regulatory agencies of other countries receive documents with less CBI than does APHIS. A previous NRC report (NRC 2000c) raised similar concerns about CBI.
In the committee’s review of public participation in the review process it was apparent that the number of comments on Federal Register notices has declined almost to zero. Committee discussions with representatives of public interest groups indicate that this decline in responses to APHIS-BBEP Federal Register notices is at least in part due to a perception that APHIS is only superficially responsive to comments. The committee finds that there is a need for APHIS to actively involve more groups of interested and affected parties in the risk analysis process while maintaining a scientific basis for decisions. As indicated above, there is a tension between use of the risk analysis process for decision support and maintenance of authority. APHIS could benefit from more attention to maintaining a balance between these two roles of its risk analyses.
In examining its day-to-day operations the committee finds that APHIS-BBEP is understaffed and questions the match between the scientific areas of staff training and their responsibilities. The committee specifically noted understaffing in the area of ecology. The committee recommends that APHIS improve the balance between the scientific areas of staff training and job responsibilities of the unit by increasing staff and making appropriate hires. In making this recommendation the committee is aware that APHIS needs help in making its hiring practices and salary ranges more flexible. Because of the large number of applications for field testing, more resources are needed in order to maintain a suitable number of well-trained APHIS officers for field inspection.
As pointed out above, the committee commends APHIS for maintaining an environment in which the decision-making process can be adjusted based on knowledge gained from past risk assessments and regulatory decisions. The committee thinks APHIS would profit from formalizing its learning process. A fault-tree analysis is one approach to such a formalized learning process.
Environmental testing of transgenic plants prior to commercialization can be effective in screening plants for many types of risks, but the committee finds there are several compelling arguments for validation-testing and ecological monitoring after commercialization of these plants.
Because APHIS has considered deregulation absolute, it does not currently conduct postcommercialization monitoring unless commercializa-