Memorandum by the Association of Anaesthetists
of Great Britain and Ireland (AAGBI) (PS 63)
1. What the risks to patient safety are and
to what extent they are avoidable, including:
"It is not you that I don't trust, it
is your species" All human beings will continue to make
errors, so we must study and manage them. Reduction of harm to
patients, not just the elimination of error, must be the primary
Human factors are reported in 75% of aircraft
incidents, and will be similar in patient safety incidents.
A feature of errors in medicine as distinct
from errors in most other situations is that the patients involved
often have coexisting medical diseases which add extra hidden
opportunities for harm to arise.
Clinical practice can never be risk-free. Processes
should be designed to reduce patient harm.
The public should be given more credit for commonsense
and information provided for them.
2. What the current effectiveness is of bodies
in ensuring patient safety:
NPSA are working with anaesthetists to pilot
new specialty specific incident reports for anaesthesia.
Evidence from the ISTC programme shows that
safety is not their first priority.
The process in "Safety First" for
investigating incidents needs further development.
Each trust should have a patient safety budget
to spend on new initiatives each year.
Funding for Standards work should be included
in future Patient Safety spending plans.
The MHRA has been very effective with the safety
and design of anaesthetic apparatus.
NHS PASA should use its purchasing power along
with the NPSA "Purchasing for safety" policy.
NCEPOD recommendations have produced significant
advances in patient safety.
AAGBI recommendations for minimal monitoring
standards resulted in all NHS patients having continuous monitoring
during anaesthesia. Such recommendations have clinical "buy
in" to achieve widespread implementation.
The Royal College Hospital Visiting Programme
played an important role in Patient Safety and should be reinstated.
The NHSLA does not cover all NHS patients having
operations in the independent sector. It is suggested in the new
Health Bill currently in process but the detailed wording needs
to guarantee it.
NHS Study Leave budget should be ring fenced,
inflation linked and not diverted to cover deficits as in 06-07.
3. What should the NHS do next regarding
Consider lead clinicians for Patient Safety
in Trusts and patient representatives on clinical governance committees
Audit/clinical governance meetings held with
similar groups from neighbouring hospitals are extremely valuable.
In connection with The World Health Organisation
(WHO) World Alliance for Patient Safety Global Challenge "
Safe Surgery Saves Lives" the NHS should routinely collect
the required data.
1. The AAGBI is the senior body representing
anaesthetists with 10,000 members including most of the anaesthetists
in the UK, the largest group in the hospital medical workforce
(16%). The organisation has been interested in Patient Safety
since its foundation in 1932, formally constituted a Safety committee
in 1976 and has initiated many safety developments in the specialty.
What the risks to patient safety are and to what
extent they are avoidable
2. The level of the risk has been estimated
nationally in several western countries. In the UK studies give
the percentage adverse event rate per hospital admission of around
10%. It is thought that about half of these adverse events can
Role of human error and poor clinical judgement
3. "to err is human" (
Alexander Pope) and it has been said "It is not you that
I don't trust, it is your species" All human beings will
continue to make errors, so we must study them, and manage them
to reduce the harm they do to patients and their lower their number.
Reduction of harm to patients, not just the elimination of error
(as there will always be some, it can never be completely prevented)
must be the primary focus.
4. Human factors (HF) are reported in about
75% of aircraft incidents, and it is likely to be similar in medical
patient safety incidents. (HF) are features that make us different
from logical, completely predictable machines and aspects that
affect our personal performance. Airline pilot Martin Bromiley's
wife Elaine died following a failed intubation in 2005 and as
a result Martin Bromiley set up a Clinical Human Factors Group
(CHFG) of experts to promote HF Training to all healthcare staff.
