Incidents in anaesthesia: past occurrence and future avoidance.
Article Type: Report
Subject: Anesthesia (Usage)
Anesthesia (Health aspects)
Perioperative care (Management)
Perioperative care (Safety and security measures)
Patients (Care and treatment)
Patients (Safety and security measures)
Authors: Catchpole, Ken
McCulloch, Peter
Pub Date: 10/01/2009
Publication: Name: Journal of Perioperative Practice Publisher: Association for Perioperative Practice Audience: Academic Format: Magazine/Journal Subject: Health; Health care industry Copyright: COPYRIGHT 2009 Association for Perioperative Practice ISSN: 1750-4589
Issue: Date: Oct, 2009 Source Volume: 19 Source Issue: 10
Topic: Event Code: 260 General services; 200 Management dynamics Computer Subject: Company business management
Accession Number: 210725235
Full Text: Introduction

This article is a revised version of an analysis of reported incidents related to anaesthesia, originally published in the journal Anaesthesia (Catchpole et al 2008a) and undertaken on behalf of the National Patient Safety Agency. The purpose was to examine the range, types, frequencies and causes of reported patient safety incidents associated with anaesthesia. First we examined anaesthetic incidents as a sub-set of the total number of reported incidents; then we examined pre-surgery assessment, epidural anaesthesia, and anaesthetic awareness incidents, as they were identified as being frequent and of potential concern. To our knowledge it was the first paper to analyse and present results of the NPSA's database in a clinical academic journal. Here, we take the opportunity to re-present and review the findings in light of subsequent progress in understanding and improving patient safety and quality of care.

Method

This study considered all patient safety incidents reported to the National Reporting and Learning System (NRLS) from January 2004 to February 2006. At the time of the study there were approximately 550,000 incident reports in the database. In order to examine events related to anaesthesia, pre-surgery assessment, epidurals, and awareness events, keyword searches were conducted to focus on reported incidents that were associated with each of these areas of interest (Table 1). Keywords were identified based on the understanding of how each problem might manifest itself in a reported event. We further analysed two sub-groups of pre-surgery assessment incidents: unexpected events in surgery and allergy or allergic reactions in surgery, as they are most likely to originate from problems in the assessment process. Existing type, cause and severity classifications were used, even though this was of variable quality and consistency. Incidents were removed that were featured more than once or were considered inappropriate following inspection.

Results

A total of 12,649 incidents (approximately 2% of all reported incidents) related to anaesthesia. Figure 1 shows the breakdown of severity and type. Most incidents did not lead to any harm, with only 2% leading to severe harm or death. Of largest cause were treatment and procedure problems, and infrastructure or equipment problems, with clinical assessment being the least likely cause. Using NRLS data to estimate the likely effect given a cause (Table 2), a number of patterns emerge regarding cause and severity. Documentation/consent issues and infrastructure/equipment issues appear to carry a lower risk: they more frequently result in no harm, than process, treatment, or clinical assessment incidents.

[FIGURE 2 OMITTED]

Examining specific types of incident, 831 reports were found to be related to the pre-surgery assessment process. A further 43 unexpected events in surgery and 161 allergy or allergic reactions in surgery were also found. Levels of harm for each are shown in figure 2, with unexpected events clearly the least frequent, but the most likely to cause injury if they occur.

Treatment/procedure failures are frequent for all three incident types (Figure 3), as are documentation and communication failures. 516 incidents were found to be associated with epidural anaesthesia, which appears to carry a higher degree of harm, with causes attributed more to treatment/procedure and medication than in the general sample (Figure 4). Finally, 18 cases of awareness events were identified (some with pain, and some without), a further seven showed feeling of the operation under regional anaesthetic, 19 suggested inadequate analgesia during surgery, and 10 incidents describe near-miss situations, where pain or awareness were only narrowly avoided. One case describes awareness of a procedure even though no procedure was performed. 42 incidents were associated with failures in the treatment or procedure (Figure 5), with seven equipment failures, four process failures and three medication failures.

Discussion

The NRLS was easily searchable using simple database and spreadsheet tools, and the data set allowed a structured approach to understanding the wide range of systemic problems faced by practitioners in ensuring the safety of their patients. As expected, most adverse events reflected a failure of multiple components within the system, rather than suggesting errors by insufficiently qualified or negligent individuals. The generally low rate of harm in this data reinforces the view that healthcare practitioners play the key role in preventing systemic weaknesses from harming the patient, and that there is a growing culture of safety reporting. However, methods of analysis and quality assurance in voluntary systems mean that the data must be used with caution. There is a bias in who reports, what is reported, and the level of harm that is reported (Kreckler et al 2009). Serious incidents usually suggest a sequence of events, and it can be difficult to examine multiple causes, or chains of causation, thus hiding the true diversity of causes of these incidents.

