Naslov
Daily data recording in a National comparative clinical database targeting continuous quality improvement of the University hospital single intensive care unit performance

Autori
Degoricija, Vesna ; Milošević, MIlan ; Sharma, Mirella, Šmit, Ivana ; Šefer, Siniša

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Izvornik
Abstracts of the 25th Annual Congress of The European Society of Intensive Care Medicine (ESICM) Lives 2012. U: Intensive Care Medicine 2012 ; 38 (Supplement 1): S1039 / European Society of Intensive Care Medicine - Lisabon : Springer, 2012, S1039-S1039

Skup
European Society of Intensive Care Medicine (ESICM) LIVES 2012 ; 25th Annual Congress

Mjesto i datum
Lisabon, Portugal, 13.10.2012. - 17.10.2012

Vrsta sudjelovanja
Poster

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
intensive care unit ; data collecting ; performance

Sažetak
Introduction: University hospital single medical intensive care unit (ICU) was enrolled in the Croatian national comparative clinical database (CROICUnet) aiming date collecting for improvements in the organization and practice of the unit, development of a self program for ICU research, managing and practicing, and improving the quality of care. The standardized mortality rate was lower than might be expected when measured against the outcomes of international best practice. The aim of the present paper was to assess the impact of a number of factors on the likelihood of ICU death outcome. Methods: Study patients were recruited from the medical ICU at the Sisters of Mercy University Hospital, Zagreb, Croatia, between 2005-2008. Demographic, clinical and laboratory date were collected by the three researches on day-by day- basis, and recorded on line in National database CROICUnet. The study included overall 1896 consecutive patients. The standard of care was measured by length of stay (LOS) and the results of care. Final outcome was ICU outcome. Results: Binary logistic regression was performed to assess the impact of a number of factors on the likelihood of ICU death outcome. The model contained 20 independent variables (age, gender, admission time, LOS, APACHE II, SOFA and GC scores, atrial fibrilation, hypertension, cardiomyopathy, coronary arthery disease, cerebrovascular disease, peripheral vascular disease, COPD, diabetes mellitus, cirrosis, chronic renal failure, malignancy, sepsis, and shock) ). The full model containing all predictors was statistically significant, X2 (21, N=1894)=783.74, P<0.001, indicating that the model was able to distinguish between influence of predictor variables on selected outcome (ICU death). The model as a whole explained between 33.9% (Cox and Snell R square) and 55.8% (Nagelkerke R squared) of the variance in symptomatology status, and correctly classified 90.1% of cases. Predictors and their influence on ICU death are shown in Fig.1 (Forrest plot). Significant contributions are marked in red. Highest odds ratio was noted for patients who were in shock (6.7 times more likely to die, controlled for all other factors in the model). Conclusion: Daily data recording in a national comparative clinical database resulted with not only continuous quality improvement of the patient care, and ICU organization and managing, but with the development of the research programs which led to implementation of the new ICU and hospital pathways in managing critically ill patients.

Izvorni jezik
Engleski

Znanstvena područja
Kliničke medicinske znanosti

POVEZANOST RADA