Measurement of Safety Performance

Modem behavior-modification programs rely on the identification and reinforcement of safe behaviors. Considerable improvements in measures of safety performance have been attributed to the introduction of these approaches (see McSween, 1993, for a petrochemical example). However, other studies have indicated that performance may return to its original level if the programs are withdrawn. It is therefore important to maintain a continuing program to ensure that the initial levels of improvements are maintained. Also, the benefits of behavior modification programs have mainly been demonstrated in the context of work activities where there is a high level of... 

Module 7 (Evaluation) consists of the feedback loop indicated in figure 4.1, and of an external measure of safety performance called the Safety Compliance Index . 

Measurement of Safety Performance. Accident/Incident Investigation and Analysis... 

From 1933 to 1999, the death rate dropped 90%. It is recognized that fatality rates are not the only measure of safety performance. Nevertheless, these numbers are significant. 

Allied to Deming s view is the work of Alphonse Chapanis, who was prominent in the field of ergonomics and human factors engineering. Representative of Chapanis writings is The Error-Provocative Situation, a chapter in The Measurement of Safety Performance, by William E. Tarrants (Tarrants, p. 119). 

Tarrants, William E. The Measurement of Safety Performance. New York Garland Publishing Company, 1980. 

For additional comment on the critical incident technique, see Safely Management by Grimaldi and Simonds, p. 248 Accident Prevention Manual Administration Programs, 12th edition. National Safety Council, p. 101 The Critical Incident Technique as a Method of Identifying Potential Accident Causes, Chapter 17 in The Measurement of Safety Performance by William E. Tarrants.)... 

Alphonse Chapanis is exceptionally well known in ergonomics and human factors engineering circles, and his writings on avoiding the design of work that is error-provocative are often cited. These are excerpts from his chapter titled The Error-Provocative Situation, which is in the book The Measurement of Safety Performance ... 

Later, while researching for Chapter 24, Measurement of Safety Performance, I observed that companies with superior OSHA rates also had lower workers compensation costs than did other companies. I then wanted to determine whether the quality of investigation of hazards-related incidents would also be superior. (For this second study, it must also be said that the methodology used would not stand the test of good science.)... 

To extend that idea, this example is quoted from an essay written by Alphonse Chapanis and titled The Error-Provocative Situation A Central Measurement Problem in Human Factors Engineering. It is in the book The Measurement of Safety Performance by Dr. William E. Tarrants. 

These are excerpts from his chapter titled The Error-Provocative Situation in the book The Measurement of Safety Performance ... 

Chapter 24, Measurement of Safety Performance, discusses performance measurement. The guidance given in the citation above is pertinent... 

According to William E. Tarrants, editor of The Measurement of Safety Performance and author of several of its chapters ... 

One must recognize that the main function of a measure of safety performance is to describe the safety level within an organization. .. in effect, then, measures of safety performance mnst... describe where and when to expect trouble and must provide guidelines concerning what we should do about the problem [p. 49]. 

In Chapter 24, Measurement of Safety Performance, it is said that precise and certain measures of safety, health, and enviromnental performance are difficult to obtain. Nevertheless, audits can be highly effective measures of the quality of hazards management in place if they are well conceived and well done. I suggest keeping in mind the observation made by Kase and Wiese concerning the purpose of a safety audit ... 

Measures of safety performance are now an often-explored subject for which there have been some interesting developments. In this chapter, comments will be made on... 

At the Professional Development Conference held by the American Society of Safety Engineers in June 2002, there were at least three sessions pertaining to leading indicators. Speakers and writers say that this new form of measurement is offered as a response to the dissatisfaction some safety practitioners have expressed concerning the use of what they say are trailing indicators — OSHA statistics, costs — as the sole measure of safety performance. 

At ORC, a study group with representation from 55 companies has been working on a publication pertaining to measurement systems. This is important work. Not since 1980 has a major, broad-ranged publication been issued pertaining to the measurement of safety performance. And the 1980 publication is out of print. Its title is The Measurement Of Safety Performance. William F. Tarrants was the editor. In 1995,/or its member companies only, the Chemical Manufacturers Association issued Program Performance Measures, about which more will be said later. 

The National Safety Council library found two relative references in its computer bank a book previously referenced here titled The Measurement of Safety Performance, for which William E. Tarrants was the editor as well as the author of some of its chapters and a paper published in August 1982 titled One Method for Evaluating Safety Performance in Working Places by Shigao Hanayas. 

Only one other book, besides the Tarrants text, was located that treats safety performance measurement in any depth. There is a chapter titled Measurement Tools for Management in Loss Control Management by Bird and Loftus. Several texts were found that give the measurement of safety performance cursory treatment. 

If major endeavors through symposia or any other type of fomm to disseminate knowledge about the measurement of safety performance have been held since 1970, their proceedings remain secluded. 

Measurement is a prerequisite for control and prediction. The main function of a measure of safety performance is to reveal the level of safety effectiveness. A second purpose is to provide continuous information concerning the safety state. Measures of safety performance must help prevent, not just record, incidents. They must indicate where hazards-ielated incidents will likely occur and provide guidelines concerning the appropriate preventive initiatives. 

Tarrants, W.E. 1980. The Measurement of Safety Performance. New York Garland STPM Press. U.S. Bureau of Labor Statistics. 2010. http //www.bls.gov/news.release/osh.niO.htm. 

---