[Federal Register Volume 67, Number 126 (Monday, July 1, 2002)]
[Rules and Regulations]
[Pages 44037-44048]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 02-16392]


=======================================================================
-----------------------------------------------------------------------

DEPARTMENT OF LABOR

Occupational Safety and Health Administration

29 CFR Part 1904

[Docket No. R-02A]
RIN 1218-AC06


Occupational Injury and Illness Recording and Reporting 
Requirements

AGENCY: Occupational Safety and Health Administration (OSHA), 
Department of Labor.

ACTION: Final rule.

-----------------------------------------------------------------------

SUMMARY: The Occupational Safety and Health Administration (OSHA) is 
revising the hearing loss recording provisions of the Occupational 
Injury and Illness Recording and Reporting Requirements rule published 
January 19, 2001 (66 FR 5916-6135), scheduled to take effect on January 
1, 2003 (66 FR 52031-52034). This final rule revises the criteria for 
recording hearing loss cases in several ways, including requiring the 
recording of Standard Threshold Shifts (10 dB shifts in hearing acuity) 
that have resulted in a total 25 dB level of hearing above audiometric 
zero, averaged over the frequencies at 2000, 3000, and 4000 Hz, 
beginning in year 2003.

EFFECTIVE DATE: January 1, 2003.

FOR FURTHER INFORMATION CONTACT: Jim Maddux, Occupational Safety and 
Health Administration, U.S. Department of Labor, Directorate of Safety 
Standards Programs, Room N-3609, 200 Constitution Avenue, NW., 
Washington, DC 20210. Telephone (202) 693-2222.

SUPPLEMENTARY INFORMATION:   

I. Background

    In January, 2001 (66 FR 5916-6135), OSHA published revisions to its 
rule on recording and reporting occupational injuries and illnesses (29 
CFR parts 1904 and 1952) to take effect on January 1, 2002. On July 3, 
2001, the agency proposed to delay the effective date of Secs. 1904.10 
Recording criteria for cases involving occupational hearing loss, and 
1904.12 Recording criteria for cases involving work-related 
musculoskeletal disorders, until January 1, 2003 (66 FR 35113-35115). 
In that notice, OSHA explained that the Agency was reconsidering the 
requirement in Sec. 1904.10 to record all cases involving an 
occupational hearing loss averaging 10 decibels (dB) or more. OSHA 
found that there were reasons to question the appropriateness of 10 dB 
as the recording criterion, and asked for comment on other approaches 
and criteria, including recording losses averaging 15, 20 or 25 dB. 
OSHA also stated that it was reconsidering the requirement in 
Sec. 1904.12 that employers check the MSD column on the OSHA Log for a 
case involving a ``musculoskeletal disorder'' as defined in that 
section.
    OSHA received a total of 77 written comments on the July 3, 2001 
proposal. After considering the views of interested parties, OSHA 
published a final rule on October 12, 2001 (66 FR 52031--52034) 
delaying the effective date of Secs. 1904.10(a) and 1904.12(a) and (b) 
until January 1, 2003, adding a new paragraph (c) to Sec. 1904.10 
establishing a 25-dB recording criterion for hearing loss cases for 
calendar year 2002, and modifying the regulatory note to paragraph 
1904.29(b)(7)(vi) to delay the language referring to privacy case 
consideration for MSD cases.
    This final rule contains amended hearing loss recording criteria 
codified at 29 CFR 1904.10(a) and 1904.10(b)(1)-(7). In a separate 
Federal Register document published today, OSHA is proposing to delay 
the effective date of Sec. 1904.10(b)(7), which requires employers to 
check the hearing loss column on the Log for hearing loss cases

[[Page 44038]]

meeting the revised recording criteria, as well as the MSD provisions 
addressed in the October 12 final rule. Additional information about 
the proposal to delay the effective date of the hearing loss column is 
contained in the section of this rule titled Adding a column to the 300 
Log, and in the separate Federal Register publication Proposed Delay of 
Effective Dates; Request for Comment, published today.

II. Recording Occupational Hearing Loss Cases

    Section 1904.10 of the January 19, 2001 final recordkeeping rule 
required employers to record, by checking the ``hearing loss'' column 
on the OSHA 300 Log, all cases in which an employee's hearing test 
(audiogram) revealed that a Standard Threshold Shift (STS) in hearing 
acuity had occurred. An STS was defined as ``a change in hearing 
threshold, relative to the most recent audiogram for that employee, of 
an average of 10 decibels or more at 2000, 3000 and 4000 Hertz (Hz) in 
one or both ears.'' The recordkeeping rule itself does not require the 
employer to test employee's hearing. However, OSHA's occupational noise 
standard (29 CFR 1910.95) requires employers in general industry to 
conduct periodic audiometric testing of employees when employees' noise 
exposures are equal to, or exceed, an 8-hour time-weighted average of 
85dBA. Under the provisions of Sec. 1910.95, if such testing reveals 
that an employee has sustained a hearing loss equal to an STS, the 
employer must take protective measures, including requiring the use of 
hearing protectors, to prevent further hearing loss. Employers in the 
construction, agriculture, oil and gas drilling and servicing, and 
shipbuilding industries are not covered by Sec. 1910.95, and therefore 
are not required by OSHA to provide hearing tests. If employers in 
these industries voluntarily conduct hearing tests they are required to 
record hearing loss cases meeting the recording criteria set forth in 
the final Section 1904.10 rule.
    The former recordkeeping rule, which remained in effect until 
January 1, 2001, contained no specific threshold for recording hearing 
loss cases. In 1991, OSHA issued an enforcement policy on the criteria 
for recording hearing loss cases, to remain in effect until new 
criteria were established by rulemaking. The 1991 policy stated that 
OSHA would cite employers for failing to record work related shifts in 
hearing of an average of 25 dB or more at 2000, 3000 and 4000 Hz in 
either ear. Subsequently, OSHA released interpretations stating that 
the employer could adjust the audiogram for aging using the tables in 
Appendix F of the Noise Standard, and that the employer was to use the 
employee's original baseline audiogram as the baseline reference 
audiogram for determining a recordable hearing loss.
    One of the major issues in the recordkeeping rulemaking was to 
determine the level of occupational hearing loss that constitutes a 
health condition serious enough to warrant recording. This was 
necessary because the final rule no longer requires recording of minor 
or insignificant health conditions that do not result in one or more of 
the general recording criteria such as medical treatment, restricted 
work, or days away from work (See, e.g., 66 FR 5931). In its 1996 
Federal Register notice OSHA proposed a requirement to record hearing 
loss averaging 15 dB at 2000, 3000 and 4000 Hz in one or both ears (61 
FR 4040). OSHA adopted the lower 10-dB threshold in the final rule 
based in part upon comments that ``(a)n age-corrected STS is a large 
hearing change that can affect communicative competence'' (66 FR 6008).

Comments on the Recording of 10-dB Shifts

    Most commenters opposed the adoption of the 10-dB threshold for 
recording hearing loss (Exs. 3-1, 3-13, 3-14, 3-19, 3-20, 3-22, 3-25, 
3-26, 3-27, 3-29, 3-34, 3-35, 3-37, 3-43, 3-45, 3-48, 3-49, 3-50, 3-54, 
3-57, 3-58, 3-59, 3-61, 3-62, 3-63, 4-3, 4-5, 5-5, 5-7). A number of 
these commenters challenged the significance of a 10-dB shift, stating 
that: 10-dB shifts are not significant--only significant health 
conditions should be captured (Exs. 3-14, 3-26, 3-48); the level 
selected must amount to a significant alteration in an employee's 
ability to hear (Exs. 3-50, 3-54, 3-59); a 10-dB shift from audiometric 
zero is a virtually imperceptible loss in hearing--10-dB shifts at 
higher levels become more important (Ex. 3-49); the medical community 
and workers' compensation do not recognize a 10 dB shift as a 
significant hearing loss (Exs. 3-19, 3-20, 3-25, 3-35, 3-43, 3-63); a 
10-dB shift is not a material impairment, so it should not be a 
recordable illness (Exs. 3-25, 3-26, 3-34, 3-50, 3-54, 3-59, 3-58, 3-
61); and, 10 dB is an early warning mechanism that is appropriate for 
the hearing standard but not for injury and illness recording--the 1904 
provisions are intended to collect data on serious injuries and 
illnesses, not potential precursors (Exs. 3-25, 3-49, 3-50, 3-54, 3-59, 
3-62). Organization Resources Counselors (ORC) remarked that:

    [a] 10 dB shift from audiometric zero is a virtually 
imperceptible loss in hearing * * * ORC understands that the finding 
of a Standard Threshold Shift (STS) to be a ``flag'' for the 
implementation of a series of actions required by the OSHA standard 
on exposure to noise. It was not intended, of and by itself, to be 
an indicator of illness, or impairment, but, rather, a sentinel 
event that triggers a series of actions that will prevent illness or 
impairment from occurring. As such a tool, it has been an effective 
indicator of employee hearing, but does not, by itself, rise to the 
level of recordability (Ex. 3-49).

