Safer Surgery 334 The high incidence of distracters suggests that both the opportunity for and tolerance of external interference was high. Instances of reduced safety consciousness were also high, which was surprising considering the well- recognized and highly undesirable effect that nosocomial infection can have on the outcome of joint replacement surgery (Gao et al. 2000). Though equivocal evidence has been offered for the effectiveness of mask discipline for control of nosocomial infection (McLure et al. 1998, Mitchell and Hunt 1991), protocols should be applied consistently and particular attention should be paid to prosthetic implantation (Woodhead et al. 2002). Controlling the use of the telephone in theatre and reinforcing safety procedures are simple and direct solutions that could and should be applied. On a number of occasions senior team members attempted to maintain standards with little success, and in one instance an individual not involved with the operation entered the theatre during the procedure, initially without a mask, made a loud and argumentative phone call without the permission of anyone in the operating team, then left theatre, violating mask protocol on the way out. Failures to control distraction and safety are therefore symptomatic of the impact of culture on surgical quality (Baird et al. 2005), and rather than blaming individuals, the aetiology of this problem should be considered at all levels, from hospital management, through professional societies and associations, to national policy. Unlike other types of surgery, which may be reliant on the scalpel-and- suture skills of the surgeon, TKR and TKR revision operations are reliant on the appropriate use of procedure-specic instruments. Achieving the correct equipment conguration is a key technical skill for this type of surgery. Though the instruments are usually reliable, the wide range of equipment required in a short space of time placed pressure on the operating team. For example, the ten- minute period from the rst incision to completion of the femoral cuts for one make of implant requires the use of ve specialist instruments (intra-medullary rod, femoral locating device, distal femoral cutting block, femoral sizing guide and anterior/posterior chamfer cutting block), as well as a range of other incidental equipment such as drills, saws, retractors, pins, hammers, scalpels, swabs, diathermy and suction. A surgeon cannot perform the operation without the scrub nurse to provide the appropriate instruments, and a scrub nurse cannot support the surgeon effectively if they are overloaded, distracted or unable to keep up with and anticipate the surgical tasks. Given the skill required for both roles, the mutual trust required and the authority gradient between them, this relationship is brittle. It was encouraging that in the present study the interaction between surgeons and scrub nurses was mutually supportive almost without exception, and remained solution-focused when things did not go according to plan. In the TKR revision operations, which were always more demanding, the active participation of the prosthetic manufacturer sales representative helped the theatre and surgical staff to appropriately manipulate the even more complex and less familiar instrumentation, as well as providing an additional resource for aiding coordination and error capture. However, while this can have exceptional Observing Failures in Successful Orthopaedic Surgery 335 safety benets, it can also have negative effects on the rest of the team, and often caused distraction, accounting for the high level of organizational/cultural threats in high risk operations. This also suggests a deciency in the design of the instruments, which seemed too complex for surgical teams to use without a considerable amount of additional, manufacturer-specic training. It also raises ethical concerns. Overall, though patient-related difculties were frequent, the difculty of the operation was more dependent upon the design and management of the equipment, the coordination of the team, and the more frequent but generally less serious cultural threats. This study provides further evidence of the value of expert observers in identifying threats to patient safety, and suggests several structured interventions to improve surgical quality. Since equipment management failures were closely associated with the effectiveness of the team, attention to non-technical skills could be immediately benecial (Fletcher et al. 2002, Healy et al. 2004). Non- technical skills which are neither formally taught nor included in competency assessments can also be useful in avoiding or capturing problems before they can accumulate (Carthey et al. 2003, de Leval et al 2000, Helmreich 2000). Indeed the provision of pre-operative briengs, post-operative debriengs and non-technical skills development may provide a highly effective and