OSHA recordable injury and illness rates are widely used to assess the effectiveness of workplace safety programs. Yet many observers say the recent five- and six-figure OSHA citations against Weyerhaeuser and General Motors for recordkeeping fraud may just be the tip of an under-reporting iceberg affecting much of U.S. industry.
Joe Fortuna, M.D., medical director of Delphi Corp., a manufacturer of mobile electronics headquartered in Troy, Mich., thinks there's a growing awareness that many companies aren't recording injuries and illnesses accurately. "I hear people saying, 'We have such a low OSHA recordable rate, but our workers' compensation costs keep going up. What's wrong with this picture?'"
Incentives for Dishonesty?
Bob Whitmore, the chief of OSHA's division of recordkeeping requirements, has been talking to people about OSHA recordkeeping (29 CFR 1904) since 1985. "When I ask people, 'Who do you keep OSHA records for?' invariably they say, 'First, for the employer, then for the employees, and third for OSHA.'" Whitmore continues, "I tell them they're wrong the records are kept for us. Period."
Whitmore explains the records are a national surveillance tool, intended to help OSHA know what's happening in U.S. workplaces so the agency can direct resources to the places that need help.
For example, OSHA's site-specific targeting program (SST) uses the data to target inspections at facilities with high injury and illness rates. While SST appears to be a logical way to make the most of OSHA's limited enforcement resources, it also means companies have an incentive to keep their recorded injuries and illnesses as low as possible.
Tying promotion or material incentives to OSHA recordable rates produces a second temptation not to record injuries, according to one consultant. "The use of OSHA data to reward anyone at any level of management or wage employees is just wrong," Thomas Durbin asserts. The former corporate director of safety and workers' compensation at PPG Industries, Durbin is now a coordinator of ORC Worldwide's safety and health consulting service. Besides giving workers and managers an incentive to hide injuries, Durbin says "the data are so variable they are an inaccurate, and therefore unfair, outcome metric." (See figure 1).
Durbin points out that accurate logs involve a two-step process. First, the injured or ill worker must report the problem to the supervisor. Next, the person in charge of the OSHA log must properly record the reported injury or illness or illness. Failures at either point mean inaccurate records.
The Bloody Pocket
This points to a third cause of under-recording: Many injuries and illnesses are never reported in the first place.
"What I think is happening across American industry is a lot of injuries are simply going unreported," says Mike Wright, director of health, safety and the environment for United Steelworkers of America.
Wright calls it the "the bloody pocket syndrome." Because of "no-fault injury and illness policies," many workers have learned they will be disciplined for injuries no matter who's at fault. "So people just suck it up, go home and patch it up."
Other causes of under-reporting are incentive programs that reward entire work teams or plants for no injuries; the result is workers won't report injuries because of peer pressure.
The very nature of the under-reporting problem makes it nearly impossible to determine how widespread it is. But Wright says, "I think the problem is growing. We run into these things every day and it afflicts even our best programs."
Finally, many recording mistakes are inadvertent. "Recordkeeping rules are more complicated than people think," says Pat Tyson, a partner in the OSHA practice of Constangy, Brooks & Smith, LLC, an Atlanta law firm that performs audits of OSHA logs at thousands of facilities. "It is rare for us to find people doing it correctly."
Many companies don't understand all the nuances of recordkeeping rules, according to experts, and haven't devoted the resources needed to comply.
Purpose of the OSHA Log
If Whitmore is right and the OSHA logs are really intended for OSHA, one might ask: Why is it so important for a company to have accurate records?
One obvious answer is the law requires it. But there are valid safety reasons as well. Steve Newell, senior consultant in ORC Worldwide's Washington, D.C., office, believes the data are the foundation of the entire safety program. "If you don't know what's going on, you can't abate the hazards," he says. "And if people believe your logs aren't accurate, just a numbers game, it can undermine your credibility on the shop floor."
Consultant Linda Ballis, owner of Toledo, Ohio-based Linda Ballis and Associates, offers another reason: Accurate records can help companies identify workplace hazards and are useful in comparing performance with other organizations performing similar operations. "For example, you might find you have more ergonomic issues than they do, and then you can ask what you are doing differently."
Eliminating cash incentives for low OSHA recordable rates is one obvious way to improve the accuracy of the logs. As Fortuna puts it, "People will go where they're 'incented' to go."
The problem isn't necessarily one of blatant fraud, according to Durbin. "I've heard safety professionals say they deal with OSHA recordkeeping the same way they deal with income taxes. They want to take advantage of every loophole they can that's legal."
Durbin laments that the time spent looking for loopholes in the complex regulations saps energy from what should be the primary focus of safety mangers: hazard abatement. He recommends companies consider doing what PPG did: Develop a separate recording system for internal use only (see sidebar).
"It's poison," agrees Newell when asked about tying incentives to OSHA recordable rates. Yet, he points to a fundamental challenge: Senior management has long believed in linking pay to performance, and the practice seems to be working everywhere it is used.
"How do you explain to them it's the wrong thing to do for safety and health?" asks Newell.