    A number of the commenters objected to recording 10-dB shifts 
because this recording level would result in the recording of too many 
``false positive'' cases, either because of audiometric testing errors, 
because the hearing loss was temporary and not persistent, or because 
the case was insufficiently work-related (Exs. 3-14, 3-19, 3-20, 3-25, 
3-26, 3-27, 3-29, 3-35, 3-37, 3-43, 3-45, 3-49, 3-50, 3-54, 3-56, 3-58, 
3-59, 3-61, 3-62, 3-63, 4-5). The issues of audiometric error, 
persistence, and work-relationship are discussed in more detail below. 
The commenters opposed to the 10-dB shift also remarked that using 10-
dB shifts will lead to overrecording (Ex. 3-37), 10 dB will result in a 
5 to 10 fold increase in hearing loss recording (Ex. 3-49), too many 
non-occupational (emphasis added) cases are captured by 10 dB (See, 
e.g., Ex. 4-5), changing to 10 dB would make the past data useless and 
make it difficult to establish trends (Ex. 3-19), and that if OSHA 
adopts 10 dB, the states may be influenced to change their workers' 
compensation standards, resulting in higher workers' compensation costs 
(Ex. 3-34).
    Some of the commenters opposed to the recording of all 10-dB shifts 
recognized a critical difference between the 25-dB criteria contained 
in the American Medical Association [AMA] Guides to the Evaluation of 
Permanent Impairment and the 25-dB level OSHA has enforced since 1991 
(Exs. 3-25, 3-49, 3-50, 3-54, 3-59, 3-62). The AMA Guides measure 
hearing loss from a baseline of audiometric zero, which represents the 
statistical average hearing threshold level of young adults with no 
history of aural pathology (ANSI S3.6-1969). The 1991 OSHA recording 
level used the individual employee's original baseline audiogram taken 
at the time the worker was first placed in a hearing conservation 
program. If an individual employee has experienced some hearing loss 
before being hired, a 25-dB shift from the original baseline will be a 
larger hearing loss than the hearing impairment recognized by the AMA 
as a disabling condition. In a single comment submitted by both

[[Page 44039]]

organizations, the National Association of Manufacturers (NAM) and the 
Can Manufacturing Institute (CMI) stated that:

    [i]t is generally accepted in the medical community that an 
average hearing level of more than 25 dB from audiometric zero (the 
hearing level of healthy young adults never exposed to high noise 
levels) at certain frequencies constitutes a material impairment. 
Accordingly, an employee with near-perfect hearing (at or near 
audiometric zero) might very well suffer a 10 or 15 dB shift in 
hearing yet continue to function within the normal range of hearing 
with no impairment whatsoever. Conversely, an employee with hearing 
on the outer edge of the normal range who experiences a 15 dB shift 
would likely suffer a material impairment. The NAM and CMI believe 
that a shift in hearing should not be recorded unless it is 
confirmed and it results in hearing levels in excess of 25 dB at the 
shift frequencies (Ex. 3-50).

    Industrial Health, Inc, a mobile hearing testing vendor, added 
that:

    [i]t is almost universally accepted in the profession that 
hearing impairment starts when hearing levels exceed 25 dB * * *. We 
believe there should be an ``impairment fence'' of 25 which must be 
crossed before a shift in hearing is required to be recorded. We 
recommend that to be recordable a shift must result in an average 
hearing level at 2000, 3000, and 4000 Hz in excess of 25 dB. This 
fence would not be adjusted for aging (however, the shift 
calculation itself should retain OSHA's allowance for aging) (Ex. 3-
62).

    A number of commenters urged OSHA to adopt the 10-dB threshold for 
recording occupational hearing loss, consistent with the January 19, 
2001 Federal Register notice (Exs. 3-3, 3-4, 3-10, 3-11, 3-15, 3-17, 3-
18, 3-21, 3-23-1, 3-24, 3-30, 3-36, 3-40, 3-47, 3-52, 3-53, 4-2, 5-2, 
5-3, 5-6). Many of these commenters argued that an age-corrected 10-dB 
shift is a large change in hearing that can affect communication 
ability (Exs. 3-3, 3-21, 3-23-1, 3-53), that a persistent 10-dB shift 
represents a permanent and irreversible loss of hearing acuity (Ex. 3-
21), that a 10-dB shift is a material impairment (Exs. 3-17, 3-23-1, 3-
53), and that real and debilitating hearing loss may not be detected if 
a higher threshold is selected (Ex. 3-3). The remarks of the Coalition 
to Protect Workers Hearing are representative:

    An age-corrected STS represents a significant amount of 
cumulative hearing change from baseline, enough to affect 
communicative competence, safety, and job productivity in the 
workplace. A confirmed, age corrected STS is not a sensitive 
indicator of early hearing damage; rather it reflects a very 
substantial permanent hearing change over time. The appropriate 
sensitive indicator of early hearing damage is a temporary threshold 
shift (TTS), which recovers quickly as the worker is noise free. 
This indicator is currently used in hearing conservation programs. 
(Ex. 2-23-1)

    Commenters also stated that use of a 10-db shift reduces 
recordkeeping and data management burdens for industry (Exs. 3-3, 3-10, 
3-23-1, 3-47, 3-53, 5-2), reduces confusion for industrial managers and 
occupational hearing conservation technicians--``[a] problem that 
occurred with OSHA's 1991 policy'' (Ex. 3-23-1), that current STS rates 
are not sufficiently high to result in an undue or inappropriate number 
of recordable events (Ex. 3-3), that many of the states (Michigan, 
North Carolina, South Carolina, Puerto Rico and Tennessee) require the 
recording of 10-db shifts with little detrimental effect on industry 
(Exs. 3-3, 3-4, 3-24), that a 10-db shift is comparable to other 
permanent injuries that are recorded on the OSHA 300 Form, such as an 
amputated finger (Ex. 3-23-1) or medical removal under the lead 
standard (Ex. 3-47), and that the 10-db shift is better for mobile and 
transient employees because the original baseline may not follow 
employees when they change jobs (Ex. 3-23-1).
    Several of the commenters argued that recording 10-db shifts would 
be more protective for workers (3-3, 3-10, 3-17, 3-18, 3-21, 3-23-1, 3-
24, 3-30, 3-47, 3-53). In a representative comment, the AFL-CIO argued 
that: ``[t]he requirement to record a 10-db hearing loss on the Log 
would aid in the early detection and prevention of occupational hearing 
loss.'' It stated that ``(r)ecording a 10-db STS on Form 300 is a 
practical and reasonable means to assist in the early detection of a 
loss in hearing so that workplace intervention measures can be 
implemented to protect workers from the hazards of noise. Having 
employers continue to record shifts in hearing of an average of 25 dB * 
* * is too high a threshold of loss in hearing acuity to be 
sufficiently proactive in preventing worker hearing loss'' (Ex. 3-24).
    Other commenters added that by recognizing disease earlier, 
employers may take preventive measures to avoid potential workers' 
compensation cases that are sometimes triggered at the 25-dB level (Ex. 
3-10), that recording triggers action on the part of employers (Ex. 3-
23-1), that 10-db shifts provide consistency for construction employers 
who are not required to test hearing (Ex. 3-10), and that the 10-db 
recording criterion is more protective and reasonable for employers who 
are not covered by the OSHA noise standard (Exs. 3-10, 3-17, 3-18, 3-
24).

Alternatives Offered

    Most of the commenters who objected to the recording of 10-db 
shifts presented alternative recording thresholds. The American 
Chemistry Council recommended a 15-db shift (Ex. 5-5), the Rubber 
Manufacturers Association recommended a 20-dB shift (Ex. 3-27), and 
Abbott Laboratories recommended recording second and subsequent 10-db 
shifts (Ex. 3-13). By far, the most common alternative offered was a 
shift of 25 dB (Exs. 3-1, 3-14, 3-19, 3-20, 3-22, 3-26, 3-29, 3-34, 3-
35, 3-37, 3-43, 3-45, 3-48, 3-50, 3-57, 3-58, 3-61, 3-63, 4-3, 4-5). 
The commenters supporting a 25-dB shift argued that 25 dB was superior 
because medical and health care professionals recommend using 25 dB 
(Exs. 3-29, 3-50, 3-54, 3-59), 25 dB is consistent with the American 
Medical Association (AMA) guidelines (Exs. 3-50, 3-54, 3-59), 25 dB is 
used for workers' compensation (Ex. 3-13), 25 dB is protective and 
provides an easily identifiable measurement for determining injuries 
(Ex. 3-35), and OSHA adopted 25 dB in 1991 because it is widely 
accepted as a meaningful loss of hearing and is well documented (Exs. 
3-37, 3-50, 3-54, 3-59).
    The National Association of Manufacturers (Ex. 3-50), the Can 
Manufacturing Institute (Ex. 3-50), and Industrial Health, Inc. (Ex. 3-
62) recommended a system where 15-db shifts would be recorded, but only 
when the shift crossed the disability boundary of 25 dB from 
audiometric zero. These commenters argued that the 15-db difference 
eliminated most shifts caused by audiometric error, and that by 
requiring them to cross the 25-dB fence, they would also clearly 
involve a hearing disability.
    Organization Resources Counselors (ORC) urged OSHA to adopt a 
``sliding scale'' recording criteria whereby the employer would record 
the first STS that exceeds 25 dB over audiometric zero, and all 
subsequent STS cases (Ex. 3-49). ORC argued that ``[t]here is no single 
objective level of hearing loss that is uniformly identifiable for 
every employee. Different employees enter the workplace with different 
levels of hearing capability, and noise affects people differently'' 
and that this concept reflects the intent of the OSH Act and the new 
rule in capturing significant injuries and illnesses.
    The American Iron and Steel Institute (Ex. 3-54), the Society for 
the Plastics Industry (Ex. 3-25) and the American Forest & Paper 
Association (Ex. 3-59) encouraged the adoption of a similar recording 
criteria where shifts would be

[[Page 44040]]

averaged over the frequencies of 500, 1000, 2000, and 3000 Hz, and the 
first shift of 10 dB over the disability fence of 25 dB would be 
recorded. This approach also set forth thresholds for the recording of 
subsequent shifts when they crossed boundaries used by various 
organizations for delineating mild, moderate, and severe hearing 
disability at the 40, 55 and 70-dB levels from audiometric zero.

