1. Principles of Construction Safety
Allan St John Holt
BA, FIOSH, RSP
Foreword by
Sir Frank Lampl

3. Principles of Construction Safety
Allan St John Holt
BA, FIOSH, RSP
Foreword by
Sir Frank Lampl

4. # 2001, 2005 Mei Wenti Ltd
Blackwell Science Ltd, a Blackwell Publishing company
Editorial Offices:
Blackwell Science Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK
Tel: +44 (0) 1865 776868
Blackwell Publishing Inc., 350 Main Street, Malden, MA 02148-5020, USA
Tel: +1 781 388 8250
Blackwell Science Asia Pty Ltd, 550 Swanston Street, Carlton, Victoria 3053, Australia
Tel: +61 (0)3 8359 1011
The right of the Author to be identified as the Author of this Work has been asserted in accordance with the Copyright,
Designs and Patents Act 1988.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988,
without the prior permission of the publisher.
First published in hardback 2001
Reissued in paperback 2005
Reprinted 2006
ISBN-10: 1-4051-3446-1
ISBN-13: 978-14051-3446-0
Library of Congress Cataloging-in-Publication Data is available
A catalogue record for this title is available from the British Library
Set in 9/14 Trebuchet
by DP Photosetting, Aylesbury, Bucks
Printed and bound in Great Britain
by TJ International Ltd, Padstow, Cornwall
The publisher's policy is to use permanent paper from mills that operate a sustainable forestry policy, and which
has been manufactured from pulp processed using acid-free and elementary chlorine-free practices. Furthermore, the publisher ensures
that the text paper and cover board used have met acceptable environmental accreditation standards.
For further information on Blackwell Publishing, visit our website:
www.blackwellpublishing.com

5. Contents
Foreword vii The contractor and CDM 41
Abbreviations viii The Principal Contractor and CDM 42
Introduction ix The Planning Supervisor and CDM 42
Competence, qualification and selection under
Part 1 Construction Safety Management 1 CDM 43
The Health and Safety File 44
1. Fundamentals 3
What is `safety' 3 5. The Safety Policy 46
Basic terms 3 Legal requirements 46
What causes accidents? 4 The policy in practice 46
Techniques of accident prevention 5 Manuals and QA systems 47
Benefits of accident prevention 6 Policy contents 47
Other safety policy considerations 49
2. Where are We Now? 8
`Off the shelf' safety policies 49
Types of injury 8
Accident causes 9
6. Assessing the Risks 50
Occupational health and hygiene 9
Benefits 50
Dangerous occurrences 10
Types of risk assessment 50
Reference 12
Contents of risk assessment 51
3. Measuring Performance and Recording Hazard evaluation 51
Information 13 Job safety analysis 52
Behaviour-based safety 13 Ranking hazards by risk 52
`No injuries Ð no problems!' 14 Decision-making 53
`All I want to know is the facts!' 14 Controlling the risks 53
Calculating rates 14 Monitoring 54
Other performance measures 15 Health surveillance 54
Accident investigation and recording 15 Information to others 54
Principles of accident investigation 16 Format of risk assessments 54
Inspections and audits 18 Project risk assessments 54
Techniques of inspection 19 Practical compliance 56
Reference 20 COSHH assessments 56
Manual handling assessments 58
4. Techniques of Construction Safety Display screen equipment assessments 63
Management 35
Objectives 35 7. Control Strategies for Construction Work 71
Benefits 35 Designing for safety and health 71
Key elements 36 Planning the work 72
And if companies do not do all this? 38 Emergency procedures 73
Is it all just a pile of paperwork? 38 Setting up the site 75
World best practice 39 Safe place of work 75
The role of the client 40 References 86

6. iv CONTENTS
8. The Health and Safety Plan 96 Control of substances hazardous to health 138
Pre-tender Health and Safety Plan 96 Measurement and reduction of energy
Developing the construction phase Health and consumption 139
Safety Plan 97 Environmental objectives and targets 139
Project safety management commitment Environmental policies 139
statement 97 References 140
Roles and responsibilities of project staff 97
Project Major Emergency Plan 100 15. Construction Hazards and Solutions 144
Access equipment 144
9. Training 104 Asbestos 148
Training needs 104 Children and third party safety 152
Training for management 107 Demolition 153
Environmental, health and safety specialists 108 Electricity 155
Legal requirements 108 Excavations 158
Falls 159
10. Meetings 109 Maintenance 160
Meeting with contractors 109 Manual handling 161
Weekly safety meetings 110 Mechanical handling 165
Monthly safety meetings 111 Noise 168
Occupational health basics 171
11. Understanding People 120 Personal protective equipment 174
Why people fail 120 Radiation 178
Ergonomics 121 Roofing work 181
Stress 122 Steel erection and decking 181
Communication 123 Transport on site 182
Reports 123 References 182
Getting the message across 124
Safety propaganda 125 16. Quick Reference Guide 184
References 125 1. Access scaffolding 185
2. Asbestos-containing materials 186
12. Joint Consultation 126 3. Bitumen boilers 187
The Safety Representatives and Safety 4. Cartridge tools 188
Committees Regulations 1977 126 5. Chainsaws 189
The Health and Safety (Consultation with 6. Clearing sites 190
Employees) Regulations 1996 128 7. Cofferdams 191
Consultation at site and project level 129 8. Confined spaces 192
9. Demolition 193
13. Access to Information 130 10. Disc cutters 194
Safety professionals 131 11. Disposal of waste materials 195
Other advice 132 12. Driving vehicles 196
13. Dumpers 197
Part 2 Environment, Health and Safety Issues 133 14. Electrical work to 415 volts 198
15. Erection of structures 199
14. Construction and the Environment 135 16. Excavations 200
Waste management and pollution control 136 17. Falsework 201
Initial ground contamination 137 18. Fire on site 202
Remediation 137 19. Fork-lift trucks 203
Spillage control 137 20. Gas welding and cutting 204
Pesticide use and control 137 21. Hand tools 205
Hazardous waste management 138 22. Joinery workshops 206

