‘Case of the month’: a novel way to learn from endoscopy-related patient safety incidents

Abstract

Patient safety incidents (PSIs) are unintended or unexpected incidents which can or do lead to patient harm. The Joint Advisory Group on Gastrointestinal Endoscopy (JAG) acknowledges that PSIs should be reviewed by endoscopy services and learning shared among staff. It is recognised that more could be done to promote shared learning as outlined by the JAG ‘Improving Safety and Reducing Error in Endoscopy’ strategy. The ‘Case of the month’ series aims to provide a broad selection of cases and subsequent learning that can be shared among services and their workforce. This review focuses on five case vignettes that highlight a variety of PSIs in endoscopy. A structured approach, based on incident analysis methodology, is applied to each case to categorise PSIs and develop learning points. Learning is directed toward the individual, team and healthcare organisation. A selection of methods to disseminate learning at local, regional and national levels are also described.

Key points.

• Patient safety incidents can be a rich source of learning for all endoscopy staff.

• Incidents do not necessarily need to be serious to be useful: learning can be derived even from minor incidents.

• A systematic approach to shared learning involves case review, identification and categorisation of patient safety incidents.

• Learning points can be directed at an individual, team or service.

• The ‘Case of the month’ series allows dissemination of learning at local, regional and national levels.

Introduction

Patient safety incidents (PSIs) can be defined as any ‘unintended or unexpected incident which could have, or did, lead to harm for one or more patients receiving healthcare’.¹ Incident reporting systems in healthcare exist to identify the causative factors behind PSIs. Governance structures should be in place to share and disseminate learning from incident analyses with the aim of preventing future harm, so-called ‘learning from error’.² It is thought that by closing the safety-feedback loop in this way, safety awareness is promoted contributing to improved patient safety.^3–5

Endoscopy services are expected to review and demonstrate learning from outcomes of adverse events, as outlined in the safety domain of the Joint Advisory Group on Gastrointestinal Endoscopy (JAG) clinical quality standards.⁶ There is a vast breadth and depth of learning from such data review, encompassing the entire patient pathway.⁷ Learning is typically disseminated at a service level, for example through governance or endoscopy user group meetings. However, learning is often not shared beyond an institution.

Recognising this gap for potentiating learning, the JAG ‘Improving Safety and Reducing Error in Endoscopy’ (ISREE) working group have developed the ‘Case of the month’ series. Hosted on the JAG website, this initiative allows users to submit, engage and reflect on safety incidents in endoscopy. This draws from the experience gained from enterprises across other specialties. The most widely known is ‘Radiology Events And Learning’ (REAL), a dedicated newsletter highlighting learning from incidents in radiology.⁸ REAL has expanded as an educational strategy through development of core standards and regular physical meetings. Initiatives like this are thought to increase engagement with safety as a concept, promoting a shift toward a more positive safety culture.⁵

The aim of ‘Case of the month’ is to improve learning from safety-related incidents for all individuals across endoscopy services. This is a novel way of publicising common or important PSIs for wider learning. In this article, we will describe five case vignettes from endoscopy, adapted from real-life scenarios, identifying issues and outlining the ways learning can be achieved through a systematic approach of case analysis.

Case vignettes

The cases that follow have been shortened for purposes of this article. All case vignettes are informed by real-life cases but have been fully anonymised and provide no identifiable information. Case summaries have been adapted for the purposes of this review article and real case data may have been amalgamated to bring out additional learning points.

Case 1: ‘Going down the rabbit hole’

Complex patients were booked onto a consultant’s list which a trainee was attending. The list was overbooked. The morning list was overrunning and the start of the list was delayed. The trainee endoscopist was not signed off for independent colonoscopy and the consultant was delayed. The trainee said he would start the list.

After the first patient was brought into the room, the nurses noted the consent form had not been completed. The trainee completed the consent in the room, verbally expressing frustration this had not already been done. Sedation was administered prior to the oxygen saturations probe being placed or oxygen being administered.

The first patient was known to have polyps but the procedure had been listed as a diagnostic colonoscopy. The insertion was challenging with suboptimal bowel preparation, complex looping and discomfort to the patient. Sedation was supplemented without checking the oxygen saturations.

