A5 : LOSS CAUSATION AND INCIDENT INVESTIGATION

PART 1 – THEORIES OF LOSS CAUSATION

Learning outcomes
Explain theories of loss causation
1. Define an accident
An accident is an unplanned, unexpected, and undesigned (not purposefully caused) event
which occurs suddenly and causes injury or loss, a decrease in value of the resources, or an
increase in liabilities. It is to be noted that an accident can be prevented if circumstances
leading up to the accident are recognized, and acted upon, prior to its occurrence. Losses
result from lack of control and are revealed by loss causing events.
2. Define an incident
An Incident is an unplanned, undesired event that hinders completion of a task and may
cause injury, illness, or property damage or some combination of all three in varying degrees
from minor to catastrophic.
3. What is a near miss?
The HSE considers a ‘Near Miss ’ to be an event that, while not causing harm, has the potential
to cause injury or ill-health. They take the term to include dangerous occurrences specified in
the Reporting of Injuries, Diseases and Dangerous Occurrences Regulations (RIDDOR) 1995.
The difference between a near miss and a fatal accident in terms of time and distance can
be very small indeed. It is therefore clear that the damage to persons or properly is not the
accident, but part of the effects of the accidents (i.e. the result or consequences).
4. What is an undesired Circumstance?
In the view of the HSE the term incident includes ‘Undesired Circumstance’, which they take to
be: a set of conditions or circumstances that have the potential to cause injury or ill-health,
e.g. untrained nurses handling heavy patients i.e. the potential for back injury.
The HSE states that an accident includes “Near miss” and Undesired circumstances””.
The HSE have set out an interesting perspective in their guidance document HSG245, which
will generate some debate. Some practitioners may feel a definition of accident that limits
itself to outcomes that harm people to be too narrow, others may find it useful to have a
focused definition, for reporting and analysis purposes. It is important to remember that
HSG245 is a guidance document and carries no direct legal duty to follow or use the
definitions contained it.
This definition encompasses events that result in a wide range of losses and has, for a long
time, helped to provide a good perspective of events with different outcomes. This has
assisted greatly in encouraging people to learn from events and the subsequent prevention of
accidents. The definition therefore, includes ‘near misses’, i.e. where no injury or damage etc.
occurs. It is important not to think of injuries, damage and other losses as accidents, but rather
as the results of accidents.
5. Give four possible consequences that may arise from a spanner falling down from a height.
The following accident model is offered to illustrate the above statement: a spanner falls from
a height. The following consequences could result:
1. The spanner falls into a pile of sand and there is no damage or injury.
2. The spanner hits an item of equipment, resulting in damage, but no injury.

1
 3. The spanner strikes a person causing a cut and bruising to the hand, this is an injury
accident.
4. The spanner strikes a person working directly underneath causing a fatality.
6. Why is it important to investigate a near miss?
An old adage says never waste an accident. Apart from being unpleasant and perhaps very
costly, every accident constitutes an opportunity to correct some problem. For this purpose, a
near miss which has the potential to cause loss is just as valuable as a serious injury/damage, in
fact even more important if we are to avoid a future loss incident, a golden opportunity not to
be missed.
As described in HSE booklet HS (G) 65 ‘Successful Health and Safety Management’, if a man
slips on a patch of spilled oil, he may be unhurt, he may damage clothing or equipment, he
may break his arm or he may fracture his skull and die.
Accident ratio studies and their use and limitations
7. Who was HW Heinrich?
HW Heinrich, an American safety engineer, proposed one of the first coherent theories of
accident causation in the mid 1920s. He suggested that accidents were not ‘acts of God’ but
were caused by the failures of people.
THE DOMINO THEORY
8. In brief, what was his theory of accident causation?.
His domino theory suggested that the series of events, which led to an injury or some other loss,
were a succession of events, which followed a logical pattern.
His domino theory suggested that the series of events, which led to an injury or some other loss,
were a succession of events, which followed a logical pattern.