long-term practitioner- driven safety and learning mechanism through which a broad range of inter- operative problems might be addressed. Further assessment of team-based non- technical skills was made during these operations, and has been independently reported (Catchpole et al. 2008). Attempts to improve the design and usability of equipment, and to reduce the number and range of specialist instruments (Nizard 2002, Nizard et al. 2004), should also be encouraged, provided appropriate human factors (Malhotra et al. 2005) and safety considerations (Lieberman and Wenger 2004) are present early in the design process. The observations of the computer- assisted system suggest that while equipment management issues are being partially addressed in new technologies, this can re-locate failures, making currently reliable and easy-to-congure instrumentation less reliable and more difcult to use in the future. Indeed, it is the opinion of the author after observing nearly 200 operations in at least six treatment centres that the systems for procuring, maintaining, and training for operating theatre equipment are either poorly managed or barely extant, compromising the safety of existing technologies, even before the increased risks associated with new technology are considered. Conclusions The most frequent failures may not always demonstrate the greatest threat to operative success or system function, but they can provide the environment in which major failures and adverse events are more likely. Examination of the fundamental properties of the system that these minor failure types display allows the identication of a small number of error reduction strategies that address the Safer Surgery 336 problem at the source. This is more advantageous than providing defences to a large number of unique deciencies, since safety systems themselves can become brittle, ineffective and add to the problems in the system if too many defences are implemented (Cook and Woods 1994). The failure source model provides a diagnosis of the likely source of threats and errors in this operating theatre and in this type of surgery. An estimate based on current failure rates (Feinglass et al. 2004, Vincent et al. 2001) and projected surgical volume (Dixon et al. 2004) suggests that in 2010 between 2000 and 9000 patients in the UK will experience some form of adverse event following a TKR. By taking a prospective approach to the observation of system failures and human error in orthopaedic surgery, this chapter has identied problems with the organization and culture of safety, problems associated with the design and organization of the equipment in theatre which become amplied in more difcult operations, and the potential for improved non-technical performance in the operating team. Attention to these deciencies in operating theatres would result in improvements in patient safety, surgical quality and system efciency. Acknowledgements This project was funded by the Department of Health Patient Safety Research Programme. Thanks to Mr Tony Giddings and Prof Marc de Leval for their comments on earlier versions of this text, to Dr Paul Godden for his assistance in the data collection, and to Trevor Dale, Guy Hirst, Peter McCulloch and Michael Wilkinson for their assistance in various aspects of these studies. The preparation of this chapter was kindly supported through a Leverhulme Trust Early Career Fellowship. References Baird, C., Nunn, T., and Gregori, A. (2005) Inadequate standards of hygiene in an operating theatre changing room. Journal of Hospital Infection 59, 268–9. Berlin, L. (2000) Hindsight bias. American Journal of Roentgenology 175, 597– 601. Bland, J.M. and Altman, D.G. (1986) Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1, 307–10. Carthey, J., de Leval, M.R. and Reason, J.T. (2001) The human factor in cardiac surgery: Errors and near misses in a high technology medical domain. Annals of Thoracic Surgery 72, 300–305. Carthey, J., de Leval, M.R ., Wright, D.J., Farewell, V.J., and Reason, J.T. (2003) Behavioural markers of surgical excellence. Safety Science 41, 409–25. Catchpole, K., Godden, P.J., Giddings, A.E.B., Hirst, G., Dale, T., Utley, M. et al. (2005) Identifying and Reducing Errors in the Operating Theatre (Rep. No. Observing Failures in Successful Orthopaedic Surgery 337 PS012). Patient Safety Research Programme. Available at <http://www.pcpoh. bham.ac.uk/publichealth/psrp/documents/PS012_Final_Report_DeLeval. pdf> [last accessed March 2009]. Catchpole, K.R., Giddings, A.E., de Leval, M.R., Peek, G.J., Godden, P.J., Utley, M. et al. (2006) Identication of systems failures in successful paediatric cardiac surgery. Ergonomics 49, 567–88. Catchpole, K.R., Giddings, A.E., Hirst, G., Dale, T., Peek, G.J., and de Leval, M.R. (2008) A method for measuring threats and errors in surgery. Cognition, Technology and Work 10, 295–304. Cook, R.I. and