Transcript Document

the heart of health
and safety
London Metropolitan Branch
Presents
Please be advised that due to technical difficulties we are unable to provide audio recording for this event
But is it significant?
Bridget Leathley
[email protected]
Time is wasted insignificant hazards are risk assessed
and unnecessary controls
HS magazines examples FAIL to control significant
hazards –death.
How do we determine what is significant? Catch -22 - 1st
assess it?
If worst happens, how can we show consider significance
of hazards organisations?
Little help from legislation - good thing? allows us to work
within risk appetite of organisations.
Significant hazard –
significant risk?
• steps - significant hazard?
significant risk? , consider the
height, depth and width
• Old people - infant school play
ground, how likely is it that a
child will jump? fall? Hurt?
injury?
Legislation
Significant findings
Significant changes
• Management Regs
• Significant finding – MHSW 3(6)a – to be documented
• 3(6) Where the employer employs five or more employees, he
shall record (a) the significant findings of the assessment
• Significant changes as a reason to review risk
assessment- MHSW 3(3)b
• 3(3) Any (risk) assessment, shall be reviewed by the employer
or self-employed person who made it if ... (b) there has been
a significant change in the matters to which it relates.
ACOP (L21)
Risks – insignificant
Risks and hazards
Risks or hazards
•
L21 - cling - dog-eared copies (or pdf files)
•
ACOP 13 – “Once the risks are assessed and taken into account, insignificant risks
can usually be ignored, as can risks arising from routine activities associated with
life in general, unless the work activity compounds or significantly alters those
risks.”
•
Catch 22 - expected to assess risk - only then can we determine insignificance and
go on to ignore them. Still doesn’t tell us how to decide insig/ sig
•
ACOP 18 tells us that “A risk assessment should (a) ensure the significant risks and
hazards are addressed” and in same para tells us (e) “should always adopt a
structured approach to risk assessment to ensure all significant risks or hazards
are addressed” – is it AND or OR?
•
Bit more clarity about what significant findings (23, 25) are – eg that they should
include controls, but not about the essential problem of what was a significant risk
(or hazard)
•
Further paras refer to sig risk or sig hazard without explanation.
HSE website
“... not trivial ...
.. insignificant ... myth of
the month”
• Neither legislation nor ACOP tell us how to determine significance or
insignificance
• Nor clear whether significance overall risk, or sign hazard (ie S v L). HSE
FAQ
• “Risks, which are significant, are those that are not trivial in nature and
are capable of creating a real risk to health and safety which any
reasonable person would appreciate and would take steps to guard
against.
• What can be considered as "insignificant" will vary ... However, we have
highlighted some areas which can be considered as insignificant through
our Myth of the Month e.g. conkers, toothpicks, hanging baskets etc”
• Again 1st - negative examples – NOT trivial – NOT conkers, hanging baskets
or toothpicks Not much of a basis for determination
HSE website
... real risk ... which any
reasonable person would
appreciate ...”
• “Risks, which are significant, are those that are ...
capable of creating a real risk to health and safety
which any reasonable person would appreciate
and would take steps to guard against.”
• “any reasonable person” - really true?
• Specialised industries expected to understand >
man or woman on Clapham omnibus?
• “reasonable person” appreciate hazards of
industrial processes, lab chemicals, legionella,
asbestos, and so on?
Significant hazard –
significant risk?
• So, returning to our steps – do we
think they were a significant hazard?
• Highly likely, but severity low?
• Explain case and original court
judgement – reasonable people
considered significant
• Porter v R 2008/07
Is this significant?
15
• Audience asked to stand , Imagine or identify a hazard
• estimated L of hazard being realised
• estimated consequences
• LxC = 15 - Is that significant?
• Sit down if you would assess that as significant – stay standing if not.
• Does everyone standing think it’s not significant?
• Sit down now if you would assess that as insignificant.
• Anyone still standing? Yes – why?
• Didn’t have enough information to determine whether or not it was
significant.
Severity
Is 15 significant?
5
5
10 15 20 25
4
4
8
12 16 20
3
3
6
9
12 15
2
2
4
6
8
10
1
1
2
3
4
5
1
2
3
4
5
Likelihood
15
• 5 x 5 - looks significant.
• ? High L / Mod S
• ?Mod L / High S Are both 15s same?
• School regarded the likelihood of a high severity outcome as
very low.. Should they have placed more emphasis on S, RP gate?
• Should we give more priority to High Sev? Others will argue
we should focus on the hi- frequency events.
• This is an important point we’ll return to when we consider an
alternative to this sort of grid.
Severity
Is 15 significant?
5
5
10 15 20 25
4
4
8
12 16 20
3
3
6
9
12 15
2
2
4
6
8
10
1
1
2
3
4
5
1
2
3
4
5
Likelihood
Severity
Is 15 significant?
