Home  »  IRAC  »  Events  »  Symposium 2004

Data Collection and Utilization in Risk Assessment and Management Decisions

SYMPOSIUM

Meeting Presentations and Agenda

Executive Summary

The Interagency Risk Assessment Consortium (RAC)sponsored the Data Collection and Utilization in Risk Assessment and Management Decisions symposium on Tuesday, September 14, 2004 at the College Park Aviation Museum. Approximately 50 Federal Government employees representing researchers, risk assessors, and risk managers attended the half-day symposium.

Objectives of Symposium: Risk assessors are challenged to analyze data in quantitative and qualitative ways, and provide useful information to address risk manager’s problem statements. Risk managers rely on research results and other data that form the underpinnings of a risk assessment when making decisions, and must understand the limitations and strength of the information provided to them. Finally data generators (researchers, field investigators, surveillance programs) are interested in designing their data collection strategies and studies in ways that provide useful and meaningful information to both risk assessors and risk managers.

The objectives of the symposium were:

  1. To provide presentations describing how researchers have designed data collection strategies with the intention of data being relevant for risk assessment and decision making (risk management).
  2. To provide presentations describing how risk assessors and risk managers have utilized available data to support decision making.

Summary of the Presentations: During the first session Dr. Robert Buchanan (Food and Drug Administration) introduced the topic of data utility. Ms. Peg Coleman (Syracuse Research Corporation) provided insight regarding potential improvements for study designs with the result being data that can more easily be incorporated into risk assessments. Dr. Martin Cole ( National Center for Food Safety and Technology) provided an overview of the statistical issues involved in sampling design. After these introductory talks, presentations focused to real-world examples of data collection. Drs. John Luchansky (United States Department of Agriculture) and Yuhuan Chen (National Food Processors Association) shared study design details regarding their past studies and lessons learned.

The second session focused on objective two. Dr. Richard Whiting (Food and Drug Administration) shared experiences from the FDA/USDA Listeriamonocytogenes in ready-to-eat food and the FDA Vibrioparahaemolyticus risk assessments with regards to data availability and usefulness. LTC Chumley (Department of Army, Veterinary Corp) shared information regarding databases and where the data comes from to populate such databases.

Conclusions of Symposium: Astute data collectors and risk assessors are questioning and learning how to best collect data that will be useful in a risk assessment and are simultaneously developing methods to help bridge the gap between data that are collected for other purposes and data needs for a risk assessment. As a scientific community we are struggling with how to define and ensure characteristics of “good and useful data.” Characteristics of “good data” were identified during the symposium as relevancy, representativeness, robustness, generalizability, external and internal validity and defensibility. It was also recognized that “good data” enhances the ability of the risk modeler to estimate variability and uncertainty.

A retrospective analysis of data collection strategies will continue to be useful from a “lessons learned” perspective. These lessons will shape the future efforts of the workgroup as it focuses on the data utility characteristics of representativeness, relevancy, robustness, validity and defensibility.

The symposium agenda, abstracts, and PowerPoint presentations are posted for public viewing here.