Douglas Noble: Tales of patient safety from the frontline for junior doctors: incident reporting

douglas nobleThe NHS has so far accumulated almost 3 million incident reports, well on the way to being as tall as the British Telecom Tower if they were all piled up one on top of the other. Many significant research studies have identified the main barrier to incident reporting as lack of feedback to the reporter. Sound familiar? Unsurprisingly, it’s increasingly difficult to get doctors to report incidents; almost 90% are reported by nurses. Even worse, the few percent doctors do contribute to this figure likely represents the least serious incidents. Add to that the lack of engagement from senior clinicians and it makes for a very curious day to day experience on the wards.

Every respectable change management paper concludes that change is only really possible if it is led from the top of organisations. In this case senior clinical leaders need to lead the way in making incident reporting the way things are done around here. Yet I’ve actually never witnessed a consultant completing an incident report form. I’m sure it must happen, I just wonder when and where.

Sadly, many perceive incident reporting as a blame tool – a way of complaining about that nurse, ward or doctor that bothers you so much. The incident report has yet to find its home as an anti-blame tool and to be seen as strengthening hospital systems. Hospital systems are fundamentally weak and littered with risk. Identifying risk is genuinely tricky.  Yet, being aware that as a junior doctor you’re only ever one step away from catastrophic error is not hard to appreciate. This is where incident reporting should seamlessly slot in; a perfect anti-blame tool allowing all healthcare professionals to notice something potentially dangerous to a patient and quickly enter that information formally into the system. A careful system of team review and feedback that is vehemently anti-individual-blame will reduce error.

However, what happens if you inject patient safety thinking, in the form of incident reporting, into a blame culture (still a reality in large parts of the NHS)? Ironically, it precipitates the blame culture, becoming a tool for retribution. Have you ever heard it: ‘I’m going to write an incident report against you?’ This phrase does not just signify the misunderstanding of incident reporting, it acts as an indicator for a much deeper problem – an existing blame culture. This may be the problem that needs to be addressed first or incident reporting will only make it worse.

Douglas Noble has worked in surgery, emergency medicine, public health and for WHO Patient Safety. From 2006 to 2008 he was clinical adviser to chief medical officer for England, Sir Liam Donaldson.

  • Christine Kang

    Dr. Noble,

    I completely symphasize with your insightful assesment of so called ‘incident report’. It IS a blame tool in everyday life in NHS and I am so sick of even hearing about incident form.

  • My org’s approach has been successful so far engaging physicians in peer review and incident reporting.

    Our model addresses the key points you mention.

    We make reporting tools available, often by integrating with the EMR at the point of care. Submitting a near-miss or event report is often a couple of clicks along with a brief narrative.

    The reports go right to my company’s web servers, a disinterested 3rd party with no ties to hospitals, insurers, or regulatory agencies.

    My company anonymizes reports before engaging the facility in a non-punitive root cause analysis and subsequent recommendation for improvement/future prevention.

    In the US, we have legal privilege as a “Patient Safety Organization” to protect the reporting individual and involved parties (patient and providers), not sure if the UK could offer such protection.

    This is not a perfect approach by any means, and is intended to lay the foundation for a proper culture of mindfulness. We protect as required under our contracts and status as a Patient Safety Organization, but the Open Source technology we deploy offers the path of slowly migrating toward a completely transparent system as the culture allows.

  • Arif Razak

    I wonder how much percentage of the reported incidents have been acted upon? It would be interesting to compare these figures to unreported incidents that have been dealt with by doctors/nurse involved.

  • Iain Yardley

    The problem Dr Noble describes of the blame culture is very real and present in my experiences.

    The other barrier to reporting is the lack of feedback. Incident forms go off into an apparent void and the individuals completing them often have no idea of the outcome. Friends and acquaintances in aviation tell me incident reports litter their coffee rooms and are a part of life. Reports into the incident where their boss wrote off a multi-million pound plane are well thumbed. Perhaps if my surgical coffee room had a few reports posted up from the last few weeks then we would be more aware of risks and the purpose of reporting them. Reporting would become the norm rather than something to be feared and a “safety culture” fostered.

