Assessment of the national antimicrobial resistance monitoring system (NARMS) and its value in critical decision-making.

Abstract This paper reviews the United States National Antimicrobial Resistance Monitoring System (NARMS). The data from this system have been used as the core basis of a recent proposal to stop the use of fluoroquinolones in poultry production, in an effort to reduce or remove a perceived threat to human health due to alleged increasing fluoro-quinolone resistance among human Campylobacter isolates. An increase in resistance has the potential to render fluoroquinolone therapy of human gastrointestinal infections less effective. This review finds no evidence in the NARMS data to suggest that a rise in resistant Campylobacter isolates has occurred since fluoroquinolones were introduced for use in poultry. In addition, this review identifies a number of shortcomings in the NARMS program. These include lack of an overall sampling design, biases in the collection of isolates from animal samples, non-compliance of state departments of public health with NARMS protocols, lack of such basic data as genotype information on the organisms isolated, and, for animal samples, lack of measurements of bacterial load. Taken together, these problems indicate that the present NARMS data very likely overestimate resistance levels, and, at best, do not provide reliable information to quantify the extent and temporal trends of antimicrobial resistance in Campylobacter and other enteric organisms from human and animal populations. Ways to improve the quantitative rigor of the NARMS program and identify other data such as ‘meta' data for human and animal samples, and data on the magnitude of other potential reservoirs of infection, including healthy humans, are suggested. Implementing these proposals would greatly enhance the value of the NARMS program as a tool for quantitative decision-making and aid in improving the quality of our food supply and human health care.