The 2007 Belgian national prevalence survey for hospital-acquired infections.

Despite ongoing targeted surveillance efforts, no overall in-hospital prevalence data for hospital-acquired infections (HAIs) have been published for Belgium. Sixty-three Belgian acute hospitals participated in a point-prevalence study among either all patients admitted in their institution or 50% of the patients in each ward. HAIs were registered bed-site at a single day per ward during the period October-November 2007. The diagnosis was made according to the Centers for Disease Control and Prevention (CDC) criteria implemented in a custom-made rule-based software expert system available on a portable computer. The total number of patients surveyed nationally was 17 343, from 543 distinct hospital wards. The overall prevalence of HAIs was 7.1% (95% confidence interval: 6.7-7.4%); 6.2% (5.9-6.5%) of the patients suffered from at least one HAI. Prevalence of HAIs on adult intensive care was 31.3%. The major proportion of HAIs was observed among patients admitted on non-intensive care unit (non-ICU) wards, mainly on the wards of internal medicine, surgery, geriatrics and rehabilitation. Urinary tract infections were the most common type of HAI at geriatric and rehabilitation wards. This study demonstrates that the use of a portable computer system with a designated expert system for diagnosing HAIs according to the CDC criteria in a large point prevalence study is feasible and may reduce the within-subject variation. In Belgium, the prevalence of HAIs in acute hospitals thus identified is similar to that of neighbouring countries. As more than 80% of all HAIs occur on non-ICU wards, preventive efforts need to extend beyond the ICU.