Jump to content
  • Who should investigate? Part 3. A series of blogs from Dr Martin Langham

    • UK
    • Blogs
    • New
    • Health and care staff, Patient safety leads


    This is part three of a series about the investigation process and human factors in healthcare. Part one looked at the why we investigate an ‘incident’. It concluded that there is only one reason to investigate – and that’s to stop the error occurring again. The idea that human factors is a science – done by science types was introduced. That facts are best collected by a minimum of two investigators. Pictures being our friend, and the cognitive interview concept was introduced.

    This part focuses on ‘Who’ should investigate and deals with the experience and expertise of the team, their roles and responsibilities in the light of the facts they will collect. 

    This blog is aimed at individual trusts and organisations rather than regulators/national bodies, etc.


    Consuming tea and cake as I write this also means I can break my ‘in healthcare rule’. This rule says never say to a medical type, “well in other industries it’s done like this”. Healthcare is very different to anything else and outsiders like me should not point at other industries and say there is a panacea of methods that healthcare should use. But, placing the cup down, deep breath – what I discuss here is based on my training in incident investigation in the police (UK, US, and EU), military operations, rail, marine, aviation and security failure.

    Overall, I’m starting with how I was trained in road death investigation methods and the senior investigating officer (SIO) murder investigation system. Two reasons for this – one, there are online manuals, two, they kind of work and are a good starting point. There is a very big ‘however’ as medicine is different (in fact all industries are different to policing). This, however, is big and came to me as a conclusion while interviewing potential applications for those applying to be investigators in rail. I did not notice, but the rest of the panel noted with some extreme words, that all the police trained investigators were there to try and convict someone rather than stop it occurring again. These blogs are not about criminal and civil liability – just prevention. Super, all police colleagues alienated!

    Given the ‘Who’ might be one person, let’s think about them and see if we can think about getting a little help to get a team together. Yes, it’s true one nurse will have many investigations to do, but in part two of this series I described that ‘making facts’ is better with two.

    When I arrive at a crash scene or an investigation, I always ask one question: “Am I the right person to do this?” It’s a bit of a chicken and egg question as you have very little detail. Most of my callouts start with the phrase “you would not believe this but”. The first phase is to take a blank sheet of paper – never a proforma as that will get you only collecting data you expect to find and be a traditional scientist. Start by looking at the scene and thinking about all the possible reasons it could have occurred. Ideally (more in the ‘When’ section), its within 24 hours and nothing has moved or changed – but its healthcare so its weeks later, and the ward has changed from neonatal to geriatric, or the theatre has been moved to the next room. 

    Before you decide who should be on the team, you should first ascertain the likely proximate (main) cause, these are: the ‘environment’, the ‘human’, the ‘equipment or system’.  Of course, these factors interact but the proximate cause should decide where to start.


    Equipment failure – or the conclusion it’s not usable – is best done with an engineer and a human factors person. Medical equipment is often poorly designed. Why a piece of equipment ‘failed’ in terms of not diagnosing the patient’s condition or keeping them safe is a little complex. In the next post on human factors we shall explore the sub disciplines of human factors and as we shall see there are two groups who specialise in computers (screens etc) and machine design. Now four areas of failure need to be thought about.

    1. The machine failed and, for example, its electronics or pump stopped. An engineer is the best person to consult, but how the failure was signalled to the operator is of interest to the human factors person.
    2. The machine failed in the sense the operator did not understand how to use it or the output was confusing. 
    3. The equipment failed in the sense that it looked similar to other equipment on the ward or the theatre but operates in a different way. Sometimes on the same ward machines that have the same function work in different ways.
    4. The machine worked well, but its outputs were not immediately comprehendible, or other machines provide outputs in a different way.

    An example of a simple bit of kit that fails in different ways:

    • A hospital trolley with no locking mechanism on the wheels – it fails as the team chase it round the ward.
    • A hospital trolley that has different drawers on each model so, in an emergency, drawers have to be searched through.
    • A trolley in the north end of the ward which looks the same as other trolleys but is not stocked with the same drugs.
    • A trolley with electronics of a similar appearance to others but where it has different user interfaces and graphical interfaces.