OSHA's Decision

    Following consideration of the comments received in response to the 
July 3, 2001 proposal to modify the hearing loss recording criteria, 
OSHA has decided to require employers to record audiometric results 
indicating a Standard Threshold Shift (STS) only when such STS cases 
also reflect a total hearing level of at least 25 dB from audiometric 
zero. The STS calculation uses audiometric results averaged over the 
frequencies 2000, 3000 and 4000 Hz, using the original baseline and 
annual audiograms required by the OSHA noise standard Sec. 1910.95. The 
rule also allows the employer to adjust the employee's audiogram 
results used to determine an STS to subtract hearing loss caused by 
aging, allows the employer to retest the workers' hearing to make sure 
the hearing loss is persistent, and allows the employer to seek and 
follow the advice of a physician or licensed health care professional 
in determining whether or not the hearing loss was work-related.
    The approach adopted in the final rule has several advantages. By 
using the STS definition from the OSHA noise standard Sec. 1910.95, the 
Sec. 1904.10 regulation uses a sensitive measure of hearing loss that 
has occurred while the employee is employed by his or her current 
employer. By requiring all STSs to exceed 25 dB from audiometric zero, 
the regulation assures that all recorded hearing losses are significant 
illnesses. OSHA received no comments suggesting that a shift of 25 dB 
from audiometric zero was anything less than a serious hearing loss 
case. While there is little consensus among the commenters concerning 
the appropriate level that should be used to record hearing loss cases, 
there is widespread agreement that a 25-dB shift from audiometric zero 
is a serious hearing loss.
    The hearing loss recording level is also compatible with the final 
rule's definition of injury or illness, ``an abnormal condition or 
disorder'' (Sec. 1904.46). Various scales used to rate hearing loss 
consider hearing levels less than 25 dB to be within the ``normal 
range'' (American Medical Association Guidelines to the evaluation of 
Material Impairment, American Academy of Family Physicians, Audiology 
Awareness Campaign). The recording level is also compatible with the 
definition of material impairment used by OSHA and MSHA in the 
development of standards for occupational noise exposure (64 FR 49548, 
48 FR 9738).
    The hearing loss recording requirements in Sec. 1904.10 differ from 
the requirements of the OSHA noise standard (Sec. 1910.95) because 
under the noise standard the employer is required to take certain 
actions (employee notification, providing hearing protectors or 
refitting of hearing protectors, etc.) for all 10-db standard threshold 
shifts while the part 1904 rule only requires the recording of STSs 
that also exceed the total 25-db level. OSHA believes that this is an 
appropriate policy, because 10-db shifts in hearing at higher levels 
(above 25 dB) are more significant. Several commenters agreed that some 
shifts are more significant than others. ORC stated that ``(a) 10-db 
shift from audiometric zero is virtually imperceptible, while 10-db 
shifts at higher levels become more important'' (Ex. 3-49). The 
American Federation of Government Employees (Ex. 3-17) argued that 
``(h)earing loss is not linear, but is exponential, and changes are 
incrementally more serious and irreversible'' and the American 
Federation of State, County and Municipal Employees remarked that 
``(additional shifts are progressively more serious in nature'' (Ex. 3-
21)).
    When audiometric testing is done, test tones are presented at 
various sound levels, usually increasing or decreasing in 5-dB steps. 
The employee is asked to respond whenever a tone is heard, with the 
goal being finding the lowest level at which the employee can 
consistently hear. The standard measurement for measuring hearing level 
is decibels, a logarithmic scale. For the first increase in hearing 
level from 0 to 10 dB, the sound intensity increases 10 fold. The next 
10 dB is a 100-fold increase. By the time a person's hearing level 
changes from 0 to 30 dB hearing level, he or she needs 1,000 times more 
sound intensity to just barely hear.
    Although the part 1904 recordkeeping regulation and the 
Sec. 1910.95 noise standard treat the STS cases differently, this has 
no effect on the noise standard's requirements and does not have any 
effect on the requirement for employers to comply with Sec. 1910.95. 
When employers detect work-related STS cases, they are required to take 
all of the follow-up actions required by the noise standard.
    Additionally, the STS measure uses existing measurements and 
calculations employers are already using to comply with the OSHA noise 
standard, resulting in less paperwork burden for employers covered by 
both rules. Employers are required to take one additional step to 
determine if the STS has also resulted in a total hearing level of 25 
dB or more, and if so, to record it. The position taken in Sec. 1904.10 
provides a reasonable compromise between the commenters' highly 
polarized views on the proper recording level. The final rule's hearing 
loss recording provisions provide a reasonable ``middle ground'' 
solution to reconcile the differences between a highly sensitive 
measure of hearing loss (all 10-db shifts) and increasingly insensitive 
measures (15, 20, or 25-db shifts).
    The approach used in this final rule is a newly developed 
alternative that was not considered in the January 2001 rulemaking 
because none of the commenters to the 1996 proposed rule suggested it. 
The approach was first suggested by Organization Resources Counselors 
in an unsolicited post-promulgation submission following publication of 
the January 2001 rule (Ex. 1-6). OSHA then solicited comment on the 
approach in the July 3, 2001 Federal Register notice requesting comment 
on the hearing loss recording issue (66 FR 35113--35115).
    OSHA believes that the Sec. 1904.10 requirements will improve the 
nation's statistics on occupational hearing loss and that more hearing 
loss cases will be entered on employers' OSHA 300 Logs. However, OSHA 
recognizes that the new requirements may not result in comprehensive 
statistics for occupational hearing loss. Employees may experience 
significant hearing loss in industries where audiometric testing is not 
required (construction, agriculture, oil and gas drilling and 
servicing, and shipbuilding industries), and is not provided 
voluntarily by the employer, and thus never be entered into the 
records. Likewise, an employee may experience gradual hearing loss 
while employed by several employers, but never work for the same 
employer long enough to allow a recordable STS to be captured. As to 
the effect on trend analysis, caution must be used when comparing 
Sec. 1904.10 hearing loss data that span the effective date of this 
rule. The new hearing loss recording rule will result in the recording 
of additional cases of hearing loss, not as a result of a change in the 
number of workers who experience hearing loss, but simply because of 
the recordkeeping change.
    OSHA finds that recording only 25-dB shifts from the employee's 
baseline audiogram is not an appropriate policy. If an employee had 
significant hearing

[[Page 44041]]

