Application — National Federation of Independent Business, et al., Applicants v. Department of Labor, Occupational Safety and Health Administration, et al.

Appendix 1

App. 1

61402

Federal Register / Vol. 86, No. 212 / Friday, November 5, 2021 / Rules and Regulations

DEPARTMENT OF LABOR

Occupational Safety and Health

Administration

29 CFR Parts 1910, 1915, 1917, 1918,

1926, and 1928

[Docket No. OSHA–2021–0007]

RIN 1218–AD42

COVID–19 Vaccination and Testing;

Emergency Temporary Standard

AGENCY: Occupational Safety and Health

Administration (OSHA), Department of

Labor.

ACTION: Interim final rule; request for

comments.

SUMMARY: The Occupational Safety and

Health Administration (OSHA) is

issuing an emergency temporary

standard (ETS) to protect unvaccinated

employees of large employers (100 or

more employees) from the risk of

contracting COVID–19 by strongly

encouraging vaccination. Covered

employers must develop, implement,

and enforce a mandatory COVID–19

vaccination policy, with an exception

for employers that instead adopt a

policy requiring employees to either get

vaccinated or elect to undergo regular

COVID–19 testing and wear a face

covering at work in lieu of vaccination.

DATES: The rule is effective November 5,

2021. The incorporation by reference of

certain publications listed in the rule is

approved by the Director of the Federal

Register as of November 5, 2021.

Compliance dates: Compliance dates

for specific provisions are in 29 CFR

1910.501(m).

Comments: Written comments,

including comments on any aspect of

this ETS and whether this ETS should

become a final rule, must be submitted

by December 6, 2021 in Docket No.

OSHA–2021–0007. Comments on the

information collection determination

described in Additional Requirements

(Section V.K. of this preamble) (OMB

review under the Paperwork Reduction

Act of 1995) may be submitted by

January 4, 2022 in Docket No. OSHA–

2021–0008.

ADDRESSES: In accordance with 28

U.S.C. 2112(a), the Agency designates

Edmund C. Baird, the Associate

Solicitor for Occupational Safety and

Health, Office of the Solicitor, U.S.

Department of Labor, to receive

petitions for review of the ETS. Service

can be accomplished by email to zzSOLCovid19-ETS@dol.gov.

Written comments. You may submit

comments and attachments, identified

by Docket No. OSHA–2021–0007,

electronically at www.regulations.gov,

which is the Federal e-Rulemaking

Portal. Follow the online instructions

for making electronic submissions.

Instructions: All submissions must

include the agency’s name and the

docket number for this rulemaking

(Docket No. OSHA–2021–0007). All

comments, including any personal

information you provide, are placed in

the public docket without change and

may be made available online at

www.regulations.gov. Therefore, OSHA

cautions commenters about submitting

information they do not want made

available to the public, or submitting

materials that contain personal

information (either about themselves or

others), such as Social Security

Numbers and birthdates.

Docket: To read or download

comments or other material in the

docket, go to Docket No. OSHA–2021–

0007 at www.regulations.gov. All

comments and submissions are listed in

the www.regulations.gov index;

however, some information (e.g.,

copyrighted material) is not publicly

available to read or download through

that website. All comments and

submissions, including copyrighted

material, are available for inspection

through the OSHA Docket Office.

Documents submitted to the docket by

OSHA or stakeholders are assigned

document identification numbers

(Document ID) for easy identification

and retrieval. The full Document ID is

the docket number plus a unique fourdigit code. OSHA is identifying

supporting information in this ETS by

author name and publication year, when

appropriate. This information can be

used to search for a supporting

document in the docket at http://

www.regulations.gov. Contact the OSHA

Docket Office at 202–693–2350 (TTY

number: 877–889–5627) for assistance

in locating docket submissions.

FOR FURTHER INFORMATION CONTACT:

General information and press

inquiries: Contact Frank Meilinger,

OSHA Office of Communications, U.S.

Department of Labor; telephone (202)

693–1999; email OSHAComms@dol.gov.

For technical inquiries: Contact

Andrew Levinson, OSHA Directorate of

Standards and Guidance, U.S.

Department of Labor; telephone (202)

693–1950; email ETS@dol.gov.

SUPPLEMENTARY INFORMATION: The

preamble to the ETS on COVID–19

vaccination and testing follows this

outline:

Table of Contents

I. Executive Summary and Request for

Comment

A. Executive Summary

B. Request for Comment

II. Pertinent Legal Authority

III. Rationale for the ETS

A. Grave Danger

B. Need for the ETS

IV. Feasibility

A. Technological Feasibility

B. Economic Analysis

V. Additional Requirements

VI. Summary and Explanation

A. Purpose

B. Scope and Application

C. Definitions

D. Employer Policy on Vaccination

E. Determination of Employee Vaccination

Status

F. Employer Support for Employee

Vaccination

G. COVID–19 Testing for Employees Who

Are Not Fully Vaccinated

H. Employee Notification to Employer of a

Positive COVID–19 Test and Removal

I. Face Coverings

J. Information Provided to Employees

K. Reporting COVID–19 Fatalities and

Hospitalizations to OSHA

L. Availability of Records

M. Dates

N. Severability

O. Incorporation by Reference

VII. Authority and Signature

I. Executive Summary and Request for

Comment

A. Executive Summary

This ETS is based on the requirements

of the Occupational Safety and Health

Act (OSH Act or Act) and legal

precedent arising under the Act. Under

section 6(c)(1) of the OSH Act, 29 U.S.C.

655(c)(1), OSHA shall issue an ETS if

the agency determines that employees

are subject to grave danger from

exposure to substances or agents

determined to be toxic or physically

harmful or from new hazards, and an

ETS is necessary to protect employees

from such danger. These legal

requirements are more fully discussed

in Pertinent Legal Authority (Section II.

of this preamble). This ETS does not

apply to workplaces subject to E.O.

14042 on Requiring Coronavirus Disease

2019 Vaccination for Federal

Contractors. In addition, OSHA will

treat federal agencies’ compliance with

E.O. 14043, and the Safer Federal

Workforce Task Force guidance issued

under section 4(e) of Executive Order

13991 and section 2 of Executive Order

14043, as sufficient to meet their

obligations under the OSH Act and E.O.

12196.

COVID–19 has killed over 725,000

people in the United States in less than

two years, and infected millions more

(CDC, October 18, 2021—Cumulative

US Deaths). The pandemic continues to

affect workers and workplaces. While

COVID–19 vaccines authorized or

App. 2

Federal Register / Vol. 86, No. 212 / Friday, November 5, 2021 / Rules and Regulations

approved by the U.S. Food and Drug

Administration (FDA) effectively protect

vaccinated individuals against severe

illness and death from COVID–19,

unvaccinated individuals remain at

much higher risk of severe health

outcomes from COVID–19. Further,

unvaccinated workers are much more

likely to contract and transmit COVID–

19 in the workplace than vaccinated

workers. OSHA has determined that

many employees in the U.S. who are not

fully vaccinated against COVID–19 face

grave danger from exposure to SARS–

CoV–2 in the workplace. This finding of

grave danger is based on the severe

health consequences associated with

exposure to the virus along with

evidence demonstrating the

transmissibility of the virus in the

workplace and the prevalence of

infections in employee populations, as

discussed in Grave Danger (Section

III.A. of this preamble).

OSHA has also determined that an

ETS is necessary to protect

unvaccinated workers from the risk of

contracting COVID–19 at work, as

discussed in Need for the ETS (Section

III.B. of this preamble). At the present

time, workers are becoming seriously ill

and dying as a result of occupational

exposures to COVID–19, when a simple

measure, vaccination, can largely

prevent those deaths and illnesses. The

ETS protects these workers through the

most effective and efficient control

available—vaccination—and further

protects workers who remain

unvaccinated through required regular

testing, use of face coverings, and

removal of all infected employees from

the workplace. OSHA also concludes,

based on its enforcement experience

during the pandemic to date, that

continued reliance on existing standards

and regulations, the General Duty

Clause of the OSH Act, 29 U.S.C.

654(a)(1), and workplace guidance, in

lieu of an ETS, is not adequate to protect

unvaccinated employees from the grave

danger of being infected by, and

suffering death or serious health

consequences from, COVID–19.

OSHA will continue to monitor trends

in COVID–19 infections and death as

more of the workforce and the general

population become fully vaccinated

against COVID–19 and the pandemic

continues to evolve. Where OSHA finds

a grave danger from the virus no longer

exists for the covered workforce (or

some portion thereof), or new

information indicates a change in

measures necessary to address the grave

danger, OSHA will update this ETS, as

appropriate.

This ETS applies to employers with a

total of 100 or more employees at any

time the standard is in effect. In light of

the unique occupational safety and

health dangers presented by COVID–19,

and against the backdrop of the

uncertain economic environment of a

pandemic, OSHA is proceeding in a

stepwise fashion in addressing the

emergency this rule covers. OSHA is

confident that employers with 100 or

more employees have the administrative

capacity to implement the standard’s

requirements promptly, but is less

confident that smaller employers can do

so without undue disruption. OSHA

needs additional time to assess the

capacity of smaller employers, and is

seeking comment to help the agency

make that determination. Nonetheless,

the agency is acting to protect workers

now in adopting a standard that will

reach two-thirds of all private-sector

workers in the nation, including those

working in the largest facilities, where

the most deadly outbreaks of COVID–19

can occur.

The agency has also evaluated the

feasibility of this ETS and has

determined that the requirements of the

ETS are both economically and

technologically feasible, as outlined in

Feasibility (Section IV. of this

preamble). The specific requirements of

the ETS are outlined and described in

Summary and Explanation (Section VI.

of this preamble).

B. Request for Comment

Although this ETS takes effect

immediately, it also serves as a proposal

under Section 6(b) of the OSH Act (29

U.S.C. 655(b)) for a final standard.

Accordingly, OSHA seeks comment on

all aspects of this ETS and whether it

should be adopted as a final standard.

OSHA encourages commenters to

explain why they prefer or disfavor

particular policy choices, and include

any relevant studies, experiences,

anecdotes or other information that may

help support the comment. In

particular, OSHA seeks comments on

the following topics:

1. Employers with fewer than 100

employees. As noted above and fully

discussed in the Summary and

Explanation for Scope and Application

(Section VI.B. of this preamble), OSHA

has implemented a 100-employee

threshold for the requirements of this

standard to focus the ETS on companies

that OSHA is confident will have

sufficient administrative systems in

place to comply quickly with the ETS.

The agency is moving in a stepwise

fashion on the short timeline

necessitated by the danger presented by

COVID–19 while soliciting stakeholder

comment and additional information to

determine whether to adjust the scope

61403

of the ETS to address smaller employers

in the future. OSHA seeks information

about the ability of employers with

fewer than 100 employees to implement

COVID–19 vaccination and/or testing

programs. Have you instituted

vaccination mandates (with or without

alternatives), or requirements for regular

COVID–19 testing or face covering use?

What have been the benefits of your

approach? What challenges have you

had or could you foresee in

implementing such programs? Is there

anything specific to your industry, or

the size of your business, that poses

particular obstacles in implementing the

requirements in this standard? How

much time would it take, what types of

costs would you incur, and how much

would it cost for you to implement such

requirements?

2. Significant Risk. If OSHA were to

finalize a rule based on this ETS, it

would be a standard adopted under 6(b)

of the OSH Act, which requires a

finding of significant risk from exposure

to COVID–19. As discussed more fully

in Pertinent Legal Authority (Section II.

of this preamble), this is a lower

showing of risk than grave danger, the

finding required to issue a 6(c)

emergency temporary standard. How

should the scope of the rule change to

address the significant risk posed by

COVID–19 in the workplace? Should

portions of the rule, such as face

coverings, apply to fully vaccinated

persons?

3. Prior COVID–19 infections. OSHA

determined that workers who have been

infected with COVID–19 but have not

been fully vaccinated still face a grave

danger from workplace exposure to

SARS–CoV–2. This is an area of ongoing

scientific inquiry. Given scientific

uncertainty and limitations in testing for

infection and immunity, OSHA is

concerned that it would be infeasible for

employers to operationalize a standard

that would permit or require an

exception from vaccination or testing

and face covering based on prior

infection with COVID–19. Is there

additional scientific information on this

topic that OSHA should consider as it

determines whether to proceed with a

permanent rule?

In particular, what scientific criteria

can be used to determine whether a

given employee is sufficiently protected

against reinfection? Are there any

temporal limits associated with this

criteria to account for potential

reductions in immunity over time? Do

you require employees to provide

verification of infection with COVID–

19? If so, what kinds of verification do

you accept (i.e., PCR testing, antigen

testing, etc.)? What challenges have you

App. 3

61404

Federal Register / Vol. 86, No. 212 / Friday, November 5, 2021 / Rules and Regulations

experienced, if any, in operationalizing

such an exception?

4. Experience with COVID–19

vaccination policies. Should OSHA

impose a strict vaccination mandate

(i.e., all employers required to

implement mandatory vaccination

policies as defined in this ETS) with no

alternative compliance option? OSHA

seeks information on COVID–19

vaccination policies that employers

have implemented to protect workers. If

you have implemented a COVID–19

vaccination policy:

(a) When did you implement it, and

what does your policy require? Was

vaccination mandatory or voluntary

under the policy? Do you offer

vaccinations on site? What costs

associated with vaccination did you

cover under the policy? What

percentage of your workforce was

vaccinated as a result? Do you offer paid

leave for receiving a vaccination? If

vaccination is mandatory, have

employees been resistant and if so what

steps were required to enforce the

policy?

(b) How did you verify that employees

were vaccinated? Are there other

reliable means of vaccination

verification not addressed by the ETS

that should be included? Did you allow

attestation where the employee could

not find other proof, and if so, have you

experienced any difficulties with this

approach? Have you experienced any

issues with falsified records of

vaccination, and if so, how did you deal

with them?

(c) Have you experienced a decrease

in infection rates or outbreaks after

implementing this policy?

(d) If you have received any requests

for reasonable accommodation from

vaccination, what strategies did you

implement to address the

accommodation and ensure worker

safety (e.g., telework, working in

isolation, regular testing and the use of

face coverings)?

5. COVID–19 testing and removal.

OSHA seeks information on COVID–19

testing and removal practices

implemented to protect workers.

(a) Do you have a testing and removal

policy in your workplace and, if so,

what does it require? How often do you

require testing and what types of testing

do you use (e.g., at-home tests, tests

performed at laboratories, tests

performed at your worksites)? What

costs have you incurred as part of your

testing and removal policies? Do you

have difficulty in finding adequate

availability of tests? How often? Have

you experienced any issues with

falsified test results, and if so, how did

you deal with them? Have you

experienced other difficulties in

implementing a testing and removal

scheme, including the length of time to

obtain COVID–19 test results? Do you

offer paid leave for testing?

(b) How often have you detected and

removed COVID–19 positive employees

from the workplace under this policy?

Do you provide paid leave and job

protection to employees you remove for

this reason?

(c) Should OSHA require testing more

often than on a weekly basis?

6. Face coverings. As discussed in the

Summary and Explanation for Face

Coverings (Section VI.I. of this

preamble), ASTM released a

specification standard on February 15,

2021, to establish a national standard

baseline for barrier face coverings

(ASTM F3502–21). Should OSHA

require the use of face coverings

meeting the ASTM F3502–21 standard

instead of the face coverings specified

by the ETS? If so, should OSHA also

require that such face coverings meet

the NIOSH Workplace Performance or

Workplace Performance Plus criteria

(see CDC, September 23, 2021)? Are

there particular workplace settings in

which face coverings meeting one

standard should be favored over

another? Are there alternative criteria

OSHA should consider for face

coverings instead of the F3502–21

standard or NIOSH Workplace

Performance or Workplace Performance

Plus criteria? Is there sufficient capacity

to supply face coverings meeting

F3502–01 and/or NIOSH Workplace

Performance or Workplace Performance

Plus criteria to all employees covered by

the ETS? What costs have you incurred

as part of supplying employees with

face coverings meeting the appropriate

criteria?

7. Other controls. This ETS requires

employees to either be fully vaccinated

against COVID–19 or be tested weekly

and wear face coverings, based on the

type of policy their employer adopts. It

stops short of requiring the full suite of

workplace controls against SARS–CoV–

2 transmission recommended by OSHA

and the CDC, including distancing,

barriers, ventilation, and sanitation. As

OSHA explained in Need for the ETS

(Section III.B. of this preamble), OSHA

has determined that it needs more

information before imposing these

requirements on the entire scope of

industries and employers covered by the

standard. OSHA is interested in hearing

from employers about their experience

in implementing a full suite of

workplace controls against COVID–19.

What measures have you taken to

protect employees against COVID–19 in

your workplace? Are there controls that

you attempted to employ but found

ineffective or infeasible? What are they?

Why did you conclude that they were

they ineffective or infeasible; for

example, are there particular aspects of

your workplace or industry that make

certain controls infeasible? Do you

require both fully vaccinated and

unvaccinated employees to comply with

these controls? Have you experienced a

reduction in infection rates or outbreaks

since implementing these controls?

8. Educational materials. Have you

implemented any policies or provided

any information that has been helpful in

encouraging an employee to be

vaccinated?

9. Feasibility and health impacts. Do

you have any experience or data that

would inform OSHA’s estimates in its

economic feasibility analysis or any of

the assumptions or estimates used in

OSHA’s identification of the number of

hospitalizations prevented and lives

saved from its health impacts analysis

(see OSHA, October 2021c)?

References

Centers for Disease Control and Prevention

(CDC). (2021, October 18). COVID Data

Tracker. https://covid.cdc.gov/coviddata-tracker/. (CDC, October 18, 2021)

Centers for Disease Control and Prevention

(CDC). (2021, September 23). Types of

Masks and Respirators. https://

www.cdc.gov/coronavirus/2019-ncov/

prevent-getting-sick/types-of-masks.html.

