Memorandum

To: Tom Mockler, OSHA

From: Chet Fenton and Carlie McCullough, ERG

Date: March 6, 2012

Subject: Analyses for Subpart V FEA

1.	ERG reviewed a set of electric utility burn accidents obtained from
OSHA’s IMIS database of fatality/catastrophe reports to assess the
potential preventability of fatalities and injuries as the result of
compliance with the flame-resistant clothing requirements of the final
Subpart V rule. This data set contained accident reports for 99
accidents involving arc-related burns associated with electrical utility
type work and covering the period 5/15/1991 to 9/8/1998. These accidents
resulted in 21 fatalities and 94 hospitalized injuries. This time period
is subsumed with the 1984 to 2001 time period encompassed by CONSAD’s
IMIS review that formed the basis for benefits assessment for the
proposed rule. At issue are arc-related burn injuries that result from
equipment failures (e.g., electrical faults resulting from a transformer
failure) or other events that are not otherwise preventable by the
clearance requirements of the final rule.

ERG used preventability criteria supplied by OSHA to judge the
preventability of each fatal or nonfatal injury. These criteria are
summarized in the table below.

Accident characteristics	Reduction of injuries	Rationale

Clothing ignition	Eliminate all burns due to clothing ignition	Rule
requires FR clothing under conditions when clothing can be ignited.

Hydraulic oil	None, except to the extent that clothing ignited	FR
clothing will not substantially reduce burns from hydraulic oil, except
to the extent that clothing ignition is prevented. 

Contact	Reduce due to clothing ignition and for parts of body > arc
distance	FR clothing prevents ignition and will reduce injury
substantially, with reduction to first-degree burns at the arc distance.

Burns to face	Eliminate if burns to other parts of body, except the
hands	If parts of the body other than the hands are burned, the energy
is likely sufficient to require face protection.

Burns to hands	Eliminate	Hand protection is required.

Third-degree burns	Reduce to 10-percent second-degree burn, except for
contact	Third-degree burns will be reduced to first-degree, except to
the extent that the employee is closer than the arc distance.

Second-degree burns	Eliminate	FR clothing will prevent nearly all
second-degree burn.

First-degree burn	None	FR clothing will likely reduce first-degree burns
substantially, but no reduction is taken.



Based on these criteria, ERG found that compliance with the final
Subpart V rule would have potentially prevented 11 of the 21 fatalities,
either by preventing the injury entirely (two cases) or by reducing the
burn severity to a nonfatal level (nine cases). This same review found
that compliance with the final rule would have potentially prevented 34
of the 94 hospitalized injuries and resulted in reduced severity for 29
others. On an annual basis, this review of the IMIS data suggests that
the final rule might potentially prevent an average of 1.14 arc-related
burn fatalities per year and 3.52 arc-related nonfatal hospitalized burn
injuries. Under the proposed rule, only 0.93 of these fatal and 1.66 of
these nonfatal burn injuries would be prevented annually.

An earlier review of IMIS electric utility accident data by CONSAD found
that in comparison to fatality data collected BLS in the Census of Fatal
Occupational Injuries (CFOI) the IMIS fatality incidence underestimated
the incidence of fatalities as reported by BLS by about 41%.  Using this
adjustment factor, ERG estimated that compliance with the final rule
would potentially prevent 1.92 fatal arc-related burn fatalities
annually. This compares to an estimate of 1.57 fatalities prevented
annually under the proposed rule.

ERG’s review also found that 23 of the 99 accidents were associated
with inadvertent direct contact with energized equipment. These
accidents would have been avoided through strict adherence to
requirements of the existing standard. According to the preventability
guidelines shown in the above table, use of FR clothing, however, would
reduce the severity of injury if the fire resulted in ignition of
clothing or burns to parts of the body within the maximum distance of
the arc. Of the injuries in the 23 contact cases, ERG judged only one
would have been would have been prevented (burns to the hand only),
while four more would have resulted in less severe burns had FR clothing
been used. None of the fatalities judged preventable were among these
direct contact-related incidents. The remaining cases involved burns
from equipment failure-related fires and other similar causes. Neither
the existing nor the final standard would prevent these accidents, but
the proper use of FR clothing might prevent or mitigate any resultant
injuries. Of the 76 accidents in this category, ERG judged that the
associated injuries were preventable in 32 cases and would be less
severe in 23 cases if FR clothing had been worn.

2.	ERG also assessed the preventability of alternative versions of the
flame-resistant clothing requirements, including a more stringent
requirement where workers would be required to wear head protection and
face shields whenever working around energized equipment and a less
stringent requirement where only clothing rated at 4 calories would be
required. The review of the same set of IMIS reports indicated that the
more stringent requirement would also prevent 1.92 fatalities (after
adjustment for undercounting), while the less stringent option would
prevent 1.40 fatalities per year.  A summary of the IMIS accident review
is shown below.