  • Douglas Carnall

    The Medical Defence Union (MDU) used to publish a very wonderful little book called “Cautionary Tales,” which contained short (100 – 300 word) anonymous summaries of cases the MDU had seen over the years. Each usually ended with the immortal words “the Union recognised the case could not be defended and settled for n pounds.”
    It was compelling reading for the final year medical student.
    Ah, they still keep the idea alive [pdf]
    The difficulty of such case reporting is anonymising the reports effectively.

  • Lilia E. Ziganshina

    Dr. Noble,

    Having been in practical clinical pharmacology since 1996 in the Russian Federation I used to think that the blame culture was the fault of our domestic health system. Now I see that this is a universal problem.

    Recently the Federal system has been set up for centralised reporting of adverse effects. However, so far the results have not proved that the efforts are worthwhile.

    Spontaneous reporting in our clinical settings has proved useful only when carefully managed by experienced practicing clinical pharmacologists in a day-to-day patient care with regular feed-back, education and knowledge sharing.

    What do you think about possibilities of practical and active involvement of clinical pharmacologists in incident reporting at the frontline of patient safety?

    Many thanks,
    Lilia E. Ziganshina

  • Martin Turner

    As an adjunct to the other suggestions, we recently resurrected mortality and morbidity meetings for our neurology department (see also Stone J. Practical Neurology 2008;8:278-279; doi:10.1136/jnnp.2008.156877), with a view to incorporating incidents/near misses raised by junior and senior medical staff in an anonymised way as far as possible.

  • J Richards

    Interestingly the Incident Form in our hospital is perceived by doctors as a tool that can be used to make managers sit up and take note. By way of an example, dissatisfaction has recently been generated amongst surgeons in our trust by a change in the supplier of sterile gloves. The new gloves were felt to perform less well than those of the previous supplier and had been introduced by the managers, who had little/no experience of using sterile gloves in the operating theatre, as a cost-saving measure. When this discontent regarding events such as tearing, diathermy burns, dropping instruments etc was expressed to the managers verbally or by email no action was taken. However following a corporate move on behalf of the surgeons to report glove-associated events using incident forms, the trust has reverted to the original supplier of gloves.
    Do you think that this is use or abuse of the Incident Form system?

    I have also noticed than when a senior doctor wishes an incident form to be completed they often delegate this task to a nurse. This may in part contribute to the high percentage of forms completed by nurses?

    I look forward to seeing how this blog unfolds!

  • Andrew McArdle

    I sympathise with the view of the risk department as a void. Despite completing several incident reports last year (in what I hope was a non-blaming manner) I had to work hard to receive any feedback on what had happened as a consequence.

    My image of the risk department is one man behind a small desk in a small office dwarfed by piles of incident reports, gently crying.

  • Andrew Robson

    Thanks Douglas for a very interesting and insightful glance at incident forms. A pile as
    high as the BT tower? I don’t fancy being the FY1 who is given the “opportunity” to do an
    audit project about that.

    In my opinion, that pile is the key – nothing clinically relevant appears to come of the
    submitted forms. The meaningful incidents are swallowed up in irrelevant tomes scribbled by sweating staff on night shifts. Years later, I still recall the consternation I caused when I refused to sign an incident form regarding a patient who, two days previously, had fallen out of bed and then got back in. (I worried for weeks that management would hear about me).

    In a positive direction, how about trusts being expected to publish and feedback to their
    staff how they have improved patient safety over a period of time?

    Too many improvements to patient care can seem to become part of the scenery without
    being acknowledged as a step forward: I visited a hospital in Bristol recently and was
    impressed by the disposable bed curtains (with dates when they were hung up), dates
    written on inserted venflons and a host of other small things that are all recent changes
    and represent an improvement in patient safety. I’d like to hear about that sort of stuff
    from management.