(CDC, September 23, 2021)

Occupational Safety and Health

Administration (OSHA). (2021c,

October). Health Impacts of the COVID–

19 Vaccination and Testing ETS. (OSHA,

October 2021c)

II. Pertinent Legal Authority

The purpose of the Occupational

Safety and Health Act of 1970 (OSH

Act), 29 U.S.C. 651 et seq., is ‘‘to assure

so far as possible every working man

and woman in the Nation safe and

healthful working conditions and to

preserve our human resources.’’ 29

U.S.C. 651(b). To this end, Congress

authorized the Secretary of Labor

(Secretary) to promulgate and enforce

occupational safety and health

standards under sections 6(b) and (c) of

the OSH Act.1 29 U.S.C. 655(b). These

provisions provide bases for issuing

occupational safety and health

standards under the Act. Once OSHA

has established as a threshold matter

that a health standard is necessary

under section 6(b) or (c)—i.e., to reduce

1 The Secretary has delegated most of his duties

under the OSH Act to the Assistant Secretary of

Labor for Occupational Safety and Health.

Secretary’s Order 08–2020, 85 FR 58393 (Sept. 18,

2020). This section uses the terms Secretary and

OSHA interchangeably.

App. 4

Federal Register / Vol. 86, No. 212 / Friday, November 5, 2021 / Rules and Regulations

a significant risk of material health

impairment, or a grave danger to

employee health—the Act gives the

Secretary ‘‘almost unlimited discretion

to devise means to achieve the

congressionally mandated goal’’ of

protecting employee health, subject to

the constraints of feasibility. See United

Steelworkers of Am. v. Marshall, 647

F.2d 1189, 1230 (D.C. Cir. 1981). A

standard’s individual requirements need

only be ‘‘reasonably related’’ to the

purpose of ensuring a safe and healthful

working environment. Id. at 1237, 1241;

see also Forging Indus. Ass’n v. Sec’y of

Labor, 773 F.2d 1436, 1447 (4th Cir.

1985). OSHA’s authority to regulate

employers is hedged by constitutional

considerations and, pursuant to section

4(b)(1) of the OSH Act, the regulations

and enforcement policies of other

federal agencies. See, e.g., Chao v.

Mallard Bay Drilling, Inc., 534 U.S. 235,

241 (2002).

The OSH Act in section 6(c)(1) states

that the Secretary ‘‘shall’’ issue an

emergency temporary standard (ETS)

upon a finding that the ETS is necessary

to address a grave danger to workers.

See 29 U.S.C. 655(c). In particular, the

Secretary shall provide, without regard

to the requirements of chapter 5, title 5,

United States Code, for an emergency

temporary standard to take immediate

effect upon publication in the Federal

Register if the Secretary makes two

determinations: That employees are

exposed to grave danger from exposure

to substances or agents determined to be

toxic or physically harmful or from new

hazards, and that such emergency

standard is necessary to protect

employees from such danger. 29 U.S.C.

655(c)(1). A separate section of the OSH

Act, section 8(c), authorizes the

Secretary to prescribe regulations

requiring employers to make, keep, and

preserve records that are necessary or

appropriate for the enforcement of the

Act. 29 U.S.C. 657(c)(1). Section 8(c)

also provides that the Secretary shall

require employers to keep records of,

and report, work-related deaths and

illnesses. 29 U.S.C. 657(c)(2).

The ETS provision, section 6(c)(1),

exempts the Secretary from procedural

requirements contained in the OSH Act

and the Administrative Procedure Act,

including those for public notice,

comments, and a rulemaking hearing.

See, e.g., 29 U.S.C. 655(b)(3); 5 U.S.C.

552, 553.

The Secretary must issue an ETS in

situations where employees are exposed

to a ‘‘grave danger’’ and immediate

action is necessary to protect those

employees from such danger. 29 U.S.C.

655(c)(1); Pub. Citizen Health Research

Grp. v. Auchter, 702 F.2d 1150, 1156

(D.C. Cir. 1983). The determination of

what exact level of risk constitutes a

‘‘grave danger’’ is a ‘‘policy

consideration that belongs, in the first

instance, to the Agency.’’ Asbestos Info.

Ass’n, 727 F.2d at 425 (accepting

OSHA’s determination that eighty lives

at risk over six months was a grave

danger); Indus. Union Dep’t, AFL–CIO v.

Am. Petroleum Inst., 448 U.S. 607, 655

n.62 (1980). However, a ‘‘grave danger’’

represents a risk greater than the

‘‘significant risk’’ that OSHA must show

in order to promulgate a permanent

standard under section 6(b) of the OSH

Act, 29 U.S.C. 655(b). Int’l Union,

United Auto., Aerospace, & Agr.

Implement Workers of Am., UAW v.

Donovan, 590 F. Supp. 747, 755–56

(D.D.C. 1984), adopted, 756 F.2d 162

(D.C. Cir. 1985); see also Indus. Union

Dep’t, AFL–CIO, 448 U.S. at 640 n.45

(noting the distinction between the

standard for risk findings in permanent

standards and ETSs).

In determining the type of health

effects that may constitute a ‘‘grave

danger’’ under the OSH Act, the Fifth

Circuit emphasized ‘‘the danger of

incurable, permanent, or fatal

consequences to workers, as opposed to

easily curable and fleeting effects on

their health.’’ Fla. Peach Growers Ass’n,

Inc. v. U. S. Dep’t of Labor, 489 F.2d

120, 132 (5th Cir. 1974). Although the

findings of grave danger and necessity

must be based on evidence of ‘‘actual,

prevailing industrial conditions,’’ see

Int’l Union, 590 F. Supp. at 751, when

OSHA determines that exposure to a

particular hazard would pose a grave

danger to workers, OSHA can assume an

exposure to a grave danger wherever

that hazard is present in a workplace.

Dry Color Mfrs. Ass’n, Inc. v. Dep’t of

Labor, 486 F.2d 98, 102 n.3 (3d Cir.

1973).

In demonstrating whether OSHA had

shown that an ETS is necessary, the

Fifth Circuit considered whether OSHA

had another available means of

addressing the risk that would not

require an ETS. Asbestos Info. Ass’n,

727 F.2d at 426 (holding that necessity

had not been proven where OSHA could

have increased enforcement of alreadyexisting standards to address the grave

risk to workers from asbestos exposure).

Additionally, a standard must be both

economically and technologically

feasible in order to be ‘‘reasonably

necessary and appropriate’’ under

section 3(8) and, by inference,

‘‘necessary’’ under section 6(c)(1)(B) of

the Act. Cf. Am. Textile Mfrs. Inst., Inc.

v. Donovan, 452 U.S. 490, 513 n.31

(1981) (noting ‘‘any standard that was

not economically or technologically

feasible would a fortiori not be

61405

‘reasonably necessary or appropriate’ ’’

as required by the OSH Act’s definition

of ‘‘occupational safety and health

standard’’ in section 3(8)); see also

Florida Peach Growers, 489 F.2d at 130

(recognizing that the promulgation of

any standard, including an ETS, must

account for its economic effect).

However, given that section 6(c) is

aimed at enabling OSHA to protect

workers in emergency situations, the

agency is not required to make a

feasibility showing with the same rigor

as in ordinary section 6(b) rulemaking.

Asbestos Info. Ass’n, 727 F.2d at 424

n.18.

On judicial review of an ETS, OSHA

is entitled to great deference on the

determinations of grave danger and

necessity required under section 6(c)(1).

See, e.g., Pub. Citizen Health Research

Grp., 702 F.2d at 1156; Asbestos Info.

Ass’n, 727 F.2d at 422 (judicial review

of these legislative determinations

requires deference to the agency); cf.

Am. Dental Ass’n v. Martin, 984 F.2d

823, 831 (7th Cir. 1993) (‘‘the duty of a

reviewing court of generalist judges is

merely to patrol the boundary of

reasonableness’’). These determinations

are ‘‘essentially legislative and rooted in

inferences from complex scientific and

factual data.’’ Pub. Citizen Health

Research Grp., 702 F.2d at 1156. The

agency is not required to support its

conclusions ‘‘with anything

approaching scientific certainty,’’ Indus.

Union Dep’t, AFL–CIO, 448 U.S. at 656,

and has the ‘‘prerogative to choose

between conflicting evidence.’’ Asbestos

Info. Ass’n, 727 F.2d at 425.

The determinations of the Secretary in

issuing standards under section 6 of the

OSH Act, including ETSs, must be

affirmed if supported by ‘‘substantial

evidence in the record considered as a

whole.’’ 29 U.S.C. 655(f). The Supreme

Court described substantial evidence as

‘‘such relevant evidence as a reasonable

mind might accept as adequate to

support a conclusion.’’ Am. Textile

Mfrs. Inst., 452 U.S. at 522–23 (quoting

Universal Camera Corp. v. NLRB, 340

U.S. 474, 477 (1951)). The Court also

noted that ‘‘the possibility of drawing

two inconsistent conclusions from the

evidence does not prevent an

administrative agency’s finding from

being supported by substantial

evidence.’’ Id. at 523 (quoting Consolo

v. FMC, 383 U.S. 607, 620 (1966)). The

Fifth Circuit, recognizing the size and

complexity of the rulemaking record

before it in the case of OSHA’s ETS for

organophosphorus pesticides, stated

that a court’s function in reviewing an

ETS to determine whether it meets the

substantial evidence standard is

‘‘basically [to] determine whether the

App. 5

61406

Federal Register / Vol. 86, No. 212 / Friday, November 5, 2021 / Rules and Regulations

Secretary carried out his essentially

legislative task in a manner reasonable

under the state of the record before

him.’’ Fla Peach Growers Ass’n, 489

F.2d at 129.

Although Congress waived the

ordinary rulemaking procedures in the

interest of ‘‘permitting rapid action to

meet emergencies,’’ section 6(e) of the

OSH Act, 29 U.S.C. 655(e), requires

OSHA to include a statement of reasons

for its action when it issues any

standard. Dry Color Mfrs., 486 F.2d at

105–06 (finding OSHA’s statement of

reasons inadequate). By requiring the

agency to articulate its reasons for

issuing an ETS, the requirement acts as

‘‘an essential safeguard to emergency

temporary standard-setting.’’ Id. at 106.

However, the Third Circuit noted that it

did not require justification of ‘‘every

substance, type of use or production

technique,’’ but rather a ‘‘general

explanation’’ of why the standard is

necessary. Id. at 107.

ETSs are, by design, temporary in

nature. Under section 6(c)(3), an ETS

serves as a proposal for a permanent

standard in accordance with section 6(b)

of the OSH Act (permanent standards),

and the Act calls for the permanent

standard to be finalized within six

months after publication of the ETS. 29

U.S.C. 655(c)(3); see Fla. Peach Growers

Ass’n, 489 F.2d at 124. The ETS is

effective ‘‘until superseded by a

standard promulgated in accordance

with’’ section 6(c)(3). 29 U.S.C.

655(c)(2).

Section 6(c)(1) states that the

Secretary ‘‘shall’’ provide for an ETS

when OSHA makes the prerequisite

findings of grave danger and necessity.

See Pub. Citizen Health Research Grp.,

702 F.2d at 1156 (noting the mandatory

language of section 6(c)). OSHA is

entitled to great deference in its

determinations, and it must also

account for ‘‘the fact that ‘the interests

at stake are not merely economic

interests in a license or a rate structure,

but personal interests in life and

health.’ ’’ Id. (quoting Wellford v.

Ruckelshaus, 439 F.2d 598, 601 (D.C.

Cir. 1971)).

When OSHA issues a standard

pursuant to section 6—whether

permanent or an ETS—section 18 of the

OSH Act provides that OSHA’s standard

preempts any state occupational safety

or health standard ‘‘relating to [the

same] occupational safety or health

issue’’ as the Federal standard. 29 U.S.C.

667(b); see also Gade v. Nat’l Solid

Wastes Mgmt. Ass’n, 505 U.S. 88, 97

(1992). A state can avoid preemption

only if it submits, and receives Federal

approval for, a state plan for the

development and enforcement of

standards pursuant to section 18 of the

Act, which must be ‘‘at least as

effective’’ as the Federal standards. 29

U.S.C. 667; Indus. Truck Ass’n v. Henry,

125 F.3d 1305, 1311 (9th Cir. 1997).

However, the OSH Act does not

preempt state laws of ‘‘general

applicability’’ that regulate workers and

non-workers alike, so long as they do

not conflict with an OSHA standard.

Gade, 505 U.S. at 107.

As discussed in detail elsewhere in

this preamble, OSHA has determined

that a grave danger exists necessitating

a new ETS (see Grave Danger and Need

for the ETS, Sections III.A. and III.B. of

this preamble), and that compliance

with this ETS is feasible for covered

employers (see Feasibility, Section IV. of

this preamble). OSHA has also provided

a more detailed explanation of each

provision of this ETS in Summary and

Explanation (Section VI. of this

preamble). In addition, OSHA wishes to

provide here some general guidance on

its legal authority to regulate COVID–19

hazards, and for particular provisions of

this ETS.

As a threshold matter, OSHA’s

authority to regulate workplace

exposure to biological hazards like

SARS–CoV–2 is well-established.

Section 6(b)(5) of the OSH Act uses

similar language to section 6(c)(1)(A):

The former sets forth requirements for

promulgating permanent standards

addressing ‘‘toxic materials or harmful

physical agents,’’ and the latter

authorizes OSHA to promulgate an ETS

addressing ‘‘substances or agents

determined to be toxic or physically

harmful’’ (as well as ‘‘new hazards’’).

OSHA has consistently identified

biological hazards similar to SARS–

CoV–2, as well as SARS–CoV–2 itself, to

be ‘‘toxic materials or harmful physical

agents’’ under the Act. Indeed, in its

exposure and medical records access

regulation, OSHA has defined ‘‘toxic

materials or harmful physical agents’’ to

include ‘‘any . . . biological agent

(bacteria, virus, fungus, etc.)’’ for which

there is evidence that it poses a chronic

or acute health hazard. 29 CFR

1910.1020(c)(13). And in addition to

previously regulating exposure to

SARS–CoV–2 as a new and physically

harmful agent in the Healthcare ETS

(see, e.g., 86 FR at 32381), OSHA has

also previously regulated biological

hazards like SARS–CoV–2 as health

hazards under section 6(b)(5), for

example in the Bloodborne Pathogens

(BBP) standard, 29 CFR 1910.1030,

which addresses workplace exposure to

HIV and Hepatitis B. The BBP standard

was upheld (except as to application in

certain limited industries) in American

Dental Association, which observed that

‘‘the infectious character’’ of the

regulated bloodborne diseases might

warrant ‘‘more regulation than would be

necessary in the case of a

noncommunicable disease.’’ 984 F.2d at

826. In addition, in the preamble to the

respiratory protection standard, 29 CFR

1910.134, which was also promulgated

under section 6(b)(5), ‘‘OSHA

emphasize[d] that [the] respiratory

protection standard does apply to

biological hazards.’’ Respiratory

Protection, 63 FR 1152–01, 1180 (Jan. 8,

1998) (citing Mahone Grain Corp., 10

BNA OSHC 1275 (No. 77–3041, 1981)).

In addition to being a physically

harmful agent covered by section

6(c)(1)(A), SARS–CoV–2 is also, without

question, a ‘‘new hazard’’ covered by

this provision, as discussed in more

detail in Grave Danger (Section III.A. of

this preamble). SARS–CoV–2 was not

known to exist until January 2020, and

since then more than 725,000 people

have died from COVID–19 in the U.S.

alone (CDC, October 18, 2021—

Cumulative US Deaths).

Turning to specific provisions of this

standard, the vaccination requirements

in this ETS are also well within the

bounds of OSHA’s authority.

Vaccination can be a critical tool in the

pursuit of health and safety goals,

particularly in response to an infectious

and highly communicable disease. See,

e.g., Jacobson v. Commonwealth of

Mass., 197 U.S. 11, 27–28 (1905)

(recognizing use of smallpox vaccine as

a reasonable measure to protect public

health and safety); Klaassen v. Trustees

of Ind. Univ., 7 F.4th 592, 593 (7th Cir.

2021) (citing Jacobson and noting that

vaccination may be an appropriate

safety measure against SARS–CoV–2 as

‘‘[v]accination protects not only the

vaccinated persons but also those who

come in contact with them’’). And the

OSH Act itself explicitly acknowledges

that such treatments might be necessary,

in some circumstances. 29 U.S.C.

669(a)(5) (providing in the Act’s

provisions on research and related

activities conducted by the Secretary of

Health and Human Services to aid

OSHA in its formulation of health and

safety standards that ‘‘[n]othing in this

or any other provision of this Act shall

be deemed to authorize or require

medical examination, immunization, or

treatment for those who object thereto

on religious grounds, except where such

is necessary for the protection of the

health or safety of others.’’ (emphasis

added)). In recognition of the health and

safety benefits provided by vaccination,

OSHA has previously exercised its

authority to promulgate vaccine-related

requirements in the COVID–19

Healthcare ETS (29 CFR 1910.502(m))

App. 6

Federal Register / Vol. 86, No. 212 / Friday, November 5, 2021 / Rules and Regulations

and the BBP standard (29 CFR

1910.1030(f)). The BBP standard

illustrates congressional understanding

that the statutory delegation of authority

to OSHA to issue standards includes

authority for vaccine provisions, where

appropriate. See Public Law 102–170,

Title I, Section 100, 105 Stat. 1107

(1991) (directing OSHA to complete the

BBP rulemaking by a date certain, and

providing that if OSHA did not do so,

the proposed rule, which included a

vaccine provision, would become the

final standard).

Additionally, OSHA’s authority to

require employers to bear the costs of

particular provisions of a standard is

solidly grounded in the OSH Act. The

Act reflects Congress’s determination

that the costs of compliance with the

Act and OSHA standards are part of the

cost of doing business and OSHA may

foreclose employers from shifting those

costs to employees. See Am. Textile

Mfrs. Inst., 452 U.S. at 514; Phelps

Dodge Corp. v. OSHRC, 725 F.2d 1237,

1239–40 (9th Cir. 1984); see also Sec’y

of Labor v. Beverly Healthcare-Hillview,

541 F.3d 193 (3d Cir. 2008). Consistent

with this authority, OSHA has largely

required employers to bear the costs of

the provisions of this ETS, including the

typical costs associated with

vaccination. The allocation of

vaccination costs to employers in this

ETS is similar to OSHA’s treatment of

vaccine-related costs in the COVID–19

Healthcare ETS and the BBP standards.

