Investigation report: A thematic analysis of HSIB’s first 22 national investigations

HSIB focuses on identifying problems within the systems and processes that determine how healthcare is delivered (Sampson et al, 2021). HSIB has been directed by the Department of Health and Social Care (2016a) to investigate safety risks that may include but are not limited to:

• risks resulting in repeated, preventable, or common occurrences of risks or harm to patients
• risks indicating a systemic problem with significant impact in more than one setting, or
• those involving new or novel forms of harm or new or novel risks of harm.

Although there are many patient safety challenges, it is important to be able to prioritise efforts to improve safety. This report provides an overview of the most significant threats to patient safety that HSIB has found so far. This analysis highlights the approaches needed to understand the different types of problems within the healthcare system.

HSIB has been directed to make safety recommendations that it considers appropriate for the purposes of addressing risks affecting patient safety at the system level, rather than focus on the activities of individuals or small teams (Department of Health and Social Care, 2016b).

The aim of investigating patient safety incidents is to learn from them by identifying new hazards and improving the management of known safety risks. The analysis of incidents also allows for the monitoring of safety performance, the evaluation of actions taken to improve safety, and the promotion of messages about safety interventions to healthcare staff. This report identifies how the healthcare system is able to support this type of learning from patient safety investigations

The NHS Patient Safety Strategy (NHS England and NHS Improvement, 2021) describes the need for a continuous improvement process that focuses on safer systems. It also identifies the crucial role that healthcare regulatory bodies play in safety management. Regulation is an important means of monitoring and improving healthcare with the aim of ensuring safe, reliable treatment for patients and a safe working environment for healthcare professionals (Oikonomou et al, 2019). The regulatory landscape is complex, and it is important to understand the different concepts and activities that are recognised as part of safety management (International Civil Aviation Organization, 2018).

This report describes safety management activities using the terms defined below; examples are provided to illustrate how these concepts relate to patient safety investigation.

• A hazard is something that has the potential to contribute to an undesirable patient safety outcome. A patient safety investigation can identify new hazards and develop an understanding of situations where these hazards are emerging.
  • For example, not being able to see the product label on a bag of intravenous fluid that is connected to a patient has the potential to contribute to a patient safety incident.
• A safety risk is the predicted probability and severity of the consequences of a hazard. Known safety risks have the potential to result in undesirable outcomes even when there are proposed ways of managing the safety risk.
  • For example, in 2008 the UK National Patient Safety Agency published a rapid response report (National Patient Safety Agency, 2008) after it identified 84 incidents, including two where patients died, where the wrong bag of fluid was connected to patients. In these cases, the product label was not visible because the bag of fluid had been placed inside another bag, known as a pressure bag, which obscured the product label. One of the report’s recommendations was that there should be a way of identifying the product in the bag of fluid, even when it has been placed inside a pressure bag. It was recommended that manufacturers should develop a universal system to address this problem.
  • Patient safety investigations can discover problems with the implementation of interventions. In the case of pressure bags, for example, manufacturers developed a sleeve to enable the product label to be viewed (Leslie et al, 2013). However, some of these sleeves used a meshed design that obscured the second line of text on the label, which was found to be critical for distinguishing between the products being used (Gupta and Cook, 2013).
  • There are multiple contributory factors that also need to be considered when managing a safety risk. In the pressure bag investigation, it was found that factors relating to the team, local working conditions, staff workload and distraction, policies, and communication systems all contributed to the safety risk (Gupta and Cook, 2013).
• A systemic safety risk is a risk that persists across the systems and processes that determine how healthcare is delivered. HSIB safety recommendations are made with the aim of addressing them nationally. These safety recommendations can identify situations where pressures on the healthcare system (financial, organisational, social, or cultural) mean that it is not possible to manage the persistent safety risk effectively.
• Safety performance evaluation is needed to assure that implemented ways of identifying safety hazards and managing safety risks are effective. There needs to be a way of monitoring the system to check safety performance – that is, whether intended safety objectives are being met and the system is achieving desired outcomes. This is achieved by following patient journeys through the system as they encounter different care providers.
• Managing change. Healthcare organisations experience both temporary and permanent change when responding to new demands and when introducing new services. Change may introduce problems that may impact on the effectiveness of care delivery. An investigation can identify unintended consequences that might affect patient safety when new ways of working are introduced.
• Safety promotion involves effectively communicating safety risks and how these risks can be managed to both staff and patients. A patient safety investigation can establish whether the relevant people are aware of the need for specific patient safety interventions, that training and education programmes are in place, and that the efficacy of these programmes has been evaluated

The first stage of the analysis involved coding the investigations by selecting sections of text that represent key investigation findings. This was carried out using qualitative data analysis software (NVivo by QSR International). The four reviewers met to discuss the codes to ensure that all of the key findings had been identified.