Woods, D. (1994) Operating at the sharp end: The complexity of human error. In M.S. Bogner (ed.), Human Error in Medicine (pp. 255–310). Hillsdale, NJ: Lawrence Erlbaum Associates Inc. Cook, R.I. and Woods, D.D. (1996) Adapting to new technology in the operating room. Human Factors 38, 593–613. de Leval, M.R., Carthey, J., Wright, D.J., Farewell, V.T. and Reason, J.T. (2000a) Human factors and cardiac surgery: A multicenter study. Journal of Thoracic Cardiovascular Surgery 119, 661–72. Dixon, T., Shaw, M., Ebrahim, S. and Dieppe, P. (2004) Trends in hip and knee joint replacement: Socioeconomic inequalities and projections of need. Annals of the Rheumatic Diseases 63, 825–30. Feinglass, J., Amir, H., Taylor, P., Lurie, I., Manheim, L. M. and Chang, R.W. (2004) How safe is primary knee replacement surgery? Perioperative complication rates in Northern Illinois, 1993–1999. Arthritis Rheumetism 51, 110–16. Fennell, D. (1998) Investigation into the King’s Cross Underground Fire (Rep. No. Cm 499). London: HMSO. Fischhoff, B. (1975) Hindsight does not equal foresight: The effect of outcome knowledge on judgement under uncertainty. Journal of Experimental Psychology: Human Perception and Performance 1, 288–99. Fletcher, G.C.L., McGeorge, P., Flin, R.H., Glavin, R.J., and Maran, N.J. (2002) The role of non-technical skills in anaesthesia: A review of current literature. British Journal of Anaesthesia 88, 418–29. Gaba, D.M. (1989) Human error in anesthetic mishaps. International Anesthesiology Clinics 27, 137–47. Gao, T., Lu, H., Zhou, D. and Guan, Z. (2000) Risk factors for nosocomial infections after total knee replacement. Zhonghua Wai Ke.Za Zhi. 38, 256–8. Healey, A.N., Undre, S. and Vincent, C. (2004) Developing observational measures of performance in surgical teams. Quality and Safety in Health Care 13, i33–i40. Helmreich, R.L. (1994) Anatomy of a system accident: The crash of Avianca Flight 052. International Journal of Aviation Psychology 4, 265–84. Helmreich, R.L. (2000) On error management: Lessons from aviation. British Medical Journal 320, 781–5. Kaplan, H.S., Battles, J.B., van der Schaaf, T.W., Shea, C.E. and Mercer, S.Q. (1998) Identication and classication of the causes of events in transfusion medicine. Transfusion 38, 1071–81. Safer Surgery 338 Kennedy, I. (2001) Learning from Bristol: The Report of the Public Inquiry into Children’s Heart Surgery at the Bristol Royal Inrmary 1984–1995. (Rep. No. Command Paper: CM 5207). London: HMSO. Kirklin, J.W., Blackstone, E.H., Tchervenkov, C.I. and Castaneda, A.R. (1992) Clinical outcomes after the arterial switch operation for transposition: Patient, support, procedural, and institutional risk factors. Circulation 86, 1501–15. Lawton, R. and Ward, N.J. (2005) A systems analysis of the Ladbroke Grove rail crash. Accident Analysis and Prevention 37, 235–44. Lieberman, J.R. and Wenger, N. (2004) New technology and the orthopaedic surgeon: Are you protecting your patients? Clinical Orthopaedics and Related Research, 338–41. Malhotra, S., Laxmisan, A., Keselman, A., Zhang, J. and Patel, V.L. (2005) Designing the design phase of critical care devices: A cognitive approach. Journal of Biomedical Informatics 38, 34–50. McLure, H.A., Talboys, C.A., Yentis, S.M., and Azadian, B.S. (1998) Surgical face masks and downward dispersal of bacteria. Anaesthesia 53, 624–6. Mitchell, N.J. and Hunt, S. (1991) Surgical face masks in modern operating rooms – a costly and unnecessary ritual? Journal of Hospital Infection 18, 242. Nizard, R. (2002) Computer assisted surgery for total knee arthroplasty. Acta Orthop.Belg. 68, 215–30. Nizard, R.S., Porcher, R., Ravaud, P., Vangaver, E., Hannouche, D., Bizot, P. et al. (2004) Use of the Cusum technique for evaluation of a CT-based navigation system for total knee replacement. Clinical and Orthopaedic Related Research 425, 180–8. Vincent, C., Taylor-Adams, S. and Stanhope, N. (1998) Framework for analysing risk and safety in clinical medicine. British Medical Journal 316, 1154–7. Vincent, C., Neale, G. and Woloshynowych, M. (2001) Adverse events in British hospitals: Preliminary retrospective record review. British Medical Journal 322, 517–19. Vincent, C., Moorthy, K., Sarker, S.K., Chang, A. and Darzi, A.W. (2004) Systems approaches to surgical quality and safety: From concept to measurement. Annals of Surgery 239, 475–82. Woodhead, K., Taylor, E. W., Bannister, G., Chesworth, T., Hoffman, P. and Humphreys, H. (2002) Behaviours and rituals in the operating theatre: A report from the Hospital Infection Society Working Group on Infection Control in the Operating Theatres. Journal of Hospital Infection 51, 241–55 Woods, D. and Cook, R.I. (1999) Hindsight biases and local rationality. In F.T. Durso, R.S. Nickerson, R.W. Schvaneveldt, S.T. Dumais, D.S. Lindsay, and M.T.H. Chi (eds), Handbook of Applied Cognition (pp. 141–71). New York: John Wiley and Sons Ltd. Chapter 20 Remembering To