[Federal Register Volume 84, Number 187 (Thursday, September 26, 2019)]
[Rules and Regulations]
[Pages 50739-50756]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 2019-20686]


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

DEPARTMENT OF LABOR

Occupational Safety and Health Administration

29 CFR Part 1910

[Docket No. OSHA-2015-0015]
RIN 1218-AC94


Additional Ambient Aerosol CNC Quantitative Fit Testing 
Protocols: Respiratory Protection Standard

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

ACTION: Final rule.

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

SUMMARY: OSHA is approving two additional quantitative fit testing 
protocols for inclusion in appendix A of the Respiratory Protection 
Standard. These protocols are: The modified

[[Page 50740]]

ambient aerosol condensation nuclei counter (CNC) quantitative fit 
testing protocol for full-facepiece and half-mask elastomeric 
respirators and the modified ambient aerosol CNC quantitative fit 
testing protocol for filtering facepiece respirators. The protocols 
apply to employers in general industry, shipyard employment, and the 
construction industry. Both protocols are abbreviated variations of the 
original OSHA-approved ambient aerosol CNC quantitative fit testing 
protocol (often referred to as the PortaCount[supreg] protocol), but 
differ from the test by the exercise sets, exercise duration, and 
sampling sequence. These protocols will serve as alternatives to the 
four existing quantitative fit testing protocols already listed in 
appendix A of the Respiratory Protection Standard and will maintain 
safety and health protections for workers while providing additional 
flexibility and reducing compliance burdens.

DATES: The final rule becomes effective on September 26, 2019.

ADDRESSES: In accordance with 28 U.S.C. 2112(a), the agency designates 
Edmund Baird, Acting Associate Solicitor of Labor for Occupational 
Safety and Health, Office of the Solicitor of Labor, Room S-4004, U.S. 
Department of Labor, 200 Constitution Avenue NW, Washington, DC 20210, 
to receive petitions for review of the final rule.

FOR FURTHER INFORMATION CONTACT: General information and press 
inquiries: Frank Meilinger, Director, Office of Communications; 
telephone: (202) 693-1999; email: [email protected].
    Technical inquiries: Natalia Stakhiv, Directorate of Standards and 
Guidance; telephone: (202) 693-2272; email: [email protected].

SUPPLEMENTARY INFORMATION: 

Table of Contents

I. Background
II. Summary and Explanation of the Final Rule
III. Procedural Determinations

I. Background

    Appendix A of OSHA's Respiratory Protection Standard (29 CFR 
1910.134) currently contains four quantitative fit testing protocols: 
Generated aerosol; ambient aerosol condensation nuclei counter (CNC); 
controlled negative pressure (CNP); and controlled negative pressure 
REDON. TSI Incorporated (``TSI'') proposed the ambient aerosol CNC 
protocol--often called the PortaCount[supreg] protocol after the CNC 
instrument manufactured by TSI--in 1987. OSHA allowed the ambient 
aerosol CNC protocol for fit testing under a compliance interpretation 
published in 1988. OSHA eventually incorporated that protocol into 
appendix A of the Respiratory Protection Standard when it revised the 
standard in 1998.
    In 2006, TSI submitted two additional quantitative fit testing 
protocols to OSHA for approval and inclusion in appendix A of the 
Respiratory Protection Standard. These protocols were modified, 
abbreviated versions of the original ambient aerosol CNC protocol 
already approved by OSHA and listed in appendix A. OSHA published a 
notice of proposed rulemaking (NPRM) on January 21, 2009 (74 FR 3526) 
to include the two protocols in its Respiratory Protection Standard, 
but later concluded that they were not sufficiently accurate or 
reliable. OSHA withdrew the proposed rule without prejudice on January 
27, 2010 (75 FR 4323), and invited the developers to resubmit the two 
protocols after addressing the issues of concern listed in the 
withdrawal notification. In 2014, TSI submitted three new quantitative 
fit testing protocols for OSHA approval. These three protocols also 
were modified, abbreviated versions of the original ambient aerosol CNC 
protocol, but different from the two protocols TSI submitted to OSHA in 
2006.
    Part II of appendix A of OSHA's Respiratory Protection Standard 
specifies the procedure for adding new fit testing protocols to the 
standard. Under that procedure, if OSHA receives an application for a 
new fit testing protocol meeting certain criteria, it must commence a 
rulemaking proceeding to consider adopting the proposed protocol. These 
criteria are: (1) A test report prepared by an independent government 
research laboratory (e.g., Lawrence Livermore National Laboratory, Los 
Alamos National Laboratory, the National Institute for Standards and 
Technology) stating that the laboratory tested the protocol and found 
it to be accurate and reliable; or (2) an article published in a peer-
reviewed industrial hygiene journal describing the protocol and 
explaining how the test data support the protocol's accuracy and 
reliability. TSI's 2014 application for approval of three new 
quantitative fit testing protocols met the second criterion. OSHA 
considers such proposals under the notice-and-comment rulemaking 
procedures specified in Section 6(b)(7) of the Occupational Safety and 
Health Act of 1970 (the ``Act'') (29 U.S.C. 655(b)(7)).

II. Summary and Explanation of the Final Rule

A. Proposed Rulemaking

    In July 2014, TSI submitted an application requesting that OSHA 
approve three new quantitative fit testing protocols for inclusion in 
appendix A of OSHA's Respiratory Protection Standard (OSHA-2015-0015-
0003). These three protocols were modified, abbreviated versions of the 
original ambient aerosol CNC protocol approved by OSHA and listed in 
appendix A, but different from the ones submitted to OSHA by TSI in 
2006. TSI's application included three peer-reviewed articles (``the 
Richardson studies'') describing the accuracy and reliability of TSI's 
proposed protocols.\1\ The application letter also included a copy of 
the 2010 ANSI/AIHA (American National Standards Institute/American 
Industrial Hygiene Association) Z88.10 ``Respirator Fit Testing 
Methods'' standard (``the ANSI standard''), which contains ``Annex A2: 
Criteria for Evaluating New Fit Test Methods'' (``the ANSI annex'') 
(OSHA-2015-0015-0007). TSI also submitted two white papers: One 
describing TSI's analysis of its talking exercise data and the second 
describing TSI's process and rationale behind the fit test exercises 
that were employed in the Richardson studies (OSHA-2015-0015-0001, 
OSHA-2015-0015-0008). OSHA determined that the information submitted in 
TSI's application met the criteria required for initiating a rulemaking 
to determine whether OSHA should approve the new protocols and add them 
to appendix A of the Respiratory Protection Standard. OSHA issued a 
notice of proposed rulemaking (NPRM) on October 7, 2016, proposing to 
add the new protocols and inviting public comments.
---------------------------------------------------------------------------

    \1\ Richardson, A.W. et al. (2014a), ``Evaluation of a Faster 
Fit Testing Method for Elastomeric Half-Mask Respirators Based on 
the TSI PortaCount,'' Journal of the International Society for 
Respiratory Protection 31(1): 9-22 (OSHA-2015-0015-0004); 
Richardson, A.W. et al. (2013), ``Evaluation of a Faster Fit Testing 
Method for Full-Facepiece Respirators Based on the TSI PortaCount,'' 
Journal of the International Society for Respiratory Protection 
30(2): 116-128 (OSHA-2015-0015-0005); Richardson, A.W. et al. 
(2014b), ``Evaluation of a Faster Fit Testing Method for Filtering 
Facepiece Respirators Based on the TSI PortaCount,'' Journal of the 
International Society for Respiratory Protection 31(1): 43-56 (OSHA-
2015-0015-0006).
---------------------------------------------------------------------------

    The three new protocols submitted by TSI in July 2014 included one 
for full-facepiece elastomeric respirators (the Fast-Full method), one 
for half-mask elastomeric respirators (the Fast-Half method), and one 
for filtering facepiece respirators (FFRs) (the Fast-FFR method). The 
authors of the Richardson

[[Page 50741]]

studies evaluated each of the three types of respirators for method 
performance separately, but the protocols for the Fast-Full and Fast-
Half methods were identical. As such, and to prevent duplicative 
regulatory text, OSHA proposed to consolidate the Fast-Full and Fast-
Half methods into a single protocol for approval: The modified ambient 
aerosol condensation nuclei counter (CNC) quantitative fit testing 
protocol for full-facepiece and half-mask elastomeric respirators. OSHA 
further proposed to approve the Fast-FFR protocol as the modified 
ambient aerosol condensation nuclei counter (CNC) quantitative fit 
testing protocol for filtering facepiece respirators. No commenters 
objected to the consolidation and naming of the protocols during the 
public comment period.
    The original ambient aerosol CNC protocol consists of eight test 
exercises, performed in the following order: Normal breathing, deep 
breathing, turning head side-to-side, moving head up-and-down, talking, 
grimace, bending over, and normal breathing again. The modified ambient 
aerosol CNC protocol for full-facepiece and half-mask elastomeric 
respirators differs as follows: (1) It includes only three of the eight 
original test exercises (bending over, head side-to-side, and head up-
and-down); (2) it adds jogging-in-place as a new exercise; and (3) it 
reduces the total test duration from 7.2 to 2.5 minutes. The modified 
ambient aerosol CNC protocol for FFRs differs from the original ambient 
aerosol CNC protocol as follows: (1) It includes only four of the eight 
original test exercises (bending over, talking, head side-to-side, and 
head up-and-down) and (2) it reduces the total test duration from 7.2 
to 2.5 minutes.
    The three Richardson studies (OSHA-2015-0015-0004, OSHA-2015-0015-
0005, OSHA-2015-0015-0006) compared the fit factors for the new 
protocols to a reference method based on the approach specified in the 
ANSI annex.\2\ This approach requires the performance evaluation study 
to administer sequential paired tests using the proposed fit testing 
method and reference method during the same respirator donning. The 
reference method consisted of the standard OSHA exercises listed in 
Section I.A.14 of appendix A of the Respiratory Protection Standard 
(which are also the eight test exercises used for the original ambient 
aerosol CNC protocol), minus the grimace exercise, in the same order as 
described in the standard (i.e., normal breathing, deep breathing, head 
side-to-side, head up-and-down, talking, bending over, normal 
breathing). Each exercise was performed for 60 seconds.
---------------------------------------------------------------------------

    \2\ A fit factor is a quantitative estimate of the fit of a 
particular respirator to a specific individual, and typically 
estimates the ratio of the concentration of a substance in ambient 
air to its concentration inside the respirator when worn.
---------------------------------------------------------------------------

    These protocols will serve as alternatives to the four existing 
quantitative fit testing protocols already listed in appendix A of the 
Respiratory Protection Standard and will maintain safety and health 
protections for workers while providing additional flexibility and 
reducing compliance burdens. This rule is a deregulatory action under 
Executive Order 13771 (82 FR 9339 (January 30, 2017)). It has 
annualized net cost savings estimated at $4.1 million. A detailed 
discussion of OSHA's estimates of the rule's benefits, costs, and cost 
savings is included in the Final Economic Analysis and Regulatory 
Flexibility Certification section.

B. Articles Supporting New Fit Testing Protocols

    TSI supported its application for adding the new protocols with the 
three Richardson studies that indicate respectively that the proposed 
Fast-Half, Fast-Full, and Fast-FFR methods can identify poorly fitting 
respirators as well as the reference method used. Each article 
described a study that compared fit test results using a reference 
method specified in the ANSI annex with results using one of the 
proposed methods. The following subsections detail the methodologies 
and findings of the three Richardson studies.
1. Evaluation of the Fast-Half Method
a. Study Methods
    The first Richardson study evaluated the Fast-Half method.\3\ The 
study authors selected three models of NIOSH-approved, half-mask air-
purifying respirators--each available in three sizes--from ``leading 
U.S. mask manufacturers'' equipped with P100 filters.\4\ Respirators 
were probed with a flush sampling probe located between the nose and 
mouth. The study included 9 female and 16 male participants.
---------------------------------------------------------------------------

    \3\ Richardson, A.W. et al. (2014a), ``Evaluation of a Faster 
Fit Testing Method for Elastomeric Half-Mask Respirators Based on 
the TSI PortaCount,'' Journal of the International Society for 
Respiratory Protection 31(1): 9-22 (OSHA-2015-0015-0004).
    \4\ The authors chose not to identify the specific respirator 
models ``because the intentional mis-sizing and lack of performing a 
user seal check would misrepresent performance of these respirators 
when used as part of a proper respiratory protection program'' 
(OSHA-2015-0015-0004).
---------------------------------------------------------------------------

    Each test subject donned a respirator for a five-minute comfort 
assessment and then performed two sets of fit test exercises, one using 
the reference method and another the Fast-Half method. The study 
authors randomized the order of the two sets of fit test exercises for 
each test subject. The reference method consisted of the eight standard 
OSHA exercises listed in Section I.A.14 of appendix A of the 
Respiratory Protection Standard, minus the grimace exercise, in the 
same order as required in the standard (i.e., normal breathing, deep 
breathing, head side-to-side, head up-and-down, talking, bending over, 
normal breathing). The study subject performed each exercise for 60 
seconds.
    The study authors explained that they decided to exclude the 
grimace exercise because it ``is intended to break the respirator seal 
to the face'' which ``potentially results in a shift of the 
respirator'' (OSHA-2015-0015-0004). TSI submitted an additional 
explanation as to why the grimace exercise was excluded in all three 
Richardson studies (OSHA-2015-0015-0008). According to TSI, ``[l]ittle 
or no support was found for the grimace exercise among respirator fit 
test experts,'' and ``[t]he most common fault expressed by a number of 
experienced fit testers and industry experts was that the grimace 
cannot be consistently applied or even defined'' (Id.). TSI further 
explained that the grimace exercise is intended to break the face seal, 
which may not reseal in the same way for subsequent exercises. As a 
result, the shift in the respirator caused by grimacing can potentially 
confound comparisons between the fit test methods. TSI finally noted 
that the fit factor from the grimace exercise (if measured) is not used 
to calculate the overall fit factor result under the original ambient 
aerosol CNC method.
    The Fast-Half method included four exercises: Bending, jogging-in-
place, head side-to-side, and head up-and-down. Each test subject took 
two breaths at each extreme of the head side-to-side and head up-and-
down exercises and at the bottom of the bend in the bending exercise.
    Although not discussed in the Richardson study, TSI explained its 
rationale for selecting the exercises that were later utilized in the 
three Richardson studies. The exercises were identified, by TSI, as 
being the most rigorous for (i.e., the best at) identifying poor 
fitting respirators in two white papers TSI prepared and submitted to 
OSHA (OSHA-2015-0015-0001, OSHA-2015-0015-0008). TSI reached its 
conclusions and selected the exercises based on a literature review, 
informal conversations with industry fit test experts, and in-house 
pilot studies.