Preventability Review of Arc-Related Burn Cases

 – IMIS Accident Reports

	Final Rule	Proposed Rule	Less Stringent Alternative	More Stringent
Alternative

Fatal Injuries



	Prevented	2	0	0	2

Less Severe	9	9	8	9

Not Prevented	9	11	12	9

Insufficient Information	1	1	1	1

Total	21	21	21	21







Nonfatal Injures





Prevented	34	16	9	43

Less Severe	29	29	32	29

Not Prevented	31	49	53	22

Insufficient Information	0	0	0	0

Total	94	94	94	94



3.	ERG also investigated whether the existence of trends in the
incidence of the types of injuries and fatalities potentially prevented
by the final rule might indicate that a revision in the baseline
estimate of the annual number of fatalities and injuries might be
appropriate. Fatality data published by the BLS in the CFOI for 2003
through 2007 fail to show, however, that any such trend exists. The
table below shows the results for the two private sector industries
segments most directly affected by the rule. While these fatality totals
do not encompass all the affected industries, any trends should be
apparent from a review of these industries. Although the fatality totals
vary year to year, no trend is apparent in these data.  

Number of Fatal Injuries by Accident Type and Year

 (Private Sector Only)

Industry	2003	2004	2005	2006	2007

Electric Power Generation, Transmission and Distribution (2211)



22	40	22	39	25

Power and Communication Line and Related Structures Construction (23713)



49	30	33	29	36

Source: BLS, Census of Fatal Occupational Injuries, 2003-2007.

4.	ERG received input on several aspects of the cost model from Charlie
Grose, a Licensed Professional Engineer who has served as a consultant
on the construction and operation of high voltage power systems since
retiring from the New York Power Authority in 1991. 

ERG first consulted Mr. Grose on September 27, 2010 about minimum
approach distances (MAD). MAD becomes relevant for work done on lines
with a line-to-line voltage above 72.5kV. Mr. Grose reported that if a
utility has a digital transient analysis program of their system
available, it can be programmed to develop data at a specific work site
location. The primary cost item is the initial programming to add a line
location. Once the programming is completed, runs for a programmed work
site location take very little time. A typical run would take less than
5 minutes.

After consulting with Mr. Grose, and OSHA, ERG concluded that the
relevant high voltage work is performed almost entirely by the utilities
themselves. Some utilities may need to use portable protective gaps
(PPGs) to ensure that the required MAD can be accommodated. Current
rates of PPG use are low, as they are typically used on compact design
lines which are found in the major population areas. At the present
time, Mr. Grose knows of only 15 utilities using PPGs on a regular
basis. The cost of a PPG varies, as each must be designed and tested to
suit a particular utility’s line configuration. A single PPG and its
grounding equipment might cost $4,000 for use on line-to-line voltage up
to and including 242 kV, or $6,500 for use on line-to-line voltage above
242 kV. If the live work only involves one phase, two PPGs are normally
used one on each side of the work site. If the live work only involves
all phases, six PPGs are normally used, with three on each side of the
work site. For lines at or below 242kV, installing or removing a single
PPG takes about 15 minutes, while installing or removing three PPGs on
each side of the work site takes about 25 minutes. For lines at and
above 262 kV, it could take up to 60 minutes to install or remove a PPG.
Installing or removing PPGs requires a two man crew and an aerial lift.
Mr. Grose also estimated that utilities in dense urban areas use PPGs
about ten percent of the time and that PPGs are normally used on compact
design lines which are found in the major population areas. Since PPGs
are less likely to be used in non-urban areas, the ten percent statistic
is a conservative measure of the extent of PPG use among all utilities
with high voltage transmission lines.

For the purposes of assisting the Agency with the estimation of
compliance costs for the Final Economic Analysis, ERG simplified the
information provided by Mr. Grose and assumed that PPG capital costs
would average $5,000 per unit. ERG also estimated that a typical
affected utility would require 24 PPG units. This would allow the
utility to work simultaneously at four sites, based on the requirement
that six PPGs per site are required for 3-phase work.

In consultation with OSHA, ERG estimated that large utilities would
require, on average, eight hours of engineering time to calculate the
maximum anticipated transient overvoltage (T) on their system, while
small utilities would require, on average, four hours of engineering
time to perform this calculation.

ERG also consulted with Mr. Grose regarding arc hazard assessment on
December 21 and 22, 2010. Mr. Grose felt that most large utilities are
already using software mentioned in the rule, such as ARCPRO, and that
conducting an in-house arc hazard assessment would take an engineer
approximately 20 hours. Smaller utilities would be more likely to use a
consultant who has ARCPRO. It would take a consultant longer to perform
the analysis than an in-house engineer, at a rate of $250 to $300 per
hour. 

