk assessment is not appropriate when CFSAN's risk managers do not need this level of sophistication to make a decision.

Circumstances that do not warrant a quantitative risk assessment would include:

• a risk that is well described by definitive data,
  
• a problem that is relatively simple, or
  
• an issue that is not of regulatory concern.

• the data describing a hazard are incomplete,
  
• the exposure system is complex, or
  
• the issue is of high regulatory or stakeholder concern.

• when a chemical fails the safety assessment threshold;
  
• when the current quantitative standard is not met or not justified;
  
• when the current standard is inconsistent with other government policies or guidelines;
  
• when the Center has been petitioned for a regulatory action;
  
• when a No-Observed Adverse Effect Level (NOAEL) cannot be established for a chemical;
  
• when a data gap analysis is needed; and
  
• when the hazard is a serious health issue, emerging pathogen and/or public health concern.

What Type of Risk Assessment is Needed?

The type of risk assessment appropriate for a specific risk management problem depends on the question to be answered and the availability of data. If adequate data are available, a quantitative risk assessment is possible, if fewer data are available qualitative and data gap analysis may be more appropriate.

Safety vs. Risk Assessments. This document is specific to the selection and conduct of risk assessments, not routine safety assessments. A difference is that risk assessment estimates risk and, in some cases, sources of risk and quantitative reductions based on various interventions. A safety assessment, on the other hand, provides a verdict of what is "safe" based on the conventions of the analysis.

Qualitative vs. Quantitative. Risk assessments can be either qualitative or quantitative in their description of the likelihood of adverse health effects, depending on the extent of the knowledge available, the complexity of the problem, and the time available to conduct the assessment. In quantitative assessments, the risk is expressed as a mathematical statement of the chance of illness or death after exposure to a specific hazard, and it represents the cumulative probabilities of certain events happening and the uncertainty associated with those events. Conversely, qualitative risk assessments use verbal descriptors of risk and severity, and often involve the aggregation of expert opinions.

The quantitative risk assessment technique used depends on the available data and on the scope and nature of the questions posed by the risk managers.

Types of Risk Assessments. Four general types of risk assessments likely to be used by the Center include risk ranking, product pathway, risk-risk and geographical. Below is a brief description of these types of assessments.

Risk ranking. Risk ranking assessments compare the relative risk among several hazards or foods. These types of assessment techniques might involve a single pathogen associated with multiple foods, a single food that has multiple pathogens, or multiple pathogens and multiple foods. Risk ranking assessments can help establish regulatory program priorities and identify the critical research needs. The FDA/USDA Listeria monocytogenes risk assessment is an example of a risk ranking assessment.

Product pathway analyses. In product pathway assessments, the factors that influence the risk associated with specific food/hazard pairs are examined. Ideally it starts at the farm and ends with consumption. These types of assessment techniques help identify the key factors that affect exposure including the impact of potential mitigation or intervention strategies on the predicted risk. The FDA Vibrio parahaemolyticus risk assessment is an example of a product pathway analysis.

Risk-risk. In risk-risk assessments, a trade off of one risk for another is considered, i.e., reducing the risk of one hazard increases the risk of another. An example of this would be a determination of the impact on public health by treating drinking water with a chemical (risk to chlorine exposure) vs. the impact of exposure to pathogenic organisms in water not treated.

Geographical. In a geographic risk assessment, the factors that either limit or allow the risk to occur are examined. The risk of introduction of disease agents through food animals or animal products (e.g. intentionally as in bioterrorism or unintentionally) can be examined. For example, the risk of introduction of variant Creutzfeldt-Jacob Disease (vCJD) in humans by the transmission of BSE from cattle through meats and animal product pathways might be examined using a geographical approach.

Who Coordinates the Identification and Selection Process?

CFSAN's Risk Analysis Working Group recommends that the risk assessment coordinating staff within Office of Science (referred to as the coordinating staff in the discussion below) be responsible for coordinating the identification and selection process for all 'major' risk assessments. This approach should be implemented for microbial risk assessments and later expanded to include chemical and other non-microbial hazards.

How Are Candidate Risk Assessments Identified?

• Stakeholders (consumers, public, industry)
  
• Consumer advocacy groups
  
• Public health institutions
  
• Follow-up from a completed risk assessment
  
• Regulatory/Management staff
  
• Researchers
  
• Other government agencies

The coordinating staff will prepare an annual All-Hands Call to be issued electronically to CFSAN staff to request the submission of ideas for risk assessments. The submitted ideas will be shared with the impacted Program Offices for evaluation.

Currently there is no established mechanism for routinely or formally obtaining ideas for risk assessments from the public, other agencies, and shareholders. The CFSAN Risk Analysis Working Group recommends that options for obtaining this input be developed. In the meantime, any ideas that are submitted in an informal manner to the Center should be shared with the impacted Program Office and with the Office of Science's coordinating staff.

Four Phases of the Identification and Selection Process

Below is a detailed description of the four phases of the proposed identification and selection process. Figure II-2 illustrates how the process uses a decision-based approach. During each phase, decisions and recommendations are made and the candidate risk assessments are systematically evaluated.

Phase 1. Concept Generation

Responsible Party. Program Offices with assistance from the coordinating staff.

Activity. Throughout the fiscal year (in time to include on the CFSAN Program Priorities list, if possible), the coordinating staff maintains a list of potential risk assessments.