[[Page 50742]]

``Talking out loud,'' ``bending,'' and ``moving head up/down'' were 
determined to be the three most critical exercises in determining the 
overall fit factor for abbreviated respirator fit test methods by 
Zhuang et al. (OSHA-2015-0015-0011).\5\ TSI's in-house pilot fit 
testing studies supported the conclusions made by Zhuang et al., 
however, additional analysis of the TSI data by TSI uncovered an 
unexpected trend within the data for the talking exercise (OSHA-2015-
0015-0001, OSHA-2015-0015-0008). TSI collected fit test data on 
subjects using consecutive sets of the seven-exercise reference method 
described above. TSI analyzed the frequency with which each exercise 
produced the lowest fit factor. Fit test data were separated into three 
groups: All fit tests, good-fitting fit tests, and poor-fitting fit 
tests. A poor-fitting fit test was defined as any test where at least 
one exercise failed, and a good-fitting fit test was defined as one 
where no exercises failed.\6\ TSI's results showed that normal 
breathing, deep breathing, and talking rarely produced the lowest fit 
factor (frequency <=3 percent) for poor-fitting full-facepiece 
respirators. On this basis, these three less rigorous exercises were 
eliminated by TSI for both the Fast-Full and Fast-Half methods. The 
bending exercise was the most rigorous exercise for poor-fitting full-
facepiece and half-mask elastomeric respirators. Talking was the 
exercise among the seven exercises that most often had the lowest fit 
factor for good-fitting full-facepiece and half-mask respirators in the 
TSI pilot study. None of the other exercises stood out for half-mask 
respirators, but TSI reasoned that there was a lack of data suggesting 
that half-mask respirator fit tests should use different exercises than 
full-facepiece respirators (OSHA-2015-0015-0008). TSI added jogging-in-
place for a fourth rigorous test exercise as part of the protocol that 
the Richardson authors would evaluate, reasoning that jogging 
``leverages the weight of the facepiece, much like bending, but on a 
different axis, and also because both OSHA and ANSI currently include 
jogging as an alternative exercise'' (OSHA-2015-0015-0008). Jogging-in-
place is an alternate (i.e., elective as opposed to required) exercise 
in the ANSI annex. The study authors stated that jogging is 
``aggressive in terms of evaluating the respirator seal'' (OSHA-2015-
0015-0004).
---------------------------------------------------------------------------

    \5\ Zhuang et al. (2004) considered those exercises that had the 
lowest fit factors as the most critical in determining the overall 
fit factor.
    \6\ Pass/fail levels were 500 for full-facepiece respirators and 
100 for half-mask elastomeric respirators and FFRs.
---------------------------------------------------------------------------

    The study authors conducted the experiments in a large chamber and 
added sodium chloride (NaCl) aerosol to augment particle 
concentrations, which they expected to range between 5,000 and 20,000 
particles/cm\3\ (target = 10,000 p/cm\3\). The study authors used a 
single CPC instrument, the PortaCount[supreg] Model 8030 (TSI 
Incorporated, Shoreview MN), for sampling and valuation. They connected 
the instrument to two equal-length sampling tubes. The first tube 
sampled particle concentrations inside the facepiece, and the second 
tube sampled the ambient particle concentration. The authors used TSI 
software to switch between sampling lines and to record concentration 
data.
    During the reference method, for each exercise, the ambient 
sampling tube was first purged for four seconds before an ambient 
sample was taken for 5 seconds, followed by an 11-second purge of the 
in-facepiece sampling tube and a 40-second in-facepiece sample. The 
reference method took a total of 429 seconds (7 minutes 9 seconds) to 
complete.
    For the reference method, the authors calculated a fit factor for 
each exercise by dividing the mean ambient concentration for that 
exercise by the in-facepiece concentration taken during each exercise 
(average of the five-second ambient measurements before and after the 
exercise). The harmonic mean of the seven exercise fit factors equaled 
the overall fit factor. During the first exercise of the Fast-Half 
method (bending over), the ambient sampling tube was first purged for 4 
seconds before an ambient sample was taken for five seconds; the in-
facepiece sampling tube was then purged for 11 seconds and a sample was 
then taken from inside the mask for 30 seconds. No ambient sample was 
taken during the next two exercises (jogging and head side-to-side)--
just one 30-second in-facepiece sample was collected for each exercise. 
For the last exercise (head up-and-down), a 30-second in-facepiece 
sample was taken, after which a 4-second ambient purge and 5-second 
ambient sample were conducted. The Fast-Half method took a total of 149 
seconds (2 minutes 29 seconds) to complete.
    For the Fast-Half method, the ambient concentration was calculated 
by taking the mean of two measurements--one before the first exercise 
and one after the last exercise. The authors calculated fit factors for 
each exercise by dividing the in-facepiece concentration taken during 
that exercise by the mean ambient concentration. As with the reference 
method, the harmonic mean of the four exercise fit factors represented 
the overall fit factor. A minimum fit factor of 100 is required in 
order to be regarded as an acceptable fit for half-mask respirators 
under appendix A of the Respiratory Protection Standard.
    To ensure that respirator fit was not significantly altered between 
the two sets of exercises, a 5-second normal breathing fit factor 
assessment was included before the first exercise set, between the two 
sets of exercises and at the completion of the second exercise set. If 
the ratio of the maximum to minimum of these three fit factors was 
greater than 100, this experimental trial was excluded from data 
analysis.
b. Richardson Study Results
    The ANSI annex specifies that an exclusion zone within one 
coefficient of variation for the reference method must be determined. 
The exclusion zone is the range of measured fit factors around the 
pass/fail fit factor of 100 that cannot be confirmed to be greater than 
100 or less than 100 with adequate confidence and, therefore, should 
not be included in evaluating performance. The study authors determined 
the variability associated with the reference method using 48 pairs of 
fit factors from 16 participants. They defined the exclusion zone as 
fit factor measurements within one standard deviation of the 100 pass/
fail value. Six pairs of fit factors were omitted by the study authors 
because the normal breathing fit factor ratio exceeded 100 and 5 pairs 
of fit factors were omitted because they were identified as outliers (> 
3 standard deviations from the mean of the remaining data points). The 
exclusion zone calculated by the study authors ranged from 82 to 123 
and did not include the five outliers. During review of the study 
methods, OSHA felt that omitting outliers to define a variability-based 
exclusion zone deviated from the usual scientific practice. Therefore, 
OSHA recalculated the exclusion zone with the outlier data included in 
the analysis (OSHA-2015-0015-0009). The recalculated exclusion zone was 
somewhat wider, ranging from 68 to 146.
    The final dataset for the ANSI Fast-Half performance evaluation 
included 134 pairs of fit factors from 25 participants. The respirator 
models and sizes were used in nearly equal proportion. The study 
authors omitted eleven pairs of fit factors because the ratio of 
maximum to minimum normal breathing fit factors was greater than 100. 
They also omitted one pair due to a methodological error (sample line

[[Page 50743]]

detached from respirator during test). As such, 122 pairs were included 
in the data analysis.
    The study authors concluded that their statistical analysis 
indicates that the Fast-Half method met the required acceptance 
criteria for test sensitivity, predictive value of a pass, predictive 
value of a fail, test specificity, and kappa statistic \7\ as defined 
in the ANSI annex (see Table 1). The same was indicated by OSHA's 
statistical analysis, utilizing the wider OSHA-recalculated exclusion 
zone, which excluded an additional three pairs for a total of nine 
pairs excluded and 119 pairs included in the analysis. OSHA therefore 
agrees with the study authors that the Fast-Half method can identify 
poorly fitting respirators at least as well as the reference method.
---------------------------------------------------------------------------

    \7\ The kappa statistic is a measure of agreement between the 
proposed and reference fit test methods. It compares the observed 
proportion of fit tests that are concordant with the proportion 
expected if the two tests were statistically independent. Kappa 
values can vary from -1 to +1. Values close to +1 indicate good 
agreement. ANSI/AIHA recommends kappa values >0.70.
---------------------------------------------------------------------------

2. Evaluation of Fast-Full Method
a. Study Methods
    The second Richardson study evaluated the Fast-Half method.\8\ The 
study authors selected three models of NIOSH-approved, full-facepiece 
air-purifying respirators from ``leading U.S. mask manufacturers'' 
equipped with P100 filters. Each model was available in three sizes. 
Respirators were probed with a non-flush sampling probe inside the nose 
cup, extending 0.6 cm into the breathing zone. The study included 11 
female and 16 male participants. The reference method, choice of 
exercises, PortaCount[supreg] instrument, test aerosol, and sampling 
sequence were identical to those used for the Fast-Half method. 
Appendix A of the Respiratory Protection Standard requires a minimum 
fit factor of 500 for full-facepiece respirators.
---------------------------------------------------------------------------

    \8\ Richardson, A.W. et al. (2013), ``Evaluation of a Faster Fit 
Testing Method for Full-Facepiece Respirators Based on the TSI 
PortaCount,'' Journal of the International Society for Respiratory 
Protection 30(2): 116-128 (OSHA-2015-0015-0005).
---------------------------------------------------------------------------

b. Richardson Study Results
    The study authors determined the variability associated with the 
reference method using 54 pairs of fit factors from 17 participants. 
The exclusion zone was defined as fit factor measurements within one 
standard deviation of the 500 pass/fail value. Five pairs of fit 
factors were omitted because the normal breathing fit factor ratio 
exceeded 100, and three pairs of fit factors were omitted because they 
were identified as outliers (> 3 standard deviations from the mean of 
the remaining data points). The exclusion zone calculated by the study 
authors ranged from 345 to 726 and did not include the three outliers. 
OSHA recalculated the exclusion zone with the outlier data included in 
the analysis (OSHA-2015-0015-0009). The recalculated exclusion zone 
determined by OSHA was somewhat wider ranging from 321 to 780.
    The final dataset for the ANSI Fast-Full performance evaluation 
included 148 pairs of fit factors from 27 participants. The respirator 
models and sizes were used in nearly equal proportion. Eleven pairs 
were omitted because the ratio of maximum to minimum normal breathing 
fit factors was greater than 100; one pair was omitted due to an 
observational anomaly (a torn piece of a cleaning wipe was observed in 
the respirator during the test); 136 pairs were included in the data 
analysis.
    The study authors concluded that their statistical analysis 
indicates that the Fast-Full method met the required acceptance 
criteria for test sensitivity, predictive value of a pass, predictive 
value of a fail, test specificity, and kappa statistic as defined in 
the ANSI annex (see Table 1). The same was indicated by OSHA's 
statistical analysis, utilizing the wider OSHA-recalculated exclusion 
zone, which excluded an additional three pairs for a total of 15 pairs 
excluded and 133 pairs included in the analysis. OSHA therefore agrees 
with the study authors that the Fast-Full method can identify poorly 
fitting respirators at least as well as the reference method.
3. Evaluation of Fast-FFR (Filtering Facepiece Respirator) Method
a. Study Methods
    The third Richardson article evaluated the Fast-FFR method.\9\ Ten 
models of NIOSH-approved N95 FFRs from six ``leading U.S. mask 
manufacturers'' were selected for study.\10\ The different models were 
selected to represent a range of styles: six cup-shaped, two horizontal 
flat-fold, and two vertical flat-fold models. No information was 
provided in the publication about whether models were available in 
different sizes. However, at OSHA's request, TSI submitted the 
following additional information regarding the choice of respirators 
(OSHA-2015-0015-0010):
---------------------------------------------------------------------------

    \9\ Richardson, A.W. et al. (2014b), ``Evaluation of a Faster 
Fit Testing Method for Filtering Facepiece Respirators Based on the 
TSI PortaCount,'' Journal of the International Society for 
Respiratory Protection 31(1): 43-56 (OSHA-2015-0015-0006).
    \10\ The authors chose not to identify the specific respirator 
models ``because the intentional mis-sizing and lack of performing a 
user seal check would misrepresent performance of these respirators 
when used as part of a proper respiratory protection program'' 
(OSHA-2015-0015-0006).

    The study plan for FFR called for 10 N95 FFR. Unlike elastomeric 
respirators, FFR designs vary widely and are typically not offered 
in different sizes. The authors felt it was important to use a 
variety of designs that represent the styles currently available in 
the US. Of the 10 models used, 6 were cup-shaped, 2 were vertical-
fold, and 2 were horizontal-fold designs. The cup-shaped style is by 
far the most common, which is why 6 of the 10 model selected have 
that fundamental design. Four flat-fold designs (2 vertical-fold and 
---------------------------------------------------------------------------
2 horizontal-fold) models are also included.