See 29 CFR 1910.502(m), (p); 29 CFR

1910.1030(f)(1)(ii)(A).

The OSH Act provides OSHA with

discretion, however, to decide whether

to impose certain costs—such as those

related to medical examinations or other

tests—on employers ‘‘[w]here [it

determines that such costs are]

appropriate.’’ 29 U.S.C. 655(b)(7). OSHA

has determined that for purposes of this

ETS, it would not be ‘‘appropriate’’ to

impose on employers any costs

associated with COVID–19 testing for

employees who choose not to be

vaccinated. For most of the agency’s

existing standards containing medical

testing and removal provisions, OSHA

has found it necessary to impose the

costs of such provisions on employers

in order to remove barriers to employee

participation in medical examinations

that are critical to effectuating the

standards’ safety and health protections.

See United Steelworkers of Am., 647

F.2d at 1229–31, 1237–38. However, as

explained in greater detail elsewhere in

this preamble (see Need for the ETS,

Section III.B. of this preamble), the

ETS’s safety and health protections are

best effectuated by employee

vaccination, not testing. Accordingly,

OSHA only requires employers to bear

the costs of employee compliance with

the preferred, and more protective,

vaccination provision, but not costs

associated with testing. The agency does

not believe it appropriate to impose the

costs of testing on an employer where

an employee has made an individual

choice to pursue a less protective

option. For the same reasons, OSHA has

also determined that it is not

appropriate to require employers to pay

for face coverings for employees who

choose not to be vaccinated.2

Finally, the Act and its legislative

history ‘‘both demonstrate

unmistakably’’ OSHA’s authority to

require employers to temporarily

remove workers from the workplace to

prevent exposure to a health hazard.

United Steelworkers of Am., 647 F.2d at

1230. And again, this is an authority

OSHA has repeatedly exercised in prior

standards, including in: COVID–19

Healthcare ETS (29 CFR 1910.502); Lead

(29 CFR 1910.1025); Cadmium (29 CFR

1910.1027); Benzene (29 CFR

1910.1028); Formaldehyde (29 CFR

1910.1048); Methylenedianiline (29 CFR

1910.1050); Methylene Chloride (29

CFR 1910.1052); and Beryllium (29 CFR

1910.1024). It is equally appropriate to

impose that obligation here.

For all of these reasons, as well as

those explained more fully in other

areas of this preamble, OSHA has the

authority—and obligation—to

promulgate this ETS.

References

Centers for Disease Control and Prevention

(CDC). (2021, October 18). COVID Data

Tracker. https://covid.cdc.gov/coviddata-tracker/. (CDC, October 18, 2021)

III. Rationale for the ETS

A. Grave Danger

I. Introduction

Section 6(c)(1) of the OSH Act

requires the Secretary to issue an ETS in

situations where employees are exposed

to a ‘‘grave danger’’ and immediate

action is necessary to protect those

employees from such danger (29 U.S.C.

655(c)(1)). Consistent with its legal

duties, OSHA is issuing this ETS to

address the grave danger posed by

occupational exposure to SARS–CoV–2,

2 OSHA notes that while the ETS does not impose

these testing or face covering costs on employers,

in some circumstances employers may be required

to pay for the costs related to testing and/or face

coverings by other laws, regulations, or collectively

negotiated agreements. OSHA has no authority

under the OSH Act to determine whether such

obligations under other laws, regulations, or

agreements might exist.

61407

the virus that causes COVID–19.3 OSHA

has determined that occupational

exposure to SARS–CoV–2, including the

Delta variant (B.1.617.2 and AY

lineages), presents a grave danger to

unvaccinated workers in the U.S., with

several exceptions explained below.4

This finding of grave danger is based on

the science of how the virus spreads, the

transmissibility of the disease in

workplaces, and the serious adverse

health effects, including death, that can

be suffered by those who are diagnosed

with COVID–19. The protections of this

ETS—which will apply, with some

limitations, to a broad range of

workplace settings where exposure to

SARS–CoV–2 may occur—are designed

to protect employees from infection

with SARS–CoV–2 and from the dire,

sometimes fatal, consequences of such

infection.

The fact that COVID–19 is not a

uniquely work-related hazard does not

change the determination that it is a

grave danger to which employees are

exposed, nor does it excuse employers

from their duty to protect employees

from the occupational transmission of

SARS–CoV–2. The OSH Act is intended

to ‘‘assure so far as possible every

working man and woman in the Nation

safe and healthful working conditions’’

(29 U.S.C. 651(b)), and there is nothing

in the Act to suggest that its protections

do not extend to hazards which might

occur outside of the workplace as well

as within. Indeed, COVID–19 is not the

first hazard that OSHA has regulated

that occurs both inside and outside the

workplace. For example, the hazard of

noise is not unique to the workplace,

but the Fourth Circuit has upheld

OSHA’s Occupational Noise Exposure

standard (29 CFR 1910.95) (Forging

Industry Ass’n v. Sec’ of Labor, 773 F.2d

1437, 1444 (4th Cir. 1985)). Diseases

caused by bloodborne pathogens,

including HIV/AIDS and hepatitis B, are

also not unique to the workplace, but

the Seventh Circuit upheld the majority

of OSHA’s Bloodborne Pathogens

standard (29 CFR 1910.1030) (Am.

Dental Ass’n v. Martin, 984 F.2d 823

(7th Cir. 1993)). OSHA’s Sanitation

3 OSHA is defining the grave danger as workplace

exposure to SARS–CoV–2, the virus that causes the

development of COVID–19. COVID–19 is the

disease that can occur in people exposed to SARSCoV–2, and that leads to the health effects

described in this section. This distinction applies

despite OSHA’s use of the terms SARS–CoV–2 and

COVID–19 interchangeably in some parts of this

preamble.

4 OSHA refers to the grave danger from

occupational exposure to SARS–CoV–2 throughout

this document. Those references are intended to

encompass exposure to SARS–CoV–2 and all

variants of SARS–CoV–2, including the Delta

variant.

App. 7

61408

Federal Register / Vol. 86, No. 212 / Friday, November 5, 2021 / Rules and Regulations

standard, 29 CFR 1910.141, which

requires measures such as cleaning,

waste disposal, potable water, toilets,

and washing facilities, addresses

hazards that exist everywhere—both

within and outside of workplaces.

Moreover, employees have more

freedom to control their environment

outside of work, and to make decisions

about their behavior and their contact

with others to better minimize their risk

of exposure. However, during the

workday, while under the control of

their employer, workers may have little

ability to limit contact with coworkers,

clients, members of the public, patients,

and others, any one of whom could

represent a source of exposure to SARS–

CoV–2. OSHA has a mandate to protect

employees from hazards they are

exposed to at work, even if they may be

exposed to similar hazards outside of

work.

As described above in Pertinent Legal

Authority (Section II. of this preamble),

‘‘grave danger’’ indicates a risk that is

more than ‘‘significant’’ (Int’l Union,

United Auto., Aerospace, & Agr.

Implement Workers of Am., UAW v.

Donovan, 590 F. Supp. 747, 755–56

(D.D.C. 1984); Indus. Union Dep’t, AFL–

CIO v. Am. Petroleum Inst., 448 U.S.

607, 640 n.45, 655 (1980) (stating that a

rate of 1 worker in 1,000 workers

suffering a given health effect

constitutes a ‘‘significant’’ risk)). ‘‘Grave

danger,’’ according to one court, refers

to ‘‘the danger of incurable, permanent,

or fatal consequences to workers, as

opposed to easily curable and fleeting

effects on their health’’ (Fla. Peach

Growers Ass’n, Inc. v. U.S. Dep’t of

Labor, 489 F.2d 120, 132 (5th Cir.

1974)). Fleeting effects were described

as nausea, excessive salivation,

perspiration, or blurred vision and were

considered so minor that they often

went unreported; these effects are in

stark contrast with the adverse health

effects of COVID–19 infections, which

are formally referenced as ranging from

‘‘mild’’ to ‘‘critical,’’ 5 but which can

involve significant illness, hospital

stays, ICU care, death, and long-term

health complications for survivors.

Beyond this, however, ‘‘the

determination of what constitutes a risk

worthy of Agency action is a policy

consideration that belongs, in the first

instance, to the Agency’’ (Asbestos Info.

Ass’n/N. Am. v. OSHA, 727 F.2d 415,

425 (5th Cir. 1984)).

In the context of ordinary 6(b)

rulemaking, the Supreme Court has said

5 See the definitions for the different levels of

severity of COVID–19 illness in the National

Institutes of Health’s COVID–19 treatment

guidelines (NIH, October 12, 2021).

that the OSH Act is not a ‘‘mathematical

straitjacket,’’ nor does it require the

agency to support its findings ‘‘with

anything approaching scientific

certainty,’’ particularly when operating

on the ‘‘frontiers of scientific

knowledge’’ (Indus. Union Dep’t, AFL–

CIO v. Am. Petroleum Inst., 448 U.S.

607, 655–56 (1980)). Courts reviewing

OSHA’s determination of grave danger

do so with ‘‘great deference’’ (Pub.

Citizen Health Research Grp. v.

Auchter, 702 F.2d 1150, 1156 (D.C. Cir.

1983)). In one case, the Fifth Circuit, in

reviewing an OSHA ETS for asbestos,

declined to question the agency’s

finding that 80 worker lives at risk

nationwide over six months constituted

a grave danger (Asbestos Info. Ass’n/N.

Am., 727 F.2d at 424). OSHA estimates

that this ETS would save over 6,500

worker lives and prevent over 250,000

hospitalizations over the course of the

next six months (OSHA, October 2021c).

Here, the mortality and morbidity risk to

employees from COVID–19 is so dire

that the grave danger from exposures to

SARS–CoV–2 is clear.

SARS–CoV–2 is both a physically

harmful agent and a new hazard (see 29

U.S.C. 655(c)(1)(A)). The majority of

OSHA’s previous ETSs addressed toxic

substances that had been familiar to the

agency for many years prior to issuance

of the ETS. OSHA’s Healthcare ETS,

issued in response to COVID–19 earlier

this year, is one notable exception. In

most cases, OSHA’s ETSs were issued in

response to new information about

substances that had been used in

workplaces for decades (e.g., Vinyl

Chloride (39 FR 12342 (April 5, 1974));

Benzene (42 FR 22516 (May 3, 1977));

1,2-Dibromo-3-chloropropane (42 FR

45536 (Sept. 9, 1977))). In some cases,

the hazards of the toxic substance were

already so well established that OSHA

promulgated an ETS simply to update

an existing standard (e.g., Vinyl cyanide

(43 FR 2586 (Jan. 17, 1978))). The

COVID–19 Healthcare ETS, which was

issued in June 2021, was the sole

instance in which OSHA issued an ETS

to address a grave danger from a

substance that had only recently come

into existence. Although that action by

the agency was challenged, the case has

not gone to briefing (see United Food &

Commercial Workers Int’l Union, AFL–

CIO, CLC and AFL–CIO v. OSHA, Dep’t

of Labor, D.C. Circuit No. 21–1143).

Thus, no court has had occasion to

examine OSHA’s authority under

section (6)(c) of the OSH Act (29 U.S.C.

655(c)) to address a grave danger from

a ‘‘new hazard.’’ Yet by any measure,

SARS–CoV–2 is a new hazard. Unlike

any of the hazards addressed in

previous ETSs, there were no

documented cases of SARS–CoV–2

infections in the United States until

January 2020. Since then, more than

725,000 people have died in the U.S.

alone (CDC, October 18, 2021—

Cumulative US Deaths). The pandemic

continues to affect workers and

workplaces, with workplace exposures

leading to further exposures among

workers’ families and communities.

Clearly, SARS–CoV–2 is both a

physically harmful agent and a new

hazard that presents a grave danger to

workers in the U.S.

Published on June 21, 2021, OSHA’s

Healthcare ETS (86 FR 32376) was

written in response to the grave danger

posed to healthcare workers in the

United States who faced a heightened

risk of infection from COVID–19. In the

healthcare ETS, OSHA described its

finding of grave danger for healthcare

and healthcare support service workers

(see 86 FR 32381–32412). OSHA now

finds that all unvaccinated workers,

with some exceptions, face a grave

danger from the SARS–CoV–2 virus.6

II. Nature of the Disease

The health effects of symptomatic

COVID–19 illness can range from mild

disease consisting of fever or chills,

cough, and shortness of breath to severe

disease. Severe cases can involve

respiratory failure, blood clots, longterm cardiovascular and neurological

effects, and organ damage, which can

lead to hospitalization, ICU admission,

and death (see 86 FR 32383–32388;

NINDS, September 2, 2021). Even in the

short time since the Healthcare ETS’s

publication in June 2021, the risk posed

by COVID–19 has changed

meaningfully. Since OSHA considered

the impact of COVID–19 when

promulgating the Healthcare ETS, over

135,000 additional Americans have died

from COVID–19, and over 933,000 have

been hospitalized, (CDC, October 18,

2021—Cumulative US Deaths; CDC,

May 28, 2021; CDC, October 18, 2021—

Weekly Review). In August 2021,

COVID–19 was the third leading cause

of death in the United States, trailing

only heart disease and cancer (Ortaliza

et al., August 27, 2021). By September

20, 2021, COVID–19 had killed as many

Americans as the 1918–1919 flu

pandemic (Johnson, September 20,

2021).

While the Healthcare ETS addresses

the risk of illness and death from

6 When OSHA refers to ‘‘unvaccinated’’

individuals in its grave danger finding, it means all

individuals who are not fully vaccinated against

COVID–19, i.e., those who are completely

unvaccinated and those who are partially

vaccinated.

App. 8

Federal Register / Vol. 86, No. 212 / Friday, November 5, 2021 / Rules and Regulations

COVID–19 as the SARS–CoV–2 virus

continues to change over time, it does

not specifically address the increases in

infectiousness and transmission, and

the potentially more severe health

effects, related to the Delta variant. The

rapid rise to predominance of the Delta

variant in the U.S. occurred shortly after

the ETS was published. At this time, the

widespread prevalence of the Delta

variant and its increased

transmissibility have resulted in

increased risk of exposure and disease

relative to the previously-dominant

strains of the SARS–CoV–2 virus.

Adding to the information covered in

the Healthcare ETS, the following

sections provide a brief review of

SARS–CoV–2 and describe the

characteristics of the Delta variant that

are different from previous versions of

SARS–CoV–2 and have changed the

risks posed by COVID–19. The agency

specifically references the material

presented in the Healthcare ETS, which

is still relevant to this analysis, to

support OSHA’s finding of grave danger.

Taken together, the information

available to OSHA demonstrates that

SARS–CoV–2 poses a grave danger to

unvaccinated workers across all

industry sectors.

a. Variants of SARS–CoV–2

Viral mutations have been a serious

concern of scientists, public health

experts, and policymakers from the

beginning of the COVID–19 pandemic.

Viral mutations can affect how a virus

interacts with a cell—altering the virus’s

transmissibility, infection severity, and

sensitivity to vaccines. The U.S.

government’s SARS–CoV–2 Interagency

Group has a variant classification

scheme that defines four classes of

SARS–CoV–2 variants: Variants Being

Monitored (VBM), Variants of Interest

(VOI), Variants of Concern (VOC), and

Variants of High Consequence (VOHC).

These variant designations are based on

their ‘‘proportions at the national and

regional levels and the potential or

known impact of the constellation of

mutations on the effectiveness of

medical countermeasures, severity of

disease, and ability to spread from

person to person’’ (CDC, October 4,

2021), with VOIs considered less serious

than VOCs and VOCs considered less

serious than VOHCs. As of early October

2021, the CDC was monitoring 10

VBMs—Alpha (B.1.1.7, Q.1–Q.8), Beta

(B.1.351, B.1.351.2, B.1.351.3), Gamma

(P.1, P.1.1, P.1.2), Epsilon (B.1.427 and

B.1.429), Eta (B.1.525), Iota (B.1.526),

Kappa (B.1.617.1), B.1.617.3, Mu

(B.1.621, B.1.621.1), and Zeta (P.2)—and

one VOC—Delta (B.1.617.2 and AY.1

sublineages)—in the U.S. (CDC, October

4, 2021). CDC defines a VOC as ‘‘[a]

variant for which there is evidence of an

increase in transmissibility, more severe

disease (e.g., increased hospitalizations

or deaths), significant reduction in

neutralization by antibodies generated

during previous infection or

vaccination, reduced effectiveness of

treatments or vaccines, or diagnostic

detection failures’’ (CDC, October 4,

2021).

While the proportions of SARS–CoV–

2 variants in the United States have

shifted over time (CDC, May 24, 2021c;

CDC, October 18, 2021—Variant

Proportions, July through October 2021),

the primary variant that drove COVID–

19 transmission in the late Winter and

Spring of 2021 was the Alpha variant.

The CDC noted that Alpha is associated

with an increase in transmission, as

well as potentially increased incidences

of hospitalization and death, compared

to the predominant variants before its

emergence (CDC, October 4, 2021;

Pascall et al., August 24, 2021; Julin et

al., September 22, 2021). As Alpha

transmission subsided in the United

States during the late Spring and early

Summer of 2021, Delta emerged and

quickly became the predominant variant

in the U.S. by July 3, 2021 (CDC,

October 18, 2021—Variant Proportions,

July through October 2021). Delta now

accounts for more than 99% of

circulating virus nationwide (CDC,

October 18, 2021—Variant Proportions,

July through October 2021).