One way to deal with the problem recommended by all the experts we spoke to: Link incentives to "leading" indicators, rather than "trailing" ones such as OSHA rates. Examples include:
- Investigations of all incidents and near misses;
- Hazard identification and remediation;
- Safety committee meetings with employee representation.
Tyson contends that any company using OSHA recordable rates as a sole indicator of safety performance should have an aggressive third-party audit program. Even without the incentive issue, audits are a good idea because it is easy to make honest recordkeeping mistakes.
Based on his many years of enforcing OSHA's recordkeeping rule, Whitmore says audits alone aren't enough. Companies must hold people accountable.
He recalls the case of a company OSHA cited some years ago for recordkeeping violations. "The man they put in charge of their audits said, 'I have a very simple process: You pass my audit, you keep your job. Now we don't have a problem with inaccurate records.'"
The GP Example
In the midst of what appears to be a growing problem of inaccurate recordkeeping, OSHA recently cited one U.S. company for doing a superb job with injury and illness reporting.
The agency selected Atlanta-based Georgia Pacific Corp. (GP) to be the first participant in the corporate pilot part of OSHA's elite Voluntary Protection Program (VPP). The idea of the corporate pilot program is that OSHA will simplify the complex VPP application for the individual sites of larger companies that have impressed the agency with their dedication to safety and health management.
According to June Brothers, GP's vice president for corporate safety, her company's injury and illness reporting system "was one of three areas of excellence" OSHA identified when it chose the tissue, packaging and building products manufacturer for the pilot program.
Brothers identified three key elements that distinguish GP's approach to recordkeeping:
- Training people at the facility level, on an ongoing basis, on what is and is not recordable;
- Having a resource for people to call if they have questions about what is recordable;
- Audit for correctness at least every 3 years.
Brothers said GP does not use outside auditors, but rather members of the corporate staff, all of whom are certified safety professionals or industrial hygienists.
Finally, GP has no incentives for low OSHA recordable rates. "We do allow recognition for safety performance. It might be a shirt or a barbecue, and it might be for activity measures," Brothers explains, "but there is no reward program for safety outcomes."
Whitmore says OSHA will soon be rolling out a variety of guidance and training tools to help employers understand how to comply with the recordkeeping rule. The guidance will include posting frequently asked questions about the recordkeeping rule on OSHA's Web site.
Will guidance without stricter enforcement be enough to address the problem? Ballis and other recordkeeping experts point out that the demand by companies for audits of OSHA logs has declined in recent years. "When we got the new recordkeeping rule [in 2002], the government went soft the first 2 years," Ballis explains. "But this year, with Weyerhaeuser and GM, the government is a lot more involved, companies are finding their logs aren't as accurate as they thought and now there is an increase in demand for outside audits."
Former OSHA Administrator John Henshaw has staked the agency's reputation on its ability to reduce "the triple bottom line" of injuries, illnesses and fatalities. While it seems reasonable to judge the agency on performance measures, some wonder if through tying its own performance to safety outcomes, OSHA has fallen into the same "incentive trap" afflicting many U.S. companies.
Tyson asks whether it is in OSHA's interest to find that many companies are keeping the records incorrectly. "OSHA's report card is based on that, too," he explains. "The only solution to these things is for OSHA to have an aggressive enforcement program."
In a recent interview, Henshaw was asked if OSHA plans to enforce the recordkeeping standard more aggressively in the near future. He replied that OSHA would follow a "balanced approach" that emphasizes voluntary methods such as outreach, education and new Web site tools. Henshaw gave no indication he favors stepped-up enforcement of recordkeeping, saying only "there will be a component of enforcement as there has been over the years."
While recorded injuries and illnesses fall year after year, in 2003 total fatalities were up from 2002. Fatalities are harder to cover up, leading some observers to question whether the other two parts of Henshaw's triple bottom line are truly declining or just not being recorded.
At the end of the day, Whitmore believes people will only take the recordkeeping program seriously if OSHA takes it seriously.
"That means we have to hold people accountable and do fair, consistent enforcement. If we're going to have a triple bottom line, I'd like to know that two-thirds of it is real."
Sidebar: Keeping a Second Set of Books?
It may sound shady, but Thomas Durbin, the former corporate director of safety and workers' compensation at PPG Industries, thinks keeping a second set of injury and illness records besides the OSHA logs can be a good idea.
When he was at PPG, a Pittsburgh-based manufacturer of glass and chemical products, Durbin developed a protocol for internal company use he believes more accurately reflects the performance of the company's safety program. The company is still using Durbin's protocol.
"For a performance metric to be valid, you need to feel the data are 80 percent accurate," says Durbin.
Because the OSHA log is intended for OSHA, companies may want to develop their own ways of measuring safety performance. For example, the OSHA log requires companies to record many injuries and illnesses that may not be under a safety manager's control, or that may be too insignificant to worry about.
In developing the internal protocol, Durbin identified four criteria to ensure meaningful and reliable data.
To be recordable, injuries and illnesses had to be:
- Difficult to manipulate;
- Significant to the organization;
- Precise and accurate;
- Able to be benchmarked.
"We took these criteria and went to types of cases we see in our experience at PPG," says Durbin. "The managers liked the new system because it helped us to focus on hazards that are important and that are true costs to the organization."