Safety at Work 6 E Part 8 potx

Risk management and behaviour modification 395
consequences of the behaviour are also listed and analysed to establish
the features they possess.
It can be seen that the avoidance of musculoskeletal injury or eyestrain
are weak consequences because they are ‘late’, ‘uncertain’ and ‘negative’
in nature. The other consequences arising from not adjusting the
workstation are ‘immediate’, certain’, and ‘positive’ to the individuals
and thus reinforce the unwanted behaviour. Whilst it is possible to debate
each component the overall balance of consequences is disproportional
and in favour of reinforcing the unwanted behaviour, namely the failure
to adjust the chair height, or alter the screen tilt, or draw the blinds so as
to prevent glare from windows etc.
The next step in the process is to state in precise terms the observable
behaviours which are desired. Then the antecedents and consequences
which influence the desired behaviour can be added. This is demon-
strated in Figure 2.8.6.
The antecedents are likely to focus around education and training
followed by ongoing reminders. The consequences would include
comments and intervention by supervisors and fellow workers. For the
consequences to be effective, supervisors would need regularly and
frequently to observe and comment on the individual’s use of, and
performance at, the workstation. The comments should be positive and
approving when the desired behaviour has occurred. Commenting only
when the desired behaviour has not occurred is far less effective.
Reducing accidents in the workplace requires that the performance
management approach is applied to all unsafe behaviours. Clearly this is
a mammoth task that should be approached systematically. It can be
achieved either by an analysis of the accident data (i.e. using historical
data) or by using Job Hazard Analysis of tasks undertaken (i.e. using
Antecedent Behaviour Consequence s/l c/u +/–
Understanding of

injury potential
Training in use
Expectation of
comment by
supervisor
Expectation of
ongoing reminders
to use workstation
correctly
Computer
workstation
adjusted by user
before use
Observation and
comment by
supervisor
Expectation of
colleagues
Observation and
comment by
colleagues
s
s
s
c
c
c
+
+
+

·Ά
s/l – soon/late c/u – certain/uncertain +/– – positive/negative
Figure 2.8.6 Revised ABC analysis of computer workstations
396 Safety at Work
predictive data). In either case the analysis is looked at from a
behavioural perspective.
Using the accidents analysis data, the first step is to group it by task
and by area, e.g. ‘fork-lift truck accidents in the distribution area’. The
second step is to examine how each accident occurred and to identify the
significant behaviours which contributed to the accident. This list is
variously termed the ‘critical behaviour list’ or the ‘key behaviour list’ for
fork-lift truck accidents in the area. The third step is to analyse each of the
critical behaviours identifying their antecedents and consequences, and
the features of each consequence (the ‘ABC’ analysis). The fourth step is
to state the desired behaviour which would avoid the accident and to
give it consequences which will reinforce the use of the desired
consequences. At this stage this analysis is complete.
Job hazard analysis begins by examining the task and listing the desired
behaviours to accomplish it safely. Each desired behaviour is examined
and the necessary antecedents and consequences added.
The analysis is the first part of the programme which then has to be
implemented. In practice this means that a number of observers have to
be trained. Their task is to understand the safety critical behaviours for
the workplace and to become skilled in identifying them.
The observers audit the work area recording the number of safety
critical behaviours observed and the number of unsafe behaviours
observed to produce a ‘percentage safe behaviour’ score as follows:
% Safe behaviour score =
number of safe behaviours observed
total number of behaviours observed

This data is plotted and posted in the workplace so that the workgroup
is encouraged to work toward a rising trend. A typical graph of the
results is shown in Figure 2.8.7.
Figure 2.8.7 Typical effect of behaviour intervention process in the workplace
Risk management and behaviour modification 397
The strength of the process is more than an analysis of actions in the
workplace and observations of activity. It works best where the
employees take a leading role in managing and implementing it.
Employees thus undertake the analysis of behaviours, add the con-
sequences necessary to achieve the safe behaviours and subsequently
audit each other. By this means greater commitment to improving safety
occurs. In addition the workgroup often know how a job is actually done
(as opposed to what the procedure for it says) and is in the best position
to draw up the list of desired safety behaviours in the first place and to
monitor compliance with it.
When accidents occur they are analysed to see what might have gone
wrong. It may be that the safety critical behaviours were not identified
correctly in the first place. Alternatively the analysis might reveal that the
consequence modifiers are ineffective and have to be rethought.
2.8.2.2 The structural feedback approach
Performance management requires that consequences are first analysed
and then restructured to encourage the preferred (safe) behaviours. The
chart demonstrating the improvement in the percentage of safe behav-
iours is a feedback tool demonstrating the gains made.
Other work by Cooper et al.
10,11,12,13
places greater emphasis upon the
feedback process. In particular they stress that publicly displaying a chart
showing how well, or otherwise, a group of employees is doing in
relation to the areas of safety in which improvement is sought is itself a

very powerful agent for change. Consequently it becomes important that
feedback charts are posted prominently and are regularly updated. In
order to achieve this it is necessary that managers adopt a particular role,
namely:
1 Champion the behavioural process and inform his workpeople of it
and of his support for it.
2 Encourage employees to become active in the process especially as
observers.
3 Allow employees the time to be involved in the training and meetings
needed for goal setting.
4 Allow each observer one observation session each working day. An
observation session should last no longer than 20 to 30 minutes.
5 Be committed to attend goal setting sessions with the observers thereby
demonstrating his support.
6 Praise employees who work safely.
7 Encourage employees to reach the safety goals.
8 Arrange for senior managers to visit the workplace each week to
encourage the safety improvement effort.
The observers, who are members of the workgroup, commence their
training by analysing local accident data. They identify contributory
factors for each accident and subdivide them into observable behaviours
or situations which are safe or unsafe. These observable data form the
398 Safety at Work
basis of a checklist. Emphasis is placed upon gaining agreement from the
workforce that the items that form the checklist of behaviours are valid.
This is an important step as the workforce is assessed and scored against
the list that has been generated. The process of gaining agreement is itself
a type of feedback which seeks to gain involvement of and ownership by
employees of the safety programme.
Scoring takes the conventional form of making observations in the

