Identifying outliers of antibiotic usage in prevalence studies on nosocomial infections

OBJECTIVE: To investigate whether the correlation between patients' antibio tic treatment (yes/no) and patients' infections (yes/no) in each hospital d epartment, described by Pearson's correlation coefficient (rho) for binary data as a measure for adequate use of antibiotics, is an appropriate qualit y indicator. DESIGN: Comparison of the results of repeated prevalence studies in differe nt hospitals with the data of a national prevalence study, comparing the ho spital (rho) and reference (rho(NIDEP)[Nosokomiale Infektionen in Deutschla nd: Erfassung und Pravention]) correlation coefficients for "use of antibio tics/presence of infections." SETTING: The data of 5,377 surgical patients were separated from the total data of a national prevalence study in 72 representative hospitals to creat e a reference correlation coefficient (rho(NIDEP)) with a reference range. Nine additional prevalence studies, involving a total of 4,984 patients, we re repeatedly performed in the surgical departments of 8 other hospitals du ring a 12-month period, whereby the correlation coefficients rho(n) for eve ry prevalence investigation were determined. RESULTS: In the national prevalence study, 15.3% of the surgical patients r eceived antibiotics on the study day. Surgical patients had a 3.8% prevalen ce of nosocomial infections and a 7.0% prevalence of community-acquired inf ections. Pearson's correlation coefficient rho(NIDEP) for correlation betwe en patients' binary data use of antibiotics and presence of infection was 0 .62. To compare the correlation coefficient of each department with the app ropriate reference range, the coefficients of the single departments were p lotted against the number of patients; in these plots, three lines indicate d the value rho(NIDEP) and the upper and lower reference ranges, depending on the number of patients. Seven of eight surgical departments investigated during the repeated prevalence studies were found to be within the referen ce range, near the reference value, in the majority of prevalence studies; only one of the departments was identified as an outlier as regards antibio tic use. CONCLUSION: The correlation between patients' antibiotic treatment (yes/no) and patients' infections (yes/no) in hospitals or departments, as describe d by Pearson's correlation coefficient rho for binary data with a definitiv e reference range depending on the number of patients, is useful for qualit y management in identifying the overall necessity for evaluating the, indic ations for antibiotic use in one's own hospital.