loss before being hired by the employer, additional hearing loss would 
not be recorded until well beyond the point of disability. This would 
not conform to the requirements of section 24 of the Act directing the 
Secretary to ``[c]ompile accurate statistics on work injuries and 
illnesses which shall include all disabling, serious, or significant 
injuries and illnesses * * *'' (emphasis added) (29 U.S.C. 673). The 
recording of 25-dB shifts in hearing acuity, measured from the 
employee's original baseline audiogram would clearly understate the 
true incidence of work-related hearing loss. Likewise, if the part 1904 
regulation were to require only the recording of 15 or 20-dB shifts, or 
categorically exclude the first STS case the rule would exclude many 
legitimate and serious hearing loss cases that should rightfully be 
entered into the records and the Nation's injury and illness 
statistics. This approach would be especially deficient at capturing 
hearing loss in those employees who change employers several times 
during their working lives.
    The Coalition to Protect Workers Hearing (Ex. 3-23) and the AFL-CIO 
(Ex. 3-24) specifically opposed the approach used in the final rule, 
which is often referred to as a ``sliding scale'' approach because it 
treats some STS cases as being more serious than others (Exs. 3-23, 3-
24). These Commenters argued that a sliding scale approach was rejected 
in 1981 because it was too complex (Exs. 3-23, 3-24), that sliding 
scales are difficult to administer and do not provide uniform 
protection for workers (Ex. 3-24), and that ``(c)ategorizing employers 
on the basis of hearing impairment is discriminatory. * * * Women and 
African Americans, both of whom tend to have better hearing 
sensitivity, might be placed in noise-hazardous jobs since they could 
develop more hearing change without crossing the line'' (Exs. 3-23-1, 
3-53).
    OSHA does not believe that these concerns are serious impediments 
to the Section 1904.10 requirements. The two-part test, an STS combined 
with a total hearing level in excess of 25 dB from audiometric zero, is 
not overly complex, and is not nearly as complex as some of the sliding 
scale approaches that were rejected during the revision of the OSHA 
noise standard in 1981. In the years since 1981, computer technology 
has become much more commonplace and is incorporated into most, if not 
all, audiometric equipment. OSHA expects that most employers and 
contractors who administer hearing tests under the provisions of the 
noise standard will use computer software to make the needed 
calculations, so the requirements will not be difficult to administer. 
OSHA has received no evidence to show that the policies in the final 
rule will encourage discriminatory behavior by employers. The 
suggestion that women or African Americans may be selected for noise 
exposed jobs in order to avoid a potential recordable hearing loss case 
is highly speculative. OSHA has seen no evidence that such 
discrimination has occurred either to avoid the requirements of the 
OSHA noise standard or to avoid workers' compensation issues.
    OSHA does not agree with the commenters who argued that because the 
function of the OSHA standards and regulations, including the part 1904 
regulation, is to protect workers, worker protection would be 
compromised by any policy other than the recording of all STS cases. 
OSHA encourages employers and employees to use the OSHA injury and 
illness records to improve workplace safety and health conditions, and 
this is one of the functions of the Part 1904 records. However, this is 
not the only function of the records. They are also used to generate 
injury and illness statistics for the Nation and for individual 
workplaces. They are used by OSHA representatives to identify hazards 
during workplace inspections, and are collected by OSHA to target its 
intervention efforts to more hazardous worksites (See 66 FR 5916-5917). 
As stated in the 2001 rulemaking, ``[n]o new protections are being 
provided by the recordkeeping rule''. Further, the OSH Act does not 
require the recording of all injuries and illnesses and specifically 
excludes certain minor injury and illness cases. This exclusion, which 
is discussed in the preamble to the January 19, 2001 final rule, 
applies to both injuries and illnesses, including hearing loss (See 66 
FR 5931-5932). It is thus entirely appropriate for the recordkeeping 
rule to exclude certain minor illness cases while capturing more 
serious cases.
    The hearing loss recording requirements of Section 1904.10 will not 
deprive employers and employees of information about noise hazards or 
diminish workers' protection against the hazards of noise in the 
workplace. The occupational noise exposure standard requires that 
employees in general industry be tested for hearing loss when noise 
exposure exceeds an 8-hour time-weighted average of 85dB, and that 
employees be informed, in writing, if a 10-dB shift has occurred. The 
audiometric test records must be retained for the duration of the 
affected employee's employment. (See 29 CFR 1910.95(g), (m)). The noise 
standard also specifies the protective measures to be taken to prevent 
further hearing loss for employees who have experienced a 10-dB shift, 
including the use of hearing protectors and referral for audiological 
evaluation where appropriate. (See 29 CFR 1910.95(g)(8)). These 
requirements, which apply without regard to the recording criteria in 
the recordkeeping rule, will protect workers against the hazards of 
noise. The modified requirements of Section 1904.10 will therefore not 
deprive employers and workers of the means to detect and prevent 
hearing loss.
    Finally, section 4(b)(4) of the OSH Act provides that ``[n]othing 
in this Act shall be construed to supercede or in any manner affect any 
workmen's compensation law or to enlarge or diminish or affect in any 
other manner the common law or statutory rights, duties, or liabilities 
of employers and employees under any law with respect to injuries, 
diseases, or death of employees arising out of, or in the course of, 
employment.'' 29 U.S.C. 653(b)(4). Accordingly, the OSHA recordkeeping 
rule will have no legal effect on state workers' compensation systems. 
There is no evidence that the states have modified their systems to 
conform to OSHA's previous hearing loss recording policies; in fact, 
the states are far from uniform in their treatment of occupational 
hearing loss (Ex. 3-24-14). Therefore, OSHA does not expect the 1904 
regulation to have any effect on state workers' compensation in the 
future.

Audiometric Error

    In its July 3, 2001 proposal, OSHA asked the public to comment on 
the variability of audiometric testing equipment and how testing 
variability should be taken into account, if at all, in the 
recordkeeping rule (66 FR 35115). Many commenters questioned the 
accuracy of audiograms, and some of them specifically questioned the 
accuracy of audiograms used to compute 10-dB shifts in hearing acuity 
(Exs. 3-5, 3-13, 3-14, 3-19, 3-20, 3-25, 3-26, 3-27, 3-29, 3-30, 3-35, 
3-37, 3-45, 3-48, 3-49, 3-50, 3-54, 3-56, 3-58, 3-59, 3-63). These 
commenters argued that 10 dB is the lowest level of detection and is 
not reliable (Exs. 3-48, 3-63); at 10 dB the precision of the 
measurement becomes an issue (Ex. 3-49); 5 to 10-dB variability is 
common, which argues for 25 dB and against 10 dB (Ex. 3-29); 10 dB is 
not effective because of the testing environment, testing procedures, 
and error of audiometric equipment (Ex. 3-27); and that at a 10-dB 
shift, there is significant uncertainty in measurement, rendering

[[Page 44042]]

a typical audiometric reading unreliable (Exs. 3-37, 3-56). Verizon 
Communications, Inc., while supporting the recording of 10-dB shifts, 
summarized the potential recording problem as follows:
    The test-retest variability inherent in properly calibrated 
audiometric equipment is  5 dB. * * * if a 10-dB recording 
threshold is adopted, the following scenario is possible:


Baseline audiogram--the threshold at 200 Hz is measured at 10 dB; 
however, the equipment is off by -5 dB, so the threshold is really 
15 dB
Follow-up audiogram--the threshold at 200 Hz is measured at 20 dB; 
however, the equipment is off by +5 dB, so the threshold is still 15 
dB
This employee would have a recordable 10-dB loss, yet, in reality, 
his/her hearing would be unchanged. This is the risk that is taken 
with a 10-dB threshold--too many false positives (Ex. 3-30).

    The International Paper Company stated that ``[a]pplying the 10-dB 
STS criterion for recordkeeping purposes would have the effect of 
recording large numbers of workers whose hearing losses may simply be 
due to testing variability'' (Ex. 3-14). The Society for the Plastics 
Industry (Ex. 3-25) cited a number of articles in the scientific 
literature to argue that measurement error in field testing as 
approximately  10 dB and the measurement error under 
laboratory conditions is  5 dB. The Specialty Steel 
Industry of North America (SSINA) and the Steel Manufacturers 
Association (SMA), in a combined comment, used information from the 
National Institute for Occupational Safety and Health (NIOSH) to argue 
that typical audiometric testing variability is 10 dB, stating that 
``(e)mployers will be required to record each occurrence of an STS at 
10 dB, using a test that has a 10-dB measurement variability. This will 
generate an overwhelming number of false positives'' (Ex. 3-37).
    In a single comment, the National Chicken Council and the National 
Turkey Federation argued that ``Lacking standardization in testing 
methods and in testing equipment, this change will mean that employers 
will likely be forced to record (or fail to record) STSs that are 
inaccurately measured'' (Ex. 3-19). The Hearing Conservation Team at 
the Naval Submarine Medical Research Laboratory (Ex. 3-56) reviewed the 
scientific literature on audiogram reliability and found that 
methodology used by various researchers varied widely, making study 
comparisons difficult. The Hearing Conservation Team recommended 
further research into the test-retest reliability of various threshold 
levels that could then be used to set an STS criterion that would 
minimize false positives.
    Another group of commenters argued that the accuracy of audiometric 
testing equipment is not a major factor (Exs. 3-15, 3-22. 3-23-1, 3-24, 
3-57, 3-58, 3-61, 5-2, 5-3). In a representative comment, the AFL-CIO 
remarked that ``The issue of audiometric test variability has been a 
settled matter since the hearing conservation amendment was promulgated 
nearly 20 years ago and is adequately addressed by the existing 
provisions contained in 1904.10'' (Ex. 3-24). The American Textile 
Manufacturers Institute commented that: ``Variability is a given in 
audiometric testing as it can never be an exact process as long as it 
relies on any given individual being tested to sense a signal and 
respond. However, variability can be minimized if there are tight 
quality controls on the test equipment, procedures, etc.'' (Ex. 3-15).
    The Coalition to Protect Workers Hearing disagreed with OSHA's 
suggestion that the 10-dB recordability criterion does not allow for 
audiometric variability, stating that ``The evaluation of work-
relatedness takes calibration shifts into account, and such audiometric 
variability occurs infrequently. When random measurement variability 
does occur, retesting reduces it'', adding that ``It is true that 
audiometric data are vulnerable to calibration differences between 
different audiometers. Calibration discrepancies may occur if the 
employer changes service providers (e.g., mobile audiometric testing, 
testing in an off-site clinic) or if the employer switches audiometers 
for in-house testing. Such change can easily affect data by 5 dB. 
However, calibration discrepancies can be minimized through careful 
procedural controls such as the use of bio-acoustic simulators and 
proper professional supervision of the audiometric monitoring program'' 
(Ex. 3-23-1).
    The Dow Chemical Company, which has voluntarily been using 10-dB 
shifts for recording loss, stated that ``In Dow's experience, following 
a standardized testing protocol (using 29 CFR 1910.95), and including 
adjustment for age and the use of a retest in 30 days, has provided 
accurate, consistent results' (Ex. 5-2). The National Institute for 
Occupational Safety and Health (NIOSH) argued that the variability of 
testing should not be taken into account in the recordkeeping rule 
because audiometric variability issues have been addressed in the OSHA 
Noise Standard 29 CFR 1910.95. NIOSH stated that they believe that 
under the OSHA Noise Standard the expected variability due to error 
will be 5 dB (Ex. 5-3).
    OSHA agrees with NIOSH that the recordkeeping rule should not take 
any actions to address the issues of audiometric variability, and finds 
that there is no need to increase the recording loss threshold to 15 or 
20 dB to account for variability. The OSHA noise standard includes 
provisions that standardize audiometric testing protocols. The 
requirements in Sec. 1910.95 (g) Audiometric Testing Program, 
Sec. 1910.95 (h) Audiometric Test Requirements, Mandatory Appendix C to 
Sec. 1910.95 Audiometric Measuring Instruments, Mandatory Appendix D to 
Sec. 1910.95 Audiometric Test Rooms, and Mandatory Appendix E to 
Sec. 1910.95 Acoustic Calibration of Audiometers, and the incorporated 
provisions of American Standard Specification for Audiometers S3.6-1969 
provide standardized methodologies for conducting hearing tests 
designed to assure, as far as possible, that audiograms are accurate. 
As discussed in the preamble to the January 2001 final rule (66 FR 
6009), following these requirements will result in audiometric test 
results with a variability of 5 dB. As the Medical 
Educational Development Institute argued in response to the 1996 
proposal, ``(t)est/re-test reliability of 5 dB is well established in 
hearing testing. For example, the Council on Accrediting Occupational 
Hearing Conservationists maintain this range of reliability in their 
training guidelines and this is recognized in American National 
Standard Method for Manual Pure-Tone Threshold Audiometry, S3.21--1978 
(R1992).'' At the  5-dB reliability level, errors of 10 dB 
will be infrequent. There is a low probability that the audiometer will 
be incorrect by -5 dB on one test and +5 dB on a subsequent test 
because many of the variables affecting reliability will remain the 
same from year to year. The employer is likely to use the same 
audiometer, in the same room, operated by the same technician from one 
test to the next. When these variables are not held constant, or a 10-
dB shift occurs due to residual random variability, the allowance for 
retesting should largely eliminate spurious shifts due to audiometric 
measurement errors. Additionally, the use of an average shift at three 
frequencies reduces the influence of random audiometric variability; 
this is one of the reasons that a frequency averaged shift was adopted 
in the Sec. 1910.95 STS definition.
    It should be noted that it is impossible to eliminate audiometric 
errors in their entirety. Any recording level, no matter how it is set, 
will be subject to some