A video presentation on the tragic incident "Just a Routine
Operation" can be seen on http://chfg.co.uk/resources.htm
5. HF training has improved air safety and
Martin Bromiley is someone the committee may wish to speak to,
the CHFG group details are on http://chfg.co.uk/index.htm
6. The extent of the role of poor clinical
judgement is also not known but one of the distinguishing features
of errors in medicine as distinct from errors in most other situations,
which usually have fixed variables, is that the patients involved
often have coexisting medical diseases which can add extra hidden
opportunities for harm to arise. Clinical judgment can be poor
in making the wrong decision about the primary diagnosis out of
lack of knowledge or turn out to be poor from its result on the
How far clinical practice can be risk-free; the
definition of "avoidable" risk; whether the "precautionary
principle" can be applied to healthcare
7. Professor James Reason has said end users
often inherit the errors of others earlier in the chain. The reports
on the Intrathecal Vincristine incidents from both Nottingham
and Great Ormond Street Children's Hospital and show that the
risks were being taken at almost every step of the pathway. A
lot of these are avoidable with careful analysis of the whole
system and an organisational culture in place which is prepared
to act and fund the appropriate measures. Currently systems in
the NHS are only subject that this type of analysis once an incident
has occurred and ideally all systems identified as high-risk should
be subject this process on a regular basis whether or not they
have failed recently. Learning from other parts of the NHS should
help, as implied in "An Organisation with a memory"
but this does still not happen. Enabling departments in neighbouring
trusts to meet together on a regional basis on their monthly audit
days would be one way of spreading information. This could easily
be enabled at no cost by just synchronising the audit dates within
a region at the beginning of the year. There are isolated areas
of this type of good practice in the NHS eg the Northwest Regional
cardiac surgical audit where the four regional cardiac surgery
units of meet and discuss their combined data on a six monthly
8. Clinical practice can never be risk-free
because it is a human activity and because of the various patient
variables that may be unknown, as mentioned above. The healthcare
processes however should be designed to reduce harm to the patient
whatever happens, so that whenever a risk becomes a reality and
threatens the patient the safety nets are in place to reduce the
effect and ameliorate any resulting harm to the patient.
The role of public perceptions of risk in determining
9. The public should be given more credit
for commonsense and information provided for them to understand
potential risks. The recent NHS policy recommendations for clothing
seemed to be driven by perceived public perceptions rather than
what evidence exists at the moment. Indeed one of the main bullet
points on page 3 said there was no evidence that clothing had
any influence on these matters! Such documents also affect healthcare
staff's "buy in" to such documents and future credibility.
Medical staff in particular are used to reading and assessing
papers published in medical journals which have been peer-reviewed
and written from a quantum higher-level when they are having to
make accountable decisions.
What the current effectiveness is of bodies in
ensuring patient safety:
10. Local and regional NHS bodies are not
generally effective in ensuring patient safety.
11. We are not aware of any Boards of NHS
bodies that have established a safety culture. Whenever a new
service initiative is proposed the first document draft usually
puts patient safety at the top. However as the iterations take
place and the potential cost becomes apparent patient safety slips
down the list and is rarely considered in the final analysis.
Sir Liam Donaldson's suggestion that safety should be first and
continue to remain so has still to be implemented.
Systems for incident reporting, risk management
and safety improvement
12. Systems for incident reporting had been
well developed in Australia. Their "Anaesthetic incident
monitoring system" (AIMS) system produces regular reports
and seems to be effective in influencing practice there and wider
afield including the UK. The National patient safety agency (NPSA)
initial reporting system is cumbersome to use and very difficult
to analyse. It has been extremely difficult to extract meaningful
data about anaesthetic incidents from the NPSA system and now
the Royal College, Association of Anaesthetists and the NPSA are
now working together to pilot then introduce a newly designed
specialty specific section for anaesthesia. So far the pilot project
is going well and starting to identify trends.
13. Risk managers appointed in trusts often
do not, or cannot, take action to address the root cause of incidents.
More training in root cause analysis or similar methods is needed
for both clinical and non-clinical managers.
The impact of the changing public-private mixing
14. The private sector has necessarily different
objectives to public bodies particularly the requirement to provide
a financial return to their investors. Evidence from the independence
sector treatment centre (ISTC) programme shows that safety is
not the first priority in their employment of staff, clinical
governance, provision of equipment, and culture. This was all
anticipated but the necessary structures to supervise and enforce
the required standards in ISTCs were not put in place. A long
established lesson in business is that outsourcing companies primarily
work for themselves, not their paymaster. This is an example of
the NHS half taking on a superficially attractive idea from business
without fully thinking it through and addressing the potential
downsides from the outset. The NHS money expended here could have
funded a lot of Patient Safety initiatives.