workplace of safe and unsafe behaviours to generate a ‘percentage safe
behaviour score’. The data are charted and posted visibly in the
workplace. Feedback of the data is not the only emphasis. The employees
are asked by their observers to establish their own goals and subgoals
against which the performance is measured
2.8.2.3 Behaviour observation and counselling techniques
Any behaviour modification technique must involve an interaction with
people. As an accident can occur at almost any time and the consequences
in terms of injury outcome are not predictable, concentrating on a list of
identified safety critical behaviours can have the following limitations:
1 The critical behaviour list may be incomplete.
2 Behaviours may appear on the list as a result of a perception of, rather
than an analysis of, an actual risk. This is more likely if the list has been
compiled from a job hazard analysis.
3 As the size of the list grows to encompass more behaviours (a result of
ongoing accident experience and the desire to eradicate all accidents by
adding more safety critical behaviours) the whole system can become
unwieldy because too may behaviours are included in the observation
process.
4 The very existence of a list may limit the focus of employees and
observers to only those behaviours which are on the list. This becomes
a more significant problem if observers are under pressure to complete
a quota of observations per week or per month. Under these
circumstances the objective can alter subtly from one of using the
technique to reduce accidents to becoming merely an exercise in
completing a checklist. The resultant quality fall-off which takes place
can undermine and discredit the entire effort.
Other approaches have been developed such as the DuPont Safety
Training Observation Program (‘STOP’) or their similar ‘Safety Manage-
ment Audit Programme’

14
. In both these programmes the approach tends
to be less analytical in defining prescribed unsafe behaviours with a
different emphasis that requires a management top-down approach in
which one level of manager, having been taught the process, subse-
quently teaches the next subordinate level. The emphasis is upon
observation of employee behaviour and immediate counselling of the
observed employees. Implementation of the process is through members
of line management from team leaders to senior managers. Each
undertakes a workplace safety behaviour audit to an agreed schedule. For
Risk management and behaviour modification 399
example, a team leader of a large workgroup may be expected to
undertake a daily audit, middle managers may do an audit each week,
and senior managers and directors an audit each month. The training
they receive assumes a degree of knowledge of the workplace and the
hazards it contains. This is not unreasonable given that many managers
will have several years’ experience and knowledge of the work areas.
Furthermore they are not necessarily expected to know in detail how safe
working on each job should be achieved. They are expected, however, to
recognise how injury might occur.
The training emphasises the skill in observing people as they work and
learning to approach and discuss safety with them in a constructive
manner. This applies to employees who are observed working safely as
well as those working unsafely. In the former case discussion can
commend the safe behaviour and be widened to encompass other tasks
the employee might do, seeking out any safety concerns arising from
them. The very fact that a person in authority is discussing safety issues
with the employee is of great importance in raising awareness and
commitment to accident-free working. Thus in these programmes, there
is greater emphasis on observation and immediate intervention than on

observation, completion of a checklist, and the posting of a chart in the
workplace. Nevertheless as employees do voice their concerns they have
expectations that remedial measures will be taken. Feedback in this case
often takes the form of a list of actions identified from the audits and a
rolling calculation of the percentage completed.
2.8.2.4 Behaviour modification and the lone worker
Behaviour modification is most easily applied where large groups of
people work in a systematic activity. Examples of applications include
factory production lines, call centres, packaging and assembly lines and
large construction sites. In these situations the tasks can be easily
identified and the safe behaviours required to perform them without
injury specified. Other jobs, where employees work alone and away from
the direct control of the supervisor present different challenges. An
example is maintenance tasks that require employees to work away from
the workshop. In these situations specific risks can arise that cannot be
identified until the job is underway. For example, maintaining a pump on
a workbench is far simpler than performing the same task while the
pump remains in its original location two miles away from the workshop.
Typical of the problems that can arise are:
᭹ Access to the pump is restricted.
᭹ Bolt fixings may be corroded and the use of a blow torch to free them
may not be permitted in the area.
᭹ Particular internal parts of the pump may need additional main-
tenance work due to unexpected wear and tear, but the spare parts are
at the workshop and the mechanic must make a decision to use the
existing part believing ‘it will last until next time’ or stop production
while the part is replaced.
400 Safety at Work
The individual employee can rarely be kept under observation as these
decisions and actions are taken. Guidance from the Health and Safety

Executive
15
recommends the use of:
᭹ structured incident reviews where the assessor lists contributing
causes and seeks clarification and comment from employees;
᭹ workforce questionnaires that seek to capture the perception of
employees in identifying which of a list of eighteen management issues
warrant improvement.
Approaches such as this are indirect behaviour modification and can be
effective where there is extensive employee participation with feedback
provided on progress made. Communication is critical to success. The
objective is to address and influence the behaviour of the remote
maintenance worker so that he can deal with the hazard at the time when
it becomes a serious risk.
2.8.2.5 Behaviour modification and employee involvement
The application of behavioural techniques to improve the control of risks
in normal employment have usually occurred as a result of a senior
management initiative. It occurs as a dictate from on high and is imposed
on the workforce. A typical response from the workforce is three-fold:
᭹ it is seen as a management lay-on by middle managers and supervisors
adding greater burden to their (the operator’s) busy working lives;
᭹ it is viewed with scepticism by employees who have seen initiatives
come and go over past years (‘flavour-of-the-month’ syndrome);
᭹ managers, supervisors and employees alike do not expect the initiative
to last beyond the lifespan of the current senior management team or
until something else comes along to distract their attention.
The result of these attitudes is that a minimum commitment is made to
the initiative until senior management show by continuous example that
the initiative is here to stay. A significant new safety initiative is likely to
require at least two years of operations to convince middle managers and