    Respirators were probed with a flush sampling probe located between 
the nose and mouth. Lightweight sample tubing and neck straps were used 
to ensure the tubing did not interfere with respirator fit. Twenty-nine 
participants (11 female; 18 male) were included in the study. The 
reference method, test aerosol, and most other study procedures were 
analogous to those used for the Fast-Half and Fast-Full methods. 
However, the Fast-FFR method employed these four exercises: Bending, 
talking, head side-to-side, and head up-and-down with the same sampling 
sequence and durations as the other test protocols. The talking 
exercise replaces the jogging exercise used in the Fast-Half and Fast-
Full methods. TSI decided not to eliminate the talking exercise for 
FFRs even though their pilot study indicated that it rarely produces 
the lowest fit factor (OSHA-2015-0015-0008). They felt from their own 
experience that jogging does not represent the kind of motions that FFR 
wearers do when using the respirator (OSHA-2015-0015-0008). TSI also 
indicated that the sampling probe configured on lightweight FFR 
respirators caused the respirator to pull down and away from the face 
during jogging creating unintentional leakage. A PortaCount[supreg] 
Model 8038 operated in the N95 mode (TSI Inc., Shoreview MN) was used 
to measure aerosol concentrations throughout the experiments. The 
particle concentrations in the test chamber were expected to be greater 
than 400 p/cm\3\. A minimum fit factor of 100 is required in order to 
be regarded as an acceptable fit for these types of respirators under 
appendix A of the Respiratory Protection Standard.
b. Richardson Study Results
    The study administered sequential paired fit tests using the Fast-
FFR

[[Page 50744]]

method and a reference method according to the ANSI annex. The study 
authors randomized the order of the two sets of fit test exercises for 
each test subject. The study authors determined the variability 
associated with the reference method using 63 pairs of fit factors from 
14 participants. They defined the exclusion zone as fit factor 
measurements within one standard deviation of the 100 pass/fail value. 
Two pairs of fit factors were omitted by the study authors because the 
normal breathing fit factor ratio exceeded 100, and six pairs of fit 
factors were omitted because they were identified as outliers (>3 
standard deviations from the mean of the remaining data points). The 
exclusion zone calculated by the study authors ranged from 78 to 128 
and did not include the six outliers. OSHA recalculated the exclusion 
zone with the outlier data included in the analysis (OSHA-2015-0015-
0009). The recalculated exclusion zone was somewhat wider, ranging from 
69 to 144.
    The final dataset for the ANSI Fast-FFR performance evaluation 
included 114 pairs of fit factors from 29 participants. The respirator 
models were used in nearly equal proportion. The authors omitted two 
pairs because the ratio of maximum to minimum normal breathing fit 
factors was greater than 100, leaving 112 pairs for the data analysis.
    The study authors concluded that their statistical analysis 
indicates that the Fast-FFR method met the required acceptance criteria 
for test sensitivity, predictive value of a pass, predictive value of a 
fail, test specificity, and kappa statistic as defined in the ANSI 
annex (see Table 1). The same was found by OSHA's statistical analysis, 
utilizing the wider OSHA-recalculated exclusion zone, which excluded an 
additional four pairs for a total of 11 pairs excluded and a 102 pairs 
included in the analysis. OSHA therefore agrees with the study that the 
Fast-FFR method can identify poorly fitting respirators at least as 
well as the reference method.

                          Table 1--Comparison of Fit Test Protocols With ANSI Criteria
----------------------------------------------------------------------------------------------------------------
                                                    ANSI Z88.10      Fast-full       Fast-half       Fast-FFR
----------------------------------------------------------------------------------------------------------------
Sensitivity.....................................         >=0.959            0.98            0.96            1.00
PV Pass.........................................          >=0.95            0.98            0.97            1.00
Specificity.....................................          >=0.50            0.98            0.97            0.85
PV Fail.........................................          >=0.50            0.98            0.93            0.93
Kappa...........................................          >=0.70            0.97        \1\ 0.89        \1\ 0.89
----------------------------------------------------------------------------------------------------------------
\1\ The kappa values in the table are those determined using the OSHA recalculated exclusion zone. The kappa
  values reported by the study authors using a narrower exclusion zone were 0.90 and 0.87, respectively, for the
  Fast-Half and Fast-FFR methods.
Other statistical values were the same for both OSHA and study author exclusion zone determinations.

C. Consensus Standards

    While appendix A of OSHA's Respiratory Protection Standard 
specifies the procedure for adding new fit testing protocols to the 
standard, it does not specify any particular method(s) or criteria for 
evaluating a new fit test. Section 6(a) of the Act directs the 
Secretary of Labor to promulgate by rule ``as an occupational safety or 
health standard any national consensus standard . . . unless he 
determines that the promulgation of such a standard would not result in 
improved safety or health for specifically designated employees.'' 29 
U.S.C. 655(a). Section 6(b)(8) of the Act further states: ``Whenever a 
rule promulgated by the Secretary differs substantially from an 
existing national consensus standard, the Secretary shall, at the same 
time, publish in the Federal Register a statement of the reasons why 
the rule as adopted will better effectuate the purposes of this Act 
than the national consensus standard.'' 29 U.S.C. 655(b)(8). And OSHA 
has a long history of considering national safety and health consensus 
standards, such as ANSI and NFPA (National Fire Protection 
Association), in developing its own standards.
    The National Technology Transfer and Advancement Act of 1995 
similarly endorses agencies' use of national consensus standards: ``all 
Federal agencies and departments shall use technical standards that are 
developed or adopted by voluntary consensus standards bodies, using 
such technical standards as a means to carry out policy objectives or 
activities determined by the agencies and departments.'' Public Law 
104-113, section 12(d), 110 Stat. 775, 783 (1996), 15 U.S.C. 272 note. 
ANSI/AIHA is such a voluntary consensus standards body, whose 
standards, including Z88.10, are widely recognized and accepted by the 
industrial hygiene community. OSHA concurs with ANSI that ``this annex 
[A2] provides a specific procedure for evaluating fit test methods 
against the current body of knowledge.'' OSHA therefore considers the 
annex's procedure to be a valid, acceptable method for evaluating new 
fit test protocols (ANSI/AIHA, 2010).

D. Comments to the Proposal

    In the October 2016 NPRM, OSHA preliminarily determined that the 
new protocols met the sensitivity, specificity, predictive value, and 
other criteria outlined in the ANSI annex and would, therefore, provide 
employees with at least as much protection as the reference method. 
That reference method consisted of the standard OSHA exercises listed 
in Section I.A.14 of appendix A of the Respiratory Protection Standard 
(which are the eight test exercises used for the original ambient 
aerosol CNC protocol), minus the grimace exercise, in the same order as 
described in the standard (i.e., normal breathing, deep breathing, head 
side-to-side, head up-and-down, talking, bending over, normal 
breathing). OSHA further concluded that it was reasonable to remove the 
grimace exercise from the reference method during the method comparison 
testing, because its inclusion would unpredictably impact respirator 
fit (see Question #10 below for a more detailed discussion). After 
having considered the comments submitted in response to the NPRM 
(discussed below), OSHA has concluded that it is appropriate to amend 
appendix A of the Respiratory Protection standard to include the 
proposed fit test protocols.
    In the NPRM, OSHA invited public comment on the accuracy and 
reliability of the proposed protocols, their effectiveness in detecting 
respirator leakage, and their usefulness in selecting respirators that 
will protect employees from airborne contaminants in the workplace. 
OSHA invited public comment on the following specific questions:
    1. Were the three studies described in the peer-reviewed journal 
articles well controlled and conducted according to

[[Page 50745]]

accepted experimental design practices and principles?
    2. Were the results of the three studies described in the peer-
reviewed journal articles properly, fully, and fairly presented and 
interpreted?
    3. Did the three studies treat outliers appropriately in 
determination of the exclusion zone?
    4. Will the two proposed protocols generate reproducible fit 
testing results?
    5. Will the two proposed protocols reliably identify respirators 
with unacceptable fit as effectively as the quantitative fit testing 
protocols, including the OSHA-approved standard PortaCount[supreg] 
protocol, already listed in appendix A of the Respiratory Protection 
Standard?
    6. Did the protocols in the three studies meet the sensitivity, 
specificity, predictive value, and other criteria contained in the 
ANSI/AIHA Z88.10-2010, Annex A2, Criteria for Evaluating Fit Test 
Methods?
    7. Are the specific respirators selected in the three studies 
described in the peer-reviewed journal articles representative of the 
respirators used in the United States?
    8. Does the elimination of certain fit test exercises (e.g., normal 
breathing, deep breathing, talking) required by the existing OSHA-
approved standard PortaCount[supreg] protocol impact the acceptability 
of the proposed protocols?
    9. Is the test exercise, jogging-in-place, that has been added to 
the Fast-Full and Fast-Half protocols appropriately selected and 
adequately explained? Should the jogging exercise also be employed for 
the Fast-FFR protocol? Is the reasoning for not replacing the talking 
exercise with the more rigorous jogging exercise in the Fast-FFR 
protocol (as was done in Fast-Full and Fast-Half) adequately explained?
    10. Was it acceptable to omit the grimace from the reference method 
employed in the studies evaluating performance of the proposed fit 
testing protocols? Is it appropriate to exclude the grimace completely 
from the proposed protocols, given that it is not used in the 
calculation of the fit factor result specified under the existing or 
proposed test methods? If not, what other criteria could be used to 
assess its inclusion or exclusion?
    11. The protocols in the three studies specify that participants 
take two deep breaths at the extreme of the head side-to-side and head 
up-and-down exercises and at the bottom of the bend in the bend-forward 
exercise. According to the developers of these protocols, the deep 
breaths are included to make the exercises more rigorous and 
reproducible from one subject to the next. Are these additional 
breathing instructions adequately explained in the studies and in the 
proposed amendment to the standard? Are they reasonable and 
appropriate?
    12. Does OSHA's proposed regulatory text for the two new protocols 
offer clear instructions for implementing the protocols accurately?
    OSHA received 27 comments from 25 separate individuals, with one 
individual submitting three separate comments (OSHA-2015-0015-0015 to 
OSHA-2015-0015-0042). In addition, TSI submitted a comment several 
months after the close of the comment period (OSHA-2015-0015-0047). 
OSHA added TSI's comment to the docket as a late submission in the 
interest of full disclosure but did not take it into account.
    Of the 27 timely comments, six did not specifically address any of 
OSHA's specific questions, but were generally in favor of the proposed 
protocols (OSHA-2015-0015-0016, OSHA-2015-0015-0018, OSHA-2015-0015-
0019, OSHA-2015-0015-0020, OSHA-2015-0015-0030, OSHA-2015-0015-0039). 
Among other things, these comments agreed that the abbreviated 
protocols would save time and resources and would increase employer 
compliance with safety and health regulations.
    OSHA addresses below the comments that addressed the NPRM's 
specific questions:
    1. Were the three studies described in the peer-reviewed journal 
articles well controlled and conducted according to accepted 
experimental design practices and principles?
    The majority of concerned comments about the proposed protocols 
related to the experimental design and methods used in the three 
Richardson studies supporting the proposed protocols. The most common 
of these criticisms was that the testing was not representative of 
``real world'' settings (OSHA-2015-0015-0022, OSHA-2015-0015-0025, 
OSHA-2015-0015-0026, OSHA-2015-0015-0027, OSHA-2015-0015-0032, OSHA-
2015-0015-0033, OSHA-2015-0015-0040, OSHA-2015-0015-0041, OSHA-2015-
0015-0042). For example, one commenter asserted that the environment of 
the test chambers used in the three Richardson studies was ``too 
controlled'' and that the studies ``did not allow for variables 
encountered by fit test providers when conducting fit testing in real 
world settings'' (OSHA-2015-0015-0026). Another commenter stated: ``In 
an uncontrolled environment many factors, including but not limited to, 
ventilation, doors being opened, and room temperature can greatly 
affect the particle count in a relatively short time'' (OSHA-2015-0015-
0040).
    Regarding these comments, OSHA would like to stress that the 
proposed protocols were evaluated using the criteria outlined in Annex 
A2 of the ANSI/AIHA Z88.10-2010 standard, which does not require 
uncontrolled testing conditions with variables such as fluctuating 
climate, temperature, elevation, air currents, ventilation, etc. OSHA 
considers the ANSI annex method to be a valid method for evaluating new 
fit test protocols.
    Many of these comments related specifically to the use of generated 
aerosols in the three Richardson studies (OSHA-2015-0015-0022, OSHA-
2015-0015-0026, OSHA-2015-0015-0033, OSHA-2015-0015-0041). For example, 
one commenter stated:

    The PortaCount[supreg] was designed and marketed to be used for 
conducting quantitative fit tests using room aerosols, whereas the 
supporting studies were conducted in a test chamber using a 
generated aerosol. Concentrations of room aerosols are typically 
about 1x10\3\ p/cc, whereas in these studies the average challenge 
concentrations were about 2x10\4\ p/cc. . . . I would recommend that 
the protocols not be accepted until these validation tests are 
conducted using ambient aerosols. . . . (OSHA-2015-0015-0033).

