United States Science Advisory EPA-SAB-IHEC-ADV-99-008
Environmental Board April 1 999
AN SAB ADVISORY: Building
Assessment and Survey
Evaluation (BASE) Study Propo
Data Analyses
PREPARED BY THE INTEGRATED HUM
EXPOSURE COMMITTEE (IHEC) OF THE
SCIENCE ADVISORY BOARD (SAB)
-------
April 22, 1999
EPA-SAB-fflEC-ADV-99-008
Honorable Carol M. Browner
Administrator
U.S. Environmental Protection Agency
401 M Street, SW
Washington, DC 20460
Subject: Advisory on the Building Assessment Survey Evaluation (BASE) Study
Proposed Data Analyses
Dear Ms. Browner:
The Integrated Human Exposure Committee (IHEC) of the Science Advisory Board met
on March 9, 1999 in Washington, DC to conduct an advisory on the proposed data analyses for
the Building Assessment Survey Evaluation (BASE) study. The BASE study was designed by the
EPA Office of Radiation and Indoor Air in response to the Agency's responsibility to gather
information and disseminate guidance regarding indoor air quality (IAQ) health risks under Title
IV of the Superfund Amendment Reauthorization Act and in response to the Government
Performance and Results Act (GPRA) Goal 4, Objective 4 which states that "By 2005, 15 million
more Americans will live or work in homes, schools, or office buildings with healthier indoor air
than in 1994." In this advisory, the Committee is providing advice on the analysis of the data
which have been recently collected.
BASE is a cross-sectional multi-year study designed to define key characteristics of IAQ
in 100 public and commercial buildings. The BASE project has four goals:
a) to collect baseline data characterizing public and commercial office buildings,
b) to establish information on important indoor air parameters for policy decisions
and guidance development,
c) to examine the relationships among parameters and between parameters and
occupants' perceptions and symptoms, and
d) to serve as a basis for hypothesis development.
-------
The Committee was charged to respond to the following Charge questions:
a) Are the proposed data analyses the most relevant?;
b) Does the Committee have advice on additional analyses that should be
considered?;
c) How should the analyses be prioritized considering the need to address relevant
scientific issues and the most important programmatic goals identified by the
Agency? In prioritizing the analyses, which analyses are essential given the
Agency's need to address relevant scientific issues and the most important
programmatic goals identified by the Agency?; and
d) Are there similar analyses (that have been conducted on other data sets) that EPA
should use as guidance in its data analysis efforts?
The ultimate goal of the BASE study is to improve public health through improvements in
indoor air quality. To reach this goal, it is necessary to establish baseline information about the
characteristics of indoor air in different buildings, in different locations, and under different
conditions. The frequency distributions of the normative data are the hallmark of this project and
should be extremely useful in supplying relevant and useful yardsticks to practitioners studying
indoor air. Specifically, the BASE data will provide the EPA with building profiles, including
distributions for the concentration of various toxicants indoors, building operational
characteristics and frequencies of various symptom complaints. Therefore, the IHEC found the
BASE survey to be extremely important and commends the Agency personnel who have steered
this complex and carefully executed data collection effort to completion.
Overall, the Committee found the proposed analyses to be the most relevant and extremely
useful in providing significant data on the contributions of indoor environments to human
exposure and adverse health. The Committee found the overall proposed analyses to be useful in
helping the Agency to meet GPRA Goal 4, Objective 4. The analyses of the study parameters can
also be useful in determining good IAQ practices and, subsequently, in helping the EPA to
achieve its GPRA goal of having 5% of the office buildings managed with good IAQ practices by
2005. The IHEC highly encourages the Agency to integrate the BASE project into the Agency's
efforts to analyze cumulative exposure (SAB, 1996) to maximize the impact of BASE on the
overall protection of public health.
-------
The IHEC strongly recommends that the Agency focus on conducting Quality
Assurance/Quality Control on the data followed by evaluating the descriptive statistics, in-depth,
in order to provide critically needed baseline information on the various parameters that have been
monitored in the 100 commercial and public buildings that were included in the study. The
Committee urges the Agency to release the information to the public as soon as the QA/QC and
descriptive statistics analysis are completed. The Committee recommends that the Agency
consider conducting more complex analyses such as testing for associations after the descriptive
statistical data are released. The Committee makes several recommendations for the subsequent
analysis of the data. The IHEC emphasized that the Agency should determine (a priori) the
acceptable power before testing for associations. As an evaluation tool to assist the Agency
further in responding to this question, the IHEC recommends that the EPA consider using a
matrix, assigning point values according to the contribution of each analysis to each GPRA or
program goal. This type of system could make it easier for the Agency to identify high priority
analyses.
The IHEC recommends that the Agency incorporate guidelines regarding the scientific
limitations in using the data. Such guidelines would reduce the likelihood that the data are
misinterpreted and that invalid associations are inferred and would minimize the likelihood of data
dredging (analyzing the data without adequately incorporating the uncertainties associated with
the data), especially given the large number of variables in the study.
In citing a few data sets with analyses that EPA may be able to use as guidance in its data
analysis efforts, the Committee emphasizes the importance of analyzing both the BASE data and
the data from the Office of Research and Development longitudinal study, the Temporal Indoor
Monitoring and Evaluation (TIME) Study. Conducted by the EPA's Office of Research and
Development, the same core parameters from BASE were collected in a smaller number of
buildings in the TIME study. However, unlike the BASE study, samples in the longitudinal study
were taken over different seasons (Fortmann, 1994; EPA, 1999). TIME has the potential, when
coupled with the BASE study, to provide valuable information on the relationships between
cross-sectional and longitudinal studies. Therefore, the Committee strongly encourages the
Agency to review and compare results from both cross-sectional and longitudinal studies
simultaneously to make sure that necessary and comparable analyses are carried out on the data of
both studies. The Committee also encourages the Agency to establish collaborative relationships
with other researchers when developing the strategy to conduct the BASE analyses and when
conducting the BASE analyses.
-------
The Committee appreciates the opportunity to provide advice to the Agency on the BASE
data analyses and looks forward to receiving a written response from the Assistant Administrator
for Air and Radiation (OAR).
Sincerely,
/signed/
Dr. Joan M. Daisey, Chair
Science Advisory Board
/signed/
Dr. Henry A. Anderson, Chair
Integrated Human Exposure Committee
Science Advisory Board
-------
NOTICE
This report has been written as a part of the activities of the Science Advisory Board, a
public advisory group providing extramural scientific information and advice to the Administrator
and other officials of the Environmental Protection Agency. The Board is structured to provide
balanced expert assessment of scientific matters related to problems faced by the Agency. This
report has not been reviewed for approval by the Agency; and hence, the contents of this report
do not necessarily represent the views and policies of the Environmental Protection Agency or
other agencies in the Federal government. Mention of trade names or commercial products does
not constitute a recommendation for use.
-------
ABSTRACT
The Integrated Human Exposure Committee (IHEC) of the Science Advisory Board met
on March 9, 1999 in Washington, DC to conduct an advisory on the proposed data analyses for
the Building Assessment Survey Evaluation (BASE) study. BASE is a cross-sectional multi-year
study designed to define key characteristics of IAQ in 100 public and commercial buildings. The
ultimate goal of the BASE study is to improve public health through improvements in indoor air
quality.
Overall, the Committee found the proposed analyses to be the most relevant and extremely
useful in providing significant data on the contributions of indoor environments to human
exposure and reported symptoms. The BASE data is expected to be normative (typical of public
and commercial buildings) because the buildings used in the study were randomly selected. The
frequency distributions of the normative data are the hallmark of this project and should be
extremely useful in supplying relevant and useful yardsticks to practitioners studying indoor air.