supervisors and four years to convince employees that the initiative is
seriously intended!
In order to gain acceptance of behaviour change initiatives, but also to
promote a wider safety culture, the involvement of employees in a
partnership is recommended. A Health and Safety Executive publica-
tion
16
suggests that workforce involvement can improve performance in
an organisation which:
᭹ does not involve the workforce in determining company policy;
᭹ does not treat the workforce or its representatives as equal partners in
the health and safety committee;
᭹ does not allow employees to set the health and safety agenda during
meetings; and
Risk management and behaviour modification 401
᭹ does not involve workers in writing safe operating procedures.
If these and other aspects show that workforce involvement is limited, it
is unlikely that a positive health and safety culture exists.
The concept of partnership is promoted in a union publication
17
that
quotes from a study by Reilly et al.
18
which shows that in-depth
consultation with employees reduces serious injury rates. Data are
reproduced in Figure 2.8.8.
2.8.2.6 Refreshing behaviour modification processes
Employee behaviour modification processes, as with all other human
processes, can become stale and ineffective. A significant proportion of
the safety improvement comes from the early interaction between the

observer and the observed. If this process becomes superficial then few
improvements in safety performance can be expected. Refreshing the
whole intervention process periodically is essential and can be achieved
by:
᭹ training new observers;
᭹ using trained observers from one department in an adjacent
department;
᭹ changing the mix of observers to include managers, supervisors and
employees in rotation;
᭹ revising the checklist of critical behaviours. If some behaviours always
remain on the critical list then generate two lists and mix the
observation sequence;
᭹ revising the checklist and discussing the changes in the light of
incident experience from within the area and from appropriate external
data.
2.8.2.7 Generic behaviour modification model
Most studies have focused upon modifying the behaviour of production
operators to improve safety performance. While variations in detail exist,
the basic generic behavioural process consists of:
Consultation model Serious injury rate
per 1000 workers
No union recognition and no joint committee 10.9
Worker representation but no joint committee 7.3
Joint committee but no trade union safety representatives 6.1 to 7.6
Full union recognition and joint committee 5.3
Figure 2.8.8 Type of consultation and serious injury rates
402 Safety at Work
᭹ Specify/know the behaviours that are necessary for safe working.
᭹ Observe. The observation step can take the form of a general
observation of an employee’s behaviour and counselling in the

appropriate target behaviour. Alternatively the observer can refer to a
previously established critical behaviour checklist and score the
number of safe and unsafe behaviours observed.
᭹ Intervene and discuss. The observer discusses with the individual
their personal safety performance. This step can take several forms.
The discussion can simply be a comparison of their performance
against the critical behaviour checklist, safe behaviours and an
emphasising of the recommended advice on the immediate corrective
action, that should be taken to remedy unsafe situations.
᭹ Follow-up action. A review of the data are collected and the action
necessary to ensure safe operator behaviour.
᭹ Feedback can take a number of forms ranging from a chart showing
the percentage of safe and unsafe behaviours to a rolling list of open
and closed corrective actions.
2.8.3 Behaviour modification for managers and
supervisors
Wrong behaviour occurs at all levels. The narrow focus on front line
employees behaviour has been criticised by labour unions
19
because it
infers:
᭹ only front line employee behaviour causes accidents;
᭹ the employee is to blame;
᭹ that a blame environment drives safety problems underground;
᭹ that capital expenditure is not authorised because incidents would be
avoided if employees did not make errors.
Figure 2.8.9 provides an illustrative example that demonstrates how
several levels and functions of management as well as front line
employees can contribute to the occurrence of an unsafe situation. It is
therefore important that behaviour modification processes are applied to

people in management and supervisory roles.
The generic behaviour modification model can be applied to managers
and supervisors as illustrated in Figure 2.8.10.
The observer of a supervisor’s behaviour can be the direct manager or
a third party, such as a safety professional. A procedure that has proved
successful comprises a list of fifty behaviours that align with the overall
safety initiative for that year (Figure 2.8.11) with the results plotted on a
‘radar screen’ chart (Figure 2.8.12). Measurement occurred in several
different ways which included:
᭹ personal self-assessment by the individual manager and
supervisor;
Risk management and behaviour modification 403
OPPORTUNITIES FOR ERRORS LEADING TO A PIPE RUPTURE WITH
CONSEQUENTIAL LOSS OF OUTPUT
The research chemist The chemist recommended in his report a minimum
operating temperature for the new compound but
failed to emphasise that it froze below that
temperature and expanded as it froze.
The design engineer The designer did not allow for extremes of
temperature and failed to specify adequate heat
tracing for a heat exchanger bypass line.
The construction contractor Because the bypass pipe was awkward to get at, did
not lay the trace heating evenly along its length.
The supervisor Wrote into the operating instructions turning on the
trace heating when the temperature of the incoming
compound dropped below a specified temperature. He
did not incorporate a check to ensure the compound
was flowing.
The plant operator Neglected to check that the trace heating was on and
that the compound was flowing.