7. CONTENTS v
23. Ladders and stepladders 207 Powers of inspectors 243
24. Lasers 208 Enforcement 243
25. Lifting equipment, general 209 The Management of Health and Safety at Work
26. Materials hoists 210 Regulations 1999 243
27. Mobile cranes 211 Summary of the Regulations relevant to the
28. Mobile elevating work platforms 212 construction industry 244
29. Mobile towers 213 Provision and Use of Work Equipment
30. Portable electrical equipment 214 Regulations 1998 and machinery safety
31. Powered tools Ð woodworking 215 requirements 246
32. Public protection 216 Some significant definitions 246
33. Pressure testing 217 Summary of the Regulations 247
34. Roadworks 218 Supply of Machinery (Safety) Regulations 1992 250
35. Sewage connections 219 References 251
36. Site transport 220 The Lifting Operations and Lifting Equipment
37. Skips 221 Regulations 1998 (LOLER) 251
38. Steam and water cleaners 222 Some significant definitions 251
39. Storage of materials on site 223 Summary of the Regulations 252
40. Storage and use of LPG 224 Construction (Health, Safety and Welfare)
41. Storage and use of HFL 225 Regulations 1996 253
42. Temporary electrical supplies 226 Some significant definitions 254
43. Working alone 227 Summary of the Regulations 254
44. Work on fragile roofs 228 Construction (Design and Management)
45. Work with flat glass 229 Regulations 1994 260
46. Working at heights, general 230 Some significant definitions 260
47. Working in occupied premises 231 Summary of the Regulations 261
48. Work near underground services 232 The Confined Spaces Regulations 1997 264
49. Work under power lines 233 Summary of the Regulations 264
50. Work over water 234 The Control of Substances Hazardous to Health
Regulations 2002 266
Part 3 Legal Requirements 235 What is a `substance hazardous to health'? 267
Summary of the Regulations 267
17. Construction Health and Safety Law 237 The Personal Protective Equipment at Work
Statute law 237 Regulations 1992 269
Types of statutory duty 238 Summary of the Regulations 269
Development of health and safety law 239 The Electricity at Work Regulations 1989 270
Common law 239 Summary of the Regulations 271
The Woolf reforms 240 The Reporting of Injuries, Diseases and
The Health and Safety at Work etc. Act 1974 240 Dangerous Occurrences Regulations 1995 272
General duties of employers 241 Summary of the Regulations 273
General duties of the self-employed 242
General duties of employees 242 18. Penalties 275
General duties of manufacturers and suppliers 242
Charges 242 Index 277
The Health and Safety Commission and
Executive 242

9. Foreword
Environment, health and safety are already crucial issues in What is needed is to embed these vital factors as values and
the upper echelons of the construction industry. Sadly, that as a part of all that our industry does, and not to regard
is the exception rather than the rule, so, if our industry is to them merely as optional extras where time allows.
prosper in the twenty-first century, everyone at every level
needs to understand the importance of these issues and The author draws upon his lifetime experience in construc-
implement the practices that will safeguard both people tion-related health and safety to provide the information
and our planet. and the background material important for a full under-
standing of the issues involved. The book is aimed at every
This book is intended to fill the gap between a technical participant in the construction industry needing information
manual and the many topic guidance notes such as those and guidance on current and future best practice.
published by the Health and Safety Executive. Uniquely,
the amount of space devoted to construction EH & S man- A positive change of attitudes among all players in our
agement is the same as that given over to techniques and to industry is vital. I welcome this book as a valuable con-
the law. Experience shows that time spent on pre-con- tribution to that goal. Allan, who has worked with us across
struction activity and planning is amply repaid during the the world to further the cause of site safety, is well placed
construction phase of work, when safety issues can be much to make it happen.
more difficult to resolve.
Sir Frank Lampl
President, Bovis Lend Lease

10. Abbreviations
ANSI American National Standards Institute MHSWR Management of Health and Safety at Work
CAWR Control of Asbestos at Work Regulations 1987 Regulations 1999
CDM Construction (Design and Management) MORT Management oversight and risk tree
Regulations 1994 NADO Notification of Accidents and Dangerous
COSHH Control of Substances Hazardous to Health Occurrences (Regulations)
Regulations NEBOSH National Examination Board in Occupational
CSCS Construction Skills Certification Scheme Safety and Health
CTA Certificate of Training Achievement PMEP Project Major Emergency Plan
FMEA Failure modes and effects analysis PPE Personal protective equipment
HSE Health and Safety Executive RIDDOR Reporting of Injuries, Diseases and Dangerous
IOSH Institution of Occupational Safety and Health Occurrences Regulations 1995
JSA Job safety analysis RPE Respiratory protective equipment
MEWPS Mobile elevating work platforms

13. Part 1
Construction Safety Management

15. 1 Fundamentals
What is `safety'? source of energy above the ability of the body or structure
to withstand it.'.
We use the word `safety' so much, often in company with
its partner `health', that it should be easy to find a defi- In normal conversation we use the word `accident' loosely,
nition. Yet the dictionaries do not offer much assistance Ð and in doing so we often couple in a sense of bad luck on the
`Safety [is] the absence of danger' one says, unhelpfully part of the injured person, and a feeling that it could not
supplying the entry for `danger' as `absence of safety'! have been foreseen. In safety management, we need to be
Others suggest `a state of protection' and `a condition not clear that the luck, or the element of chance, is only con-
involving risk'. Perhaps the best we can do is to agree that cerned with the physical outcome of the incident, which is
there is no arbitrary state of `absolute safety', as there is `that sequence of events or actions resulting in the unde-
always a chance Ð a risk Ð of something going wrong, sired consequences'. For ease of reading, this book uses the
however small that chance may be. word `accident' to describe injury events, except where an
important distinction has to be made between `accident'
In the same way, a little thought about `health' brings the and `incident'.
same conclusion Ð it is a relative notion, in the sense that in
any population there will be those in varying states of An injury is thus a consequence of an incident Ð but not the
wellness. But this does not stop us using the word in an only possible one. It has been shown that hundreds of
everyday sense to convey the idea that, in the workplace at incidents occur in the construction industry for every one
least, the aim should be that workers do not leave their that causes injury or loss. But all have the potential to do
work less `healthy' than when they arrived. so. That is why it is important to look at all incidents as
sources of information on what is going wrong. Relying on
The management of workplace health and safety is done injury records only allows a review of a minority of inci-
together, and in the same way, so that often in speech the dents Ð those which happened to result in a serious injury
word `safety' is used to mean both. In recent years, it has consequence. We can make some reasonable estimates
been recognised that environmental issues also need to be about the likelihood of, say, failure of a lifting appliance.
managed, and again often by using the same techniques Only chance will decide whether an injury rather than, or as
and practices. So, for reasons of space and ease of under- well as, property damage will occur on a particular occa-
standing, in this book the reader will often find the word sion, and how severe either will be.
`safety' used alone although the presence of its natural
partners `health' and `environment' should be understood. Hazard means `the inherent property or ability of some-
thing to cause harm Ð the potential to interrupt or inter-
Basic terms fere with a process or person'. Hazards may arise from
interacting or influencing components, for example two
An accident is an incident plus its consequences; the end chemicals interacting to produce a third.
product of a sequence of events or actions resulting in an
undesired consequence (injury, property damage, inter- Risk is `the chance or probability of loss', an evaluation of
ruption, delay). An accident can be defined more formally the potential for failure. It is easy to confuse the terms
as `an undesired event, which results in physical harm and/ `hazard' and `risk', but a simple way to remember the
or property damage, usually resulting from contact with a difference is that `hazard' describes potential for harm,

