APPLICATION OF EXPONENTIAL SMOOTHING FOR NOSOCOMIAL INFECTION SURVEILLANCE

Detection of outbreaks of infection or increases in bacterial resistan ce to antimicrobial agents is an essential component of hospital infec tion control surveillance, The authors applied the method of exponenti al smoothing to microbiology data from 1987-1992 to investigate a susp ected outbreak of gentamicin resistance among Pseudomonas aeruginosa b acteria at the Department of Veterans Affairs Medical Center, San Fran cisco, California, in 1991-1992, The years 1987-1990 were used to deve lop the baseline for the forecast model, Application of the model indi cated that two observed prominent peaks in the annual cumulative incid ence of gentamicin-resistant P. aeruginosa were within the upper bound s of their respective 95% confidence intervals as estimated by the for ecast model-i.e., that no epidemic was in progress, This prediction wa s supported by investigations by the hospital's infection control team which indicated that the apparent increases were due to readmission o f patients previously known to harbor these organisms, In contrast, ap plication of a typically employed method that ignores the time series data structure indicated that there were 6 months in which incidence r ates exceeded the upper bounds of their respective 95% confidence inte rvals, thereby erroneously suggesting that an epidemic was in progress , Recursive algorithms and some simplifying assumptions that do not af fect the validity of inferences make the application of this method pr actical for nosocomial infection control programs.