The maintenance mechanic Failed to report or repair damage to the trace heating
caused during earlier maintenance.
The plant manager Delayed activating the trace heating system to save
energy but failed to recognise the cooling effect of a
cold spell.
The corporate director Cut the plant budget causing staff shortages that
prevented all the safety checks being carried out,
particularly to the trace heating system.
Result Compound temperature dropped and it froze,
expanded and fractured the bypass pipe. Process shut
down for 6 hours while a repair was effected. No
injuries but much lost production.
Figure 2.8.9 Error opportunities at different organisational levels (adapted from
Lorenzo
20
)
Figure 2.8.10 Generic behavioural check for managers and supervisors
TASK EXAMPLE
Specify the supervisors behaviour that is
required
To hold a weekly toolbox talk on the
previous weeks safety issues
Observe that the behaviour is occurring Check/attend/sample the toolbox talks
Intervene to commend the activity or
counsel if improvements are needed
Provide comment to the supervisor on
the content and the impact of the
toolbox talk
Follow-up on actions that the observer
has undertaken

If the supervisor needs training in running
group meetings ensure it is arranged
Feedback on overall performance
periodically
Discuss the safety behaviours at the
annual performance review
BEHAVIOR
DIMENSION A
BEHAVIOR
DIMENSION B
BEHAVIOR
DIMENSION C
BEHAVIOR
DIMENSION D
BEHAVIOR
DIMENSION E
Management
Commitment,
Involvement and
Leadership
Training and Education Employee Involvement Hazard Identification and
Analysis
Hazard Prevention,
Elimination and Control
ᮀ I have an HSE policy,
supported by annual goals
and objectives, which is
written and shared with
my direct reports
ᮀ The rules for Personal

Protective Equipment
(PPE) in my area are
posted and the safety
signage is unambiguous
ᮀ My areas have safety
committees which meet
at least quarterly
ᮀ My area investigates all
significant incidents
immediately (whether a
person is injured or a
serious near-miss occurs)
ᮀ I ensure that
individuals responsible for
correcting hazards are
identified and are set
clear timelines to achieve
the corrective action
ᮀ I have reviewed
performance against my
policy, goals and
objectives with my direct
reports and my HSE staff
this quarter
ᮀ My area has a process
to review work
procedures with all
employees regularly (not
just the HSE people) and
to ask them if changes

are necessary
ᮀ I chair my safety
committee
ᮀ A line manager or
supervisor always leads
the incident reviews, not
an HSE staff person
ᮀ I keep track of the
corrective actions in my
area
ᮀ My safety policy states
what I expect the line
organization to do and
what I expect the HSE
staff organization to do
ᮀ I have attended the
Basic Law training class
ᮀ My safety committee is
comprised primarily of
line people, with a
minimum of HSE staff in
attendance
ᮀ I make clear my
expectation that an
injured person must go
to Medical even if they
also wish to see their
private physician
ᮀ I visit work areas and
make it a point to

comment on hazards I
have identified and safety
improvements achieved
ᮀ I energize the HSE
support function to
behave and act as a
‘coach’ rather than a ‘do-
er’
ᮀ I have attended an
employee absence Case
Management training class
ᮀ All actions arising from
my safety committees
have agreed-upon actions
and timelines
ᮀ My area uses the new
accident model form
during incident reviews
to brainstorm all possible
contributions to an
incident
ᮀ I recognize and reward
work areas when they
achieve safety
performance milestones
ᮀ I use HSE issues to
demonstrate a leadership
style that is co-operative,
participative, and inclusive
as distinct from an

autocratic or adversarial
style
ᮀ I have assisted in
delivering the ‘Incident
2001’ training class to my
direct reports
ᮀ The safety meeting
minutes are freely
available. Departmental
level safety committee
minutes are posted on
local bulletin boards
ᮀ My area conducts
incident reviews in a
positive, open, no-blame
environment to ensure all
views and opinions are
discussed
ᮀ The area’s safety data
is rolled up by the line
organization and not by
an HSE staff person
ᮀ I make clear an
expectation of zero
incidents because
‘professional’ individuals
and ‘professional’
workgroups do not have
incidents
ᮀ My HSE staff person

has reviewed my personal
safety training plan
ᮀ Each safety bulletin
board has an assigned
owner who updates it
frequently
ᮀ I fulfill my monthly
Behaviour audit goals
ᮀ My area has an
Emergency Preparedness
program which has been
tested within the last 12
months
ᮀ I promote an
expectation of personal
responsibility and
accountability to avoid
incidents in my work
group
ᮀ My area has a safety
training plan for each
employee
ᮀ As a manager, I
encourage participation
by seeking my employees’
input and actively listen
to their concerns
ᮀ My area uses data
from incident reviews to
focus the Behaviour

audits on topics of
concern
ᮀ During this quarter, I
have undertaken a spot-
check on a key
maintenance or
inspection procedure
(e.g., lock-out/tag-out)
that is critical to control
a hazard in my area
Figure 2.8.11
BEHAVIOR
DIMENSION A
BEHAVIOR
DIMENSION B
BEHAVIOR
DIMENSION C
BEHAVIOR
DIMENSION D
BEHAVIOR
DIMENSION E
ᮀ I promote feelings of
ownership among team
members for their
collective safety
ᮀ My area has a process
to review all workplace
changes so that safety
issues are addressed
ᮀ I communicate my

expectation of an overall
‘partnership’ in safety
matters that creates an
equal responsibility
between the manager, the
supervisor, or the
employee
ᮀ The work groups in
my area get feedback on
incidents and audits
ᮀ I insist that hazards
and safety incidents are
reported to me by the
line organization, not the
HSE staff person
ᮀ I seek each day an
opportunity to verbally
and personally reinforce
the safety values with my
direct reports
ᮀ I encourage my HSE
staff to increase the
training they give to my
managers, supervisors,
and employees to do the
activities that the HSE
staff currently do
ᮀ I use my HSE
representative to review
and guide in improving

the areas ‘partnership in
safety’
ᮀ I use incidents and
behaviour audits to
promote personal
responsibility for the
‘system’ (manager’s
behaviour), the
‘workplace’ (supervisor’s
behaviour), and the ‘work
activity’ (employee’s
behaviour)
ᮀ I set a personal
example by following
local safety rules and
procedures
ᮀ Safety is one of the
top 3 agenda items at my
group and individual
meetings with my direct
reports
ᮀ Each person in my
area has received the
Safety Calendar for 2001
ᮀ My area exhibits a high
standard of ‘orderliness’
knowing that to achieve
it requires a high
standard of workplace
discipline and motivation

