Patient safety

In November 2005 the National Audit Office (NAO) produced a report on the current state of patient safety in the UK entitled ‘A safer place for patients: Learning to improve patient safety’ (National Audit Office 2005). The statistics presented in this report make startling reading. For example, in 2003–2004 hospital, mental health and ambulance trusts in the UK reported 885 832 patient safety incidents or near misses. In 2004–2005, the number was 974 000. In general, this figure does not include those arising from hospital-acquired infections. Of these, 2181 incidents are reported as resulting in death, although it is acknowledged in the report that these figures may well be under-reported. The cost to the National Health Service (NHS) of patient safety incidents and near misses is given as £2 billion a year in extra bed days, with the value of outstanding clinical negligence claims being in excess of £2 billion to the end of 2003–4. By anyone's standards, this is a huge amount of money, not to mention the individual suffering, pain and anxiety that inevitably accompanies such incidents. Thus, patient safety is an issue in which the government rightly takes an ever-increasing interest.

The NAO report also gives some insight into the source of patient safety errors and near misses. Of these, over 30% are related to injury, including falls. Seven percent are related to medication errors, and 4% to equipment-related problems. Translated into sheer numbers, there were therefore somewhere between 250 000 and 300 000 incidents involving injuries such as slips, trips and falls in clinical environments.

Traditionally, ergonomists and human factors specialists are concerned with these problems on a broader scale and have much to offer in terms of reducing the level of patient safety incidents. Interest in the ergonomics of patient care is not of course new to ergonomists. In 1997 Francois Daniellou started the Hospital Ergonomics Technical Committee (HETC) for the International Ergonomics Association (IEA) with 56 members from 14 countries. In 2002 Sue Hignett took over as chair of HETC9, renaming it Healthcare Ergonomics to reflect the range of activities and the membership has grown to 98 from 17 countries. All areas of the healthcare industry are represented for clinical and non-clinical healthcare working environments in both acute and community (social) settings, and interests range from product and architectural design and musculoskeletal issues, to accident analysis and organizational systems. There is a growing awareness of the knowledge and expertise that ergonomics can offer and this is seen clearly at both specialist conferences (e.g. Healthcare Ergonomics Systems and Patient Safety 2005 in Florence) and healthcare ergonomics streams at major congresses. For IEA 2006 there will be over 80 papers in the Healthcare Ergonomics stream and a second major Healthcare Ergonomics and Patient Safety international conference is planned for 2008 in Edinburgh.

The heightened interest in patient safety at a government level has also meant that opportunities for ergonomics research in this area have increased accordingly with the inception of bodies such as the National Patient Safety Agency, research networks such as the Patient Safety networks and grant programmes such as the National Patient Safety Research Programme. The five UK Patient Safety Networks were set up in response to a joint research councils call for greater numbers of and better quality research proposals in patient safety. Pump-priming funding was allocated to these networks and they have each extended their work and role in patient safety, and new funding found as a result.

It is clear that ergonomics has a vital role to play in the improvement of patient safety. This role ranges from involvement in the evaluation and implementation of organizational strategies, which encourage the reporting of patient safety incidents and near misses, through measurement and recording of those incidents themselves (and the measurement of variables known or thought to impact on patient safety), to direct interventions aimed at improving patient safety. In this special issue a set of articles is presented that covers this range of activity.

An open invitation to submit to this special issue was given, being advertised in the August 2004 edition of this journal as well as at relevant patient safety and other ergonomics and human factors conferences. The papers presented in this issue represent more or less the full range of areas in which ergonomists might contribute in the area of patient safety, from macro-ergonomic work systems analysis at one end to the micro-ergonomic at the other, where specific, relatively contained issues are considered.