Do Things Later and Resuming Interrupted Tasks: Prospective Memory and Patient Safety Peter Dieckmann, Marlene Dyrløv Madsen, Silke Reddersen, Marcus Rall and Theo Wehner Introduction In this chapter we take a closer look at tasks requiring some form of prospective memory (PM) which allows people to remember to do something in the future or to resume interrupted tasks. Examples of Prospective Memory from the Operating Room Example 1 – Treatment in Time A patient is scheduled for surgery, requiring peri-operative antibiotics to prevent infection. It is absolutely critical that the antibiotics are applied some time before the surgeon cuts the skin. It is believed that allergic reactions are more frequent if antibiotics are given to conscious patients. So, when the patient arrives with the antibiotic taped to his chart, the anaesthesia team rst induces anaesthesia and then has to remember to give the antibiotic, whilst preoccupied with many other tasks before the surgery starts. Example 2 – Recurrent Measurements A patient with insulin-dependent diabetes has to undergo major surgery, lasting several hours. After taking an initial blood glucose level measurement before anaesthesia, the anaesthetist must then repeatedly measure the blood glucose of the anaesthetized patient. Severe brain damage may result in the case of prolonged low blood glucose. There will be several occasions during the surgical procedure where the ‘countdown timer’ will beep, but the anaesthetist will not be able to respond, because of other ongoing critical tasks (e.g., blood transfusion, hemodynamic management). It will be a repeated challenge not to forget the Safer Surgery 340 checking of the blood glucose, which might be postponed for a while but must not be forgotten for too long. Example 3 – Dynamic Change of Plans A patient with a history of several laparotomies due to a renal disability in his childhood is admitted to the clinic with an acute abdomen. In the CT scan nothing specic can be found, so the patient is scheduled for an explorative laparoscopy and it is decided to perform an appendectomy, as ‘no one would like to have to operate on this patient’s abdomen again’. The operation is rather difcult due to multiple adhesions. There is a danger of abdominal injury, so the surgeon really has to concentrate on his task. Sometime during the operation he has to remember to also perform the appendectomy. We will concentrate our discussion mainly on PM and its failures in the operating room (OR). We will explain the concept of prospective memory in some detail and explain how it is relevant for medical care and patient safety. We will use the PM theoretical framework (Brandimonte et al. 1996) to analyse the examples above and use data from a questionnaire and a simulator study to illustrate the relevance of PM for acute medical care. Finally, we will discuss implications for PM-related research and medical practice and patient safety. The longer-term goal of an improved understanding of this error form is to identify how best to deal with this challenge: removing error prone situations systematically, implementing protective measures and training people to optimize performance in PM situations. So far, we have only begun to investigate PM and its failures in acute care settings and it is too early to give sound advice on effective protection and countermeasures. Relevance in the Operating Room (OR) and Other Acute Care Settings The examples above emphasize the relevance and face value of prospective memory (PM) for medical practice. With rationalization, cost and time pressure, healthcare professionals are responsible for more than one and, sometimes several patients. Interruptions and delays are frequent in the different healthcare settings (Manser et al. 2003, Healey et al. 2006, Chisholm et al. 2000, Dieckmann et al. 2006, Dyrløv Madsen and Schou 2008). Interruptions in medical care have been found to be related to a preference for direct (synchronous) communication modes in medicine, for example face-to-face communications. Whilst using synchronous channels might help to decrease memory burden (e.g., by delegating tasks), it also increases the potential for interruptions when compared to asynchronous communication channels, like email or voice mails (Parker and Coiera 2000). Delays and interruptions in the OR stem from task inherent difculties (complications) and, the social, technical and organizational environment (Healey et al. 2006). For example, tasks are begun and nished, usually in a desynchronized Remembering To Do Things Later and Resuming Interrupted Tasks 341 fashion, by the separate crews (Gaba et al. 2001) involved in the treatment in the OR (e.g., anaesthesia crew, surgical crew, cleaning crew, etc.). While an individual starts a certain task (e.g., the anaesthetist places a nerve block before surgery to improve post-surgery care and comfort), the surgical