ᮀ In my area, less
significant incidents are
investigated and reported
at a local level
ᮀ I always intervene if
another person is
behaving in a way that
could result in injury
Figure 2.8.11 (Continued)
Risk management and behaviour modification 407
᭹ personal self-assessment which was reviewed with the direct
superior;
᭹ the divisional management team set the minimum behaviours expec-
ted with achievement against or beyond these targets being recognised
in the annual performance review;
᭹ direct reports of the manager or supervisor scoring the performance of
the senior manager.
The particular method selected must depend on the maturity, con-
fidence and interpersonal skills of the people involved. The fifty
behaviours are organised into five dimensions each containing ten
desirable behaviours. In this example, the behaviours are not ranked in
any order and each behaviour bears the same value. The number of
behaviours that a manager or supervisor demonstrates in each dimen-
sion is plotted along the appropriate axis of the radar screen. It is then
possible to see at a glance what behaviours are being achieved and
what behaviours should be undertaken. The radar screen was chosen in
one organisation for a particular reason – that other production data
was also displayed in the same format. Consequently ‘safety’ was seen
as an integral part of the manager’s or supervisor’s work in the same
way that production was.

Figure 2.8.12 Safety leadership radar screen
408 Safety at Work
2.8.4 Applying behaviour concepts to incident
investigation
Incident investigation is more effective when it is structured to engage
people in a way that promotes a sense of personal responsibility. This can
be achieved by modifying the well known domino accident model as
illustrated in Figure 2.8.13
In a normal application of the domino incident model, the user works
back from the ‘loss’ to identify the ‘root cause’. The process by its very
terminology seeks to identify a single causative factor (seldom is the
analysis referred to as a ‘root cause’ analysis) and usually focuses upon
system deficiencies. Applying a behavioural perspective is achieved by
asking a different question that changes the focus of the investigation.
The question asked is who could have done what to prevent the incident
occurring or reduce the likelihood of it occurring, or reduce the consequence when
it occurred. The investigation process should first focus on the managers’
behaviours (the system controllers), then the supervisors’ behaviours (the
workplace controllers) and finally the activity controller(s) (the oper-
ators’). If this sequence is reversed it is easy to give an impression of
trying to blame the operator. Once all the possible contributory factors
have emerged, suitable remedial actions can be chosen according to the
circumstances, the risks and the costs. Several actions can usually be
identified and the behavioural responses addressed through the behav-
ioural audit process.
When carried out within a mutually supportive culture, the process
promotes greater personal responsibility, accountability and commitment
to improve.
2.8.4.1 Stress and using the incident model
A previous chapter used an example of Joe falling from a stepladder (see

Figure 2.6.12). That analysis was undertaken using the Incident Model
Figure 2.8.13 Domino incident model related to behavioural activity
Risk management and behaviour modification 409
and it was discussed in terms of the organisational responsibilities for
health and safety.
The example provided in Figure 2.8.14 relates to the condition of stress
and demonstrates how the model can be applied to chronic health and
safety issues. It shows how stress incidents can be analysed in such a way
to influence the behaviours of managers, supervisors and operators.
Stress has been the subject of much recent research and comment. Life
itself is not without stress but the addition of work-related stress can
contribute to a situation where the individual is emotionally and
psychologically overcome. Stress affects the individual’s decision-making
ability and also his actions (behaviours) and can lead to acute accidents.
Research by the Health and Safety Executive
21
indicates that a risk
Figure 2.8.14 Behavioural incident model applied to a stress incident
410 Safety at Work
assessment approach to managing stress can stimulate new and innova-
tive ways of dealing with the problem.
The model can be used with a checklist of behaviours expected from
each contributor to avoid stress problems. Only a partial list is shown.
2.8.5 Behaviour concepts and the safety management
system
In chapter 2.6 the elements of safety management systems were
discussed. Systems as systems can amount to no more than a list of sterile
procedures and general expectations that deliver little improvement. In
order to put ‘flesh on the bones’ of a system thought must be given to the
behaviour that is necessary to support it and the system must be

constructed empathetically with those behavioural aims. A safety
management system based on the model suggested by the Health and
Safety Executive
22
can be designed to generate the practical behaviours
that make the system effective. An example is provided in Figure 2.8.15.
The process demonstrates the following features:
᭹ expectations are specific;
᭹ expectations have an assigned owner;
᭹ expectations are measurable.
The whole system must be visible to the operator workgroup and their
representatives. The process is most effective where the workgroup play
a large role in assessing performance. In large organisations the system
elements and behavioural expectations can be established and measured
level by level down the chain of command. Each manager or supervisor
should ask their work group to monitor its performance and report
upwards.
2.8.6 Risk, behaviour, leadership and commitment
Changing peoples’ behaviour to avoid hazards and reduce risks can
become a transactional process and increasingly ineffective. A transac-
tional style of leadership can be defined as a rule-following approach that
is mainly concerned with achieving a task. By contrast, a transformational
leadership style is mostly focused upon people and is more effective in
changing behaviours. An HSE research report
23
quotes a study
24
which
concluded that a transformational leadership style had a strong positive
impact on the safety performance of those individuals who generally

were otherwise less committed to safety. Some differences between a
transactional and transformational leadership style are listed in Figure
2.8.16.
The behavioural processes that have been described in this chapter can
be applied with either a transactional or a transformational style. For
example if a behavioural checklist becomes a rule-based observation and
intervention activity, then it is being applied in a transactional manner. In
Risk management and behaviour modification 411
Figure 2.8.15 Example of behaviour shaping associated with a safety management
system
HSG65 ELEMENT SYSTEM
REQUIREMENT
MEASURE
POLICY 1. The board to review
and reissue the policy
every 3 years
A policy less than three
years old
2. Each board member to
ensure that the policy is
reviewed in cascaded
discussions throughout
the organisation
At least 80% of a sample
of employees to confirm
their knowledge of the
new policy within 3
months of its issue
3. Independent auditors
to assess that the policy