15. The objectives set out in "Safety
First", which is a good document, are appropriate. The process
for investigating incidents was unclear and more progress needs
to be made on this.
Whether past spending on patient safety has been
sufficient and cost-effective and what future spending should
16. We are not aware of figures on past
spending on patient safety; however the feeling is that it has
been insufficient and patient safety initiatives have repeatedly
failed through lack of funding at trust level and above, for example:
17. 10 years ago reports in Australia showed
that if a single patient use breathing filter were not used there
was a risk of transmitting hepatitis between patients on a surgical
list. One trust presented the business case for single use filters
to their chief executive. He said that the anaesthetic department
should find the additional funding from savings in their own budget.
This was not possible so it did not take place. Later a patient
with tuberculosis potentially infected a whole list of patients.
All patients were recalled, counselled and tested and one patient
later sued the trust for the distress caused whilst they waited
six weeks for what turned out to be a negative test result. None
of the patients had become infected. Single patient use breathing
filters later became the national standard.
18. Each trust should have a patient safety
budget to spend on new initiatives each year which often only
require pump priming. They should have to account for what they
have spent it on in the trust annual report.
19. In parallel with our Safety Committee
the AAGBI holds a Standards Committee. Every anaesthetist who
sits on a British or International Standards Committee for anaesthetic
related equipment attends and this is the only forum where they
can meet to discuss common issues. In this way independent clinicians
can feed into the standards groups which draw up the standards
equipment manufacturers must follow in future making sure they
fit for NHS and Patient Safety requirements. A recurring problem
however for these 12 individuals is obtaining the travel expenses
to attend BS and ISO standards meetings. This money was held and
distributed centrally but in the 1990s it was "devolved"
down to trust level and as a consequence individual trusts do
not understand this or the highly specialized activity that these
doctors do often in their own spare time. Reverting to the previously
successful arrangement where expenses (not a large sum) could
be allocated centrally seems the only reliable solution. This
activity is very cost effective for the clinical input and influence
it provides and the AAGBI would ask that it be included in future
Patient Safety spending plans.
20. National targets should be used sparingly.
The AAGBI have produce over 50 documents nicknamed "glossies"
which give detailed guidelines and recommendations for providing
a safe service. These are freely available on the website www.aagbi.org.
Audit standards for local comparison are also available in the
Royal College of Anaesthetists audit recipe book "Raising
National Patient Safety Agency (NPSA)
21. The NPSA has become more effective recently,
issuing guidance notes etc. It misses some opportunities to be
more effective by not being more robust when appropriate. Its
recent recommendations on design of drug packaging recommend printing
the drug name on the five sides of the box, why not all six! Then
whichever way the box is randomly placed in a cupboard the name
can be read.
The Medicines and Healthcare Regulatory Agency
22. The MHRA has been very effective with
safety and design of anaesthetic apparatus. They investigate incidents
promptly and issue Hazard notices when required which get down
to the front line and are discussed. They have a track record
of competence, dealing sensibly with industry and an organisation
that the committee may like to contact.
The NHS Purchasing and supply agency (PASA)
23. The NPSA has recommended a "Purchasing
for safety" policy for the NHS: whenever a choice has to
be made between two pieces of equipment or pharmaceutical products
the safest one should always be purchased. The NHS rarely uses
its purchasing power as it could and should. PASA is now setting
up product councils involving clinicians and this may help. It
could also invite tenders or produce for itself difficult to source
items or "orphan" drugs eg. the old but still essential
vasoconstrictor drug "Metaraminol" is now produced by
the Devon Healthcare Trust for the whole NHS.
The National Confidential Enquiry into Patient
Outcome and Death and (NCEPOD)
24. NCEPOD is an independent body originally
set up by the AAGBI and the Association of Surgeons. It has gained
the confidence of the profession through its confidential processes
and periodic reports. As a result of its recommendations patient
safety has seen significant advances. Each NHS trust now has a
dedicated emergency theatre and out of hours about operating has
largely been reduced to only essential life and limb saving procedures.