crew might simply wait or ll the waiting time by beginning other tasks or taking a break – potentially outside the OR. The coordination of the interdependent tasks is at times imprecise (e.g., only rough or understated estimations of how long a task will take), leading to prolonged delays. While the above anaesthetist might underestimate the task duration (‘I will be nished in ve minutes’ – where the task might actually take 15 minutes), the surgical crew might underestimate the duration of the task they do in between (‘Anaesthesia will never nish in ve minutes, so let’s start our task – that might take 20 minutes’). After nishing placing the nerve block and waiting for the surgical crew to nish, the anaesthetic crew might then begin their next task to ll the waiting time, and so on. To exit this vicious circle, the interdisciplinary team would have to be very well coordinated, accept unproductive waiting time for some team members, or have to interrupt or speed up a task. So far there are few data that directly link prospective memory and its failures to the quality and safety of care, but there is some implicit evidence, also from other disciplines, especially aviation, where PM failures were found to be important (Dismukes and Nowinski 2006). Some evidence within medicine stems from incident reporting systems. The MedMARx incident reporting system (<www.usp.org/medmarx>) collects information on medication errors. In a report, 6224 cases were analysed for underlying causes and contributing factors (US Pharmacopeia 2000). Interruptions of actions were found as contributing factors in 14 percent of the cases with negative consequences. Also, safety culture (Dyrløv Madsen et al. 2007, Itoh et al. 2007) seems to have an impact on interruptions. In an investigation with 2000 participants, a Danish safety culture questionnaire study showed that 69 percent of the participating healthcare professionals agree or highly agree that interruptions are reasons or contributing factors for adverse events (Dyrløv Madsen and Schou 2008). Finally, there is much anecdotal evidence from healthcare professionals who often respond enthusiastically to the concept of PM, claiming that it describes very common experiences of daily practice. Interruptions as Situations with High PM Burdens Many different situations place a high burden on PM: interruptions, distractions, parallel tasks, delays, etc. In the following section we focus on interruptions to narrow the scope of the discussion, allowing for more thorough analysis. Interruptions and distractions are known to have a negative impact on performance, as they inuence and minimize the professionals’ ability to stay focused on a specic task (intention) (Beyea 2007). By analysing PM errors in interruption situations, it will also be possible to take a work analytical stance. Investigating PM errors in interruptions will help in understanding the personal Safer Surgery 342 redenition of tasks. What do professionals see as an interruption (from the outside, e.g., by colleagues, alarms, telephone), a distraction (from the inside, e.g., sudden thoughts or ideas, difculties in concentrating, inner resistances) and a task-inherent challenge (e.g., difculties during intubation or nding the correct differential diagnosis)? This perspective is especially important in the multiple- crew, interdependent task context of most acute medical care settings. What does one crew accept as a legitimate task of the other crew? What do they see as a necessary interruption? What is accepted and what is not? These factors might inuence the crew and team climate, the willingness to cooperate and thus, also patient safety. Analysing PM errors in the interruption context will allow for supplementing the cognitive perspective with social aspects. Most professionals both expect and accept interruptions as part of their normal work day (Beyea 2007, Parker and Coiera 2000). From the cognitive perspective, those interruptions might be helpful to decrease one’s own memory load interrupting a colleague and delegating the task frees one’s own cognitive resources however, this interruption may increase that colleague’s cognitive load. From the social perspective, interruptions might be related to reciprocal acceptance: accepting being disrupted and asked for help might, in turn, legitimize oneself interrupting others later. Interruptions are likely to be related to hierarchy as well: they could be seen as a status symbol (‘I am important and needed – people interrupt me and I can help solve their problems’) but also as an integral part of a certain position (by denition and job description, a consultant is a centre in a complex network of information ow with many ‘inputs’ and ‘outputs’ and thus interruptions) (Groopman 2007, Jauhar 2007). A recent observational study of nurses and doctors in an oncological ambulatory unit in a Danish hospital revealed