August 18, 2006

FINAL DISPOSITION OF PEER REVIEW COMMENTS FOR 

CHARACTERIZATION OF DUST LEAD LEVELS AFTER RENOVATION, REPAIR, AND
PAINTING ACTIVITIES:

RESPONSES TO PEER REVIEW CHARGE QUESTIONS FROM

EPA REVIEWERS

Prepared By

BATTELLE

505 King Avenue

Columbus, Ohio  43201

EPA Contract No. EP-W-04-021

Work Assignment 2-10

Prepared For

Sineta Wooten, Project Officer

John Schwemberger, Work Assignment Manager

Samuel Brown, Deputy Work Assignment Manager

Program Assessment and Outreach Branch

National Program Chemicals Division

Office of Pollution Prevention and Toxics

U.S. Environmental Protection Agency

1200 Pennsylvania Avenue NW (7404T)

Washington, D.C.  20460

RESPONSES TO PEER REVIEW CHARGE QUESTIONS FROM EPA REVIEWERS

Table of Contents

  TOC \o "1-3" \h \z \t "BHLevel4,4,BHLevel5,5"    HYPERLINK \l
"_Toc138483662"  Table of Contents	  PAGEREF _Toc138483662 \h  1  

  HYPERLINK \l "_Toc138483663"  Responses to Peer Review Charge
Questions	  PAGEREF _Toc138483663 \h  2  

  HYPERLINK \l "_Toc138483664"  General Questions	  PAGEREF
_Toc138483664 \h  2  

  HYPERLINK \l "_Toc138483665"  Question 1:	  PAGEREF _Toc138483665 \h 
2  

  HYPERLINK \l "_Toc138483666"  Question 2:	  PAGEREF _Toc138483666 \h 
5  

  HYPERLINK \l "_Toc138483668"  Question 3:	  PAGEREF _Toc138483668 \h 
6  

  HYPERLINK \l "_Toc138483669"  Question 4:	  PAGEREF _Toc138483669 \h 
7  

  HYPERLINK \l "_Toc138483670"  Specific Questions	  PAGEREF
_Toc138483670 \h  8  

  HYPERLINK \l "_Toc138483671"  Question 1:	  PAGEREF _Toc138483671 \h 
8  

  HYPERLINK \l "_Toc138483672"  Question 2:	  PAGEREF _Toc138483672 \h 
9  

  HYPERLINK \l "_Toc138483673"  Question 3:	  PAGEREF _Toc138483673 \h 
10  

  HYPERLINK \l "_Toc138483674"  Question 4:	  PAGEREF _Toc138483674 \h 
11  

  HYPERLINK \l "_Toc138483675"  Question 5:	  PAGEREF _Toc138483675 \h 
12  

  HYPERLINK \l "_Toc138483676"  Question 6:	  PAGEREF _Toc138483676 \h 
13  

  HYPERLINK \l "_Toc138483677"  Question 7:	  PAGEREF _Toc138483677 \h 
14  

 

Note: Federal (EPA) Reviewers 6 and 7 did not submit any responses to
the peer review charge questions.  Thus, only comments from Federal
Reviewers 8 and 9 are presented.Responses to Peer Review Charge
Questions

General Questions

Question 1:

A risk assessment will be conducted to be used in the benefits analysis
for the Economic Analysis for the Lead Renovation, Repair, and Painting
Program final rule.  This assessment will estimate the risks to children
and adults resulting from renovation activities both with and without
the rule requirements.  EPA anticipates that the data generated by this
study design will add to the data on lead dust levels currently
available in the literature for use in the risk assessment.  

To what extent do you think this study design ensures that appropriate
data are being collected for the risk assessment, and sufficient number
of samples for each parameter to address measurement variability?

Input from Reviewer FR8:

It seems to me that exposure is current exposure – reduction in
exposure due to clean up +increase in exposure due to RRP activity. The
exposure due to clean up should have mainly a short term increase in
exposure and a small long term component. I don’t think the study
provides evidence about anything but the increase in exposure due to RRP
activity. 