[[Page 44043]]

level of false positive and false negative errors. However, OSHA 
believes that the audiometric testing requirements of Sec. 1910.95, if 
followed, will provide reasonably accurate audiometric data for the 
administration of the OSHA noise standard, and for the recording of 
occupational hearing loss. As the Dow Chemical Company (Ex. 5-2) 
commented: ``(f)ollowing a standardized testing protocol (using 29 CFR 
1910.95), and including adjustments for age and the use of a retest in 
30 days, has provided accurate, consistent results.'' OSHA believes 
that the provisions allowing the employer to age adjust audiograms, 
seek advice from a physician or other licensed health care professional 
for determining work-relationship, retest within 30 days, and remove 
cases later found not to be persistent provide reasonable checks 
against false positive results being recorded on the 300 Log.

Age Correction

    The final rule carries forward the January 19, 2001 rule's 
conceptual framework allowing, but not requiring, the employer to age 
adjust an employee's annual audiogram when determining whether or not a 
10-dB shift in hearing acuity has occurred. There were no comments 
objecting to the age-correction of audiometric results when evaluating 
Standard Threshold Shifts in hearing. However, the American Iron and 
Steel Institute (Ex. 3-54), the Society for the Plastics Industry (Ex. 
3-25) and the American Forest & Paper Association (Ex. 3-59), in 
support of a recording criteria similar to that adopted in the final 
rule, recommended that, ``[b]ecause of the recognized contribution of 
aging to hearing loss, all hearing loss determinations would be age-
adjusted in accordance with Appendix F to 29 CFR 1910.95''.
    While the final rule allows the employer to age-correct the STS 
portion of the recording criteria, there is no allowance for age 
correction for determining a 25-dB hearing level. The AMA Guides 
specifically state that total hearing loss should not be age adjusted, 
and there is no recognized consensus method for age adjusting a single 
audiogram. The method used in Appendix F of Sec. 1910.95 is designed to 
age correct STS, not absolute hearing ability. The 25-dB criteria is 
used to assure the existence of a serious illness, and reflects the 
employee's overall health condition, regardless of causation. Age 
correcting the STS will provide adequate safeguards against recording 
age corrected hearing loss. Therefore, it would be inappropriate and 
unnecessary to age correct the 25-dB hearing level.

Persistence

    Although OSHA did not specifically ask for comment on the topic, 
several commenters raised the issue of how to verify that recorded 
hearing loss cases are persistent. The OSHA noise standard addresses 
the issue of temporary hearing losses by allowing the employer to 
retest the employee's hearing within 30 days (1910.95(g)(7)(ii)). The 
2001 rule adopted the same 30 day retest option at Sec. 1904.10(b)(4) 
by allowing the employer to delay recording if a retest was going to be 
performed in the next 30 days.
    A number of commenters stated that OSHA should record only 
permanent shifts in hearing (Exs. 3-23-1, 3-25, 3-26, 3-37, 3-48, 3-50, 
3-58, 3-61, 3-62). In a representative comment, Industrial Health Inc. 
remarked that ``[n]o shift, regardless of the number of dB, should be 
recorded unless it is found to be persistent in a second audiogram 
taken at a later time, which we believe should be no less than 60 days 
and preferably 6 months or more after the initial audiogram which 
revealed the shift'' (Ex. 3-62).
    The National Association of Manufacturers and the Can Manufacturing 
Institute, in a combined comment, argued that 30 days does not allow 
enough time to resolve transient conditions such as colds or allergies, 
and the retest period should be extended to one year (Ex. 3-50). The 
Coalition to Protect Workers Hearing recommended that ``(a)t the 
discretion of the reviewing professional, within 15 months of the 
initial identification of the STS, any STSs which are not confirmed by 
subsequent retesting or otherwise found not to be work related, may be 
lined out on Form 300. Documentation justifying line outs must be 
provided and should be retained with the employees' records'(Ex. 3-23).
    OSHA agrees with these commenters that the goal of the rule is to 
record only persistent hearing loss cases, and to help accomplish that 
goal, the Agency has carried forward the 30 day retest provision. 
However, OSHA has decided not to allow a longer retesting period. A 
longer retesting period would increase the likelihood that the employer 
would lose track of the case and therefore inadvertently fail to record 
the case. These errors would have a detrimental effect on the accuracy 
of the records and run counter to OSHA's goal of improving the quality 
of the injury and illness data. The Agency also believes that using 
different time periods for retesting in the part 1904 and Sec. 1910.95 
rules would result in increased confusion for employers.
    The Agency has also rejected the suggestion that all hearing loss 
cases must be confirmed prior to recording them. Waiting for one year 
or longer to record an occupational hearing loss would move the 
recording to a year in which the original hearing loss was not 
initially discovered, would be administratively more complex for 
employers, and would have a detrimental effect on the hearing loss 
data. Many legitimate hearing loss cases could go unrecorded simply 
because the employee did not receive a subsequent audiogram due to job 
changes or some other circumstance that might occur before the next 
annual audiogram required by the noise standard.
    In order to make it clear to employers that they may remove any 
cases that are found to be temporary, the final rule has adopted the 
removal option recommended by the Coalition to Protect Workers Hearing, 
with three modifications. First, the final rule does not include the 15 
month time limit. OSHA does not believe that a time limit is needed 
because any future audiogram that shows an improvement in hearing and 
refutes the recorded hearing loss would indicate a temporary hearing 
loss that should be removed from the records. Second, the regulatory 
text does not specify that the removal must be at the discretion of the 
reviewing professional. The OSHA noise standard, at Sec. 1910.95(g)(3), 
requires that:

    Audiometric tests shall be performed by a licensed or certified 
audiologist, otolaryngologist, or other physician, or by a 
technician who is certified by the Council of Accreditation in 
Occupational Hearing Conservation, or who has satisfactorily 
demonstrated competence in administering audiometric examinations, 
obtaining valid audiograms, and properly using, maintaining and 
checking calibration and proper functioning of the audiometers being 
used. A technician who operates microprocessor audiometers does not 
need to be certified. A technician who performs audiometric tests 
must be responsible to an audiologist, otolaryngologist or 
physician.

    Because the noise standard already requires audiograms to be 
conducted by, or under the supervision of, a qualified professional, 
subsequent audiograms that may refute the persistence of a recorded 
hearing loss will be reviewed by the appropriate professional. The 
Sec. 1904.10 simply cross-references the need for the audiograms to be 
obtained pursuant to the requirements of Sec. 1910.95, so there is no 
need for the Sec. 1904.10 rule to repeat the review requirement. Third, 
the rule does not

[[Page 44044]]

require the employer to maintain documentation concerning the removal 
of cases. Section 1910.95(m)(2) of the noise standard requires the 
employer to keep records of all audiometric tests that are performed, 
and those records will be available, should they be needed for future 
reference. As a result, there is no need to add a duplicative paperwork 
burden in the Sec. 1904.10 rule. Therefore, Sec. 1904.10(b)(4) states 
that ``If subsequent audiometric testing indicates that an STS is not 
persistent, you may erase or line-out the recorded entry''. OSHA has 
added this additional regulatory language to minimize the recording of 
temporary hearing loss cases while capturing complete data on the 
incidence of hearing loss disorders.