1
3 5
2 4

9. Which series of events leads to an accident according to him?

According to Heinrich, the series of events which led to accidents were:
1. Social or family defects
2. Personal defects
3. Unsafe conditions or behaviour
4. Accidents
5. Casualties

2
If any piece falls down, it will cause a chain reaction of dominoes, and result in an accident.
The accident may not occur if the third piece of unsafe conditions or behaviour is removed.

In 1969, Frank E Bird analyzed 1,753,498 accidents, in 21 industries reported by 297 companies,
and he showed that there is a fixed ratio between losses of different severity (and accidents
where no loss occurred, i.e. near misses).
The Frank Bird accident ratio study can be demonstrated with a pyramid model:

1 Serious / Disabling Injury

10 Minor Injuries

30 Property Damage

600 Near Misses

There have been several versions of the accident pyramid, with some in HSE publications, e.g.
HSG 96 “The costs of Accident at work”.

Over-3-day injury accident

1

Minor injuries (e.g. first aid only)

56+

Non-injury accidents
3570+

Accident ratio pyramid. Source: HSG96 the costs of accidents at work.

In 1974, Tye and Pearson studied about one million accidents in British industry and suggested
that for 1 fatal or serious injury, there would be 3 minor injuries with up to 3-day work absence,
50 injuries requiring first aid treatment, 80 property damage accidents and 400 non-
injury/damage accidents.

USE
10. What conclusions can be drawn from the accident triangle or pyramid?
The models illustrate that if limited interest is taken in the full range of events that occur, such
that only those resulting in injury are considered, many opportunities to learn about what goes
wrong are being missed. If near misses are also studied they can provide more opportunities to
learn and possibly prevent some of the events that result in injury.
The Bird model includes property damage in addition to near misses, and again if measured,
analysed and acted on this will help to prevent the injuries events.

3
The models help to convince people of the value of reporting a wide range of events and
show that there are usually more near misses than injury events, which provide more
opportunities to learn and improve health and safety.

LIMITATIONS
11. What are the limitations of using these accident triangles or pyramid?
These results show that Bird’s findings are not uniform throughout industry and in fact differ from
industry to industry. This may be mostly due to the range of risk involved, but there could be
cultural differences, and the level of reporting.
The accident ratio studies may not necessarily show the extent of the loss to the organisation.
For example, the “property damage” category may include extensive damage to large plant
and equipment. Also, shown from Bird’s and HSE’s examples, there are no universally agreed
of each subset of accident type.
Statistical analysis of loss events relies on large numbers, comparable work and worker skills
over the measured time-frame, to be effective. In smaller organisations, the first recorded
accident may be the top event, i.e. fatality or major injury. Near miss reporting may mean
different things to different people, for example, an office employee who enters a production
plant will have a higher perception of danger than a worker who has been at work there for
several years.

Amended Domino and multi-causality theories

Further research by the International Loss Control Institute (ILCI) (and Frank E Bird) into accident
causation led them to put forward a modified domino theory:

12. How was the Heinrich Domino theory modified by Bird and his team?
Bird considered that there were 5 stages represented by 5 standing dominos, namely:
1. Lack of Management Control
2. Indirect Causes
3. Immediate Causes (Symptoms)
4. Event (Contact)
5. Injury/ Damage (Loss)

Let us consider each stage of the Frank Bird – LLCI domino separately (starting from the last),
namely:
“Loss”
13. List four components which fall under “Loss”.
This is the consequence of the accident and can be measured in terms of people (injuries),
property (damage) or loss to the process (failed telecommunication) or the potential for any
of these and hence loss of profit.

“Event (accident or incident)”

The event producing the loss involving contact with a substance or source of energy above
the threshold limits of the body or structure.

14. What do you understand by immediate causes.

“Immediate (direct) causes” – relate to what happened (Premises, plant, procedures, People)
These are the substandard (unsafe) acts (e.g. using tools and equipment for tasks they were
not designed to do) and substandard (unsafe) conditions (e.g. a trailing telephone cable in
an office), which give rises to an accident. These are symptoms, which can be observed.
Whilst these symptoms cannot be ignored, action solely at this level will not, by itself, ensure

4
that recurrence is not prevented. Unsafe acts and conditions may be considered as
workplace hazards.