The trainee’s communication with the nurses and the patient diminished. The nurses recognised the trainee was in difficulty and indirectly proposed solutions but these suggestions were met with silence, followed by hostility. The trainee was distracted during the procedure as his mobile phone rang on multiple occasions (the trainee was also on call). The trainee was uncertain about caecal landmarks but recorded adequate caecal views had been obtained. The withdrawal time was rapid and a polypectomy was attempted and then abandoned. The trainee was unable to communicate the outcome of the procedure to the patient due to time pressures.

Case 2: ‘Mind the patch’

A patient with metastatic oesophageal cancer and a palliative oesophageal stent presented with a recurrent food bolus obstruction.

The patient was cachectic, lived independently at home, had an existing ‘Do Not Attempt Cardiopulmonary Resuscitation’ decision and was already known to palliative care services. As the patient was tolerating saliva, a decision was made to perform an oesophago-gastroduodenoscopy (OGD) under sedation instead of general anaesthetic in light of their comorbidities. Informed consent was obtained with the risk of aspiration explained to the patient.

The procedure was performed in the X-ray screening room. Prior to the procedure, Xylocaine throat spray, fentanyl 25 μg and midazolam 1 mg were administered.

Within 1 min of administrating the drugs, there was a rapid desaturation to 60%. High-flow oxygen, naloxone and flumazenil were administered. There was a short delay in administration of reversal agents as while the drugs were available, the needles and syringes required were unavailable in the procedure room. Additionally, when moving the patient forwards to examine their chest, a buprenorphine patch was identified on the patient’s back.

The patient was diagnosed clinically with left-sided aspiration and investigations later confirmed aspiration pneumonia. The patient was managed supportively and made a good recovery.

Case 3: ‘Forget me not’

A patient underwent an OGD for persistent reflux, which was performed by a senior gastroenterology trainee. Endoscopic intubation was poorly tolerated. The trainee was uncertain of the endoscopic view and called for a consultant’s advice who was concerned about a high oesophageal perforation. The procedure was abandoned. A CT scan confirmed a perforation. The patient was transferred for surgical repair to another centre and after a protracted recovery, left hospital several weeks later.

A full investigation was carried out and outcomes discussed at the local governance meeting. The trainee was fully certified for independent practice and the correct action had been taken following the complication. However, the patient made a formal complaint raising the following concerns:

1. Why was a ‘trainee’ performing a procedure unsupervised?

2. Why were they not consented for a serious complication?

3. Why had they not received any communication following transfer?

The investigational findings were shared with the patient and they were invited to discuss them with the clinical team. The consent form clearly documented risks but no copy had been given to the patient. Although a letter had been sent to the general practitioner (GP), the patient had not been written to directly.

Case 4: ‘Pass the baton’

A patient was referred for endoscopic retrograde cholangiopancreatography (ERCP) following an episode of jaundice and an ultrasound scan which demonstrated a dilated common bile duct with a 16 mm stone.

The procedure was performed under conscious sedation. The main bile duct was grossly dilated but no stone was seen. A sphincterotomy was performed, with a plan to conduct a balloon trawl. A larger balloon (18 mm) was required to make sure small stones were captured. (An 18 mm balloon can be inflated to two measurements: 15 mm or 18 mm. The volume of an 18 mm balloon is almost twice than that of the 15 mm balloon. There is a ‘mark’ on the air syringe attached to the balloon, up to which the endoscopy assistant pushes the plunger to achieve 15 mm inflation.)

The endoscopist placed the balloon in a narrow part of the bile duct and asked the endoscopy assistant, who had just swapped into the procedure due to the arrangement of breaks, to inflate it ‘to the mark’. The balloon was inflated all the way, reaching 18 mm in diameter. There was an immediate suspicion that damage may have been done to the bile duct, and injection of contrast revealed a leak. A stent was inserted to encourage bile to drain down through the liver. A CT scan was performed on the same day and appeared to confirm a bile duct injury.

The patient was informed of the incident on the day, an apology was given and the incident was reported. The patient recovered and was discharged after a few days. A follow-up ERCP was performed and the bile duct injury had completely healed.