5.	The PRIA was based in part on a report prepared by CONSAD, which was
based on 1997 NAICS and SIC codes. The Final Economic Analysis was
updated with the assistance of ERG, using data from the U.S. Census
Bureau’s 2007 Statistics of U.S. Businesses (SUSB). The numbers CONSAD
developed for small, large, and total establishments were based on the
1997 U.S. Economic Census, which used some NAICS classifications which
are now obsolete. In order to be analytically consistent, however, ERG
maintained the older NAICS categories, but updated them using the U.S.
Census Bureau’s 1997 NAICS and 1987 SIC Correspondence Tables, 1997
NAICS to 2002 NAICS Correspondence Tables, and 2002 NAICS to 2007 NAICS
Correspondence Tables to match CONSAD’s NAICS and SIC categories to
2007 NAICS. In many cases a single 1997 NAICS code maps to multiple 2007
NAICS codes.  In these cases ERG averaged the wages, number of
employees, number of establishments, profit rates, or other data for the
2007 NAICS categories to produce a single updated estimate for the 1997
NAICS category. Data for Industrial Power Generators, Major Publicly
Owned Utilities, and Line-Clearance Tree Trimmers are not included in
the SUSB data, and ERG estimated data for these sectors from other
sources (as detailed in the FEA).

6.	Following the algorithm developed by CONSAD, ERG projected the number
of projects per year for a given industry as equal to the number of
crews (i.e., the number of power workers divided by the crew size)
multiplied by the number of projects per crew day (1) multiplied by the
number of work days per year (250). For most industries a crew consists
of 3 power workers at small establishments and 6 power workers at large
establishments. For Ornamental Shrub and Tree Services (SIC 0783),
however a crew consists of 2 power workers at a small establishment and
4 power workers at a large establishment. The following table shows the
estimated number of projects per year for each of the affected
industries.

Estimated Number of Power Projects

Industry	Power Projects per Year

Water, Sewer, and Pipeline Construction (NAICS 234910)	65,078

Power and Communication Transmission Line Construction (NAICS 234920)
1,701,656

Industrial Nonbuilding Structure Construction (NAICS 234930)	78,017

All Other Heavy Construction (NAICS 234990)	410,541

Electrical Contractors (NAICS 235310)	1,247,104

Structural Steel Erection Contractors (NAICS 235910)	21,066

Building Equipment and Other Machine Installation Contractors (NAICS
235950)	19,739

All Other Special Trade Contractors (NAICS 235990)	62,701

Electric Power Generation (NAICS 221110)	1,582,025

Electric Power Transmission, Control, and Distribution (NAICS 221120)
2,689,805

Major Publicly Owned Utilities (NAICS 2211)	360,869

Industrial Power Generators	723,820

Ornamental Shrub and Tree Services (SIC 0783)	990,830



 Some arc-related burn injuries are caused by equipment failures in
which heated fluids are released, burning nearby workers. Burn injuries
from these types of accidents are not prevented by use of the
flame-resistance clothing required by the Subpart V rule. Six of the 99
accidents in the dataset reviewed were of this type.

 CONSAD’s review of fatality data from the BLS Census of Fatal
Occupational Injuries and from the OSHA’s IMIS database concluded that
“656 fatalities occurred due to accidents between 1992 and 2000
involving power generation, transmission, and distribution work
(including line-clearance tree trimming work) that would be covered by
the proposed, revised 1926 and 1910 standards” or 72.9 fatalities per
year [CONSAD, p. 4.23]. A review of comparable fatal accidents contained
in the IMIS database for the period from March 1984 through October 2001
found 759 fatal cases or 43.2 fatalities per year [CONSAD, p. 4.12].
Thus, CONSAD concluded that the IMIS fatalities undercount the true
number of fatalities by 40.7% (43.2 = (1 - 0.407)*72.9).

 ERG again relied on preventability criteria for different types of burn
accidents under the alternative regulatory options as supplied by OSHA
for this analysis.

 The 4-digit NAICS code 2211 encompasses the entire electric utility
industry including generation, transmission, and distribution.

 NAICS 23713is the 2007 (and 2002) NAICS equivalent of the 1997 NAICS
234920 used in the FEA.

 U.S. Census Bureau. 2009a. 1997 NAICS and 1987 SIC Correspondence
Tables. Available at    HYPERLINK
"http://www.census.gov/epcd/www/naicstab.htm" 
http://www.census.gov/epcd/www/naicstab.htm . U.S. Census
Bureau. 2009b. 1997 NAICS to 2002 NAICS Correspondence Tables.
Available at    HYPERLINK
"http://www.census.gov/eos/www/naics/concordances/" 
http://www.census.gov/eos/www/naics/concordances/ 
1997_NAICS_to_2002_NAICS.xls. U.S. Census Bureau. 2009c. 2002 NAICS to
2007 NAICS Correspondence Tables. Available at 
http://www.census.gov/eos/www/naics/concordances/2002_to_2007_NAICS.xls.

  PAGE   \* MERGEFORMAT  1 