    How about staff being required to put a priority score on their incident form that
    determines whether it makes its way to the top of the BT tower or moulders earnestly
    at the bottom?

  • Douglas Noble

    Thanks for such great and interesting responses to this blog.

    One of my biggest frustrations with reporting on the frontline is the derision it seems to get from doctors. Everyone just feels there’s no point filling in an incident report and even if they do it just gets put in a drawer somewhere and they never hear about it again. I remember the first incident report I filled in, which concerned a patient who had almost been given the wrong blood because the normal double checking mechanism was bypassed. I dutifully filled in a report. I never heard about it again. So, I sympathise with some of your views about this from personal experience.

    Addressing blame issues may be helped by anonymous reporting (as suggested by Martin Turner). Mandatory reporting also offers a possibility of alleviating blame, especially if filling in an incident report becomes the norm in everyday working life. Ian Yardley makes the good point that it is like this in the aviation industry. The downside is that mandatory reporting could drive all the problems underground, or ironically make the blame problem worse. Tackling a blame culture is a whole other topic, and perhaps we’ll get onto that in a future blog.

    The experience of Mike O’Meera shows what can be done when an organisation puts its mind to fixing problems and getting everyone involved. Very impressive. Your story reminds me of the Microsoft error reporting system which sends vast amounts of information to Microsoft at a single click and then, at least with Windows Vista, the performance system sends you back solutions.

    As for the role of pharmacologists, I think it’s essential especially as drug errors make up approximately a third of all patient safety incidents. Interesting letter from Jeff Aronson in last week’s HSJ on that topic. Thanks to Lilia E. Ziganshina for updating us on what’s happening in Russia.

    I’m fascinated by Andrew Robson’s insights into falls out of bed. There’s something about falls reporting that meets none of the other generalisations about incident reporting. Reporting falls out of bed is culturally normal in the NHS. It’s slightly weird that almost a third of all incident reports in the NRLS are about falls. What more can we learn? Is this because this is a non-blame area? I’d be interested to hear other’s views.

    Jenny Richards starts to touch on something critically important about formally logging error so it can be acted upon. I had the same type of incident once when a patient was put on a temporary ward, dropped their BM to 1 (almost arrested) and there was no emergency drugs available. A mad panic ensued with a student nurse being sent to a neighbouring ward to get some IV glucose stat! I foolishly didn’t fill in an incident report, choosing instead to report it verbally to one of the matrons and senior consultants. To my knowledge nothing happened. The lesson: fill in the incident forms because they are levers for action and ultimately will make patients safer.

  • Hello Douglas and everyone,

    As a practising consultant anaesthetist with a passion for this subject, I believe that if all healthcare workers and administrators better understood the concept of ‘human factors’ – the study of how humans perform in work environments -they would be better able to put blaming in context and see reporting for what it is – an essential tool for improvement.

    Human factors is in my view one of the ‘basic sciences’ of patient safety. Caring for patients without understanding it is a bit like performing surgery without knowing anatomy. Yet it is only a recent addition to some medical education curricula, and to my knowledge an uncommon component of ongoing inservicing programmes in public hospitals.

    What has human factors taught me?

    I has taught me that all humans make mistakes. Occasionally the worst mistakes are made by the best people. In the vast majority of cases this does not mean they are incompetent, rather the worker has found themselves dangling at the end of a chain of systemic errors and conditions that existed long before the adverse event occurred. These are the ‘accidents waiting to happen’ that most workers and administrators are aware of. And because individuals, however well trained and motivated, can never mount an impregnable line of defence against these latent errors, organisations would be better offinvesting their scant resources in improving the work environment rather than blaming the worker for imperfect performance. The organisation cannot learn,and individuals themselves cannot learn, if people do not report and reflect upon critical incidents. So in my view incident reporting is not optional.