16. 4 PRINCIPLES OF CONSTRUCTION SAFETY 1
risk is the likelihood that harm will result in the particular n Leaving equipment in a dangerous condition
situation or circumstances, coupled with a measure of the n Using equipment at the wrong speed
degree of severity of that harm. Comparisons between risks n Disconnecting safety devices such as guards
can be made using simple numerical formulae. n Using defective equipment
n Using equipment the wrong way or for the wrong tasks
What causes accidents? n Failure to use or wear personal protective equipment
n Bad loading of vehicles
Accidents are the direct results of unsafe activities and
n Failure to lift loads correctly
conditions, both of which can be controlled by manage-
n Being in an unauthorised place
ment. Management is responsible for the creation and
n Unauthorised servicing and maintaining of moving
maintenance of the working environment and tasks, into
equipment
which workers must fit and inter-react. Control of workers
n Horseplay
and their behaviour is more difficult. They have to be given
n Smoking in areas where this is not allowed
information, and the knowledge that accidents are not
n Drinking alcohol or taking drugs
inevitable but are caused. They need training to develop
skills and recognise the need to comply with and develop
Some of the reasons why people fail Ð to behave safely, to
safe systems of work, and to report and correct unsafe
conform to policies and procedures, for example Ð are
conditions and practices. Their safety awareness and atti-
discussed in Chapter 8.
tudes require constant improement, and the social envir-
onment of the workplace Ð the safety climate Ð must be
one which fosters good safety and health practices and Unsafe conditions
conditions, not one which discourages them.
n Inadequate or missing guards to moving machine parts
n Missing platform guardrails
On investigation, and after a little thought, it can be seen
n Defective tools and equipment
that accidents are relatively complex events. A man falls
n Inadequate fire warning systems
off a ladder. It seems straightforward Ð the ladder was not
n Fire hazards
tied and witnesses say that it was set at the wrong angle
n Ineffective housekeeping
and not secured against slipping. This incident could be put
n Hazardous atmospheric conditions
down to carelessness on the part of the man, having failed
n Excessive noise
to appreciate the physical situation. Carelessness, though,
n Not enough light to see to do the work
is rarely either a good or an adequate explanation of events
like accidents.
These are all deviations from required safe practice, but
they must be seen as the symptoms of more basic under-
Unsafe acts and unsafe conditions are often referred to as
lying indirect or secondary causes which allow them to
immediate or primary causes of accidents, because they
exist and persist.
are the most obvious causes and because they are usually
directly involved or present at the moment the accident
happens. Secondary causes are also important, although Secondary causes of accidents
they are usually harder to seek out and identify. They are
n Management system pressures
the failures of the management system to anticipate, and
N financial restrictions
include lack of training, maintenance, adequate job plan-
N lack of commitment
ning and instruction, and not having safe systems of work in
N lack of policy
place.
N lack of standards
N lack of knowledge and information
Some examples of unsafe acts and conditions are given
N restricted training and selection for tasks
below.
N poor quality control systems resulting from the
above
Unsafe acts
n Social pressures
n Working without authority N group attitudes
n Failure to warn others of danger N trade customs

17. 1 FUNDAMENTALS 5
N industry tradition 2. Tackle risks at source
N society attitudes to risk-taking
Design and specification can make a significant difference
N `acceptable' behaviour in the workplace
to site conditions. Design is likely to dictate the way the
N commercial/financial pressures between con-
construction work is carried out on site, and particularly to
tractors
force contractors to work in potentially unsafe ways. A
review at the design stage repays the time spent hand-
The primary causes of accidents in the construction
somely because of later savings in time and money, and
industry have been the target of safety law for many years
possibly even lives.
Ð specifying details of scaffolding and ladders, for
example. Relatively recently, legal requirements in several
Examples: Designing floor slabs with fewer voids removes
countries, notably the Member States of the European
the need to control the risk of falling through them by
Union, and Australia, have begun to address the secondary
setting up barriers or covering them. Avoid dust-producing
causes as well, forcing attention to be paid to all organi-
processes: specify off-site finishing.
sational aspects of safety management.
3. Adapt work to the individual when designing
Techniques of accident prevention work areas and selecting methods of work
Accident prevention in construction is not just a matter of
Ergonomics aims to improve the interface between people
setting up a list of rules and making safety inspections,
and their workspaces, by seeking to adapt the workspace
although both of these have their place. What is required is
rather than the person. Thought given to layout can
a system for managing health and safety which meets the
improve working conditions and reduce risks.
needs of the business and complies with the law. A dis-
cussion of the ideas involved in safety management can be
Example: Locating equipment such as a saw bench in a
found in Chapter 4, and most of the law on construction
corridor could block access for others, may cause lighting
safety is covered in Part 3 of this book.
difficulties and allow offcuts to pile up, increasing the
risks. Asking how much room the carpenter needs and
There are seven principles to be observed in setting up
planning a suitable spot for a powered saw bench takes
strategies for control and management of health and safety
little time.
at work in the construction industry. If they are followed,
accident prevention is more likely to be achieved.
4. Use technology to improve conditions
1. If possible, avoid a risk altogether by Keeping up to date with new developments can bring a
eliminating the hazard safety benefit when plant is being replaced, or when work
operations can be mechanised.
It is always more effective to remove a problem altogether
rather than to establish a control strategy, especially one Examples: Specifying a quieter design of machine when
which relies upon people to work in the correct way. This ordering replacement equipment. Use of a wheeled man-
and the next principle demonstrate the fundamental hole lifter to replace hand hooks or other hand tools brings
importance of design and planning in safety management. gains in productivity as well as minimising the chance of a
back injury.
Examples: Do not specify fragile roofing materials through
which people can fall. This is obviously more effective than
5. Give priority to protection for the whole
specifying solutions designed to minimise the risks from
workplace rather than to individuals
falls through fragile material. Specification of lighter
materials, such as blocks and bags, is preferable to Reliance on personal protective equipment (PPE) as a sole
arranging ways to handle heavy materials on site. Inevi- means of risk control is rarely acceptable. This is because
tably, at some stage the latter will result in someone no PPE is 100% effective for 100% of the time for 100% of the
having to lift manually a load heavier than it could have people who use it (see Chapter 15). One of many reasons for
been. Avoid using hazardous substances at all where pos- this is that it may not always be possible to identify
sible, or substitute those known to be less hazardous. everyone at risk and issue the PPE to them.

