Over 20 years ago, Philip Brachman advised us that “… the surveillance of all institutionally associated infections is important in order to minimize the risk of infection to all patients entering the institution and to members of the community.” Seven years later, in 1970, other staff members at the Centers for Disease Control (CDC) offered us more specific surveillance objectives:
A. To determine the frequency and kinds of endemic nosocomial infections, in order to identify deviations from the baseline so that infection control personnel can:
1. Determine where studies are needed.
2. Ascertain where control measures (long-term and emergency) need to be established and how effective new control measures are.
3. Establish policy.
B. To provide the patient and personnel (and in some instances the community) with all possible protection from infections of nosocomial origin.
C. To meet the requirements of the Joint Commission on Accreditation and the medical-legal guidelines of “accepted standards of patient care.”
D. To provide the medical and nursing staff with meaningful data on the level of nosocomial infection in their work areas.
If each single nosocomial infection represented sufficient deviation from the baseline occurrence, then analysis of surveillance data would be quite straightforward. However, that is not the case. A 1976 project report for the National Center for Health Statistics identified various “sentinel health events” whose occurrence should trigger “… scientific search for remediable underlying causes.”