FDA authorized and approved

COVID–19 vaccines currently work well

against all of these variants; however,

there are differences in various variants’

ability to spread and the likelihood of

infection to cause severe illness. Data on

the Beta and Gamma variants do not

indicate that infections from these

variants caused more severe illness or

death than other VOCs. Data on the

Alpha variant does indicate its ability to

cause more severe illness and death in

infected individuals. And some data on

the Delta variant suggests that the Delta

variant may cause more severe illness

than previous variants, including Alpha,

in unvaccinated individuals (CDC,

October 4, 2021).

The emergence of the Delta variant,

along with other VOCs, has resulted in

a more deadly pandemic (Fisman and

Tuite, July 12, 2021). While the Delta

variant is the most transmissible SARS–

CoV–2 variant to date, the possibility

remains for the rise of future VOCs, and

even more dangerous VOHCs, as the

virus continues to spread and mutate.

Inadequate vaccination rates and the

abundance of transmission create an

environment that can foster the

development of new variants that could

61409

be similarly, or even more, disruptive

(Liu and Rocklov, August, 4, 2021). In

this context, it is critical that OSHA

address the grave danger from COVID–

19 that unvaccinated workers are

currently facing by requiring

vaccination and the other measures

included in this rule, in order to

significantly slow the transmission of

COVID–19 in workers and workplaces

and mitigate the rise of future variants.

b. Transmission

SARS–CoV–2 is a highly

transmissible virus, regardless of

variant. Since the first case was detected

in the U.S., there have been close to 45

million reported cases of COVID–19,

affecting every state and territory, with

thousands more infected each day (CDC,

October 18, 2021—Cumulative US

Cases), and some indication that these

numbers continue to underestimate the

full burden of disease (CDC, July 27,

2021). According to the CDC, the

primary way the SARS–CoV–2 virus

spreads from an infected person to

others is through the respiratory

droplets that are produced when an

infected person coughs, sneezes, sings,

talks, or breathes (CDC, May 7, 2021).

Infection could then occur when

another person breathes in the virus.

Most commonly this occurs when

people are in close contact with one

another in indoor spaces (within

approximately six feet for at least fifteen

minutes) (CDC, August 13, 2021).

Additionally, airborne transmission may

occur in indoor spaces without adequate

ventilation where small respiratory

particles are able to remain suspended

in the air and accumulate (CDC, May 7,

2021; Fennelly, July 24, 2020). While

scientists’ understanding of the Delta

variant’s virology is evolving and

remains at the frontier of science,

current data shows that the routes of

transmission remain the same for all

currently-identified SARS–CoV–2

variants. In addition, all variants can be

transmitted by people who are presymptomatic (i.e., people who are

infected but do not yet feel sick) or

asymptomatic (i.e., people who are

infected but never feel any symptoms of

COVID–19), as well as those who are

symptomatic. Pre-symptomatic and

asymptomatic transmission continue to

pose serious challenges to containing

the spread of COVID–19. For more

extensive information on transmission

routes, as well as pre-symptomatic and

asymptomatic transmission, see the

preamble to the Healthcare ETS (86 FR

App. 9

61410

Federal Register / Vol. 86, No. 212 / Friday, November 5, 2021 / Rules and Regulations

32392–32396), which is hereby

included in the record of this ETS.7

The Delta variant is transmitted from

infectious individuals via the same

routes as previous variants, but is much

more transmissible. Specifically, Delta

differs from previous dominant variants

of SARS–CoV–2 in terms of the

amplification of viral particles expelled

from infected individuals. Testing of

Delta-infected individuals indicates that

their viral loads are—on average—

approximately 1,000x greater than those

of the SARS–CoV–2 variants from the

first COVID–19 wave in early 2020. This

finding suggests much faster replication

of viral particles during early infection

with the Delta variant, resulting in

greater infectiousness (contagiousness)

when compared to earlier versions of

SARS–CoV–2 (Li et al., July 12, 2021).

The transmissibility of viruses is

measured in part by the average number

of subsequently-infected people (or

secondary cases) that are expected to

occur from each existing case (often

referred to as R0). Several comparisons

of the transmissibility of the initial

SARS–CoV–2 variants to the Delta

variant have shown that Delta is

approximately twice as transmissible

(contagious) as previous versions of

SARS–CoV–2 (CDC, August 26, 2021;

Riou and Althaus, January 30, 2020; Li

et al., July 12, 2021; Liu and Rocklov,

August, 4, 2021), likely the result of

higher initial viral loads during the presymptomatic phase (Li et al., July 12,

2021). In addition, as described further

below, data on Delta shows that both

unvaccinated and vaccinated

individuals are more likely to transmit

Delta than previous variants (Liu and

Rocklov, August, 4, 2021; Eyre et al.,

September 29, 2021), making it

especially dangerous to those who

remain unvaccinated.

c. Health Effects

COVID–19 infections can lead to

death. As reported in the Healthcare

ETS, by May 24, 2021, there had been

587,432 deaths and 32,947,548 million

infections in the U.S. alone (CDC, May

24, 2021a; CDC, May 24, 2021b). At that

point in the pandemic, 1.8 out of every

1,000 people in the U.S. had died from

COVID–19 (CDC, May 24, 2021a). Since

then, reported cases have increased to

44,857,861 and the number of deaths

has increased to 723,205 (CDC, October

18, 2021– Cumulative US Cases;

Cumulative US Deaths). By September

2021, an astounding 1 in 500 Americans

had died from COVID–19 (Keating,

7 This adoption includes the citations in the

referenced section of the Healthcare ETS, which are

also included in the docket for this ETS.

September 15, 2021). Updated mortality

data 8 currently indicate that people of

working age (18–64 years old) now have

a 1 in 202 chance of dying when they

contract the disease, with the risk much

higher (1 in 72) for those aged 50–64

(CDC, October 18, 2021—Demographic

Trends, Cases by Age Group; CDC,

October 18, 2021—Demographic Trends,

Deaths by Age Group). For a more indepth description of the health effects

resulting from SARS–CoV–2 infection,

see the preamble to the Healthcare ETS

(86 FR 32383–32392), which is hereby

included in the record of this ETS.9

Apart from fatal cases, COVID–19 can

cause serious illness, including longlasting effects on health. Many patients

who become ill with COVID–19 require

hospitalization. Indeed, updated CDC

hospitalization and mortality data

indicate that working age Americans

(18–64 years old) now have a 1 in 14

chance of hospitalization when infected

with COVID–19 (CDC, October 18,

2021—Demographic Trends, Cases by

Age; Total Hospitalizations, by Age).

Those who are hospitalized frequently

need supplemental oxygen and

treatment for the disease’s most

common complications, which include

pneumonia, respiratory failure, acute

respiratory distress syndrome (ARDS),

acute kidney injury, sepsis, myocardial

injury, arrhythmias, and blood clots.

One study, which included 35,502

inpatients nationwide, determined that

the median length of hospital stay was

6 days, unless the cases required ICU

treatment. For those cases, ICU stays

were on median 5 days in addition to

the time spent hospitalized outside of

the ICU (Rosenthal et al., December 10,

2020). Another study that assessed

hospital length of stay for COVID–19

patients in England estimated that a

non-ICU hospital stay averaged between

8 and 9 days, but those estimates ranged

from approximately 12 to 18 days when

patients were admitted to the ICU

(Vekaria et al., July 22, 2021). Moreover,

given that SARS–CoV–2 is still a novel

virus, the severity of long-term health

effects—such as ‘‘post-COVID

conditions’’—are not yet fully

understood.

Many members of the workforce are at

increased risk of death and severe

disease from COVID–19 because of their

age or pre-existing health conditions.

8 Risk of death is based on averages from reported

CDC data. Risks of hospitalization and death are

much higher in unvaccinated individuals, as

discussed further in Grave Danger, Section III.A.IV.

Vaccines Effectively Reduce Severe Health

Outcomes from and Transmission of SARS–CoV–2.

9 This adoption includes the citations in the

referenced section of the Healthcare ETS, which are

also included in the docket for this ETS.

The comorbidities that further

exacerbate COVID–19 infections are

common among adults of working age in

the U.S. For instance, 46.1% of

individuals with cancer are in the 20–

64 year old age range (NCI, April 29,

2015), and over 40% of working age

adults are obese (Hales et al., February

2020). Disease severity is also likely

exacerbated by long-standing healthcare

inequities experienced by members of

many racial and economic

demographics (CDC, April 19, 2021).

Recent data suggests that Delta variant

infections may result in even more

severe illness and a higher frequency of

death than previous COVID–19 variants

due to Delta’s increased transmissibility,

virulence, and immune escape (Fisman

and Tuite, July 12, 2021). Symptomatic

Delta variant infections do occur in fully

vaccinated people (Mlcochova et al.,

June 22, 2021; Musser et al., July 22,

2021); however, as reported by the CDC

(CDC, August 26, 2021), the vast

majority of the continuing instances of

severe and fatal COVID–19 infections

are occurring in unvaccinated persons

(discussed further in Grave Danger,

Section III.A.IV. Vaccines Effectively

Reduce Severe Outcomes from and

Transmission of SARS–CoV–2). An

assessment of Delta-related hospital

admissions in Scotland found that

hospitalizations were approximately

doubled in patients with the Delta

variant when compared to the Alpha

variant (Sheikh et al., June 4, 2021). A

similar study conducted using a

retrospective cohort in Ontario, Canada

compared the virulence of novel SARS–

CoV–2 variants and found that the

incidences of hospitalization, ICU

admission, and death were more

pronounced with the Delta variant than

any other SARS–CoV–2 variant (Fisman

and Tuite, July 12, 2021). A large

national cohort study that included all

Alpha and Delta SARS–CoV–2 patients

in England between March 29 and May

23, 2021 found a ‘‘higher hospital

admission or emergency care attendance

risk for patients with COVID–19

infected with the Delta variant

compared with the Alpha variant,’’

suggesting that Delta outbreaks—

especially amongst unvaccinated

populations—may lead to more severe

health consequences and an equivalent

or greater burden on healthcare services

than the Alpha variant (Twohig et al.,

August 27, 2021). However, one more

recent study examining data from

several U.S. states demonstrated a

significant increase in hospitalization

from the pre-Delta to the Delta period,

which may be related to increased

transmissibility of Delta rather than

App. 10

Federal Register / Vol. 86, No. 212 / Friday, November 5, 2021 / Rules and Regulations

more severe health outcomes (Taylor et

al., October 22, 2021).

III. Impact on the Workplace

SARS–CoV–2 is readily transmissible

in workplaces because they are areas

where multiple people come into

contact with one another, often for

extended periods of time. When

employees report to their workplace,

they may regularly come into contact

with co-workers, the public, delivery

people, patients, and any other people

who enter the workplace. Workplace

factors that exacerbate the risk of

transmission of SARS–CoV–2 include

working in indoor settings, working in

poorly-ventilated areas, and spending

hours in close proximity with others.

Full-time employees typically spend 8

hours or more at work each shift, more

time than they spend anywhere else but

where they live. Employees work in

proximity to others in workplaces that

were not originally designed to keep

people six feet away from other people

and that may make it difficult for

employees to perform work tasks while

maintaining a six-foot distance from

others. Even in the cases where workers

can do most of their work from, for

example, a private office within a

workplace, they share common areas

like hallways, restrooms, lunch rooms

and meeting rooms. Furthermore, many

work areas are poorly ventilated (Allen

and Ibrahim, May 25, 2021; Lewis,

March 30, 2021). An additional factor

that exacerbates the risk of transmission

of SARS–CoV–2 is interacting with or

caring for people with suspected or

confirmed COVID–19; this was a

primary driver of OSHA’s determination

of grave danger for healthcare workers

in the Healthcare ETS (see 86 FR 32381–

32383). In recent weeks, the majority of

states in the U.S. have experienced what

CDC defines as ‘‘high or substantial

community transmission,’’ indicating

that there is a clear risk of the virus

being introduced into and circulating in

workplaces (CDC, October 18, 2021—

Community Transmission Rates).

Although COVID–19 is not

exclusively an occupational disease, it

is evident from research accrued since

the beginning of the pandemic that

SARS–CoV–2 transmission can and

does occur in workplaces, affecting

employees and their lives, health, and

livelihoods. This continues to be true

for the Delta variant, with its increased

transmissibility and potentially more

severe health effects. This section

describes some of the clusters,

outbreaks, and other occurrences of

workplace COVID–19 cases that

government agencies, researchers, and

journalists have described, and the

widespread effects of SARS–CoV–2 in

industry sectors across the national

economy. While the focus is on more

recent data reflecting the impact of the

Delta variant, evidence of workplace

transmission that occurred prior to the

emergence of the Delta variant is also

presented.

The workplace-based clusters

described below provide evidence that

workplaces in a wide range of industries

have been affected by COVID–19, that

many employees face exposure to

infected people in their workspaces, and

that SARS–CoV–2 transmission is

occurring in the workplace, including

during the recent period where the Delta

variant has predominated. Although the

presence of a cluster on its own does not

necessarily establish that the cluster is

work-related (i.e., a result of

transmission at the worksite), many

state investigation reports and

published studies provide evidence that

transmission is work related by

documenting that infections at a

workplace occurred within 14-days (the

incubation period for the virus) of each

other and ruling out the possibility that

transmission occurred outside the

workplace. In addition, the information

below demonstrates that exposures to

SARS–CoV–2 happen regularly in a

wide variety of different types of

workplaces.

The basis for OSHA’s grave danger

finding is that employees can be

exposed to the virus in almost any work

setting; that exposure to SARS–CoV–2

can lead to infection (CDC, September

21, 2021); and that infection in turn can

cause death or serious impairment of

health, especially in those who are

unvaccinated (see Section III.A.IV.

Vaccines Effectively Reduce Severe

Health Outcomes from and

Transmission of SARS–CoV–2). The

information described in this section

supports OSHA’s finding that

employees who work in spaces shared

by others are at risk of exposure to

SARS–CoV–2. The degree of risk from

droplet-based transmission may vary

based on the duration of close proximity

to a person infected with SARS–CoV–2,

including the Delta variant, but the

simple and brief act of sneezing,

coughing, talking, or even breathing can

significantly increase the risk of

transmission if controls are not in place.

SARS–CoV–2, including the Delta

variant, might also be spread through

airborne particles under certain

conditions, particularly in enclosed

settings with inadequate ventilation,

which are common characteristics of

some workplaces.

The peer-reviewed scientific journal

articles, government reports, and news

61411

articles described below establish the

widespread prevalence of COVID–19

among employees, beginning with a

description of the recent impact from

the Delta variant. OSHA’s findings are

based primarily on the evidence from

peer-reviewed scientific journal articles

and government reports. However, peer

review for scientific journal articles and

the assembly of information for

government reports and other official

sources of information take time, and

therefore those sources do not always

reflect the most up-to-date information

(Chan et al., December 14, 2010). In

addition, while state and local health

departments can report workplace

outbreaks to CDC, the agency does not

provide summary statistics by

workplace so that those outbreaks can

be tracked on a national level. In the

context of the COVID–19 pandemic,

given the recent impacts due to the

Delta variant and the emergence of new

information on a daily basis, it is critical

for OSHA to rely on the most up-to-date

information available. Therefore, OSHA

has occasionally supplemented peerreviewed data and government reports

with additional information on

occupational outbreaks contained in

other sources of media (e.g.,

newspapers, digital media, and

information submitted to or obtained by

private organizations).10 The reported

information from other sources can

provide further evidence of the impact

of an emerging and changing disease,

especially for industries that are not

well represented in the peer-reviewed

scientific literature. Together, these

sources of information represent the

best available evidence of the impact on

employees of the pandemic thus far.

The information described herein

illustrates a significant number of

infections among employees in a variety

of industries, with virtually every state

continuing to experience what CDC

defines as high or substantial

community transmission related to the

recent surge of the Delta variant. The

industries and types of workplaces

described are not the only ones in

which a grave danger exists. The science

of transmission does not vary by

industry or by type of workplace. OSHA

therefore expects transmission to occur

in diverse workplaces all across the

country (see Dry Color Mfrs. Ass’n, Inc.

v. Dep’t of Labor, 486 F.2d 98, 102 n.3

(3d Cir. 1973) (holding that when OSHA

determines a substance poses a grave

10 OSHA did not make findings based solely on

non-peer-reviewed sources such as news articles,

but the agency found that those sources can

sometimes provide useful information when

considered with more robust sources.

App. 11

61412

Federal Register / Vol. 86, No. 212 / Friday, November 5, 2021 / Rules and Regulations

danger to workers, OSHA can assume an

exposure to a grave danger exists

wherever that substance is present in a

workplace)). In addition, the severity of

COVID–19 does not depend on where

an employee is infected; an employee

exposed to SARS–CoV–2 might die

whether exposed while working at a

meat packing facility, a retail

establishment, or an office (see Grave

Danger, Section III.A.V.b. Employees

Who Work Exclusively Outside, below,

for a discussion of the risk of exposure

in outdoor workplaces).

a. General Impact on Workers

Data on SARS–CoV–2 infections,

illnesses, and deaths among employees

in general industry, agriculture,

construction, and maritime support

OSHA’s finding that COVID–19 poses a

grave danger to employees in these

sectors across the U.S. economy. This

section summarizes studies and reports

of COVID–19 illness and fatalities in a

wide range of workplaces across those

industry sectors. Not all workplace

settings are discussed; nor is the data

available to do so. However, the

characteristics of the various affected

workplaces—such as indoor work

settings; contact with coworkers, clients,

or members of the public; and sharing

space with others for prolonged periods

of time—indicate that exposures to

SARS–CoV–2 are occurring in a wide

variety of work settings across all

industries. Therefore, most employees

who work in the presence of other

people (e.g., co-workers, customers,

visitors) need to be protected.