18. 6 PRINCIPLES OF CONSTRUCTION SAFETY 1
Example: Extensive work on a flat roof may require the The cost in human suffering, physical pain and hardship
presence of a large number of workers from different resulting from death and disability is impossible to quantify
employers, in addition to supervisory staff, clients, etc. In Ð we know that there are hundreds of lives lost each year in
these circumstances, the appropriate protection would be construction and related industries, with tens of thousands
provided by perimeter barriers, rather than by giving of serious injuries and countless numbers of more minor
`everyone' a safety harness. Installing permanent edge injuries. We can only guess at the disruption to lives of
protection during the main construction process instead of workers and their families which these cause, but we know
at the end gives protection to both constructors and end that construction safety is not an impossible dream; it is an
users. achievable goal.
Moral reasons stem from a developing public awareness
6. Ensure everyone understands what they have that something needs to be done to raise the quality of life
to do to be safe and healthy at work at work. Attention is focusing on the ability of employers
and project managers in the industry to handle a wide
Safety training is not just a matter of handing out booklets
variety of issues, previously seen only as marginally rele-
Ð it is unwise to assume anything about people's previous
vant to the business. Environmental affairs, pollution,
experiences in construction work, or even their ability to
design safety, maintainability and other matters are now
read and understand instructions and information. Safety
commonly discussed. There is a growing belief that it is
awareness is not inherited, and induction training on sites is
morally unacceptable to put the safety and health of others
required to make sure that everyone knows what the
(inside or outside the construction site) at risk, for profit or
hazards and the control measures are.
any other reason.
Example: Specific site induction must include the local
Worker morale is strengthened by active participation in
emergency evacuation procedure, and understanding may
accident prevention programmes, and is weakened by
need to be confirmed by holding regular practice fire
accidents. Adverse publicity affects the fortunes of the
drills.
organisation both internally in this way and externally, as
public confidence may weaken local community ties,
7. Make sure health and safety management is market position, market share, shareholder value and
accepted by everyone, and that it applies to all reputation generally.
aspects of the organisation's activities
Legal reasons are contained in statute law, which details
A single contracts manager, joinery manager, or other steps to be taken and objectives to be met, and which
member of senior management who believes that the carries the threat of prosecution or other enforcement
company safety management system does not apply to action as a consequence of failure to comply. Civil law
situations where time is short can destroy the safety enables injured workers and others to gain compensation
climate overnight. If someone is injured as a result, the either as a result of breach of statutory duties or because a
penalty can be severe for that person, and possibly for the reasonable standard of care was not provided under the
company as well. particular circumstances. The cost in terms of money and
adverse publicity of a prosecution or civil claim can be very
Examples: Failure of a senior manager to wear safety high, and there is the potential for a prison sentence in
footwear and safety helmet on a site visit gives the some circumstances. For a discussion of these issues, see
impression to the workforce that the rules do not apply to Chapter 17.
senior management. Failure to carry out risk assessments
because of pressure of work could lead to criminal Financial reasons for accident prevention ensure the con-
prosecution of an individual. tinuing financial health of a business and avoid the costs
associated with accidents. These include monetary loss to
employers, community and society from worker injuries
Benefits of accident prevention
and ill-health, damage to property and production delays.
There are generally said to be five main reasons why acci- Some, but not all of these costs are insurable and these are
dent prevention in construction is worthwhile. known as direct costs. They include the cost of compen-

19. 1 FUNDAMENTALS 7
sation (for which insurance is a legal requirement). were not the case. The same amount and quality of infor-
Increased premiums will be a consequence of claims, so an mation is potentially available for each incident, yet we
increase in overheads is predictable following accidents. frequently limit investigations to those incidents where
Indirect costs include: injury or damage is serious Ð at the tops of the triangles Ð
and thus miss the chance of obtaining a lot more informa-
n Uninsured property and material damage tion about what is going wrong. This is why counting and
n Delays investigating `near miss' incidents is useful.
n Overtime costs and temporary labour
n Management time spent on investigations Finally, a good safety record and documented safety
n Decreased output from those replacing the injured management system can more than repay the time spent on
worker(s) it because of its value in gaining new business. Many clients
n Clerical work and project management operations have extensive vetting
n Fines procedures to identify those contractors and suppliers who
n Loss of expertise and experience are competent in safety matters. The vetting may be a
requirement of their quality programme under BS EN 9000
A study carried out by the Health and Safety Executive into accreditation, but in any event in the UK the competence
the costs of accidents showed that for the construction site issue is at the heart of the Construction (Design and Man-
under review, the direct costs were a small proportion of agement) Regulations 1994 (CDM).
the total and produced a direct:indirect ratio of 1:11. This
ratio is commonly illustrated as an `iceberg', because of Conversely, inability to satisfy requirements for compe-
the invisible hidden costs below the waterline. On the site tence in safety under CDM can result in loss of significant
studied over a period of 18 weeks, 120 people were work- contracts as well as public reputation. An electrical con-
ing, and in that time there were 56 minor first-aid injuries tractor working for a local authority was successfully pro-
and no lost-time injuries. But there were also 3570 non- secuted together with the authority for contravening the
injury accidents. The results for major, minor and non- Control of Asbestos at Work Regulations when hand-drilling
injury accidents are often reproduced in the form of an holes in domestic cupboards for wiring. The fine was rela-
`accident triangle'. tively modest, but the contractor's hard-won local repu-
tation for workmanship and reliability suffered, and the
The conclusion to be drawn from this and other accident conviction had to be disclosed on every subsequent appli-
triangles is that serious injuries are much less frequent than cation to join a tender list.
less serious ones, and of course it would be strange if that

20. 2 Where are We Now?
It is widely recognised that health and safety injury statistics On average, a worker is killed in the industry every three
offer only the depressing and not especially useful prospect working days, and a member of the public is killed every
of counting our failures to control injury and ill-health. What two months by construction activities. There are now about
is needed is an agreed system of positive measures of how 80 fatalities a year to workers in the construction industry,
well controls are working Ð but there is no sign of agreement including the self-employed and trainees. This number has
or even substantive moves towards agreement about the remained roughly the same since the early 1990s (Table
nature of the measures that might provide answers. The 2.1). Taking the long-term view for all industries, the
prospect of common use of standards such as the British accident rates overall are a quarter of those reported in the
Standards OHSAS 18000 safety management model is early 1960s, and less than a half of those in the early 1970s.
brighter, but opinion internationally is divided on whether This is thought to be due only partly to changes in
the objectives of continuous improvement and accreditation employment patterns.
should be included in such a standard.
Table 2.1: Fatalities to employees in the construction
Meanwhile, international statistical comparisons mean lit-
industry, sample years
tle, as there are so many measuring tools and so many
different criteria for measurement. There is no general Year Number
agreement on how to calculate frequency rates, or on what 1961 272
counts as a reportable injury. Even in the UK, the rules 1971 156
change every 5 to 10 years, and invalidate previous data 1981 105
sets. And there is good evidence that many injuries go 1990/91 96
unreported. 1995/96 62
1996/97 66
Recent studies by the Health and Safety Executive using the 1997/98 58
Labour Force Surveys (which send questionnaires to 1998/99* 48
households) indicate that only 55% (1997/98) of all non- * Data for 1998/99 were provisional at the time of writing
fatal reportable injuries in construction are actually
reported to the Executive and local authorities. This is
Types of injury
actually an improvement Ð in 1989/90 the figure was as low
as 38%, rising to 46% in 1994/95. Commentary on the dis- The most common source of fatalities in recent years has
parity between reality and what is reported is presented been the head injury, accounting for almost one-third of
annually in the Health and Safety Commission's Statistical the total. There is claimed to be a marked reduction, of
Review. The information can be regarded only as an indi- about 25%, in the head injuries rate overall, following the
cation of the position for any given year. introduction of the Construction (Head Protection) Regu-
lations on 1 April 1990. In most years, at least 40% of all
An improved picture can be drawn from a study of fatal- construction fatalities have been falls from a height. The
ities, which are less easy to ignore, although these too may figure for falls as a percentage of total fatalities has been
be under-reported because of poor diagnosis of ill-health remarkably constant over past years, justifying the atten-
exposures, for example, and also where a significant time tion given to fall protection in its own right in the Con-
may elapse between injury and consequent death. struction Regulations (Table 2.2).

