Abstract/Summary

Background EPrescribing was developed to reduce medication errors and adverse drug events (ADEs) in intensive care units (lCU). Even though ePrescribing is effective in reducing medication errors and ADEs in ICUs, it is clear from the literature that ePrescribing may introduce new errors which could have been caused by the design and interface of the systems. Therefore, it is important to explore ePrescribing systems to improve their design and efficiency, and improve patient safety based on the assistance that such systems provide to professionals in making clinical decisions from a multi-level perspective: at drug level, process level and system level. Methods A mixed methods design was used to address the multi-level approach. In a quantitative phase, a systematic review was conducted to identify drugs/drug groups and errors most commonly associated with preventable adverse drug events (PADEs) in ICU patients at a drug level, and an eDelphi study to identify high-risk prescribing safety indicators specific to ICU with experts in pharmacotherapy/adverse drug events to address the process level. In a qualitative phase, semi-structured interviews were conducted with staff members to explore the benefits of using the ePrescribing system, problems encountered with the use of the ePrescribing system and elicit users' recommendations to improve the efficiency of the ePrescribing system in the ICU of Royal Berkshire National Health Services Foundation Trust at the system level. Results Sedatives and analgesics were the drugs/drug groups most frequently associated with pADEs in ICU patients, and dosing errors were the errors most commonly associated with pADEs in ICU patients. Eleven high-risk prescribing safety indicators specific to the ICU and which are amenable to clinical decision support were identified. The prescribing indicators list could be used as a standardised tool for the routine collection of prescribing error data from both paper based and electronic systems. The prescribing indicators list could also be used to evaluate the impact of safety improvement initiatives such as implementing ePrescribing with or without clinical decision support. The benefits of the ePrescribing system reported in this study included accuracy ofthe ePrescribing system and ease ofaccess using the ePrescribing system. Problems encountered with the ePrescribing system in this study included the mixed system (both paper and electronic) of prescribing and inefficiencies (incomplete drug database and absence of automatic functions) of the system. The benefits and issues about the use of ePrescribing reported in this study were consistent with the current literature. New issues were also noted in this study and these include: co-signing of drugs, deskilling of professionals and changes to pharmacists' work practices. This could be addressed by improving the design of the systems in response to feedback from the users of the system and also involving users at every stage of the design and refinement processes. Therefore, a user-centred systems approach that takes into account the needs of the users and their work environment should be used when designing and implementing ePrescribing systems. Conclusions Errors can be understood at the drug, process and system levels, to inform the design of strategies to reduce ADEs and medication errors at a whole system level. The findings from this study could inform the basis for designing and developing strategies to improve patient safety and utilisation of the ePrescribing system.