Enterococci
% resistant to vancomycin
Basic statistics
(January-June 2001)

TABLE 1. Clusters of the hospitals based on resistance of enterococci to vancomycin

A. ALL WARDS 
Cluster
Mean % Resistance ± SD
Hospital
"Low Resistance"
2% ± 2%
GR001, GR004, GR005, GR007, GR009, GR012, GR013, GR014, GR015, GR018, GR024, GR027, GR028, GR032, GR033, GR035, GR038, GR039, GR040, GR042, GR046
"High Resistance"
26% ± 11%
GR026, GR044

B. ICU          
Cluster
Mean % Resistance ± SD
Hospital
"Low Resistance"
4% ± 6%
GR001, GR005, GR012, GR015, GR024, GR027, GR033, GR035, GR038, GR039, GR040, GR046
"High Resistance"
38% ± 6%
GR007, GR013, GR014

Figure 1. Multi resistance by hospital.
Dot lines represents 95% CI of the overall mean
resistance to vancomycin

A. All WARDS

B. ICU


Specific comments. The applied methodology of clustering the hospital wards and ICUs using the calculated multi-resistance is similar to the analysis of variance (ANOVA) "in reverse." In particular, we started with k random clusters, and then move cases (i.e. hospitals) between those clusters with the goal to:

  1. minimize variability within clusters
  2. and maximize variability between clusters.
This is analogous to "ANOVA in reverse" in the sense that the significance test in ANOVA evaluates the between group variability against the within-group variability when computing the significance test for the hypothesis that the means in the groups are different from each other. However it should be mentioned that the small number of the available information, in several cases, increases the uncertainty of our analysis and limits our findings.