21. 2 WHERE ARE WE NOW? 9
handling, for example. Others occur relatively infre-
Table 2.2: Fatal falls to employees in the construction
quently, but when they do there is a higher than usual
industry over 12 years
chance of not surviving them Ð becoming trapped by col-
All employee Falls as % of lapse or overturning and electrocution are examples of
Year Falls fatalities all fatalities this. Table 2.4 compares the proportionate outcomes of
1987/88 47 99 47.48 various causes of injury.
1988/89 49 101 48.52
1989/90 53 100 53.00 We need this information in order to estimate risk, which is
1990/91 45 96 46.88 a measure combining the chances of something happening
1991/92 37 83 44.58 with the potential outcome in terms of injury. Too much
1992/93 27 69 39.13 reliance should not be placed on the data because of the
1993/94 35 73 47.95 under-reporting factor mentioned above.
1994/95 24 56 42.86
1995/96 21 62 33.87 Although the total numbers engaged in some occupations
1996/97 33 66 50.00 can be expected to fluctuate with time, together with the
1997/98 29 58 50.00 amount and type of work available in the construction
1998/99* 22 48 45.83 industry, it is hard to believe that there has not been an
* Data for 1998/99 were provisional at the time of writing
improvement in health and safety standards in the last 10
years. The improving data are unlikely to be due solely to
changes in the pattern of work and numbers employed.
In 1998/99 the percentage of falls in the `non-fatal major
injury' category was 37%, with an additional 21% as falls Information on rates is not easy to acquire. The Health
from the same level (slips, trips and falls); 35% of report- and Safety Commission and Executive use comparisons
able, over-3-day injuries in the industry were due to between industries based upon accidents recorded per
manual handling. 100 000 workers, rather than upon hours worked. For con-
struction in 1998/99 the provisional rate is 399.0 for
fatalities and major injuries combined per 100 000 work-
Accident causes
ers, and for all reported injuries the rate is given as 1254.
Canadian studies have shown that active involvement in The previous year's rates were 388.1 and 1354.4, respec-
safety management by the most senior levels in a con- tively.
struction company is directly correlated with reductions in
numbers of accidents and injuries. Cross-industry comparisons are of little significance to
individual employers unless their business covers several
Knowledge of causation patterns provides a starting point employment sectors. The best information of this kind is to
for focusing particular preventive measures. Case studies be found by benchmarking against other similar businesses,
and descriptions of accidents can be used to give informa- and comparing internal figures over similar periods.
tion about prevention techniques Ð the Health and Safety
Executive's publication Blackspot Construction is still
Occupational health and hygiene
recommended reading, although now out of print. It com-
mented that in a sample studied, 90% of fatalities were Traditionally the construction industry's high level of injury
found to be preventable, and in 70% of cases positive accidents has received the attention of enforcement,
management action could have saved lives. The three worst media publicity and management action. Arguably, the size
task areas found by the study (75% of all deaths) were of that problem has led to a neglect of the less tangible
maintenance (42%), transport and mobile plant (20%) and consequences of occupational hygiene and health prob-
demolition/dismantling (13%). lems, apart from well-publicised topics such as asbestos.
There is little general awareness of just how big the occu-
Table 2.3 shows the distribution of causes of fatalities in pational health risks are in construction; compare the
the years 1997/98 and 1998/99, considering all workers in numbers already discussed for conventional injuries with
the industry. Some activities, of course, are frequent the fact that mesothelioma, a form of cancer linked spe-
sources of injury, but rarely result in a fatality Ð manual cifically to asbestos exposure, kills around 1400 people

22. 10 PRINCIPLES OF CONSTRUCTION SAFETY 2
Table 2.3: Fatalities in the construction industry by causation
Categories of 1997/98 1998/99
accident causation Employees Self-employed Total % of total Employees Self-employed Total % of total
Falls: >2 m 29 12 41 51.25 21 14 35 53.03
Falls: <2 m Ð 1 1 1.25 Ð 3 3 4.55
Falls: unknown Ð 4 4 5.00 1 Ð 1 1.52
heights
Total falls from 29 17 46 57.50 22 17 39 59.09
heights
Contact with moving 3 Ð 3 3.75 2 Ð 2 3.03
machinery
Struck by moving or 11 1 12 15.00 7 1 8 12.12
falling object
Struck against fixed 1 Ð 1 1.25 Ð Ð Ð Ð
or stationary object
Struck by moving 5 ± 5 6.25 9 Ð 9 13.64
vehicle
Lifting, handling, 1 Ð 1 1.25 Ð Ð Ð Ð
carrying
Trapped by collapse 3 1 4 5.00 3 Ð 3 4.55
Asphyxiation or Ð Ð Ð Ð 1 Ð 1 1.52
drowning
Exposure to harmful Ð Ð Ð Ð 1 Ð 1 1.52
substances
Explosion Ð Ð Ð Ð 1 Ð 1 1.52
Electrocution 5 2 7 8.75 2 Ð 2 3.03
Other Ð 1 1 1.25 Ð Ð Ð Ð
Total 58 22 80 100.00 48 18 66 100.02
each year. A good proportion of these exposures are related requirements) imposed by the Noise at Work Regulations
to construction work. 1989.
A general downgrading of normal health can also occur.
Dangerous occurrences
Reports suggest that construction workers age prematurely
due to hypothermia caused by working in the cold and wet. `Dangerous occurrences' are sets of circumstances which
Respiratory diseases such as bronchitis and asthma are also must be reported to the enforcing authorities if they occur,
thought to occur at above average levels in construction and which are defined within the current set of regulations
workers. containing reporting requirements for injuries. Currently
the relevant regulations are the Reporting of Injuries,
Experience with the implementation of the Control of Diseases and Dangerous Occurrences Regulations 1995
Substances Hazardous to Health (COSHH) Regulations in (RIDDOR). They were introduced in approximately their
construction shows that there is little awareness of the current format in 1980 as the Notification of Accidents and
principles of assessment, or significant appreciation of the Dangerous Occurrences Regulations, or NADO, which
risks to workers from substances brought onto the site Ð introduced the concept of dangerous occurrences for the
and especially from those created there. Also, there is said first time. Since the list of circumstances which are to be
to be a disappointing response from the industry to the reported has changed over the years, and again because of
noise controls (mostly managerial action and measurement significant under-reporting, statistical comparisons are