The Committee found the overall proposed analyses to be useful in helping the Agency to meet
GPRA Goal 4, Objective 4, which states that "By 2005, 15 million more Americans will live or
work in homes, schools, or office buildings with healthier indoor air than in 1994." The analyses
of the study parameters can also be useful in determining good IAQ practices and, subsequently,
in helping the EPA to achieve its GPRA goal of having 5% of the office buildings managed with
good IAQ practices by 2005. The IHEC highly encouraged the Agency integrate the BASE
project into the Agency's efforts to analyze cumulative exposure in order to maximize the impact
of BASE on the overall protection of public health.
The IHEC strongly recommended that the Agency focus on conducting Quality
Assurance/Quality Control on the data and then conduct an in-depth evaluation of the descriptive
statistics in order to provide critically needed baseline information on the various parameters that
have been monitored in the 100 commercial and public buildings that were included in the study.
The Committee urged the Agency to release the information to the public as soon as the QA/QC
and descriptive statistics analyses are completed. It was recommended that more complex
analyses, such as testing for associations, be considered after the baseline data are released. The
IHEC provides several recommendations for the subsequent data analyses. The IHEC
emphasized that the Agency should determine (apriori) the acceptable power before testing for
associations.
The IHEC recommended that the Agency incorporate guidelines regarding the scientific
limitations in using the data. Such guidelines would reduce the likelihood that the data are
misinterpreted or that invalid associations are inferred and would minimize the likelihood of data
-------
dredging, especially given the large number of variables in the study. The Committee cited a few
data sets with analyses that EPA may be able to use as guidance in its data analysis efforts and
emphasized the importance of analyzing both the BASE data and the data from the Office of
Research and Development longitudinal study, the Temporal Indoor Monitoring and Evaluation
Study (TIME). The Committee also encouraged the Agency to establish collaborative
relationships with other researchers when developing the strategy to conduct the BASE analyses
and while conducting the BASE analyses.
Keywords: Building Assessment Survey and Evaluation Study (BASE); indoor air; indoor air
quality (IAQ); indoor environments; human exposure; Government Performance
and Results Act (GPRA); and cumulative exposure; Temporal Indoor Monitoring
and Evaluation Study (TIME).
in
-------
U.S. Environmental Protection Agency
Science Advisory Board
Integrated Human Exposure Committee
Roster - IHEC Building Assessment Survey and Evaluation (BASE) Advisory Panel
Chair
Dr. Henry A. Anderson, Chief Medical Officer, Wisconsin Bureau of Public Health,
Madison, WI
Members
Dr. Annette Guiseppi-Elie, Senior Consultant, Dupont Engineering, Corporate Remediation,
Wilmington, DE
Dr. Michael Jayjock, Senior Research Fellow, Rohm and Haas Co., Research Laboratories,
Spring House, PA
Dr. Lovell Jones, Director, Experimental Gynecology-Endocrinology, Department of
Gynecologic Oncology; and Professor of Gynecologic Oncology; MD Anderson Cancer
Center, Houston, TX (Dr. Jones did not participate in this advisory)
Dr. Michael D. Lebowitz, Professor of Medicine and Epidemiology, Arizona Prevention
Center, University of Arizona College of Medicine, Tucson, AZ
Dr. Kai-Shen Liu, Epidemiologist, California Department of Health Services,
Berkeley, CA
Dr. Thomas McKone, Staff Scientist and Professor, School of Public Health,
University of California, Berkeley, CA
(Dr. McKone did not participate in this advisory.)
Dr. Jerome O. Nriagu, Professor, University of Michigan, School of Public Health,
Department of Environmental and Industrial Health, Ann Arbor, MI
Dr. Barbara Petersen, President, Novigen Sciences, Inc., Washington, DC
Dr. David Wallinga, Senior Scientist, Natural Resources Defense Council,
Washington, DC
Dr. Charles Weschler, Senior Scientist, Bell Communication Research,
Red Bank, NJ
IV
-------
Radiation Advisory Committee Liaison
Dr. Jill Lipoti, Assistant Director for Radiation Protection Programs, Division of Environmental
Safety, Health and Analytical Programs, New Jersey Department of Environmental
Protection, Trenton, NJ
Science Advisory Board Staff
Ms. Roslyn Edson, Designated Federal Officer, U.S. Environmental Protection Agency,
Science Advisory Board (1400), 401 M Street, SW, Washington, DC
Ms. Wanda R. Fields, Management Assistant, U.S. Environmental Protection Agency,
Science Advisory Board (1400), 401 M Street, S.W., Washington, DC
-------
TABLE OF CONTENTS
1. EXECUTIVE SUMMARY 1
2. INTRODUCTION 3
2.1 Background 3
2.2 The Review and Charge 5
3. RESPONSE TO CHARGE QUESTIONS 6
3.1 General Findings 6
3.2 Charge Question 1: Relevance of the Proposed Data Analyses 7
3.3 Charge Question 2: Additional Analyses That Should Be Considered 9
3.4 Charge Question 3: Prioritizing the Data Analyses 13
3.5 Charge Question 4: Similar Analyses for Guidance in BASE Data Analyses ... 16
4. SUMMARY OF RECOMMENDATIONS AND CONCLUSIONS 20
REFERENCES CITED R-l
APPENDIX A - ACRONYMS AND ABBREVIATIONS A-l
APPENDIX B - BASE Core Parameters B-l
APPENDIX C - Proposed BASE Analyses C-l
APPENDIX D - GPRA Strategic Goal 4, Objective 4 D-l
VI
-------
1. EXECUTIVE SUMMARY
The Integrated Human Exposure Committee (IHEC) of EPA's Science Advisory Board,
supplemented by a radon expert (a liaison from the SAB Radiation Advisory Committee), met on
March 9, 1999 to review the proposed data analyses for the Building Assessment Survey and
Assessment (BASE) study. The purpose of BASE is to help fill the significant data gap regarding
baseline indoor air quality (IAQ) in public and commercial office buildings in the United States.
This study was conducted by the EPA Office of Radiation and Indoor Air (ORIA).
The Committee addressed the following Charge questions:
a) Are the proposed data analyses the most relevant! - Overall, the Committee found
the proposed analyses to be the most relevant and extremely useful in providing
significant data on the contributions of indoor environments to human exposure
and reported symptoms. The Committee also found the overall proposed analyses
to be useful in helping the Agency to meet GPRA Goal 4, Objective 4, which
states that "By 2005, 15 million more Americans will live or work in homes,
schools, or office buildings with healthier indoor air than in 1994." The analyses
of the study parameters can also be useful in determining good IAQ practices and,
subsequently, in helping the EPA to achieve its GPRA goal of having 5% of the
office buildings managed with good IAQ practices by 2005. The IHEC highly
encouraged the Agency integrate the BASE project into the Agency's efforts to
analyze cumulative exposure (SAB, 1996) in order to maximize the impact of
BASE on the overall protection of public health.
b) Does the Committee have advice on additional analyses that should be
considered? - The Committee recognized the significant effort that the Agency has
undertaken in performing this study. As mentioned in the previous section,
overall, IHEC found that the data analysis being proposed is adequate and
comprehensive. In an effort to facilitate getting the normative information in the
published literature, the Committee hesitated to recommend additional analyses
that may be useful to perform since those analyses are not as critical in the
near-term. The BASE data is expected to be normative (typical of public and
commercial buildings) because the buildings used in the study were randomly
selected. Thus, in performing the current analyses, the Committee recommended a
focus on the QA/QC of the available data that will facilitate publishing quality data
that may be used by the Agency and others in future evaluations. The IHEC
provided advice on the subsequent data analysis and commented on: the
1
-------
aggregation of the data; uncertainty analysis; the classification of chemicals; testing
for associations; confounding factors; psychosocial stress; the building symptoms
index; and the sampling protocol. The Committee encourages the Agency to work
with other organizations to define and perform these additional analyses.