many disturbances and interruptions in the daily workow (Dyrløv Madsen and Schou 2008). Their study documented a large number of interruptions, especially for nurses, which often resulted in staff forgetting to nish tasks. Nevertheless, during post-observation interviews about the observed events, the professionals, on the whole, found them appropriate, legitimate and were able to justify their frequency. The interviewees claimed the interruptions were appropriate for the task at hand and/or to improve teamwork. On further reection and discussion with the interviewer the professionals, however, recognized that many of the interruptions were a problem for patient safety and that they probably could be minimized or avoided if everyone was dedicated to do so. Therefore, how interruptions are perceived seems to be inherent to the professional culture. Finally, interruptions are relevant for PM research as it might be easier to identify the underlying intention, as compared to other situations like delays in having the opportunity to execute a certain intention. We will discuss this aspect in more detail below. In summary, by studying interruptions in relation to PM in acute medical care, we hope to advance the knowledge about this error type. Remembering To Do Things Later and Resuming Interrupted Tasks 343 Denition of Prospective Memory and its Failures Prospective memory is dened as a psychological processes which enable humans to execute previously formed intentions during an appropriate but delayed ‘window of opportunity’ (Harris and Wilkins 1982) without being explicitly reminded to do so (Dieckmann et al. 2006; for an overview see Brandimonte et al. 1996). Not resuming the interrupted task, resuming the interrupted task at the wrong spot (missing a step or wrongly performing a step more than once) or not executing an intended task might be related to errors of PM. However, in order to make this ‘error diagnosis’, differential diagnoses need to be considered. Not resuming or executing a task might be related to a conscious decision (which might in itself be right or wrong), it might be because the window of opportunity was not recognized, or that other cognitive and emotional processes prevented the execution (e.g., unconscious resistances). Weimer (1925) suggested that one needs to distinguish between error forms (here: missing the execution of an intention, failing to resume a task or resuming it at the wrong spot) from the underlying error types, i.e., the underlying psychological processes (e.g., forgetting, denying, deciding against it). What might look like the same error from the ‘outside’ (error form: an intended task is not executed), might be very different from the ‘inside’. Focusing on PM in interruptions as a situation type/error form helps in narrowing the search space for underlying error types. In general, one can distinguish between different classes of PM tasks, depending on how the window of opportunity is to be described: activity-based PM tasks (i.e., do X after you are nished with M, see Example 3, ‘Dynamic change of plans’, above), event-based PM tasks (i.e., do X when Y occurs, see Example 1, ‘Treatment in time’) and time-based PM tasks (i.e., do X at time Z, see Example 2, ‘Recurrent measurements’). Within the PM framework, ve phases are distinguished that belong to a full PM cycle (Ellis 1996): In the beginning, the intention is formed and encoded, containing three components: that, what and when. The that component can be seen as energizing the inner system, preparing the person to act at all (Goschke and Kuhl 1993, Lewin 1926). The what component species what needs to be done in more detail. The when component helps in anticipating the window of opportunity in a temporal and conditional sense. During the retention phase the intention is kept in memory, more or less consciously, while the duration of the interval does not seem to have a large impact on the retention of the intention and its execution (Kvavilashvili and Ellis 1996). Short interruptions might be enough to hinder the execution of intentions. The window of opportunity begins and the intention could be executed. It is a question of whether the person intending to act recognizes the beginning of the window of opportunity and is also able to retrieve the intention. For 1. 2. 3. . transfusion medicine. Transfusion 38, 1071–81. Safer Surgery 338 Kennedy, I. (2001) Learning from Bristol: The Report of the Public Inquiry into Children’s Heart Surgery at the Bristol Royal Inrmary. allows the identication of a small number of error reduction strategies that address the Safer Surgery 336 problem at the source. This is more advantageous than providing defences to a large. and Reason, J.T. (2001) The human factor in cardiac surgery: Errors and near misses in a high technology medical domain. Annals of Thoracic Surgery 72, 300–305. Carthey, J., de Leval, M.R .,