    Human factors has not made me a ‘no-blame’ person. I believe all healthcare workers should expect that if they behave in a certain way in a given circumstance (e.g operating while drunk) they should be held to account for their actions, just as they would in their lives outside healthcare. What I have learned from aviation’s experience is that organisations have to separate ‘the people with the magnifying glass’ – the people investigating incidents – from ‘the people with the stick’. When the two groups ae mixed it becomes impossible to seperate analysis for learning from analysis for performance management. Aviation has also learned to treat vexatious or mischievous ‘incident’ reporting as a serious matter in itself. What we should expect is that people reporting incidents will be treated fairly and respectfully.

    For the last 10 years my organisation has been running 2-day courses on Human Error and Patient Safety (HEAPS) in Australia and overseas. The course was devised and run by practicising clinicians. Initially we had input from an aviation CRM trainer, but there is more than enough literature on human factors in healthcare to address all the key topics using relevant clinical examples- the psychology of error, teamwork, communication, situation awareness, control/command/leadership concepts, incident analysis and disclosure of adverse events. About 30% of our attendees are doctors, the rest mostly nurses, with a smattering of administrators and allied health personnel.

    We’ve run to course to well over 1200 workers now, and get particiapants to fill out an evaluation after each course. Over 98% of participants find the content useful and relevant, and 93% believed it would change the way they work, and that all their colleagues should do it. Now, of course this gives no indication of actual change in the workplace, not does it account for preaching to the converted, but perhaps with a properly constructed prospective pilot study in the UK, one might be able to determine whether such an intervention could actually alter attitudes to incident reporting, both among frontline clincians and administrators.

    Stavros Prineas (

  • The following website may be of interest to UK bloggers interested in human factors.

  • Dear Douglas and others,

    A very interesting discussion. Just over six years ago I had a significant delay in the diagnosis of breast cancer. An action was later raised and the case was settled out of court.

    Over the past couple of years I have taken an interest in and learnt a lot about patient safety and how despite the government report “An Organisation with a Memory” (2000) the medical profession has still not got to grips with learning from mistakes.

    I think that the way forward is to have a no-blame system with mandatory, anonymous reporting. I also think that any system of reporting must be integrated: anonymous reporting by the medical profession plus information from the medical defence unions plus information from GMC tribunals plus information from patients plus information from litigation. The last can be very informative. For example in the case of delay in breast cancer diagnosis there are the hundreds of actions raised over the last ten years to analyse. (Medical negligence lawyers will tell you that many cases they deal with are frighteningly similar.) There are also the major incidents that come to the attention of the press from time to time, for example: the East Devon Breast Screening Service (1990s); St Margaret’s Hospital Breast Screening Unit in Epping(2003-2004); Trafford Hospital Manchester and North Manchester General Hospital (2003-2005); Altnagelvin Hospital and Belfast City Hospital (2002-2005); Inverclyde Royal Infirmary in Greenock (2004-2006); Barrington’s hospital in Limerick (2005-2007); St Vincent’s University Hospital, Dublin (2008); Accrington Victoria Community Hospital (2008-2009).

    The playwright Tom Stoppard said that: ” It’s not the voting that’s democracy; it’s the counting. ” I would say that it isn’t just the reporting of mistakes that will ensure patient safety but the counting. At present the NPSA is awash with a complexity of data and is struggling to carry out any sort of meaningful analysis. We need to have a much more thoughtful report form with agreed terminology and key words.

    So where do we go from here? I am talking to BBC Newsnight in the hope that the programme journalists will examine the issues around learning from mistakes. We (doctors, patients, politicians, insurers, lawyers and even journalists) need to have an open, honest and constructive discussion about learning from medical mistakes. I think that it’s particularly important to get junior doctors on board. It may be that they haven’t made a serious error yet but they will, even the best of them. they will also be the senior medics of the future.

    So, Please talk to your colleagues and organisations about this. Ask them to contact Newsnight. Let’s spread the word, get the ball rolling, have a proper debate and learn from the mistakes.