7
6
5
4
3
2
1
7 14 21 28 35 42 49
6 12 18 24 30 36 42
5 10 15 20 25 30 35
4 8 12 16 20 24 28
3 6 9 12 15 18 21
2 4 6 8 10 12 14
1 2 3 4 5 6 7
1 2 3 4 5 6 7
Likelihood
• But did I say we were using a 5x5
matrix? What if we were using a
7x7 matrix? Is 15 significant now?
• Where I’ve put the boundaries, it
certainly looks less significant than
before
Is 15 significant?
Severity
10 10 20 30 40 50 60 70 80 90 100
9
9 18 27 36 45 54 63 72 81 90
8
8 16 24 32 40 48 56 64 72 80
7
7 14 21 28 35 42 49 56 63 70
6
6 12 18 24 30 36 42 48 54 60
5
5 10 15 20 25 30 35 40 45 50
4
4
8 12 16 20 24 28 32 36 40
3
3
6
9 12 15
15 18 21 24 27 30
2
2
4
6
8 10 12 14 16 18 20
1
1
2
3
4
5
6
7
8
9 10
1
2
3
4
5
6
7
8
9 10
Likelihood
•And what about 10x10?
Now, with scores of 100
possible, 15 might be in
the acceptable region

what we’ve showed so far may be quite obvious:
 If you don’t agree on the grid you’re using, you have no basis to assess
whether a hazard is significant or not.

But how many of your organisations have sat down and agreed what the
matrix should look like?

As we’ll see, the question is not just how many categories we should have.
Significant risk
Significant
Insignificant
Routine
Material
Unprecedented
Obvious
Fanciful Hypothetical
Appreciable
Trivial Negligible
Foreseeable
Reasonably foreseeable
Problem with judges is that they don’t always use the same language as the
regulations. Whilst we are thinking about :
Significant
Foreseeable
and
or
Insignificant
and considering what is
Reasonably foreseeable
Judges seem to prefer a range of other terms
Routine; Unprecedented; Material; Obvious; Appreciable; Hypothetical;
Negligible; Trivial
And my favourite, least legal sounding but frequently used term “Fanciful”
Significant risk
Material
Obvious
Fanciful Hypothetical
Appreciable
Trivial Negligible
Foreseeable
HSE
Highly unlikely
Slightly harmful
Unlikely
Harmful
Likely
Extremely harmful
In FAQ section of HSE website
Admire simplicity - 3 x 3 grid
I don’t think you would ever get a bunch of risk assessors in a room to agree
what ‘slightly’ and ‘extremely’ mean.
Experiment - Masters Class - research methods
asked about a range of terms used to describe L
Context of mother - child with downs syndrome.
Range of answers for ‘highly unlikely’ was 1 in a million births to 1 in 10 births.
Clearly these are terms that even the educated will misunderstand.
Severity
1
First-aid up to 7 days
2
7 days+
3
Single death or permanent
disablement
4
Multiple fatalities
Instead – define cats meaningful.
The severity is probably. easier to picture and describe
Assume simple matrix for our example – 4 x 4
?Realistically - kill one person at a time, might use 4 = single death this
example assume multiple.
Often grids no head room for multiple deaths – single deaths (eg a roadside
worker) treated the same as a multi-car pile-up leading to many deaths and
injuries
Could argue – SHOULD argue
NOT presenting this today as a scheme that everyone should use. hope you
see why this won’t work - ?need an extra cat? Fewer cats?
DISCUSS - AGREE
Likelihood
1
Once in 100 years or less
2
> 100 years, < 10 years
3
> 10 years, < 1 year
4
More than once a year
Important to decide Likelihood refer
• per task
• per event
• per day
• Per lifetime.
Task going wrong is 1:100 each time the task is done, but task 100
times a day across an organisation suddenly your likelihood sounds a
lot higher.
In example, I’ve compounded information about how often a task takes
place into a simple measure of how often we would expect the
hazardous event to be realised across the organisation
So, now we have a scale, let’s try and apply it
Is the risk significant?
• Even then, its difficult to get agreement
• Is the risk of one of your drivers having an
accident whilst driving for work Significant?
• Let’s consider 3 risk assessors sitting in a
room discussing this using our new grid.
Option 1
4x1=4
• Hazardous event – multi-car pile up – using our
earlier scheme, severity 4 (multiple deaths)
• Likelihood – fairly low – looking at company records
over a period of time
• no one has actually been involved with that sort of
accident despite the miles driven,
• let’s say we don’t expect it to happen more than
1:100 years – call that a 1
• 4x1 = 4
Option 2
1x4=4
But second risk assessor says ah, most likely accident
in our record is a shunt in a town, with whip lash as
the consequence, back at work within a couple of days
That’s a 4 for likelihood – across the fleet it happens
around once a week.
But only a 1 for consequence – That’s 1 x 4 = 4
Option 3
3x3=9
3rd assessor
“but under estimates average risk of driving for work
-frequent minor accidents
- infrequent major accidents
– but what’s the most likely worst-case scenario?”