Nothwithstanding such classification or hindsight bias (Woods & Cook 1999, Tamuz et al 2002) failures in treatment and procedure appear to be the most frequent and are among the highest levels of harm. Process failures also appear to be both frequent, and potentially harmful. Medication failures can be harmful and are moderately frequent, but rarely result in severe harm or death. Less harmful, though not always infrequent are consent and documentation failures, with equipment and infrastructure failures both relatively infrequent, and rarely associated with harm. This is supported by other studies (Fasting & Gisvold 2002, Goldhill & Waldmann 2006) and has been replicated in a similar assessment of the NHS-Litigation Authority (Clinical Negligence Scheme for Trusts) incident database (Catchpole et al 2006).

The key challenge in improving safety is to help practitioners bring about institutional change. Suggestions for future improvement are found in Table 3. Aspects under the control of the individual anaesthetist such as new techniques, equipment or improving technical skills might be of benefit, but bring new classes of problem and need to be considered with care (Cook & Woods 1996. Steffek et al 2004). The role of the practitioner as the last line of defence suggests that there may be a particular benefit in improving understanding of why errors occur, the recognition of potentially harmful situations, and the anticipation of such events in advance. Such 'non-technical' skills are rarely trained outside crisis management, even though they have been recognised as being important in anaesthesia for some time (Schaefer et al 1995, Fletcher et al 2000, 2003).

Recent advances

Since the completion and original publication of this study, there has been substantial progress in addressing front-line safety issues. The NRLS itself now contains well over 2 million incidents. In part as a consequence of these studies, the Royal College of Anaesthetists has now taken ownership of the reporting process, to allow better analysis, feedback, and empowerment by the specialty. This will also lead to better rates of reporting, improved use of incident reports and more focused assessments of specific safety issues.

Two complementary models of causation have been proposed both of which emphasise the critical role of the situation in which clinicians find themselves. The first model, popularly known as the 'Swiss Cheese' model (Reason 1990, 1997) emphasises the importance of providing a supporting system that recognizes and reduces the many chances of causing injury, rather than relying wholly on individuals to avoid or rescue. The second model is the escalation or 'Snowball' model (Helmreich 2000, Helmreich & Musson 2000, Catchpole et al 2005, 2007, 2008b), which describes how serious harm can arise from the accumulation of small, otherwise innocuous and unconnected sequences of events. The more problems there are, the greater the erosion of spare human capacity, and the more difficult it becomes to analyse and control the situation. Performance is degraded, thus increasing the chances of further harmful errors. The 'Snowball' model provides a model of accelerating risk, where the chances of patient injury can increase or decrease with moment-to-moment changes. Together, these models help us to understand more accurately why errors happen in the operating theatre (Catchpole et al 2008c, Healey et al 2008).

There is also increased awareness of the importance of safety management (Healthcare Commission 2009) and in the uptake of safety practices by clinicians, and the evidence of their efficacy (Haynes et al 2009). The use of safety cultural surveys is increasing, and the success of care bundles (Resar et al 2005), checklists (Haynes et al 2009), human factors non-technical safety training (Moorthy et al 2006, Haller et al 2008, McCulloch et al 2009), and new models of service safety and quality improvement (Young & McClean 2008) are informing and changing approaches to safety. We are also learning the value of feedback, empowerment of front-line staff, and the need to integrate prospective process and retrospective outcome data to understand and address systemic weaknesses.

Conclusions

There is no universal panacea for improving safety in anaesthesia. Incident reporting systems are one method with which to learn, anticipate and avoid future problems, and the true value of these systems is only just beginning to be realised. However, what is most clear is the importance of front-line staff in responding to new and unanticipated problems, identifying areas of systemic weakness, taking ownership of the problems, and being empowered by management to address and resolve them. The application of human factors knowledge--both through the training of generic teamwork skills and through specific expertise in systems design and evaluation--will be central to these future developments.

KEYWORDS Safety / Error / Incident reporting / Teamwork / Anaesthesia

Acknowledgements

The authors would like to thank all those who have taken the time to report these incidents, and the NPSA for funding and supporting this work. In particular, we would like to thank Sara Johnson (NPSA), Dominic Bell (Leeds Teaching Hospitals and NPSA), and Mark Boult (Det Norske Veritas) for their contributions to the earlier study. Dr Catchpole is gratefully supported by a Leverhulme Trust early career fellowship.