The reviewers then classified all of the codes across all of the investigations using the Safety Incident ResearCH (SIRch) framework (see SIRch framework box below). The framework was used to identify how the design of work systems can impact on the safety of patients. Patient safety incidents can arise from conflicting, incomplete, or suboptimal systems of care, which patients are a part of, and interact with. This approach aimed to identify the problems within the systems and processes that determine how healthcare is delivered.

The second stage involved a thematic analysis of the systems and processes that were identified using the SIRch framework. Similar themes were compared across the 22 HSIB investigations. The aim was to find recurring patient safety themes across the different healthcare settings that HSIB has investigated, by generating thematic maps. These maps were discussed and reviewed with the investigators to ensure that they captured the main findings of investigations, and a consensus was reached. Maps representing each of the identified themes can be seen in the next section of this report.

The third stage involved an analysis of the 85 HSIB safety recommendations that were made by the 22 HSIB investigations. Similar safety recommendations were classified by recommendation type, which captured the reason for making the safety recommendation and where in the healthcare system the safety recommendation was targeted. The following categories were chosen as they represent the fundamental safety management activities used across safety-critical industries (International Civil Aviation Organization, 2018):

1. identification of patient safety hazards
2. improving the management of known patient safety risks
3. monitoring of patient safety performance
4. evaluation of patient safety interventions
5. training and education for patient safety
6. promotion of patient safety.

The reports’ safety recommendations were put into one or more of these categories. The categories used in this analysis are recognised components of a safety management system. This is a deliberate choice to see if this provides an insight into whether these principles of safety management could improve how safety recommendations are made and acted upon in healthcare

Theme definition

Access to care and transitions of care: If services are available, then the opportunity to obtain healthcare exists. However, barriers also exist, and these can prevent appropriate access to care. These barriers often become apparent during transitions of care (that is, when patients move between care providers or care settings). These barriers can be physical, financial, organisational, social, or cultural.

Thematic map

Eight HSIB investigations were strongly linked to problems associated with access to care (see figure 1). Problems were most apparent when patients transitioned or transferred between different care providers. For example, harm occurred when:

• processes failed for patients requiring access to mental health services (Investigations 2 and 5)
• transferring acutely unwell patients between hospitals (Investigation 7) and chronically unwell people between prisons (Investigation 15)
• discharging patients from hospital who required new medication
• the healthcare system did not direct patients who required emergency care in a timely way (Investigation 13) • arrangements for patients who required urgent follow-up were not made (Investigations 12 and 18).

Mental health

A lack of provision and capacity in mental healthcare services can be a factor that affects access to care for all individuals. Two HSIB investigations found that there are also systemic problems associated with the continuity of mental healthcare and when patients attempt to access care from a new provider.

Transition from child and adolescent mental health services to adult mental health services (Investigation 2).

• The investigation found that young people using child and adolescent mental health services would benefit from a flexible, managed transition to adult mental health services. This should be carefully planned with the young person and should provide continuity of care and follow-up after transition. A period of shared care would help to ensure continuity for the young person.

Provision of mental health care to patients presenting at the emergency department (Investigation 5) •

• The investigation found that when assessing a patient’s physical health, emergency departments had established processes for triage and initial assessment, but lacked a consistent approach when considering a patient’s mental state.

Care of acutely unwell patients

Acute illnesses are those that are of short duration. They may be minor or serious. Sometimes acutely unwell patients require access to specialist services, for example, when transferring a patient to a hospital that provides specialist treatment of disorders of the heart.

Transfer of critically ill adults (Investigation 7)

• The investigation found that there were no consistent guidelines for the transfer of acutely unwell patients for either emergency or planned situations. There was variation in how Critical Care Operational Delivery Networks, whose role is to co-ordinate patient pathways between healthcare providers, are set up and governed.