While there is no comprehensive

source of nationwide workplace

infection data, reports from states and

communities on outbreaks related to

workplaces provide key, up-to-date data

that illustrate the likelihood of

employee exposure to SARS–CoV–2 at

workplaces throughout the U.S. OSHA

identified a number of recent reports

from various regions of the country that

together demonstrate the impact that

SARS–CoV–2 can have on a variety of

workplaces, including in service

industries (e.g., restaurants, grocery and

other retail stores, fitness centers,

hospitality, casinos, salons), corrections,

warehousing, childcare, schools, offices,

homeless shelters, transportation, mail/

shipping/delivery services, cleaning

services, emergency services/response,

waste management, construction,

agriculture, food packaging/processing,

and healthcare. Deaths are reported in

many studies performed prior to the

emergence of the Delta variant but,

because the Delta outbreak is so recent

and deaths can occur weeks after

infection, the number of deaths from

recent infections might be

underestimated. Some of the reports

include cumulative data representing

various phases of the pandemic,

beginning prior to the availability of

vaccines and continuing through the

recent surge of the Delta variant. In

addition, some studies report

investigations of recent outbreaks,

which provide insight on the impact of

the Delta variant as well as impacts

associated with the current vaccination

status of workers.

The Washington State Department of

Health (WSDH) reports outbreaks

occurring in non-healthcare workplaces

(WSDH, September 8, 2021). In nonhealthcare workplaces, outbreaks are

defined as two or more laboratory

confirmed cases of COVID–19, with at

least two cases reporting symptom onset

within 14 days of each other, and

plausible epidemiological evidence of

transmission in a shared location other

than a household. As of September 4,

2021, WSDH reported 5,247 outbreaks

in approximately 40 different types of

non-healthcare work settings. During

the week of August 29 through

September 4, 2021, WSDH identified

137 separate workplace outbreaks. The

types of non-medical workplace settings

that represented more than 5% of the

total outbreaks during that week

included food service/restaurants,

childcare, schools, retail, grocery, and

shelter/homeless services. Other types

of non-healthcare settings where

outbreaks occurred recently included

non-food and food manufacturing,

construction, professional services/

office based, agriculture/produce

packing, transportation/shipping

delivery, government agencies/facilities,

leisure hospitality/recreation,

corrections, utilities, warehousing,

facility/domestic cleaning services,

youth sports/activities, camps, and

public safety. Over the course of the

pandemic, outbreaks have also been

observed at bars/nightclubs, hotels, and

fishing/commercial seafood vessels.

The Oregon Health Authority (OHA)

publishes a weekly report detailing

outbreaks directly related to work

settings. OHA epidemiologists consider

cases to be part of a workplace outbreak

when clusters form with respect to

space and time, within a plausible

incubation period for the virus, and

their investigation does not uncover an

alternative source for the outbreak. For

privacy reasons, OHA only reports

outbreaks with 5 or more cases in

workplaces with 30 or more people.

OHA reported a total of 26,013 cases

and 135 deaths related to workplace

outbreaks as of September 1, 2021. As

of September 1, 2021, OHA was

investigating more than 124 active

workplace outbreaks (OHA, September

1, 2021). Those outbreaks occurred in a

wide variety of industries including

correctional facilities, emergency

services, waste management, schools

and child care, retail and grocery stores,

restaurants, warehousing, agriculture,

food processing/packaging,

construction, healthcare, mail and

delivery services, office locations,

utilities, transportation, and others.

Tennessee Department of Health was

investigating 557 active COVID–19

clusters as of September 8, 2021 (TDH,

September 8, 2021). Clusters are defined

as two or more laboratory confirmed

COVID–19 cases linked to the same

location or event that is not a household

exposure. The clusters occurred in 13

types of settings, 10 of which were

workplace settings. Outbreaks at

workplaces represented more than half

of the total active outbreaks in the state

at that time. Settings comprising more

than 5% of total clusters included

assisted care living facilities, nursing

homes, and correctional facilities. Other

types of workplaces where outbreaks

occurred included bars, construction,

farms, homeless shelters, and industrial

settings.

The North Carolina Department of

Health and Human Services reports

cumulative numbers of clusters, cases,

and deaths for workers in poultry

processing facilities (beginning in April

of 2020) and other types of workplaces

(beginning in May of 2020) (NCDHHS,

August 30, 2021). Clusters are defined

as a minimum of 5 cases with illness

onset or initial positive results within a

14-day period and plausible

epidemiological linkage between the

cases. Plausible epidemiological linkage

means that multiple cases were in the

same general setting during the same

time period (e.g., same shift, same

physical area) and that a more likely

source of exposure is not identified (e.g.,

household contact or close contact to a

confirmed case in another setting).

During that time period of April/May

2020 through August 30, 2021,

workplaces 11 were associated with

nearly 80% of the 1,969 clusters and

27,097 cases observed and nearly 40%

of the 167 deaths related to the clusters.

Cumulative numbers of clusterassociated deaths were highest in meat

and poultry processing (25 of 5,351

cases), followed by healthcare (10 of

1,036 cases), government services and

manufacturing (5 of 1,048 cases and 5 of

11 NCDHHS identifies a ‘‘workplace’’ category in

their report (e.g., agriculture, construction), but

OSHA includes other settings where employees

would be present (e.g., retail, restaurants, childcare,

healthcare).

App. 12

Federal Register / Vol. 86, No. 212 / Friday, November 5, 2021 / Rules and Regulations

1,856 cases, respectively), and

restaurants and childcare (3 of 421 cases

and 3 of 1,943 cases, respectively).

Recently, in July of 2021, the number of

cases associated with workplace clusters

began increasing in several different

types of work settings, including meat

processing, manufacturing, retail,

restaurants, childcare, schools, and

higher education.

Colorado Department of Public Health

& Environment/Colorado State

Emergency Operations Center (CDPHE/

CSEOC, September 8, 2021) reported

5,584 resolved workplace-related

outbreaks involving 40,156 employee

cases and 79 employee deaths since

May of 2020. The agency’s current

investigations, as of September 8, 2021

included 291 active outbreaks (not

defined), with 2,865 staff cases

(assumed to be cases in employees). The

majority of active outbreaks were

reported in childcare, schools,

healthcare, and corrections. Active

outbreaks were also reported in

construction, retail, homeless shelters,

casinos, restaurants, hotels, offices, law

enforcement, manufacturing, delivery

services, and warehouses. Other types of

work settings that were affected in

resolved outbreaks included

warehouses, bars, government locations,

waste management, utilities, salons,

emergency services, meat processing/

packaging, and postal services. From

June 21, 2021 (the date the healthcare

ETS was published) through September

8, 2021, 1,469 staff cases associated with

outbreaks were reported, for an average

of approximately 19 cases per day.

Similar reporting is available from

Louisiana’s Department of Health (LDH,

August 24, 2021), with 1,347 outbreaks

and 9,130 cases reported as of August

24, 2021. LDH defines an outbreak as 2

or more cases among unrelated

individuals who visited a site within a

14-day period. More than three quarters

of outbreaks through that date were

associated with workplaces. Workplace

settings in Louisiana that experienced

more than 5% of outbreaks included

day care facilities, bars, restaurants,

retail settings, industrial settings, and

office spaces. Other types of workplace

settings or industries where outbreaks

occurred included casinos, gyms/fitness

centers, banks, automotive services,

construction, and ships/boats.

In addition to the state data above,

some published studies and government

reports provide information on recent

workplaces outbreaks. For example, 47

people, including 3 of 11 staff members,

23 gymnasts, and 21 household

contacts, contracted COVID–19 from an

outbreak linked to an Oklahoma

gymnastics facility during April 15

through May 3, 2021 (Dougherty et al.,

July 16, 2021). All 21 of the virus

samples sequenced were determined to

be the Delta variant. The majority of the

infected individuals (85%) were

unvaccinated. Infections were reported

in 16 adults aged 20 years or older; two

adults were hospitalized and one

required intensive care.

The state of Hawaii defines clusters as

three or more confirmed or probable

cases linked to a site or event within 14

days, with no outside exposure of cases

to each other (Hawaii State, August 19,

2021). The state reported a COVID–19

cluster in July associated with a concert

at a bar that affected 16 people,

including employees, band members,

and concert attendees; infections also

spread to 7 household members. Band

members had performed while sick.

Four of the initial 16 people and none

of the household members who tested

positive for COVID–19 were fully

vaccinated. The concert cluster was

linked to clusters at another workplace

and another concert. The report lists

additional clusters investigated in the

two weeks prior to the report; those

clusters were observed in workplace

locations such as correctional facilities,

bars and nightclubs, restaurants,

construction/industrial sites, travel/

lodging/tourism, schools, food

suppliers, and gyms.

Additional evidence that employees

are at risk of exposure to SARS–CoV–2

in the workplace is available from

published, peer-reviewed studies that

were conducted before the Delta variant

emerged. Those studies demonstrate

that employees have been at risk of

infection, illness, and death throughout

the COVID–19 pandemic. Because the

Delta variant is more transmissible and

likely causes more severe disease than

previous variants, there is even greater

potential for unvaccinated employees to

become seriously ill or die as a result of

exposure to the Delta variant.

Contreras et al. (July, 2021) examined

workplace outbreaks (excluding

healthcare settings, homelessness

services, and emergency medical

services) in Los Angeles county from

March 19 through September 30, 2020.

Workplace outbreaks were defined as 5

or more suspected or laboratory

confirmed COVID–19 cases (prior to

May 29) or 3 or more laboratory

confirmed cases (after May 29)

occurring within 14 days. Nearly 60% of

the 698 identified outbreaks occurred in

three sectors—manufacturing (184,

26.4%), retail trade (137, 19.6%), and

transportation and warehousing (73,

10.5%). Also notable were the 71

outbreaks in the accommodation and

food services industry, which

61413

represented 10.2% of the outbreaks. The

study authors concluded that outbreaks

were larger and lasted longer at facilities

with more onsite staff.

Outbreaks in Wisconsin from March 4

through November 16, 2020 were also

examined (Pray et al., January 29, 2021).

Non-household outbreaks were defined

as two or more confirmed COVID–19

cases that occurred within 14 days in

persons who attended the same facility

or event and did not share a household.

During the period from March 4 through

November 16, 2020, the largest

percentages of cases were associated

with outbreaks in long-term care

facilities (26.8% of cases), correctional

facilities (14.9% of cases), and colleges

or universities (15% of cases). Also

notable were the substantial number of

cases associated with outbreaks in food

production or manufacturing facilities

(including meat processing and

warehousing; 14.5% of cases) and

schools and childcare facilities (10.6%

of cases).

Bui et al. (August 17, 2020) analyzed

data from the Utah Department of

Health’s COVID–19 case surveillance

system, which included data on

workplace outbreaks. Outbreaks were

defined as two or more laboratory

confirmed cases occurring within a 14

day period among coworkers in a

common workplace (e.g., same facility).

During the time period between March

6 and June 5, 2020, 277 COVID–19

outbreaks were reported, of which 210

(76%) occurred in workplaces. The 210

workplace outbreaks occurred in 15 of

20 industry sectors, and the industry

sectors of manufacturing (43 outbreaks,

20%), construction (32 outbreaks, 15%),

and wholesale trade (29 outbreaks, 14%)

together represented nearly half of

workplace outbreaks. Other sectors that

represented more than 10% of total

outbreaks were retail trade (28

outbreaks, 13%) and accommodation

and food services (25 outbreaks, 12%).

Incidence rates of COVID–19 over the

period of March 6 through June 5, 2020

were 339/100,000 workers in

manufacturing, 122/100,000 workers in

construction, 377/100,000 workers in

wholesale trade, 68/100,000 workers for

retail trade, and 78/100,000 workers for

accommodation and food services. For

COVID–19 cases associated with

workplace outbreaks in which

hospitalization and severity status were

known (1,382 and 1,155, respectively),

the number in all sectors who were

admitted to the hospital was 85 (6%)

and the number with severe outcomes

(intensive care unit admission,

mechanical ventilation, or death) was 40

(3%).

App. 13

61414

Federal Register / Vol. 86, No. 212 / Friday, November 5, 2021 / Rules and Regulations

The impact of SARS–CoV–2

exposures on employee infection,

illness, and death has also been

demonstrated in studies focusing on

specific types of industries, such as

those where employees have frequent

contact with each other and the public

(e.g., grocery stores, bars, fitness

facilities, schools, and law enforcement/

corrections). For example, a study by

Lan et al. (September 26, 2020)

demonstrates the risk of infection in

service industries. The cross-sectional

study examined the risks of SARS–CoV–

2 exposure and infection for employees

in a Boston, Massachusetts-area retail

grocery store market. The study tested

104 grocery store employees, of whom

20% (21 employees) were positive for

COVID–19; 76% of confirmed cases did

not have symptoms. After adjusting for

gender, smoking, age, and the

prevalence of COVID–19 in the

employees’ residential communities,

employees who had direct customer

exposure (e.g., cashiers, sales associates,

cart attendants) were 5.1 times more

likely to have a positive test for COVID–

19 than employees without direct faceto-face customer exposure (e.g.,

stockers, backroom, receiving and

maintenance). The infection rate of 20%

among all employees was significantly

higher than the rate in the surrounding

community.

In February of 2021, an event at an

Illinois bar that accommodates

approximately 100 people resulted in a

COVID–19 outbreak that affected 46

people, including 3 (10%) staff

members, 26 (90%) patrons, and 17

secondary cases (Sami et al., April 9,

2021). People at the event included an

asymptomatic person diagnosed with

COVID–19 on the previous day and 4

symptomatic people who were later

diagnosed with COVID–19. The

outbreak resulted in a school closure

and the hospitalization of a resident at

a long-term care facility.

In Minnesota, 47 COVID–19 outbreaks

were detected at fitness facilities from

August through November of 2020

(Suhs et al., July 23, 2021). One

outbreak at a fitness facility during

October through November of 2020

resulted in 23 COVID–19 cases

including 5 (22%) employees and 18

(78%) members. A genetic analysis of

specimens from 3 employees and 10

members identified 2 distinct genetic

subclusters, indicating two distinct

chains of transmission among members

and employees.

School-related outbreaks were

examined from December 1, 2020

through January 22, 2021 in eight public

elementary schools of a Georgia school

district (Gold et al., February 26, 2021).

A COVID–19 case was determined to be

school-related if (1) symptom onset or a

positive test was consistent with the

incubation period of the virus following

contact with an index case or a schoolassociated case, (2) close contact

occurred with the index case or schoolassociated case while that person was

infected, and (3) no known contact

occurred with an infected community or

household contact in the two weeks

prior to a positive test for COVID–19.

The investigators identified nine

clusters of three or more

epidemiologically linked COVID–19

cases that involved 13 educators and 32

students in six of the eight elementary

schools. Approximately half of the

school-associated cases involved two

clusters that began with probable

transmission between educators,

followed by educator to student

transmission. Eighteen of 69 household

members tested received positive

results.

A number of studies demonstrate the

impact of COVID–19 in law enforcement

and related fields such as corrections.

For example, a study examining

COVID–19 antibodies in employees

from public service agencies in the New

York City area from May through July of

2020, found that 22.5% of participants

had COVID–19 antibodies (Sami et al.,

March, 2021). The percentage of

correctional officers found to have

COVID–19 antibodies (39.2%) was the

highest observed among all the

occupations. The percentages of police

dispatchers, traffic officers, security

guards, and dispatchers found to have

COVID–19 antibodies (29.8 to 37.3%)

were among the highest levels observed

in all the occupations. The study

authors noted that those jobs involve

frequent or close contact with the public

or are done in places where employees

work in close proximity to their

coworkers.

Wallace et al. (May 15, 2020)

evaluated data on COVID–19 cases and

deaths among correctional facility

employees and inmates from January 21

to April 21, 2020. Data were reported to

CDC by 37 (69%) of 54 state and

territorial health department

jurisdictions. Of these 37 jurisdictions,

32 (86%) reported at least one COVID–

19 case from a correctional facility. Of

the 420 facilities with a case, 221 (53%)

reported cases only among staff

members. In total, 4,893 COVID–19

cases among incarcerated or detained

persons and 2,778 cases among staff

members were reported (total tested not

provided). Among staff member cases,

79 hospitalizations (3%) and 15 deaths

(1%) were reported. The study authors

noted that ‘‘correctional and detention

facilities face challenges in controlling

the spread of infectious diseases

because of crowded, shared

environments and potential

introductions by staff members and new

intakes.’’

Ward et al. (June 2021) analyzed

COVID–19 prevalence among prisoners

and staff in 45 states from March 31,

2020 through November 4, 2020. During

that time period, COVID–19 cases in

staff were 3 to 5 times higher compared

to the U.S. population. Average daily

increases in cases were 42 per 100,000

prison employees, 61 per 100,000

prisoners, and 13 per 100,000 U.S.

residents. On November 4, 2020,

COVID–19 prevalence for prison staff

was 9,316 cases per 100,000 employees,

which was 3.2 times greater than

prevalence in the U.S. population (2,900

cases per 100,000).

Kirbiyik et al. (November 6, 2020)

analyzed movement through a networkinformed approach to identify likely

high points of transmission within the

Cook County Jail in Chicago, IL. At that

facility, over 900 COVID–19 cases were

reported across 10 housing divisions in

13 buildings from March 1–April 30,

2020. Staff members were required to

report symptoms of COVID–19

(probable cases) or receipt of a positive

test result (confirmed cases). A total of

2,041 staff members (77% of staff) were

included in the network analysis

because information was available about

their shift and division assignments,

and 198 (9.7%) of those staff members

had COVID–19 during the two-month

study period. Connections between staff

members who had COVID–19 were

higher than expected, suggesting likely

transmission among staff members.

Fewer connections than expected were

observed among detained persons with

SARS–CoV–2 infections, suggesting the

effectiveness of medical isolation at

reducing transmission.

The Officer Down Memorial Page,

which tracks police officer fatalities

determined to be occupationally related,

reported that the majority of officer

deaths for 2021 (157 of 269) were

related to COVID–19 (ODMP, September

14, 2021). For the 269 officers who died,

causes of death were not reported for

each month, but the highest numbers of

monthly deaths, 52 in January and 65 in

August (compared to 16 to 34 deaths on

other reported months), were consistent

with the winter surge of COVID–19 and,

more recently, the surge caused by the

Delta variant.