Frequencies

    Some commenters urged OSHA to measure hearing loss at frequencies 
other than 2000, 3000 and 4000 Hz (See, e.g., Exs. 3-25, 3-54, 3-57, 3-
58, 3-59, 3-61). Alabama Power (Ex. 3-61) and the Southern Company (Ex. 
3-58) recommended using 500, 1000, and 2000 because ``these are the 
frequencies where most communication occurs''. Another group of 
commenters recommended the use of 500, 1000, 2000 and 3000 Hz because 
these are the frequencies specified by the American Medical Association 
and the American Academy of Otolaryngology-Head and Neck Surgery, Inc. 
(Exs. 3-25, 3-54, 3-57, 3-59).
    OSHA has decided to continue to use the frequencies used in the 
Sec. 1910.95 OSHA noise standard (2000, 3000, and 4000 Hz). While 
``most'' communication occurs at lower frequencies, these are clearly 
audible frequencies where some speech occurs, and where hearing loss 
can have a significant impact on workers' lives outside of verbal 
communication. Using these frequencies reduces the burden on employers 
that would be created by requiring separate calculations of audiometric 
results, and, as Industrial Health, Inc. stated ``(w)ith regard to the 
early effects of noise exposure, it seems reasonable to extend the 
definition across the standard shift frequencies 2000, 3000, and 4000 
Hz'' (Ex. 3-62).

Baseline Reference and Revision of Baseline

    In its July 3, 2001 Federal Register notice OSHA asked the public 
to comment on the appropriate benchmark against which to measure 
hearing loss, e.g., the employee's baseline audiogram, audiometric 
zero, or some other measure (66 FR 35115). One commenter, Eric Zaban 
with the State of Michigan, suggested using audiometric zero as the 
appropriate benchmark (Ex. 4-1). The vast majority of the commenters 
who addressed this issue supported using the employee's baseline 
audiogram (Exs. 3-15, 3-20, 3-21, 3-22, 3-23-1, 3-24, 3-25, 3-27, 3-29, 
3-30, 3-37, 3-47, 3-49, 3-50, 3-53, 3-54, 3-57, 3-58, 3-59, 3-61, 3-62, 
3-63, 4-2, 4-5, 5-2, 5-3, 5-5 ). Alabama Power remarked that:

[T]he appropriate benchmark against which to measure hearing loss is 
the employee's original baseline. Using the employee's original 
baseline ensures that employers are not held responsible for any 
prior hearing loss the employee may have suffered. Comparing an 
employee's audiogram to audiometric zero would not take into account 
any previous hearing loss that may have occurred prior to employment 
(Ex. 3-61).

    The AFL-CIO agreed, stating that ``Using the original baseline 
takes into account any hearing loss that a worker may have experienced 
while employed by a previous employer'' and ``Using the baseline 
ideogram (audiogram) will assist in preventing the recording of cases 
of non-occupational hearing loss' (Ex. 3-24).
    The two-part test for recording that is being adopted in the final 
rule uses the baseline audiogram as the reference point for determining 
whether or not the employee has had a change in hearing while employed 
by his or her current employer, and then uses audiometric zero as the 
reference point for determining the overall hearing ability of the 
affected employee. OSHA agrees that the employee's baseline audiogram 
is a superior reference point for measuring a change of hearing, a 
Standard Threshold Shift. Using the baseline audiogram taken upon 
employment reduces the effect of any prior hearing loss the employee 
have experienced, whether it is non-occupational hearing loss or 
occupational hearing loss caused by previous employment. Therefore, the 
final rule uses the employee's original baseline audiogram as the 
reference for the STS component of an initial hearing loss cases, and 
uses the revised baseline audiogram from that initial case as the 
reference for future cases.
    The 25-dB total hearing level component of an OSHA recordable 
hearing loss uses a reference of audiometric zero. This portion of the 
recording criteria is used to assure that the employee's total hearing 
level is beyond the normal range of hearing, so it does not exclude 
hearing loss due to non-work causes, prior employment, or any other 
cause. The measurement simply reflects the employee's current hearing 
ability as reflected in the most recent audiogram. This comparison to 
audiometric zero is a simple matter, because audiometers are designed 
to provide results that are referenced to audiometric zero. The hearing 
level at each frequency is oftentimes printed by the equipment, so 
there is rarely a need to perform manual calculations.

Work Relationship

    The final rule published on January 19, 2001 included a presumption 
of work-relatedness when employees are exposed to loud noise at work, 
relying on the OSHA noise standards criteria of an 8-hour 85 dBA 
exposure level, or a total noise dose of 50 percent. The preamble 
discussion of the work-relatedness presumption was that:

    [I]n line with the overall concept of work relationship adopted 
in this final rule for all conditions, an injury or illness is 
considered work related if it occurs in the work environment. For 
workers who are exposed to the noise levels that require medical 
surveillance under Sec. 1910.95 (an 8-hour time-weighted average of 
85 dB(A) or greater, or a total noise dose of 50 percent), it is 
highly likely that workplace noise is the cause of or, at a minimum, 
has contributed to the observed STS. It is not necessary for the 
workplace to be the sole cause, or even the predominant cause, of 
the hearing loss in order for it to be work-related (66 FR 6012).

    Several commenters discussed the difficulties of determining the 
work-relatedness of hearing losses, and many argued that the 8-hour 85 
dBA presumption was invalid (Exs. 3-2, 3-3, 3-13, 3-20, 3-23-1, 3-25, 
3-27, 3-29, 3-37, 3-43, 3-48, 3-50, 3-54, 3-63, 4-3). In a 
representative comment, the Coalition to Protect Workers Hearing (Ex. 
3-23-1) remarked that:

    [W]ork relatedness should not be presumed solely on the basis of 
an exposure to time-weighted averages (TWAs) of 85 dBA or higher; 
instead it should be evaluated on a case-by-case basis. Presumption 
of work-relatedness based on equivalent 8-hour exposure alone is 
unsatisfactory because it presumes that the employer's hearing 
conservation program is completely ineffective and does not take 
into account other factors such as hearing protector fit and use 
compliance. Presumption of work-relatedness is a disincentive for 
employers to develop successful programs and to implement noise 
control because they receive no credit for their efforts. The 
audiologist or physician reviewing the audiometric record should 
make a determination regarding whether the OSHA STS is work-related 
and should do so when the 10-dB STS occurs.

    Other commenters suggested that if an employer has an active and 
enforceable hearing conservation program in effect, then the 
recordkeeping rule should presume that a hearing loss case is non-work-
related (Exs. 3-37, 3-50); that the rule needs to take non-work noise 
exposure into account (Exs. 3-29, 3-37, 3-50); and that the rule should 
only

[[Page 44045]]

consider a hearing loss to be work-related if work contributed more 
than 50% (Ex. 3-63). Several commenters made the same argument as the 
Coalition to Protect Workers Hearing, arguing that each case should be 
evaluated on its merits (Exs. 3-29, 3-43, 3-50, 3-63). The American 
Foundry Society argued that ``[w]ork-relatedness should be evaluated by 
a health care professional with experience in occupational health. Low 
level occupational noise exposure or documented regular use of hearing 
protection devices (HPDs) in noisy areas should mitigate against the 
presumption of work-relatedness' (Ex. 3-63).
    OSHA agrees with these commenters that it is not appropriate to 
include a presumption of work-relatedness for hearing loss cases to 
employees who are working in noisy work environments. It is possible 
for a worker who is exposed at or above the 8-hour 85-dBA action levels 
of the noise standard to experience a non-work-related hearing loss, 
and it is also possible for a worker to experience a work-related 
hearing loss and not be exposed above those levels. Therefore, the 
final rule states that there are no special rules for determining work-
relationship and restates the rule's overall approach to determining 
work-relatedness--that a case is work-related if one or more events or 
exposures in the work environment either caused or contributed to the 
hearing loss, or significantly aggravated a pre-existing hearing loss.
    The final rule's approach to determining work-relatedness differs 
from the January 2001 rule for three reasons. First, although it is 
likely that occupational exposure to noise in excess of 85 dBA will be 
a causal factor in hearing loss in some cases, a presumption of work-
relatedness is not justified in all cases. Further evaluation is needed 
to make this determination. Second, the policy in the final rule is 
consistent with the general principle in Sec. 1904.5 that work-
relatedness is to be determined on a case-by-case basis. Third, the 
approach used in the January 2001 rule is not supported by comments to 
the docket. None of the commenters supported the presumption, while 
many opposed it.
    The final rule also continues the 2001 rule's policy allowing the 
employer to seek the guidance of a physician or other licensed health 
care professional when determining the work-relatedness of hearing loss 
cases. Paragraph (b)(6) of the rule states that if a physician or other 
licensed health care professional determines that the hearing loss is 
not work-related or has not been significantly aggravated by 
occupational noise exposure, the employer is not required to consider 
the case work-related, and therefore is not required to record it.
    When evaluating the work relatedness of a given hearing loss case, 
the employer should take several factors into account. The Coalition to 
Protect Workers Hearing recommended that employers consider prior 
occupational and non-occupational noise exposure, evaluation of 
calibration records and the audiometric environment, investigation of 
related activities and personal medical conditions, and age correction 
before presuming that hearing loss is work related (Ex. 3-23-1). One 
important factor to consider is the effectiveness of the hearing 
protection program. When employees are exposed to high levels of noise 
in the workplace, and do not wear appropriate hearing protection 
devices, a case of hearing loss is more likely to be work-related. If 
an employee's hearing protection devices are not appropriate for the 
noise conditions, if they do not fit properly, or if they are not used 
properly and consistently, they may not provide enough protection to 
prevent workplace noise from contributing to a hearing loss case.