The collection under the plastic sheeting does not seem appropriate. 

It seems it should be done before the work is started. 

The question about sufficient samples to address measurement error seems
hard to answer. Some justification in using 0.75, 1.0 and 1.25 for the
sigma due to unexplained error should be given. I think 80% power should
be used and in that case if only two replicates are used the power is
63.5% for work area bare floors as shown in Table A2a. If time
limitations do not permit, then only two replicates have to be used.

In Table 6A it looks like Post Clean Smooth Floors have larger half
widths than Work Area Hazard or Adjacent Room Hazard, and therefore the
text at the bottom of the third page of the Power Study section is
incorrect. 

Replicates are being used to increase the precision of the estimates.
When gutting a kitchen, is the kitchen half gutted, cleaned up, and then
the replicate is taken? I wonder how independent the replicates are. 

Response:

The study is designed to assess the distribution of lead dust during RRP
activities across the four protection/clean up routines.  Samples taken
in the non-work rooms will be used to examine the efficacy of the
containment proposed by the rule and the spread of dust to other areas
of a unit during each of the clean up routines.

We will continue to collect dust under the plastic sheeting, as these
data were requested for the risk assessment to determine how much lead
dust is released when the plastic is rolled up.  We will also be
collecting soils samples both before and after all exterior experiments,
and will be collecting dust wipe samples from dust collection trays
prior to the start of the work to measure background levels on the
exterior.

Sigma values of .75, 1.0 and 1.25 were used to evaluate sensitivity of
model power across a range of possible levels of unexplained variability
in the model. The Multiplicative half-widths of the 95% confidence
interval are wider for the post-cleaning samples than the post-activity
samples, because the power study assumes additional variability being
introduced by each worker’s cleaning habits. 

Input from Reviewer FR9:

A. The number and types of interior samples to be collected for the
housing units (HUs) appear adequate. The design for the child-occupied
facilities (COFs) would be improved by conducting two jobs, rather than
one, from each job category. 

Response:

We agree that performing more jobs at child occupied facilities would
lead to a more complete characterization of lead dust generation;
however, budget constraints prevent us from increasing our sample size
to include more COFs.

B. An estimate of potential lead cross-contamination from adjacent
buildings, roadways, air-entrained soil, etc. needs to be added to the
exterior sample collection design.  Prior to beginning work and setting
up containment, dust collection trays should be placed in the same
locations that will be used during the renovations. The length of
exposure should equal the time that the dust collection trays will be
exposed during the renovation. If no additional exterior samples can be
added to this study, I would recommend the substitution of these
cross-contamination check samples for the soil samples

Response:

The study protocol has been modified to include dust wipe samples from
dust collection trays before the study begins.  Soil samples will still
be collected before and after all exterior experiments.

C. There are no samples being collected during the renovation that will
provide lead concentrations of soil/dust tracked into the house from the
outside or from other parts of the building that might contribute to the
hallway lead concentrations. 

Response:

Before all interior experiments, soil samples will be taken near the
path traveled by the workers in order evaluate the potential track-in
from the exterior.  All areas that will not be sealed off prior to the
experiment will be cleaned and cleared, thus eliminating any track-in
from other parts of the building. In addition, all study personnel are
required to wear booties upon entering the property.   Hallways will be
sampled after the RRP activities have been completed.

Question 2:

Does this study design allow for a quantitative determination of lead
dust levels likely to occur without requiring the plastic coverings and
specialized cleaning in the rule requirements?  Does this study design
allow for a quantitative determination of lead dust levels likely to
occur with requiring the plastic protective coverings and specialized
cleaning in the rule requirements?  Please explain the basis for your
answers.