covers all risks etc.
An audit report to be
reviewed at the senior
safety committee
ORGANISING 1. Each job will have
specified safety
responsibilities which are
reviewed every 3 years
Job responsibilities to be
reviewed/checked by the
safety department
2. All work procedures
will be reviewed every
three years
Work procedures to be
reviewed by the
workplace representative
3. Safety committees at
appropriate organisational
levels will meet bi-
monthly etc.
Records of meeting
minutes to be kept under
review
PLANNING AND
IMPLEMENTING
1. Employees will receive
a copy of the safety plan
Operator reaction to be
sampled by the safety

reps
2. Suitable training will be
undertaken for key plan
personnel etc.
The safety department to
specify the training and
ensure delivery of it
MEASURING
PERFORMANCE
1. Safety committees will
monitor their area’s part
in the overall plan
Data from safety
committees to be
summarised for the
board
2. Incidents will be
reviewed and key findings
shared widely etc.
Safety department to
monitor compliance
AUDIT & REVIEW 1. The safety department
will provide a review of
plan progress and
incidents each quarter
A report will be sent to
the board, members of
the management team
and safety committee
members

2. An external audit will
be performed every 3
years etc.
– the audit will be
available to employees
412 Safety at Work
these circumstances behavioural processes will be less successful. A study
by Griffin et al.
25
concluded that safety performance improved where the
manager adopted a supportive and caring style. This is not surprising
because a good manager will seek to explain and understand a poor or
‘at-risk’ behaviour before applying a workplace safety rule. Specific rules,
generally speaking, do not cover every situation in and nuance of
workaday life. Different levels of manager have different impacts.
O’Dea’s study
24
indicates that senior managers influence the general
climate and expectations for proactive safe behaviour while middle
managers and supervisors influence the adherence to local procedures
and rules. Simard et al.
26
found that low accident rates occurred where
the supervisor adopted a participatory leadership style whereas higher
accident rates occurred where a hierarchical style was used. A study of
safety on construction sites
27
concludes that management commitment is
vitally important. Duff et al.
28

noted that the best performing construction
sites in their study were those where managers showed their
commitment.
Leadership style and committed application at all levels, but especially
from managers and supervisors, are essential to the success of behaviour
modification programmes. Experience has shown that success occurs
where senior managers create the fertile soil in which safety professionals
can plant appropriate seeds for supervisors to tend and bring to full
bloom. A flowery metaphor, but apposite!
2.8.7 Behaviour modification processes: the hazards
Behaviour modification processes can fail to have the aimed-for impact
for a number of reasons including:
1 Behaviour processes have not been placed within the context of the
wider safety management system but have been viewed as an
alternative to it.
2 The behaviour modification process designed does not include
managers or supervisors, only front-line operators.
TRANSACTIONAL
LEADERSHIP STYLE
TRANSFORMATIONAL
LEADERSHIP STYLE
Managers command Managers involve
Managers ‘have the answers’ Managers ask employees for answers
Managers resist change – conform to
‘same old way’
Managers seek out and consider new
ideas – prepared to change
Managers communicate one-way Managers listen and encourage everyone
to communicate
Managers ignore failings Managers right wrongs

Figure 2.8.16 Features of transactional and transformational leadership styles
Risk management and behaviour modification 413
3 The cost and time commitment has been underestimated especially in
regard to initial training, refresher training, and employee time taken in
observation and intervention.
4 There is a lack of willingness to make a sustained commitment over
several years. Behaviour modification processes are seen as a ‘quick fix’
to boost safety performance.
For success, the organisation must be ready to adopt a behaviour
modification process and the factors which indicate this is so include:
᭹ Accident levels have stopped falling.
᭹ Management is frustrated at the lack of improvement in safety
performance and is willing to embrace a new initiative.
᭹ Poor improvements in safety performance have resulted from capital
expenditure.
᭹ Incident analyses indicate that behaviour modifications by employees
could have avoided many of the accidents.
It must be understood that behaviour modification processes are
inappropriate where failure will result in a serious or fatal injury. Human
beings by their very nature are not perfect – they fail. If the consequence
of a single failure is so serious, the risk must be controlled by other
means.
2.8.8 Behaviour and safety culture
A positive safety culture is often seen as the endpoint of the systematic
efforts to improve safety. Once attained any hazards will be controlled in
a coherent, supportive, constructive, even happy environment. But what
is a positive safety culture? There are many answers. The Confederation
of British Industry
29
lists eleven features:

1. Leadership and commitment from the top which is genuine and
visible. This is the most important feature.
2. Acceptance that it is a long term strategy which requires sustained
effort and interest.
3. A policy statement of high expectations and conveying a sense of
optimism about what is possible, supported by adequate codes of
practice and safety standards.
4. Health and safety should be treated as other corporate aims, and
properly resourced.
5. It must be a line management responsibility.
6. ‘Ownership’ of health and safety must permeate at all levels of the
workforce. This requires employee involvement, training and
communication.
7. Realistic and achievable targets should be set and performance
measured against them.
8. Incidents should be thoroughly investigated.
414 Safety at Work
9. Consistency of behaviour against agreed standards should be
achieved by auditing and good safety behaviour should be a
condition of employment.
10. Deficiencies revealed by an investigation or audit should be remedied
promptly.
11. Management must receive adequate and up-to-date information to be
able to assess performance.
Research by Pidgeon
30
implies three major features constitute a good
safety culture:
᭹ the existence of procedures and rules reinforced by high expectations
of compliance;