Care of chronically unwell patients

A chronic condition is an illness that is long-lasting in its effects. Patients managing chronic conditions sometimes encounter barriers that prevent them from routinely accessing care on a timely basis.

Management of chronic health conditions in prisons (Investigation 15)

• The investigation found that where there is only one authorised prescriber on site in a prison healthcare department, particularly during core hours when transfers between prisons occur, this creates pressures on the system that may put prisoners with medication requirements at risk.

Medication systems

When a patient changes care providers, for example when they are discharged from hospital, the healthcare system must ensure that any ongoing medication needs have been accurately specified and communicated to the patient and the ongoing provider of care. This is challenging because different medication systems are used in different settings. Issues relating to transitions of care can prevent timely access to care or sometimes result in unintended errors, for example, a patient being prescribed inappropriate medication.

Electronic prescribing and medicines administration systems and safe discharge (Investigation 16)

• The investigation found a lack of interoperability (the capacity to exchange, interpret and store data to common standards) between primary and secondary care electronic prescribing systems, between secondary care facilities, between secondary and tertiary care, and between secondary care and community pharmacy.

Emergency care pathways

Patients with certain health conditions need to be treated as an emergency and offered treatment without delay to improve the chance of a good outcome. One example is testicular torsion, which occurs when a person’s testicle rotates, twisting the spermatic cord that brings blood to the testicle. It requires emergency surgery and, if treated quickly, the testicle can usually be saved.

Management of acute onset testicular pain (Investigation 13)

• The investigation found that the forwarding of clinical information from a face-to-face consultation is straightforward, as GPs can provide it directly to the patient to pass on to the emergency department. The IT systems did not enable clinical information from a telephone consultation to be shared across all providers. A lack of clinical information when a patient attends an emergency department increases the risk of a missed diagnosis.

Urgent follow-up

After patients have undergone medical tests or been diagnosed with health conditions that need to be monitored, it is important that the system is able to ensure that these patients are followed up appropriately.

Failures in communication or follow-up of unexpected significant radiological findings (Investigation 12)

• The investigation found that there was wide variation in practice in how unexpected significant radiological findings (findings from scans and X-rays) were communicated to clinicians. There was also considerable variation in how findings were acknowledged by clinicians (if they were at all). There was very little assurance that the actions indicated by the findings had been taken.

Lack of timely monitoring of patients with glaucoma (Investigation 18)

• The investigation found there was not enough capacity in hospital eye services to meet the demand for glaucoma services. Capacity can be maximised by ensuring referrals to, and follow-ups by, hospital eye services are appropriate and by introducing new ways of working.

Learning from patient safety investigations

Table 2 provides examples of HSIB safety recommendations that are about access to care and transitions of care. The safety recommendations were classified by the type of safety management activity that was required to respond to them (see Method section).

Theme definition

Communication and decision making: The delivery of healthcare depends on timely communication and effective decision making. There are many situations where pressures on the healthcare system can impact on the reliability of communication and decision-making processes. These processes often need to be supported by carefully designed communication and decision support tools, relevant and usable guidelines, and evidence based education. The analysis of this theme identified that communication and decision making are dependent on each other, i.e., decision making processes are constrained by ineffective communication.

Thematic map

Six HSIB investigations were strongly linked to problems associated with communication and decision making (see figure 2). Risks became apparent when it was discovered that serious health conditions had not been detected, or clinical concerns had not been properly communicated to others. Problems occurred when:

• needing to urgently refer (or escalate) the care of patients to specialist teams (Investigations 10 and 21)
• making the necessary adjustments in line with patient needs (Investigations 20 and 22)
• encountering rare conditions that required an immediate response (Investigations 11 and 19)

Highlights from investigations

The escalation of care and referral systems

Healthcare staff use tools such as early warning scores and referral systems to guide decisions about how a patient should be cared for. An early warning score is a guide used by clinicians to help alert them to deterioration in a patient’s condition. It is represented by a numerical value, and if a threshold is reached, a patient’s care should be escalated to a specialist team. Referral systems can include a means of prioritising patient care in situations where there is a risk that serious conditions could be undetected and further tests are urgently required.