Case 5: ‘Wrong site endoscopy’

A female patient was referred by her GP with rectal bleeding and a routine outpatient flexible sigmoidoscopy was arranged. The procedure was performed by a new consultant endoscopist.

Following a digital rectal examination, the patient was noted to have anal spasm. The endoscopist waited for the anal tone to normalise and proceeded to insert the scope. As was her normal practice, the endoscopist froze the endoscopic image, so that the patient would not see their own anus. When the image was made live, the anatomy was not correctly identified by endoscopist or the attending nurses: the scope had been mistakenly introduced into the patient’s vagina. The patient’s cervix was visualised on the endoscopy screen, which the consultant assumed to be a rectal mass. She proceeded to take biopsies.

There had been difficulty in obtaining biopsies due to the texture of the tissue and the endoscopist questioned whether the scope was in the colon. She checked the patient’s perineum and realised that the scope was in the vagina and removed it. She placed the scope into the rectum and proceeded to perform the sigmoidoscopy which was unremarkable. At the end of the procedure, the endoscopist requested the nursing staff to discard the cervical biopsies but then changed her mind and retrieved the biopsies from the sharps bin.

Following the procedure, in recovery, the endoscopist gave the patient a full explanation of the error and an apology. She entered a report onto the incident reporting system following advice from the nursing staff. The patient was discharged home after recovery without any further complication. The incident was reported as a ‘never event’ to National Health Service (NHS) England and through the trust’s internal system.

A systematic approach to sharing learning

Each of the cases submitted from hospitals across the UK illustrate various details with distinct safety issues raised. Therefore, it is important to have a framework for reviewing cases for the purposes of developing learning.

The ‘framework for analysing risk and safety’ forms the basis of root cause analysis (RCA) and provides the traditional model of analysing safety incidents.⁹ The RCA approach is often time-intensive and labour-intensive, and learning points may not necessarily be fed back purposively to front-line staff.^{10 11} In fact, poor feedback to staff is often cited as a reason that some incidents are not reported in the first instance.¹² Additionally, an RCA is usually reserved for the most serious of incidents and may not be suitable for some of the cases described.

For all incident types, a more concise methods of PSI analysis is likely better suited to development of learning points.¹³ JAG have outlined a systematic approach to sharing learning based on principles from these methods of safety analysis. The aim is to create a set of learning points directed at staff and services that are actionable and readily applicable. This process is not meant to replace current governance structures, but rather complement them. The individual steps are summarised in figure 1.

Figure 1.

Systematic approach to reviewing cases, identifying patient safety incidents and developing learning points.

PSI identification and categorisation

Having summarised each case, the next step is to identify PSIs. This is achieved with use of a framework to categorise incidents. Within the UK, the most commonly used framework for PSI categorisation is defined by the National Reporting and Learning System (NRLS)¹⁴ (see table 1). Categories are designed to encompass a variety of healthcare settings, allowing a comparable overview of incidents. However, there can be a loss of specificity for individual specialities. For the purposes of developing targeted learning points, we have used an endoscopy-specific PSI categorisation system, informed by a previous observational study (see table 1).¹⁵

Table 1.

Patient safety incident (PSI) categories defined by National Reporting and Learning System (NRLS) and endoscopy-specific classifications (in alphabetical order)

NRLS PSI categories	Endoscopy PSI categories
Access/admission/transfer/discharge Clinical assessment Consent, communication, confidentiality Documentation Implementation of care and ongoing monitoring/review Infrastructure Medical device/equipment Medication Patient accident Treatment/procedure Other	Consent Distractors and time management Documentation and reporting errors Drug errors Histology and sampling errors Non-technical skills and training Oxygen monitoring Sedation, intravenous access and monitoring Technical skills and equipment Other

For the cases described, PSIs were categorised using this framework. PSIs were assessed for severity based on the actual or potential impact to the patient and adherence to established clinical guidance, as per previous methodologies,¹⁵ and informed by NRLS guidance.¹⁶ Severity is defined as minor, moderate or severe based on these criteria.

Figure 2 shows the categorised PSIs in each case highlighted by a severity rating.

Figure 2.

Patient safety incidents (PSIs) identified in each case and categorised severity. Colour key: Yellow=minor, orange=moderate, red=severe (where multiple PSIs within category were identified, the most severe grading is shown).

