The benefits from comprehensive measurement of performance within a health and safety management system can aid elimination of occupational illness and injuries, and it can also verify legal compliance and measure continuous improvement.

Measurement is necessary to understand how efficient the company is against its objectives, its policy, and its rivals. Without measurement, it is diificult for the organisation to know what progress has been made in relation to its targets and goals. The information gained from such measurement will provide a base to prove adherence to standards, best practice, the law, corporate governance and defence from litigation.

Without an understanding of the benefits that can be gained, and the pitfalls that are associated with measurement of health and safety performance (as well as those that can arise from the statistics themselves), full advantage will not be taken from the measurement and valuable resources will be wasted. ROSPA (2001 P2 Part E) indicates that those wasted resources can arise from costs incurred during auditing, maintenance of record systems, monitoring and analysis of the data, editing and publishing. Indirect costs can arise from negative impressions of the organisation safety performance, negative media coverage, and used as evidence in civil proceedings.

Successful health and safety management and the benefits of measurement are adequately defined in the following quote from HS(G) 65 1991. p 55:

"Organisations achieving success in health and safety measure their performance against pre-determined plans and standards, the implementation and effectiveness of which they assess as a basis for taking appropriate remedial action. This leads them to establish, operate and maintain systems, which ensure that performance is measured objectively."

The mission

IOHA (2001 Report P3 Appendix B) confirms that a health and safety policy is a starting point and a foundation document from which an organisation's goals, objectives and performance measures are developed. It also expresses the organisation's mission, targets and values. Those values should continue to be developed into defined health and safety objectives, which starts with a manual.

The manual should state areas of responsibility, lines of communication, consultation, documentation, training, information and performance measurement that is essential to eliminating injury and verify continual improvement with systems that are adequate for the size and complexity of the organisation.

"The predecessors of an organised manual are a wide assortment of memoranda, instructions, marked prints, and other bits of information. These have multiple origins and end up in desk drawers, filing cabinets, people's heads, and wastebaskets. As these things multiply, so does the job of finding authoritative information. Finally, a state is reached where the company concludes that a more orderly approach is needed."
Juran 1974. P6-25.

Identifying risk

The next step from this development are the procedures which should develop performance measures that are consistent with the policy, goals, objectives, and measures both preventive and training performance indicators associated with all identified risks within the organisation.

The need to develop and integrate management systems has been stated by the Department of Environment, Transport and the Regions (DETR) (2001 p18, paragraph vi):

"This culture must be further supported through the full integration of health and safety within general management systems."

Without those policies, objectives and targets to act as a baseline it can be very difficult to justify measurement, as indicated by Peat and Warwick (1992, p18). "Unless a company is clear about where it is and where it aims to go, performance measurement is unlikely to be effective."

Evaluate and review

A comprehensive baseline evaluation or review of an organisation's health and safety management system, practices and hazards are necessary before an occupational health and safety management system can be designed or implemented and therefore measured.

The evaluation must identify organisational hazards and their associated risks clearly; this information is essential to develop training need's analysis, training standards, hazard control systems, and emergency preparedness and response systems.

Failures to properly identify, categorise, prioritise, hazards and associated risks can lead to inadequate employee protection or protective measures that are excessive.

Control systems should set and assess improvement targets within the context of continual improvement and aim to reduce or eliminate hazards, typical control methods are written safe working instructions, surveys and documentation such as permits to work, personal protective equipment, engineering controls or hazard specific training.

Defining and measuring the hazards within a company and the methods used will depend upon the risks associated with that organisation and the size/complexity of the system or organisation itself. The system must be practical and adequate for the size of the organisation and may range from the use of HSG 183 5 Steps to risk assessment, to carrying out Hazard and Operability Studies, Hazard Analysis, or Event Tree Analysis as indicated by Scarman Centre (2001. M4 p21-24). The benefit of carrying out adequate measurement and categorisation, is that once they are quantified, they can provide priorities that may prove action taken was timely and adequate, so far as was reasonably practicable.

Often the first measurements taken by inexperienced safety professionals and companies who wish to manage their health and safety system are derived from accident statistics. One of the reasons for this is that national information has been readily available for years, is categorised and published on a regular basis and the statistics can be directly compared with performance within specific industry sectors.