15. What do you understand by underlying causes?

Underlying (indirect or root) causes (relating to why it happened – P, RA, 4C, M, R)
These are the underlying or root causes of accidents. Identifying these causes will explain why
the substandard act happened or the condition arose. They are not always easy to identify.

16. What are the 3 major categories under which underlying causes fall?
Underlying causes fall into three major categories:
 Organisational Factors (Procedural).
 Job Factors (Technical).
 Personal Factors (Behavioural).

17. List the items which falls under organisational factors

Organisational factors
 Work standards and procedures
 Communication
 Co-ordination
 Supervision

18. List the items which falls under job factors

 Design of equipment and layouts
 Maintenance
 Purchase of materials and equipment

19. List the items which falls under personal factors.

Personal factors include
 Physical capability
 Mental capability
 Physical stress
 Mental stress
 Knowledge
 Information
 Skill
 Motivation

20. What do you understand by Lack of Management control?

“Lack of management control”
This is the initial stage, centred on the management functions of:
 Policy
 Planning
 Organizing
 Controlling
 Monitor
 Review

In the HSE’s guidance document on the investigation of accidents and incidents, HSG245,
their perspective of the domino theory is set out. “Each domino represents a failing or error
which combines with other failings and errors to cause an adverse event. Dealing with the

5
immediate cause (B) will only prevent this sequence. Dealing with all causes, especially root
causes (A) can prevent a whole series of adverse events”. This is an important aspect of loss
causation/prevention which is why thorough investigation to determine root causes is
necessary.
In this guidance document the HSE explain their view on what is an immediate, underlying and
root case, HSG245 should be read with care as some of the data on what is meant by these
different causes appears in more than one place and is contradictory.
The meanings that follow good practice conventions have been selected from the document
and are set out below:

21. Distinguish between immediate causes, underlying causes and root causes.
“Immediate cause: the most obvious reason why an adverse event happens, e.g. the guard is
missing”.
“Underlying cause: the less obvious ‘system’ or ‘organisational’ reason for an adverse event
happening, e.g. pre-start-up machinery checks are not carried out by supervisors”.
“Root cause: an initiating event or failing from which all causes or failings spring. Root causes
are generally management, planning or organisational failings”.

MULTI-CAUSALITY THEORY
22. How is the Multi-causality theory become relevant in accident investigation?
Modern trend in accident investigation and analysis accepts the principle that accidents are
usually the result of a number of interrelated causes, none of which can be described as the
sole cause. The concept leads to a systems approach to accident investigation and
prevention rather than the concentration on one specific causal relationship. Thus, each of
the multiple causation factors may be seen as one domino in its own line of dominoes (just as
the roots of a tree branch out).

2a Cause 1a

2b Cause 1b Accident

2c Cause 1c

Multiple accident causes. Source: RMS

23. Give an example of how the Multi-Causality theory can explain an accident.
For example – consider a tired gas filter; he does not check his work equipment before leaving
home and is delayed by heavy traffic on his way to work. On arrival at work, he finds a co-
worker has not arrived on time, so starts a two-man job on his own. The work is to be carried
out of height. On checking, he finds that his ladder is not on his van, so uses a drum for access
to a high level pipe, slips and injures his leg.

Latent and active failures – Reason’s model of accident causation

Reason’s original work (1990) involved operators of a nuclear power plant; he was particularly
concerned with latent (hidden) human failures which could jeopardise the overall safety of
the system at some future time.
Reason proposed four levels of human failure, each influencing the next.

6
24. What is the first level of lhuman failure according to Reason?
Considering an example of an accident involving a train crash we can consider each of these
levels of human failure. Working backwards from the accident the first level that Reason
identifies is unsafe acts of persons, which ultimately lead to the accident. These unsafe acts
were considered to relate to the errors, such as the skill-based slips/lapses and mistakes
identified in Rasmussen’s work in 1987. These were called active failures. Reason argued that
there was little advantage in blaming the operator for errors when there were many latent
failures that led to error. In the case of a train driver, it may be too easy to blame the cause of
the accident on ‘driver error’ in circumstances where there were latent failures that lead to
the error.

