Communication vital among antimicrobial surveillance programs

Data sharing and publishing findings can facilitate clinical practice.

Issue: August 2008

BOSTON — Data sharing is of the utmost importance to ensure the success of antimicrobial surveillance programs, according to experts who spoke at the 108th General Meeting of the American Society for Microbiology, held here.

The benefits of conducting surveillance research and publishing findings can be seen from the laboratory to the clinic, according to the experts.

“If we conduct surveillance properly, devise mechanisms to share our information better and coordinate surveillance programs on an international and national basis, we can achieve reductions in some of the infections we are dealing with today,” said Daryl Hoban, PhD, clinical microbiologist, diagnostic services of Manitoba and professor in the department of medicine, microbiology & infectious diseases at the University of Manitoba Health Sciences Center, in Winnipeg, Canada.

North American programs

Hoban spoke about the Canadian Antimicrobial Resistance Alliance, the Canadian Bacterial Surveillance Network, the Etude Quebecoise des pathogènes Respiratoires Project in Quebec and a study of antimicrobial resistance in isolates throughout hospitals in Canada known as the CANWARD study. He discussed the influence of these large surveillance operations in the vast country.

“The volume of antimicrobial prescriptions in Canada has dropped dramatically, particularly in pediatrics,” Hoban said. “We have approached a 50% reduction in pediatric antimicrobial prescriptions and between 10% and 25% in adults in the past 10 years. I think surveillance has helped that. It has allowed us to develop better treatment guidelines, allowed us to educate physicians on a national basis as to how to use various agents in a more appropriate fashion. That would not have been possible without surveillance.”

Hoban also addressed the potential drawbacks of so many programs. “Canada is a country with 10 provinces, and each has its own medical system intact inside,” he said. “Formulary decisions are made on a provincial basis and not on a national basis. Each province has the right or authority to have drugs listed under formulary for patients, while other provinces may not.”

But even this roadblock can be helpful, Hoban said. “You can see changes in jurisdictions where surveillance is taking place and try to relate that back to overall drug use.”

There are multiple active surveillance programs in the United States, as well. Those programs also experience difficulties.

David G. White, PhD, acting director, Office of Research Center for Veterinary Medicine for the FDA spoke about the National Antimicrobial Resistance Monitoring System (NARMS). The system is a collaboration between the FDA’s Center for Veterinary Medicine, the USDA and the CDC. The objectives of the program include providing data on antimicrobial drug susceptibility of foodborne bacteria to veterinarians and physicians.

“We are putting data from all three arms together side by side to look at trends in resistance,” said White. “But it is complicated because not all trends from the three arms correlate with each other. Also, they used to publish all of their data reports separately. The problem was that if you wanted to find specific data, you had to go through three reports that were several hundred pages each. So we started looking at all three arms and finding the most interesting, related data which is now published in an annual executive report.”

Compatibility and volume

Michael Feldgarden, PhD, research director of the Reservoirs of Antibiotic Resistance project, expanded on the communication and compatibility problems addressed by Hoban and White.

“Different researchers use different methods to approach the same problems,” said Feldgarden. “We need to standardize methods across studies. When you combine data across data sets, you gain power in terms of number of isolates, but because the data is amalgamated, you lose power in terms of resolution. What we have learned is that if you do not have common portable methods of surveillance, you find that even when study A and study B found the same result, often what happens is that study A and study B were conducted using different methods, so it becomes difficult to draw any strong conclusions. Structure and sample design should be coordinated whenever possible.”

The other problem most of the speakers addressed was the sheer volume of information coming in from different studies.

Ronald N. Jones, MD, president and CEO of JMI Laboratories spoke about the amount of information in the Sentry surveillance program. “We have published more than 200 peer-reviewed manuscripts over the last 10 years, and we consider ourselves failures,” he said. “The reason for that is the amount of information in our databases has only been tapped about 10%. We have much more information that should be getting into the public domain.”

The natural consequence of such volume is lag time between when the information comes in and when it is published. White and the NARMS program are addressing this problem as directly as possible. “We are working to improve the timeliness of our reports,” Jones said.

While the overwhelming amount of research can make it difficult for clinicians to sort out pertinent or useful information, most researchers agree that the more data available, the better. Scott K. Fridkin, MD, MPH of the CDC, spoke on the growth of the National Healthcare Safety Network.

“In the beginning, the network was entirely voluntary,” Fridkin said. “But mandates began to become implemented in different states, and the number of facilities reporting data has increased. We started off with about 500 facilities reporting data. It is up above 1,500 today, and we expect over 3,000 facilities to be reporting by the end of next year. This has really had dramatic effects on participation in the network. Historically, reporting data on resistant infections has been completely voluntary, but that is changing.”

This kind of mandatory participation has brought antimicrobial resistance and surveillance data into sharper focus. Physicians prescribing drugs are much more aware of trends in resistance than they have ever been, and this translates into patient outcomes.

Clinical implications

Though much of the material in the presentations addressed strictly research and epidemiology issues, the researchers said the overall objective is better results in patients.

“These programs have been able to detect resistance to penicillin, macrolides, and fluoroquinolones,” Hoban said. “We have seen a pattern in which resistance increases, but then it stabilizes and ultimately tapers off due to surveillance. If you look at tables of minimum inhibitory concentrations over time, you can detect trends and shifts in resistance levels.” This indicates that physicians are paying attention to the research, he said.

Many of the researchers stressed that this pattern will continue if clinicians continue to be aware of surveillance data. “Resistance does not start or end at the hospital door,” said Feldgarden.

Though many of the problems with surveillance programs still need to be addressed, the researchers agreed that the trends are moving in the right direction. Even the small percentage of data that has seen the light of day has been helpful.

“All of this information has helped physicians make better antimicrobial choices,” said White. – by Rob Volansky

For more information:

Hoban D. Canadian surveillance programs.

White D. National Antimicrobial Resistance Monitoring System (NARMS).

Jones R. Sentry antimicrobial surveillance program.

Fridkin S. Overview of antimicrobial resistance data associated with the National Healthcare Safety Network (NHSN) of CDC.

Feldgarden M. APUA Global projects on antimicrobial resistance.

All presented at: The 108th General Meeting of the American Society for Microbiology; June 1-5, 2008; Boston.