c) How should the analyses be prioritized considering the need to address relevant
scientific issues and the most important programmatic goals identified by the
Agency? In prioritizing the analyses, which analyses are essential given the
Agency's need to address relevant scientific issues and the most important
programmatic goals identified by the Agency? - The IHEC strongly recommended
that the Agency focus on conducting Quality Assurance/Quality Control on the
data followed by evaluating the descriptive statistics, in-depth, in order to provide
critically needed baseline information on the various parameters that have been
monitored in the 100 commercial and public buildings that were included in the
study. The Committee urged the Agency to release the information to the public
as soon as the QA/QC and descriptive statistics analyses are completed. The
Committee recommended that the Agency consider conducting more complex
analyses such as testing for associations after releasing the data on the descriptive
analysis. The IHEC emphasized that the Agency should determine (a priori) the
acceptable power before testing for associations.
d) Are there similar analyses (that have been conducted on other data sets) that EPA
should use as guidance in its data analysis efforts?- There are many similar
analyses that have been conducted on other data sets, essentially too numerous to
list. There are multiple studies in the literature (including proceedings from Indoor
Air & Healthy Buildings) of which the EPA staff in the Office of Air and
Radiation, Office of Radiation and Indoor Air, Indoor Environments Division are
aware. The IHEC cited four data sets with analyses that may be useful as guidance
in EPA's data analysis efforts. Those data sets are from the following studies are:
the California Healthy Building Study, the European Audit Project to Optimize
Indoor Air Quality and Energy Consumption in Office Buildings, the Japanese
Office Building Survey and the TEAM study. The Committee also emphasized the
importance of reviewing and comparing results from both BASE and the Temporal
Indoor Monitoring and Evaluation Study (TIME).
-------
2. INTRODUCTION
2.1 Background
BASE is a cross-sectional, multi-year study designed to define key characteristics of IAQ
in 100 public and commercial buildings. The BASE project has four goals:
a) to collect baseline data characterizing public and commercial office buildings,
b) to establish information on important indoor air parameters for policy decisions
and guidance development (a list of the parameters is included in Appendix B),
c) to examine the relationships among parameters and between parameters and
occupants' perceptions and symptoms, and
d) to serve as a basis for hypothesis development.
Buildings were randomly selected from cities with a population of at least 100,000 in 10
climatic regions. Businesses were also selected randomly using business listings obtained for a
given city. In order for a business to be included in the BASE study, the building
owner/management had to be willing to participate, the building could not be highly publicized as
a "sick" or "problem" building, and the building had to meet the study area criteria. In the BASE
study, the study area criteria was defined as a building which is served by no more than two air
handlers, houses a minimum of 50 employees, and has a maximum of 3 floors.
In the 100 buildings selected for the BASE study, specific environmental measures were
taken, building and heating, ventilation, and air-conditioning (HVAC) characteristics were
defined, and occupant questionnaires were completed. A list of the specific core parameters and
measurements taken are provided in Attachment B. Examples of environmental measures
included temperature, relative humidity and carbon dioxide measurements. Some of the building
characteristics that were recorded included building use, occupancy and smoking policy. The
Indoor Environmental Quality Questionnaire included questions on job characteristics, health and
well-being of the occupant, and work place environmental conditions such as the cleanliness of
the workspace and the lighting conditions (EPA, 1994). The sampling sites and the schedule of
measurements and equipment were also described in the standardized BASE protocol (EPA,
1994). All measurements were taken over the course of a week, from Monday to Friday.
Specifically, some environmental measures such as relative humidity, carbon dioxide and carbon
monoxide were sampled continuously from Tuesday to Thursday. For those samples, the Agency
recorded 5 minute averages. Integrated sampling was used to measure the concentration of
-------
VOCs, particles, and formaldehyde, resulting in 8-9 hour exposure levels. Also, the bioaerosols
were only sampled for 2 minutes and 5 minutes twice on Wednesday.
The EPA has completed field measurements on 100 public and commercial buildings.
Some summary statistics have been completed for VOCs (volatile organic compounds), fungi, and
particulate matter (PM). Some of the data analysis that is underway includes: QA/QC for the data
for the last 14 buildings included in the study; the evaluation of the representativeness of buildings
(compared with the Department of Energy (DOE) survey-DOE, 1995); the evaluation of the
precision and accuracy of the data; and the calculation of ventilation rates of the buildings that
were included in the survey.
The Agency developed a proposed plan for the data analysis that was designed to best
address relevant scientific issues and the most important programmatic goals identified by the
Agency. The Agency's proposed data analysis is included as Appendix C and is listed in order of
priority. Specifically, the data analysis plan was developed based on the Government
Performance and Results Act (GPRA) goals, especially GPRA Goal 4, Objective 4 which is
included in Appendix D. In the proposed BASE analysis plan, EPA has identified and prioritized
six types of analyses. Those analyses include: a) quality assurance/quality control; b)
representativeness of the building samples and weighting determinations; c) ventilation rate
calculations - % outdoor air, air exchange rate and the amount of cubic feet per minute of air per
occupant; d) frequency distribution (for normative data); e) associations; and f) indices and
measures. A description of those analyses is provided below.
a) Quality Assurance/Quality Control - EPA plans to conduct a quality
assurance/quality control analysis first to review the data for errors, needed
changes, or missing data. In this QA/QC procedure, the EPA also plans to
determine the accuracy and precision of the data.
b) Representativeness of Building Samples and Weighting Determinations -
The representativeness of the buildings samples was selected as the second
analysis. The Agency plans to compare regional frequency distributions of
building characteristics such as occupancy, building age, gross floor area, and the
number of floor to those found in the DOE study; to examine any potential biases
resulting from the building selection process; and to develop nationally
representative distributions (DOE, 1995).
c) Ventilation Rate Calculations - % outdoor air, air exchange rate and
CFM/occupant - After conducting analyses on the representativeness of the
-------
buildings, the Agency plans, as the third analysis, to calculate the ventilation rates
to determine the percent of outdoor air, air exchange rate and the cubic feet per
minute (CFM) of air per occupant.
d) Frequency Distribution (normative data) - The fourth analysis, the frequency
distributions (on normative data) would be determined for several of the core
parameters, including concentrations of environmental measurements, symptoms
reported on the occupant questionnaire, building maintenance practices, occupant
demographics, comfort parameters (such as continuous temperature, relative
humidity, sound and light), and sources (such as furnishings, computers, cleaning
materials, and cigarette smoke). (The BASE data is expected to be normative
(typical of public and commercial buildings) because the buildings used in the study
were randomly selected.)
e) Associations - In its fifth analysis, the Agency proposes to test for associations
between the core parameters such as the relationships among symptoms,
demographics, environmental parameters, and building and HVAC characteristics.
f) Developing Indices and Measures - Finally, in its sixth analysis, the EPA proposes
to develop indices and measures including indices for building symptoms, indoor
pollutants, and building system quality.
2.2 The Review and Charge
On March 9, 1999, the Integrated Human Exposure Committee met in Washington, DC to
conduct an advisory on the proposed data analyses for the Building Assessment Survey and
Evaluation (BASE) project. The IHEC was charged to respond to four questions. These
questions and the responses by IHEC are presented in the next section.