Input from Reviewer FR8:

**No input was included from reviewer FR8.**

Input from Reviewer FR9:

Yes the study design will provide the quantitative data needed. Air
sampling will provide the airborne lead dust levels during the
renovation as well as during the clean-up activities. The post-work dust
samples will provide lead dust concentrations generated by the various
job types so comparisons can be made across job types,

The four phases will provide data to characterize the residual lead
after the various protection/clean-up routines. The settled dust samples
collected in the observation rooms will provide information on the
spread of the dust to rooms outside of the work area. Additionally, the
hallway dust samples will provide lead concentrations to determine if
lead is being tracked from the work area. However, the hallway lead
levels may be difficult to interpret if no “track in” samples are
collected. There are no samples being collected during the renovation
that will provide lead concentrations of soil/dust tracked into the
house from the outside or from other parts of the building that might
contribute to the hallway lead concentrations. 

Response:

Before all interior experiments, soil samples will be taken near the
path traveled by the workers in order to evaluate the potential track-in
from the exterior.  All areas that will not be sealed off prior to the
experiment will be cleaned and cleared, thus eliminating any track-in
from other parts of the building. In addition, all study personnel are
required to wear booties upon entering the property.   Hallways will be
sampled after the RRP activities have been completed.

Question 3:

Is the study design optimized to provide useful information for the risk
assessment?

Input from Reviewer FR8:

The risk assessment’s use of information from this study seems to be
the estimated increase in lead exposure due to dust levels after
renovation activities. Is there a statement in the risk assessment of
how accurate this data needs to be? There should be.  

Response:

EPA’s risk assessment requires information related to the anticipated
dust-lead hazards generated by specific RRP activities that are being
incorporated into the exposure analysis and economic/benefits analysis,
as well as the likely impact of cleanup and containment.  The study is
collecting as much data as is possible within the time frame and budget
for the study. The power study indicates the study as planned will
provide informative results

Input from Reviewer FR9:

Yes. See responses to question 2 and responses to question 1 B and C.

Response:

Agreed.

Question 4:

Do you believe that the steps taken in the study are sufficient to
protect future occupants of the buildings in this study, the renovation
workers and sample collection workers, and the occupants of any
neighboring buildings from exposure to lead dust that may be generated
by the study?

Input from Reviewer FR8:

The steps taken seem very thorough. I would say yes for future
occupants. The worker protection was addressed in Appendix C. It seems
adequate also. 

Response:

Agreed.

Input from Reviewer FR9:

Very thorough (HUD protocol) cleaning is being performed as well as
clearance testing so future occupants will be protected. If the
guidelines provided in this study for the renovation workers and the
sample collection workers are followed, they will be protected. The
occupants in the neighboring buildings will only be protected from
debris that fall on the ground but may not be protected from airborne
lead dust that can be generated by scraping and sanding.

Response:

Vertical exterior containment will be used when the exterior experiments
are generating large amounts of dust.  Wind speed and direction will be
taken into account when constructing the vertical containment.  To
protect the surrounding areas, work will not be performed when the wind
speed is in excess of 20 mph.  If there are occupied houses or buildings
in the surrounding area, the occupants will be notified of the work to
be done and warned to keep children away from the work stie.

Specific Questions

Question 1:  

Are the houses contemplated for this study design representative of
target housing (see definition of target housing below)?  Are the jobs
representative of the renovation jobs that will be undertaken in target
housing?  

Target housing is defined as any housing constructed prior to 1978,
except housing for the elderly or persons with disabilities (unless any
one or more children age 6 years or under resides or is expected to
reside in such housing for the elderly or persons with disabilities), or
any 0-bedroom dwelling.

Input from Reviewer FR8:

It would be good to choose the houses randomly. However, a small random
sample may not be representative. Perhaps, the available houses of each
type could be listed, and a random sample could be taken of them.  This
would be a step toward a representative sample. However, if the type of
house depends of the amount of lead-based paint, and this has to be
measured, then this approach would seem expensive. However, one could
still sample randomly to determine the order in which the houses are
measured for lead based- paint.  