Another commenter questioned why the study authors used generated 
aerosol in a test chamber when their goal was to prove the 
acceptability of a new ambient aerosol test protocol (OSHA-2015-0015-
0041).
    None of the three Richardson studies, however, employed a 
``generated aerosol'' atmosphere as described in the ANSI/AIHA Z88.10 
standard; instead, they used ``the ambient laboratory aerosol which was 
augmented by NaCl particles from a TSI Model 8026 Particle Generator'' 
(OSHA-2015-0015-0004, OSHA-2015-0015-0005, OSHA-2015-0015-0006). This 
approach is allowed by ANSI/AIHA in Annex A2, which states that ``a 
proposed modification to an accepted QNFT [quantitative fit testing] 
protocol can be evaluated using the accepted protocol for that 
instrument as the reference standard.'' As some commenters noted (OSHA-
2015-0015-0031, OSHA-2015-0015-0041), it is often necessary to augment 
the ambient environment when using the original OSHA-approved ambient 
aerosol CNC fit test method in a relatively clean office environment. 
The TSI particle generator is one of several approaches commonly used 
(OSHA-2015-0015-0051, OSHA-2015-0015-0050). In fact, as noted by one 
commenter, technicians sometimes burn candles or incense in order to 
reach and

[[Page 50746]]

maintain ambient particle counts (OSHA-2015-0015-0032). OSHA has 
concluded that there is no material difference between the experimental 
atmosphere employed in the three Richardson studies and the atmosphere 
commonly used for quantitative fit testing with the ambient aerosol CNC 
method.
    Other commenters expressed concerns that the ambient and purge 
times were too short (OSHA-2015-0015-0022, OSHA-2015-0015-0026, OSHA-
2015-0015-0027, OSHA-2015-0015-0032, OSHA-2015-0015-0033, OSHA-2015-
0015-0036, OSHA-2015-0015-0038, OSHA-2015-0015-0041, OSHA-2015-0015-
0042). For example, one commenter recommended that the proposed 
protocols ``should provide for suitable ambient and respirator purge 
durations to address the full range of particle concentrations that the 
device is recommended for use in instead of selecting a duration based 
on the optimum conditions that were selected for the studies. . . .'' 
(OSHA-2015-0015-0026). Several commenters were also concerned that each 
ambient sample conducted at the beginning and end of the new protocols 
lasted only five seconds (OSHA-2015-0015-0032, OSHA-2015-0015-0036, 
OSHA-2015-0015-0042).
    Regarding these comments, OSHA notes that for every exercise 
(except the grimace), the original OSHA-approved ambient aerosol CNC 
protocol involves a 4-second ambient purge, a 5-second ambient sample, 
and an 11-second mask purge, followed by a 40-second mask sample. A 
final 4-second ambient purge and 5-second ambient sample occur after 
the last 40-second exercise (normal breathing) mask sample. The 
proposed protocols employ the same 4-second ambient purge, 5-second 
ambient sample, and 11-second mask purge, followed by 4 consecutive 30-
second mask samples during each of the 4 exercises, and a final 4-
second ambient purge and 5-second ambient sample. The ambient purge and 
sample times are the same. The new protocols differ from the original 
OSHA-approved sampling protocol in these ways: The ambient environment 
is measured only at the beginning and end of the exercises and not 
between each exercise, mask purging occurs just once (after the first 
ambient sample), and mask sampling time is 30 seconds rather than 40 
seconds. Additionally, requirements for conducting the fit test in an 
environment with an adequate particle concentration also did not 
change; they have been standard practice for the ambient aerosol CNC 
fit test method since its inception and approval by OSHA.
    Regarding ambient measurements, the only difference between the new 
protocols and the original OSHA-approved protocol is that the new 
protocols take measurements at the beginning and end of the exercises, 
while the original protocol does so between each exercise. Because the 
total duration of the new protocols is much shorter than the original--
2.5 minutes versus 7.2 minutes--OSHA has concluded that there is no 
need to take periodic samples between exercises. In particular, the 
time between the two ambient samples in the proposed protocol is 2 
minutes 15 seconds, compared to 55 seconds between each ambient sample 
in the original protocol. This minor difference is unlikely to 
introduce any significant errors if fit testers follow standard 
practice: (1) Ensure the aerosol concentration falls between 1,000 and 
30,000 particles/cm\3\ (p/cm\3\) for filters with a NIOSH designation 
of N/R/P-99 or 100, and 30 to 1,500 p/cm\3\ for filters with a N/R/P-95 
designation; and (2) do not augment the ambient environment if the 
concentration exceeds 8000 p/cm\3\ or 800 p/cm\3\ for the 99/100 or the 
95 filters, respectively (OSHA-2015-0015-0049).
    Two commenters expressed concern over eliminating purging between 
exercises altogether (OSHA-2015-0015-0022, OSHA-2015-0015-0038). But 
there is no reason for purging between the different exercises in the 
proposed protocol because the instrument continues to sample from the 
same environment (inside the facepiece) throughout the exercises. 
Particles measured during the first few seconds of transition from one 
exercise to the next will have almost no influence on the average 
concentration over a 30-second exercise sampling period.
    Purging ensures that the sensing volume evaluates particles from 
the correct environment and is only necessary when switching between 
ambient and mask samples or vice versa. The proposed protocols do not 
switch between ambient and mask sampling during the exercises, so 
purging is not required.
    Some commenters requested further review of the methodology of the 
three Richardson studies or further validation testing by a ``third 
party'' (OSHA-2015-0015-0029, OSHA-2015-0015-0040). OSHA notes that the 
studies were conducted by a third party, Battelle Memorial Institute, 
and the study methods were approved by Battelle's Institutional Review 
Board. In addition, NIOSH stated that their ``review determined that 
the three methods met the criteria contained in the ANSI/AIHA Z88.10-
2010, Annex A2'' (OSHA-2015-0015-0031). And one commenter who had some 
concerns about the proposed protocols conceded that the ``referenced 
peer-reviewed articles in J. of Respiratory Protection appear to meet 
the mathematical and statistical criteria we expect'' (OSHA-2015-0015-
0024). Finally, the publication of the three Richardson studies in a 
peer-reviewed industrial hygiene journal suggests they were well-
controlled and conducted according to accepted experimental design 
practices and principles. In summary, OSHA determined that the public 
comments did not identify any significant shortcomings in the 
experimental design and methodology used in the three studies.
    2. Were the results of the three studies described in the peer-
reviewed journal articles properly, fully, and fairly presented and 
interpreted?
    Although critical of the fact that the studies were conducted in a 
test chamber as opposed to a real world setting, one commenter stated 
``that under the specific set of conditions that the tests were 
performed that they were presented well'' (OSHA-2015-0015-0026). But 
another commenter expressed that it was ``impossible to determine if 
the articles were properly, fully, and fairly presented and 
interpreted'' because the articles did not provide data tables listing 
``respirator make, model, style, size, individuals tested, and the 
paired results of the new test and the reference test'' as outlined in 
the ANSI annex (OSHA-2015-0015-0038). The annex recommends--but does 
not require--such tables, and it is often difficult to publish a peer-
reviewed article containing a complete dataset. Regardless, OSHA 
reviewed the full datasets provided by TSI as part of the review of the 
study protocols, and no commenters asked to see the datasets. In 
summary, OSHA finds that the public comments did not identify any 
significant shortcomings in the way that the results of the three 
journal articles were presented or interpreted.
    3. Did the three studies treat outliers appropriately in 
determination of the exclusion zone?
    While OSHA disagreed with the studies' omissions of outliers in 
calculating exclusion zones, OSHA recalculated exclusion zones with the 
outlier data included, and the results of the re-analysis did not 
change any of the studies' conclusions. In addition, NIOSH considered 
the study authors' identification of outliers to be ``a reasonable 
method for diagnosing/identifying outliers'' (OSHA-2015-

[[Page 50747]]

0015-0031). Finally, no commenters expressed concern about the 
treatment of outliers. OSHA concludes that the treatment of outliers in 
the studies did not undermine any of the studies' results or 
conclusions.
    4. Will the two proposed protocols generate reproducible fit 
testing results?
    Some commenters questioned the reproducibility of fit testing 
results using the two proposed protocols (OSHA-2015-0015-0022, OSHA-
2015-0015-0042), but did not offer any compelling data or research 
suggesting their non-reproducibility. One of these commenters had 
concerns based on NIOSH's recommendation that OSHA (OSHA-2015-0015-
0042) conduct additional research to gather evidence for a more 
informed decision. The commenter stated:

    With this recommendation OSHA should not accept a protocol that 
still needs further evidence to show it will produce reproducible 
fit testing results. There are too many respirators and employees in 
hazardous conditions to allow a protocol to move forward that isn't 
fully vetted and accurate (OSHA-2015-0015-0042).

OSHA believes this commenter took NIOSH's comment out of context. The 
NIOSH response to this question--in its entirety--is the following:

    The studies used the OSHA-accepted ambient aerosol condensation 
nuclei counter (CNC) quantitative fit testing protocol as the 
reference method. This method has been shown to produce reproducible 
fit testing results [Zhuang et al. 1998; Coffey et al. 2002]. Using 
the procedures and requirements of ANSI Z88.10-2010, the abbreviated 
methods provided results comparable to the reference method. 
Therefore, the proposed protocols are anticipated to generate 
reproducible results. NIOSH recommends that additional research be 
conducted to provide evidence for a more informed decision (OSHA-
2015-0015-0031).

    While additional research is always valuable, OSHA agrees with 
NIOSH that the proposed protocols are anticipated to generate 
reproducible results. The proposed protocols were evaluated based on 
the approach specified in the ANSI annex, which provides a specific 
procedure for evaluating fit test methods ``against the current body of 
knowledge'' and is considered a valid method by much of the industrial 
hygiene community. Having met the criteria of the ANSI annex, OSHA 
concludes that the proposed protocols will generate reproducible fit 
testing results.
    5. Will the two proposed protocols reliably identify respirators 
with unacceptable fit as effectively as the quantitative fit testing 
protocols, including the OSHA-approved standard PortaCount[supreg] 
protocol, already listed in appendix A of the Respiratory Protection 
Standard?
    Several commenters questioned not only the acceptability of the 
proposed protocols, but also the validity of the original ambient 
aerosol particle counting quantitative method already accepted by OSHA 
and listed in appendix A (OSHA-2015-0015-0022, OSHA-2015-0015-0026, 
OSHA-2015-0015-0027, OSHA-2015-0015-0029). Some of these commenters 
were also of the opinion that the CNP-based fit testing methods are 
superior to other quantitative fit testing methods. One commenter 
(OSHA-2015-0015-0042) stated that the following NIOSH ``statement 
raises major concerns to the ability & proven accuracy of this proposed 
protocol to identify respirators with unacceptable fit'':

    Evidence is not available in the literature to assess whether 
the two proposed protocols reliably identify respirators with 
unacceptable fit as effectively as the other accepted quantitative 
fit testing protocols (generated aerosol and controlled negative 
pressure (CNP)). It is recommended that further side-by-side studies 
be conducted to test the equivalency of the new PortaCount Fast-Fit 
methods in identifying poorly fitting respirators as effectively as 
the OSHA-accepted CNP testing; potentially, tests using other 
``generated aerosols'' would be needed to determine whether the 
methods are equivalent (OSHA-2015-0015-0031).

Although NIOSH recommended future research, it nonetheless recommended 
that OSHA accept the proposed protocols. In its review of the three 
Richardson studies, NIOSH also determined that the proposed protocols 
conform to the requirements of the ANSI annex.
    The validity of the original OSHA-approved ambient aerosol CNC fit 
testing protocol was never under question in this rulemaking. Appendix 
A of OSHA's Respiratory Protection Standard states that quantitative 
fit testing using ambient aerosol as the test agent and appropriate 
instrumentation (condensation nuclei counter) to quantify the 
respirator fit has ``been demonstrated to be acceptable.'' In addition, 
the members of the ANSI/AIHA Z88.10 ``Respirator Fit Testing Methods'' 
committee, who represent many of the nation's leading respiratory 
protection experts, opted to retain, rather than reject, this method as 
an acceptable quantitative fit testing method when they updated the 
national consensus standard in 2010. Furthermore, the proposed 
protocols were evaluated using the method described in the ANSI annex, 
which does not require a statistical comparison against the CNP method 
(OSHA-2015-0015-0007). Likewise, OSHA's Respiratory Protection Standard 
does not require that a new fit testing protocol be compared to the CNP 
method, or any other specific fit testing method. Moreover, just as 
OSHA does not rank specific makes and models of respirators, OSHA also 
does not rank fit testing methods. Each fit testing method has its own 
advantages and disadvantages.
    In summary, OSHA determined that the new protocols met the 
sensitivity, specificity, predictive value, and other criteria outlined 
in the ANSI annex and will, therefore, provide employees with 
protections comparable to protections afforded to them by the reference 
method, which consisted of the standard OSHA exercises listed in 
Section I.A.14 of appendix A of the Respiratory Protection Standard, 
minus the grimace exercise, in the same order as described in the 
standard (i.e., normal breathing, deep breathing, head side-to-side, 
head up-and-down, talking, bending over, normal breathing). These are 
the same test exercises, minus the grimace, that are utilized for both 
the CNC and CNP protocols. OSHA concluded that it was reasonable to 
remove the grimace exercise from the reference method during the method 
comparison testing, because its inclusion would unpredictably impact 
respirator fit within each pair of data comparing the current and new 
fit test protocols (see Question #10 below for a more detailed 
discussion).
    6. Did the protocols in the three studies meet the sensitivity, 
specificity, predictive value, and other criteria contained in the 
ANSI/AIHA Z88.10-2010, Annex A2, Criteria for Evaluating Fit Test 
Methods?
    One commenter stated that evaluating the sensitivity of the new 
protocols ``presents a quandary because the sensitivity of the standard 
PortaCount protocol has itself not been established'' (OSHA-2015-0015-
0022). As discussed under question #5, the validity of the original 
OSHA-approved ambient aerosol CNC fit testing protocol is not at issue 
in this rulemaking.
    OSHA's evaluation of the proposed protocols determined that they 
met the criteria outlined in the ANSI annex (see sections A-B above). 
In addition, NIOSH stated that their ``review determined that the three 
methods met the criteria contained in the ANSI/AIHA Z88.10-2010, Annex 
A2'' (OSHA-2015-0015-0031). Another commenter agreed that ``the 
submitted request has followed the defined procedures and the results 
fit within the statistical limits set forth in ANSI Z88.10-2010'' 
(OSHA-2015-0015-0035). Furthermore, OSHA

