FDA PHYSICALLY EXAMINES IMPORTS ONLY ABOUT 9% OF THE TIME
FDA PHYSICALLY EXAMINES IMPORTS ONLY ABOUT 9% OF THE TIME, compared with 20% 10 years ago, the General Accounting Office said in a recent study. In an April 28 report entitled "Opportunities to Improve FDA's Inspection Program," GAO says that the number of physical inspections by FDA of imported entries dropped from 20% in the mid-1970's to 9% by 1987. Since "physical inspections of imports are the primary means of determining product quality," GAO states, "the reduction in the proportion of imported products being inspected is of serious concern to the agency." FDA attributes the fall-off to the fact that imports have surged and staffing has not kept pace. Between the mid-1970's and 1987, the annual number of imported entries subject to FDA inspection tripled from 500,000 to 1.5 mil., according to the report. The agency has achieved the 9% level "only by reprogramming resources from other areas." Of the total number of imported products falling under FDA authority, approximately 84%, or $20 bil. worth, are food items. The rest of the entries consist mostly of drugs and medical devices. GAO staff studied "how FDA staff responsible for inspecting imported products spent their time and identified areas where the efficiency of inspection activities could be improved." The study consisted of evaluating questionnaires distributed to all 226 staff at FDA's 20 district offices during the week of Feb. 29, 1988. GAO also interviewed Customs officials about their inspection procedures and practices. The questionnaires asked FDA district office staff involved in import inspection tasks to record the time they spent on 17 tasks grouped into five categories: paperwork processing, physical inspections, travel, meetings, and supervisory and clerical tasks. The report found that, on average, FDA staff spent 38% of their import-inspection time in paperwork processing; 13% in travel to and from inspection sites; 22% in physical inspections; 14% in clerical support and supervision tasks; and 13% in meetings. Paperwork processing, the most time-consuming category, includes the manual review of paperwork along with FDA notices, alerts and other information on products and importers. The time spent on it varied widely among respondents, from 20% to 57%. The paperwork examination determines whether FDA inspectors decide to release the product for marketing or to perform a physical inspection. Rep. Dingell (D-Mich.) recently suggested to FDA that one means of reducing paperwork would be to share the U.S. Customs Service's automated information processing capabilities ("The Pink Sheet" May 22, p. 7). However, FDA has rejected the offer because Customs' data system was not developed to take into account the variables FDA must consider, an FDA spokesperson said. The Customs system is focused instead on the need to handle invoices and to collect duty on imports. Moreover, according to the spokesperson, because Customs' data system uses codes unfamiliar to FDA, the agency inspection staff would lose valuable inspection time decoding the data. FDA told GAO that it agreed the agency should implement its own automated paperwork review system. That process is already underway, the agency says. In 1987, FDA established the Import Alert Retrieval System, which transmits abbreviated alert notices to district offices, and the agency is currently implementing the design specifications for an automated entry review data system, the Import Support and Information System (ISIS), which would make import data available in a real-time situation. A test pilot of the basic data entry component of the system is tentatively scheduled for March 1990, according to an FDA spokesperson. A planned component of the ISIS system, which would not be included in the test pilot, is a data interface between FDA and Customs. FDAers estimate that an electronic linkup between Customs and FDA is still several years away.
Sign in to continue reading.
New to Pink Sheet?
Start a free trial today!
Register for our free email digests: