Patient safety is defined as the avoidance, prevention, and amelioration of adverse outcomes or injuries stemming from the process of health care . Preventing medical errors is one of the most important aspects of providing safe and high quality care within health care systems.
Leape  defines error as an unintended act of either commission or omission that did not achieve its intended outcome. Significant research has been undertaken over the past few decades in studying issues around errors and adverse events. The delivery of medication to patients is the end result of a complex process comprising many steps, each vulnerable to error. Therefore in the acute hospital setting, preventing medical, and in particular medication errors, through the use of Information and Communication Technologies (ICT) is widely advocated as one of the most effective means of achieving such improvements [3, 4]. Anticipated benefits include, for example, the reduction of legibility errors thus making prescribing safer, the increase in efficiency through eliminating lost notes or charts, improvements in communication between care providers and the increase in capacity at a time of workforce and financial constraint .
There is growing evidence from countries such as the United States where the use of health care ICT has a longer history, that even though electronic prescribing systems with inbuilt clinical decision support may contain features that protect against error and thus enhance patient safety , they may also introduce new risks of their own that are not always immediately obvious . Some of the issues that have been identified are cognitive overload; loss of overview of the clinical situation; errors in data entry and retrieval; excessive reliance on electronically held data; disruptions of established workflow patterns and the tendency to infer that data entry equates to communication within and among health care teams [7, 8]. These new risks are the unintended and unanticipated consequences associated with the introduction of the electronic prescribing system. There is a distinction to be drawn between the unintended and the unanticipated consequences in that "unintended" implies lack of purposeful action or causation, while "unanticipated" means an inability to forecast what eventually occurred . Unintended errors can be exemplified by situations such as picking a wrong option from a drop down list or typing 100 instead of 10, which we consider to be akin to a slip which is a failure of attention leading to inadvertent actions. Lapses are failures of memory and often occur as a result of distractions or interruptions, whereas mistakes are caused by inappropriate clinical reasoning and failures in planning or problem solving . In purely technological terms, an unanticipated consequence can occur when there are incorrect rules within a computer system or some other technical failure. Furthermore, unanticipated errors can occur if a user deliberately deviates from the standard procedures, recommendations or guidelines thus committing a technical violation  often referred to as a 'work-around' when there is a glitch in the way information systems are used relative to initial intentions [12, 13]. While previous studies, predominantly North American, have produced some promising insights, given the fundamental differences in health care delivery, it is likely that they are not easily transferable into the UK context. Understanding the contribution of health care ICT to improving patient safety as well as its unanticipated and unintended consequences in the local and UK context is vital in order to facilitate the design of interventions at individual, team and organisational level to mitigate these potential new threats and thus further improve patient safety.
According to Shekelle and Goldzweig , computerised systems should be considered as a complex intervention with four key components: technical, human, project management, and organisational and cultural change all of which must be systematically studied. However, as Greenhalgh et al.  conclude in their extensive review of the literature of electronic patient records, research has only scratched the surface of what the introduction of clinical decision support systems means, at the level of fine-grained detail, for a health care organisation and the staff and patients who practise and interact in that setting. We propose an approach which acknowledges the mutual influence between technology and its social and organisational context [15, 16] in which research focuses on the sociotechnical aspects (issues involving the interplay of organisational, individual, social and technical components) of patient safety and risk management in hospitals. New technology is not simply integrated into current practice and ways of working, but has a profound impact on organisational arrangements, professional work, and medical practice. This social view of technology recognises that individual systems are anchored to a variety of other practices as well as broader organisational conditions  which we investigated with respect to this locally developed system.
The purpose of this study was to explore the types of possible unintended and unanticipated consequences as well as the nature of their effects within the sociotechnical and organisational context at an acute hospital. To date, few hospitals in the UK have implemented a comprehensive IT system with the aim of reducing error and enhancing patient safety, and there is as yet little knowledge on the implications of such a highly computerised environment for working practices and care processes. In order to find out if the electronic prescribing system had unintended consequences in creating new errors, we accessed the existing risk management system to collect data in the form of reported incidents which were related to the medication process (i.e. prescribing, preparation, supply and administration of medicines). From a sociotechnical perspective, capturing this subjective contextual data was useful insofar as we sought to identify the dynamics between technology and the social, professional, and cultural environment in which it was used which in turn can be useful in developing preventative strategies . However, many incidents may not be reported or may not be reported in enough detail regarding contributory factors and preceding events [19, 20], or may contain inaccurate information.
Some recent UK based studies [21–23] have investigated medication errors related specifically to electronic prescribing systems by retrospectively reviewing electronic records for a random selection of patients, or by retrospectively analysing all medication orders generated in a specified period of time. In contrast, we worked with the hospital's risk management team, collecting all medication related clinical incidents, irrespective of their association with the electronic prescribing system, reported by hospital staff. Our objectives were to describe the range of medication incidents reported by hospital staff, identify the proportion of incidents that relate to sociotechnical factors and explore the nature and characteristics of reported sociotechnical incidents.
Based on the fact that clinical incident reports are not by their nature neutral descriptions when reported by individual members of health care staff, one of the foci of our analysis was the manner in which sociotechnical problems were described and presented in the incident reports, and what views and perceptions about the system they revealed. The study also allowed us to gain an insight into hospital staff's views of the system and its perceived role in mediating or preventing medication related errors. We did not seek to evaluate the technical performance of the system or investigate the validity of concerns expressed by clinical staff related to perceived technical problems.