The NAO report (National Audit Office 2005) reports that the safety culture within NHS trusts is now more open and fair and is slowly moving away from the ‘blame’ culture that used to exist, although this ethos has not been entirely cast aside. Compared with other industries, such as aviation, the Health Service has traditionally been slow to acknowledge that errors will occur and that, rather than blaming the perpetrator of those errors, steps should be taken to reduce them where possible. This can only be achieved through effort at the organizational level. The first two articles deal with those organizational issues. The article by Fogarty and McKeon used structured equation modelling in order to investigate the impact of organizational climate and unsafe medication administration practices in rural Australia. They show that nurses' level of distress and morale is influenced by organizational climate and that these factors in turn influence the level of unsafe practice. Following in the same vein, the second article by Elfering, Semmer and Grebner looked at the relationship between workplace stress and safety-related incidents. The authors used observational techniques, to measures nurses' work stressors, and self-report to measure the occurrence of safety-related incidents. The results suggest that safety-related incidents occurred because of incomplete or incorrect documentation, medication errors, delays in the delivery of patient care and violent patients. The study also shows that high levels of job demand and low levels of control are associated with such incidents. Thus, both the Fogarty and Mckeon paper and the Elfering et al. paper demonstrate the effects of negative workplace factors on the incidence of patient safety incidents.

A third paper looking at the broader issues of the role of organizations and the individuals within it is that by Carayon, Hundt, Alvarado, Springman and Ayoub. Using quantitative and qualitative approaches the authors gathered data from healthcare providers and focused on the main quality and safety of care issues concerning communication with patients, coordination of reports and forms, patient and staff time pressures and standards of care. The results showed that, overall, the healthcare providers reported high quality of care provided by themselves (98%) and their surgery centre (96%). With regard to patient safety (i.e. cancellations of surgeries, patient safety problems and serious mistakes), there was a clear difference in perceptions reported by the physicians vs. the nurses and other staff. Nurses and other staff were more likely to report patient safety problems than physicians.

‘If you can't measure it, you can't manage it’ is a message from the National Audit office report (p. 23) and in the fourth paper Kostopoulou describes the process of developing a taxonomy of patient safety in general practice. In this paper, the sources of patient safety events in five West Midlands practices were traced back to their root causes and a taxonomy for those events developed. The taxonomy has multiple levels, with the events being classified at three levels: information processing; whether the immediate causes were internal or external to the individual; and remote causal factors. A fifth paper by Kanse, van der Schaaf, Vrijland and Mierlo followed up 31 near-miss patient safety incidents in a hospital pharmacy and explored in detail the ways in which potential errors were spotted. Using confidential reports and follow-up interviews, the authors identified again the important role of an appropriate organizational culture in following up and dealing with patient safety incidents, areas in which procedures could be improved and the need for support of unplanned recovery.

The single most important underlying concept in safety, including patient safety, is that of error, its detection, and recovery from error. In the sixth paper, Nyssen and Blavier look at the problem of error detection, which they point out is seriously under-researched compared with many other error-related issues such as measurement, classification, consequences, etc. They gathered data from a set of reports on anaesthetic incidents or near misses, which had been generated through a reporting system developed in two Belgian hospitals, and then looked at the pattern of error detection. They found that the majority of errors were detected using regular monitoring of the equipment (the ‘standard check’) but that different types of errors were detected in different ways and that the types of error committed were dependent on the training level of the anaesthetist. The next paper, by Dieckmann, Reddersen, Wehner and Rall looks specifically at prospective memory failures (failure to carry out an action in the future) as a possible factor in errors and patient safety, another under-researched issue. One feature of this study is that the authors showcase the potential use of patient simulators as a methodology for investigating medical error. They carried out studies where they varied the subjective importance of participants' intention and also the type of intention. Over one-quarter of all intentions were not actually carried out, but there were also interactions between intention importance and type of intention.

