ABSTRACT
BACKGROUND: Physical restraint is frequently used in the intensive care setting but little is known regarding its clinical scenario and effectiveness in preventing adverse events (AEs), defined as device removal. METHODS: We carried out a prospective observational study in three Intensive Care Units on 120 adult high-risk patients. The effectiveness of physical restraint was evaluated using the propensity score methodology in order to obtain comparable groups. RESULTS: Physical restraint was applied in 1371 of 3256 (43%) nurse shifts accounting for 120 patients. Substantial agitation, the nurse's judgement of insufficient sedation and sedative drug reduction were positively associated with physical restraint, whereas the presence of analgesics at admission, increased disease gravity and the treating hospital as the most substantial variable showed a negative association. Eighty-six AEs were observed in 44 patients. Quiet (SAS=1-4), unrestrained patients accounted for 40 cases, and agitated (SAS≥5) but physically restrained patients for 17 cases. The presence of any type of physical restraint had a protective effect against any type of AE (OR=0.28; CI 0.16-0.51). The observed AEs showed a limited impact on the patients' course of illness. No physical harm related to physical restraint was reported. CONCLUSION: Physical restraint efficiently averts AEs. Its application is mainly driven by local habits. Typically, the almost recovered, apparently calm and hence unrestrained patient is at greatest risk for undesirable device removal. The control/interpretation of the patient's analgo-sedation might be inappropriate.