Recognising and responding to critically unwell patients (Investigation 10)

• The investigation found that staff may overly rely on tools such as early warning scores, especially when working in a busy and complex environment. There tended to be a focus on the latest physiological observations (such as heart rate, temperature and blood pressure readings). This could provide false reassurance in some patients, especially when done in isolation, without listening to the concerns of a relative or considering the clinical judgement of the healthcare professional. Furthermore, the design and presentation of clinical information did not support staff in making a complete and accurate assessment of patients.

The diagnosis of ectopic pregnancy (Investigation 21)

• The investigation found that referral systems should include standardised information that supports triage and decision making by early pregnancy services. There may also be benefits in standardising the information leaflets given to women in early pregnancy who are discharged from an emergency department.

Consideration of patient needs

Communication and decision making are likely to be enhanced if the needs of different patient groups are better understood and processes are adjusted in response to those needs to reduce patient safety risks.

Potential under-recognised risk of harm from the use of propranolol (Investigation 20)

• The investigation found that there is a specific group of patients who may be at an increased risk of using propranolol for self-harm because they have coexisting migraine, depression, or anxiety. Awareness of the potential harm of propranolol in overdose was limited and hindered the ability of prescribers to exercise clinical judgement when they chose to prescribe propranolol.

Undiagnosed cardiomyopathy in a young person with autism (Investigation 22)

• The investigation found that children with autism, learning disabilities and/ or learning difficulties often find clinical environments distressing, which may be reflected in their physiological observations. This may result in diagnostic overshadowing, where problems such as autism (or a medical condition) are attributed as the cause of other new problems, rather than considering other underlying causes, thereby leaving other co-existing conditions potentially undiagnosed.

Rare conditions

Rare health conditions can remain undetected if they are not suspected by healthcare professionals and where there is limited diagnostic decisionmaking support available.

Undetected button and coin cell battery ingestion in children (Investigation 11)

• The investigation found that although the risks associated with batteries and specifically button/coin cells have been published by the battery industry for the past 10 years, the severity of harm caused by such batteries becoming lodged in a young child’s oesophagus is not widely understood.

Delayed recognition of acute aortic dissection (Investigation 19)

• Aortic dissection is a serious condition in which there is a tear in the wall of the artery carrying blood out of a person’s heart. The investigation found that aortic dissection is a rare cause of chest pain, particularly in comparison to a heart attack. Staff in non-specialist hospitals may be unfamiliar with the condition and its signs and symptoms, as aortic dissection is seen rarely and symptoms can vary or be confusing.

Learning from patient safety investigations

Table 3 provides examples of HSIB safety recommendations that are about communication and decision making. Safety recommendations were classified by the type of safety management activity that was required to respond to them (see Method section).

Theme definition Checking at the point of care: Many routine activities require healthcare workers to check that the intended treatment is being delivered correctly. The point of care is where healthcare workers and medical devices are in direct contact with patients. Examples of errors that occur at the point of care include selecting the wrong treatment, administering the treatment in the wrong way, or forgetting a required step. The burden of avoiding or mitigating these errors is often on the healthcare workers, who are required to check and sometimes double-check. Instead, the aim should be to reduce the reliance on checking by developing systems and procedures that are designed to mitigate known risks.

This is one of the fundamental principles in human factors and ergonomics (Carayon et al, 2014). Some opportunities for error can be mitigated by better design and the focus should be on improved design rather than on healthcare workers trying to be ‘more vigilant’. Good design encompasses the engineering of devices and equipment, the development of evidence-based ways of working to minimise the likelihood of error, and effective training and education.

Thematic map

Eight HSIB investigations were strongly linked to problems associated with checking at the point of care (see figure 3). These problems became apparent when:

• selecting components used during surgery (Investigations 1 and 4)
• administering medication (Investigations 3 and 9)
• using medical devices and equipment without noticing that they will not work in the intended way (Investigations 6 and 8)
• not having access to required resources while working in a busy environment (Investigation 14)
• having to use equipment that is not fit for purpose (Investigation 17).

Highlights from investigations

Selecting surgical components

Processes for verifying that the correct components are used during surgery can be improved. There are often opportunities to consider how data should be entered when using IT systems so that the likelihood of error is mitigated.