-------
3. RESPONSE TO CHARGE QUESTIONS
3.1 General Findings
Overall, the Committee found the proposed analyses to be the most relevant and extremely
useful in providing significant data on the contributions of indoor environments to human
exposure and reported symptoms. The BASE data is expected to be normative (typical of public
and commercial buildings) because the buildings used in the study were randomly selected. The
frequency distributions of the normative data are the hallmark of this project and should be
extremely useful in supplying relevant and useful yardsticks to practitioners studying indoor air.
The Committee also found the overall proposed analyses to be useful in helping the Agency to
meet GPRA Goal 4, Objective 4. The analyses of the study parameters can be useful in
determining good IAQ practices which can ultimately help the EPA to achieve their GPRA goal
of having 5% of the office buildings managed with good IAQ practices by 2005. The IHEC
recommends that the Agency integrate the BASE project into the Agency's efforts to analyze
cumulative exposure.
The IHEC strongly recommends that the Agency first focus on conducting Quality
Assurance/Quality Control on the data and then focus on evaluating the descriptive statistics, in-
depth, in order to provide critically needed baseline information on the various parameters that
have been monitored in the 100 commercial and public buildings that were included in the study
since this is the heart of BASE. The Committee urges the Agency to release the information to
the public as soon as the QA/QC and descriptive statistics analysis are completed. The
Committee recommends that the Agency considers conducting more complex analyses such as
testing for associations only after the data on the descriptive analysis have been released. The
IHEC emphasizes that the Agency should determine (apriori} the acceptable power before
testing for associations.
The IHEC strongly recommends that the Agency, for additional analyses, incorporate
guidelines regarding the scientific limitations in using the data, e.g., to reduce the likelihood that
the data are misinterpreted or that invalid associations are inferred. It will also be necessary to
develop procedures to minimize the likelihood of data dredging, especially given the large number
of variables in the study. This may be particularly important in using the data on symptoms in
conjunction with the data on the building characteristics. The Committee also cites a few data
sets with analyses that EPA may be able to use as guidance in its data analysis efforts, and
emphasizes the importance of analyzing both the BASE data and the data from the Office of
Research and Development longitudinal study, the Temporal Indoor Monitoring and Evaluation
-------
(TIME) study. The Committee also encourages the Agency to establish collaborative
relationships with other researchers when developing the strategy to conduct the BASE analyses
and while conducting the BASE analyses.
3.2 Charge Question 1: Relevance of the Proposed Data Analyses
Are the proposed analyses the most relevant?
The Committee found the proposed analyses to be the most relevant and extremely
useful to those who are concerned with the contribution of indoor environments to human
exposures. IHEC commends the Agency personnel who have steered this complex and
carefully executed effort to completion. The Committee also found the overall proposed
analyses to be useful in helping the Agency to meet GPRA Goal 4, Objective 4. For example,
the associations between symptoms, environmental parameters and building and HVAC
characteristics could be useful in determining good IAQ practices. These practices could then
be publicized to help building owners to improve their IAQ practices and to help EPA to
achieve their GPRA goal of having 5% of the office buildings managed with good IAQ
practices by 2005. The IHEC strongly recommends that the Agency integrate the BASE
project into its cumulative exposure efforts such as the National Human Exposure Assessment
Survey (NHEXAS) project and the Cumulative Exposure Project (SAB, 1996). As an
evaluation tool to assist the Agency further in responding to this question, the IHEC
recommends that the EPA consider using a matrix, assigning point values according to the
contribution of each analysis to each GPRA or program goal. This type of system could make
it easier for the Agency to identify high priority analyses.
The Committee recommends that the Agency modify the prioritization of the data
analyses to begin with Quality Assurance/Quality Control analyses, followed by an analysis of
frequency distributions, and then an analysis of ventilation rate calculations. The frequency
distribution analysis should also include an examination of the shape of the distribution, in
addition to an examination of medians and interquartile ranges. Before proceeding with the
frequency distribution analysis, the Agency should reevaluate non-detect values using more
recent approaches (including simulation/Monte Carlo analyses). The IHEC recommends that
the analyses of the associations and indices be placed at the bottom of the priority list of the
data analyses. More specific advice on the prioritization of the data analyses is provided in the
Committee's response to Charge question 3 in Section 3.4 of this report.
a) Associations - There was some concern that there may be a temptation to
select the types of associations based on curiosity as opposed to significance,
-------
especially given the large number of associations that the Agency could
analyze.
b) Averaging the data - There was also a concern that important information on
outliers and sensitive populations would be lost by averaging the data. Specific
recommendations regarding these concerns are provided in the Committee's
response to Charge question 2 in Section 3.3 in the discussion on the
aggregation of the data.
c) Reliance on subjective reports of symptoms - There was a concern that the
current study depends solely on subjective reports of symptoms, rather than
actual signs of health effects. The occupant questionnaire included several
subjective questions about employee health and well-being. Inclusion of
analyses on both signs and symptoms would have provided a more complete
picture of indoor air quality. Therefore, the Committee recommends that in
future studies on indoor air quality, the Agency consider including more
objective health data such as physical exams and biomarkers. For example, a
short physical exam could include observations for dermatitis or wheezing.
Also, several biomarkers exists for determining the presence of various
chemicals or their metabolites in the body, including several VOCs and
pesticides.
d) Potential sources of indoor air quality lacking adequate data in the BASE
study - The Committee recommends that, in future studies, the Agency include
more information on pesticide exposure, allergens, and cleaning agents in order
to provide a more complete picture of their significance as sources for IAQ
complaints and lAQ-related health effects. For example, the EPA Indoor
Environmental Quality Survey, does not mention pesticides or pesticide
exposures. The monitoring checklist on page D-4 of the supplement does
include pesticides. However, this informal monitoring, apparently performed
only twice during one day of the study, depends on actual observations of
pesticide application which would be unlikely during the hours when the study
was done. The IHEC was also concerned that the BASE protocol does not
include specific information about cleaning agents or specific allergens. Some
cleaning agents are potential sources of indoor exposure to hazardous
chemicals via aersolization and dermal exposure. The building maintenance
workers, who typically clean at night, would, in general, have more detailed
knowledge of these sources than building occupants.
-------
3.3 Charge Question 2: Additional Analyses That Should Be Considered
Does the Committee have advice on additional analyses that should be considered?
The Committee recognizes the significant effort that the Agency has undertaken in
performing this study. As mentioned in the previous section, overall, the data analysis being
proposed is adequate and comprehensive. In an effort to facilitate getting the normative
information in the published literature, the Committee hesitates to recommend additional
analyses that may be useful to perform since those analyses are not critical in the near-term.
Thus, in performing the current analyses, the Committee recommends a focus on the QA/QC
of the available data, which will facilitate publishing quality data that may be used by the
Agency and others in future evaluations. The Committee encourages the Agency to work
with other organizations to define and perform these additional analyses.
The Committee provides several recommendations on some of these subsequent
analyses. However, the contextual framework and the issues to be explored using the data
will have a strong influence on whether additional analysis will be required. Since the
analytical models are still being developed, the Committee recommends that the following
factors be considered in the data analysis.
a) Aggregation of the data - There are concerns that in aggregating data,
valuable information may be lost. The need to provide the normative frequency
distributions is viewed as more important than developing nationally
representative distributions, particularly for example, for any parameters that
are best described on a regional basis (e.g., construction type).
Another concern is that by averaging the data into frequency distributions, one
could lose the power of looking at the outlier data. Thus, in developing
distributions, there should be a discussion of the statistical rigor, or at a
minimum the level of confidence/uncertainty, with which any distribution is
developed. The shape of the distributions should be examined, not just the
medians and interquartiles. Also, the relevance of outliers should be addressed.