Adding a Column to the 300 Log

    Section 1904.10(a) of the January 2001 rule required that employers 
check a hearing loss column on the Log when recording a hearing loss 
case. OSHA is issuing a separate Federal Register document proposing to 
delay the effective date of the hearing loss column requirement until 
January 1, 2004, and asking for comment on issues related to the 
hearing loss column. The 1996 proposed recordkeeping rule did not 
contain a hearing loss column requirement, and did not ask for comment 
on whether a column should be added. In the 2001 final rule, OSHA 
explained that it was adding a hearing loss column to the 300 Log so 
that BLS could produce more reliable statistics on occupational hearing 
loss cases (66 FR 6005). OSHA's July 3, 2001 Federal Register notice 
sought comment on alternative criteria for recording occupational 
hearing loss, but did not mention the hearing loss column as an issue.
    OSHA does not believe that the existing record provides an adequate 
basis to determine the need for the hearing loss column. OSHA believes 
that interested parties should be allowed to comment on the issue. 
Accordingly, OSHA is publishing a separate Federal Register document 
today, proposing to delay the effective date of the hearing loss 
requirement until January 1, 2004 while the Agency reconsiders the 
column requirement in light of public comment. To facilitate public 
comment, OSHA has separated the requirement from Sec. 1904.10(a) and 
placed it in a separate paragraph at Sec. 1904.10(b)(7), which asks 
``How do I complete the 300 Log for a hearing loss case?'' and answers 
``When you enter a recordable hearing loss case on the OSHA 300 Log, 
you must check the 300 Log column for hearing loss illnesses.'' To 
further help assure that the public is informed about this additional 
rulemaking activity, OSHA is adding a regulatory note to 
Sec. 1904.10(b)(7) explaining that OSHA is delaying the applicability 
of Sec. 1904.10(b)(7) until further notice while the Agency reconsiders 
the hearing loss column.

Miscellaneous Hearing Loss Issues

    OSHA received one miscellaneous comment that is worthy of 
discussion. The International Chemical Workers Union Council (Ex. 3-53) 
remarked that ``[i]t is difficult for workers and their representatives 
to gain access to audiometric exams or summaries of those exams''. 
Several of OSHA's rules provide access rights to audiometric data. 
Section 1910.95(g)(8) of the noise standard requires employers to 
inform employees, in writing, that they have experienced a standard 
threshold shift. OSHA's rule for access to employee exposure and 
medical records (Sec. 1910.1020) requires employers to provide access 
to medical records, exposure records, and analyses of records to 
employee's and their designated representatives. Finally, the part 1904 
regulation requires employers to provide employee access to the OSHA 
injury and illness data.

Economic Analysis

Costs of the Revisions to the Hearing Loss Recording Provisions

    OSHA has determined that the total cost of this action is 
$1,049,650 per year and, thus, that it is not an economically 
significant regulatory action within the meaning of Executive Order 
12866. The methodology that OSHA has used for computing costs for the 
new rule is presented in the next two sections.

Changes in Coverage

    Under the 2002 rule, employers were required to record all hearing 
loss cases that involved a work-related Standard Threshold Shift (STS) 
of an average of 25 dB or more at 2000, 3000 and 4000 hertz (Hz) in 
either ear, compared to the employee's original baseline audiogram. The 
new rule requires recording all

[[Page 44046]]

hearing loss cases that involve a work-related STS of an average of 10 
dB or more if the accumulated loss of hearing is at least 25 dB above 
audiometric zero. (The use of the tables in Appendix F of the Noise 
Standard to adjust for aging remains unchanged.)
    OSHA estimates that approximately 40,000 hearing loss cases would 
have to be recorded under the 2002 rule, as opposed to approximately 
145,000 hearing loss cases under the new rule. Thus, the new rule 
increases the number of recordable hearing loss cases by approximately 
105,000. (In the Final Economic Analysis of the 2001 revisions to the 
rule, OSHA estimated that there would be 275,000 additional hearing 
loss cases (66 FR 6121), but the new rule has a narrower definition of 
hearing loss cases than the 2001 rule.)

Estimating the Number of Recordable Hearing Loss Cases

    To estimate the number of cases that would be recorded, OSHA used 
the same estimation methodology as in the January 19, 2001 final rule. 
First, OSHA estimated the number of employees that would receive 
audiometric tests. OSHA's noise standard Sec. 1910.95 requires 
employers to provide baseline and annual audiograms (and take other 
actions) when employees are exposed to certain noise levels. OSHA 
believes that approximately 23% of workers in the manufacturing sector 
are covered by the OSHA noise standard. Therefore, the number of 
covered manufacturing workers is 4,255,000 (18,500,000 manufacturing 
workers x .23). OSHA estimates that an additional 10% of workers are 
covered in other general industry sectors (such as transportation and 
utilities) or receive audiograms in industries not required to perform 
audiometric testing under the OSHA noise standard (such as construction 
and agriculture). Therefore, the total number of covered workers is 
estimated to be approximately 4,680,500 (4,255,000 x 1.1).
    OSHA then reviewed a National Institute for Occupational Safety and 
Health (NIOSH) database of audiograms to determine the proportion of 
audiograms meeting the recording criteria. 3.09% of audiograms met the 
final rule's criteria for recording hearing loss, and 0.83% met the 
2002 recording criteria (25 dB). Applying this percentage to the number 
of employees receiving annual audiograms results in 144,627 (4,680,500 
x 0.0309) estimated hearing loss cases under the final rule, and 38,848 
(4,680,500 x .0083) estimated hearing loss cases recorded under the 
2002 rule.
    Therefore, OSHA estimates 105,779 (144,627 - 38,848) additional 
cases of occupational hearing loss will be captured by the final 
section 1904.10 regulation, and has rounded this figure to 105,000 for 
cost estimation purposes.

Annual Costs of Maintaining Records

    The additional hearing loss cases will require additional entries 
on the OSHA Form 300 Log and Summary of Occupational Injuries and 
Illnesses and the OSHA Form 301 Injury and Illness Incident Report. 
Access of employees and their representatives to the additional Form 
301s will also involve costs.
    OSHA estimates that employers will incur for each additional 
hearing loss case a cost of 15 minutes for the Log entry.
    As explained in the 2001 Final Economic Analysis, based on data 
collected during approximately 400 recordkeeping audit inspections, 
OSHA estimates that 82 percent of incidents will be recorded on forms 
other than Form 301, such as workers' compensation forms. The remaining 
18% of additional hearing loss cases will take 22 minutes for the 
filling out the Form 301.
    Assuming that an individual with the skill level of a Personnel 
Training and Labor Relations Specialist will do the recordkeeping 
required by this rule, an hourly wage of $30.02 is used to compute 
cost. (The average hourly wage for a Personnel Training and Labor 
Relations Specialist as reported in the Bureau of Labor Statistics 
Occupational Employment Statistics Survey for Year 2000 was $21.71; 
benefits are computed at 38.3 percent of the hourly wage.)
    Thus, employers will incur, for each additional hearing loss case, 
data entry costs of 15 minutes for the Log entry plus, for 18% of the 
cases, 22 minutes for the Form 301. The total annual cost is estimated 
to be $996,064 [= (105,000 Cases) x (15 Minutes/Case) x ($30.02/Hour) + 
(18,900 Cases) x (22 Minutes/Case) x ($30.02/Hour)].
    As in the Year 2001 Final Economic Analysis, OSHA assumes that (a) 
at one-tenth of covered establishments, one employee would request 
access to his or her own Form 301 (10,500 instances), and (b) at one 
percent of covered establishments, a union representative would request 
access to all Form 301s at the establishment. Using the same estimation 
method as the 2001 Economic Analysis, OSHA estimates union 
representative access will result in an additional 10,500 forms being 
provided by employers. OSHA assumes that, for each of the 21,000 forms 
being provided (10,500 + 10,500), employers would require five minutes 
to pull, copy (at $0.05), and replace the relevant Form 301.
    The estimated total cost of providing access to additional hearing 
loss records would thus be $47,110 [= (21,000 Forms) x (5 Minutes x 
($30.02/Hour) + $.05/Copy)]. Thus, according to the above analysis, the 
total annual cost of this regulatory action is $1,049,650.