    I use these examples from my own research. One trust had 17(+) different types of trolley at one site, all with unique set ups and contents, yet all were labelled as having the same function. The experimental test was the question “In what drawer is the adrenalin?” A 100% failure rate was found.

    More when we talk about human factors. And thanks to Emma for pointing out this major issue – yay for reviewers!


    People who know the environment best are those who work in it. Typically, the environmental issues to measure and understand relate to lighting, stressors like heat and cold, ergonomic physical space and, in my experience, noise levels. In many other industries a noise level of 120dB means closure of the site and ear defenders, in healthcare a noise level typically found in a factory is ignored. In the forthcoming human factors posts I’ll cover sound. 

    Environmental factors are not as easy to correct. Telling the hospital that the anaesthetics room is so small it can’t be used will cause death threats from the facilities management team! Lighting changes – from fluorescent or incandescent to LED always causes issues. However, the issues of lighting – and by implication colour rendition – can all be simply measured by tools in the typical human factor types backpack (designer of course). Light meter, dB meter etc. Human factor people should be able to chat about the physical layout and how the fact us slowly evolving cave dwellers are physically weak should be discussed. Sometimes the physical design of an environment is very poor. I’ve heard the comment “With one hand the operator controls this, with the other these controls, and the other hand interacts with the screens”.


    95% is the number you should be familiar with. That to say the most common factor – or contributory factor – is the human element. Research from the late Barbara Sabey’s seminal work points out that in 95% of cases it is the human that’s caused the problem. Irrespective of industry, location or activity that number (or very close to it) comes up. So, a human factors person should be part of the team. It takes a postgraduate qualification and lots of experience so hire one early on. They do drink lots of tea and although they are doing 95% of the work – they are only one of your team.


    It’s true I should say it’s a human factors person that does this bit, but nurses are the best I’ve found. By the phrase ‘system’ we mean in terms of system theory, which 'Professor' Wikipedia says is

    “Systems theory is the interdisciplinary study of systems. A system is a cohesive conglomeration of interrelated and interdependent parts that is either natural or man-made. Every system is delineated by its spatial and temporal boundaries, surrounded and influenced by its environment, described by its structure and purpose or nature and expressed in its functioning. In terms of its effects, a system can be more than the sum of its parts if it expresses synergy or emergent behaviour.”

    I should describe psychology systems theory (part of human factors) – but the above is good enough for now and there are many sub fields.

    In essence, look at how information flows and how many stages people need to go through to achieve a task. Each stage or function adds another source of error. Walk through the process and critically examine how each stage can fail. My work in the military tells me every stage that can fail will. Remember humans are fallible and the system or method of working should fail-to-safe. Some within the NHS say fail to safe is a new concept. It’s part of the 1832 Railways act.

    If we mention a system it might be worth saying the phrase – task analysis. Task analysis is something human factors types are trained to do. It breaks down a task or system into each of its constitute part and looks at the order or dependencies that are needed to complete each main and sub task. The classic training is making a cup of tea, the victim, I mean student, lists all the stages and the master (teacher) points out that there are lots more stages. The student says fill cup with hot water. The master says – but where does the cup come from? Student and teacher hate each other for years after but a list of things to do to achieve the task is formalised. It’s a useful tool but often it’s the most junior member of an human factors team that does it and it’s a rite of passage to say you are a human factors person. The method should be used to help the investigation. Buy me a drink and I’ll explain why the doors of a tram – when its coupled to another tram, and that tram is in a certain station, and the third emergency button in carriage one is pressed while the driver is closing the doors in carriage two while another part of the tram the smoke detector is activated within 0.25 of a second of another alarm… well you get the picture.

    Four in a team then?

    With four headings this kind of implies four people in the team. The team should have those skills but what about the personal qualities and above all the relationship to the injured party.