᭹ attitudes toward safety that are constructive and positive;
᭹ an ability, capacity and willingness to consider and learn from
experience from within and without the organisation.
A technique for assessing where the organisation stands in respect of
safety behaviours at any point in time is to ask those most exposed to the
hazards – the operators. The deliberations in the boardroom lead to
decisions that can affect the health and safety of employees. If asked,
employees can indicate the impact of these high level decisions and show
whether their intentions and objectives are being achieved. This reflects
the ‘culture in being’.
Understanding employees’ perceptions and opinions can be considered
as a ‘reality check’ and can help to focus on concerns that they see as
impediments to good safety performance. Psychologists such as Stanton
and Glendon
31
have developed structured survey tools for this purpose.
Another is available from the HSE
32
. This latter publication asks 71
questions covering the following 10 factors:
1 Organisational commitment and communication.
2 Line management commitment.
3 Supervisor’s role.
4 Personal role.
5 Workmates’ influence.
6 Competence.
7 Risk-taking behaviour and some contributory influences.
8 Some obstacles to safe behaviour.
9 Permit-to-work.
10 Reporting of accidents and near misses.

Using the tool requires that managers, supervisors and their operators
each provide their opinion on these 10 topics. This permits the views that
managers hold to be compared with the views of other levels in the
organisation. The survey can be undertaken in other areas of the
company and at other dates enabling comparison across an organisation
and over a period of time.
Measurement is only the first step in using survey techniques.
Collected data, including accident and incident data, should be used for
Risk management and behaviour modification 415
discussion with operators. Utilising the data in this way is as important
as getting it in the first place. The involvement of operators in this way
raises expectations – operators expect more from managers and managers
expect a greater contribution from operators.
The process measures opinions, and it also generates a focus on, and
commitment to, improved safety performance, behaviours that even-
tually produce a positive safety culture.
2.8.9 Conclusion
The experience of many established companies is reflected in Figure
2.8.17.
Accidents have been decreasing over the last fifty years but are
tending to reach a trough. In early post-war years the emphasis
remained on physical safeguarding. This gave way to the systems-of-
work approaches in the 1970s which continues today in the form of risk
assessments. The behaviour-based approach is becoming increasingly
important as a technique to improve safety at work for two reasons.
First, the other approaches have been implemented but there still
remains room for considerable improvement. Second, the nature of
work is changing. The emergence of service sector employment and the
increasing freedom from a conventional workplace, which modern
communication and computer systems allow, have caused written

procedures and close supervision to become less easy to apply as an
effective means of exerting control. Effective safety measures will
emerge from better education and training and the implementation of
motivational theories to change attitudes and behaviour.
Whilst an understanding of the links between attitude, behaviour and
the consequences of the behaviour are still developing, enough is known
Figure 2.8.17 Safety emphasis and post-war accidents trends
416 Safety at Work
to establish that certain key elements of a behaviour approach are more
likely to yield success. These key elements include:
1 Management commitment. This should take the form of providing
encouragement, being supportive and offering coaching to the work-
group. A ‘command and control’ approach is not a recipe for lasting
success.
2 Workforce involvement. It is essential that the employees, who
themselves suffer the accidents, are thoroughly immersed in the
process. Their involvement is crucial because they are the people who
know the unsafe acts that are committed and the reasons for them.
Clearly the workgroup themselves are in the best position to know the
antecedents and consequences which operate and how they can be
adjusted to promote safer working.
3 Effective feedback is essential and should be twofold in content. First,
there should be immediate interaction with the employee being
observed irrespective of whether safe or unsafe working is noticed.
Second, there should be feedback to the work group as a whole for
comparison with the agreed checklist of safety critical behaviours. This
can give a behaviour score which can be compared with goals agreed
by the work group. The result can be used to measure progress and can
be posted on a chart in the workplace.
Behavioural processes may have limitations as to their effectiveness.

Most examples come from areas where the hazards are acute and
understood and the behaviour is observable, such as the wearing of
personal protective equipment. It may be harder to apply these
techniques to work systems such as ‘permits-to-work’ or where the risk is
low in probability but high in consequence and where precautions are
taken but effects not immediately visible. An example might include the
maintenance of pressure systems where poor work may not become
visible until an explosion occurs much later.
The objective of a behaviour-based approach is to change work habits
for the better. There is evidence to show that success does result from
such programmes. However, they require considerable resources and
commitment which might not always be forthcoming. Little work has
been undertaken to show if improvements in performance continue to
accrue or can even be maintained when resources are scaled down, for
example, by undertaking less frequent audits. It may be that the
imposed effort can be scaled down because attitudes have effectively
and permanently altered in favour of ongoing safe working. The
evidence from major companies which have reputations for con-
tinuously superior safety performance seems to be that the formality can
be reduced but there must be an ongoing management focus. In
practical terms this involves being seen to be committed to safe
working, walking the workplace, coaching employees in safety behav-
iours, and encouraging their full participation into what is a stated
major workplace value, namely an ongoing reduction in accidents and
ill-health at work. When this point is attained, a culture change has
been achieved.
Risk management and behaviour modification 417
References
1. Skinner, B.F., The Behaviour of Organisms, Appleton-Century-Crofts, New York (1938)
2. Health and Safety Executive, Guidance publication no. HSG 48. Reducing Error and

