Objective: Describe the incidence of non-infectious adverse events associated with invasive procedures and care in intensive care pediatric units.
Methods: A prospective, multi-center 3 ICU's pediatric tertiary care hospitals (28 beds) of the State of Minas Gerais, Brazil, September 2006 to June 2009. Defined by the compulsory reporting by professionals, data validation by the senior management of ICU and sending monthly e-on central administration in 2006, becoming voluntary reporting in 2009. The classification and definitions have been proposed by Pedrosa and Couto (2003), classifying the event for the affected systems as the device / procedure associated, skin and soft tissues and others. Descriptive statistical techniques and graphical inspection of the temporal evolution of the data for the choice of mathematical models explaining was used. Table 1 – Pooled adverse events rates per pediatric ICU, january 2007 through june 2009, Fhemig, Brazil. 2006 Conclusions: The system of notification of adverse events temporally proposed to assess the impact on quality of care capabilities and protocols. The overall increase in rates suggests that removing the incentive of notification does not seem to be affecting the institutional culture of surveillance of adverse events noninfectious established.
Results: There was a significant exponential increase in the incidence of reported events per 100 admissions from 2006 to 2009 (p = 0.039). The incidence density approached the quadratic model of progression (p = 0.09). The frequencies and rates per ICU are presented in table 1. 2007 2008 2009 Global Fhemig R² p AE (n) 99 233 199 227 758 Patient-days 3150 10533 8411 5292 27386 Admissions 274 508 371 236 1389 Global rate per 1000 patient-days 31,4 22,1 23,7 42,9 27,7 97,5 0,09 Per 100 admissions (%) 36,1 45,9 54 96 54,6% 88,4 0,04 Rates per ICU (per 1000 patient-days) Pediatric trauma 28,4 5,5 10,6 31,1 Pediatric 1 15,4 15,2 14,2 27,8 Pediatric 2 69 46,5 51,7 77,7