[[Page 50748]]

determined that the public comments did not provide any substantive 
data or information suggesting that the proposed protocols in the three 
studies did not meet the sensitivity, specificity, predictive value, 
and other criteria contained in the ANSI annex.
    7. Are the specific respirators selected in the three studies 
described in the peer-reviewed journal articles representative of the 
respirators used in the United States?
    One commenter questioned the ``very small sample of the wide range 
of tight sealing respirators that were used in the [studies]'' (OSHA-
2015-0015-0029), and another expressed that ``the small sample size of 
respirators chosen for testing lends itself to being less than ideal'' 
(OSHA-2015-0015-0040). However, neither commenter provided specific 
recommendations or statistical data regarding the numbers and types of 
respirators that should have been selected or why. Further, the 
industrial hygiene research community does not require a specified 
sample size of respirators to assess fit testing protocols. Finally, 
had the respirator sample size been too small to produce reliable 
results, the studies likely would not have been accepted for 
publication in a peer-reviewed journal.
    One commenter questioned why the Richardson studies included only 
filtering facepiece respirators without exhalation valves, noting that 
many users opt to wear filtering facepiece respirators with exhalation 
valves for comfort reasons (OSHA-2015-0015-0026). But an exhalation 
valve does not affect respirator fit. While the study authors did not 
explain how they selected the respirator models and designs, OSHA has 
determined that the public comments did not identify any significant 
shortcomings in respirator selection and believes that the models and 
designs selected for the three experiments were appropriately 
representative.
    8. Does the elimination of certain fit test exercises (e.g., normal 
breathing, deep breathing, talking) required by the existing OSHA-
approved standard PortaCount[supreg] protocol impact the acceptability 
of the proposed protocols?
    Several commenters expressed concern over removing certain fit test 
exercises (OSHA-2015-0015-0021, OSHA-2015-0015-0024, OSHA-2015-0015-
0025, OSHA-2015-0015-0029, OSHA-2015-0015-0032, OSHA-2015-0015-0033, 
OSHA-2015-0015-0038, OSHA-2015-0015-0041), but did not provide any 
peer-reviewed data or published research to support their opinions. 
Three commenters (OSHA-2015-0015-0021, OSHA-2015-0015-0025, OSHA-2015-
0015-0032) expressed concern about removing the talking exercise, 
because they had experienced fit test failures during the talking 
exercise when fit testing workers. Another commenter felt that ``it 
doesn't make sense to eliminate [the talking] exercise simply because 
it wasn't the worst contributing exercise with poor fitting 
respirators'' (OSHA-2015-0015-0033). A third suggested retaining the 
head side-to-side, head up-and-down, and talking exercises because he 
believes they are currently the most rigorous exercises (OSHA-2015-
0015-0024).
    Another commenter suggested that ``the conclusion to eliminate 
Normal Breathing 2 (NB2) from the Fast Full Protocol is extremely 
subjective'' and questioned how ``NB2 [normal breathing #2] could be 
eliminated and UD [moving head up and down] kept if there is no 
correlation with the study data?'' (OSHA-2015-0015-0038). This 
commenter suggested increasing the purge time to improve the ability of 
the NB2 exercise to detect poor fits. Regarding this question, OSHA has 
concluded that TSI properly excluded the second normal breathing 
exercise. In TSI's study of the Fast-Full method, the second normal 
breathing exercise had the lowest fit factor 19% of the time for poor-
fitting respirators. While this score normally indicates an exercise 
was effective at detecting poor-fitting respirators, TSI concluded that 
score was anomalous because the corresponding score for the first 
normal breathing (NB1) exercise was 0%. TSI reasoned the 19% score was 
a result of particles introduced into the facepiece during the 
preceding (bending over) exercise that were not purged (OSHA-2015-0015-
0008). Increasing the purge time to clear such particles would not, as 
the commenter suggests, improve the ability of the NB2 exercise to 
detect poor fits. Instead, NB2 would likely be as ineffective as NB1, 
which was never the lowest fit factor for any poor-fitting respirators. 
This is also supported by the fact that the NB1 and NB2 exercises 
produced the lowest fit factors only 2% and 5% of the time, 
respectively, for good-fitting respirators.
    One commenter noted that ``[e]limination of the normal breathing, 
deep breathing, and talking fit test exercises from the proposed Fast 
protocols has significant potential for adverse impact on PortaCount 
fit test results in the real world'' (OSHA-2015-0015-0022). With 
respect to normal breathing and talking, the commenter noted that 
several studies not mentioned by the three Richardson studies indicate 
that the first normal breathing exercise fit factor is typically lower 
than fit factors from all subsequent exercises and that the talking 
exercise also often results in a lower fit factor. But this commenter 
did not provide any basis to believe eliminating these exercises will 
put workers at risk. Indeed, he conceded that ``respirator donning has 
a greater effect on respirator fit than do fit test exercises'' and 
``the lower fit factors produced by the talking exercise appear to be 
more consistent with sampling artifact than with actual exercise 
dynamics.'' And, as TSI explained, fit factors for the second normal 
breathing exercise are likely to be contaminated by prior exercises 
(OSHA-2015-0015-0008). Finally, this commenter offered no data or 
published information that suggest deep breathing is more rigorous than 
other exercises or that eliminating deep breathing will put workers at 
risk.
    One commenter (OSHA-2015-0015-0029) stated that ``our experience 
strongly suggests that the Deep Breathing and Talking Exercises are 
frequently the exercises that see the lowest fit factors calculated and 
often are `THE Exercises' which determine whether a respirator wear 
will achieve a Pass or Failure following the completion of the fit test 
series of exercises.'' He further suggested ``a more thorough 
evaluation of this change by a third party such as NIOSH-NPPTL. . . .'' 
Another commenter requested that a review of the studies be performed 
by an independent third party (OSHA-2015-0015-0040). NIOSH/NPPTL did in 
fact review and evaluate the studies. In the comments NIOSH submitted 
to OSHA, NIOSH did not express any concern over the removal of the 
talking exercise and ultimately ``recommend[ed] that OSHA accept the 
three protocols'' (OSHA-2015-0015-0031).
    Regarding all these comments, the industrial hygiene community has 
not come to a consensus as to which test exercises must be used in a 
new fit testing protocol. Neither the ANSI annex nor OSHA's appendix 
requires any specific test exercise(s) be used in a new fit testing 
protocol. Further, in 2004, OSHA approved an abbreviated version of the 
CNP protocol, called the CNP REDON protocol, which excludes the deep 
breathing and talking exercises, and includes only the facing forward 
(same as normal breathing), bending over, and head shaking exercises. 
In sum, the information submitted in the public comments did not 
convince OSHA that the elimination of the deep breathing and talking 
exercises adversely impacted the acceptability of the proposed 
protocols,

[[Page 50749]]

which met the sensitivity, specificity, predictive value, and other 
criteria contained in the ANSI annex.
    9. Is the test exercise, jogging-in-place, that has been added to 
the Fast-Full and Fast-Half protocols appropriately selected and 
adequately explained? Should the jogging exercise also be employed for 
the Fast-FFR protocol? Is the reasoning for not replacing the talking 
exercise with the more rigorous jogging exercise in the Fast-FFR 
protocol (as was done in Fast-Full and Fast-Half) adequately explained?
    One commenter was of the opinion that ``[t]he jogging exercise, 
while rigorous, is not representative of real-life civilian 
activities'' (OSHA-2015-0015-0024). NIOSH stated that it would have 
liked to have seen references to support that the jogging-in-place 
exercise used in the protocols for elastomeric respirators was 
aggressive in evaluating the respirator seal. However, this did not 
prevent NIOSH from recommending that OSHA approve the proposed 
protocols (OSHA-2015-0015-0031). Furthermore, as stated above under 
question #8, the industrial hygiene community has not come to a 
consensus as to which test exercise(s) must be included in new fit 
testing protocols. More importantly, neither the ANSI annex nor OSHA's 
appendix requires that any specific test exercise(s) be used in a new 
fit testing protocol.
    10. Was it acceptable to omit the grimace from the reference method 
employed in the studies evaluating performance of the proposed fit 
testing protocols? Is it appropriate to exclude the grimace completely 
from the proposed protocols, given that it is not used in the 
calculation of the fit factor result specified under the existing or 
proposed test methods? If not, what other criteria could be used to 
assess its inclusion or exclusion?
    One commenter (OSHA-2015-0015-0026) stated that he ``seriously 
question[s] the choice of the study and protocol authors in removing 
the Grimace exercise.'' While he ``concur[s] with their statement that 
it cannot be consistently applied and with their statement that the fit 
factor if measured should not be used in calculation of the fit 
factor,'' his ``interpretation is that the importance of the grimace is 
not in the fit factor achieved during this step of the protocol but 
instead in the ability of the mask to re-seal after this exercise which 
goes to the respirator['s] proper fit.''
    While NIOSH (OSHA-2015-0015-0031) ``recommends that the grimace 
test be included in the abbreviated protocols when used in the 
workplace since it is part of the currently accepted protocols,'' NIOSH 
agrees that the new ``protocols provide a valid reason for not 
including [the grimace] in the method comparison testing since it would 
add a non-controlled variable.'' Similarly, another commenter stated:

    The Grimace exercise is intended to break the face seal and then 
measure the recovery of the seal in the following exercises. By 
breaking the seal in the Grimace exercise during the reference 
protocol you have now altered the original fit of the mask and 
compromised the second fit test data. Therefore it makes logical 
sense that this exercise was eliminated from the test procedure for 
both the reference test and the proposed test. The fit of the mask 
as originally donned is consistent for both the reference test and 
the proposed protocol test (OSHA-2015-0015-0035).

    OSHA agrees that it is reasonable to remove the grimace exercise 
from the reference method during the method comparison testing, because 
its inclusion would unpredictably impact respirator fit. Some 
respirator fit test protocols include the grimace exercise because it 
is believed that it will unseat the respirator facepiece; whether this 
occurs is assessed, however, only during the subsequent exercise--fit 
measured during the grimace exercise is not included in the calculation 
of overall fit. Because method comparison requires a range of fit 
factors (from poor- to well-fitting respirators), OSHA believes that 
excluding the short grimace exercise allows for a more consistent 
assessment of fit between the reference and new fit test protocols.
    Finally, neither the ANSI annex nor the OSHA appendix specifies 
which exercises must be used in a new fit testing protocol. The 2010 
ANSI Z88.10 standard specifically considers the grimace exercise to be 
elective for the particle-counting instrument quantitative fit test 
procedure that it describes (see Table I). And although OSHA requires 
the grimace exercise as part of the original ambient aerosol CNC 
protocol, OSHA approved an abbreviated CNP REDON protocol in 2004 that 
excluded the grimace exercise among four other exercises. As such, OSHA 
concludes that it is not necessary to add the grimace exercise to the 
proposed protocols.
    11. The protocols in the three studies specify that participants 
take two deep breaths at the extreme of the head side-to-side and head 
up-and-down exercises and at the bottom of the bend in the bend-forward 
exercise. According to the developers of these protocols, the deep 
breaths are included to make the exercises more rigorous and 
reproducible from one subject to the next. Are these additional 
breathing instructions adequately explained in the studies and in the 
proposed amendment to the standard? Are they reasonable and 
appropriate?
    OSHA received no comments regarding these questions, which suggests 
that the breathing instructions were adequately explained in both the 
studies and in the proposed amendment to the standard, and that 
stakeholders were not concerned about this issue.
    12. Does OSHA's proposed regulatory text for the two new protocols 
offer clear instructions for implementing the protocols accurately?
    Neither TSI nor any commenters expressed concern about the clarity 
of OSHA's proposed regulatory text instructions for implementing the 
protocols. In the absence of such comments, the only changes that OSHA 
has made to the proposed regulatory text include an expansion of the 
titles of Tables A-1 and A-2 to match the names of the new protocols 
exactly. OSHA did this solely for clarity, so employers correctly 
correlate these two new tables with the two new proposed protocols.
    Several commenters expressed miscellaneous concerns that did not 
fall directly under any of OSHA's specific questions for public 
comment. OSHA addresses each in turn. One commenter was not in favor of 
any quantitative fit testing methods because, in his view, qualitative 
fit tests are more convincing to the respirator wearers themselves 
(OSHA-2015-0015-0017):

    [p]assing quantitative measurements may be literally orders of 
magnitude apart. If the machine says a 13 is passing, and a 400 is 
passing as well, how are the wearers of the respirators supposed to 
feel when they compare their numbers? (I have literally seen those 
numbers before entering a CBRN Defense Training Facility (CDTF) with 
live nerve and mustard agent; each individual was concerned that 
his/her mask was not as ``good'' as the other's, as they had no idea 
what the numbers meant.

As an initial matter, this rulemaking was not intended to compare 
qualitative fit tests to quantitative fit tests--employers are free to 
choose such tests as appropriate under appendix A of the Respiratory 
Protection Standard. The two new protocols will serve only as 
additional quantitative fit testing options to employers. That said, 
qualitative fit testing is not appropriate for certain respirators. In 
fact, the individuals described by the commenter could not have used 
qualitative fit testing because proper protection against CBRN 
(chemical, biological, radiological and nuclear) exposures

[[Page 50750]]

requires a full-facepiece, which must be fit tested using a 
quantitative method.\11\
---------------------------------------------------------------------------

    \11\ Qualitative fit tests are limited to negative pressure air-
purifying respirators that must achieve a fit factor of 100 or less, 
i.e., they may only be used to fit test half-mask, not full-
facepiece, respirators. 29 CFR 1910.134(f)(6).
---------------------------------------------------------------------------

    Another commenter was concerned about shortening the protocols to 
less than an eight-minute period, because she thought that symptoms of 
claustrophobia/panic attacks might not manifest before eight minutes 
(OSHA-2015-0015-0021). However, the risk of claustrophobia/panic 
attacks is already addressed when the wearer is required, under Sec.  
1910.134(e)(1) of the Respiratory Protection Standard, to undergo a 
mandatory medical evaluation ``to determine the employee's ability to 
use a respirator, before the employee is fit tested or required to use 
the respirator in the workplace.'' And the mandatory medical 
questionnaire in Appendix C of the standard includes a question 
regarding claustrophobia. In addition, OSHA is unaware of this having 
been an issue for respirator wearers fit tested using the CNP REDON 
protocol, which also lasts less than eight minutes and was approved by 
OSHA in 2004.
    Two commenters who favored shorter protocols expressed interest in 
making the new protocols available on all ambient aerosol CNC-based fit 
testing instruments, particularly the older PortaCount[supreg] (model 
8020) machines (OSHA-2015-0015-0028, OSHA-2015-0015-0030). OSHA notes 
that the new protocols are not restricted to any particular testing 
instrument because OSHA only approves fit testing protocols, not 
specific fit testing machines.\12\ OSHA has no authority to require 
specific fit testing machines or models for new protocols. Employers 
must contact the manufacturers of CNC fit testing machines to determine 
which models support the new protocols.
---------------------------------------------------------------------------

    \12\ TSI informed OSHA that the new protocols would not be 
available on the now-discontinued 8020 models (OSHA-2015-0010).
---------------------------------------------------------------------------

E. Conclusions

    After reviewing the comments submitted to the record, OSHA finds 
that the two proposed modified ambient aerosol CNC quantitative fit 
testing protocols are supported by peer-reviewed studies that were 
conducted according to accepted experimental design practices and 
principles and that produced results that were properly, fully, and 
fairly presented and interpreted. In addition, based on the peer-
reviewed studies and comments submitted to the record, OSHA finds that 
the two proposed protocols meet the sensitivity, specificity, 
predictive value, and other criteria contained in the ANSI annex. 
Moreover, the proposed protocols met the criteria of the ANSI annex, 
and in the absence of any compelling data or research in the record 
that would suggest that the proposed protocols would not generate 
reproducible fit testing results, OSHA concludes that the proposed 
protocols will generate reproducible fit testing results. In summary, 
OSHA concludes that the two proposed protocols are sufficiently 
accurate and reliable to approve and include in appendix A of its 
Respiratory Protection Standard.