Companies have had to report certain categories of accidents as defined within the Reporting of Injury Disease Dangerous Occurrences Regulations (RIDDOR) 1995 and record accidents within BI 510, 1988. This book satisfies the recording requirements of RIDDOR and the Social Security Administration Act 1992.

Consequently, much information is already available to collate and compare; paradoxically it is also one of the hardest performance measures to manage. The reliance purely on accident statistics as a safety performance measurement can lead to decisions being made from unreliable information.

An industry or small company with few accidents will inevitably experience large variations in their reporting rate; random fluctuations will mean the number of accidents cannot be completely relied on as an indicator of performance.

The following quote by ROSPA (2000, p4 Part D ) sums this up quite succinctly. "In other words, the variability of accident rates in the UK industry is so large that the probability of making an error in the interpretation of the results is nearly 100%!"

Boom and bust

To measure accidents is to measure failures; accident figures can also reflect problems that have already been removed. Without care and an understanding of the results, measurement can lead to valuable resources being directed to remedial action for regular negligible accidents, instead of preventing fatalities which occur less often. It can also lead to a boom and bust cycle caused by management concentrating on one particular key performance indicator, such as high accident statistics, until the action taken reduces the figures.

Management attention is then redirected to another problem until the figures return to a level that causes concern again, this type of crisis management and reactive monitoring fosters cynicism and lack of commitment to the organisations ideals and goals.

Safety management performance can only be improved through measurement of indicators that are linked to the end product, if accidents are the end result (symptom), measurements of linked problems that can be identified before an accident occurs (cause) should be taken.

Investigation of the accident chain will identify parameters such as inadequate risk assessment, near miss incidents, poor/lack of training, behavioral problems, poor moral, absenteeism, ill health, housekeeping, tripping hazards and other indicators that are able to provide a wealth of information, which will indicate safety performance adequately.

Any relevant, quantifiable indicator can be chosen, unfortunately each indicator gives rise to its own particular problems and they are harder to collect and measure than simple accident statistics. A further difficulty can arise from the fact that few managers have responsibility for all these areas within an organisation, this can provide conflict from managers who feel threatened by people questioning or measuring what they and their departments do.

Near-miss data can be an indicator of future accident statistics but is a difficult parameter to define and measure. Once the initial problem of convincing people to report incidents, which did not actually result in an accident, cause any injury, damage to buildings, machinery, etc, is overcome, there are the problems associated with the reports themselves.

Some have to be redefined, due to being dangerous occurrences. Some are not actually near-miss incidents, but are the result of the reporting person's perception; this can be due to their dread factor or lack of experience, which is defined in Scarmen Centre (2001M1 p5 22-24). Some are safety improvement requests and some can be a general moan, which may be an indicator of poor/low moral.

Each has to be addressed, reported on and action taken with that action being made available to the person who reported, failure to do so may cause that person not to report again. The benefits gained from collecting near-miss data are not only the data itself but the provision of a platform for buy in by employees, who feel they are an important part of the system by reporting.

Hannagan (1998. p322-324) indicates that data from behavioral measurement has several advantages: it can be targeted by managers carrying out employee appraisals and used within goal setting techniques to develop commitment of employees to safety. This can also be measured and quantified by setting specific targets which have to be achieved by the next appraisal (results based) that helps managers create a culture of safety in a positive target based way.

Employees must be set safety objectives as well as performance and training objectives, to prevent safety being viewed as the job of a particular profession or level of management. There are many benefits, which can arise from behavioral measurement; the main pitfall is that it takes a great deal of time and dedication from all levels within the organisation, without that commitment, it cannot succeed.

The bigger picture

Having shown that the measurement of single outcome accident statistics are not an adequate indicator of an organisations performance in health and safety and that measurement of many other indicators suitably cross referenced are required to gain meaning from those statistics. It is appropriate to progress to the bigger picture and measurement of the management system as a whole.

Sayle (1985 p29) defines quality assurance as: "Integrated management systems that provide an assurance that the contractual and legal obligations of the company to the customer and the community are efficaciously fulfilled".

He also indicates on p30 that quality assurance involves planning how a job is to be done and planning for the four elements needed (person, item, equipment, documents) for each task, to execute the job and getting the correct resources together in order to execute the plan. This is the basic requirement of a health and safety management system designed to prevent injury or company loss. The measurement method used within quality assurance is the management audit.

Sayle (1985 Pix) states: "A key technique for assuring that obligations are being met is the audit performed on behalf of management, the management audit."