Response:

For housing units, a number of criteria have to be met for a housing
unit to be a candidate for the study.  Not every job can be completed at
every unit, so complete randomization of job assignment is not feasible.
 Additionally, property owners must agree to the job to be performed on
their property.  The phases of a job will be assigned randomly and will
not always follow the same order.  

Input from Reviewer FR9:

Yes the housing is representative.

Yes, the jobs are representative.

Response:

Agreed.

Question 2:  

The study calls for initially testing with a portable XRF to identify
components with lead-based paint, and for collecting paint chip samples
in the buildings selected for the study to obtain a laboratory value for
those components that likely have lead-based paint as shown from the
portable XRF readings.  Is this approach efficient and accurate for the
purposes of the study?

Input from Reviewer FR8:

I am not a chemist and don’t believe I could helpfully answer this
question.

Response:

Noted.

Input from Reviewer FR9:

Yes.

Response:

Agreed.

Question 3:

How much lead-based paint should be required in the buildings in the
study, considering both interior and exterior jobs and the four
different phases for interior jobs?

Input from Reviewer FR8:

I assume this question is what fraction of the area of the walls of the
house are covered by lead-based paint, and not a question of the
concentration of lead in the paint. It would seem that this question
should be addressed by estimating the exposure associated with fractions
of the wall are covered by lead-based paint. However, no health-effect
threshold has been detected for blood lead. Therefore, I would suggest
selecting levels of lead-based paint in the range that are easiest to
deal with and provide less unexplained error in the fitting of the
statistical models. 

Response:

Based on our power analysis and EPA estimates of dust generation, we
feel that the existing study design is sufficient to produce
statistically significant results.

Input from Reviewer FR9:

A lead range typical of US housing should be used.  Lead concentrations
from 1 mg/cm2 (5,000   SEQ CHAPTER \h \r 1 μg/g) to very high levels
will provide the most useful information and regressions may also
indicate if there may be contamination concerns for renovation
activities conducted on surfaces containing lead at less than 1 mg/cm2
(5,000   SEQ CHAPTER \h \r 1 μg/g).

An additional consideration when pairing the level of lead with the RRP
job and phase is the total lead that will be disturbed as some RRP jobs
will disturb more area than others resulting in potentially more
lead-containing dust.

Response:

Agreed.

Question 4: 

Portions of buildings in this study will be sealed off and the portions
needed for the study will be cleaned prior to the start of the study and
between different phases of the study.  Are these steps adequate to
produce lead levels that can be attributable to the interior renovation,
repair, and painting jobs in the study?  

Input from Reviewer FR8:

Possibly also including the time to complete the job as a covariate in
the statistical model is also necessary.  

Response:

The time to complete the job will be recorded for possible use in the
statistical model.

Input from Reviewer FR9:

Yes with one exception. Track-off mats adjacent to the study area could
prevent tracking in of soil/dust into the house from the outside or from
other parts of the building that might contribute to the hallway lead
concentrations.

Response:

All study personnel will be required to wear protective booties upon
entering the property to help eliminate contributions of exterior
soil/dust to the interior lead concentrations.  

Question 5:

What steps would you recommend to ensure that renovators do not carry
lead dust from one phase of a job to another in this study?

Input from Reviewer FR8:

It seems that avoiding this problem would be difficult. If they dusted
off at the site of the last phase they would increase the amount of dust
there.  The worker decontamination procedure could be followed when a
worker walks between areas of different phases.

Response:

The current protocol calls for all personnel to wear clean, disposable
protective clothing, including a protective suit made of Tyvek or other
suitable material and booties, upon entering a property.  These items
must be removed and disposed of before leaving the property.

Input from Reviewer FR9:

Thorough cleaning of renovation equipment and changing of protective
clothing and booties should prevent lead dust carry-over from one phase
to another.

Response:

Agreed.  The current protocol calls for all personnel to wear clean,
disposable protective clothing, including a protective suit made of
Tyvek or other suitable material and booties, upon entering a property. 
These items must be removed and disposed of before leaving the property.
 All tools and supplies will be cleaned prior to removal from the study
site.

