Guest blog post by Professor Peter Hibbert, Patient Safety and Appropriateness of Care Research Stream Lead – Australian Institute of Health Innovation, Macquarie University, Australia.
Patient safety continues to be a major public health problem across the world. From Imperial College’s National State of Patient Safety published in 2022, we know that incidents resulting in harm to patients (adverse events) which are preventable have improved in some areas, such as healthcare-associated infections (HAIs), compared to the turn of the century when the landmark report, An Organisation with a Memory was published. However, rates of preventable adverse events remain unacceptably high in England with, for example, maternal care for minority groups being a particular policy focus.
One of the key strategies to reduce harm to patients in healthcare and other safety-critical industries is to undertake investigations into what went wrong and why and propose actions to reduce recurrence of similar adverse events. In healthcare, these investigations have traditionally used the root cause analysis methodology, however additional methods are being introduced. They are generally undertaken locally within NHS trusts, led by clinicians or staff employed in safety and quality or clinical governance units.
A systematic review of the limited research undertaken shows that root cause analysis is not particularly useful for determining effective measures to prevent the recurrence of patient harm. It should be acknowledged that doing research in this area is quite difficult because the clinical procedural changes proposed in root cause analyses and other investigation methods are highly varied. Additionally, each type of adverse event occurs infrequently locally at the level of trusts, meaning that measuring the impact of investigation recommendations on rates of adverse events is problematic.
Given this research, other reports, and ongoing feedback from the healthcare sector, the Australian government via the National Health and Medical Research Council (NHMRC) (the Australian NIHR equivalent), and four state and territory governments have recognised that these investigations – as one tactic across a whole response to patient safety adverse events – need to be more effective in reducing recurrence of adverse events. They have invested in our research program to understand how investigations can be improved via a four-year project which commenced in 2023.
Our researchers have been speaking with clinicians, safety and quality staff, patients, experts across the world, and investigators at the Health Services Safety Investigations Body (HSSIB). In total, we have spoken to 97 people thus far. What they have been telling us is that there is little sharing (between healthcare organisations) of and learning from investigations, that the quality of the investigations needs to be significantly improved, and that additional methodologies for investigations are necessary. This builds on our previous award winning research on the effectiveness of recommendations in patient safety investigations which proposed that there should be more use of direct observations of clinical workflow and processes, greater involvement of human factors, more aggregated analyses of multiple adverse events, more independence of investigations, and more analyses and making recommendations at a national level.
Given these contemporary findings in Australia, it is worth reflecting on some of the contributions of the HSSIB, particularly as Australia does not have a comparative organisation and function. In the next section, I discuss three areas in which I believe HSSIB significantly contributes to improved patient safety – a national approach, professionalisation of investigations, and introducing other investigation methodologies and data sources. I then briefly discuss HSSIB’s education offering and then suggested area for improvement for HSSIB in relation to research and building an evidence base.
The contributions of HSSIB
A national approach to complement local investigations and sharing of lessons
Trusts undertaking effective investigations can improve local clinical processes and therefore reduce adverse events re-occurring. The remit of the recommendations or actions that flow from these investigations tends to be local, i.e., confined to the trust where the adverse events occurred, such as improving clinical workflow or communication or handovers between clinical departments, modifying a local procedure, or adding a checklist to make a clinical process more reliable.
What tends not to be tackled from these investigations are broader systems issues that can impact on patient safety. These can be policies from Departments of Health or regulators, clinical practice recommendations from Royal Colleges, or changes to medications, equipment, or packaging. These types of changes are not seen as within the control or power of trusts and therefore largely considered as out of scope.
Another problem is that findings of local investigations tend not to be shared between trusts. Therefore, the learning tends to be local, and repeated investigations of the same type of adverse event may be undertaken at multiple trusts, duplicating effort.
The HSSIB makes a significant contribution to address these two issues. The remit of the HSSIB is to make safety recommendations to organisations and bodies best placed to take action to address a risk to patient safety at the national level. Therefore, they can make recommendations directly to organisations such as regulators, Royal Colleges, and suppliers to the NHS. These recommendations can complement and indeed strengthen the actions made locally by investigations undertaken by trusts and can impact multiple different types of adverse events, at a population level. For example, HSSIB in their report on The selection and insertion of vascular grafts in haemodialysis patients recommends that NHS England reviews system requirements for barcode scanning technology, in order to support local organisations to reduce the risk of incorrect selection and insertion of prostheses/implants. Such a recommendation is clearly likely to have a much greater impact than a similar recommendation by an investigation by one trust, and is also likely to promote supply chain changes across the NHS in relation to new technology.
Unlike local investigations, the HSSIB reports are public. The findings are shared, lessons can be learnt, duplication reduced, and standardisation promoted across the NHS and beyond. Indeed, in 2022, I was a member of an Executive Clinical Governance Committee in an Australian hospital with a paediatric inpatient service. The HSSIB’s report Weight-based medication errors in children was tabled by the Committee and an audit was undertaken within the hospital on the local-level learning actions proposed by the report. So, learning from, and sharing of HSSIB patient safety investigation reports may extend beyond the borders of England.
