On Jan. 1, 2002, OSHA's new regulations for recording occupational injuries and illnesses went into effect. The new regulations are the welcome culmination of a long and difficult rulemaking process. As the preamble shows, OSHA's professional staff tackled head-on many nagging and difficult problems with the old regulations, which, in the words of one OSHA official, had "more land mines per square inch than any regulation I've ever seen."

OSHA succeeded in fixing the clear majority of those problems and, as a result, the new regulations overall are much clearer, better written and much easier to apply than the old rules. The new forms, for example, are especially an improvement over the old ones.

No human endeavor is perfect, however, and some important problems with the regulations remain. This article discusses a few of those problems.

When Is Administration of Oxygen not Recordable?

OSHA's position under the old regulation was that the administration of oxygen is first aid. When OSHA proposed the new regulations, it proposed that the "[s]imple administration of oxygen" continue to be first aid, but it asked for comments on whether "more severe procedures" (such as intermittent positive pressure breathing) should be considered medical treatment. Some comments filed during the rulemaking urged OSHA to consider oxygen as medical treatment. The American Red Cross commented, for example, that "no administration of oxygen is 'simple.'"

Surprisingly, in the final rule's preamble, OSHA stated that it had changed its mind. The new regulations omit administration of oxygen entirely from the exclusive list of first aid in 1904.7(b)(5)(ii), thus rendering it "medical treatment." OSHA explained that it was convinced that "oxygen administration is a treatment that can only be provided by trained medical personnel, uses relatively complex technology and is used to treat serious injuries and illnesses." Since the publication of the new regulations, OSHA has received more complaints about the new oxygen rule than any other new "medical treatment" provision.

The problem with the new oxygen rule is that its premises are sometimes contrary to fact: Although oxygen is often used to treat serious injuries and illnesses, not all administrations of oxygen are in response to serious injuries or illnesses, and some are not treatments at all. Nearly every reader of Occupational Hazards can tell of plant nurses who administer oxygen to patient-employees not as a treatment, but simply to put them at ease by making them feel that they are being treated. Indeed, some ambulance protocols require the administration of oxygen (whether medically necessary or not) for anyone who is transported by ambulance, even for a sprained ankle. (An OSHA official intimately familiar with the new regulations told us that the case involving the ambulance would be recordable because an overtreatment, even if completely unnecessary, is still a treatment.)

The National Association of Manufacturers (NAM) challenged this aspect of the new rules in court, but withdrew the challenge in return for the following passage, which OSHA published Dec. 27 in the Federal Register (66 Fed. Reg. 66943-66944) and incorporated into Compliance Directive CPL 2-0.131 (Jan. 1): "If an employee is exposed to chlorine or some other substance at work and oxygen is administered as a purely precautionary measure, is the case recordable? Answer: If oxygen is administered as a purely precautionary measure to an employee who does not exhibit any symptoms of an injury or illness, the case is not recordable. If an employee exposed to a substance at work exhibits symptoms of an injury or illness, the administration of oxygen makes the case recordable."

A problem with this language is that, even though plant medical personnel may administer oxygen in response to symptoms, they do not always administer it as a treatment for those symptoms. As noted above, sometimes oxygen is administered in response to symptoms but without a judgment that it will treat an injury or illness.

Another problem is that employers may misread this language to exempt all purely precautionary uses of oxygen. It exempts such use only if the employee "does not exhibit any symptoms of an injury or illness," and "injury or illness" is broadly defined in new section 1904.46 as merely any "abnormal condition or disorder."

That OSHA does not see the settlement language as significantly narrowing the oxygen rule is indicated by its view of the following hypothetical case: A firefighter trains vigorously under a hot sun. He becomes hot and tired, and his breathing becomes somewhat labored. His symptoms fall short of a heat stroke, but oxygen is administered anyway to ease his discomfort and as a precaution to prevent the development of heat stroke. OSHA sees the case as recordable for, although the oxygen was not administered to treat a medically diagnosed illness, it was administered in response to signs or symptoms of an "abnormal condition," which the regulations define as constituting an "illness."

The result of OSHA's new oxygen rule is already becoming clear. Plant nurses and company physicians who are aware that the administration of oxygen will create a recordable case will no longer administer oxygen as a matter of course. Instead, they will wait for affirmative evidence of a medical need to use oxygen as a treatment. Is this really best for employees?

What can an employer do if a plant nurse administers oxygen not as a treatment but, for example, merely to make employees think that they are being treated and, thereby, to ease their anxiety? Must the employer record?

First, the employer might argue that the case falls within the interpretation in the compliance directive (i.e., that the oxygen was given purely as a precaution to employees who have not yet shown symptoms of an illness or injuries). Second, the employer can argue that a particular administration of oxygen does not fall within the definition of "medical treatment" in 1904.7(b)(5)(i) (i.e., "the management and care of a patient to combat disease or disorder"). The employer could argue that the oxygen was administered not "to" combat disease or disorder but, for example, to ease an employee's anxiety. Although the argument may strike an OSHA area director as unfamiliar, it might well, depending on the facts, convince a judge.