    The investigation teams need to be independent and not be familiar with anyone involved. Humans are not rational data collecting machines and knowledge about others and the environment will cloud how you collect the facts – even if you try hard. The ideal lead investigator knows perhaps a little about the hospital or care place but does not know those involved. They should know a little about the equipment or machines, but not be so familiar that they know the ‘work arounds’ to get it to function. The lead investigator should handle discussions with outside organisations. Now if you are following carefully you note I’ve just added another team member – the senior investigator or lead investigator. This person organises the team of those with different skills and is responsible for the output and keeping the team from the pressures of the outside world, far enough away so the team can collect data. They should be cognisant of not just the investigation team but also those who were involved in the incident. This is where healthcare is very different to any other industry – the care of those involved.

    Care of those involved is, in my experience, very strange in healthcare. If you attend an incident you ensure that all those involved get home safely and have a contact point. In healthcare the fact that it’s highly unlikely that the action that led to a fatality was deliberate and the nurse, doctor, administration person was not trying their best appears to be overlooked. I attended an incident where the team were in tears and visibly shaking, they were told to drive home and come back tomorrow (next shift and list already given to them). It should not be the role of the investigator to organise a taxi and lend a phone so family can be called for support.

    Now the ‘Who’ may suddenly, radically change. The question first in your mind should be – is this an accident? You may recall in part one – ‘why investigate’ we defined an accident or incident. Essentially, we decided together that it’s a random rare unforeseen event with lots of variables coming together in one moment in time. Well at this point you should know if it’s ‘accidental’ or deliberate, and if its random or not. The who investigates or leads the investigation could change. If the test of accidental is not met, then it’s a criminal matter and the lead investigator or police senior investigating officer will take over and “Oh well done you are using a recognised method” should be heard.

    Investigations involve a small team. There is lots of data and the experience of just one person is not enough. A lead investigator can deal with the media, families and ensure the investigation team collects the facts. Those who investigate need to be able to collect the facts needed and be scientific in their approach. The person collecting evidence from, say, CCTV needs to have some experience of using it, and the person considering the medical records needs an understanding of how that particular ward uses the charts. Or, in my experience where they are, which is different to the ward on the floor below. No one has all the experience needed, and we all need to learn. My first recovery of CCTV images took two days. Strange the operator (now manager) and I are still friends many years later, but approach the matter with a smile and a look of “I’m lost”.

    Time off from the day job is vital. Investigations are hard work and mentally and physically exhausting. How long should the team be together? Well that depends and only comes apparent after a day. I’ve had a purchase order for 6 weeks investigation services and finished at lunchtime on day one. I’ve been asked to spend 6 weeks and then two years later still going. One of my team did the whole thing in 30 minutes… saving the 12 weeks of funding for something else.

    When I’m part of an investigation team I deal with humans and the systems, I’ve no (well little) knowledge of the engineering side or the other facets. Human factors is part of the story and the skill is for the lead investigator to bring all the views together. A human factors type should not lead the investigation. More colleagues alienated.


    Investigations should be conducted with a small team. Small teams need a leader and that can be the person who is also charged with the welfare of those involved and the communication with those outside the team. The team should have time enough from their ‘day jobs’ to do it. How long do they need off? Well that’s down to the incident. The best bit of equipment to take is a blank sheet of paper, and there is a vital need to effectively communicate the findings.

    That’s three blog posts on the investigation method. Still lots to do.

    Read the other blogs in this series

    About the Author

    Martin is topic leader for the hub.

    He founded the Human Factors group at the University of Sussex (1999), which became User Perspective Ltd in 2003. Martin, User Perspective MD and Chief Scientist, aided by his team, has undertaken almost 600 research and forensic investigation projects. He is interested in human error and human factors.

    Martin is a research auditor for the UK government, EU academic networks and many governments worldwide. Within healthcare he has investigated matters as diverse as neonatal safety in transport, unexplained injuries in the hospital mortuary, sepsis diagnosis and retained instruments. Martin co-authored the very first Healthcare Safety Investigation Branch (HSIB) report that investigated orthopaedic surgery in the UK and Europe. His interest in the law and justice extends to his voluntary role as a justice of the peace (JP) in the Magistrate and Crown courts.

    0 reactions so far

    1 Comment

    Recommended Comments

    Create an account or sign in to comment

    You need to be a member in order to leave a comment

    Create an account

    Sign up for a new account in our community. It's easy!

    Register a new account

    Sign in

    Already have an account? Sign in here.

    Sign In Now
  • Create New...