Question 6:  

A set of approaches has been developed to obtain comparable data for
each interior job for each of the four phases, and achieve the study
goals of assessing the impact of plastic sheeting versus no plastic
sheeting, and the impact of specialized cleaning versus normal cleaning.
 Are there any simpler ways to achieve the study goals?  Are there any
approaches regarding phases that should be eliminated from the study
design?

Input from Reviewer FR8:

I would not use a simpler method that is not plausible. I think the
proposed method of statistical analysis should be used. I have no
comment about eliminating approaches regarding the phases. 

Response:

Agreed.

Input from Reviewer FR9:

The interior study design appears adequate except for no measurements to
estimate potential track-in. There are no samples being collected during
the renovation that will provide lead concentrations of soil/dust
tracked into the house from the outside or from other parts of the
building that might contribute to the hallway lead concentrations.

Response:

Before all interior experiments, soil samples will be taken near the
path traveled by the workers in order to evaluate the potential track-in
from the exterior.  All areas that will not be sealed off prior to the
experiment will be cleaned and cleared, thus eliminating any track-in
from other parts of the building.  In addition, all study personnel are
required to wear booties upon entering the property.   Hallways will be
sampled after the RRP activities have been completed.

Question 7: 

We understand that more sampling is always better, but in the interest
of keeping the costs of this study in line, we would like your
assistance in identifying the most valuable samples to collect.  Within
that context, please evaluate the protocol to identify any sampling that
you feel is not being conducted that you would find valuable in
addressing the underlying questions of this study.  Please also
identify samples that you think might provide limited additional
information within the context of the questions being asked.  For
instance:

a.	The current protocol calls for samples to be collected on window
sills after interior jobs.  Is this a good use of our limited number of
samples?

b.	The current protocol calls for no samples to be collected on the
exterior of the structure when an interior job is conducted. What would
be the added value of adding exterior samples when an interior job is
conducted?

c.	The current protocol calls for no samples to be collected on the
interior of the structure when an exterior job is conducted.  What would
be the added value of adding interior samples when an exterior job is
conducted?

Input from Reviewer FR8:

a. No, it is not if the windows are kept closed.

b. The exterior lead concentration might be used as a covariate in the
statistical analysis and reduce unexplained error. I wouldn’t think so
however.

c. The interior lead concentration might be used as a covariate in the
statistical analysis and reduce unexplained error. I wouldn’t think so
however.

Response:

a.  The windows are being kept closed.  For interior experiments,
however, that will not prevent the settling of lead dust on the
windowsill.  These samples will remain in the protocol.

Before all interior experiments, soil samples will be taken near the
path traveled by the workers in order to evaluate the potential track-in
from the exterior.

c.  Noted.

Input from Reviewer FR9:

a. Yes, these are valuable samples for inclusion in this study.

b. Limited additional information might be obtained by exterior dust
sampling during the jobs of replacing a window from the inside, adding a
window, or replacing a window with a door.

c. The interior samples would indicate if the exterior renovation
protocols were protective.

Response:

a. Agreed.

b. In the cases where interior jobs will present potential hazards to
the exterior, such as those examples provided by the reviewer, dust tray
samples will be taken from outside of the work area.  Soil samples will
be collected from the ground nearest where the work is being completed
for all interior experiments before and after the experiment.

c. Noted. 

		

5

6

7

8

T

U

V

W

h

i

j

„

Š

ö

„

…

†

‡

ˆ

‰

Š

‹

Œ

¨

©

ª

«

Ô

Õ

Ö

ð

ñ

ò

ó

ô

õ

ö

÷

ø

4

‚

×

	

 

 

.

/

0

1

2

3

4

5

6

R

S

T

U

`

a

b

|

 |

}

~



€

‚

ƒ

„

 

¡

¢

£

µ

¶

·

Ñ

Ò

Ó

Ô

Õ

Ö

×

Ø

Ù

õ

ö

÷

ø

j¸

j;

j¬

j/

 of 15

August 18, 2006