References

Candido KD, Stevens RA 2003 Post-dural puncture headache: pathophysiology, prevention and treatment Best Practice and Research Clinical Anaesthesiology 17 (3) 451-469

Catchpole K, Godden PJ, Giddings AEB et al 2005 Identifying and reducing errors in the operating theatre Patient Safety Research Programme Available at http://pcpoh.bham.ac.uk/publichealth /psrp/publications.htm

Catchpole K, Bell MD, Johnson S, Boult M 2006 Reviewing the evidence of patient safety incidents in anaesthetics Report to the National Patient Safety Agency London, NPSA

Catchpole KR, Giddings AE, Wilkinson M, Hirst G, Dale T, de Leval MR 2007 Improving patient safety by identifying latent failures in successful operations Surgery 142 (1) 102-110

Catchpole K, Bell MD, Johnson S 2008a Safety in anaesthesia: a study of 12,606 reported incidents from the UK National Reporting and Learning System Anaesthesia 63 (4) 340-346

Catchpole KR, Giddings AE, Hirst G, Dale T, Peek GJ, de Leval MR 2008b A method for measuring threats and errors in surgery Cognition, Technology and Work 10 (4) 295-304

Catchpole K, Mishra A, Handa A, McCulloch P 2008c Teamwork and error in the operating room: analysis of skills and roles Annals of Surgery 247 (4) 699-706

Cook RI, Woods DD 1996 Adapting to new technology in the operating room Human Factors 38 (4) 593-613

Fasting S, Gisvold SE 2002 Equipment problems during anaesthesia--are they a quality problem? British Journal of Anaesthesia 89 (6) 825-831

de Filho GR, Gomes HP, da Fonseca MH, Hoffman JC, Pederneiras SG, Garcia JH 2002 Predictors of successful neuraxial block: a prospective study European Journal of Anaesthesiology 19 (6) 447-451

Fletcher GCL, Flin RH, McGeorge P 2000 Review of Human Factors Research in Anaesthesia The identification and measurement of anaesthetists' non-technical skills SCPMDE Project: RDNES/991/C Industrial Psychology Group, University of Aberdeen.

Fletcher GCL, McGeorge P, Flin RH, Glavin RJ, Maran NJ 2002 The role of non-technical skills in anaesthesia: a review of current literature British Journal of Anaesthesia 88 (3) 418-429

Fletcher GCL, Flin RH, Glavin RJ, Maran NJ, Patey R 2003 Anaesthetists' Non-Technical Skills (ANTS): Evaluation of a behavioural marker system British Journal of Anaesthesia 90 (5) 580-588

Goldhill D, Waldmann C 2006 Excellent anaesthesia needs patient preparation and postoperative support to influence outcome Current Opinions in Anaesthesiology 19 (2) 192-197

Haller G, Garnerin P, Morales MA et al 2008 Effect of crew resource management training in a multidisciplinary obstetrical setting International Journal for Quality in Health Care 20 (4) 254-263

Haynes AB, Weiser TG, Berry WR et al 2009 A surgical safety checklist to reduce morbidity and mortality in a global population New England Journal of Medicine 360 (5) 491-499

Healey A, Catchpole K, Yule S 2008 Enhancing surgical systems Cognition, Technology and Work 10 (4) 251-254

Healthcare Commission 2009 Investigation into Mid Staffordshire NHS Foundation Trust London, Commission for Healthcare Audit and Inspection

Helmreich RL 2000 On error management: lessons from aviation British Medical Journal 320 (7237) 781-785

Helmreich RL, Musson DM 2000 The University of Texas threat and error management model: components and examples British Medical Journal Web Site Available from: http://www.bmj.com/misc/bmj.320.7237.781

Hofer TP, Hayward RA 2002 Are bad outcomes from questionable clinical decisions preventable medical errors? A case of cascade iatrogenesis Annals of Internal Medicine 137 (5) 327-333

Horn J, Bell MD, Moss E 2004 Handover of responsibility for the anaesthetised patient--opinion and practice Anaesthesia 59 (7) 658-663

Kluger MT, Bullock MF 2002 Recovery room incidents: a review of 419 reports from the Anaesthetic Incident Monitoring Study (AIMS) Anaesthesia 57 (11) 1060-1066

Kreckler S, Catchpole K, McCulloch P, Handa A 2009 Factors influencing incident reporting in surgical care Quality and Safety in Health Care 18 (2) 116-120

Lennmarken C, Bildfors K, Enlund G, Samuelsson P, Sandin R 2002 Victims of awareness Acta Anaesthesiologica Scandinavica 46 (3) 229-231