The next three papers look in detail at ways of measuring likely global sources and causes of error, and each of these demonstrate the kind of detailed and thorough approach that will need to be taken if detailed objective evidence is to be obtained with regard to patient safety in these areas. The first, by Hallock, Alper and Karsh from the USA, provides an exploratory perspective on the complexities of a healthcare system, an out-patients facility and a macro-ergonomic work system analysis of the diagnostic testing process. Their paper highlights the many steps in the diagnostic testing process, from ordering to result dissemination. Their key results indicate a large number of potential hazards within the diagnostic testing process (for example, test order, sample collection, sample delivery and so on through to results dissemination), which all contributed to delays in the process. They further note that the majority of potential hazards occurred across the boundaries of different systems and were related to poor or absent feedback structures.

The paper by Catchpole, Giddings, de Leval, Peek, Godden, Utley, Gallivan, Hirst and Dale investigated the types and sources of system failures in paediatric cardiac surgery in the UK. They used checklists, notes and video recordings to observe 24 successful operations, collecting data on a total of 366 failures. They conclude that adverse events are likely to be associated with errors relating to, for example, communication, equipment, the safety culture, patient-related problems and perfusion-related problems. A further paper by Healey, Sevdalis and Vincent demonstrates the development of an observational tool aimed at recording distraction and interruption in the operating room during surgery. They assigned the observed events to pre-defined categories, which were weighted in terms of the involvement of team members and other factors affecting the power of each factor to distract. This study demonstrates quite a startling number of distractions during operations (an average of 0.3 events per min over the whole set of cases observed). They also found that the overall level of interference correlated with the number of door openings observed and that there were many sources of interruption and distraction, ranging from those intrinsic to the operation to extrinsic factors such as bleepers, phone calls and external staff entering and leaving the operating theatre.

Of course, ergonomics is sufficiently well-developed as a discipline to be able to provide direct advice on the kinds of interventions that are very likely to be successful in improving patient safety, and the last two papers review two areas where existing knowledge might inform current practice where significant problems are known to exist. The first paper, by Hignett & Masud, looks at the issue of in-patient falls, which the NAO has highlighted as being the single most significant cause of patient safety incidents and near misses (National Audit Office 2005). The authors review existing knowledge of risk factors and conclude that although there have been many recommendations made about the management of environmental hazards, there is supporting objective evidence only for the use of bed rails, and three systematic reviews have failed to provide supporting evidence for many of the recommendations currently being made. Worryingly, the available evidence on bed rails suggests that bed rails not only fail to reduce the frequency of falls, they may also exacerbate the severity of injury. Hignett and Masud also present a framework for the development of an environmental hazard assessment model. The final paper, by Hellier, Edworthy, Derbyshire and Costello presents a review of warning and labelling research applied to the issues and problems surrounding patient medication labelling, a significant factor in the second most cited cause of patient safety incidents, medication errors. Evidence in the medical domain (and the associated guidance and legislation) is often piecemeal. The review paper presents evidence-based guidance on medication labelling, focusing on effects that have been found to be robust and systematic in relation to design characteristics, such as font size, colour, the use of signal words and linguistic variables, and their effects on crucial variables such as compliance, understandability and discriminability.

Patient safety is a big public concern and ergonomists have a lot to offer across a broad range of skills. It is believed that this special issue will highlight some of those concerns as well as some of the skills and knowledge that ergonomists have to offer. There is also considerable scope for ergonomists not currently working in patient safety to become involved in the future.

Finally, we would like to thank all of the authors for their contributions, as well as the many reviewers of papers we would like to list here. Our thanks go to Janet Anderson, Marilyn Sue Bogner, James Brown, Peter Buckle, Ken Catchpole, Pascale Carayon, Bryony Dean, Peter Dieckmann, Achim Elfering, Alistair Gale, Georgina Fletcher, Andrew Healey, Pam Jacobs, Ray Jones, Ben-Tzion Karsh, Olga Kostopoulou, Rebecca Lawton, Jan Noyes, Anne Sophie Nyssen, Fabrice Parmentier, Isabel Smith, Neville Stanton, Charles Vincent, James Ward and Yan Xiao for their contributions as reviewers of the papers presented.