However, it may be appropriate to group some types of parameters, including
chemicals, to reduce the total number of analyses. Any effort to do so should
be carefully examined to ensure that the grouping is appropriate as described
below.
-------
b) Analyzing classes of chemicals and individual chemicals - An effort should be
made to classify the chemicals into a small number of categories in order to
facilitate the use of the data in risk analysis. There are several classification
schemes that relate the chemical properties of compounds to the toxicity
potential. However, the IHEC also recommends that the Agency analyzes
individual chemicals in addition to classes of chemicals since both may be
found have a significant effect on indoor air quality. The IHEC recommends
that the Agency start with structure-activity relationships for the health
outcomes that were monitored during the study.
Although the use of "toxicity equivalence units" has inherent flaws, the fact
that people are exposed to a mixture of indoor air contaminants at any given
time should be realized. In September 1998, the IHEC reviewed the
disproportionate impact methodologies that the Agency was proposing to use
to help it respond to complaints filed under Title VI of the Civil Rights Act of
1964 that allege discriminatory effects from the issuance of pollution control
permits by states or other governmental bodies that receive financial assistance
from EPA. Those methodologies included the use of "toxicity equivalence
units." (SAB, 1999).
c) Uncertainty analyses - The uncertainty in data for any collected parameter
should be addressed, as far as possible. The Committee found the proposal for
performing the stated QA/QC analyses to be appropriate. However, the IHEC
recommends that the Agency include the level of confidence in the reported
data for factors (e.g., smoking policy) that cannot be absolutely defined. It
was also recommended the non-detect values be re-evaluated using more
current approaches like Monte Carlo simulation methods.
d) Associations - Although categories of associations are listed, no specific ones
are defined. After providing the normative data, this type of analyses is
considered the most useful. Methods to perform the analyses include simple
pattern analysis using the raw data, development of building profiles, and
multivariate and stratified analyses. When conducting multivariate and
stratified analyses, the Agency should consider using demographic risk factors
and building risk factors as covariates and/or effect modifiers. (Risk indices
could be calculated for individuals and for buildings, as well as the other
indices indicated. These risk indices could then be used in the analyses). This
is especially important in looking at the simplistic symptoms (and indices) in
10
-------
relation to environmental parameters (where other "complaints" are treated as
confounders) and/or building/HVAC characteristics. Before conducting
multivariate analyses, the Agency should first determine if there is colinearity
between any of the core parameters such as between contaminants and
between contaminants and HVAC characteristics. However, care should be
exercised in performing these analyses because of the danger of defining cause
and effect interpretations where none exist.
e) Confounding factors - Confounding factors must be considered in any effort to
establish causal relationships in the data. For instance, in analyzing the data,
the Agency should control for the effects of existing medical conditions when
trying to establish the relationship between exposure and reported symptoms
since the symptoms themselves may not be independent variables. The
Committee expressed some concern about the utility of the occupant
questionnaire given the possibility of a healthy worker bias since sick workers
may not have been captured during the questionnaire process. The study was
designed to address some confounders and cofactors. Therefore, it is
particularly important that the EPA clearly document the limitations in use of
the data in the exposure-response assessment. For future analyses of the
worker well-being, the IHEC stresses the importance of obtaining information
from employees on sick leave.
f) Psychosocialfactors - The psychosocial factors (work stressors, at-home
stressors, ergonomic factors of lighting, office comfort and proximity to
windows) should be included in the assessment of relationships between
exposure and health outcomes (symptoms). The association of psychosocial
stressors with the reported symptoms may be particularly important in the
population under study.
11
-------
g) Building Symptoms Index (BSI) - The Building Symptoms Index is based on
six self-reported symptoms that improve when occupants are away from work.
Those symptoms include: dry, itching or irritated eyes; headache; sore or dry
throat; unusual tiredness, fatigue or drowsiness; stuffy or runny nose, or sinus
congestion; and dry or itchy skin. The IHEC found that the concept of
Building Symptoms Index (BSI) needs further development to increase its
power as a risk assessment tool. The BSI is calculated by averaging, for each
building, the Personal Symptoms Index (PSI) values. Each PSI value is
calculated by adding the number of the six symptoms reported by each
occupant (Brightman, 1996). The assumption that each factor contributes
equally to the "total" symptom may be unrealistic in many instances. The
IHEC recommends that the Agency consider using a weighting that takes into
account the importance of key risk factors (in the indoor environment) and the
severity of the symptoms.
Since the "symptoms" are so non-specific, the development of "indices" that
may be subsequently used in cause and effect relationships should be carefully
evaluated. Any subsequent interpretations using these "indices" should be
based on analyses that relate effects not only to exposure (because of the
presence of the material) but should be compared to "environmental levels"
that would be expected to produce "effects" consistent with symptoms or other
adverse effects.
h) Sampling Schedule - Some environmental measures such as relative humidity,
carbon dioxide and carbon monoxide were sampled continuously from Tuesday
to Thursday. For those samples, the Agency has recorded 5 minute averages.
However, integrated sampling was used to measure the concentration of
VOCs, particles, and formaldehyde, resulting in an 8-9 hour exposure level.
Given the limitations of current technologies for bioaerosol sampling, the
bioaerosols were only sampled for 2 minutes and 5 minutes twice on
Wednesday. The Committee was concerned that measurements without
continuous data, such as the VOCs and particles would be of limited utility in
testing for associations with acute health effects (e.g., asthma which is a
chronic condition that is aggravated by acute attacks).
The Committee offers the following general discussion to assist the EPA in putting
context around additional studies as well as specific recommendations on additional/alternate
12
-------
analyses, should the Agency have the resources to perform these. The Committee
recommends that the Agency consider:
a) analyzing in other indoor environments, including residences and day care
centers;
b) identifying one environmental parameter that is a good indicator of overall
building air quality so that all environmental parameters do not have to be
measured;
c) analyzing indoor air to protect "sensitive" populations as well (the current
analyses are designed to protect the "average" person);
d) determining the correlation between outdoor air quality and indoor air quality;
e) conducting longitudinal studies, e.g., evaluating overlapping buildings in the
ORD study that was carried out over time; and
f) determining whether the buildings with the highest level of concern for a given
environmental parameter are also found to be the same buildings with the
highest levels of concern for another parameter (e.g., determining if buildings
with the highest levels of fungi were also the same buildings with the greatest
amount of water damage and if the buildings with the highest levels of VOCs
are the same buildings with the most reported symptoms).
3.4 Charge Question 3: Prioritizing the Data Analyses
How should the analyses be prioritized considering the need to address relevant
scientific issues and the most important programmatic goals identified by the
Agency? In prioritizing the analyses, which analyses are essential given the Agency's
need to address relevant scientific issues and the most important programmatic goals
identified by the Agency?
