Paola E. Cogo Daniele Poole Daniela Codazzi Corinne Boniotti Anna Capretta Martin Langer Davide Luciani Carlotta Rossi Guido Bertolini Outcome of children admitted to adult intensive care units in Italy between 2003 and 2007 Received: 3 November 2009 Accepted: 27 April 2010 Published online: 20 May 2010 Ó Copyright jointly held by Springer and ESICM 2010 For the GiViTI group Gruppo Italiano per la Valutazione degli Interventi in Terapia Intensiva (Italian Group for the Evaluation of Interventions in Intensive Care Medicine). A complete list of study participants appears in the Appendix. P. E. Cogo ( ) ) Á A. Capretta Department of Pediatrics, Pediatric Intensive Care Unit, University of Padova, Via Giustiniani 3, 35128 Padova, Italy e-mail: cogo@pediatria.unipd.it Tel.: ?39-049-8211477 Fax: ?39-049-8213502 D. Poole Department of Anesthesia and Intensive Care, Intensive Care Unit, San Martino Hospital, Belluno, Italy D. Codazzi Pediatric Intensive Care Unit, Ospedali Riuniti, Bergamo, Italy C. Boniotti Pediatric Intensive Care Unit, Ospedale dei Bambini, Spedali Civili, Brescia, Italy M. Langer Fondazione IRCCS Istituto Nazionale Tumor, Universita ` degli Studi di Milano, Milan, Italy D. Luciani Unit of Clinical Knowledge Engineering, Laboratory of Clinical Epidemiology, Mario Negri Institute for Pharmacological Research, Milan, Italy C. Rossi Á G. Bertolini GiViTI Coordinating Center, Laboratory of Clinical Epidemiology, Mario Negri Institute for Pharmacological Research, Ranica (BG), Italy Abstract Background: Centrali- sation of critically ill children to paediatric intensive care units is supported by a strong rationale, but evidence is not overwhelming. Objective: To compare the outcome of children admitted to adult intensive care units (ICUs) in Italy between 2003 and 2007 with that of children admitted to paediatric intensive care units (PICUs) in Italy between 1994 and 1995. Methods: Prospective, multicenter cohort study and histori- cal controls. Risk of ICU mortality was assessed with the PRISM score in both study and historical control groups. Descriptive statistics, stan- dardized mortality ratios (SMRs) with their 95% confidence intervals, and the calibration plots were reported. Results: A total of 1,265 children admitted to 124 adult ICUs between 2003 and 2007 were compared with an historical control group formed by 1,533 children admitted to 26 PICUs between 1994 and 1995. The PRISM score slightly underestimated hospital deaths for low-risk patients in both groups. The overall SMR was 1.11 (95% CI 0.91–1.31) for adult ICUs and 1.04 (95% CI: 0.88–1.19) for PICUs. Conclusions: The level of care provided nowadays to children admitted to adult ICUs in Italy is similar to that provided by Italian PICUs 10 years earlier. On the other hand, there is evidence that Italian PICUs have improved the level of care in the same period. These find- ings, if confirmed, suggest a better quality of care for children admitted to PICUs as compared to adult ICUs and support the indication, when possible, of early referral to more specialized units in countries where paediatric intensive care is not centralised. Keywords Intensive care units Á Critically ill children Á Delivery of health care Á Health care surveys Á Outcome assessment Á Child care Intensive Care Med (2010) 36:1403–1409 DOI 10.1007/s00134-010-1914-5 PEDIATRIC ORIGINAL