Professionalisation of investigations and integration of systems and human factors
A well-recognised problem with some local patient safety investigations is their lack of professionalisation of investigations. Safety is a science, with a strong foundation of peer-reviewed literature and theoretical underpinnings, and robust academic and operational debates occurring over the optimal methods of investigations to use. There is also an increasing recognition of the complexity of health, which may make the task of finding contributing factors to adverse events and developing effective recommendations even more challenging. Indeed, the contrast has been made between biomedical sciences and patient safety: the former is tackled via independent research institutions with highly trained scientists; patient safety problems, which are arguably just as complex, are expected to be solved with fewer resources, using part-time staff with little task-specific experience, at a local healthcare organisation level. These complex patient safety problems require appropriate investments in expertise and effort analogous to the requirements for research into biomedical problems.
Generally, there is a stark contrast between qualifications and experience of the investigators in other safety critical industries, such as transport and nuclear and those undertaking investigations in healthcare. In healthcare, short courses and development of on-the-ground knowledge tends to the norm. This may be one of the factors limiting the effectiveness of patient safety investigations and their findings which may focus on blaming clinicians involved in adverse events and development of superficial, generally ineffective recommendations such as educating staff.
In other safety critical industries, investigators are generally professionally qualified and experienced in the art and science of investigations. Their skills include cognitive interviewing, gathering and collation of data and evidence, and knowledge of theories and frameworks of organisational dynamics and culture, human factors and complexity science. From our contact with HSSIB, their investigators also possess a similar array of skills. The investigator team has a varied set of qualifications in health and other areas, such as engineering, psychology, human factors, systems science, law, and aviation, allowing a diverse range of perspectives. These are likely to contribute to investigations that will produce effective and sustainable actions that reduce harm to patients.
Using different methodologies and data sources
HSSIB is introducing and trialling additional methodologies into their investigations. These include Systems Engineering Initiative for Patient Safety (SEIPS), AcciMap, Systems-Theoretic Accident Mode and Processes/Systems-Theoretic Process Analysis (STAMP/STPA), and Functional Resonance Analysis Method (FRAM). They also incorporate methods developed in other industries such as the Australian Transport Safety Bureau (ATSB) investigation analysis model. Their use will inform the evidence base on the suitability and acceptability of these methods in healthcare.
Many local patient safety investigations rely on two main data sources – documentation, including the medical record and policies and procedures, and interviews with staff (and potentially patients and families) involved in the adverse event. These data sources tend to be reasonable at describing and identifying the facts of the adverse event, however they are less use in understanding the systems stresses on staff that occur every day. These systems stresses are likely to provide rich data on why adverse events occur to inform effective and sustainable recommendations.
HSSIB are not just relying on these data sources but including others such as observations of clinical work to understand how healthcare is delivered in practice. Including observations of normal work in investigations is considered best practice in the safety science literature. The benefits are that the investigators gain a deep understanding of the normal stresses on clinicians, which then inform the actions proposed to improve care to patients.
Pleasingly, HSSIB are also including speaking with patients and families impacted by adverse events in their investigations. Their first-hand experiences can be highly informative, for example, how communication, consent, and monitoring deterioration can contribute to adverse events occurring, which can inform patient-centred recommendations.
Education and leadership
As well as undertaking patient safety investigations, the HSSIB has an education arm. Their well-regarded education programme is available to those in health and social care settings in the NHS in England. The courses are delivered by professional healthcare safety investigation experts with the flagship course – A systems approach to investigating and learning from patient safety incidents – accredited by the Chartered Institute of Ergonomics and Human Factors (CIEHF). The education offering is high quality and a welcome addition to the landscape of patient safety in England.
As well as this education, an important role for HSSIB is modelling good practice in investigations. This includes all that I have mentioned above – professionalisation of investigations and incorporating human factors and systems thinking and new methodologies and data sources into investigations. The HSSIB patient safety investigation reports set expectations and demonstrate how investigations can be done to a high standard, thereby providing trusts and other healthcare organisations from across the world with a model of what is possible.
Evaluating the impact of HSSIB’s investigations
Improvements in patient safety are difficult to measure. This is because most types of adverse events at the level of trusts occur relatively infrequently. Definitions of individual adverse events are often not agreed upon and the data is often not aggregated across trusts hampering measurement. For a national body such as the HSSIB which puts out recommendations for trusts or national institutions to do, evaluation of effectiveness can be fraught – for example, attribution of a recommendation’s effectiveness in a complex world where there are multiple influences on organisations and clinicians, and understanding whether recommendations are effective or whether they have been implemented or even known about at trust level, can be challenging to measure.
However, this is not an excuse for not attempting to evaluate the impact of HSSIB’s patient safety investigation reports. One suggestion for HSSIB to improve would be to engage more in the space of evaluation and publishing these findings. Not only is the impact of their recommendations of interest to the patient safety research, policy, operational and clinical communities, but also their ways of working and use of additional investigation methodologies. The learning is relevant not just for England but for the wider international patient safety community.
Related articles