The Committee recommends that the Agency first discard incomplete and unreliable
data before it analyzes the data. The Agency's proposed data analysis is included as Appendix
C and is listed in order of priority. The Committee concurs with the Agency's placement of
the QA/QC at the top of the list. However, the Committee's recommendations for the
prioritization of the other data analyses is different from that proposed by the Agency. The
IHEC recommends that the EPA focuses on the outlined frequency distribution first, after
13
-------
conducting the QA/QC. The IHEC recommends analysis on the frequency distribution for the
second analysis since baseline data is the heart of BASE. This analysis provides baseline
information on the various parameters that have been monitored in the 100 commercial and
public buildings that were included in the study. After conducting the analysis on the
frequency distributions, the Agency should then calculate the ventilation rates. The
Committee places the analyses on the representativeness of the buildings and the analyses on
association at the bottom of the list of priorities. If, at some point, regional weighting factors
are developed and employed, the Committee urges the Agency to be quite explicit when
presenting data summaries to indicate whether the data are weighted or unweighted. As noted
in the Committee's response to Charge question 2, the Agency should consider analyzing
indoor air, in the future, to protect "sensitive" populations.
a) Frequency Distributions - The frequency distributions of the normative data
are the hallmark of this project and should be extremely useful in supplying
relevant and useful yardsticks to practitioners studying indoor air. It is the
Committee's understanding that the overall shape and central portions (25th to
75th percentiles) of these distributions are reasonably well defined and will
provide much of the value of this study. The majority of the Committee was
of the opinion that the Agency should direct a lesser, somewhat modest effort
and level of attention to the tails of these distributions. They could be studied
for the lessons they may hold. Indeed, these tails or "outliers" may represent a
fundamentally different population and this could be important information. A
reasonable level of sensitivity analysis could be conducted to provide more
information and insight relative to these tails. There was a minority opinion
that the Agency could lose significant information on subpopulations if it does
not analyze the tails of the distributions.
b) Ventilation Rate - The Committee recommends that the Agency calculate the
ventilation rates rather than have numerous users of the database repeat the
exercise and possibly make mistakes. There may be some issues of relevancy
with regard to the ventilation rate calculations and their association with
occupant symptoms. Clearly the "core" zones receive less fresh air than
"perimeter" zones but the anonymity of the survey data may prevent the
placement of these occupants in either zone type. Also, the critical details of
the treatment of infiltration and the issue of whether the representative area
was in a "core" or "perimeter" area should be sorted out.
14
-------
The Committee anticipates that there will be significant uncertainties in
the ventilation rate calculations and recommends that these uncertainties
should be clearly stated. The issues and problems associated with using CO2 as
a surrogate for ventilation rate are well established and should be a well-
documented caveat in the reports that will describe the BASE data. In those
situations where it is possible, both comparisons between CO2-derived
ventilation rates and the actual air flows measured in buildings and
comparisons between ventilation rates calculated using the CO2 approach
versus the temperature approach would be useful. As an aside for future
consideration, one IHEC Member has suggested the possibility of using an
incidental outdoor air contaminant as a tracer penetrating within the building to
directly measure infiltration. Possible outdoor air contaminants that would
serve this purpose, and not usually found indoors (as in BASE) buildings are:
the aliphatic hydrocarbons and some of the chlorinated hydrocarbons (e.g.,
trichloroethylene). Trace metals have also been used for this purpose.
c) Associations - As mentioned previously, the Committee agreed with the
Agency's placement of the analyses of the associations and indices toward the
bottom of the priority listing of analysis. Clearly, it will be most challenging to
focus on the associations that have the greatest impact for public health; that is,
those areas with the greatest practical significance. They are worthy and
relevant projects but should only be implemented with the best in managerial
and statistical acumen to first assign the level of acceptable power before it
tests for associations.
If reasonably well-documented and adequately powered associations
are established between symptoms, environmental parameters and building and
HVAC characteristics then such calculations could be useful in determining
good IAQ practices. These can then be publicized to help building owners
improve their IAQ practices and help EPA achieve their GPRA goal of having
5% of the office buildings managed with good IAQ practices by 2005. Some
potential associations to include in the analyses are:
1) water damage vs. biological contamination as an indicator of
biologicals;
2) biologicals vs. asthma incidence (and biologicals vs. other symptoms);
15
-------
3) type of filtration vs. PM2 5 and type of filtration vs. PM10; and
4) indoor volatile organic compounds (VOC) levels vs. ventilation rate.
As mentioned above, the appropriate and meaningful analyses of the associations will
be very challenging given the uncertainties and other limitations of the data. The Committee
commends the EPA for sharing the data set and exploring the possible associations with
individuals outside the Agency.
3.5 Charge Question 4: Similar Analyses for Guidance in BASE Data Analyses
Are there similar analyses (that have been conducted on other data sets) that EPA should use
as guidance in its data analysis efforts?
There are many similar analyses that have been conducted on other data sets, essentially too
numerous to list (including several studies described in the proceedings for Indoor Air &
Healthy Buildings). The IHEC acknowledges that the EPA personnel associated with the
BASE study are well aware of most of this literature and cites four studies with analyses that
may be particularly useful as guidance to the EPA for its data analyses. Those studies are: the
California Healthy Building Study, the European Audit Project to Optimize Indoor Air Quality
and Energy Consumption in Office Buildings, the Japanese Office Building Survey and the
TEAM study.
a) The California Healthy Building Study - The California study included
investigations on the relationships between the type of ventilation system, VOC
levels, and office worker symptoms in 12 office buildings in the San Francisco
Bay Area (Mendell, 1996). Using data from this study, Ten Brinke, Daisey
and co-workers from the Lawrence Berkeley National Laboratory tested seven
VOC exposure metrics in terms of their ability to predict complaints among
office workers (Ten Brinke, et.al, 1998). Although some of the metrics were
not statistically significant predictors of symptoms, the analysis of the data
resulted in at least one statistically significant predictor of symptoms. The
BASE study may lend itself to a similar examination.
b) The European Project - The European project compared IAQ parameters
across different countries using 56 office buildings in 9 European countries
(Bluyssen, 1995). The Agency may find some of the analyses used in the
European project useful for analyzing the regional differences of the IAQ core
parameters included in the BASE study. However, in it's recommendations on
16
-------
the prioritization of the data analyses, the IHEC placed analyses on the
representativeness of the buildings at the bottom of the list of priorities. Also,
the Committee urges that the Agency exercise caution in conducting such an
analysis to insure that such regional differences in IAQ measurements and
questionnaire data are not "averaged" out. The IHEC also recommends that
the Agency indicate whether data is weighted or unweighted if at some point,
regional weighted factors are developed and employed.
c) The Japanese Study - The Japanese study included 131 office buildings in four
major cities. The proportion of buildings with indoor environmental
measurements exceeding acceptable levels according to Japanese guidelines
(e.g. 1200 ppm for CO2, 10 ppm for CO, temperature within 17ฐC-28ฐC,
relative humidity within 40-70%, etc.) were assessed (Building Management
Education Foundation, 1988). There are no comparable IAQ guidelines in this
country. However, once the frequency distributions of measured variables
have been determined, it would be easy to assess the proportion of
measurements exceeding certain levels (e.g., outdoor air standards or indoor
levels recommended by other national or international organizations).
d) The EPA Team Study - Some of the analyses used in the latter stages of the
TEAM study should be useful as guidance (and are probably being used for
this purpose). The TEAM studies of volatile organic compounds in several
U.S. cities and the Particle (PTEAM) studies in California were important in
characterizing the normal ranges of residential indoor and personal air
exposures for US populations (EPA, 1996; 1997). In the latter stages of the
TEAM study, the VOC data were fitted to log-normal distributions. The fits
were reasonably good and the approach provided a convenient way to
summarize a large amount of data. A similar approach may be useful in
summarizing selected environmental measurements in the BASE data set.
e) Comparing Data Sets- The buildings included in these European and Asian
studies are, in many ways, quite different from the buildings in the BASE data
set. These differences include design, construction materials, furnishings, and
the types and manner of operation of the HVAC systems. Nonetheless, besides
using the above-mentioned studies for guidance in the developing the analyses
plans for BASE, using the European and Asian studies to compare the results
with those from BASE could also generate some insight into the understanding
of building problems. For example, the comparison of questionnaire data
17
-------
between BASE study and the National Institute for Occupational Safety and
Health (NIOSH) investigations of complaint buildings has revealed some
factors associated with complaints in office buildings (Brightman, 1997). Also,
the data from the various studies can be compared to outdoor air standards or
recommended indoor levels by various national or international organizations.