23. 2 WHERE ARE WE NOW? 11
Table 2.4: Reported injuries to all construction workers, 1997/98 and 1998/99*
Categories of 1997/98 1998/99
accident Fatalities Non-fatal major >3 day Total Fatalities Non-fatal major >3 day Total
causation injuries injuries reportable injuries injuries reportable
Falls: >2 m 41 754 395 1190 35 813 360 1208
Falls: <2 m 1 734 752 1487 3 824 843 1670
Falls: unknown 4 163 179 346 1 120 147 268
heights
Total falls from 46 1651 1326 3023 39 1757 1350 3146
heights
Contact with 3 138 237 378 2 136 283 421
moving machinery
Struck by moving 12 858 1913 2783 8 122 1745 1875
or falling object
Struck against 1 137 542 680 Ð 153 474 627
fixed or stationary
object
Struck by moving 5 104 155 264 9 132 142 283
vehicle
Lifting, handling, 1 370 3633 4004 Ð 360 3324 3684
carrying
Slip, trip or fall, Ð 790 1694 2484 Ð 929 1655 2584
same level
Trapped by 4 55 49 108 3 62 52 117
collapse or
overturn
Asphyxiation or Ð 9 12 21 1 14 6 21
drowning
Exposure to Ð 78 218 296 1 86 200 287
harmful
substances
Fire Ð 13 37 50 Ð 7 17 24
Explosion Ð 4 21 25 1 14 18 33
Electrocution 7 55 96 158 2 75 78 155
Animal injury Ð 2 36 38 Ð Ð 27 27
Acts of violence Ð 14 39 53 Ð 15 41 56
Other 1 46 252 299 Ð 57 187 244
Total 80 4324 10260 14664 66 3919 9599 13584
* Data for 1998/99 were provisional at the time of writing
likely to be misleading. The intention is to bring to the various kinds, 13% advised of substance escape as defined
notice of the enforcing authority those incidents and con- within RIDDOR, and 11% notified pipeline failures. These
ditions which are sufficiently serious to be likely to cause an dominant categories and percentages have remained gen-
injury, even though such an injury did not necessarily erally constant over time since 1995. The construction
occur. industry is likely to be involved in reporting lifting
machinery failures of several kinds, as well as the unin-
In 1998/99, 4173 reports were made of dangerous occur- tended collapse of buildings.
rences. Some 25% dealt with failures of lifting machinery of

24. 12 PRINCIPLES OF CONSTRUCTION SAFETY 2
Acknowledgement: The Health and Safety Executive's Reference
assistance with the provision of statistical information is
(Health and Safety Commission) Statistical Review (published
gratefully acknowledged.
annually). HSE Books, London.

25. Measuring Performance and
3 Recording Information
There are many good reasons for measuring safety perfor- such as first-aid treatments and lost workday cases.
mance; some of them are often forgotten. Measuring can The US standard is to count `disabling injuries', which
enable management to: are those that cost a full person-day. The trend is
towards lowering the severity thresholds for reporting,
n Identify the causal factors involved in injury and loss but this tends in turn to result in under-reporting.
n Locate areas where controls are not working ade-
n How can different levels of risk and exposure variations
quately
be allowed for? Using numbers of hours worked as the
n Have a basis for comparing trends
baseline for exposure measurement does not reflect
n Describe the level of safety within the organisation
differences in risk. What can be learned from com-
n Predict future safety problems
paring the injury data per million person-hours worked
n Evaluate the success of the control programme
between, say, carpet fitters and roofers? The roofers'
n Maximise cost-effectiveness of decisions on the allo-
injury rate might be expected to be much higher,
cation of resources
because they appear to be at greater risk, but that
n Assess the costs of injuries and losses
takes no account of the amount of time spent, the
n Benchmark against other similar organisations
relative exposure to risk, whilst at work. The best that
can be done is to compare like with like where pos-
Before any useful conclusions can be drawn from any set of
sible, and to include severity rates where available.
data, the information collected must be both reliable and
meaningful. Are the right things being recorded? Do the
For several reasons, measuring tools which are purely
numbers give the whole picture? How many injuries and
injury-related do not do a good job of representing the
incidents are unreported? What distortions may be present?
quality of the performance effort or the safety climate in
And what gives information about `safety performance'?
an organisation.
Until recently, the only measuring tools available were the
Behaviour-based safety
counting of failures (lost-time injuries however defined,
first-aid cases, property damage incidents and `near A set of techniques known generally as `behaviour-based
misses'), and attempts at measuring the financial costs of safety' has introduced ideas and methods from the beha-
losses resulting from failures to control safety, health and vioural sciences into performance measurement and safety
the environment. All of these involve studying the evidence management. A full discussion of the techniques used is
of failures in one form or another, rather than the perfor- beyond the scope of this book, but they are based on the
mance achieved. And there are difficulties in collecting the claim that measuring the frequency of safe behaviour
evidence, for example: generates more, and more accurate, predictive data,
allows for precise reinforcement and provides positive
n How severe must an injury be to be counted at all? As accountability. The general principle involves sampling,
mentioned in the previous chapter, this definition recording and publicising the percentage of safe (versus
varies widely between countries. The UK minimum unsafe) behaviours noted by observers drawn from work-
reporting requirement is set at 3 days' absence from force and management, and specially trained. This gives
work, although many of the larger construction com- more data on potential system and individual failures than
panies now measure lower severity level indicators could be obtained from a study of accident records.

26. 14 PRINCIPLES OF CONSTRUCTION SAFETY 3
Staff at the University of Manchester Institute of Science nificant information and is recorded for analysis in a sui-
and Technology (UMIST) have extensive experience in the table format. Basic training is also required, to ensure that
application of behaviour-based safety systems to the con- those given the responsibility are fully committed to the
struction industry. In the United States, the work of Dr goals of investigating and recording all significant inci-
Thomas Krause and his colleagues at Behavioral Science dents, and that the techniques involved are understood.
Technology Inc, California, is particularly well known. It
has shown that a behaviour-based approach to health and The way in which injuries, potential injuries (`near-
safety management can be an effective tool for increasing misses'), occupational ill-health exposures and environ-
safety on construction sites and elsewhere, despite some mental incidents are investigated and the statistics are
practical problems of implementation. collected should be written into the organisation's safety
policy or equivalent document. This gives the process a
Employers investing in these techniques say they have mandate from the person in charge, without which no
found that the involvement of workers in the measuring efforts of significance in safety are likely to be successful.
process generates interest and improved commitment to
the employer's safety objectives. The results are said to be One test of the efficiency of a data collection system is
significant in that the techniques lead to a reduction in loss- whether `near-miss' incidents are reported and recorded.
producing incidents as well as to the improved assessment Their investigation will provide the same information on
of performance by the positive step of measuring workers' causation as `real' injury incidents.
safe actions.
`All I want to know is the facts!'
For the purpose of this chapter, it is noted that the
attraction is that the system offers a method of measuring In addition to distortions caused by under-reporting and
the potential for harm, independent of the accident factors introduced by the way the data are collected,
record. Disadvantages may include the need to achieve an whatever data are provided to management may be subject
altered safety climate for both management and workforce to misinterpretation by anyone who does not fully under-
to adopt the techniques, and employee suspicion of hidden stand the nature of what is being presented to them. For
motives for the observations. example, it has been reliably estimated that about 375 000
people were killed by accidents in the United States during
World War II. About 408 000 were killed by war action. It has
`No injuries Ð no problems!'
been claimed as a result that it was nearly as dangerous to
Because the numbers of recorded incidents and injuries are stay at home as it was to be in the armed services.
relatively low in most companies, they produce a limited
amount of information about risk and there is a temptation A moment's thought should lead to questions about rates
to believe that all is well. The argument often put forward rather than numbers. It turns out that the death rate in the
by managers Ð `We haven't had any accidents, therefore US armed forces during World War II was about 12 per
we must be safe' Ð takes no account of the potential for thousand men per year, which compares with the overall
injury, or risk, which must be evaluated when deciding on civilian accidental death rate of about 0.7 per thousand per
appropriate measures to take. year.
When things do go wrong, the information which can be Other influencing factors include the ages of those exposed
obtained from recording and investigating incidents can be to the hazards, and the duration and type of exposure to
both substantial and useful, depending to a large degree on the hazard under analysis. The raw numbers do not, in fact,
the methods used to collect information on individual tell us much at all about the chances of survival at the time.
incidents and throughout the organisation. The success of
any collection method also depends on the commitment of
Calculating rates
individuals to supplying the information in a timely and
appropriate way. This means that organisations need to The simple formulae for calculating frequency and severity
have a system for reporting and recording injuries and rates which follow produce the rates used most often in the
other losses which is seen by all as reasonable under all the industry. In most countries, there are no standard or formal
circumstances, is sufficiently thorough to capture all sig- requirements for these formulae.