F. N95-CompanionTM Technology

    The original TSI PortaCount[supreg] machine (model 8020) could only 
be used to fit test respirators equipped with >=99% efficient filter 
media (i.e., N-, R-, or P-99 and 100 NIOSH filter designations). In 
1998, TSI introduced the N95-CompanionTM Technology, which, 
when combined with the PortaCount[supreg] 8020 model, could be used to 
fit test respirators equipped with <99% efficient filter media (e.g., 
N95 NIOSH filter designation). TSI no longer manufactures the 8020 
model, which was replaced by a second generation of PortaCount[supreg] 
instruments (models 8030 and 8038). TSI introduced a third generation 
of PortaCount[supreg] instruments (models 8040 and 8048) in November 
2017. Models 8030 and 8040 can only test the most efficient filters 
(i.e., 99 and 100 NIOSH filter designations), while models 8038 and 
8048, which include the N95 CompanionTM Technology already 
built into the machine, can test any type of filter by selecting the 
appropriate operating mode. Because employers are sometimes confused by 
this distinction, OSHA considered using this rulemaking to propose 
additional language to Part I.C.3 of appendix A of the Respiratory 
Protection Standard to reflect this technological development. The 
additional language proposed by OSHA did not alter the fit testing 
protocol or impose any new requirements on employers; it was merely 
intended for clarification purposes.
    One commenter expressed concern over the use of the brand name 
``Portacount[supreg]'' within the regulatory text, stating that 
``[t]his seems to exclude other potential CNC providers'' (OSHA-2015-
0015-0024). Regarding this comment, the original OSHA-approved ambient 
aerosol CNC protocol is often commonly referred to as the 
PortaCount[supreg] protocol because of the name of the CNC machines 
manufactured by the company (i.e., TSI) that proposed the original 
protocol. OSHA is aware of only one other manufacturer that produces 
CNC instrumentation that is sold in the U.S. at this time. This new CNC 
instrumentation was only recently introduced into the market, so OSHA 
estimates that the overwhelming majority of the CNC instruments used in 
the U.S. at this time are still TSI PortaCount[supreg] machines. As 
such, OSHA determined that it is in the best interests of worker health 
and safety to retain the PortaCount[supreg] name within the regulatory 
text, as it has appeared in appendix A since 1998. This language is not 
intended to be exclude other manufacturers. It is intended merely to 
reflect that TSI's machines are those typically used for this test at 
this point in time. OSHA does not approve any safety equipment or 
require employers to use specific brands of safety equipment. However, 
it does sometimes refer to company or brand names when it is in the 
interest of safety and health. For example, appendix A of the 
Respiratory Protection Standard also includes the brand name (i.e., 
Bitrex[supreg]) for the substance (i.e., denatonium benzoate solution 
aerosol) overwhelmingly used for one of the OSHA-approved qualitative 
fit testing protocols. In addition, appendix A refers to the name of 
the company (i.e., Occupational Health Dynamics) that proposed the 
original CNP protocol and manufacturers CNP instrumentation.
    OSHA has, however, decided not to add the clarifying information 
about the different types of PortaCount[supreg] machines, due to 
commenter concerns that the inclusion of such information could create 
the appearance of a product endorsement. Since OSHA approves fit 
testing protocols rather than machines, OSHA feels that employers can 
contact fit testing instrument manufacturers for product specificity 
and capabilities.

III. Procedural Determinations

A. Legal Considerations

    OSHA's Respiratory Protection Standard is based on evidence that 
fit testing is necessary to ensure proper respirator fit for employees, 
which protects them against excessive exposure to airborne contaminants 
in the workplace. Employers covered by this revision already must 
comply with the fit testing requirements specified in paragraph (f) of 
OSHA's Respiratory Protection Standard at 29 CFR 1910.134.
    OSHA has determined that the additional modified ambient aerosol 
CNC protocols provide employees with protection that is comparable to 
the protection afforded them by the existing fit testing provisions. 
The additional

[[Page 50751]]

modified ambient aerosol CNC protocols do not replace existing fit 
testing protocols, but instead are alternatives to them. Therefore, 
OSHA finds that the final standard does not directly increase or 
decrease the protection afforded to employees, nor does it increase 
employers' compliance burden. The additional modified ambient aerosol 
CNC protocols reduce the total fit test duration, and therefore may 
reduce the compliance burden for employers that elect to use one of 
these protocols.

B. Final Economic Analysis and Regulatory Flexibility Certification

    The rule is not economically significant under Executive Order 
12866 (58 FR 51735) or a ``major rule'' under Section 804 of the Small 
Business Regulatory Enforcement Fairness Act of 1996 (5 U.S.C. 804). 
The rule imposes no additional costs on any private- or public-sector 
entity and is not a significant or major rule under Executive Order 
12866 or other relevant statutes or executive orders. This rulemaking 
increases employers' flexibility in choosing fit testing methods for 
employees, and the final rule does not require an employer to update or 
replace its current fit testing method(s) if the fit testing method(s) 
currently in use meets existing standards. Furthermore, because the 
rule offers additional options that employers would be expected to 
select only if those options did not impose any net cost burdens on 
them, the rule will not have a significant impact on a substantial 
number of small entities.
    OSHA received several comments in response to the NPRM related to 
the time savings anticipated by the proposal. As discussed in the 
``Summary and Explanation,'' a number of commenters noted that time 
savings of the proposed fit testing protocols would increase efficiency 
and be substantial when aggregated across a large number of employees 
(OSHA-2015-0015-0018, OSHA-2015-0015-0020). No comments indicated that 
the time savings estimates would be significantly different from those 
put forth in the Preliminary Economic Analysis (PEA).\13\ As a result, 
OSHA has not changed its methodology for calculating the potential cost 
savings of implementing the new protocols.
---------------------------------------------------------------------------

    \13\ As discussed in the ``Summary and Explanation,'' several 
comments (OSHA-2015-0015-0022, OSHA-2015-0015-0032, OSHA-2015-0015-
0042) expressed concern about the estimated decrease in total 
ambient test time included as part of the protocol. The ``Summary 
and Explanation'' explains why this test time is reasonable and 
sufficient in this context. However, the comments did not question 
the total estimated time savings for the new protocols, per se.
---------------------------------------------------------------------------

    The new quantitative fit testing (QNFT) protocols will provide 
employers additional options to fit test their employees for respirator 
use. While OSHA approves fit testing protocols rather than fit testing 
machines, OSHA understands that, currently, the market for fit testing 
machines using the original ambient aerosol CNC protocol is dominated 
by TSI's PortaCount[supreg] machines (Models 8020, 8030, 8038, 8040, 
8048).\14\ As such, OSHA's Final Economic Analysis (FEA) focuses 
specifically on TSI's PortaCount[supreg] machines. Employers already 
using the original ambient aerosol CNC protocol with a 
PortaCount[supreg] machine (with the exception of the now-discontinued 
8020) may switch from the original ambient aerosol CNC protocol to the 
new protocols. OSHA estimates switching saves approximately 5 minutes 
per fit test, and grants the employer corresponding cost savings.
---------------------------------------------------------------------------

    \14\ TSI indicated that as of the beginning of 2018, there were 
no active competitors, but that at least one company may be entering 
the market later in the year (OSHA-2015-0015-0046).
---------------------------------------------------------------------------

    According to TSI, ``[e]xisting owners of the PortaCount[supreg] 
Respirator Fit Tester Pro Model 8030 and/or PortaCount[supreg] Pro+ 
Model 8038 will be able to utilize the new protocols without additional 
expense. It will be necessary for fit testers to obtain a firmware and 
FitPro software upgrade, which TSI will be providing as a free 
download. As an alternative to the free download, PortaCount[supreg] 
Models 8030 and 8038 returned for annual service will be upgraded 
without additional charge. Owners of the PortaCount[supreg] Plus Model 
8020 with or without the N95-CompanionTM Technology (both 
discontinued in 2008) will be limited to the current 8-exercise OSHA 
fit test protocol'' (OSHA-2015-0015-0010).\15\ There are approximately 
12,000 Model 8030 or 8038 units in the field.\16\ Existing 
PortaCount[supreg] users may adopt the new protocols with minimal 
effort: The fit tester will be able to select the new protocol after 
taking an estimated less than five minutes to download TSI's firmware 
and software updates. The individual being fit tested is also likely to 
learn the new protocols with minimal time. In fact, information about 
the new protocols could be imparted during the annual training mandated 
by OSHA's respiratory protection rule (OSHA-2015-0015-0012). As a 
practical matter, the new protocols contain fewer exercises requiring 
mastery. And Part I.A.12 of appendix A of OSHA's Respiratory Protection 
Standard already requires the fit tester to describe the fit test to 
the respirator wearer, regardless of which fit test it is or how often 
it is used. Thus, there should be no additional burden to the employer 
or employee.
---------------------------------------------------------------------------

    \15\ TSI later confirmed this information still applied in 2018, 
even after the introduction of their new models (OSHA-2015-0015-
0046).
    \16\ As indicated by TSI in 2015 (OSHA-2015-0015-0012). As 
explained later on in this FEA, the aggregate cost savings were 
based on estimates of current use of the 8030 and 8038 models. As 
the market is now being augmented with the 8040 and 8048 models, it 
is likely a conservative estimate of the potential cost savings.
---------------------------------------------------------------------------

    OSHA anticipates many employers who currently use the original 
ambient aerosol CNC protocol will adopt the new protocols because they 
could be adopted at negligible cost to the employer and would take less 
time to administer. OSHA expects that the new protocols are less likely 
to be adopted by employers who currently perform fit testing using 
other quantitative or qualitative fit tests because of the significant 
equipment and training investment that they already have made to 
administer these fit tests. For example, OSHA estimates, based on 
information from TSI, that switching from qualitative to quantitative 
fit testing would require upfront costs of $8,700 to $12,000 per 
machine (OSHA-2015-0015-0012).
    OSHA has estimates of the number of users of the PortaCount[supreg] 
technology at the establishment level, both from the manufacturer and 
from the 2001 NIOSH Respirator Survey. However, what is not known is 
how many respirator wearers, that is, employees, are fit tested using a 
PortaCount[supreg] device. As described in the PEA, OSHA expects that 
economies of scale will apply in this situation--larger establishments 
will be more likely to encounter situations needing QNFT, but will also 
have more employees over which to spread the capital costs. OSHA 
received no comments about its understanding of employer size in 
relation to QNFT use. Once employers have invested capital in a 
quantitative fit testing device, they have more of an incentive to 
perform QNFT in a given situation, even if not technically required to 
use QNFT in every situation. Also, some QNFT devices are acquired by 
third parties, or ``fit testing houses,'' that provide fit testing 
services to employers. In short, as put forth in the PEA, OSHA believes 
that employers using PortaCount[supreg] QNFT will process more 
respirator wearers than the average establishment. OSHA received no 
comments about this conclusion.
    As set forth in the PEA, if one started with an estimate of 12,000 
establishments using PortaCount[supreg]

[[Page 50752]]

models 8030 and 8038 annually for all of their employees and assumed an 
average of 100 respirator wearers fit tested annually per 
establishment, this yielded an estimate of 1.2 million respirator 
wearers that could potentially benefit from the new QNFT protocols.\17\ 
Alternatively, as also set out in the PEA, a similar estimate would 
have been obtained if one assumed, employing data from the 2001 NIOSH 
Respirator Survey, that 50 percent of the devices requiring QNFT (such 
as full-facepiece elastomeric negative pressure respirators) use 
PortaCount[supreg] currently, as well as 25 percent of half-mask 
elastomeric respirators, and 10 percent of filtering facepieces.\18\ 
These estimates in the PEA were not questioned in public comment. In 
the intervening period between the PEA and the FEA, the total number of 
employees and estimated respirator wearers increased somewhat, raising 
the estimated number of respirator wearers affected by the rulemaking, 
based on survey data, to approximately 1.3 million.
---------------------------------------------------------------------------

    \17\ TSI estimated the number of users of their devices at over 
12,000 establishments (OSHA-2015-0015-0012). As indicated in the 
PEA, this was consistent with data from the 2001 NIOSH respirator 
survey (OSHA-2015-0015-0045), which, if benchmarked to a 2012 count 
of establishments (OSHA-2015-0015-0048) and containing fit testing 
methods to include ambient aerosol, generated aerosol, and a 
proportionally allocated percentage of the ``don't know'' 
respondents, would provide an estimate of 12,458 establishments 
using PortaCount[supreg] currently. Based on information from TSI, 
the large majority of these are estimated to be the newer 8030 and 
8038 devices.
    \18\ Based on the 2001 NIOSH respirator survey (OSHA-2015-0015-
0045), benchmarked to 2015 County Business Patterns (OSHA-2015-0015-
0048), OSHA estimates 1,273,616 (or approximately 1.3 million) 
employees will be affected by the rulemaking. These estimates are 
based only on private employers. Accounting for governmental 
entities would result in an even larger number of total estimated 
respirator users affected.
---------------------------------------------------------------------------