The risk of COVID–19 has also been

examined in industries where

employees have little contact with the

public, such as construction, and food

processing, and where most exposure to

App. 14

Federal Register / Vol. 86, No. 212 / Friday, November 5, 2021 / Rules and Regulations

SARS–CoV–2 likely comes from other

workers. Pasco et al. (October 29, 2020)

examined the association between

construction work during the COVID–19

pandemic and community transmission

and construction worker hospitalization

rates in Austin, Texas from March 13 to

August 20, 2020. A ‘‘Stay Home-Work

Safe’’ order enacted on March 24, 2020,

limited construction to only critical

infrastructure and excluded commercial

and residential work. One week later,

the Texas governor lifted the restriction

for essential workers and allowed all

types of construction work to resume,

while keeping the order in place for

other workers. The authors found that

resuming construction during the

shelter-in-place order led to an increase

in community transmission, an increase

in hospitalizations among community

members, and an increase in

hospitalizations of construction

workers. By mid-July, Austin Public

Health identified at least 42 clusters (not

defined) of COVID–19 cases in the

construction industry; 515 individuals

were hospitalized for COVID–19

illnesses acquired as part of these

clusters, and 77 of those reported

working in construction. The study

found that construction workers had a

nearly 5-fold increased risk of

hospitalization in central Texas

compared with workers in other

occupations. The authors’ model

predicted that allowing unrestricted

construction work would be associated

with an increase in COVID–19

hospitalization rates from 0.38 per 1,000

residents to 1.5 per 1,000 residents

overall, and from 0.22 per 1,000

construction workers to 9.3 per 1,000

construction workers for the

construction industry specifically. The

authors concluded that stringent

workplace safety measures could

significantly mitigate risks related to

COVID–19 in the industry.

The meat packing and processing

industries and related agricultural and

food processing sectors have also been

impacted by COVID–19. Waltenburg et

al. (January, 2021) reported COVID–19

cases in employees from meat and

poultry processing facilities in 31 states

from March 1 through May 31, 2020. As

reported in Table 2 of that report, 28,364

employees in those facilities were

confirmed to have COVID–19 by

laboratory testing and 132 died. Among

the 20 states that reported total numbers

of employees, 11.4% of the workers

were diagnosed with COVID–19 (with a

range of 3.1 to 27.7% of workers in

individual states). For states that

reported at least one COVID–19-related

death, the percentages of employees

who died in each state ranged from 0.1

to 2.4% of those with COVID–19. The

authors found a high burden of disease

in persons employed at these facilities

who were racial or ethnic minorities.

Higher incidence in these populations

might be due to the likelihood of these

employees working in areas in the plant

where transmission risk is higher.

Steinberg et al. (August 7, 2020)

reported that attack rates (i.e., the

number of individuals who are infected

in comparison to the total number at

risk) among production employees in

the Cut (30.2%), Conversion (30.1%),

and Harvest (29.4%) departments of a

meat processing plant (where spacing

between employees is less than 6 feet)

were double that of salaried employees

(14.8%) whose workstations had been

modified to increase physical distancing

from others.

Waltenburg et al. (January, 2021) also

evaluated COVID–19 incidence in food

manufacturing and agricultural settings

(e.g., manufacturing or farming

involving fruits, vegetables, dairy, baked

goods, eggs, prepared foods), as reported

in 30 states from March through May

2020. In food manufacturing and

farming of fruits, vegetables, dairy, and

other items, 742 workplaces were

affected, including 8,978 infections and

55 fatalities. For states that reported

total numbers of employees, the

proportion of employees who developed

COVID–19 in each state ranged from 2.0

to 43.5%. For states that reported at

least one death, the percentages of

deaths among cases ranged from 0.1 to

3.8%.

Porter et al. (April 30, 2021) reported

that 13 COVID–19 outbreaks occurred at

Alaska seafood processing facilities and

vessels (both of which were described as

high density workplaces) during the

Summer and early Fall of 2020. The 13

outbreaks involved 539 COVID–19

cases, with 2–168 cases per outbreak.

Attack rates in facilities and offshore

vessels ranged from less than 5% to

75%. Outbreaks were also reported in

entry quarantine groups. Because of

these outbreaks, it was determined that

vaccination of these essential workers is

important and requirements for COVID–

19 prevention were updated to include

smaller quarantine groups, serial testing,

and testing before transfers from one

facility or vessel to another.

Finally, two published studies

analyzed death records to determine

how mortality rates among individuals

in various types of workplaces had

changed during the pandemic. Chen et

al. (June 4, 2021) analyzed records of

deaths occurring on or after January 1,

2016 in California and found that

mortality rates in working aged adults

61415

(18–65 years) increased 22% during the

COVID–19 pandemic period of March

through November 2020 compared to

pre-pandemic periods. Relative to prepandemic periods, the groups of

employees experiencing the highest,

statistically significant increases in

relative excess mortality were those in

food/agriculture (39% increase),

transportation/logistics (31% increase),

facilities (23% increase), and

manufacturing (24% increase). Other

groups that also experienced excess,

statistically significant mortality

compared to pre-pandemic periods were

health or emergency workers (17%

increase), retail workers (21% increase),

and government and community

workers (17% increase). The study

authors concluded that certain

occupational sectors were impacted

disproportionally by mortality during

the pandemic and that essential work

conducted in-person is a likely avenue

of infection transmission.

Hawkins et al. (January 10, 2021)

examined death certificates of

individuals who died in Massachusetts

between March 1 and July 31, 2020. An

age-adjusted mortality rate of 16.4 per

100,000 employees was determined

from 555 death certificates that had

useable occupation information.

Employees in 11 occupational groups

had particularly high mortality rates:

healthcare support; transportation and

material moving; food preparation and

serving; building and grounds cleaning

and maintenance; production,

construction and extraction;

installation/maintenance/repair;

protective services; personal care

services; arts/design/entertainment;

sports/media; and community and

social services. The study authors noted

that occupational groups expected to

have frequent contact with sick people,

close contact with the public, and jobs

that are not practical to do from home

had particularly elevated mortality

rates.

b. Healthcare Workers

As explained in the Healthcare ETS,

COVID–19 presents a grave danger to

workers in all U.S. healthcare settings

where people with COVID–19 are

reasonably expected to be present (86

FR 32381). Healthcare settings covered

by the Healthcare ETS primarily include

settings where people with suspected or

confirmed COVID–19 are treated,

exacerbating the risk present in most

workplaces. To control the higher level

of risk in those settings, OSHA

determined that a suite of workplace

controls was necessary to protect all

employees, whether they are vaccinated

or unvaccinated. As explained further

App. 15

61416

Federal Register / Vol. 86, No. 212 / Friday, November 5, 2021 / Rules and Regulations

below, OSHA now finds that

unvaccinated healthcare workers in

healthcare settings not covered by the

Healthcare ETS are also at grave danger

from exposure to SARS–CoV–2, just like

unvaccinated workers in other

industries. Data continue to be collected

and reported for healthcare workers,

and a small number of peer-reviewed

studies demonstrate the potential

impact of the Delta variant on

healthcare workers.

CDC continues to provide updates for

COVID–19 cases and deaths among

healthcare personnel. However,

information on healthcare personnel

status continues to be reported for only

a fraction (18.91%) of total reported

cases, and death status was reported for

only 82.16% of healthcare personnel

cases as of October 18, 2021 (CDC,

October 18, 2021—Healthcare

Personnel). Given incomplete reporting,

the data from this source represent only

a fraction of actual healthcare cases and

deaths. Nevertheless, CDC reported

666,707 healthcare personnel cases

among the 6,754,306 reported cases that

included information on healthcare

personnel status (9.9%) and 2,229

fatalities among the 547,769 cases that

included death status (0.4%) for

healthcare employees as of October 18,

2021. This is a 26% increase in the

number of cases and a 27% increase in

the number of deaths since the May 24,

2021 data reported in the ETS (CDC,

October 18, 2021—Healthcare

Personnel). The Delta variant is likely

responsible for the majority of those

deaths. No healthcare worker deaths

were reported by CDC during the weeks

of May 30 through June 13, 2021;

however, as the Delta variant’s

prevalence rose after June 20, healthcare

worker deaths began increasing; they

peaked during the period of August 15

through September 12, 2021, when 34 to

36 healthcare worker deaths were

reported per week (CDC October 18,

2021—Healthcare Personnel, Deaths by

Week). Independent reporting by Kaiser

Health News and The Guardian reported

more than 3,600 fatalities in health care

workers as of April 2021 (Spencer and

Jewett, April 8, 2021). That number is

expected to be higher at this time since

the earlier figure did not include the

most recent 5 months of the pandemic,

which includes the period of Delta

variant predominance.

Published studies also demonstrate

that healthcare workers, especially those

who are unvaccinated, remain at risk of

being infected with SARS–CoV–2 (see

Section III.A.IV. Vaccines Effectively

Reduce Severe Health Outcomes from

and Transmission of SARS–CoV–2).

Routine testing of health care personnel,

first responders, and other frontline

workers in eight U.S. locations in six

states from December 14, 2020 through

August 14, 2021 revealed 194 infections

in 4,136 unvaccinated participants

(89.7% symptomatic) and 34 infections

in 2,976 fully vaccinated participants

(80.6% symptomatic) (Fowlkes et al.,

August 27, 2021). During time periods

when the Delta variant represented

more than 50% of viruses sequenced, 19

infections were detected in 488

unvaccinated participants (94.7%

symptomatic) and 24 infections were

detected in 2,352 vaccinated

participants (75% symptomatic).

Monthly COVID–19 cases in

healthcare workers were reported

during the period from March 1 to July

31, 2021 at the University of California

San Diego (UCSD) health system, which

is a healthcare provider that includes

primary care services such as family

medicine and pediatrics (Keehner et al.,

September 1, 2021; UCSD, 2021). During

that time period, a total of 227 health

care workers tested positive for COVID–

19. One hundred and nine of 130 fully

vaccinated workers who tested positive

(83.8%) were symptomatic and 80 of 90

unvaccinated workers (88.9%) were

symptomatic; one unvaccinated person

was hospitalized for COVID–19

symptoms. By July of 2021, after the end

of California’s mask mandate on June 15

and after the Delta variant became

dominant, the number of cases detected

dramatically increased; the Delta variant

accounted for more than 95% of SARS–

CoV–2 viruses sequenced by the end of

that month. During July of 2021,

symptomatic infections were detected in

94 of 16,492 fully vaccinated workers

and 31 of 1,895 unvaccinated workers.

Attack rates in July of 2021 were 5.7 per

1,000 fully vaccinated workers and 16.4

per 1,000 unvaccinated workers.

In Finland, a Delta variant infection

from a hospitalized patient spread

throughout the hospital and to three

primary care facilities, infecting 103

individuals, including 45 healthcare

workers (Hetemäki et al., July 29, 2021).

Twenty-six of the healthcare workers

were infected at the hospital and 19

were infected at primary care facilities.

The affected health care workers

included 28 with direct patient contact

(11 who were not fully vaccinated), 8

unvaccinated healthcare worker

students, and 9 other staff, including

hospital cleaners and secretaries (of

whom 6 were not fully vaccinated).

According to study authors, ‘‘There was

high vaccine coverage among permanent

staff in the central hospital, but lower

for HCW in primary healthcare

facilities. . .’’ Study authors estimated

that vaccine effectiveness against the

Delta variant in healthcare workers was

approximately 88–91%, suggesting how

much more extensive the outbreak

could have been if a high percentage of

healthcare workers were not fully

vaccinated.

In the UK, a Delta variant infection in

a healthcare worker resulted in an

outbreak in a care home that affected 16

of 21 residents and 8 of 21 staff

(Williams et al., July 8, 2021). One staff

member was hospitalized. Attack rates

were 35.7% in staff who were partially

vaccinated (i.e., received their second

dose of vaccine on the day that the

index case was diagnosed with COVID–

19 or had only received one vaccine

dose) and 40% in staff who were not

vaccinated.

Recent news stories demonstrate that

outbreaks affecting staff members are

still occurring in U.S. healthcare

facilities. An outbreak that began in

August, 2021 at a Washington State

nursing center resulted in infections in

22 staff members and 52 residents. In an

unrelated outbreak, a nursing facility in

Hawaii reported infections in 24

employees and 54 patients (Wingate,

September 24, 2021). Vaccination rates

were reported at 64.5% of residents and

37.1% of staff in the Washington State

facility and 91% of staff and more than

80% of patients at the Hawaii facility.

COVID–19 cases were also observed

in staff at ambulatory care settings prior

to emergence of the Delta variant. Over

an 11-week period beginning on March

20, 2020, 254 tests for SARS–CoV–2

were performed on employees who had

potential exposures at an outpatient

urology center in New York State

(Kapoor et al., 2020). Positive test rates

in employees correlated with rates in

New York State, declining over time,

from 26.1% in the early stage to 7.3%

in the late stage of the study. According

to study authors, the positive test results

coincided with the implementation of

infection control procedures (e.g.,

symptom screening, masking,

distancing, and hygiene). Positivity rates

were similar in administrative and

clinical staff and the study authors

concluded that ‘‘administrative staff in

an outpatient setting were equally—if

not more—vulnerable to SARS–CoV–2

transmission when compared with

clinical staff who were more directly

exposed to patients.’’ The study authors

speculated that possible reasons for the

findings were that clinical staff were

more familiar with PPE and that

administrative staff, especially in checkin and check-out points, tend to work

close to each other.

App. 16

Federal Register / Vol. 86, No. 212 / Friday, November 5, 2021 / Rules and Regulations

c. Conclusion for Employee Impact

The evidence described above

provides examples of the impact that

exposures from SARS–CoV–2, including

those involving the Delta variant, have

had on employees in general industry,

agriculture, construction, maritime, and

healthcare settings. It demonstrates that

SARS–CoV–2 has spread to employees

in these industries and, in many cases,

infection was linked to exposure to

infected persons at the worksite (WSDH,

September 8, 2021; OHA, September 1,

2021; TDH, September 8, 2021;

NCDHHS, August 30, 2021; Hawaii

State, August 19, 2021; Pray et al.,

January 29, 2021; Sami et al., April 9,

2021; Suhs et al., July 23, 2021; Gold et

al., February 26, 2021; Porter et al.,

April 30, 2021; Hetemäki et al., July 29,

2021; Williams et al., July 8, 2021). The

documentation of so many workplace

clusters suggests that exposures to

SARS–CoV–2 occur regularly in

workplaces where employees come into

contact with others. This prevalence of

clusters, combined with some evidence

that many infections occurred within

the 14-day incubation period for SARS–

CoV–2 and that exposures to infected

persons outside the workplace were

frequently ruled out, supports the

proposition that exposures to and

transmission of SARS–CoV–2 occur

frequently at work. Multiple studies

demonstrate high rates of COVID

infections, illnesses, and fatalities in the

wide range of occupations that require

frequent or prolonged close contact with

other people, indoor work, and work in

crowded and/or poorly ventilated areas

The large numbers of infected

employees suggest that SARS–CoV–2 is

likely to be present in a wide variety of

workplaces, placing unvaccinated

workers at risk of serious and

potentially fatal health effects.

IV. Vaccines Effectively Reduce Severe

Health Outcomes From and

Transmission of SARS–CoV–2

During the course of the SARS–CoV–

2 pandemic, different variants have

emerged with different characteristics

that better enable transmission and

potentially cause more severe outcomes.

However, vaccines remain very effective

at reducing the occurrence of COVID–

19-related severe illness, disability and

death.12 The Delta variant is more

transmissible than previous variants,

might cause more severe illness than

previous variants in unvaccinated

12 A discussion of vaccination rates, as well as

OSHA’s rationale for why vaccination is a critical

means of protecting workers from the grave danger

described in this section, can be found in Need for

the ETS (Section III.B. of this preamble).

people, and has led to hospitalization of

individuals in numbers similar to those

of the November 2020 to February 2021

surge. These changes in characteristics

have provided a clearer realization of

the continuing capacity for SARS–CoV–

2 to present a grave danger to workers.

However, it is well evident that even

given these changed characteristics of

Delta, serious disease and death

continue to occur overwhelmingly in

unvaccinated individuals while the

vaccinated are afforded great

protection.13

a. Impact of Vaccination on Severe

Health Outcomes

There are currently three vaccines

that are approved or authorized for the

prevention of COVID–19 in the U.S.:

The Pfizer-BioNTech COVID–19 vaccine

(FDA approved for ages 16 and above;

authorized for ages 12 and above), the

FDA-authorized Moderna COVID–19

vaccine (authorized for ages 18 and

above), and the FDA-authorized Janssen

COVID–19 vaccine (also known as the

Johnson & Johnson vaccine; authorized

for ages 18 and above.) Pfizer-BioNTech

and Moderna are mRNA vaccines that

require two primary series doses

administered three weeks and one

month apart, respectively. Janssen is a

viral vector vaccine administered as a

single primary vaccination dose (CDC,

September 15, 2021). The vaccines were

shown to greatly exceed minimum

efficacy thresholds in preventing

COVID–19 in clinical trial participants

(FDA, December 11, 2020; FDA,

December 18, 2020; FDA, February 26,

2021). Data from clinical trials for all

three vaccines and observational studies

for the two mRNA vaccines clearly

establish that fully vaccinated persons

have a greatly reduced risk of SARS–

CoV–2 infection compared to

unvaccinated individuals. This includes

severe infections requiring

hospitalization and those resulting in

death. For more information about the

effectiveness of vaccines as of late

Spring 2021, see 86 FR 32397, which

OSHA hereby includes in the record for

this ETS.14

Vaccines remain highly effective

against hospitalization and death. A

study evaluating vaccine effectiveness at

preventing hospitalization among those

with SARS–CoV–2 infections in New

13 While mild cases of COVID–19 are included in

the grave danger presented by COVID–19, as stated

in the Healthcare ETS (see 86 FR 32382), OSHA is

focusing on the most severe health effects, i.e., cases

requiring hospitalization and cases resulting in

death, in this new rulemaking effort in order to

prevent the gravest of consequences to workers.

14 This adoption includes the citations in the

referenced section of the Healthcare ETS, which are

also included in the docket for this ETS.