Any cross-study analysis should include information on indoor sources of
exposure. Such data may provide insight on those parameters that account for
differences in indoor air quality found across studies.
f) Complementary Longitudinal Study - Along with the BASE study, there is a
complementary longitudinal study referred to as the Temporal Indoor
Monitoring and Evaluation (TIME) study. (Fortmann, 1994; EPA, 1999).
Conducted by the EPA's Office of Research and Development, the longitudinal
study collected the same core parameters in a smaller number of buildings.
However, unlike the BASE study, samples in the longitudinal study were taken
over different seasons. A portion of the buildings were included in both studies
to ensure comparability and to provide some information on the relationship
between cross-sectional and longitudinal measurements. The TIME study has
the potential, when coupled with the BASE study, to provide valuable
information on the relationships between cross-sectional and longitudinal
studies. Therefore, the Committee strongly encourages the Agency to review
and compare results from both cross-sectional and longitudinal studies
simultaneously to make sure that necessary and comparable analyses are
carried out on the data of both studies.
g) Study on Water Damaged Buildings - Swedish researchers Jan Sundell and
Carl-Gustof Bornehag have assembled a database that contains studies of
water damaged buildings (Sundell, and Bornehag, 1998). The Committee
recommends that the EPA investigators contact Sundell and Bornehag for
guidance on examining associations between water damage and building
complaints for consideration in the analysis of the BASE data set.
h) Studies from other data sets - Studies from other data sets, not necessarily
building studies, are also relevant in terms of guidance for data analyses. The
EPA should consider reevaluating non-detect values using some of the more
recently described approaches, including simulations and Monte Carlo methods.
The Agency should also identify outliers and decide on a consistent procedure for
treating them. If there is co-linearity between contaminants or between
18
-------
contaminants and HVAC characteristics, these should be considered before all
parameters are included in multivariate analyses.
In examining potential associations, the EPA should also consider some
of the associations that have been reported in other building studies. However,
the EPA should be selective; it should first consider those potential associations
with the largest practical significance. When multivariate and stratified analyses
are performed, demographic and building risk factors might be included as
covariates and/or effect modifiers. (Risk indices could be calculated for
individuals and for buildings, as well as the other indices that have been
mentioned.) This is especially important in looking at the relatively simple
symptom categories in relation to environmental parameters and/or
building/HVAC characteristics.
19
-------
4. SUMMARY OF RECOMMENDATIONS AND CONCLUSIONS
In this report the IHEC has made a number of recommendations for the BASE data
analyses plan:
a) The IHEC concurs with the Agency's placement of Quality Assurance/Quality
Control as a first priority for the data analysis.
b) After Quality Assurance/Quality Control, the Agency should analyze the
descriptive statistics, in-depth, to provide a baseline of information about the
characteristics of indoor air in the 100 commercial and pubic buildings included
in the study.
c) The baseline data containing the descriptive statistics should be released to the
public as soon as the statistical analysis has been completed.
d) The BASE study should be integrated into the Agency's other efforts to analyze
cumulative exposure in order to maximize the impact of BASE on the overall
protection of pubic health.
e) Before testing for associations, the Agency must first assign the level of
acceptable power before it tests for associations.
f) The EPA should incorporate guidelines regarding the scientific limitations in using
the data to reduce the likelihood that the data are misinterpreted and that invalid
associations are inferred and to reduce the likelihood of data dredging, especially
given the large number of variables in the study.
g) The Agency should review and compare results from both the cross-sectional
study (BASE) and the longitudinal study (TIME) simultaneously to make sure
that necessary and comparable analyses are carried out on the data of both
studies.
REFERENCES CITED
R-20
-------
*Brightman, H.S., Womble, S.E., Ronca, E.L., and Girman, J.R. 1996. Baseline
Information on Indoor Air Quality in Large Buildings (BASE '95). Proceedings of
Indoor Air '96. Vol. 3, pp. 1033-1038.
*Brightman, H.S, Womble, S.E., Girman, J.R., Sieber, W.K., McCarthy, J.F., Buck, R.J.,
and Spengler, J.D. 1997. Preliminary Comparison of Questionnaire Data From Two IAQ
Studies: Occupant and Workspace Characteristics of Randomly Selected Buildings and
"Complaint" Buildings. Presented at the Healthy Buildings/IAQ 97 Conference in
Washington, DC, September 1997.
Bluyssen, P.M., Fernandes, EDO, Fanger, P.O., Groes, L., Clausen, G., Roulet, C.A.,
Bernhard, C.A., and Valbjorn, O. 1995. European Audit Project to Optimize Indoor
Air Quality and Energy Consumption in Office Buildings. TNO Building and
Construction Research, March 1995.
Building Management Education Foundation. 1988. Report of a Survey of Building
Hygienic conditions (Japanese). Building Management Education Foundation, March
1988.
DOE, 1995. Commercial Buildings Energy Consumption Survey. U.S. Department of
Energy. Website location: http://www.eia.doe.gov/emeu/cbecs/char95.
March 29,1999.
EPA, 1987, The Total Exposure Assessment Methodology (TEAM) Study: Summary and
Analysis: Volume 1, USEPA Office of Research and Development, Washington, DC,
EPA/600/6-87/002a, June 1987.
EPA, 1994, A Standardized EPA Protocol for Characterizing Indoor Air Quality in Large
Office Buildings, Indoor Air Division, Office of Radiation and Indoor Air, USEPA,
Washington, DC and Atmospheric Research and Exposure Assessment Laboratory,
Office of Modeling, Monitoring Systems, and Quality Assurance, USEPA, Research
Triangle Park, North Carolina. June 1, 1994.
R-l
-------
EPA, 1996. The Particle Team, PTEAM Study: Analysis of the Data, Final Report,
Volume 3, USEPA Office of Research and Development, Washington, DC,
EPA/600/R-95/098, August 1996.
EPA, 1999. TIME Study - Overview and Summary Data for Six Initial Buildings. Website
location: http://www.epa.gov/iaq/base/summary.html
Fortmann, R., Clayton, R., Highsmith, V.R., and Nelson, CJ. 1994. The U.S. EPA/ORD
Large Building Study: Results of the Initial Survey of Randomly Selected GSA
Buildings. Presented at the 1994 Air and Waste Management Association
Symposium. Website location: http://www.epa.gov/iaq/base/awma.html.
March 29, 1999.
*Girman, J.R, Womble, S.E., and Ronca, E.L. 1995. Developing Baseline Information on
Buildings and Indoor Air Quality (BASE '94): Part II - Environmental Pollutant
Measurements and Occupant Perceptions. Presented at Healthy Buildings '95,
September 11-14, 1995.
*Hadwen, G.E., McCarthy, J.F., Womble, S.E., Girman, J.R., and Brightman, H.S. 1997.
Volatile Organic Compound Concentrations in Office Buildings in the Continental
United States. Presented at the Healthy Buildings/IAQ 97 Conference in
Washington, DC, September 1997.
Mendell, M.J., Fisk, W.J., Deddens, J.A., Seavey, W.G., Smith, A.H., Smith, D.F., Daisey,
J.M., and Goldman, L.R. 1996. Elevated Symptom Prevalence Associated with
Ventilation Type in Office Buildings. Epidemiology, Vol. 7, Issue 6, pp. 583-589
SAB, 1996. An SAB Report: The Cumulative Exposure Project, A Review of the Office of
Planning, Policy and Evaluation's Cumulative Exposure Project (Phase 1) by the
Integrated Human Exposure Committee, September 1996, EPA-SAB-fflEC-ADV-96-
004. Washington, DC.
SAB, 1999. An SAB Report: Review on Disproportionate Impact Methodologies, A Review
by the Integrated Human Exposure Committee (IHEC) of the Science Advisory Board
(SAB). EPA-SAB-IHEC-99-007, Washington, DC.