    If applied to approximately 1.3 million respirators wearers, an 
estimated savings of 5 minutes per respirator wearer would equal over 
100,000 hours of employee time saved annually. Consistent with 
Department of Labor policy for translating the labor time savings into 
dollar cost savings for this FEA, OSHA included an overhead rate when 
estimating the marginal cost of labor in its primary cost calculation. 
Overhead costs are indirect expenses that cannot be tied to producing a 
specific product or service. Common examples include rent, utilities, 
and office equipment. Unfortunately, there is no general consensus on 
the cost elements that fit this definition. The lack of a common 
definition has led to a wide range of overhead estimates. Consequently, 
the treatment of overhead costs needs to be case-specific. OSHA adopted 
an overhead rate of 17 percent of base wages, consistent with overhead 
rates used for other regulatory compliance rules.\19\ For example, this 
is consistent with the overhead rate used for sensitivity analyses in 
the 2017 Improved Tracking FEA and the FEA in support of OSHA's 2016 
final standard on Occupational Exposure to Respirable Crystalline 
Silica. For example, in this case, to calculate the total labor cost 
for a typical respirator wearer, based on the mean worker wage, three 
components are added together: Base wage ($23.86) + fringe benefits 
($10.42--43.7% of $23.86); \20\ and the applicable overhead costs 
($4.06--17% of $23.86). This results in an hourly labor cost of a 
respirator wearing employee to $38.34. This implies an estimated cost 
savings of $4.1 million attributable to the adoption of the new fit 
testing protocols.
---------------------------------------------------------------------------

    \19\ The methodology was modeled after an approach used by the 
Environmental Protection Agency. More information on this approach 
can be found at: U.S. Environmental Protection Agency, ``Wage Rates 
for Economic Analyses of the Toxics Release Inventory Program,'' 
June 10, 2002. This analysis itself was based on a survey of several 
large chemical manufacturing plants: Heiden Associates, Final 
Report: A Study of Industry Compliance Costs Under the Final 
Comprehensive Assessment Information Rule, Prepared for the Chemical 
Manufacturers Association, December 14, 1989.
    \20\ Mean wage rate of $23.86 (OSHA-2015-0015-0043), assuming 
fringe benefits are 30.4 percent of total compensation (OSHA-2015-
0015-0043), or by extension, 43.7% of base wages (1/(1-bw)).
---------------------------------------------------------------------------

    Because the $4.1 million represents annual cost savings, the final 
estimate is the same when discounted at either 3 or 7 percent. For the 
same reason, when the Department of Labor uses a perpetual time horizon 
to allow for cost comparisons under E.O. 13771, the annualized cost 
savings of the final rule are also $4.1 million with 7 percent 
discounting. As indicated earlier, this final estimate includes an 
overhead factor in the labor costs. This is estimated to add an 
additional savings of approximately 12%, or over $400,000, on what 
would have been an estimated savings of $3.6 million.
    In addition to costs related to the respirator wearer's time, there 
will also likely be time savings for the person administering the fit 
tests. However, OSHA did not include this cost savings element in the 
PEA because it lacked specific empirical information on this point at 
the time of the proposal. OSHA requested comment on this question, but 
did not receive any. While OSHA believes this element of the cost 
savings is potentially substantial, it is not a critical element for 
the FEA, as it is simply a question of how large the cost savings are, 
and not required, for example, to determine economic feasibility. 
Therefore, OSHA is maintaining in the final analysis the same 
analytical approach used in the PEA.\21\
---------------------------------------------------------------------------

    \21\ For example, in the PEA OSHA posited that the time saved 
may potentially be as much as a 1:1 ratio between the tester and 
those being tested. But, for purposes of argument, if the ratio was 
only 1:4 (or the equivalent of 1 minute 15 seconds of tester's time 
per employee tested), OSHA estimates the cost savings related to the 
tester would be an additional $1.3 million.
---------------------------------------------------------------------------

    In addition, as discussed, this FEA does not account for potential 
conversions from testing methods other than the original ambient 
aerosol CNC protocol. While such conversions could further increase 
time and cost savings, OSHA cannot predict the number of conversions 
with confidence. In short, while certain factors could change the 
precise cost savings estimates in the FEA, OSHA believes its estimates 
reasonably capture the direction and order of magnitude of the 
rulemaking's economic effects.
Regulatory Flexibility Certification
    In accordance with the Regulatory Flexibility Act, 5 U.S.C. 601 et 
seq. (as amended), OSHA has examined the regulatory requirements of the 
final rule to determine whether these requirements will have a 
significant economic impact on a substantial number of small entities. 
This rule will impose no required costs and could provide a cost 
savings in excess of $4 million per year to regulated entities. While 
measureable in the aggregate, these savings will be dispersed widely, 
and therefore are not estimated to have a substantial economic impact 
on any small entity, although the impacts are estimated to be positive. 
The Assistant Secretary for Occupational Safety and Health therefore 
certifies that the final rule will not have a significant economic 
impact on a substantial number of small entities.

C. Paperwork Reduction Act

Overview
    The Paperwork Reduction Act (PRA) requires that agencies obtain 
approval from OMB before conducting any collection of information (44 
U.S.C. 3507). The PRA defines ``collection of information'' to mean 
``the obtaining, causing to be obtained, soliciting, or requiring the 
disclosure to third parties or the public, of facts or opinions by or 
for an agency, regardless of form or format'' (44 U.S.C. 3502(3)(A)).
    In accordance with the PRA, 44 U.S.C. 3506(c)(2), OSHA solicited 
public comments on proposed revisions to the Respiratory Protection 
Standard Information Collection Request (ICR) (paperwork burden hour 
and cost

[[Page 50753]]

analysis) for the information collection requirements associated with 
the Additional PortaCount[supreg] Quantitative Fit-Testing Protocols: 
Amendment to Respiratory Protection Standard proposed rule (81 FR 
69747). The Department submitted this ICR to OMB for review in 
accordance with 44 U.S.C. 3507(d) on October 7, 2016. A copy of the ICR 
for the proposed rule is available to the public at: https://www.reginfo.gov/public/do/PRAViewICR?ref_nbr=201511-1218-005.
Solicitation of Comments
    On November 22, 2016, OMB issued a Notice of Action withholding its 
approval of the ICR. OMB requested that, ``[p]rior to publication of 
the final rule, the agency should provide a summary of any comments 
related to the information collection and their response, including any 
changes made to the ICR as a result of comments. In addition, the 
agency must enter the correct burden estimates.''
    No public comments were received specifically in response to the 
proposed ICR submitted to OMB for review. However, several public 
comments submitted in response to the NPRM, described earlier in this 
preamble, substantively addressed provisions containing collections of 
information and included information relevant to the burden hour and 
costs analysis. These comments are addressed in the preamble, and OSHA 
considered them when it developed the revised ICR associated with this 
final rule. See the comment analysis in section II.D above.
    Under the PRA, a Federal agency cannot conduct or sponsor a 
collection of information unless it is approved by OMB under the PRA, 
and the collection of information notice displays a currently valid OMB 
control number (44 U.S.C. 3507(a)(3)). Also, notwithstanding any other 
provision of law, no employer shall be subject to penalty for failing 
to comply with a collection of information if the collection of 
information does not display a currently valid OMB control number (44 
U.S.C. 3512). The revised information collection requirements found in 
the final rule are summarized below.
    The Department of Labor has submitted the final ICR concurrent with 
the publication of this final rule. The ICR contains a full analysis 
and description of the burden hours and costs associated with the 
information collection requirements of the final rule to OMB for 
approval. A copy of the ICR is available to the public at https://www.reginfo.gov/public/do/PRAViewICR?ref_nbr=201904-1218-002. OSHA will 
publish a separate notice in the Federal Register announcing the 
results of OMB's review. That notice will also include a list of OMB-
approved information collection requirements and the total burden hours 
and costs imposed by the final rule.
    The additional protocols adopted in this final rule revise the 
information collection in a way that reduces existing burden hours and 
costs. In particular, the information collection requirement specified 
in paragraph (m)(2) of OSHA's Respiratory Protection Standard, at 29 
CFR 1910.134, states that employers must document and maintain the 
following information on quantitative fit tests administered to 
employees: The name or identification of the employee tested; the type 
of fit test performed; the specific make, model, style, and size of 
respirator tested; the date of the test; and the test results. The 
employer must maintain this record until the next fit test is 
administered. While the information on the fit test record remains the 
same, the time to obtain the necessary information for the fit test 
record is reduced since the additional PortaCount[supreg] protocols 
will take an employer less time to administer than those currently 
approved in appendix A of the Respiratory Protection Standard. As a 
result, the total estimated burden hours decrease by 201,640 hours, 
from 7,622,100 to 7,420,460 hours. This decrease is a result of the 
more efficient protocols established under the final rule. OSHA 
accounts for this burden under the Information Collection Request, or 
paperwork analysis, for the Respiratory Protection Standard (OMB 
Control Number 1218-0099). Note that OSHA cannot require compliance 
with the information collection requirements for the new information 
collection in this final rule until OMB has approved the information 
collection requirements.
    Title of Collection: Respiratory Protection Standard (29 
CFR1910.134).
    OMB Control Number: 1218-0099.
    Affected Public: Private Sector--business or other for-profits.
    Total Estimated Number of Respondents: 24,710,469.
    Total Estimated Number of Responses: 25,042,236.
    Total Estimated Annual Time Burden Hours: 7,420,460.
    Total Estimated Annual Other Burden: $316,906,665.

D. Federalism

    OSHA reviewed this rulemaking according to the Executive Order on 
Federalism (E.O. 13132, 64 FR 43255, Aug. 10, 1999), which requires 
that Federal agencies, to the extent possible, refrain from limiting 
state policy options, consult with states before taking actions that 
would restrict states' policy options and take such actions only when 
clear constitutional authority exists and the problem is of national 
scope. The Executive Order provides for preemption of state law only 
with the expressed consent of Congress. Federal agencies must limit any 
such preemption to the extent possible.
    Under section 18 of the Occupational Safety and Health Act (the 
``Act,'' 29 U.S.C. 651 et seq.), Congress expressly provides that 
states may adopt, with Federal approval, a plan for the development and 
enforcement of occupational safety and health standards (29 U.S.C. 
667). OSHA refers to states that obtain Federal approval for such a 
plan as ``State Plan states.'' Occupational safety and health standards 
developed by State Plan states must be at least as effective in 
providing safe and healthful employment and places of employment as the 
Federal standards. Subject to these requirements, State Plan states are 
free to develop and enforce under state law their own requirements for 
occupational safety and health standards. With respect to states that 
do not have OSHA-approved plans, OSHA concludes that this standard 
conforms to the preemption provisions of the Act. Section 18 of the Act 
prohibits states without approved plans from issuing citations for 
violations of OSHA standards. OSHA finds that the rule does not expand 
this limitation. Therefore, for States that do not have approved 
occupational safety and health plans, the rule will not affect the 
preemption provisions of Section 18 of the Act.
    OSHA's rulemaking to adopt additional fit testing protocols under 
its Respiratory Protection Standard at 29 CFR 1910.134 is consistent 
with Executive Order 13132 because the problems addressed by these fit 
testing requirements are national in scope. OSHA concludes that the fit 
testing protocols adopted by this rulemaking provide employers in every 
state with procedures that will assist them in protecting their 
employees from the risks of exposure to atmospheric hazards. In this 
regard, the rule offers thousands of employers across the nation an 
opportunity to use additional protocols to assess respirator fit among 
their employees. Therefore, the rule provides employers in every state 
with an alternative means of complying with the fit testing 
requirements specified by paragraph (f) of OSHA's Respiratory 
Protection Standard.
    Section 18(c)(2) of the Act (29 U.S.C. 667(c)(2)) requires State 
Plan states to

[[Page 50754]]

adopt an OSHA standard, or to develop and enforce an alternative that 
is at least as effective as the OSHA standard. However, the new fit 
testing protocols adopted by this rulemaking provide employers with 
alternatives to the existing fit testing protocols specified in the 
Respiratory Protection Standard; therefore, the alternative is not, 
itself, a mandatory standard. Accordingly, states with OSHA-approved 
State Plans are not obligated to adopt the additional fit testing 
protocols adopted here. Nevertheless, OSHA strongly encourages them to 
adopt the final provisions to provide additional compliance options to 
employers in their states.
    In summary, this rulemaking complies with Executive Order 13132. In 
states without OSHA-approved State Plans, this rulemaking limits state 
policy options in the same manner as other OSHA standards. In State 
Plan states, this rulemaking does not significantly limit state policy 
options.

E. State Plan States

    Section 18(c)(2) of the Act (29 U.S.C. 667(c)(2)) requires State 
Plan states to adopt mandatory standards promulgated by OSHA, or to 
develop and enforce an alternative that is at least as effective as the 
OSHA standard. However, as noted in the previous section of this 
preamble, states with OSHA-approved State Plans are not obligated to 
adopt the provisions of this final rule. Nevertheless, OSHA strongly 
encourages them to adopt the final provisions to provide compliance 
options to employers in their States. In this regard, OSHA concludes 
that the fit testing protocols adopted by this rulemaking provide 
employers in the State Plan states with procedures that protect the 
safety and health of employees who use respirators against hazardous 
airborne substances in their workplace at least as well as the 
quantitative fit testing protocols in appendix A of the Respiratory 
Protection Standard.
    There are 28 states and U.S. territories that have their own OSHA-
approved occupational safety and health programs called State Plans. 
The following 22 State Plans cover state and local government employers 
and private-sector employers: Alaska, Arizona, California, Hawaii, 
Indiana, Iowa, Kentucky, Maryland, Michigan, Minnesota, Nevada, New 
Mexico, North Carolina, Oregon, Puerto Rico, South Carolina, Tennessee, 
Utah, Vermont, Virginia, Washington, and Wyoming. The following six 
State Plans cover state and local government employers only: 
Connecticut, Illinois, Maine, New Jersey, New York, and the Virgin 
Islands.