It is important to understand the relevance of PSIs to each case in more detail. Figures 3 and 4 demonstrate the interplay between individual PSIs and the overall effect on the care of the patient.

Figure 3.

Incident timelines (cases 1 and 2)—visual representation of patient safety incidents and relationship to patient encounter.

Figure 4.

Incident timelines (cases 3, 4 and 5)—visual representation of patient safety incidents and relationship to patient encounter. OGD, oesophago-gastroduodenoscopy.

Learning points

Learning points are derived from a review of the case alongside the identified PSIs. Specific actionable learning points are derived from each case, designed to educate staff and inform service provision. Where appropriate, links to relevant literature or guidance are provided. For the purposes of this review, the learning points from all five cases have been collated and categorised into individual, team and service-level points for ease of dissemination. These are displayed in table 2.

Table 2.

Learning points from cases, split by individual, team and service-specific issues

	Individual	Team	Service
Case 1	Trainee endoscopists should not perform procedures unsupervised if they are not JAG certified—this puts the patient and trainee at increased risk of an adverse event. Recognising when a task is beyond your skill set and acting on it is key to patient safety.	Having good situational awareness and effective communication with the nursing team may prevent ‘tunnel vision’ at stressful times and may avert a crisis. Nursing staff should be empowered to challenge endoscopists if they feel patient safety is compromised and seek further support if necessary. Using checklists enhances the likelihood of adhering to basic patient safety measures particularly at times of stress when cognitive workload is increased. A team briefing would have identified the overbooked complex list and alternative measures put into place proactively.	System errors are prevalent in the organisation of procedures and the clinician is often the inheritor of risk. An overbooked list and parallel on-call commitment should be escalated with the endoscopy clinical lead.
Case 2	If opiate and benzodiazepine sedation are used together should be at least a 1 min gap between administration. Aspiration risk increases with concomitant use of lignocaine throat spray and intravenous sedatives. Consider avoiding this in high-risk patients. Consider single-agent intravenous sedation in elderly and cachectic patients. Combination of fentanyl and midazolam even in low doses may cause profound respiratory depression and sedation in these patients.	Oesophageal obstruction is associated with a high risk of aspiration. Seek anaesthetic review and support even if patients may not be for resuscitation or intubation. Careful consent for high-risk procedures and/or patients with advanced comorbidity. If possible, include family and next of kin in discussions. Equipment, drug stock and administration devices should be checked prior to every case. Careful review of the drug chart should be performed prior to endoscopy and form part of the team briefing/checklist.	Reversal agents with the relevant drawing up equipment should be kept together and be easily accessible.
Case 3	The term ‘trainee’ maybe confusing and misleading to patients and we ought to consider other terms in view of competence.	Documentation by the team involved is crucial following an adverse event. Those involved should write contemporaneous notes so that if there is an investigation or a complaint at a later date there is a full record of what happened.	Following an adverse event or complication, communication with the patient should continue until the issue is fully resolved. The patient should be invited to discuss what happened as soon as they have recovered. Patients should be offered a copy of their consent form to keep a record of benefits, risks and discussions around a procedure.
Case 4	The new assistant may not have understood the instruction to inflate to the desired limit as a result of ineffective handover and/or inexperience with equipment. It is good practice for the endoscopy assistant to repeat instructions back to the endoscopist for confirmation. This reflects the concept of ‘closed-loop communication’ where the ‘sender’ of messages ensures that they are conveyed and understood by ‘recipients’.	Team focus can be impaired by staff swapping into procedures or handing over while it is in process. Ideally, staff should not swap during advanced, therapeutic endoscopic procedures. If it is necessary, the endoscopist should be warned at an appropriate moment, and asked to confirm that they are happy for this to go ahead. The incoming assistant must have time to be briefed on the stage of the procedure and what the therapeutic intention is.	Two-stage balloons may carry an inherent risk and services should make enquiries into ordering balloons with fixed diameter syringes so that accidental overinflation cannot occur.
Case 5	Following rectal intubation, the perineum should be inspected to ensure that the correct orifice has been used prior to further progress being made and/or biopsies being taken. Avoid the practice of freezing the endoscopic image while the scope is being inserted. Use direct endoscopic views as well as direct perineal examination to guide insertion. Retrieving discarded items from a sharps bin should never be undertaken and is a serious hazard.	Use the expertise of the endoscopy nurses if an unusual view is obtained. Ensuring the procedure room is well lit prior to the insertion of the endoscope, to ensure the perineum is visualised. Maintain clear communication between team members in relation to the management of biopsies. Use the team brief at the start of the list to familiarise team members with new endoscopists.	Prompt and effective communication is necessary at the time of an incident, including full duty of candour, as was done in this case. There should be appropriate escalation and follow-up by senior clinicians of relevant specialties with subsequent reassurance if no lasting harm has been caused by the incident.