R-2
-------
Sundell, J. and Bornehag, C. 1998. (Personal communication with C. Weschler in 1998)
Ten Brinke, J.T., Selvin, S., Hodson, A.T., Fisk, W. J., Mendell, M.J., Koshland, C.P., and
Daisey, J.M. 1998. Development of New Volatile Organic Compounds (VOC)
Exposure Metrics and Their Relationship to "Sick Building Syndrome" Symptoms.
Indoor Air, Vol. 8, pp. 140-152.
*Womble, S.E., Axelrad, R., Girman, J.R., Thompson, R., and Highsmith, R. 1993. EPA
BASE Program - Collecting Baseline Information on Indoor Air Quality. Proceedings
of Indoor Air '93, Vol. 1, pp. 821-825.
*Womble, S.E., Girman, J.R., Ronca, E.L., Axelrad, R., Brightman, H.S., and McCarthy, J.F.
1995. Developing Baseline Information on Buildings and Indoor Air Quality (BASE
'94): Part I- Study Design, Building Selection, and Building Descriptions. Presented at
Healthy Buildings '95, September 11-14, 1995.
*Womble, S.E., Ronca, E.L., Girman, J.R., and Brightman, H.S. 1996. Developing Baseline
Information on Buildings and Indoor Air Quality (BASE '95) in IAQ 96, Paths to
Better Building Environments, Ed. Kevin Y. Teichman. American Society of Heating,
Refrigerating and Air-Conditioning Engineers, Inc.
Material reviewed by the Committee prior to the meeting.
R-3
-------
APPENDIX A - ACRONYMS AND ABBREVIATIONS
BASE
BSI
CFM
CO
CO2
CV
DOE
GPRA
HVAC
IAQ
fflEC
NHANES
NHEXAS
NIOSH
ORD
PM
PTEAM
QA/QC
SAB
TEAM
TIME
VAV
VOCs
Building Assessment Survey Evaluation
Building Symptoms Index
cubic feet per minute
carbon monoxide
carbon dioxide
constant volume
Department of Energy
Government Performance and Results Act
Heating, Ventilation and Air-Conditioning
Indoor Air Quality
Integrated Human Exposure Committee
National Health and Human Nutrition Examination Survey
National Human Exposure Assessment Survey
National Institute for Occupational Safety and Health
Office of Research and Development
particulate matter
Particle Total Exposure Assessment Methodology
Quality Assurance/Quality Control
Science Advisory Board
Total Exposure Assessment Methodology
Temporal Indoor Monitoring and Evaluation Study
variable air volume
Volatile Organic Compounds
A-l
-------
APPENDIX B - BASE Core Parameters
BASE Core Parameters
Environmental Measures
CONTINUOUS
SAMPLING
Temperature (1)
Relative Humidity (1)
Carbon Dioxide (1)
Sound (indoors only) (1)
Light (indoors only) (1)
Carbon Monoxide (1)
Radon (indoors only)
INTEGRATED
SAMPLING
Particles (3)
S PM10, PM25
VOCs (3)
Formaldehyde (3)
Bioaerosols (4)
S air,
S visible growth
Building Characteristics
Use
. Age(w)
Floor Area
# of Floors
Occupancy
Geographical
Location (w)
Ventilation (w)
S equipment
S operation
S schedule
Outdoor Sources (w)
Special Use Spaces
Smoking Policy (W)/(T)
Water Damage (W)/(T)
Fire Damage (W)/(T)
Renovation (wy(A)
Pest Control (W)/(T)
Cleaning Practices (W)/(T)
HVAC Characteristics
Type
Specifications
S air handler
S exhaust fans
Filtration
Air Cleaning
Systems
Air Washers
Humidification
Systems
Maintenance
Schedule
Inspection Schedule
Supply Air Flow
Rate <6>
Percent Outdoor Air
(6)
Outdoor Air Intake
Rate <6>
Supply Air
S temperature (1)
S relative
humidity (1)
Exhaust Fan Rates (6)
Local Ventilation
Performance (7)
Natural Ventilation
Measurements
(if needed)
Occupant
Questionnaireฎ
Workplace Physical
Information
Health and Well-
being
Workplace
Environmental
Conditions
Job Characteristics
(1) Tuesday AM to Thursday PM (5 min. averages)
(2) Monday PM to Thursday PM
(3) 8-9 hrs on Wednesday
(4) 2 min. and 5 min. sampling twice (AM and PM) on Wednesday
(5) Distributed Thursday AM. Collected until Thursday PM and Friday AM.
(6) Tuesday AM, Wednesday and Thursday AM and PM
(7) Tuesday PM
-------
APPENDIX C - Proposed BASE Analyses
Analysis*
Quality Assurance/Quality Control
Representativeness of Building
Samples/Weighting Determinations
Ventilation Rate Calculations-% outdoor
air, air exchange rate and CFM/occupant
Frequency Distribution (normative data)
Associations
Develop Indices and Measures
Item/Description
- Review for errors, needed changes, or missing data points
- Determine accuracy and precision
- Compare regional frequency distribution of building characteristics
(occupancy, gross floor area, building age, # of floors) to DOE
- Examine any potential biases resulting from building selection
process
- Develop regional weighing factors to use in developing nationally
representative distributions
- Tracer gas (continuous CO2 in outside air, supply, return, and
occupied space)
- Temperature differences (outside air, supply, return, and occupied
space)
- Concentration
- Symptoms
- Comfort parameters (continuous temperature and relative
humidity, sound and light)
- Building and HVAC characteristics
- Building maintenance practices
- Sources (furnishings, special use, copiers, computers, smoking,
cleaning materials, pesticides...)
- Occupant demographics
- Symptoms
- Environmental parameters
- Building and HVAC characteristics
- Building symptom indices
- Indoor pollutant indices
- Building system quality indices
*Prior to the SAB review, the Agency planned to study the items in the first column in the order in which they
appear.
This table was originally prepared by USEPA Office of Air and Radiation, Office of Radiation and Indoor Air, Indoor
Environments Division.
C-l
-------
C-2
-------
APPENDIX D - GPRA Strategic Goal 4, Objective 4
By 2005, 15 million more Americans will live or work in homes, schools, or office buildings
with healthier indoor air than in 1994.
More specifically, to reduce lung cancer, respiratory diseases, and other health problems,
11.5 million more Americans will be exposed to healthier indoor air in their homes by the
mitigation of 700,000 homes with high radon levels, the construction of one million homes
with radon-resistant construction techniques, a reduction in the proportion of households in
which children 6 and under are regularly exposed to smoking from 27% in 1994 to 15%, and
a reduction in the number of children and low-income populations exposed to indoor air
pollutants which worsen or trigger asthma episodes. To reduce health problems in the nearly
10 million children made ill annually from indoor air problems in schools, 15% of the nation's
schools will adopt good IAQ practices consistent with EPA's "Tools for Schools" guidance.
To reduce lAQ-related illness from contaminated air in the workplace, 5% of office buildings
will be managed with good IAQ practices consistent with EPA's "Building Air Quality"
guidance.
D-l
-------
DISTRIBUTION LIST
Administrator
Deputy Administrator
Assistant Administrators
Deputy Assistant Administrator for Science, ORD
Director, Office of Science Policy, ORD
EPA Regional Administrators
EPA Laboratory Directors
EPA Headquarters Library
EPA Regional Libraries
EPA Laboratory Libraries
Library of Congress
National Technical Information Service
Congressional Research Service
------- |