F. Unfunded Mandates Reform Act

    OSHA reviewed this rulemaking according to the Unfunded Mandates 
Reform Act of 1995 (UMRA) (2 U.S.C. 1501-1507) and Executive Order 
12875 (58 FR 58093 (1993)). As discussed above in section III.B of this 
preamble (``Final Economic Analysis and Regulatory Flexibility 
Certification''), OSHA has determined that the rule imposes no 
additional costs on any private-sector or public-sector entity. The 
substantive content of the rule applies only to employers whose 
employees use respirators for protection against airborne contaminants, 
and compliance with the protocols contained in the final rule are 
strictly optional for these employers. Accordingly, the final rule does 
not require additional expenditures by either public or private 
employers. Therefore, this rulemaking is not a significant regulatory 
action within the meaning of Section 202 of the UMRA, 2 U.S.C. 1532.
    As noted above under Section E (``State Plan States'') of this 
preamble, OSHA standards do not apply to state or local governments 
except in states that have voluntarily elected to adopt an OSHA-
approved State Plan. Consequently, this final rulemaking does not meet 
the definition of a ``Federal intergovernmental mandate'' (see 2 U.S.C. 
658(5)). Therefore, for the purposes of the UMRA, the Assistant 
Secretary for Occupational Safety and Health certifies that this 
rulemaking does not mandate that state, local, or tribal governments 
adopt new, unfunded regulatory obligations, or increase expenditures by 
the private sector of more than $100 million in any year.

G. Applicability of Existing Consensus Standards

    Section 6(b)(8) of the Act (29 U.S.C. 655(b)(8)) requires OSHA to 
explain ``why a rule promulgated by the Secretary differs substantially 
from an existing national consensus standard,'' by publishing ``a 
statement of the reasons why the rule as adopted will better effectuate 
the purposes of the Act than the national consensus standard.'' The 
American National Standards Institute (ANSI) developed a national 
consensus standard on fit testing protocols (``Respirator Fit Testing 
Methods,'' ANSI Z88.10-2001) as an adjunct to its national consensus 
standard on respiratory protection programs. ANSI/AIHA updated the 
Z88.10 standard in 2010 (``Respirator Fit Testing Methods,'' ANSI 
Z88.10-2010) (OSHA-2015-0015-0007).
    Paragraph 7.2 of ANSI/AIHA Z88.10-2010 specifies the requirements 
for conducting a particle-counting-instrument (e.g., 
PortaCount[supreg]) quantitative fit test. The modified CNC protocols 
adopted by the final rule are variations of this national consensus 
standard's particle counting-instrument quantitative fit test 
procedures: The new protocols require the same 30-second duration for 
fit testing exercises, but not the same exercises as ANSI/AIHA. 
However, Annex A2 of ANSI/AIHA Z88.10-2010 recognizes that a 
universally accepted measurement standard for respirator fit testing 
does not exist and provides specific requirements for evaluating new 
fit testing methods. OSHA has concluded that the modified CNC protocols 
submitted by TSI meet the evaluation criteria outlined in ANSI/AIHA 
Z88.10-2010, Annex A2.

H. Advisory Committee for Construction Safety and Health (ACCSH) Review 
of the Proposed Standard

    The Contract Work Hours and Safety Standards Act (Construction 
Safety Act) (40 U.S.C. 3704), OSHA regulations governing the Advisory 
Committee for Construction Safety and Health (ACCSH) (i.e., 29 CFR 
1912.3), and provisions governing OSHA rulemaking (i.e., 29 CFR 
1911.10) require OSHA to consult with the ACCSH whenever OSHA proposes 
a rule involving construction activities. Specifically, 29 CFR 1911.10 
requires that the Assistant Secretary provide the ACCSH with ``any 
proposal of his own,'' together with ``all pertinent factual 
information available to him, including the results of research, 
demonstrations, and experiments.''
    The addition of two quantitative fit test protocols to appendix A 
of OSHA's Respiratory Protection Standard affects the construction 
industry because it revises the fit testing procedures used in that 
industry (see 29 CFR 1926.103). Accordingly, OSHA provided the ACCSH 
members with TSI's application letter, supporting documents, and other 
relevant information, prior to the December 4, 2014 ACCSH meeting. OSHA 
explained its proposal to add new protocols to the ACCSH at that 
meeting, and the ACCSH unanimously approved proceeding with a proposed 
rule.

List of Subjects in 29 CFR Part 1910

    Fit testing, Hazardous substances, Health, Occupational safety and 
health, Respirators, Respiratory protection, Toxic substances.

[[Page 50755]]

Authority and Signature

    Loren Sweatt, Acting Assistant Secretary of Labor for Occupational 
Safety and Health, U.S. Department of Labor, authorized the preparation 
of this document pursuant to Sections 4, 6, and 8 of the Occupational 
Safety and Health Act of 1970 (29 U.S.C. 653, 655, 657), 29 CFR part 
1911, and Secretary's Order 1-2012 (77 FR 3912).

    Signed at Washington, DC, on September 19, 2019.
Loren Sweatt,
Principal Deputy Assistant Secretary of Labor for Occupational Safety 
and Health.

Amendments to the Standard

    For the reasons stated in the preamble, the agency amends 29 CFR 
part 1910 as follows:

PART 1910--[AMENDED]

Subpart I--[Amended]

0
1. Revise the authority citation for subpart I of part 1910 to read as 
follows:

    Authority:  29 U.S.C. 653, 655, 657; Secretary of Labor's Order 
No. 12-71 (36 FR 8754), 8-76 (41 FR 25059), 9-83 (48 FR 35736), 1-90 
(55 FR 9033), 6-96 (62 FR 111), 3-2000 (65 FR 50017), 5-2002 (67 FR 
65008), 5-2007 (72 FR 31160), 4-2010 (75 FR 55355), or 1-2012 (77 FR 
3912), as applicable, and 29 CFR part 1911.


0
2. Amend Part I in appendix A to Sec.  1910.134 as follows:
0
a. Revise Section A.14(a) introductory text;
0
b. In Section C.3:
0
i. Revise the introductory text; and
0
ii. Remove the terms ``PortacountTM'' and ``Portacount'' and 
add in their place the term ``PortaCount[supreg]'';
0
c. Redesignate Sections C.4 and 5 of as Sections C.6 and 7;
0
d. Add new Sections C.4 and 5; and
0
e. In newly redesignated Section C.7:
0
i. Revise paragraph (a) and paragraph (b) introductory text; and
0
ii. Redesignate Table A-1 as Table A-3; and
    The revisions and additions read as follows:


Sec.  1910.134  Respiratory protection.

* * * * *

APPENDIX A to Sec.  1910.134--FIT TESTING PROCEDURES (MANDATORY)

Part I. OSHA--Accepted Fit Test Protocols

A. Fit Testing Procedures--General Requirements

* * * * *
    14. Test Exercises. (a) Employers must perform the following 
test exercises for all fit testing methods prescribed in this 
appendix, except for the two modified ambient aerosol CNC 
quantitative fit testing protocols, the CNP quantitative fit testing 
protocol, and the CNP REDON quantitative fit testing protocol. For 
the modified ambient aerosol CNC quantitative fit testing protocols, 
employers shall ensure that the test subjects (i.e., employees) 
perform the exercise procedure specified in Part I.C.4(b) of this 
appendix for full-facepiece and half-mask elastomeric respirators, 
or the exercise procedure specified in Part I.C.5(b) for filtering 
facepiece respirators. Employers shall ensure that the test subjects 
(i.e., employees) perform the exercise procedure specified in Part 
I.C.6(b) of this appendix for the CNP quantitative fit testing 
protocol, or the exercise procedure described in Part I.C.7(b) of 
this appendix for the CNP REDON quantitative fit testing protocol. 
For the remaining fit testing methods, employers shall ensure that 
the test exercises are performed in the appropriate test environment 
in the following manner:
* * * * *

C. Quantitative Fit Test (QNFT) Protocols

* * * * *
    3. Ambient aerosol condensation nuclei counter (CNC) 
quantitative fit testing protocol.
    The ambient aerosol condensation nuclei counter (CNC) 
quantitative fit testing (PortaCount[supreg]) protocol 
quantitatively fit tests respirators with the use of a probe. The 
probed respirator is only used for quantitative fit tests. A probed 
respirator has a special sampling device, installed on the 
respirator, that allows the probe to sample the air from inside the 
mask. A probed respirator is required for each make, style, model, 
and size that the employer uses and can be obtained from the 
respirator manufacturer or distributor. The primary CNC instrument 
manufacturer, TSI Incorporated, also provides probe attachments (TSI 
mask sampling adapters) that permit fit testing in an employee's own 
respirator. A minimum fit factor pass level of at least 100 is 
necessary for a half-mask respirator (elastomeric or filtering 
facepiece), and a minimum fit factor pass level of at least 500 is 
required for a full-facepiece elastomeric respirator. The entire 
screening and testing procedure shall be explained to the test 
subject prior to the conduct of the screening test.
* * * * *
    4. Modified ambient aerosol condensation nuclei counter (CNC) 
quantitative fit testing protocol for full-facepiece and half-mask 
elastomeric respirators.
    (a) When administering this protocol to test subjects, employers 
shall comply with the requirements specified in Part I.C.3 of this 
appendix (ambient aerosol condensation nuclei counter (CNC) 
quantitative fit testing protocol), except they shall use the test 
exercises described below in paragraph (b) of this protocol instead 
of the test exercises specified in section I.C.3(a)(6) of this 
appendix.
    (b) Employers shall ensure that each test subject being fit 
tested using this protocol follows the exercise and duration 
procedures, including the order of administration, described in 
Table A-1 of this appendix.

   Table A-1-- Modified Ambient Aerosal CNC Quantitative Fit Testing Protocol for Full Facepiece and Half-Mask
                                             Elastomeric Respirators
----------------------------------------------------------------------------------------------------------------
           Exercises \1\                            Exercise procedure                   Measurement procedure
----------------------------------------------------------------------------------------------------------------
Bending Over.......................  The test subject shall bend at the waist, as if  A 20 second ambient
                                      going to touch his/her toes for 50 seconds and   sample, followed by a 30
                                      inhale 2 times at the bottom \2\.                second mask sample.
Jogging-in-Place...................  The test subject shall jog in place comfortably  A 30 second mask sample.
                                      for 30 seconds.
Head Side-to-Side..................  The test subject shall stand in place, slowly    A 30 second mask sample.
                                      turning his/her head from side to side for 30
                                      seconds and inhale 2 times at each extreme \2\.
Head Up-and-Down...................  The test subject shall stand in place, slowly    A 30 second mask sample
                                      moving his/her head up and down for 39 seconds   followed by a 9 second
                                      and inhale 2 times at each extreme \2\.          ambient sample.
----------------------------------------------------------------------------------------------------------------
\1\ Exercises are listed in the order in which they are to be administered.
\2\ It is optional for test subjects to take additional breaths at other times during this exercise.

    5. Modified ambient aerosol condensation nuclei counter (CNC) 
quantitative fit testing protocol for filtering facepiece 
respirators.
    (a) When administering this protocol to test subjects, employers 
shall comply with the requirements specified in Part I.C.3 of this 
appendix (ambient aerosol condensation nuclei counter (CNC) 
quantitative fit testing protocol), except they shall use the test 
exercises described below in paragraph (b) of this protocol instead 
of the test exercises specified in section I.C.3(a)(6) of this 
appendix.
    (b) Employers shall ensure that each test subject being fit 
tested using this protocol follows the exercise and duration 
procedures, including the order of administration, described in 
Table A-2 of this appendix.

[[Page 50756]]



 Table A-2-- Modified Ambient Aerosal CNC Quantitative Fit Testing Protocol for Filtering Facepiece Respirators
----------------------------------------------------------------------------------------------------------------
           Exercises \1\                            Exercise procedure                   Measurement procedure
----------------------------------------------------------------------------------------------------------------
Bending Over.......................  The test subject shall bend at the waist, as if  A 20 second ambient
                                      going to touch his/her toes for 50 seconds and   sample, followed by a 30
                                      inhale 2 times at the bottom \2\.                second mask sample.
Talking............................  The test subject shall talk out loud slowly and  A 30 second mask sample.
                                      loud enough so as to be heard clearly by the
                                      test conductor for 30 seconds. He/she will
                                      either read from a prepared text such as the
                                      Rainbow Passage, count backward from 100, or
                                      recite a memorized poem or song.
Head Side-to-Side..................  The test subject shall stand in place, slowly    A 30 second mask sample.
                                      turning his/her head from side to side for 30
                                      seconds and inhale 2 times at each extreme \2\.
Head Up-and-Down...................  The test subject shall stand in place, slowly    A 30 second mask sample
                                      moving his/her head up and down for 39 seconds   followed by a 9 second
                                      and inhale 2 times at each extreme \2\.          ambient sample.
----------------------------------------------------------------------------------------------------------------
\1\ Exercises are listed in the order in which they are to be administered.
\2\ It is optional for test subjects to take additional breaths at other times during this exercise.

* * * * *
    7. Controlled negative pressure (CNP) REDON quantitative fit 
testing protocol.
    (a) When administering this protocol to test subjects, employers 
must comply with the requirements specified in paragraphs (a) and 
(c) of part I.C.6 of this appendix (``Controlled negative pressure 
(CNP) quantitative fit testing protocol,'') as well as use the test 
exercises described below in paragraph (b) of this protocol instead 
of the test exercises specified in paragraph (b) of part I.C.6 of 
this appendix.
    (b) Employers must ensure that each test subject being fit 
tested using this protocol follows the exercise and measurement 
procedures, including the order of administration described in Table 
A-3 of this appendix.
* * * * *
[FR Doc. 2019-20686 Filed 9-25-19; 8:45 am]
 BILLING CODE 4510-26-P