Development of a risk assessment tool for deliberate self-extubation in intensive care patients

Objective. To develop a risk stratification scheme for deliberate self-extubation in intensive care patients. Design. A nested case-control study. Setting. Four surgical ICUs, one medical ICU, one coronary care unit, and one emergency department of a tertiary care center. Measurement. In a 3-month period, the number of ventilation periods, ventilation days, and unplanned extubations were recorded. Potential determinants of unplanned extubation were assessed with a translated (English to Dutch) and modified version of the "Unplanned Extubation Data Collection Tool." Patients. Clinical and demographic characteristics and circumstances of the 26 unplanned extubations were compared with those of 48 randomly selected control patients who did not experience unplanned extubation. Results. The incidence of unplanned extubation was 4.2%, corresponding to 0.68 unplanned extubations per 100 ventilation days. The incidence was substantially lower at surgical ICUs (2.6%) compared with that at medical ICU/CCUs (9.5%). Multiple logistic regression analysis revealed that patients with a low sedation level (Bloomsbury Sedation Score) and a higher degree of consciousness (Glasgow Coma Scale) were at higher risk for deliberate self-extubation. The explained variance of this model including these factors was 67.3%. Conclusion. Based on the risk factors identified, a risk assessment tool was developed. Systematic administration of the Bloomsbury Sedation Score and the Glasgow Coma Scale, and the use of the stratification scheme, allows identification of patients at risk. Appropriate reduction of sedative drugs during weaning, a timely extubation, and increased surveillance in patients identified to be at risk are possible interventions to diminish the number of unplanned extubations.