The Association of Anaesthetists of Great Britain
and Ireland (AAGBI)
25. In 2007 the AAGBI 75th anniversary strap
line was "75 years advancing patient safety". The Safety
committee was formally founded in 1976 and remains the national
forum for patient safety connected with anaesthesia having representatives
from the RCA, MHRA, NPSA, and MDOs. The AAGBI has produced over
50 patient safety guidelines. In 1986 the AAGBI recommendations
for minimal monitoring standards resulted in all NHS patients
having continuous pulse oximetry, blood pressure, ECG and end
tidal carbon dioxide monitoring used on them during anaesthesia.
This was achieved over the following 5 years and verified by the
Royal College of Anaesthetists (RCA) Hospital Visiting Programme.
This demonstrated one of the values of these RCA visits and they
should be reinstated
26. In 2004 the AAGBI recommended the use
of syringe labels for intravenous drugs in the international colour
codes. A national audit in 2005 showed over 95% of NHS hospitals
using them. Such patient safety recommendations by the AAGBI as
a professional body have credibility and clinical "buy in"
to achieve widespread implementation quickly at low cost.
NHS litigation authority (NHSLA)
27. This could be used to provide more information
about patient safety if it was able to analyse its database and
publish closed claims reports. Currently its database was not
designed for this purpose but could be changed in future.
28. Currently the NHSLA does not cover all
NHS patients having operations in the independent sector. The
AAGBI, BMA, surgical and patient organisations agree that it should.
It is in the new Health Bill currently in process but the detailed
wording needs to guarantee it.
Education for health professionals
29. Patient safety is now a topic in this.
NHS Study Leave funding budgets should be ring fenced; inflation
linked and not be removed to cover deficits elsewhere in the NHS
as occurred in 2006-07.
What should the NHS do next regarding patient
Whether the measures taken to improve patient
safety are supported by adequate evidence regarding their clinical
effectiveness and cost effectiveness
30. The medical profession is rightly concerned
and trained from an early stage to always seek out the necessary
evidence of clinical effectiveness. Some observers think that
the medical profession almost unique in this respect but in serious
questions of patient safety some common sense action should be
taken pending the arrival, if ever, of such evidence. This is
particularly applicable to rarely occurring hazards eg the safe
administration of Vincristine. Making Vincristine only available
in 50 ml minibags would significantly decrease the chances of
it ever being given intrathecally. Local anaesthetic toxicity
is another example. In 2007 the AAGBI recommended the use of Intralipid
20% during resuscitation based on rats experiments in 1998. There
are now several articles in the world literature reporting that
lives have been saved after overdoses using this technique.
How to determine best practice and ensure it is
spread throughout the whole of the NHS
31. Setting up pilot projects to try new
practice and then reporting the findings in the usual way still
works. MHRA and NPSA notices must be brief and not overused to
overload clinicians get ignored. Having a lead clinician for Patient
Safety in appropriate areas that can sift the plethora of information
and direct it may help.
How to ensure that learning is implemented
32. Regular audits are a way and every trust
/ specialty could be asked to do one or two specific audits every
year. Also audit/ clinical governance meetings held with similar
groups form neighbouring hospitals are extremely valuable.
What should be measured and assessed and what
data should be published.
33. Any data published should be of good
quality and validated. The information published in connection
with cardiac surgery in the United Kingdom is appropriate and
may be extended to other specialties. National safety audits proposed
earlier could be made public when they are topical for that particular
34. In connection with The World Health
Organisation (WHO) World Alliance for Patient Safety Global Challenge
" Safe Surgery Saves Lives" the NHS should routinely
collect the required data. to add NHS data to the WHO database
and make meaningful comparisons internationally
What incentive should there be to improve patient
35. The medical profession does not require
great incentives to improve patient safety but there is obviously
some inhibition taking place in the management structure. It may
be that any savings could be returned to the appropriate units
after implementing safety changes.
How patients and the public can be involved in
ensuring that services are safe.
36. Some trusts have patient representatives
on their clinical governance committees and this experience should
be extended. Patient information could be developed to explain
that some processes such as having to wait untreated and then
being treated in a published order are actually safety measures
and defences against them being harmed. Being asked six times
by different members of staff and which leg they wish to have
removed is not necessarily an incompetent system just a safety
process of multiple checking.
37. We would ask the Health Committee to
take this evidence into account in their inquiry into Patient
Safety which we continue to see as a priority for healthcare worldwide.