25. What is the second level of human failures according to Reason?

Reason identified three further levels of human failure that comprised latent failures. The
second level of human failure involves preconditions that lead to unsafe acts taking place.
This level involves preconditions such as mental fatigue, poor communication and poor work
practices, such as a train driver passing a faulty signal when it shows danger.

26. What is the third level of human failures according to Reason?

In many instances, these preconditions can be traced back to instances of unsafe supervision,
the third level of human failure identified by Reason. In the example of the train crash, this may
relate to a driver being asked to do extra driving time to cover for the absence of others or an
acceptance of the practice of passinf signals at danger if the drivers believe it to be faulty.

27. What is the fourth level which leads to human failures?

Importantly, Reason’s identified that causation did not stop at the supervisory level. He
recognised that the fourth level, the organisation itself, can impact on performance at all
levels. The fallible decisions of designers, higher level managers and others can have adverse
consequences. For example, in the case of the train crash, if supervisory level line
management are provided with limited resources for driver training signal maintenance and
they are under pressure to maintain train service this can lead to active failures at the lower
level of human failure.

The schematic below represents barriers at each level, with circles of potential failure which
may vary and change over time, but do not result in an active failure. However, when all the
circles of failure in each barrier are in alignment then the path to active failure will exist and
the accident will occur.

LESSON REVISION QUESTIONS

1. Define an accident
2. Define an incident

3. What is a near miss?

4. What is an undesired Circumstance?

5. Give four possible consequences that may arise from a spanner falling down from a height.

6. Why is it important to investigate a near miss?

7. Who was HW Heinrich?

7
8. In brief, what was his theory of accident causation?.

9. Which series of events leads to an accident according to him?

10. What conclusions can be drawn from the accident triangle or pyramid?
11. What are the limitations of using these accident triangles or pyramid?

12. How was the Heinrich Domino theory modified by Bird and his team?

13. List four components which fall under “Loss”.

14. What do you understand by immediate causes.
15. What do you understand by underlying causes?

16. What are the 3 major categories under which underlying causes fall?

17. List the items which falls under organisational factors

18. List the items which falls under job factors

19. List the items which falls under personal factors.

20. What do you understand by Lack of Management control?

21. Distinguish between immediate causes, underlying causes and root causes.
22. How is the Multi-causality theory become relevant in accident investigation?

23. Give an example of how the Multi-Causality theory can explain an accident.
24. What is the first level of lhuman failure according to Reason?
25. What is the second level of human failures according to Reason?

26. What is the third level of human failures according to Reason?

27. What is the fourth level which leads to human failures?
28. What is the learning outcome of this part of the element?

EXAM TYPE QUESTIONS

1. An employee has tripped whilst carrying a box of stationery through an office.
Note down approximately 10 QUESTIONS, to form a checklist to help you start your
investigation and determine the CAUSES of the incident. Consider general questions and
those specific to this particular scenario.

2. A delivery van has collided with a skip in your warehouse yard.

Note down approximately 10 QUESTIONS, to form a checklist to help you start your
investigation and determine the CAUSES of the incident. Consider general questions and
those specific to this particular scenario.

3. The wheel on a portable grinder being used by a maintenance employee has shattered,
pieces narrowly missing several employees.
Note down approximately 10 QUESTIONS, to form a checklist to help you start your
investigation and determine the CAUSES of the incident. Consider general questions and
those specific to this particular scenario.

8
4. A chain sling, being used to lift a new piece of equipment into place, has snapped.
Note down approximately 10 QUESTIONS, to form a checklist to help you start your
investigation and determine the CAUSES of the incident. Consider general questions and
those specific to this particular scenario.