27. 3 MEASURING PERFORMANCE AND RECORDING INFORMATION 15
Frequency rate percentage of subcontractors working on a project that had
actually been pre-qualified. The percentage of sub-
Number of injuries  100 000 contractors who had submitted a competent method
Fˆ
Total number of hours worked
statement prior to starting work could be another.
In the USA, figures of 1 000 000 or 200 000 often replace the
These internal yardsticks can also be given targets to
100 000 in the numerator, depending upon the collecting
achieve, so that continuous improvement can be made by
agency.
raising the targets year on year. The setting of such key
performance indicators is one of the features that distin-
Severity rate
guishes those companies interested in establishing them-
selves as `world class' rather than being merely
This is essentially a weighted frequency rate, allowing the
`compliance orientated'.
days lost due to temporary total disability to be recorded,
and also a notional number of days to be recorded for
fatalities and permanently disabled cases. The notional Accident investigation and recording
days often used are 6000 (20 working years at 300 days per
The elements of an accident recording system consist of:
year) for a fatality and 1800 for loss of an eye. The Amer-
ican National Standards Institute (ANSI) has developed an
n Report form(s)
arbitrary schedule of notional days in relation to particular
n Investigation reports Ð format
permanently disabling injuries, of which the foregoing are
n Summary analysis forms used by the data collector
two examples.
n Statistical analysis
Total days lost plus notional days charged  100 000 n Summary reports for management
Sˆ
Employee hours of exposure
In large organisations, or those with widely-spread sites,
The same numerator should be selected as in the frequency the system may also use a fax or email notification system
rate calculation. giving early warning to senior management that an incident
has occurred which requires their attention. Care should be
The frequency rate can be improved `artificially' by con- taken to avoid making personal comments in emails, as they
trolling temporary total absence cases (often by manage- are likely to become `discoverable' in legal proceedings Ð
ment policy, and by making jobs available for temporary visible to everybody. A good rule to follow is: `If you don't
workers convalescing). Similarly, the severity rate can want your opinion or comment known to potentially hostile
fluctuate wildly because the schedule of notional days can strangers, don't send the email'. Deleting an email from
impose a severe notional time penalty for some injuries. the system usually has no effect on its viability on a server,
somewhere.
Other performance measures
Standard report format
Many companies are now introducing positive measures of
performance in the field of environment, health and Use of a suitable standard report form allows the collection
safety. Once a safety management programme has been of information in a uniform way. The design of the report
developed, key points in it can be identified and measured form is important. As the forms will usually be completed at
to provide information on whether the system is working site level it will assist site staff if they are only asked to give
properly. For example, it might be decided that one of the information which is likely to be readily available to them Ð
critical features in such a programme is the need to ensure social security numbers and other personnel information
that necessary safety information is supplied to sub- may be restricted, or held elsewhere. The penalty for using
contractors prior to tendering. Whether this happens or not a format calling for answers that site staff cannot provide
can be measured by examining project documentation, at can be delays in the return of forms. Whatever detail is
perhaps quarterly intervals and a score produced based on asked for, and whatever the final design of a report form, it
the percentage of projects where this is being done. Or, if a will be helpful to require at least those answers to be given
company operates a pre-qualification scheme for sub- which are required by local or national authorities, when
contractors, measures could be developed to show the notifying them in turn.

28. 16 PRINCIPLES OF CONSTRUCTION SAFETY 3
Report forms should not require the senders to: types of incident and their consequences so that common
causal factors can be isolated, future problem areas can be
n Assign blame for the incident predicted, training can be focused and trends assessed.
n Make comments which the senders cannot sub- Depending upon the size of the organisation, it may be
stantiate, or necessary to collect data site by site and collate that into a
n Draw conclusions which may be beyond their level of monthly executive summary report.
competence
If benchmarking is carried out against other organisations
This is because the form may be required as evidence in or against industry records, it will be helpful to design
legal proceedings, where liability is an issue and state- summary forms which use the same categories as the
ments of this type may be found to constitute an admission benchmark targets. Otherwise, the format used by the
of liability where none was intended. regulatory authorities to present data can be used.
Speed is of the essence Ð it may be desirable to design an
Statistical analysis
initial report form to be faxed to senior management as
soon as practicable, to be followed by a more detailed It is always tempting to compare the results from the
report when time and circumstances permit. The detail organisation with national or industry data, but the relative
may be completed later, following a deeper investigation sizes of the samples must be remembered when doing so.
and interviews with witnesses. Also, industry frequency and severity rates, where pub-
lished, are often based on guesswork on the hours worked,
and can take no account of under-reporting of injuries to
Investigation report format
the authorities. Generally, the best comparison to use is a
The person or team carrying out an investigation into an previous time period within the same business or bench-
incident should record their comments in writing. This may marking partners.
or may not involve the completion of an internal investi-
gation report form. Again, care should be taken to avoid
Summary reports for management
making statements or comments which are not factual.
Recommendations to prevent a recurrence should be made Monthly or quarterly summaries of injury figures should be
in the form of a letter attachment rather than on a report presented in a format which makes valid comparisons easy,
form, again for legal reasons. and with a short written account to provide a summary of
selected incidents. The use of pie charts and other pre-
When completing an investigation report, it is important to sentation aids should be considered. It is important that the
be aware that a potential reader may know nothing about person designated as the competent person for the pur-
circumstances or techniques which the investigator may poses of the Management of Health and Safety at Work
take for granted. For this reason, no assumptions should be Regulations 1999 is given a summary report at regular
made about the level of knowledge possessed by the intervals. A clutch of statistics alone rarely provides enough
report's readership. Sketches, plans, drawings and espe- meaningful information for senior management. A written
cially photographs should be included to amplify the writ- summary of lessons learned, actions taken and the current
ten report. status of the organisation's rolling safety programme
should also be included.
The use of a standard layout in a typed report can assist the
investigator, because it can help to clarify thought. A sui-
Principles of accident investigation
table format for a written report is discussed later in this
chapter. The hardest lessons to be learned in accident prevention
come from the investigation of accidents and incidents
which could have caused injury or loss. Facing up to those
Summary analysis forms used by the data
lessons can be traumatic for all concerned, which is one
collector
reason why investigations are often incomplete and sim-
In order to make the best use of data supplied, the system plistic. Nevertheless, the depth required of an investiga-
of performance measurement needs a way of recording the tion must be a function of the value it has for the