JAG, Joint Advisory Group on Gastrointestinal Endoscopy.

Although each case provides unique learning points, there are three themes that are common to most cases, described as follows.

1. Preprocedural planning

Many of the PSIs identified could have potentially been avoided through adequate planning and team briefing. Checklists, together with team briefings, form part of local safety standards for invasive procedures, based on national safety standards for invasive procedures.¹⁷ The utilisation of checklists is vital in maximising recognition of potential errors preprocedurally.¹⁸ Additionally, they are a useful adjunct in preprocedural planning, for example, patient-specific sedation planning.¹⁹

Implementation of the WHO surgical safety checklist has seen improvements in patient outcomes. However, incidents can still occur despite checklist use. The concept of ‘checklist fatigue’, where checklist use invokes an apathy and lack of team engagement, may contribute to this. Constant training and feedback should aim to reinforce checklist engagement promoting effective use.²⁰

2. Non-technical skills (NTS, individual and team)

NTS comprise skills including communication, situational awareness, judgement, decision-making and leadership.²¹ These skills are crucial to the functioning of endoscopy teams.²² Historically, breakdown in NTS has been implicated in complications and adverse patient outcomes in endoscopy.²³ There is now an increasing understanding of the effect of team-based NTS on patient outcomes. NTS training appears to have a beneficial impact on team performance and patient outcomes.^{24 25}

The endoscopic non-technical skills framework was developed to provide a guide for the training and assessment of endoscopists.²⁶ However, there should be an increasing drive to develop team-based NTS training through modalities such as simulation.²⁷

3. Patient communication and managing expectations

Lack of communication, documentation and poor consent may contribute to lack of patient understanding. Open communication and maintaining duty of candour is important throughout the patient journey.²⁸ This may help to alleviate fears and manage expectations, preventing misunderstanding or complaint further down the line.

The patient is an active observer of their care and patient feedback, including complaints, is important to incorporate into service improvement. This reflects the notion of a ‘just safety culture’ where the patient voice is listened to and valued.²⁹ This is also mirrored in the recent Independent Medicines and Medical Devices Safety Review which highlights the need for a ‘patient safety commissioner’—an independent leader that ‘champions the value of listening to patients’.³⁰

Disseminating learning

A core message of the ISREE strategy is sharing learning. A recent endoscopy safety workshop conducted by JAG identified the ways in which this can be achieved in a practical sense (see figure 5). These items were derived from focused discussions involving endoscopists, nurses, trainees and patient safety/human factors experts.

Figure 5.

How individuals and services can share learning from ‘Case of the month’.

Summary

PSIs can be a useful source of learning. Incident analysis can tease apart the complexity of cumulative error to gain a better understanding of safety. This can be translated into learning through a systematic approach with widespread dissemination. The JAG ‘Case of the month’ aims to share safety messages across endoscopy staff and services and promote a shared organisational memory. This should complement other safety initiatives underway, including improvement of incident reporting systems.^{31 32}

The NHS patient safety strategy highlights how reviewing incidents can be powerful in collective learning, but there should also be a focus on learning from positive experiences to prevent error.⁵ While it is recognised that this is an important future direction, it requires a shift toward development of a positive safety culture. To enable this, ongoing engagement from all levels of staff within endoscopy is required. As the ‘Case of the month’ series expands, we welcome contributions from a much wider audience. We hope this method of case review becomes the norm, complementing safe endoscopy for all.

For more information, please visit https://www.thejag.org.uk/ISREE

Ethics statements

Patient consent for publication

Not required.