Benefits

    Hearing loss cases result in substantial disability and lead to 
safety accidents as well. OSHA believes that aligning the recording 
threshold for such cases with the STS criterion in the Agency's Noise 
Standard will simplify recording for many employers who are already 
familiar with this criterion and provide more opportunities for 
employers to intervene to prevent other hearing loss cases.
    As explained in the 2001 Final Economic Analysis, possession of 
information about events and exposures will increase the ability of 
employers and employees to identify hazardous conditions and to take 
remedial action to prevent future illnesses. If this enhanced ability 
to identify (and thus address) hazards translates into a reduction even 
as small as 0.5 to 1 percent of the estimated number of additional 
recordable cases, it would mean the prevention of 525 to 1,050 
illnesses per year [= (.005 to .01 x 105,000].
    The revisions in the rule will also make the injury and illness 
records more useful to OSHA, as well as to employers and employees. 
Improvements in the records being kept by employers would enhance 
OSHA's capacity to focus compliance outreach efforts on the most 
significant hazards; identify types or patterns of illness whose 
investigation might lead to regulatory changes or other types of 
prevention efforts, such as enforcement strategies, information and 
training, or technology development; and set priorities among 
establishments for inspection purposes.
    Employers and employees both stand to benefit from the more 
effective use of OSHA's resources. The enhanced ability of compliance 
officers to identify patterns of illness will enable OSHA to focus on 
more serious problems. Identification of such patterns will also 
increase the ability of employers to control these hazards and prevent 
other similar illnesses. To the extent that employers take advantage of 
this information, the burden of OSHA inspections should be reduced in 
the long run. Employees clearly also will

[[Page 44047]]

benefit from these reductions in illnesses.

Regulatory Flexibility Certification

    The 2001 revisions of the recordkeeping rule, which were much more 
extensive, did not have a significant impact on a substantial number of 
small entities (66 FR 6121). In the Final Economic Analysis for those 
revisions, OSHA estimated that over the entire range of SICs affected, 
the average cost per small firm was only $31.63. The impacts of those 
revisions on sales and profits did not exceed 1 percent for small firms 
in any covered industry (66 FR 6108).
    Even if all the additional hearing loss cases estimated to result 
from this year's revisions were distributed among the 541,988 small 
firms that keep the injury and illness records (as OSHA identified in 
its Year 2001 Final Economic Analysis) the average cost of the current 
revisions per small firm would be less than two dollars.
    OSHA hereby certifies that the current revision to the hearing loss 
recording provisions, with an estimated annual cost of just over a 
million dollars, will not have a significant impact on a substantial 
number of small entities.

Unfunded Mandates

    For the purposes of the Unfunded Mandates Reform Act of 1995, as 
well as Executive Order 12875, this rule does not include any Federal 
mandate that may result in increased expenditures by State, local, and 
tribal governments, or increased expenditures by the private sector of 
more than $100 million in any year.

Federalism

    This rule has been reviewed in accordance with Executive Order 
13132 (52 FR 41685), regarding Federalism. Because this rulemaking 
action involves a ``regulation'' issued under section 8 of the OSH Act, 
and not a ``standard'' issued under section 6 of the Act, the rule does 
not preempt State law, see 29 U.S.C. 667(a). The effect of the rule on 
States is discussed in the State Plans section of this preamble.

Paperwork Reduction Act

    OSHA will modify its previously approved information collection 
requirements prior to the January 1, 2003 effective date.

State Plans

    The 26 States and territories with their own OSHA-approved 
occupational safety and health plans must adopt a comparable regulation 
within six months of the publication date of this final regulation. 
These states and territories are: Alaska, Arizona, California, Hawaii, 
Indiana, Iowa, Kentucky, Maryland, Michigan, Minnesota, Nevada, New 
Mexico, North Carolina, Oregon, Puerto Rico, South Carolina, Tennessee, 
Utah, Vermont, Virginia, Virgin Islands, Washington, and Wyoming. 
Connecticut, New Jersey, and New York have OSHA approved State Plans 
that apply to state and local government employees only.
    A few commenters urged OSHA to make sure that the State Plan States 
have the same recording criteria as federal OSHA (see, e.g., Exs. 3-22, 
3-30, 3-49, 3-55). During 2002, the State Plan States were allowed to 
maintain their policies for the recording of hearing loss to maintain 
their former requirements, while OSHA reconsidered what the appropriate 
recording criteria should be. In the Federal Register document 
announcing the one year delay and the interim policy for year 2002, 
OSHA stated that when it issues a final determination for the recording 
of occupational hearing loss for calendar years 2003 and beyond, the 
states would be required to have identical criteria (66 FR 52033). Now 
that OSHA has issued its final determination, the States are required 
to promulgate identical criteria.

Executive Order

    This document has been deemed significant under Executive Order 
12866 and has been reviewed by OMB.

Authority

    This document was prepared under the direction of John L. Henshaw, 
Assistant Secretary for Occupational Safety and Health, U.S. Department 
of Labor, 200 Constitution Avenue, NW., Washington, DC 20210. It is 
issued pursuant to section 8 of the Occupational Safety and Health Act 
of 1970 (29 U.S.C. 657).

    Signed at Washington, DC, this 25th day of June, 2002.
John L. Henshaw,
Assistant Secretary of Labor.

    For the reasons stated in the preamble, 29 CFR part 1904 is amended 
as follows:

PART 1904--[AMENDED]

    1. The authority citation for part 1904 continues to read as 
follows:

    Authority: 29 U.S.C. 657, 658, 660, 666, 673, Secretary of 
Labor's Order No. 3-2000 (65 FR 50017), and 5 U.S.C. 533.


    2. Revise Sec. 1904.10 to read as follows:


Sec. 1904.10  Recording criteria for cases involving occupational 
hearing loss.

    (a) Basic requirement. If an employee's hearing test (audiogram) 
reveals that the employee has experienced a work-related Standard 
Threshold Shift (STS) in hearing in one or both ears, and the 
employee's total hearing level is 25 decibels (dB) or more above 
audiometric zero (averaged at 2000, 3000, and 4000 Hz) in the same 
ear(s) as the STS, you must record the case on the OSHA 300 Log.
    (b) Implementation.
    (1) What is a Standard Threshold Shift? A Standard Threshold Shift, 
or STS, is defined in the occupational noise exposure standard at 29 
CFR 1910.95(g)(10)(i) as a change in hearing threshold, relative to the 
baseline audiogram for that employee, of an average of 10 decibels (dB) 
or more at 2000, 3000, and 4000 hertz (Hz) in one or both ears.
    (2) How do I evaluate the current audiogram to determine whether an 
employee has an STS and a 25-dB hearing level?
    (i) STS. If the employee has never previously experienced a 
recordable hearing loss, you must compare the employee's current 
audiogram with that employee's baseline audiogram. If the employee has 
previously experienced a recordable hearing loss, you must compare the 
employee's current audiogram with the employee's revised baseline 
audiogram (the audiogram reflecting the employee's previous recordable 
hearing loss case).
    (ii) 25-dB loss. Audiometric test results reflect the employee's 
overall hearing ability in comparison to audiometric zero. Therefore, 
using the employee's current audiogram, you must use the average 
hearing level at 2000, 3000, and 4000 Hz to determine whether or not 
the employee's total hearing level is 25 dB or more.
    (3) May I adjust the current audiogram to reflect the effects of 
aging on hearing?
    Yes. When you are determining whether an STS has occurred, you may 
age adjust the employee's current audiogram results by using Tables F-1 
or F-2, as appropriate, in Appendix F of 29 CFR 1910.95. You may not 
use an age adjustment when determining whether the employee's total 
hearing level is 25 dB or more above audiometric zero.
    (4) Do I have to record the hearing loss if I am going to retest 
the employee's hearing?
    No, if you retest the employee's hearing within 30 days of the 
first test, and the retest does not confirm the recordable STS, you are 
not required to record the hearing loss case on the

[[Page 44048]]

OSHA 300 Log. If the retest confirms the recordable STS, you must 
record the hearing loss illness within seven (7) calendar days of the 
retest. If subsequent audiometric testing performed under the testing 
requirements of the Sec. 1910.95 noise standard indicates that an STS 
is not persistent, you may erase or line-out the recorded entry.
    (5) Are there any special rules for determining whether a hearing 
loss case is work-related?
    No. You must use the rules in Sec. 1904.5 to determine if the 
hearing loss is work-related. If an event or exposure in the work 
environment either caused or contributed to the hearing loss, or 
significantly aggravated a pre-existing hearing loss, you must consider 
the case to be work related.
    (6) If a physician or other licensed health care professional 
determines the hearing loss is not work-related, do I still need to 
record the case?
    If a physician or other licensed health care professional 
determines that the hearing loss is not work-related or has not been 
significantly aggravated by occupational noise exposure, you are not 
required to consider the case work-related or to record the case on the 
OSHA 300 Log.
    (7) How do I complete the 300 Log for a hearing loss case?
    When you enter a recordable hearing loss case on the OSHA 300 Log, 
you must check the 300 Log column for hearing loss.

    Note to 1904.10(b)(7): The applicability of paragraph (b)(7) is 
delayed until further notice.


[FR Doc. 02-16392 Filed 6-28-02; 8:45 am]
BILLING CODE 4510-26-P