Surveillance of hospital infections: at the bedside or at the bacteriological laboratory?

A prevalence study concerning wound infections in the surgical service at a university hospital was made during a 7-month period. One ward was visited every week during the wound dressing rounds and samples for bacteriologic culture were taken from clinical infections. The prevalence of infection varied between 0 and 21% (median value 8.2%) for surgical wounds and between 0 and 47% (median value 5.3%) for other wounds. Staphylococcus aureus made up 16% of the bacteria isolated from surgical wounds and the Enterobacteriacae 32.6%. For other wounds Staph. aureus and the Enterobacteriacae made up 22.6% each. A comparison of the number of cultures taken by investigators and that taken by the ward showed that the culturing frequency by the ward staff was so high that surveillance of infection by review of bacteriological culture records only, would give an almost equivalent degree of information concerning infection rates. This system is less time consuming and also has the advantage of giving epidemiological information. The drawback with the system is that little or no patient data are given on the requisitions from the ward. This makes it impossible to know from what kind of infections samples are taken.