If a vessel entry procedure were managed within a total quality assurance context, then the operation of the system would require periodical examination by audits. The auditor would look for non-conformities such as permits not completed correctly, pre-entry analysis for quantity, quality and adequacy, rescue equipment availability, adequacy of training and co-operation between disciplines.

Continual evaluation of the total system and procedures that measures conformity enables the problems to be identified and corrected. One of the pitfalls, which can arise from this type of total quality management viewpoint, is that systems and paperwork can, without care, become overwhelming and unusable due to its size and complication.

The following quote by Smith et al (1998 p4) states the requirements. "The aim should be to address those OH&S issues present, managing them appropriately for the size and nature of the organisation and the level of risk that exist."

There are many benefits and pitfalls from measuring performance using auditing principles and techniques. The first consideration should be the qualifications, knowledge and experience of the auditors. An auditor should be trained in how to carry out an audit and given practical experience by observing an experienced auditor, before being allowed to carry out an audit alone or as part of a team.

An inexperienced auditor can do untold damage within a department as the following quote by Sayle (1985 P5) indicates: "Such auditors are unlikely to produce constructive analyses; often they leave a trail of destructive criticism and mayhem in their wake."

An auditor should be able to converse and report confidently to employees at all levels within an organisation. An auditor lacking confidence to deal with upper management may be brow beaten into not recording minor infringements. The ability to converse confidently is necessary as a great deal of information is gained through talking to employees as indicated by Hoyle (1998 p98): "The auditor obtains much of the information through interviewing people and this is not a simple task, especially when such people would rather not disclose their weaknesses or that they had been told not to by their manager."

Someone who has a vested interest in the department or system being audited should never be allowed to carry out an audit; their observation and conclusions may reflect preconceived ideas from the management of the department being audited.

This point is reflected in the following quote by Sayle (1985 p1): "It is common for senior and middle management to see their own operation through rose-coloured spectacles, some times to such an extent that any similarity between the official procedure and actual practice is purely coincidental."

When measuring performance within a department; it should not be assumed that the same measurement would produce the same results in another department, the manager's style and commitment in individual departments may produce different results.

The audit process should be suitably defined and documented indicating its frequency and parameters, those parameters should include the success criteria. The following quote by Hoyle (1998 P58) suggests why success should be included within the audit parameters; "Both the audit team and company need to agree on the success criteria, otherwise there will be dispute and argument".

The frequency will need to be defined in such a manner that it does not state that all systems will be audited every twelve months, this does not allow auditing according to the dangers inherent within the system, a particular process or an increase in incidents.

Without the parameters adequately defined commitment from the auditing team and management will not be gained as indicated by Hoyle (1998 p1): "For audits to be useful they have to be against standards, which the organization is committed to meeting, otherwise they will not be taken seriously."

During the audit, the skill and expertise of the auditor becomes apparent by their ability to gain agreement from the departments being audite. This is often gained by showing that they are looking for conformity to requirements not looking to discredit they or their department.

It is often the case that management provides a memorandum for employees instructing them to carry out a particular task and even provide a tick list or form for them to fill in to eliminate a particular problem. When at a later date, the reason for filling in the form has been forgotten and management does not appear to inspect the form, it will usually fall into disuse - this is often the fate of good intention not brought into an official management system.

Once the audit has been completed, the result should be able to be defined as by the following quote:

"...establish, by unbiased means, factual information on some aspect of performance..., audits are performed as a safeguard against deterioration in standards. Some variation may be tolerable, other variation may be unacceptable and if left to continue may well signify loss of business, credibility and customer confidence."
Hoyle. 1998 p1

The organisation must then to address any nonconformance according to its policy. The audit may not have established the root cause of the nonconformance and a similar approach should be made to any nonconformance raised by an audit, as is taken when investigating an accident. The chain of events that has caused the nonconformance must be considered - it may be the symptom and not the cause which is reported. Once defined, the issues need to be collated, categorised and addressed according to importance to the company policy, goals and targets.

The resulting information and data allows quality techniques such a statistical process control to be used to detect trends. The advantage of using this method to gain meaning out of statistics is that it makes use of minimum resources to achieve maximum effect. The resultant information, which has been measured and collated, must then be presented in a suitable manner to ensure it is used to its full potential; failure to do so will result in wasted effort and resources. Oakland (1986 p17) points out that: "Much of this potential information can lie dormant or not be used to the full, often due to the lack of visual presentation."

