Copyright © 2004 Business Insurance

"Metrics Improve Safety Through Better Statistics June 14, 2004

by MEG FLETCHER

Increasing numbers of safety managers are striving to develop statistically-based systems to help prevent injuries and illnesses, rather than taking a reactive approach to safety analysis.

In their work to assess their safety efforts, safety managers are putting more emphasis on measuring and reporting "leading" indicators, such as compliance with machine guard use requirements because statistics show that fewer workers are injured when such guards are in place.

The change in emphasis from historical analysis of loss data to proactive analysis of the effectiveness of safety measures is significant, experts say.

Previously, safety managers focused more on "lagging" indicators-such as statistical reports on the number of worker injuries or their lost workdays-which reflect the reporting requirements of the Occupational Safety and Health Administration, according to Christopher A. Janicak. He is an associate professor of safety at Indiana University of Pennsylvania in Indiana, Pa.

"The historical health and safety measurement mindset is one of tracking failure, showing loss avoidance and not a positive contribution to business," said Paul Esposito, president of STAR Consultants Inc., which is based in Annapolis, Md.

The new trend in prevention may require the use of a safety metrics program, which consists of quantifying and analyzing data to determine if an organization's safety goals are being met, said Mr. Janicak, who has written a book on the subject. For example, a safety manager asked to determine the appropriate height to install an emergency shut-off switch might use stndard tables of workers' average reaching height and then factor in normal distributions and standard deviations to determine a height that 98% of all workers could be expected to reach, he said.

The goals of an individual company's safety program determine the type of safety data that it needs to collect and analyze, said Lon Ferguson, chairperson of the Safety Sciences Department at Indiana University of Pennsylvania. Companies should not use "a cookie-cutter approach," he said.

Detroit Edison Co.'s Fermi 2 nuclear plant uses safety metrics to track both leading and lagging indicators in its effort "to drive continuous improvement in our safety performance by establishing `stretch goals' targeted for best-in-class" operations, said David A. Varwig, principal technical specialist-safety, assessment and planning at the Newport, Mich., plant. The facility generates electrical power for southeastern Michigan.

The utility periodically tracks, reports and evaluates 47 measures, including 26 that relate to safety, by comparing its performance against benchmarks, including ones established by the industry-rated top 10% and top 25% of similar nuclear power plants.

"Personnel safety measures include tracking a worker's exposure to radiation. They also include tracking human performance errors and incidents," Mr. Varwig added. Other safety measures track the operation of the nuclear reactor as well as public safety measures, which include emergency plans and monitoring of effluents.

The Fermi 2 reactor plant uses a mix of reporting formats, including scorecards and trend graphs, he said.

Using safety metrics has had a positive effect on operations, because "we have advanced beyond the `what gets measured, gets done' slogan to formalizing expectations that facilitate involvement and promote ownership, acceptance and accountability for our safety results at all levels of the organization," Mr. Varwig said.

Everyone at the plant understands his or her safety roles and how he or she can contibute to plant safety, he said.

The negative aspect of using such metrics, however, is "that the more you learn about your operations and further specify your safety metrics, the more you increase the possibilities of data/information overload...increasing the potential for `paralysis of over-analysis,"' he said.

Among other safety metric measures that some employers use is a scorecard that Mr. Esposito said he is helping a federal agency develop to rate supervisors. It will evaluate them on their use of the agency's inspection, hazard analysis and accident investigation programs.

Other companies are taking simpler approaches than analyzing comprehensive injury statistics. But even simpler means of leveraging information can help improve workplace safety efforts.

For example, safety professionals working for the San Francisco Public Utilities Commission's $3.6 billion capital improvement program participate in daily "toolbox" meetings, during which project managers discuss the work to be done that day, including safety issues.

Safety professionals also visit the worksite during the day to make sure that workers are performing appropriately, said Marge P. Layne, manager of the owner-controlled insurance program for the project, which seeks to improve delivery of water to the area.

"It's simply a day-to-day process and there is no follow-up record keeping" unless an accident occurs, Ms. Layne said. Any accident would be investigated and a report would be made, primarily to educate safety personnel and workers about hazards to help prevent future injuries and illnesses, she said.

While interest in injury analysis is growing, safety consultant Eddie Greer estimates that "only about 30%" of safety professionals routinely emphasize collection of data and analysis of leading injury indicators.

Mr. Greer, who operates his own safety consulting firm in Sugar Land, Texas, is a former president of the Des Plaines, Ill.-based American Society of Safety Engineers.

One reason safety managrs and companies have not adopted the approach is that "they know that it will take an investment of time," though establishing such a system would ultimately save a safety manager time in the long run, Mr. Esposito said.

 

© Copyright Business Insurance 2004