Implantation of wrong prostheses during joint replacement surgery (Investigation 1)

• The investigation found that orthopaedic joint replacement surgery (for example, hip replacement surgery) involves combinations of components, and processes for selecting and verifying the correct prostheses (artificial body parts). The prosthesis checking process did not always prevent implantation of the wrong prosthesis. Opportunities for improvement associated with the design of labels, how data about joint replacement surgery is entered in the National Joint Registry and the use of barcode scanning were identified.

Insertion of an incorrect intraocular lens (Investigation 4)

• The investigation found that the processes used by organisations at each stage of the patient pathway for cataract surgery (where intraocular lenses are used) varied significantly. It was found that the design of the software interface used in the operating theatre meant it was prone to usability issues, even for those who were familiar with it.

Administering medication

There are many guidelines and safety standards associated with how medication should be administered to patients. There is a need to understand work as it is done in practice and evaluate how useful the guidelines and safety standards are in mitigating the likelihood of errors.

Administering a wrong site nerve block (Investigation 3)

• The investigation found that guidance for the nerve block procedure (the injection of an anaesthetic) that was intended to minimise the likelihood of errors did not provide sufficient clarity or direction on how it should be used in practice. Consequently, there was significant variation in practice and its uptake among clinical staff. There was no evaluation of how the guidance was being used in practice and whether local variations or alternative approaches reduce risks.

Inadvertent administration of an oral liquid medicine into a vein (Investigation 9)

• The investigation found that unfamiliarity with different types of syringes can mean that the safety barriers designed to prevent some types of medication errors are ineffective. Many national organisations play a role in medicines safety. However, there was a lack of clarity of the roles of these organisations and their responsibilities for system-wide implementation of safety standards and the dissemination of messages about safety.

Medical devices and equipment

Medical devices and equipment require engineering and design so that healthcare workers can determine that treatments are being selected and delivered as intended.

Design and safe use of portable oxygen systems (Investigation 6)

• The investigation found that portable oxygen systems currently used across the NHS in England do not provide clear and timely feedback that oxygen is flowing to the patient. There are various design issues with current portable oxygen systems that may lead users to interpret that oxygen is flowing when it is not.

Piped supply of medical air and oxygen (Investigation 8)

• The investigation found that safety alerts are issued to reduce the risk of errors. An example of this would be unintentionally using an air outlet instead of an oxygen outlet to support a patients breathing. However, local assurance processes relating to the implementation of these alerts were found to be ineffective. The ability to nationally record trusts’ specific actions in response to alerts existed but was found to require further enhancement to support trusts and national oversight.

Error-prone systems and design

Opportunities to use new technology to improve the reliability of processes and treatments within healthcare often exist but are not always identified or implemented.

Wrong patient details on blood sample (Investigation 14)

• The investigation found that staff are required to adapt the way they take and label blood samples to account for the environments and circumstances in which they work. It was found that there was a risk that systems that use labels and handwriting on blood samples were error prone. There was evidence to support the use of electronic systems that could minimise the risk of incidents and make the process of blood sampling and labelling more efficient.

Detection of retained vaginal swabs and tampons following childbirth (Investigation 17)

• The investigation found that national guidance was mainly focused on changing the way in which people work. The investigation considered that this was less likely to be effective than changing the design of vaginal swabs and tampons and provided weaker mitigation for preventing these products being unintentionally left inside patients. Factors were identified which made it more difficult to detect the presence of vaginal swabs and tampons, including the way in which they are designed.

Learning from patient safety investigations

Table 4 gives examples of HSIB safety recommendations that are about checking at the point of care. Safety recommendations were classified by the type of safety management activity that was required to respond to them (see Method section).

A safety management system is an organised approach to managing safety (see below). It sets out the necessary system-wide processes needed for proactive and reactive safety management. Safety management systems seek to proactively mitigate threats to safety before they result in undesirable outcomes. Through the implementation of safety management systems, stakeholders can integrate their safety activities. This enables the system to prioritise actions to address safety issues and manage its resources more effectively.

HSIB’s work so far suggests that the NHS may benefit from exploring how a safety management system approach could be built on the foundations developed by the NHS Patient Safety Strategy (NHS England and NHS Improvement, 2021).