Mayberry LJ, Clemmens D, De A 2002 Epidural analgesia side effects, co-interventions, and care of women during childbirth: a systematic review American Journal of Obstetrics and Gynecology 186 (5) S81-S93

McCulloch P, Mishra A, Handa A, Dale T, Hirst G, Catchpole K 2009 The effects of aviation-style non-technical skills training on technical performance and outcome in the operating theatre Quality and Safety in Health Care 18 (2) 109-115

Moorthy K, Munz Y, Forrest D et al 2006 Surgical crisis management skills training and assessment: a simulation[corrected]-based approach to enhancing operating room performance Annals of Surgery 244 (1) 139-147

Ormrod G, Casey D 2004 The educational preparation of nursing staff undertaking pre-assessment of surgical patients--a discussion of the issues Nurse Education Today 24 (4) 256-262

Phillips JM, Stedeford JC, Hartsilver E, Roberts C 2002 Epidural abscess complicating insertion of epidural catheters British Journal of Anaesthesia 89 (5) 778-782

Rai MR, Pandit JJ 2003 Day of surgery cancellations after nurse-led pre-assessment in an elective surgical centre: the first 2 years. Anaesthesia 58 (7) 692-699

Reason JT 1990 Human Error Cambridge, University Press

Reason J 1997 Managing the Risks of Organisational Accidents Aldershot, Ashgate

Resar R, Pronovost P, Haraden C, Simmonds T, Rainey T, Nolan T 2005 Using a bundle approach to improve ventilator care processes and reduce ventilator-associated pneumonia Joint Commission Journal on Quality and Patient Safety 31 (5) 243-248

Schaefer HG, Helmreich RL, Scheidegger D 1995 Safety in the operating theatre--part 1: interpersonal relationships and team performance Current Anaesthesia and Critical Care 6 48-53Sigalovsky N 2003 Awareness under general anesthesia. American Association of Nurse Anaesthetists Journal 71 (5) 373-379

Steffek M, Owczuk R, Szlyk-Augustyn M, Lasinska-Kowara M, Wujtewicz M 2004 Total spinal anaesthesia as a complication of local anaesthetic test-dose administration through an epidural catheter Acta Anaesthesiologica Scandinavica 48 (9) 1211-1213

Tamuz M, Thomas EJ, Franchois KE 2002 Defining and classifying medical error: lessons for patient safety reporting systems Quality and Safety in Health Care 13 13-20

Trim J, Fordyce F, Dua S 2003 Using clinical governance to standardise an epidural service Nursing Standard 18 (9) 43-45

Weinger MB, Englund CE 1990 Ergonomic and human factors affecting anesthetic vigilance and monitoring performance in the operating room environment Anesthesiology 73 (5) 995-1021

Woods D, Cook RI 1999 Hindsight biases and local rationality pp141-171 In: Durso FT et al eds Handbook of Applied Cognition Chichester, John Wiley and Sons Ltd

Young TP, McClean SI 2008 A critical look at lean thinking in healthcare Quality and Safety in Health Care 17 (5) 382-386

Ken Catchpole

BSc(Hons), PhD

Leverhulme Research Fellow, Quality, Reliability,

Safety and Teamwork Unit (QRSTU), Nuffield

Department of Surgery, University of Oxford

Peter McCulloch

FRCS

Reader in Surgery, Head of the Quality, Reliability, Safety and Teamwork Unit (QRSTU), Nuffield Department of Surgery, University of Oxford

Correspondence address: Ken Catchpole, Level 6, The John Radcliffe, Headington, Oxford, OX3 9DU. Email: ken.catchpole@nds.ox.ac.uk
Table 1: Incidents identified for analysis

Topic         Keyword Search    Hits    Comment

All           Anaest * OR       12649   All incidents
reported      anest *                   considered
anaesthetic
incidents     Anaes * AND       1155    828 incidents selected
              assess *                  for further analysis

Pre-surgery   Anaes * AND        110    43 incidents selected
assessment    unexpected                for further analysis

              Anaes * AND        187    161 relevant incidents
              Allergy

Epidural      Epidural          2009    516 selected for
                                        further analysis

incidents     Epidural AND       30     27 selected far
              headache                  further analysis

              Epidural AND       16     8 selected for further
              puncture                  analysis

              Epidural AND        1     Suitable for analysis
              back ache

              Epidural AND        2     1 suitable; 1 added
              abscess                   from analysis
                                        elsewhere

              Anaesthetic        59     Pressure sores not
              AND sore           40     considered any
              Epidural                  further.
              AND sore

Awareness     Anaes * AND        59     No relevant cases
during        aware *                   18 relevant
anaesthesia                             incidents; 40
              Anaes * AND        609    other incidents
              pain                      of interest