There are few voices raised against the total management systems viewpoint but the following quote shows concern from people who believe that we can overburden ourselves with systems to the extent we are no longer able to react quickly. A viewpoint with which the author agrees, but also believes we should adapt and address the systems, not throw them out. Without systems there is anarchy, lack of control and accountability, a recipe for disaster in any organisation.

"...an entrepreneurial organisation, driven by competition, adapts in real time to a constantly changing environment. A "standard" or "family of standards" may in fact not add value to face such daily challenges."
ICSA. 2001

In summary, safety, health, environment law, corporate governance and quality standards have similar requirements; all require adequate management and systems. Those systems must provide responsibilities and an audit trail that proves measurement and action taken to remedy inadequacy and provide for improvement. A quality issue inspected in the light of health, safety or environmental regulation can often provide concern and become a safety, environmental or health problem. Because of the links, similar approaches can be taken and similar tools used.

Measurement of performance, as has been shown, has many advantages and pitfalls. The measurement of a particular outcome can provide information that cannot be relied upon unless measurement of the variable parameters, which affect the outcome, is also measured.

Measurement of health and safety performance and measurement of the management system requires to be measured as a whole. The collection of individual performance statistics which cannot be related to the complete organizations performance, are at best individual and at worst misleading.

A wider, more consistent approach, taking a holistic view of the safety management systems and using planned active monitoring and unplanned spot checks, suitably quantified, increases the perceived importance of safety and gives more sustainable credible results.

Dealing with the entire organisation, instead of separate systems designed to achieve a particular end or accreditation provides efficiency and reduction of man-hours, paperwork and systems.

The fact is that audits can be carried out against any measurable target from contractor performance on site, to waste management; the important consideration must be defining the parameters and defining what constitutes a successful audit.

The measurement of the health and safety management system using audits provides further information and knowledge of the systems to the auditor, the manager of the area being audited and the employees who are questioned on their activities and understanding.

Constant and regular auditing of the systems reduces the amount of training that needs to be given on the systems and can give impetus to employee buy in to a system which they may not have had any input into when first devised.

Inspections and tours, etc, suitably integrated into the management system, provide a cohesive and defined method of categorising the whole organisation's risks. As explained in this article, these need to be quantified to provide adequate measurement. Quantification not only provides statistics that management can quickly understand in the form of graphs, it also provides measurement which a simple yes or no to an inspection question fails to do.

Index

DETR 2001 June. Revitalising Health and Safety Strategy Statement. Accessed June 2001. http://www.detr.gov.uk/pubs/index.htm

Hannagan Tim. 1998 Second Edition. Management Concepts & Practices. Pearson Education Limited Essex.

Hoyle David. 1998. ISO 9000 Quality System Assessment Handbook. MPG Books Ltd, Bodmin, Cornwall.

HS(G) 65. 1991. Sixth edition. Successful health and safety management. HSE Books Sudbury.

HS(G) 183. 1999. 5 steps to risk assessment. Case studies. HSE Books Sudbury.

ICSCA news. Accessed 4-9-2001. http://www.icsca.org.au/

IOHA Report to ILO on an International OHSMS. Accessed 19-9-01
http://www.ilo.org/public/english/protection/safework/cis/managment/ioha/app_b.htm
Juran. 1974. Third edition.. Quality Control Handbook. Kingsport Press, United States of America.

Oakland J S. (1986) Statistical Process Control. Heinmann Professional Publishing. London.

Peat and Warwick. September 1992. Business in the Environment and KPMG. Blueberry London

Rimington John 1991 Forward within HS(G) 65. 1991. Sixth edition. Successful health and safety management. HSE Books Sudbury.

ROSPA 2001. Direct Action on Safety and Health. Measuring and Reporting on Corporate Health and Safety Performance. (Final Draft) Royal Society for the Prevention of Accidents. Accessed 17-9-01.
http://www.rospa.co.uk/ohs/dash/part_e.htm
http://www.rospa.co.uk/ohs/dash/part_d.htm

Sayle J Allan. 1985. Management Audits. The assessment of quality management systems. Robert Hartnoll Ltd, Bodmin, Cornwall.

Scarman Centre. 2001. MSc/Postgraduate Diploma. University of Leicester.

Smith David. Hunt Geoff. Green Clive. 1998. Managing Safety the Systems Way. BSI London