61417

York found that effectiveness did not

change from May 3 to July 25, 2021 as

the Alpha variant gave way to the Delta

variant (91.9–96.2% range; Rosenberg et

al., August 27, 2021). Grannis et al. used

data from 187 hospitals in nine states

from June to August 2021 to evaluate

the efficacy of vaccines against

hospitalization when Delta had emerged

as the predominant variant causing

SARS–CoV–2 infections (September 17,

2021). This study found that vaccines

were 89% effective at preventing

hospitalization in individuals aged 18 to

74. Similarly, vaccines were also found

to be 89% effective in preventing

hospitalization in a study collecting

data from five Veteran Affairs Medical

Centers from July 1 to August 6, 2021,

a time when most transmission was

attributed to the Delta variant (Bajema et

al., September 10, 2021).

Two other studies found that,

although the level of protection

provided by vaccination has decreased

somewhat with the emergence of the

Delta variant, vaccines continue to

provide high levels of protection against

hospitalization. In a U.S. study,

researchers found that while the

Moderna and Janssen vaccines mostly

maintained their effectiveness at

preventing hospitalization (going from

93% to 92% after more than 120 days

post-vaccination and 71% to 68% after

more than 28 days post-vaccination,

respectively) from March to August

2021, the effectiveness of the PfizerBioNTech vaccine at preventing those

severe outcomes decreased from 91% to

77% after more than 120 days postvaccination (Self et al., September 17,

2021). An Israeli study on infections

documented between July 11 and July

31, 2021 found a significant decrease in

vaccine efficacy for the Pfizer-BioNTech

vaccine against severe outcomes in

relation to when an individual was

vaccinated, but the absolute difference

was much less than what was observed

in the U.S. study (e.g., 98% effective for

40–59 year olds vaccinated in March

versus 94% effective for those in the

same age group who were vaccinated in

January) (Goldberg et al., August 30,

2021).

Vaccines also remain extremely

effective at preventing death. A UK

study evaluated the effectiveness of the

Pfizer-BioNTech vaccine against death

and found it to be 96.3% effective

against the Alpha strain and 95.2%

protective against the Delta strain

(Andrews et al., September 21, 2021).

Two Israeli studies, Haas et al. and

Saciuk et al., performed during time

periods where Alpha was predominant,

found the Pfizer-BioNTech vaccine to be

96.7% and 91.1% effective,

App. 17

61418

Federal Register / Vol. 86, No. 212 / Friday, November 5, 2021 / Rules and Regulations

respectively, against death (Haas et al.,

May 15, 2021; Saciuk et al., June 25,

2021). A California study found that the

Moderna vaccine was 97.9% effective

against death (Bruxvoort et al.,

September 2, 2021). A study on patients

served by the Veterans Health

Administration found that PfizerBioNTech and Moderna vaccines

provided 99% effectiveness against

death (Young-Xu et al., July 14, 2021).

The risks of hospitalization and death

appear to have increased for

unvaccinated individuals since the

Delta variant became a common source

of infections. A study of Los Angeles

County SARS–CoV–2 infections found

that vaccinations reduced

hospitalization risk by a factor of 10 on

May 1, 2021, when the Alpha variant

was dominant, but that the risk of

hospitalization was even more greatly

reduced (by a factor of 29.2) on July 25,

2021, when the Delta variant was

dominant (Griffin et al., August 27,

2021). This difference suggests both that

vaccines continue to provide a high

level of protection against disease that

results in hospitalization and that risk

has increased for those who are

unvaccinated. Similar increased risk for

unvaccinated individuals was reported

in a study that evaluated hospitalization

and death data from 13 U.S.

jurisdictions between June 20 and July

17, 2021, a period when the Delta

variant gained prominence (Scobie et

al., September 17, 2021). For

unvaccinated 18 to 49 year olds, the risk

of hospitalization was 15.2 times

greater, and the risk of death was 17.2

times greater, than the risks for

vaccinated people in the same age

range. For unvaccinated 50 to 64 year

olds, the risk of hospitalization was 10.9

times greater, and the risk of death was

17.9 times greater, than for those who

are vaccinated. These studies illustrate

that vaccination is an extremely

effective control measure to minimize

severe outcomes resulting from Delta

variant infections.

b. Impact of Vaccination on Infection

and Transmission

Vaccines continue to provide robust

protection for vaccinated individuals

against SARS–CoV–2 infections, even

though several studies indicate that

vaccine efficacy against infection may

have decreased somewhat with the

emergence of the Delta variant (Fowlkes

et al., August 27, 2021; Rosenberg et al.,

August 27, 2021; Nanduri et al., August

27, 2021; Seppala et al., September 2,

2021; Bernal et al., August 12, 2021).

For example, vaccination was observed

to reduce the risk of infection by a factor

of 8.4 on May 1, 2021, when the Alpha

variant was predominant in Los Angeles

county (Griffin et al., August 27, 2021).

However, the level of protection had

fallen to a factor of 4.9 by July 25, 2021,

when Delta made up 88% of infections

in the county. The findings from this

study indicate that while vaccines

maintain robust protection against

severe outcomes, protection against

infection has fallen with the increased

circulation of the Delta variant. A

broader study using data from 13 U.S.

jurisdictions had similar findings,

observing that the protection vaccines

afforded against infection decreased

from a factor of 11.1 (i.e., vaccinated

people were 11.1 times less likely than

unvaccinated people to become

infected) between April 4 and June 19,

2021, to a factor of 4.6 between June 20

and July 17, 2021 (Scobie et al.,

September 17, 2021). An additional

study noted, however, that the decrease

in vaccine protectiveness against

symptomatic infection from the Delta

variant could be due to the waning of

immunity specifically in older

populations. Andrews et al. (September

21, 2021) found that while the PfizerBioNTech vaccine effectiveness

decreased from 94.1% to 67.4% in those

65 years old and older, vaccine

effectiveness for those 40 to 64 years old

only decreased from 92.9% to 80.6%.

While infections themselves do not

normally result in serious illness for

those who are vaccinated, evidence

shows that vaccinated individuals who

become infected with the Delta variant

can transmit the disease more easily to

others than with previous variants. This

development poses a great concern for

the unvaccinated, who generally do not

have the protections against severe

outcomes that vaccination affords.

Before Delta, vaccinated individuals

were shown to have lower estimated

viral loads when infected than those

who were unvaccinated, which

suggested that infected vaccinated

individuals were likely not a major

concern for transmission (LevineTiefenbrun et al., March 29, 2021).

Transmission studies prior to the

emergence of Delta appear to bear this

out. A Scottish study performed during

a time period when the Alpha variant

was predominant in the region, showed

that a fully vaccinated individual was

3.2 times less likely than an

unvaccinated individual to transmit the

virus to unvaccinated family members

(Shah et al., September 10, 2021;

supplementary appendix). A

population-based study from the

Netherlands found that vaccination

decreased secondary transmission to

household members from 31% to 11%

(de Gier et al., August 5, 2021).

Additionally, a study from the UK

found that household transmission

decreased by as much as 50% when the

infected individual was vaccinated

(Harris et al., June 23, 2021).

More recent research suggests that the

Delta variant may have reduced the

level of protection vaccination affords

against transmission of the virus to

others, but still significantly reduces

transmission risk in comparison to

infected unvaccinated individuals. A

UK study found that fully vaccinated

individuals infected by the Delta variant

are able to transmit the virus to both

vaccinated and, to a greater degree,

unvaccinated persons (Singanayagam et

al., September 6, 2021). Still, the rate at

which transmission to unvaccinated

individuals occurred was nearly double

the rate of transmission to vaccinated

individuals (35.7% compared to 19.7%).

Similarly, Eyre et al., (September 29,

2021) found that during the

predominance of Alpha, full vaccination

with the Pfizer-BioNTech vaccines

resulted in a significant reduction in

transmission to others (an adjusted

Odds Ratio (aOR) of 0.18, meaning that

being unvaccinated increased the odds

of transmission by over five times). With

the rise of the Delta variant, that

reduction in transmission to others was

less than with the Alpha variant, but

still significantly more than for

unvaccinated individuals (aOR of 0.35,

meaning that being unvaccinated

increased the odds of transmission by

almost three times).

The greater ability for vaccinated

individuals to transmit the Delta variant

of SARS–CoV–2 to others (compared to

previous variants) appears to be linked

to the generation of similar viral loads

(as estimated by Ct threshold) in the

vaccinated compared to the

unvaccinated (Ct threshold is the

number of RT–PCR cycles that need to

be run in order to amplify the RNA

enough to be detected—fewer cycles

means a greater initial amount of virus

was collected) (Singanayagam et al.,

September 6, 2021). This observation

has been made in several studies. A

study from Israel observed that viral

loads among those infected with the

Delta variant were only decreased in

people who had been vaccinated

recently (within the past two months) or

in those who had recently received a

booster dose (Levine-Tiefenbrun et al.,

September 1, 2021). In a study of SARS–

CoV–2 infections in Los Angeles

County, performed when the Delta

variant was predominant, vaccination

status did not appear to affect the

estimated viral loads, suggesting that

infected individuals who are vaccinated

App. 18

Federal Register / Vol. 86, No. 212 / Friday, November 5, 2021 / Rules and Regulations

may be just as likely to transmit the

virus (Griffin et al., August 27, 2021).

Additionally, estimated viral loads did

not appear to be significantly different

with respect to vaccination status in a

Wisconsin study (Riemersma et al., July

31, 2021). Regardless of viral loads in

vaccinated and unvaccinated

individuals, the fact remains clear that

unvaccinated people pose a higher risk

of transmission to others than

vaccinated people, simply because they

are much more likely to get COVID–19

in the first place.

These studies, however, appear to

overstate increases in transmission risk

from vaccinated individuals related to

the Delta variant. From May to July

2021, UK researchers tested individuals

at random to better characterize viral

load estimates in people with

asymptomatic as well as symptomatic

infections; they found that vaccination

was associated with a significantly

lower estimated viral load (Elliott et al.,

September 10, 2021). This more

comprehensive study (i.e., Elliott et al.,

September 10, 2021) may have been able

to better characterize the course of

infection and to incorporate vaccinated

individuals whose viral loads were

decreasing quickly. The findings in

Elliott et al. are consistent with studies

observing that viral load may fall more

quickly in vaccinated individuals,

resulting in a shorter infectious period

and possibly fewer transmission events

(Chia et al., July 31, 2021; Eyre et al.,

September 29, 2021).

c. Conclusion for the Impact of Vaccines

The studies discussed above indicate

that vaccines continue to effectively

protect vaccinated individuals against

SARS–CoV–2 infections, while the risk

of infection, hospitalization, and death

increased among unvaccinated people

as the Delta variant became

predominant in the U.S. The Delta

variant is even more dangerous to

unvaccinated individuals than previous

variants because of the higher

transmission potential from both

unvaccinated and vaccinated people.

Because unvaccinated individuals are at

much higher risk of severe health

outcomes from infection with SARS–

CoV–2, and also pose a greater

transmission risk to those around them,

it is critical to assure that as many

people as possible are fully vaccinated

in order to prevent transmission at

work.

V. Coverage of OSHA’s Grave Danger

Finding

Based on the information discussed

above, OSHA finds that many

unvaccinated workers across the U.S.

economy are facing a grave danger of

severe health effects or death from

exposure to SARS–CoV–2. Fully

vaccinated workers are not included in

this grave danger finding because, as

described throughout this section, those

who are fully vaccinated are much

better protected from the effects of

SARS–CoV–2 and, in particular, the

most severe effects, than are those who

are unvaccinated.15 Beyond that,

OSHA’s grave danger determination

exempts several categories of workers

based on characteristics of their work or

workplace: (1) Workers who do not

report to a workplace where other

individuals are present or who telework

from home; and (2) workers who

perform their work exclusively

outdoors. The basis for these

exemptions is explained below. In this

section, OSHA also addresses the basis

for OSHA’s grave danger finding for

workers who are unvaccinated yet had

a prior COVID–19 infection, and

explains the Agency’s more nuanced

grave danger finding in the healthcare

industry.

a. Employees Who Telework and

Employees Who Do Not Report to a

Workplace Where Other People Are

Present.

Employees who report to workplaces

where no other people are present face

no grave danger from occupational

exposure to COVID–19 because such

exposure requires the presence of other

people. For those who work from their

homes, or from workplaces where no

other people are present (such as a

remote worksite), the chances of being

exposed to SARS–CoV–2 through a

work activity are negligible. Therefore,

OSHA is exempting those workers who

do not come into contact with others for

work purposes from its grave danger

finding as well as the scope of the ETS

(for more information, see the Summary

and Explanation for Scope and

Application, Section VI.B. of this

preamble).

b. Employees Who Work Exclusively

Outside

Employees who work exclusively

outside face a much lower risk of

15 The exclusion of vaccinated workers from this

grave danger finding does not mean that vaccinated

workers face no risk from exposure to SARS–CoV–

2. The best available evidence clearly shows that

vaccination provides great protection from infection

and severe outcomes, but breakthrough infections

do occur and vaccinated individuals can still

transmit the virus to others. In some cases, the level

of risk to vaccinated workers may even rise to the

level of a significant risk, the standard OSHA must

meet for promulgation of a permanent standard

under section 6(b)(5) of the OSH Act (29 U.S.C.

655(b)(5)).

61419

exposure to SARS–CoV–2 at work,

because their workplaces typically do

not include any of the characteristics

that normally enable transmission to

occur (e.g., indoors, lack of ventilation,

crowding). Bulfone et al. attributed the

lower risk of transmission in outdoor

settings (i.e., open air or structures with

one wall) to increased ventilation with

fresh air and a greater ability to

maintain physical distancing (November

29, 2020). While the best available

evidence firmly establishes a grave

danger in indoor settings, the CDC has

stated that the risk of outdoor

transmission is ‘‘low’’ (CDC, September

1, 2021) and OSHA is unable to

establish a grave danger in outdoor

settings from exposure during normal

work activities.

OSHA recognizes that outdoor

transmission has been identified in a

few specific incidents (e.g., 2 of 7,324

cases, Qian et al., October 27, 2020).

However, general reviews of

transmission studies that include largescale and high-density outdoor

gatherings indicate that indoor

transmission overwhelmingly is

responsible for SARS–CoV–2

transmission. Additionally, the lack of

evidence tied to specific case studies

illustrating outdoor transmission in

comparison to the bevy of case studies

on indoor transmission makes it

difficult to support a conclusion that

outdoor transmission rises to the level

of a grave danger.

Bulfone et al. reviewed a collection of

SARS–CoV–2 studies that evaluated

infections in outdoor and indoor

settings (November 29, 2020), and found

that transmission is significantly less

likely to occur in outdoor settings than

in indoor settings. The studies overall

found that the risk of outdoor

transmission was less than 10% of the

risk of transmission in indoor settings,

with three of the studies concluding risk

was 5% or less of the risk of

transmission in indoor settings. While

acknowledging significant gaps in

knowledge, the authors of a different

study suggested that increases in

transmission related to large events such

as the Sturgis motorcycle rally may be

related to lack of local efforts to prevent

transmission indoors (e.g., requiring the

wearing of masks, closing indoor

dining), rather than the outdoor setting

for the rally (Dave et al., December 2,

2020). In contrast, transmission rates

did not increase as expected following

the Summer 2020 protests on racial

injustice. This outcome was attributed,

in part, to participants having been less

likely to enter indoor commercial

establishments.

App. 19

61420

Federal Register / Vol. 86, No. 212 / Friday, November 5, 2021 / Rules and Regulations

Weed and Foad (September 10, 2020)

found that transmission of SARS–CoV–

2 related to large scale outdoor

gatherings could be largely attributed to

individual behaviors related to that

event, such as communal travel and

indoor congregation at other facilities

(e.g., restaurants, shared

accommodations), rather than to the

time spent outdoors at those gatherings.

Similarly, a Public Health England

evaluation of the literature on SARS–

CoV–2 and surrogate respiratory viruses

(December 18, 2020) also concluded that

when transmission does occur at

outdoor events, outdoor activities were

mixed with indoor setting use. Public

Health England concluded that the vast

majority of transmission happens in

indoor settings, with very little evidence

for outdoor transmission.

A systemic review of SARS–CoV–2

clusters identified 201 events through

May 26, 2020 (Leclerc et al., April 28,

2021), only 4 of which occurred at

predominantly outdoor settings. For

those 4 clusters, the authors noted that

they were not able to evaluate specific

transmission events and attributed it to

local health agencies being

overwhelmed by the pandemic. OSHA

notes that the designations of settings in

this study are somewhat generic, as

outdoor construction sites will often

have indoor locations, such as mobile

offices, or locations with reduced

airflow, such as areas with a roof or

ceiling and two or more walls.

Regardless, this study illustrates the

comparable abundance of evidence

available to evaluate SARS–CoV–2

transmission in indoor settings versus

outdoor settings.

Cevik et al. (August 1, 2021) reviewed

studies on the transmission dynamics of

SARS–CoV–2 infections from large

scale, contact-tracing studies. The

authors recommended that, based on the

evidence that outdoor transmission

dynamics resulted in significantly fewer

infections than in indoor settings,

public health entities should greatly

encourage use of outdoor settings. The

researchers highlighted a study by

Nishiura et al. (April 16, 2020), who

evaluated 110 cases in Japan at the

beginning of the pandemic and found

that outdoor settings reduced

transmission risk by 18.7 times and

reduced the risk of super-spreader

events by 32.5 times.

Agricultural workplace settings have

experienced significant SARS–CoV–2

infections. However, transmission in

these settings is difficult to characterize

because many jobs in this sector include

both outdoor and indoor activities.