5. Ladder fall
A storekeeper fell from a ladder whilst attempting to remove a light item from some racking at
a height of 3 metres. The storekeeper was working alone at the time of the accident and the
ladder was not footed and it slipped. There was no risk assessment in place for this area of
work. The ladder was found to be in a poor condition, one of the adjustable feet was missing.
A safety representative informed you that the issue of use of ladders had been raised at the
safety committee, although there were no minutes of this meeting.
Based on the information provided, separate out the immediate and underlying causes.
You should be able to identify one significant immediate cause and a number of relevant
underlying causes. Remember to note down positive aspects as well as the weaknesses in the
management system. What further areas of questioning would you need to complete your
investigation?
6. Knife injury
A 17-year-old trainee badly lacerated hand from cutting open a large cardboard box using a
‘Stanley’ knife. He had been warned about using this tool 2 weeks previously by his supervisor
but had then brought a further knife in from home as he felt it was the only way to open the
boxes. A risk assessment was in place covering general office safety. A number of work
experience 16 and 17-year-old workers had been taken on in the past 2 months, but there had
been no revision to the risk assessment. Induction office safety training had been carried out
for all new starters and records were available.
Based on the information provided, separate out the immediate and underlying causes.
You should be able to identify one significant immediate cause and a number of relevant
underlying causes. Remember to note down positive aspects as well as the weaknesses in the
management system. What further areas of questioning would you need to complete your
investigation?

7. Explain how the ‘domino theory’ of accident causation, as developed by Bird and Loftus,
could be used to structure an investigation of a typical workplace accident. (10) D1 June 2007

8. (a) Explain the purposes and benefits of collecting ‘near miss’ incident data. (7)
(b) Outline the factors to consider when developing and implementing a system for reporting
‘near miss’ incidents.(7)
(c) Outline the limitations of using accident/incident data as a means of measuring health
and safety performance. (6) D1 Dec 07

Question 9
A national campaign aimed at improving safely standards in a particular industry has been
deemed a failure due to a significant increase in the rate of reported accidents over the
period of the campaign.
Explain why accident rates may have provided a poor measure of the campaign’s
effectiveness and identify other measures that might have been used. (10)

9
Question 10: Jul 05
Outline ‘domino’ and ‘multi-causality’ theories of accident causation, showing their respective
uses and possible limitations in accident investigation and prevention. (10)
Question 11 Jul 2014
10.a Explain theories of loss causation
10.b. Explain the quantitative analysis of accident/incident and ill-health data, limitations of
their application, and their presentation in numerical and graphical form
10.c. Explain loss investigations; the requirements, benefits, the procedures, the
documentation, and the involvement of and communication with relevant staff and
representatives
Question 12 Jul 2011
(a) The results of accident / incident ratio studies are often depicted as a triangle.
(i) Explain why the outcomes are often depicted as a triangle. (2)
(ii) Explain how raw accident / incident data can be converted into the type of results which
are normally shown in an accident / incident ratio study triangle. (3)
(iii) Explain the reasons why, in practice, the ratios of accident / incident outcomes in an
organisation always follow this triangular pattern. (4)
(iv) Explain the implications of accident / incident ratio studies for accident and incident
investigation arrangements and resourcing. (5)
(b) A business has undertaken a study of the numbers of different types of accidents and
incidents reported in a 12 month period. The results are shown below as raw numbers:
Incident outcome Numbers
Major injury 4
Lost time accident 10
First-aid treatment 26
Property damage 25
Near miss 45
Outline the conclusions that might be drawn from this data when compared with the results of
published accident/ incident ratio studies. (6)
Question 13 (July 2013)
Explain the 'domino' and multi-causality theories of accident causation, including their
respective uses and possible limitations in accident investigation and prevention. (10)

Question 14
Four previously independent companies, all manufacturing the same type of product, have
been bought by a holding company. It is intended that the companies will continue to
operate largely autonomously but will be required to comply with corporate standards,
particularly in relation to health, safety, environmental and quality issues. The lost-time
accident rates of the four companies currently vary considerably, with the rates of one
company consistently more than double those of the one appearing to be the best performer.
(i) Describe the range of factors that might account for the variation in accident rates. (14)
(ii) Review the strengths and weaknesses of accident rates as a measure of health and safety
performance (6)

10