A second major domestic disease surveillance effort is the National Nosocomial Infections Surveillance System (NNISS), which currently gathers data from approximately 120 sentinel hospitals and is operated by CDC's Hospital Infection Program. Although nosocomial diseases constitute an important share of the burden of disease in this country, the system has several major limitations. For example, it cannot correct for differences among participating hospitals in diagnostic testing, intensity of surveillance, and provisions for postdischarge surveillance. The requirement that NNISS member hospitals have at least 100 beds, and the relatively small sample of hospitals that are included in the system, are also potential sources of bias. Current plans call for improvements in the dissemination of NNISS data, the inclusion of a surveillance component for immunosuppressed patients, and the addition of more sentinel hospitals, among other efforts.

The committee recommends that additional resources be allocated to the Centers for Disease Control to enhance the National Nosocomial Infections Surveillance System (NNISS) in the following ways:

1. Include data on antiviral drug resistance.

2. Include information on morbidity and mortality from nosocomial infections.

3. Increase the number of NNISS member hospitals.

4. Strive to make NNISS member hospitals more representative of all U.S. hospitals.

5. Evaluate the sensitivity and specificity of nosocomial infection surveillance activities performed in NNISS member hospitals.

6. Determine the reliability of antimicrobial susceptibility testing performed in NNISS member hospitals.

Considerable effort and resources are being expended on the various surveillance activities in which U.S. government agencies and the private sector participate. Much of this information, however, is not readily accessible. There is currently no single database from which a physician, researcher, health care worker, public health official, or other interested party can obtain information on disease incidence, antibiotic drug resistance, drug and vaccine availability, or other topics that might be relevant to infectious disease surveillance, prevention, treatment, and control.

The committee recommends that the U.S. Public Health Service develop a comprehensive, computerized infectious disease database. Such a database might consolidate information from more specialized sources, such as the National Nosocomial Infections Surveillance System (NNISS), the National Electronic Telecommunications System for Surveillance (NETSS), and the influenza surveillance system; it could also include