Last September, ISHN printed a survey on this page asking you about your opinions and experiences regarding behavior-based safety. It's part of Virginia Tech's two-year research grant from the National Institute for Occupational Safety and Health to study the critical success factors for behavior-based safety. Before the end of 1996, 162 completed surveys were returned to us by mail or fax.

An overall respect for behavioral safety was shown by 80% of the respondents answering "yes" to the question: "Do you believe behavior-based safety is a viable approach for reducing at-risk work behaviors and activities?" Plus, more respondents answered "no" (48%) than "yes" (34%) to the question of whether a safety program should focus more on attitudes than behaviors. This shows a shift from the traditional attitude-oriented approach to safety education.

But our survey was not random, however, and was likely biased toward the behavioral approach. Most of you who took the time to answer the questions are probably more knowledgeable and confident about behavioral strategies than the average reader of ISHN. Still, answers revealed some misperceptions about behavioral safety that can limit its potential for safety improvement. For example, the three descriptions selected most often by respondents to describe behavior-based safety were: "An intervention approach for increasing safe behavior" (selected by 88%); "An observation and feedback process" (80%); and "A tool for managing safety" (70%).

But only 26% consider behavior-based safety useful for investigating injuries. 54% feel it's useful for evaluating safety achievement, and 61% see it as a way of decreasing unsafe behavior.

I suspect many people have a rather narrow viewpoint about behavior-based safety. It's actually "a general philosophy that can be applied to many aspects of safety management"-the most accurate definition listed in our survey, and one selected by 71% of respondents.

Many sources, including my recent book, The Psychology of Safety, explain how behavior-based safety can be applied effectively to injury investigation, education and training, evaluation, ergonomics, and the design of incentive/reward programs. Three basic principles are worth repeating:

Target observable behavior

Organizational and community research has shown over and over that it's easier and more cost-effective to effect large-scale change by focusing on behaviors rather than attitudes. Attitude and other subjective feeling states are changed indirectly when behavior-based techniques are applied appropriately.

Focus on positive consequences

Behavior-based interventions manipulate external or environmental factors to motivate change. Events that occur before behavior-such as signs, memos, demonstrations, and group discussions-can direct our actions, but we actually perform the behavior for the consequences we expect to receive. In other words, antecedent events tell us what to do in order to receive a motivating consequence.

Interestingly, antecedent strategies turn out to be the most popular technique used by survey respondents, with policies, posted safety signs, demonstrations, and lectures leading the list. Behavior-oriented goal-setting and feedback for individuals and groups are the most cost-effective interventions, yet they are currently used at less than half of the sites represented by responses. The most common goals, feedback, and incentive/reward programs are used to focus on outcome ("accidents or injuries") rather than process ("safety-related behaviors or activities").

Monitor behavior trends

In an observation and feedback process, safety coaches work with a checklist to record safe and at-risk behaviors. The completed summary is used to give feedback to the person or group observed. Data can be compiled across individuals or groups per day or week with a "percent safe behavior" metric. This diagnostic measure shows how a particular intervention program influences different safety-related behaviors, and it can illustrate fluctuations in progress across weeks, months, or years. Monitoring trends tells you when to focus on different behaviors, or when to change an intervention approach or protocol. This is the key to continuous improvement. Traditional safety outcome measures, such as OSHA recordables and lost-time injuries, are reactive and have practically no diagnostic value. Since many factors influence the reporting (or non-reporting) of injuries, changes in these numbers provide no useful information. This is certainly not a revolutionary point for ISHN readers. But while readers know current safety measurements are insufficient, they are still the most popular.

Only 15% of respondents monitor "percent safe behavior" to assess the success of their safety programs. Traditional outcome metrics are most popular, with 77% using OSHA recordables, 75% using lost-time accidents, 42% using total recordable injury rates, and 44% using total recordable rates, including illnesses.

I think the gap between understanding and applying some of these behavior-based principles has more to do with management systems than personal inconsistencies. It shows us that shifting from an outcome-based, reactive safety process to one focused on critical up-stream activities is easier said than done.