Sedation and Patient Safety Debora Simmons, RN, MSN, CCRN, CCNS The Institute for Healthcare Excellence, The University of Texas MD Anderson Cancer Center, 1515 Holcombe, Unit 141, Houston, TX 77030 – 4437, USA The Eindhoven classification system provides a basic framework for approaching safety within systems. The Eindhoven classification was originally developed for industrial applications and has been used extensively in the chemical industry and has been cited within Institute of Medicine reports [1].A medical modified Eindhoven classification scale is used for the root causes identification in the Medical Event Reporting System for Transfusion Medicine transfusion event reporting system [2].Causal codes, such as used in the Eindhoven scale, are useful as a framework for gathering data, standardizing inves- tigations, and collecting aggregates of data to examine clusters of events [3]. In addition, the simplicity of the Eindhoven codes allows for use of readily available information for nurses and facilitates the analysis. According to the Eindhoven scale, root causes are classified into three major categories or domains: (1) technical (equipment, software, forms); (2) orga- nizational (policies, procedures, and protocols); and (3) human causes (knowledge-based, rule-based, and skill-based). The three domains are useful in classi- fying contributing factors and organizing causes. Because the initial reaction in an error event is to focus on the human factors, it is important to consider the first two domains before looking at human factor contributions (Table 1). Technical risks to safety The Eindhoven classification considers technical failures as being related to the ‘‘hardware’’ of care including mechanical devices, forms, and software. Medical devices are a major safety concern both in consideration of a primary failure of the device and as a failure in human-machine interactions. The Food and Drug Administration (FDA) main- tains surveillance over medical devices and receives over 80,000 reports annually regarding failures, serious injuries, and deaths related to medical devices [4]. The Institute of Medicine in Canada reviewed 425 incident reports of parental, epidural, insulin, and patient- controlled analgesia pumps and found 23 deaths and 135 injuries related to technical features of pumps [5]. Both groups found similar problems related to devices. The FDA MedWatch program collects suspicious and confirmed device-related problems and provides a free public service news alert [6]. There must be caution in reviewing device failures in that reports of device failures are not necessarily the readily apparent cause [7]. A careful investigation into causes is always suggested, and to do so, a complete review of factors is necessary. Despite increasingly strenuous testing and device design requirements, failures related to human inter- action with devices continue. Broad categories of fail- ures include factors related to devices, external factors, support system failures, user errors, and tampering or sabotage [7]. Commonly cited failures include 1. Lack of standardized devices in a facility (eg, the on switch on one device may be the off switch on a similar device of a different model or manufacturer) [4] 2. Lack of safe default settings so alarms that are suspended may not return to audible levels (software default settings) [4,8] 3. Designs that allow for free flow of medica- tions [5,8] 0899-5885/05/$ – see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ccell.2005.04.009 ccnursing.theclinics.com No funding support was received for this paper. E-mail address: dsimmon@mdanderson.org Crit Care Nurs Clin N Am 17 (2005) 279 – 285