Lessons from WHONET: The Interdependence of Surveillance, Sustainable Essential Microbiology and Patient Care
By Thomas F. O'Brien and John M. Stelling, WHONET
WHONET is based in the Department of Medicine and World Health Organization Collaborating Centre for Surveillance of Antimicrobial Resistance, Brigham and Women's Hospital and Harvard Medical School, Boston, MA. We have been working for three decades to develop and distribute software to support the WHO goal of multilevel global surveillance of antimicrobial resistance. Our software and its data conversion utility, WHONET 5.4 and BACLINK 2 respectively, may be downloaded at no cost from a WHO websitei.
These tools enable any microbiology laboratory to put its test results into a database and conduct analyses to support local infection control and antibiotic use. Laboratories can also upload files created by WHONET, to feed into national or other multi-center surveillance networks and to inform drug policyii. Such surveillance programs are now in place in many countries. In supporting these countries we have learned many lessons. Specifically, we have identified some major obstacles to the achievement of global health goals; and in many cases, devised innovations to surmount them.
One lesson is that little can be done by public health agents about any kind of infection without surveillance. As we are seeing now with swine fever, viral influenza's peculiar clinical picture and spread allow for syndromic surveillance. However, to identify or track many other infections requires microbiological analysis, and this is generally undertaken in medical facilities where sick people bring their pathogens. Global surveillance to protect us from new threats – a basic public health task – thus relies on the world's patient care microbiology laboratories. These laboratories exist for and are paid for by patients, but they often lack awareness of their role in the broader public health system and may not be incorporated into it. Across much of the world, patient care laboratories are too few and less functional than what is required to sustain infectious disease surveillance.
A related lesson is that using microbiology for even minimal public health surveillance, and especially for patient care, may require a different model in resource-limited settings. Using the affluent-world model has not worked. For these settings, we need to create sustainable, essential microbiology that would identify and susceptibility test the most deadly pathogens using cheaper methods, a simpler supply list, and a workforce trained to use a web-based operating system overseen by a remote supervisor. Testing by caregivers can then be done more frequently and more cheaply, resulting in enhanced surveillance and better patient care.
We have also learned that improving any of the components of public health surveillance improves the others. For example, when we started our work three decades ago, some concern was expressed about surveillance efforts starting before laboratory test performance was perfected. Over time, however, we have seen that ongoing data analysis coupled with collegial oversight has created a continuous quality improvement processiii. Data use and analysis is the strongest driver of laboratory test improvement, and in turn strengthens local, national, regional and global public health surveillance, as well as patient care.
Finally, we have learned that information technology-based public health tools, such as WHONET, follow Moore's Law. As our program has evolved from computer punch cards to the Internet, its capabilities have likewise expanded at a faster pace. The data we gather is now being extended to all microbiology test results. The system currently supports automated outbreak detection and reporting, and is developing automated phenotype discrimination linked to selective genotyping to enhance epidemiological specificity. These and other measures we are taking will continue to overcome the obstacles thrown at us by ever-adapting pathogens.
The WHONET experience suggests that solid drug resistance data can be collected and analyzed in resource-constrained settings, using core microbiology, if local laboratories are given appropriate support. Strengthening these laboratories is therefore a potentially cost-effective contribution to both treating drug resistant disease and preventing its further spread.
ihttp://www.who.int/drugresistance/whonetsoftware
iiO'Brien, T. F. and J. M. Stelling (1995). "WHONET: an information system for monitoring antimicrobial resistance." Emerg Infect Dis 1(2): 66.
iiiO'Brien, T. F., M. A. Eskildsen, et al. (2001). "Using internet discussion of antimicrobial susceptibility databases for continuous quality improvement of the testing and management of antimicrobial resistance." Clin Infect Dis 33 Suppl 3: S118-23.