              Anaes * AND         2     No relevant
              awake                     incidents

              Anaes * AND       None    NBA
              psycho *

              Anaes * AND       None
              PTSD

              Anaes * AND        12     4 partially
              traumatic                 relevant
                                        incidents

Table 2: Probability of outcome given incident cause

                            Total    No      Low or     Severe
                            Number   Harm   Moderate   or Death

Process                      1067    69%      28%         4%
Clinical assessment          524     76%      21%         3%
Documentation/Consent        2273    87%      13%         1%
Medication                   1120    79%      20%         1%
Infrastructure/ Equipment    2791    83%      15%         1%
Treatment/ Procedure         3856    67%      30%         3%
Other                        1018    64%      33%         4%

Table 3: Areas to investigate for improvement

Incident Type   Potential improvement

Pre-Surgery     Better defined range of information
                (Kluger & Bullock 2002)

Assessment      Better elicitation of information

Incidents       Maintain flexibility for patient
                needs

                Standardised & comprehensive
                documentation

                * Improved design of consent form
                Methods for accurate & timely
                communication

                * Standardised handovers (Horn et
                al 2004)

                * Pre-operative briefings (Hofer &
                Hayward 2002)

                Better defined competencies for
                conducting assessment (Rai & Pandit
                2003, Ormrod & Casey 2004)

Epidural        Minimise risk of infection
Incidents       (Phillips et al 2002)

                * Use of care bundles (REF)

                Consider use of ultrasound (REF)

                Standardise practices (Trim et al
                2003)

                Consider use of test dose (Steffek
                et al 2004)

                Consider use of small-gauge needles
                (Candido & Stevens 2003)

                Ensure optimal patient positioning
                (de Filho et al 2002)

                Better understanding of
                expertise/experience (de Filho et
                al 2002)

                Better understanding of
                organisational factors (Mayberry &
                Clemmens 2002)

                Offer of blood patch to high-risk
                patients (Candido & Stevens 2003)

Awareness       Recognise and respond early
Incidents       (Lennmarken et al 2002)

                * Psychiatric assessment

                * Treatment

                * Long-term follow up

                Consider equipment techniques for
                identification (Sigalovsky 2003)

                Maintain vigilance &reduce
                distraction (Weinger & Englund
                1990)

                Non-technical skills training
                (Fletcher et al 2002)

                Better classification of awareness
                incidents

                Better grading for severity

                Timely root cause analysis

Figure 1: Severity and type of all anaesthetic incidents

Severe,                       181, 1%
Death,                         88, 1%
No Harm,                    9531, 75%
Low,                        1977, 16%
Moderate,                     872, 7%

Process,                     1067, 8%
Clinical assessment,          524, 4%
Documentation/Consent,      2273, 18%
Medication,                  1120, 9%
Infrastructure/Equipment,   2791, 22%
Treatment/Procedure,        3856, 31%
Other,                      1018,  8%

Note: Table made from pie chart.

Figure 3: Causes of assessment failures

Allergy incidents

Other                          2%
Process                        4%
Assessment                     1%
Documentation/Communication   49%
Medication                    19%
Infrastructure/Equipment       0%
Treatment/Procedure           25%

Unexpected incidents

Process                        5%
Assessment                     2%
Documentation/Communication    5%
Infrastructure Equipment       0%
Medication                     0%
Treatment/Procedure           62%
Other                         26%

All Assessment incidents

Other                          2%
Process                        8%
Assessment                    54%
Documentation/Communication   12%
Medication                     2%
Infrastructure/Equipment       4%
Treatment/Procedure           18%

Note: Table made from pie chart.

Figure 4: Severity and causes of epidural incidents

Death,                        2, 0%
No Harm,                   340, 66%
Low,                        94, 18%
Moderate,                   71, 14%
Severe,                       9, 2%
Process,                    50, 10%

Clinical Assessment,         11, 2%
Documentation/Consent,      53, 10%
Medication                  83, 16%
Infrastructure/Equipment,   83, 16%
Treatment/Procedure,       202, 39%
Other,                       34, 7%

Note: Table made from pie chart.

Figure 5: Causes of awareness during anaesthesia

Other,                         0, 0%
Process,                       4, 7%
Assessment,                    1, 2%
Documentation/Communication,   1, 2%
Medication,                    3, 5%
Infrastructure/Equipment.     7, 12%
Treatment/Procedure,         42, 72%

Note: Table made from pie chart.
Gale Copyright: Copyright 2009 Gale, Cengage Learning. All rights reserved.