Work-Relatedness: the Case of the Collapsed High Heel

Suppose an employee goes dancing one Saturday night in her high-heel shoes. The dancing is so vigorous that it weakens the high heel on one of her shoes to the point that it could collapse at any moment. As she prepares for work on Monday morning, the employee decides to wear the same shoes she wore on Saturday night. While driving to work, she sees the local mall and considers stopping to quickly buy a few items. Ultimately, she decides against the detour, however, and drives to work. She parks her car in the company lot, which, like the lot in the mall, is perfectly paved and level. After walking a few steps in the lot, her shoe's heel collapses and she breaks her ankle.

A materials engineer testifies that the dance-weakened heel would have collapsed at the mall parking lot if the employee had gone there instead. Is her broken ankle work-related? (Remember that OSHA has eliminated the old parking lot exception and replaced it with a rule for automobile accidents in parking lots while commuting.)

Let's look at new 1904.5, titled "Determination of work-relatedness." The first sentence of paragraph (a), titled "Basic requirement," states: "You must consider an injury or illness to be work-related if an event or exposure in the work environment either caused or contributed to the resulting condition or significantly aggravated a pre-existing injury or illness."

Read literally, one might argue that the case is work-related, for there was an "event ... in the work environment" (the collapse of the heel in the company's parking lot) that caused or contributed to the broken ankle, even though the cause of the event had nothing to do with work. Even OSHA realizes that this is too broad, for not everything that happens at work is caused by work. So OSHA provided a second sentence that qualified and clarified the first sentence.

The second sentence states: "Work-relatedness is presumed for injuries and illnesses resulting from events or exposures occurring in the work environment, unless an exception in 1904.5(b)(2) specifically applies" (italics added). Let us take this sentence one piece at a time. Inasmuch as we have an "event" "occurring in the work environment," work-relatedness is "presumed" unless an exception applies. OSHA would likely say that none of the nine exceptions in 1904.5(b)(2) from the presumption of work-relatedness apply. For example, there is an exception in 1904.5(b)(2)(ii) if an "injury or illness involves signs or symptoms that surface at work but result solely from a nonwork-related event or exposure that occurs outside the work environment." OSHA would likely say that this exception is inapplicable because the injury's symptoms did not merely "surface" at work and because the event causing them did not occur outside the work environment. So this leaves the "presumption" of work-relatedness standing. Presumptions are still rebuttable, however, and the facts above, particularly the testimony of the materials engineer, flatly rebut it here. So the case is not recordable, right?

Not according to OSHA. The authors recently posed the above hypothetical case to two OSHA officials intimately familiar with the new regulations. They stated that, inasmuch as none of the nine exceptions apply, the above case is work-related.

Not so fast, we urged. That no exception applies means only that the presumption of work-relatedness still stands. That does not mean that the presumption cannot then be rebutted. There is no point in having a presumption if it is not rebuttable by the facts, as it is here. The OSHA officials disagreed, stating that the only way to rebut the presumption is through the exceptions, which is stated nowhere in the regulations.

OSHA's position causes the regulation to generate irrational results for, as the above hypothetical shows, there are cases that do not fall within any of the nine exceptions but are still not work-related. The NAM settlement, which requires that the workplace be a "discernable" cause of an injury, is consistent with this observation. More importantly, OSHA's reading eliminates the word "presumed" from the regulation. One OSHA official to whom we spoke stated that, as he reads the provision, the second sentence of 1904.5(b) might as well state, "Injuries and illnesses resulting from events or exposures occurring in the work environment are work-related unless an exception in 1904.5(b)(2) specifically applies."

Indeed, OSHA recently posted on its Web site a question-and-answer pair that shows that OSHA does not read the exception in 1904.5(b)(2)(ii) literally or sees the presumption arising under 1904.5(a) as rebuttable. Question 5-8 of OSHA's Frequently Asked Questions (www.osha.gov/recordkeeping/entryfaq.html) asks whether a broken arm suffered in a fall resulting from an epileptic seizure is work-related. OSHA answers that it is not because it falls within the above-quoted exception in 1904.5(b)(2)(ii). The passage reads as follows:

"Question 5-8. If an employee's pre-existing medical condition causes an incident that results in a subsequent injury, is the case work-related? For example, if an employee suffers an epileptic seizure, falls and breaks his arm, is the case covered by the exception in section 1904.5(b)(2)(ii)?

"Neither the seizures nor the broken arm are recordable. Injuries and illnesses that result solely from nonwork-related events or exposures are not recordable under the exception in section 1904.5(b)(2)(ii). Epileptic seizures are a symptom of a disease of nonoccupational origin, and the fact that they occur at work does not make them work-related. Because epileptic seizures are not work-related, injuries resulting solely from the seizures, such as the broken arm in the case in question, are not recordable."

Under the old regulations, OSHA took the opposite position and held that, while the unconsciousness was not work-related, the broken arm was.

What is fascinating about OSHA's new answer is that it cannot be squared with the actual words of 1904.5(b)(2)(ii). That provision makes nonoccupational an "injury or illness [that] involves signs or symptoms that surface at work but result solely from a nonwork-related event or exposure that occurs outside the work environment."