Influencing Behaviour, HSE Books, Sudbury (1999)
3. Komaki, J., Barwick, K.D. and Scott, L.R., A behavioural approach to occupational
safety: Pinpointing and reinforcing safe performace in a food manufacturing plant. J.
Appl. Psychol., 63: 434–445 (1978)
4. Komaki, J., Heinzman, A.T. and Lawson, L., Effect of training and feedback: component
analysis of a behavioural safety program. J. Appl. Psychol., 65: 261–270 (1980)
5. Komaki, J.D., Collins, R.L. and Penn, P., The role of performance antecedents and
consequences in work motivation. J. Appl. Psychol., 67: 334–340 (1982)
6. Suzler-Azaroff, B., The Modification of Occupational Safety Behaviour. J. Occupational
Accidents, 9: 177–197 (1987)
7. Nasanen, M. and Saari, J., The effects of positive feedback on housekeeping and
accidents at a shipyard. Journal of Occupational Accidents, 8: 237–250 (1987)
8. Daniels, A.C. and Rosen, T.A., Performance Management: Improving Quality and
Productivity through Positive Reinforcement. Performance Management Publications, Inc.,
Tucker, Georgia (1987)
9. Krause, T.R., Hidley, J.H. and Hodsen, S.J., The Behaviour-Bases Safety Process. Van
Nostrand Reinhold, New York (1990)
10. Cooper, M.D., Makin, P.J., Phillips, R.A. and Sutherland, V.J., Improving safety in a
large, continuous shift, production plant using goal setting and feedback: benefits and
pitfalls. Brit. Psychol. Soc. Annual Occ. Psychol. Conference, Brighton, Jan. 3–5 (1993)
11. Cooper, M.D., Goalsetting for safety. The Safety and Health Practitioner, November 1993,
32–37.
12. Cooper, M.D., Implementing the behaviour based approach, a practical guide. The Safety
and Health Practitioner, November 1994, 18–23.
13. Cooper, M.D., Phillips, R.A., Sutherland, V.J. and Makin, P.J., Reducing accidents using
goal setting and feedback: A field study. J. Occ. & Org. Psychol., 67, 219–240, (1994)
14. Safety Training Observation Program. E.I. du Pont de Nemours and Company,
Wilmington, Delaware, 1989.
15. Health and Safety Executive, Guidance publication, Improving Maintenance: a Guide to
Reducing Human Error, HSE Books, Sudbury (2000)

16. Health and Safety Executive, Guidance publication no. HSG 217, Involving Employees in
Health and Safety, HSE Books, Sudbury (2001)
17. The Trades Union Congress, Partners in Prevention: Revitalising Health and Safety in the
Workplace, Trades Union Congress, London
18. Reilly, Paci and Hall, British Journal of Industrial Relations, 33(2): June (1995)
19. Howe, J., Warning, Behavior Based Safety Can Be Hazardous To Your Health And Safety
Program!, Union of Automotive Worker, International Union, September 1993
20. Lorenzo, D.K., A Manager’s Guide to Reducing Human Errors, Chemical Manufacturers
Association, USA (1990)
21. Health and Safety Executive, Contract Research Report no. CRR 435/2002, Interventions
to Control Stress at Work in Hospital Staff, HSE Books, Sudbury (2002)
22. Health and Safety Executive, Guidance publication no. HSG 65, Successful Health and
Safety Management, 2nd edn, HSE Books, Sudbury (1999)
23. Health and Safety Executive, Contract Research Report no. CRR 430/2002, Strategies to
Promote Safe Behaviour as part of a Health and Safety Management System, HSE Books,
Sudbury (2002)
24. O’Dea, A. and Flin, R., Site managers, supervisors and safety in the offshore oil industry,
Academy of Management symposium, Canada (2000)
25. Griffin, M., Burley, I. and Neal, A., The impact of supportive leadership and
conscientiousness on safety behaviour at work. Academy of Management symposium,
Canada (2000)
26. Simard, M. and Marchand, A., Workgroup’s propensity to comply with safety rules: the
influence of micro-macro organisational factors. Ergonomics, 40(2): 172–188 (1997)
27. Health and Safety Executive, Contract Reasearch Report no. CRR 299/1999, Improving
Safety on Construction Sites by Changing Personal Behaviour, HSE Books, Sudbury
(1999)
418 Safety at Work
28. Duff, A.R., Robertson, I.T., Phillips, R.A. and Cooper, M.D., Improving safety by the
modification of behaviour. Construction Management and Economics, 12: 67–78 (1994)
29. Confederation of British Industry, Developing a Safety Culture, CBI, London (1991)

30. Pidgeon, N.F., Safety culture and risk management in organisations, Journal of Cross-
Cultural Psychology, 22: 129–140.
31. Stanton, N. and Glendon, I., Safety Culture Questionnaire, Griffith University, Australia
and University of Southampton, England, 1996 (Private communication)
32. Health and Safety Executive, Health and Safety Climate Survey Tool (Diskette), HSE Books,
Sudbury (1998)
Part III
Occupational health and hygiene
Chapter 3.1 The structure and functions of the human body
(Dr T. Coates) 421
Chapter 3.2 Occupational diseases (Dr A. R. L. Clark) 447
Chapter 3.3 Occupational hygiene (Dr C. Hartley) 492
Chapter 3.4 Radiation (Dr A. D. Wrixon and updated by Peter Shaw
and Dr M. Maslanyj) 524
Chapter 3.5 Noise and vibration (R. W. Smith) 543
Chapter 3.6 Workplace pollution, heat and ventilation
(F. S. Gill) 568
Chapter 3.7 Lighting (E. G. Hooper and updated by Jonathan
David) 581
Chapter 3.8 Managing ergonomics (Nick Cook) 594
Chapter 3.9 Applied ergonomics (John Ridley) 617
In his work, the safety adviser may be called upon to recommend
measures to overcome health problems that have been identified by the
doctor or nurse. Part of his duties may include the identification of
processes and substances that are known to give rise to health risks and
advising on the procedures to be followed for their safe use.
The advice he can give will be more pertinent if the safety adviser has
an understanding of the nature of the substance and the manner in which
it affects the functioning of the human body.
This Part explains the functions of the major organs of the body,

considers the characteristics and hazards of a range of commonly used
substances and processes and discusses the techniques that can be
employed to reduce the effects of those risks on the health and well-being
of the workpeople.