Impact of structural reform on nosocomial infection rates of three adult ICUs

Background: Device - related nosocomial infection (NI) is a substantial concern in adult intensive care units (ICUs) and, due to its internal and external comparativeness, remains the best alternative when evaluating the impact of NI preventive measures.

Objective: To elucidate the impact of structural renovation of three adult ICUs on incidence of NI.

Methods: An analytical-observational study based on the data obtained from the ICU-HELICS Nosocomial Infection Surveillance System was conducted at the Miguel Servet University Hospital. In January 10^th, 2009, three new ICUs were opened offering several structural advantages for patient outcomes such as individual rooms with individual hand washing sinks and High Efficiency Particulate Air filters.  
All patients admitted into the ICUs for more than 48 hours between January 1^st, 2008 and June 2009 were included in the study. The bivariate and multivariate logistic regression models were performed using SPSS software to evaluate the impact of the new facilities on device - associated NI in the three adult ICUs adjusting by different risk factors.

Results: The total number of patients admitted in the ICUs during the study period was 1415 and were divided into two groups: 790 patients admitted during the period previous to the structural reform (group A) and the other 425 during the period post reform (group B). Infection rates and device utilization ratios were: central line-associated primary bloodstream infections per 1000 central line days/Central line utilization ratio: group A 1.210/0.957 and group B 1.926/0.993; Ventilator-associated pneumonias per 1000 ventilator days/Ventilator utilization ratio: group A 27.499/0.398 and group B 13.746/0.495; and catheter-associated urinary tract infections (UTI)/ urinary catheter utilization ratios: group A 1.998/0.902 and group B 3.735/0.845.
In the final multivariate model considering Ventilator-associated pneumonia as the outcome, the variables period of study (groups A and B), days of ventilator use and the age at admission were considered risk factors. 0,392 (95% CI 0.234-0.657) was the adjusted odds ratio obtained for group B versus group A.

Conclusions: After structural changes in ICUs, ventilator-associated pneumonia rates decreased whereas BSI and UTI rates increased but not significantly.
Probably, the reform had also a positive impact on the staff outcomes that contributes in the decreasing of the device – related NI in the three adult ICUs.