Dealing with endoscopic disaster

The results of the National Colonoscopy Audit, presented at the British Society of Gastroenterology’s plenary session, highlighted the fact that there has been significant improvement in recent years.1 In 2004, a study by Bowles et al reported disappointing results with poor caecal intubation rates and higher than expected complication rates.2 However, since this study was undertaken, there has been increased investment in endoscopic training, a quality assurance framework for endoscopy units has been established, and the National Bowel Screening Programme has been implemented. Conducted over a two-week period in 2011, the first national audit of colonoscopy practice included data on over 20,000 colonoscopies and 2,681 colonoscopists, from over 300 units. The overall caecal intubation rate was reported to be 92.3%. However, when adjusted for impassable strictures and poor bowel preparation, the rate was 95.8%. The polyp detection rate was 32.1%, 11.7% for significant polyps (>1 cm diameter), while some 92.3% of resected polyps were retrieved.1 The audit also found that 90.2% of procedures achieved acceptable levels of patient comfort. With regards to incidents, a total of eight perforations and 52 significant haemorrhages were reported – with eight patients undergoing surgery as a consequence of a complication.1 The authors concluded that there has been an improvement in performance, since the initial study carried out by Bowles et al, which reported a caecal intubation rate of 76.9% compared to 95.8%.

 Preventing safety incidents

While these latest results offer some cause for celebration, speakers also identified concerns over continuing safety incidents – a theme that was addressed within the BSG Endoscopy Section Symposium: Dealing with endoscopic disasters. The safety of endoscopy was discussed by Dr Manmeet Matharoo – a research fellow at The Wolfson Unit for Endoscopy at St. Mark’s Hospital affiliated with The Centre for Patient Safety and Service Quality (CPSSQ) Imperial College London. She presented the results of an evaluation of patient safety incidents and ‘Never Events’, based on observations carried out in a single tertiary endoscopy unit.3 On the subject of patient safety, Dr Matharoo commented: “We know that medical errors are common – Charles Vincent’s influential paper cites 10% of patients receiving medical healthcare within the NHS suffer an adverse event, based on a review of over 1,000 patient records. Up to half of these were deemed to be avoidable, while a third led to significant disability or death. An extra one billion pounds per year is spent on additional days in hospital alone, as a direct result of safety errors, but how does this relate to endoscopy? “The National Confidential Enquiry into Patient Outcome and Death (NCEPOD) looked at 30-day mortality data. Many of the recommendations were focussed on non-technical aspects, rather than technical expertise – such as monitoring and sedation. With the expansion of bowel cancer screening, we know that an ever increasing proportion of elderly patients will be undergoing therapeutic procedures with all the risks that this entails.” Great strides have been made to improve the quality of endoscopy services, Dr Matharoo pointed out. However, she added that it is important to understand the multifaceted reasons why patient safety incidents continue to occur. A significant proportion of adverse events are deemed to be preventable and often arise from multiple systems failures as described in Reason’s ‘Swiss Cheese’ model.4 “The ‘holes’ within the cheese represent ‘latent failures’ and, when the holes ‘line up’, these latent failures can lead to a major incident,” explained Dr Matharoo. She cited an example of where poor communication occurred within an endoscopy unit, with regards to changes to the list: “This in itself could have been viewed as inconsequential; however, it was coupled with an inadequate patient identification check within the waiting area. This led to further teamwork failures in the endoscopy room, where the patient’s identity, once again, was not checked. The endoscopist failed to review all of the clinical information, which led to a patient being sedated unnecessarily for the wrong procedure.” The aim, according to Dr Matharoo, is to ‘build defences’ to prevent the ‘holes’ from lining up and, ultimately, culminating in a serious incident. She went on to explain that the Department of Health’s Never Events are defined as ‘serious but preventable patient safety incidents’. In 2009, there were 111 such events, including patient misidentification and wrong site surgery. The list has now been expanded from 8 to 25 Never Events. With regards to this list, she pointed out that ‘patient misidentification’ is relevant to endoscopy. In addition, while ‘wrong site surgery’ may not appear relevant, it also applies to endoscopic procedures. Overdose of midazolam during conscious sedation, along with failure to monitor/respond to oxygen saturation, are also among the new Never Events – both of which are highly relevant to endoscopy. To further improve quality and safety, Dr Matharoo and her colleagues identified the need for structured analysis to determine the type, severity and rate of safety incidents. This is particularly important in view of the NHS National Bowel Cancer Screening Programme, which means that increasing numbers of asymptomatic individuals will undergo endoscopic procedures. For the purposes of the study, carried out at a single tertiary endoscopy unit, patient safety incidents were defined as ‘Never Events’, ‘adverse events’ or ‘near misses’. The incidents were categorised according to severity (on a scale of 1 to 3), the type of incident, and whether they were ‘clinical’, ‘process’ or ‘human’ errors. These were agreed by expert consensus. Ninety procedures, by 16 endoscopists, over 22 lists, were evaluated and 41 patient safety incidents were identified. This equated to a rate of 45%. Half of these were categorised as severe safety incidents, while nearly a quarter (24%) had the potential to be ‘Never Events’. “Interestingly, during this study period, there was only one clinical incident form completed, so clearly there is a discrepancy between the observed patient safety incidents and those that were reported,” Dr Matharoo commented. “Unless these incidents are actively sought out, they will go unnoticed. Needless to say, this represents the tip of the iceberg.” Some examples of safety incidents observed included: patient misidentification (a patient with a similar sounding name stood up when called for in the waiting room); the incorrect procedure was performed (a colonoscopy was attempted instead of flexible sigmoidoscopy, with sedation and an inappropriate consent form); a sedated patient was left in a corridor unmonitored; sedation was administered to a patient with obstructive sleep apnoea without an oxygen saturation monitor; polypectomy was carried out with no IV access. “Although there was no actual patient harm observed within the study, we know that multiple small errors can lead to endoscopic disaster. In conclusion, we found that errors occur frequently. Moreover, they occur frequently in a high-end, high-volume screening unit. Patient safety incidents can be with or without consequence. However, unless there is an immediate consequence, these safety incidents often go unnoticed. Near misses are our golden opportunity to intervene. We should not wait for the disaster to happen – these incidents represent a latent risk and must be addressed.” Dr Matharoo concluded that the focus for adverse events should shift from that of ‘reporting’ to ‘understanding’ the multifaceted reasons why a patient safety incident occurred, as well as prevention. “We are now focussing on root cause analysis and are currently investigating and validating whether these errors can be averted, by using an ‘endoscopy safety checklist’,” she revealed. “The aim is to build on the success of the World Health Organization’s surgical safety checklist.”