    S Mitchell

  • Douglas Noble

    Thanks to Susan for this addition to the blog. Mandatory reporting is increasingly common, see table below:

    Mandatory reporting Voluntary reporting
    Holland Denmark
    Australia Ireland
    Slovenia Czech Republic

    It’s been documented that doctors tend to prefer voluntary reporting. There are problems with this, not limited to the epidemiological bias it presents. Getting an accurate idea of the magnitude of error is confounded. A parliamentary report indicated that at least 20-40% of errors in the NHS go unreported.

    Those in favour of a voluntary system are attracted to the depth of analysis. For example, near misses are probably more likely to be reported voluntarily. These are often the incidents that help organisations gain a tacit feel for risk and put barriers in place before adverse events occur.

    My personal view is that mandatory reporting is the way forward now that, for the most part, voluntary reporting is not succeeding. We have to watch out for error going underground, although we should be able to benchmark the information from a mandatory system against the data from the current system, at least ensuring that we are not loosing information.

    There is also a cultural oddity in the NHS, such that when it becomes absolute policy to act in a certain way everyone often does.

  • Sukhmeet S Panesar

    Dear Douglas and all,

    Thank you for encouraging a useful discussion (and debate) around patient safety, and in particular incident reporting.The National Patient Safety Agency houses and manages the largest voluntary, self-reporting database of patient safety incidents (PSIs), and as you point out we now have in excess of three million reports. No reporting system will be perfect, be it mandatory or voluntary, and the question we need to ask is whether a database of PSIs can act as a barometer of patient safety or will it only highlight rare cases. I do confess my conflict of interest here but will gladly admit that as an orthopaedic trainee (a year ago), I developed an almost anaphylactic response to clinical incident forms. The usual statements came to mind -”yet another piece of paper” and ”no one cares.”

    I have been extremely fortunate to work at the NPSA and all I find myself saying these days is ”If only doctors would report more and to a higher standard.” We can only act on the information we are given. Incident analysis and indeed patient safety as a discipline is its infancy as a science and it will take some time before we can change culture and indeed develop more robust analytical techniques.

    Let us remember that analysis of PSIs can indeed yield useful outputs such as our rapid response report on chest drains and the efficacy of hip cement. However, we have also been able to instigate significant change through small tweaks in the system. There was a time that there were >27 different crash call numbers and we indeed advocated and ensured that there is a standard number now – 2222.

    On a personal note, and it would be interesting to see how we can attract junior doctors (not a handful but most!) to the subject of patient safety. Perhaps:

    a) we need to have some mandatory teaching on patient safety in all hospitals and across all grades

    b) perhaps the structure of mortality and morbidity meetings needs to be altered and each team should have a system for discussing patient safety issues

    c) we need to create a Patient Safety in Action (PSA) team consisting of a risk manager, a senior clinician and a junior doctor who act as the patient safety anchorpoint for a team e.g. in orthopaedics, a consultant + ST + foundation + nurse could have an afternoon every week where they go through PSIs and areas of concern highlighted by their speciality in that hospital. The team should be able to fix some problems there and then – a common issue may be the lack of an ”SHO” to cover the ward and be on-call simultaneously. Simple, they could either try and distribute manpower evenly or get a locum! Some incidents will be serious and these need to be take further. So I suppose we should have swift local risk identification and reduction strategies and a copy of all these should go via the traditional route to national reporting databases

    d) we seem to love audits (I doubt most are meaningful as they are an exercise in compromise). Why can’t we get junior doctors to take on quality improvement projects and give them a sense of belonging to the team? I wish my consultant would say – ”We have a problem with DVT prophylaxis on this ward. I would like you to develop a project and the seniors will help you. ” You could develop a simple list or reminder whenever a drug chart is written and indeed a confidential register to report missed cases

    e) should we have an online dashboard or website showcasing the success stories of junior doctor involvement in patient safety and an open blackboard where we advertise opportunities for juniors to get involved

    These are fascinating times for junior doctors to engage with patient safety as a whole. Radical, thought provoking ideas are always welcome.