Let us suppose that the broken arm were the "injury" involving signs or symptoms that "surfaced" at work. Even though it "result[ed] solely from a nonwork-related event" (the fall or the seizure), that event did not "occur[] outside the work environment," which means that OSHA's interpretation is contrary to the words of 1904.5(b)(2)(ii).

Now suppose instead that the "event" is whatever nonoccupational event caused the epilepsy. This raises an even greater problem for OSHA: It means that one may look behind the immediate cause of an event and determine whether its underlying cause was work-related, which is logically equivalent to asking whether the presumption in 1904.5(a) is rebuttable.

Let's go back to the broken ankle resulting from the weakened high heel: Would not OSHA's reasoning in the case of the epileptic seizure mean that this case too is not recordable?

Must a "Significant" Aggravation Be Significant?

Under the old regulations, any work-related aggravation of a nonwork-related injury or illness could make a case recordable. So if Joe Goodfella, for example, played tennis on Sunday, got a sore elbow, came to work, used the elbow only slightly but aggravated it enough to require medical treatment or a work restriction, the case was recordable even if the employee's work-related movements alone could not have caused a sore elbow.

Employers complained mightily about this rule and, in the preamble to the new rule, OSHA stated that it had changed it. The preamble states that the new regulations require that "work 'significantly' aggravate a pre-existing injury or illness before the case is recordable." It promises that "minor aggravation of pre-existing conditions" need not be recorded. It states that the definition requires that "the contribution of the aggravation at work is such that it results in tangible consequences that go beyond those that the worker would have experienced as a result of the pre-existing injury or illness alone, absent the aggravating effects of the workplace." Clearly, OSHA's rulemaking staff believes in good faith that the new aggravation provision has worked a change in the old regulation.

The problem is that a lawyer representing OSHA could argue that the new aggravation provision changes nothing. The new provision is in 1904.5(b)(4), which reads in part:

"A pre-existing injury or illness has been significantly aggravated ... when an event or exposure in the work environment results in any of the following: ...

"(iii) One or more days away from work, or days of restricted work, or days of job transfer that otherwise would not have occurred but for the occupational event or exposure;

"(iv) Medical treatment in a case where no medical treatment was needed for the injury or illness before the workplace event or exposure, or a change in medical treatment was necessitated by the workplace event or exposure."

Let's take Joe Goodfella with tennis elbow again and see how his case fares under this new language. Although his elbow is sore, it is just shy of being so sore as to require a work restriction. Then, at work, Joe merely picks up a hand drill. Although this increases his soreness ever so slightly, the difference sends his elbow over the edge and he requires a work restriction. Now look at 1904.5(b)(4)(iii). Inasmuch as the restriction "would not have occurred but for the occupational event or exposure" (i.e., his lifting of the hand drill), the aggravation is deemed "significant" and, therefore, is recordable. In other words, nothing seems to have changed.

Other persons to whom we have spoken have been unable to construct a case of "minor" aggravation that would have been recordable under the old regulations but would not be recordable under the above definition. And OSHA has already received complaints from industry that the new rule changes nothing.

Hearing Shifts and Exposure to More Than 85 dBA

Section 1904.10(b)(5) states a special rule for determining the work-relatedness of hearing loss: a loss of hearing acuity (of at least 25 dB for at least the year 2002; 10 dB in some states) "is presumed to be work-related if the employee is exposed to noise in the workplace at an eight-hour, time-weighted average of 85 dBA or greater, or to a total noise dose of 50 percent, as defined in 29 CFR 1910.95" (the noise standard).

Suppose an employee suffers a hearing loss of 25 dB. The employer investigates the noise levels to which the employee had been exposed and finds that it was usually 80 dBA or below, except that once or twice a year, during a clean-out operation, the noise rose to 90 dBA for a day. No physician would say that such limited exposures could have caused the hearing loss. Does this case nevertheless meet the special test for work-relatedness? How frequent must exposure to 85 dBA be? Will only a single exposure over 85 dBA per year suffice?

The new regulations do not answer the question. This is surprising, for employers often ask the question, and even OSHA's field enforcement personnel have asked it. For example, a regional administrator in 1994 told OSHA's national office that he "needs specific direction involving the requirement for audiometric testing (i.e. after what length of exposure: One day? 30 days? One year?) are we going to require [audiometric testing] of an employer)?"

Evidently, OSHA has not come up with an answer. It probably would agree that the above case is not recordable and that, while even one day's exposure per month to noise over 85 dBA is not enough to make a hearing loss recordable, more than one day per week might be enough. Nevertheless, employers still have no clear rule to follow. OSHA's failure to answer this question is annoying and likely illegal. OSHA should stop dodging it and provide some guidance to employers.

About the authors: Robert C. Gombar is a partner in the Washington office of McDermott, Will & Emery and head of the firm's OSHA practice. Arthur G. Sapper is a partner in the OSHA practice group. They are contributing editors to Occupational Hazards.