Miller et al. (April 30, 2021) evaluated

an outbreak among farmworkers in

Washington State. The researchers

found that 28% of workers with

predominantly indoor tasks where they

were unable to maintain physical

distance were infected, compared to 6%

of workers who performed

predominantly outdoors tasks in the

orchards. Conversely, a study on

farmworkers in Monterey County,

California found a significant

correlation between evidence of

infection and individuals who worked

in the fields as opposed to indoor work

(Mora et al., September 15, 2021). The

paper noted that infections were

predominant in individuals who lived

in crowded conditions, commuted

together to the fields, and spoke at home

in indigenous languages, which is

important as written health messages

are often not available in all worker

languages. These papers cannot identify

where or when infections occurred in

order to discern causation. The

associations observed may indicate that

SARS–CoV–2 infections may be more

related to aspects related to indoor

exposures outside of the work activities

(e.g., crowded living conditions) or

potentially overlooked indoor aspects

connected to outdoor work (e.g., shared

commuting).

Several studies discussed below in

more detail have evaluated outdoors onfield transmission from infected

participants during football, soccer, and

rugby matches. These events include

repeated close physical contact between

players, without PPE or physical

distancing, over the course of fairly long

events, with increased exertion leading

to greater respiratory effort and

production of respiratory droplets.

These events also include opposing

cohorts who only interact during onfield activities. Therefore, these studies

provide some evidence for the low

likelihood of outdoor transmission in

other workplace activities greatly

impacted by the pandemic, such as in

construction.

Mack et al. (January 29, 2021) detailed

the National Football League’s complex

program to assess and prevent

transmission, which included devices

that recorded distance and duration of

interactions with others, for the purpose

of improving identification of

individuals with high-risk exposures.

Although 329 positive cases were

identified among roughly 11,400 players

and staff, there were no reported cases

of on-field transmission by infected

players. The results led the NFL to focus

more on reducing transmission in

indoor settings, including

transportation.

Egger et al. (March 18, 2021) reviewed

three soccer matches involving 18

players who had SARS–CoV–2; one

match involved a team where 44% of

the players were infected. Video

analysis was used to determine the type

of contact between players, such as

contact to face or hand slaps. None of

the existing cases were associated with

on-field play and no secondary

transmission from on-the-field contacts

was observed. Jones et al. (February 11,

2021), evaluated four rugby Super

League matches involving eight players

who were found to be infected with

SARS–CoV–2. Using video footage and

global positioning data, the researchers

were able to identify 28 players as highrisk contacts with the infected players.

These high-risk players together had as

many as 32 tackles and were within two

meters of infected players as often as

121 times during the four matches. Of

the 28 players noted as high-risk

contacts, one became infected with

SARS–CoV–2. However, researchers

determined that the transmission

resulted from internal team outbreaks

and not from exposure on the field.

OSHA acknowledges that the risk of

transmission of SARS–CoV–2 in

outdoor settings is not zero, and that

there may be some low risk to workers

performing general tasks exclusively in

outdoor settings. However, where

studies have been able to differentiate

between indoor and outdoor exposures,

they indicate that indoor exposures are

the much more significant drivers of

SARS–CoV–2 infections. Therefore, the

best available evidence at this time does

not provide OSHA with the information

needed to establish SARS–CoV–2 as a

grave danger for general work activities

in outdoor settings (see Int’l Union,

United Auto., Aerospace, & Agr.

Implement Workers of Am., UAW, 590

F. Supp. at 755–56, describing a ‘‘grave

danger’’ as a risk that is more than

‘‘significant’’). Therefore, OSHA has

excluded employees who work

exclusively outdoors from the scope of

this ETS (see the Summary and

Explanation for Scope and Application,

Section VI.B. of this preamble).

c. Employees in Healthcare

Because OSHA issued a separate

grave danger determination several

months ago for some healthcare

workers, some explanation of how its

current finding applies to healthcare

workers is necessary. In June 2021,

OSHA issued its Healthcare ETS (86 FR

32376) after determining that some

healthcare workers faced a grave danger

of infection from SARS–CoV–2. This

grave danger determination, along with

the protections of the Healthcare ETS,

applied to healthcare and healthcare

support workers in settings where

App. 20

Federal Register / Vol. 86, No. 212 / Friday, November 5, 2021 / Rules and Regulations

people with suspected or confirmed

cases of COVID–19 are treated, and was

based on the increased potential for

transmission of the virus in such

settings (see 86 FR 32411–32412). These

workers are currently covered by the

protections of the Healthcare ETS (29

CFR 1910.502). OSHA does not have

data to demonstrate that unvaccinated

workers in settings covered by the

Healthcare ETS face a grave danger from

SARS–CoV–2 when the requirements of

that standard are followed. However, if

the Healthcare ETS were no longer in

effect, OSHA would consider the

workers who were covered by it, and

who remain unvaccinated, to be at grave

danger for the reasons described in this

ETS.

OSHA’s new finding of grave danger

applies to healthcare and healthcare

support workers who are not covered by

the Healthcare ETS, to the extent they

remain unvaccinated. In this ETS, as

discussed in this section, OSHA has

made a broader determination of grave

danger that applies to most

unvaccinated workers, regardless of

industry. OSHA’s current finding of

grave danger supporting this ETS does

not depend on whether a workplace is

one where people with suspected or

confirmed COVID–19 are expected to be

present. Therefore, the finding of grave

danger applies to unvaccinated workers

in healthcare settings that are not

covered by 29 CFR 1910.502 to the same

extent it applies to unvaccinated

workers in all other industry sectors.

d. Employees Who Were Previously

Infected With SARS–CoV–2

OSHA has carefully evaluated the

effectiveness of previous SARS–CoV–2

infections in providing protection

against reinfection. This section

provides a detailed description of the

current scientific information in order to

ascertain what the best available

scientific evidence on this topic

indicates regarding the risk to

individuals with previous COVID–19

infections from exposure to SARS–CoV–

2. While the agency acknowledges that

the science is evolving, OSHA finds that

there is insufficient evidence to allow

the agency to consider infectionacquired immunity to allay the grave

danger of exposure to, and reinfection

from, SARS–CoV–2.

To determine whether employees

with infection-induced immunity from

SARS–CoV–2 (i.e., those who were

infected with SARS–CoV–2 but have not

been vaccinated) face a grave danger,

OSHA reviewed the scientific evidence

on the protective effects of vaccineinduced SARS–CoV–2 immunity versus

infection-induced immunity. Individual

immunity to any infectious disease,

including SARS–CoV–2, is achieved

through a complex response to exposure

by the immune system. This response

consists of disease-specific antibody

production guided and augmented by

certain types of immune cells, such as

T and B cells, which work together to

neutralize or destroy the disease-causing

agent. Immune responses to viruses like

SARS–CoV–2 can be measured in

several ways. For instance, blood serum

can be taken and exposed to specific

proteins found on the SARS–CoV–2

virus, in order to measure the presence

of antibodies in the blood. Another

antibody test, the neutralization test,

measures the ability of the antibodies

present in a serum to neutralize

infectivity and prevent cells from being

infected. T cell immunity can be

measured using techniques that target a

specific biomolecule that is specific to

SARS–CoV–2.

A considerable number of individuals

who were previously infected with

SARS–CoV–2 do not appear to have

acquired effective immunity to the virus

(Psichogiou et al., September 13, 2021;

Wei et al., July 5, 2021; Cavanaugh et

al., August 13, 2021). The level of

protection afforded by infectioninduced immunity appears to depend

on the severity of individuals’

infections. In a study from Greece,

immunogenicity was compared between

healthcare workers who were

vaccinated with Pfizer-BioNTech and

unvaccinated patients who acquired a

natural infection (Psichogiou et al.,

September 13, 2021). The researchers

found that the immune response in

unvaccinated individuals correlated to

the severity of their disease. Fully

vaccinated healthcare workers had

immune responses (measured as

antibody levels specific to SARS–CoV–

2) that were 1.3 times greater than

patients who had critical cases of

COVID–19 cases, 2.5 times greater than

patients who had moderate to severe

cases, and 10.5 times greater than

patients who had asymptomatic/mild

illnesses. Similarly, another study found

that 24.0% (1,742 of 7,256) of

individuals who had a previous SARS–

CoV–2 infection were seronegative (i.e.,

did not produce antibodies in response

to the virus), suggesting that the

previous infection provided insufficient

protection against future infection (Wei

et al., July 5, 2021). Individuals who

were seronegative were typically older,

had lower viral burdens when infected,

and were more likely to be

asymptomatic. The authors posited that

the immunity of those who were

seropositive (i.e., did produce

61421

antibodies in response to the virus)

would provide some measure of

protection, but that these individuals

would benefit from a vaccination

booster. This position appears to be

validated by a study that compared the

reinfection rates of individuals in

Kentucky based on their post-recovery

vaccination status (Cavanaugh et al.,

August 13, 2021). Unvaccinated

individuals with previous infection

were found to be 2.3 times more likely

to be reinfected than those who were

vaccinated after their prior infection.

These studies demonstrate not only that

those with milder infections may not be

protected against future infection, but

that it is difficult to tell, on an

individual level, which individuals

might have had prior infections that

conveyed protection equivalent to that

provided by vaccination.

A number of other studies indicate

that fully vaccinated individuals may be

better protected against future infection

than those with previous infections. A

study in Massachusetts concluded that

the immunity conveyed from a previous

SARS–CoV–2 infection was effectively

equivalent to the immunity of an

uninfected individual who has had only

one dose of an mRNA vaccine

(Naranbhai et al., October 13, 2021). The

authors found that fully vaccinated

individuals have an immune response

(i.e., antibodies and neutralization) well

above the levels observed in

unvaccinated, previously-infected

individuals. German researchers found

that individuals who were fully

vaccinated with Pfizer-BioNTech had a

significantly greater immune response

(as measured by antibody levels) than

unvaccinated individuals who had

infections, concluding that vaccination

would be needed for those unvaccinated

individuals to have similar protection

against infection (Herzberg et al., June

13, 2021). Similarly, a Dutch study

observed that vaccination greatly

improved the immune response (as

measured by antibodies and virusspecific T cells) of individuals who had

recovered from COVID–19 (Geers et al.,

May 25, 2021). Planas et al. (August 12,

2021) also noted that immune response

(as measured by neutralization) to the

Alpha, Beta, and Delta (B.1.617.2)

variants in unvaccinated, previouslyinfected individuals was considerably

less than the immune response in

individuals five weeks after their second

Pfizer-BioNTech dose. When

unvaccinated, previously-infected

individuals were vaccinated, their

immune response (as measured by

neutralization) increased by more than

an order of magnitude. Likewise, Wang

App. 21

61422

Federal Register / Vol. 86, No. 212 / Friday, November 5, 2021 / Rules and Regulations

et al. (July 15, 2021) found that the

immune response (as measured by

neutralization) of those with previous

SARS–CoV–2 infection increased by

more than an order of magnitude against

Alpha (B.1.1.7), Beta (B.1.351), Iota

(B.1.526), and Gamma (P.1) variants

when they were vaccinated. These

studies show that infection-induced

immunity may not equal the protection

afforded by vaccination and that

vaccination greatly improves the

immune response of those who were

previously infected.

The aforementioned studies indicate

that immunity acquired through

infection appears to be less protective

than vaccination. There are also a

number of epidemiological studies that

provide some evidence that infectionacquired immunity has the potential to

provide a significant level of protection

against reinfection. As OSHA discusses

in greater detail below, these studies

suffer from methodological limitations

that render them inconclusive about the

level of immunity conferred by

infection, and therefore OSHA is unable

to establish that such immunity

eliminates grave danger. This

determination is based in three parts.

First, the epidemiological literature

OSHA reviewed generally suffers from

selection bias to a degree that it serves

as an unreliable basis on which to reach

a robust conclusion on whether

previous infection removes workers

from grave danger. In general, the

studies described below do not account

for people who had mild COVID–19

infections, leading to study findings

regarding the level of protection

afforded by prior infection that are not

generally applicable. Second, the tests

employed in the studies are being used

in ways that they were not originally

designed to be employed. These tests

are powerful tools, but there are

limitations to their use in determining if

a specific individual is, in fact,

protected from the grave danger of

SARS–CoV–2. Particularly problematic

is the lack of established thresholds to

determine full protection from

reinfection or even a standardized

methodology to determine infection

severity or immune response. Thus,

while these studies broadly establish

some increase in protectiveness against

SARS–CoV–2 among the studied

populations, they as yet are unable to

provide a reasonable degree of certainty

on whether the degree of protection

afforded any particular individual from

their prior infection is sufficient to

eliminate the grave danger from

reinfection (see Milne, et al., October 21,

2021.) Third, while the research

methodology itself creates difficulties in

the context of OSHA’s grave danger

inquiry, the implications of trying to

apply investigative research

methodology to clinical practice are

even more challenging. The need for the

development of standardized methods

and criteria for establishing sufficient

immunity preclude the application of

the studies’ findings to robust and

reliable clinical practice. These three

rationales for OSHA’s finding are

described in more detail below.

Several epidemiological studies used

previous RT–PCR positive cases to

define previous infections (Hansen et

al., March 27, 2021; Pilz et al., February

11, 2021; Vitale et al., May 28, 2021;

Pouwels et al., October 14, 2021; Braeye

et al., September 15, 2021; Hall et al.,

April 17, 2021). RT–PCR tests,

particularly in the beginning of the

pandemic, were given high priority to

discern who seeking medical care was,

in fact, infected. For instance, the

progression of testing from medical

needs to more of a community

perspective is illustrated in Denmark

(Vrangbaek et al., April 29, 2021).

Denmark, considered one of the gold

standard countries for its

comprehensive testing program, missed

five infections for every one it identified

in the spring of 2020 (Espenhaim et al.,

August 22, 2021). Hansen et al. (March

27, 2021) depended greatly on these first

surge infection definitions to determine

that survivors had protection of 80.5%

effectiveness during the second surge in

Denmark from September through

December, 2020. By only noting RT–

PCR positives from the spring when

testing was limited and highly focused

on health care needs, it seems apparent

that the study excluded many less

severe cases (which are less likely to

result in an effective immune response

against reinfection), leading to results

that may suggest greater protection is

afforded by infection than in actuality.

Even by December of 2020, it appears

Denmark’s gold standard

comprehensive testing approach was

only able to capture roughly half of all

infections. Similar systemic

undercounts have also been determined

to be true in the United States where

approximately three out of four

infections have never been reported

(CDC, July 27, 2021b).

It is important to recognize that RT–

PCR testing was not implemented to

find every infection, but was used

instead to assist in determining when

medical and community interventions

were necessary. Infections without

symptoms or with mild symptoms likely

would not require medical intervention

and, therefore, would likely not be

identified via testing. The absence of

this population that is more vulnerable

to reinfection, in these studies,

undercuts their usefulness in OSHA’s

grave danger analysis, because they may

overestimate the protectiveness of

immunity acquired through infection.

Several other studies in regions less

known for their sampling approach than

Denmark also were heavily dependent

on early, limited pandemic RT–PCR

testing. An Austrian study found a

roughly ten-fold decrease in reinfection

in survivors of reported infections from

February to April 30, 2020 in

comparison with the general public

(Pilz et al., February 11, 2021). The

authors noted that ‘‘infections in the

first wave are likely to have been far

more common than the documented

ones’’ and referred to their results as a

‘‘rough estimate.’’ Researchers at the

Cleveland Clinic also found a reduced

rate of reinfection in those who had a

reported previous infection compared

with those with no prior infection

(13.8% infection rate for those

previously uninfected and 4.9%

infection rate for those previously

infected), but noted that testing was

limited in that the ‘‘Cleveland Clinic did

not test asymptomatic patients unless

they were admitted to hospital or

undergoing a procedure/surgery’’

(Sheehan et al., March 15, 2021). These

criteria for testing create uncertainty in

determining the level of effectiveness

previous infection provides against

SARS–CoV–2 because many individuals

with asymptomatic infections would not

have been tested. Similar issues are also

found in studies on populations in Italy,

Belgium, and the UK (Vitale et al., May

28, 2021; Braeye et al., September 15,

2021; Pouwels et al., October 14, 2021).

To avoid the well-known problems

with RT–PCRs defining previous

infection, other studies have defined

previous infection as testing positive for

antibodies specific for SARS–CoV–2

(Lumley et al., February 11, 2021; AbuRaddad et al., April 28, 2021; Hall et al.,

April 17, 2021). As noted above,

previous infection does not necessarily

result in a seropositive outcome; one

study indicated that nearly a quarter

(24%) of those infected with SARS–

CoV–2 subsequently showed no sign of

an immune response in SARS–CoV–2specific antibody testing (Wei et al., July

5, 2021). Therefore, studies only

considering seropositive individuals are

in essence studying only the individuals

most likely to have protection from

reinfection. Lumley et al. (February 11,

2021) found that those having a

seropositive response had almost an

order of magnitude fewer infections

(e.g., 0.11 adjusted incidence rate ratio).

Likewise, Abu-Raddad et al. (April 28,

App. 22

Federal Register / Vol. 86, No. 212 / Friday, November 5, 2021 / Rules and Regulations

2021) found that seropositive

individuals were reinfected less (0.7%)

during their study period in comparison

to seronegative individuals (3.09%). In

addition to the bias associated with

using antibodies to determine previous

infection, the authors also noted that

there may have been issues with being

able to document cases with mild or no

symptoms.

Hall et al. (April 17, 2021) cast a

wider net by defining previous infection

to include both positive RT–PCR tests

and seropositivity. The researchers

found that those who were considered

previously infected had an 84% lower

risk of infection compared to those who

were unvaccinated with no record of

infection. While the study does attempt

to capture as many previously-infected

individuals as possible, this does not

actually address the weaknesses of each

method. Those with less severe

infections were less likely to have

sought out or been able to get an RT–

PCR test during the first surge, which is

when an overwhelming number of the

previous infections were recorded in

this study (March through May, 2020).

Additionally, the less severe infections

that are most likely underrepresented in

the study appear to be the ones that are

less likely to produce seropositivity.

Shenai et al. (September 21, 2021)

pooled several studies with the above

issues and concluded that immunity

acquired through a previous infection

from SARS–CoV–2 may be as protective

as, or more protective than, the

immunity afforded by vaccination to an

individual without previous infection.

However, authors of several of those

underlying studies used in the analysis

noted that their studies were limited by

not having the capability to fully

accou

This text is long and has been trimmed here. Open the source document for the complete record.