Information collected includes:

• specific hazards/commodities
  
• risk management question(s)/regulatory needs
  
• risk assessment question(s)

• purpose of the risk assessment
  
• scope of the problem (such as commodities, disease agent, impact on public health, distribution of risk in the population)
  
• importance and need for the Center (including current management practices to control the risk)
  
• how the assessment results would be used by the Center

Objective. During Phase 1, the focus should be on the nature of the risks to be evaluated and not specifically a consideration of the data available for conducting the risk assessment, as that is the objective of Phase 3.

Criteria for Identifying Candidate Risk Assessments.

• characteristics and importance of the hazard of concern
  
• magnitude (presence, prevalence, concentration of hazards) and severity (impact on public health) of the risk
  
• urgency of the situation
  
• populations of concern
  
• other factors associated with specific hazards (such as processing, cooking, cross contamination)

Phase 2. Problem Identification

Responsible Party. CFSAN leadership and management teams

Activity. At a regularly scheduled weekly team meeting, the leadership team reviews the justifications for each candidate risk assessment for the purpose of whether the assessment meets CFSAN's needs. The leadership team makes a recommendation for each candidate assessment and provides the recommendation to the management team.

Example recommendations include:

• The candidate risk assessment should be conducted.
  
• The candidate risk assessment should not be conducted. Either because it is not relevant to CFSAN's mission or it is not needed to make a risk management decision.
  
• The need for this risk assessment could not be determined based on the information available for the candidate risk assessment. Either request the coordinating staff to provide additional information, or if new data are needed, determine the need for research to be conducted before a decision is made.

Objective. Develop justification for conducting risk assessment. The resources needed to conduct the risk assessment should be considered in Phase 2, but a complete assessment of resource needs and available should be addressed in Phase 4.

Criteria. Same as above for Phase 1 plus consideration of risk management/ policy issues.

Phase 3. Data Feasibility Determination (Evaluation and Recommendation)

Responsible Party. Coordinating staff and short-term detailees, as needed.

Activity. Under the direction of the coordinating staff, short term detailees review published and unpublished literature to determine the availability of data needed to conduct a candidate risk assessment that will address the risk management question(s) and answer the corresponding risk assessment question(s). [In most cases, the detailees will be from the program office or group that requested the risk assessment.]

Using the results of the review of data availability, the coordinating staff prepares a recommendation concerning further actions.

Recommended actions may include:

• conduct a quantitative risk assessment
  
• conduct a qualitative risk assessment
  
• modify risk management and/or assessment question(s)
  
• more research needed [Short-term research projects to fill gaps in the information needed to determine feasibility could be commissioned, if necessary.]

Objective. Determine appropriate action, specifically whether the stated risk management questions can be answered through a risk assessment; if data are available; or if more research needs to be conducted.

Criteria for Data Feasibility Evaluation. The evaluation of the feasibility of conducting a risk assessment must consider the following:

• availability of data,
  
• the uncertainty created by lack of data, and
  
• other legitimate factors specific to individual hazards.

Examples of the type of information collected during a feasibility evaluation are provided in Appendix E. Developing a conceptual model (including overall structure, data inputs, and flow of calculations) may be helpful in determining the types of data and information needed.

Phase 4. Disposition (Selection)

Responsible Party. Management team

Activity. Each fiscal year (in time to include on the priority list if possible), select risk assessments to be conducted based on technical merit, data feasibility, resource availability, and other factors deemed appropriate by the Center.

Objective. Using the results of the recommendations from the feasibility determination (Phase 3) as an aid, the management team selects specific risk assessments meriting further action.

Further actions might include:

• request a specific risk assessment
  
• select a hazard for which additional research is needed
  
• determine whether a risk assessment should be reconsidered at a later date (i.e., when additional research is available or changes in priorities permit)
  
• conduct a more comprehensive or follow up risk assessment
  
• other appropriate actions

• technical merit
  
• Center's regulatory needs
  
• timeframe for making a risk management decision
  
• resource needed/available
  
• expertise needed
  
• data availability

Timeline				PHASE 1 Coordinating Staff Maintains list of risk assessment (RA) ideas and assists Program Offices with preparation of justification for the candidate projects Requests leadership team review of candidate RAs justifications.
			January
			February	PHASE 1 Leadership Team Meets to review each candidate RA project justification Requests the coordinating staff to provide additional information, if needed Makes recommendation to management team
			March	PHASE 2 Management Team Reviews Leadership team recommendation(s) Decides action for each candidate RA
				Reconsider at later date Retain on a list of potential RAs	Conduct data feasibility study	Not needed
			April to May	PHASE 3 Coordinating Staff and short-term detailees Conducts data feasibility study Makes recommendation of data feasibility to management team
			June	PHASE 4 Management Team Based on results of data feasibility study, resource availability, and other factors decides whether risk assessment should be conducted.
				Data or resources are inadequate; Retain on list of potential RAs.	Conditional	Data and resources are adequate; RA meets Center's needs
					Modify assessment question(s) Conduct qualitative assessment
						Program Offices Include in annual CFSAN Program Priorities Commit resources needed to conduct RA
In time for listing in program priorities
Figure II-2. A proposed decision-based approach to identify and select 'major' risk assessments