1/ year driver knocks over a pedestrian, occasionally l> death, mostly injury.
Severity = 3 and Likelihood = 3 Risk = 9
Consider publicity – public opinion appears to accept motorists killing or
injuring other motorists.
But if one of your livery vehicles knocks over a pedestrian – particularly a child
or the elderly – that has more implications beyond a simple harm x likelihood
equation.
3 opinions
Severity Likelihood
Risk
4
1
4
1
4
4
3
3
9
Let’s summarise those opinions
Are these 2 4’s equally as significant (or insignificant)?
If we listened to risk assessor 1 and 2 we might decide driving for
work was insignificant. If we listened to assessor 3 we might have a
different opinion.
Should we do more about one than the other? The problem with
numbers is that they don’t tell us anything about the detail.
Small change = big impact
Severity Likelihood
Risk
4
1
4
2
4
8
3
3
9
What if we discovered in most cases the whip
lash led to more than 7 days off work?
Does that mean the risk is now doubled? No,
because these numbers don’t mean that – an 8
isn’t twice the value of a 4, just more (probably)
than 4.
Really a difference in risk between 8 and 9?
Severity
The problem with numbers
4
4
8
12
16
3
3
6
9
12
2
2
4
6
8
1
1
2
3
4
1
2
3
4
Likelihood
Of course, it depends where I put the
boundaries. I could have chosen 4 as the
cut off rather than 6.
Similarly at the top end I could have chosen
9 instead of 12. But this is the point – these
decisions are so arbitrary, do they really
help us to assess if a risk is significant?
Severity
R=S+L
4
5
6
7
8
3
4
5
6
7
2
3
4
5
6
1
2
3
4
5
1
2
3
4
Likelihood
Why Severity x Likelihood?
Why not add them together?
Boundaries in right place -> the same grid as S x L
Does not take as very far other than realising there is
nothing special or sacrosanct about multiplying them the
way we do.
Or weight one over the other – Charities commission
Severity
R = S x L + 2S
4
12
16
20
24
3
9
12
15
18
2
6
8
10
12
1
3
4
5
6
1
2
3
4
Likelihood
Why Severity x Likelihood?
Weight one over the other – Charities commission.
Low likelihood high severity events
weighted more
Hi likely low severity events
Small chance of killing a member of the public is considered organisationally
more significant than the high likelihood of many staff suffering from MSD as a
result of manual handling or DSE use.
Grid that works for your organisation?
numbers are still arbitrary.
Why go to the difficulties of using numbers and all they promise but can’t deliver?
Let’s get rid of the numbers altogether
Return to Porter case – although jury considered it
significant, both original judge and appeal judge didn’t,
and eventually teacher was acquitted.
But it shows us we can’t rely on “reasonable person”
judgement, and that we have to be having these
conversations within our organisation.
No numbers
Multiple
deaths
Severity
Single
death/ dis
7 day +
First-aid
< 7days
<100 10-100
1-10 yrs 1 yr+
yrs
yrs
Likelihood
Let’s put in our descriptive categories instead of
the numbers
don’t have to agree - surprised, nay
disappointed, if all of you did disagree and
argue, constructively, in your organisations
What is significant across your organisation?
Is it significant?
Multiple
deaths
Single
death/ dis
7 day +
First-aid
< 7days
<100 10-100
1-10 yrs 1 yr+
yrs
yrs
Quickly decide some - not going to spend long controlling hazards where the
consequences are first aid – 1:100 years.
How often FA before significant? chain of shops - tagging gun incidents every day -> sig
argue anything between is to be controlled where RP
Deal with any multiple deaths, however infrequent
1:100 year flood conditions two years in a row.
accept a single death or disability every 100 years, but no more
want to do something about it - middle region
How often we accept 7 day injuries depend on your industry
Construction - every year
Other industries - sign if > 1: 10 years.
Judgement - think about examples in organisation.
whip lash – couple of years – acceptable miles driven?
APPETITE for incidents v willingness to spend £££ preventing
More appetite...
Multiple
deaths
Single
death/ dis
7 day +
First-aid
< 7days
<100 10-100
1-10 yrs 1 yr+
yrs
yrs
Higher risk appetite might look like this
Important - result doesn’t have to be symmetrical
Significance - other business risks – reputation
What is Significant?
Questions?
• Not defined for us by HSE – and perhaps it is best that it isn’t
• Face the facts - courts inconsistent
• Good news for sensible health and safety is that the head teacher was
acquitted on appeal – a judge could see... Quotes from case summary
here...
• Away from numbers, back to what we’re trying to assess.
• What significance if a given hazard occurred.
• Despite acquittal staff at the school where Kian died would still rather
have prevented the accident that led to his hospital treatment that led to
his death. It’s not just about compliance.
• Thank you