29. 3 MEASURING PERFORMANCE AND RECORDING INFORMATION 17
organisation and other bodies which may make use of the Documentation
results, such as enforcement agencies. Conducting one can
Information obtained during investigations is given verb-
be expensive in time.
ally, or provided in writing. Written documentation should
be gathered to provide evidence of policy or practice fol-
Purpose lowed on site, and witnesses should be talked to as soon as
possible after the accident. The injured person should also
The number of purposes is large; the amount of detail be seen promptly and interviewed.
necessary in the report depends upon the uses to be made
of it. Enforcement agencies look for evidence of blame, Key points to note about investigations are:
claims specialists look for evidence of liability, trainers
look for enough material for a case study. From the view- n Events and issues under examination should not be
point of prevention, the purpose of the investigation and prejudged by the investigator
report is to establish whether a recurrence can be pre- n Total reliance should not be placed on a single source
vented, or its effects lessened, by the introduction of of evidence
safeguards, procedures, training and information, or any n The value of witness statements is proportional to the
combination of these. amount of time which passes between the events or
circumstances described and the date of a statement
or written record. (Theorising by witnesses increases
Procedure as memory decreases)
There should be a defined procedure for investigating all n The first focus of the investigation should be on when,
accidents, however serious or trivial they may appear to where, to whom and the outcome of the incident
be. The presence of a form and checklist will help to con- n The second focus should be on how and why, giving the
centrate attention on the important details. The manage- immediate cause of the injury or loss, and then the
ment team of the project where the accident occurred will secondary or contributory causes
be involved; for less serious accidents they may be the only n The amount of detail required from the investigation
people who take part in the investigation and reporting will depend upon (a) the severity of the outcome and
procedure. Workers' representatives may also be involved (b) the use to be made of the investigation and
as part of the investigating team. report
n The report should be as short as possible, and as long as
necessary for its purpose(s).
Equipment
The following are essential tools in the competent inves-
The report
tigation of accidents and damage/loss incidents:
For all purposes, the report which emerges from the
n Report form, possibly a checklist as a routine prompt investigation must provide answers to the following ques-
for basic questions tions. Only the amount of detail provided should vary in
n Notebook or pad of paper response to the different needs of the recipients.
n Tape recorder for on-site comments or to assist in
interviews n What was the immediate cause of the accident/injury/
n Camera Ð Polaroid instant-picture cameras are useful loss?
(but further reproduction of the results may be diffi- n What were the contributory causes?
cult and expensive, or they may be of poor quality). n What is the necessary corrective action?
Their improved performance and the ability to insert n What system changes are either necessary or desirable
photographs into text now makes the use of digital to prevent a recurrence?
cameras attractive n What reviews are needed of policies and procedures
n Measuring tape, which should be long enough and (for example, risk assessments)?
robust, like a surveyor's tape
n Special equipment in relation to the particular inves- It is not the task of the investigation report to allocate
tigation, e.g. meters, plans, video recorder individual blame, although some discussion of this is

30. 18 PRINCIPLES OF CONSTRUCTION SAFETY 3
almost inevitable. Reports are usually `discoverable'; this Inspections
means they can be used by the parties to an action for
damages or criminal charges. It is a sound policy to assume Inspections should be based on a positive approach, seeking
that accident investigation reports will be seen by solicitors to establish what is good and well done as well as what is
acting on behalf of the injured party. They are entitled to not. Too often the `inspection' process has a negative
see the factual report, and this will include anything writ- implication associated with fault-finding. The inspection of
ten in it or in connection with it (which might later prove sites and premises has three main objectives:
embarrassing), so certainly it should not contain comments
on the extent of blame attaching to those concerned, or n Identification of hazards, triggering the corrective
advice given to management. It is appropriate, necessary process
and quite proper that professional advice is given, but it n Improving conditions and reducing risks
should be provided in a covering letter or memorandum n Measuring safety performance
suitably marked `confidential'. Changes in the rules gov-
erning the civil liability claims process mean that `side Some common system should be followed for every
letters' and other formerly acceptable means of conveying inspection to make sure that everything relevant has been
concern `off the record' have become potentially dis- covered. Checklists can be used, and an adequate reporting
coverable. system must be present so that a record is made of what
needs attention, and management can be advised of the
Whether the report is made on a standard form, or specially results of the inspection. Ideally, inspections should be
written, it should contain the following headings: measurable so that comparisons can be made with stan-
dards elsewhere in the business.
n A summary of what happened
n An introductory summary of events prior to the acci- Inspection for health and safety purposes often has a
dent negative implication, associated with fault-finding. A
n Information gained during investigation positive approach based on fact-finding will produce better
n Details of witnesses results, and co-operation from all those taking part in the
n Information about injury or loss sustained process.
n Conclusions
n Recommendations There are a number of types of inspections, for example:
n Supporting material (photographs, diagrams to clarify)
added as appendices n Statutory Ð for compliance with health and safety
n The date, and it should be signed by the person or legislation
persons carrying out the investigation n External Ð by enforcement officials, insurers, con-
sultants
Sample standard report forms are included at the end of n Executive Ð senior management tours
this chapter (Figs 3.1 and 3.2). n Scheduled Ð planned at appropriate intervals, by
supervisors
n Introductory Ð check on new or reconditioned equip-
Inspections and audits
ment
Audits look at systems and the way they function in prac- n Continuous Ð by employees, supervisors, which can be
tice, inspections look at physical conditions. So, while formal and preplanned, or informal
inspections of a site, or particular items of equipment,
could (or possibly should) be done formally at least For any inspection, knowledge of what is being looked at
weekly, an audit of the inspection system throughout an is required, also knowledge of applicable regulations,
organisation would look at whether the required inspec- standards and Codes of Practice. Some system must be
tions were themselves being carried out, the way they followed to ensure that all relevant matters have been
were being recorded, who received copies of the considered, and an adequate reporting system must be in
report, whether action was taken promptly as a result, place so that remedial actions necessary can be taken
and so on. More information on audits can be found in the and that the results of the inspection are available to
next chapter. management.