This article was downloaded by: [Bibliothèques de l'Université de Montréal]
On: 12 March 2013, At: 06:41
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK
Journal of Occupational and Environmental Hygiene
Publication details, including instructions for authors and subscription information:
http://oeh.tandfonline.com/loi/uoeh20
Occupational Exposure to Silica in Construction
Workers: A Literature-Based Exposure Database
a
Charles Beaudry , Jérôme Lavoué
c
a b
a
a
, Jean-François Sauvé , Denis Bégin , Mounia
d
Senhaji Rhazi , Guy Perrault , Chantal Dion
a e
& Michel Gérin
a f
a
Université of Montréal, Department of Environmental and Occupational Health, Montreal,
Quebec, Canada
b
Université de Montréal Hospital Research Center (CRCHUM), Montreal, Quebec, Canada
c
INRS-Institut Armand-Frappier, Unit of Epidemiology and Biostatistics, Laval, Quebec,
Canada
d
Consultant, Laval, Quebec, Canada
e
Institut de recherche Robert-Sauvé en santé et en sécurité du travail, Montreal, Quebec,
Canada
f
Université de Montréal, Institut de recherche en santé publique de l'Université de Montréal
(IRSPUM), Montreal, Quebec, Canada
Accepted author version posted online: 08 Nov 2012.Version of record first published: 19 Dec
2012.
To cite this article: Charles Beaudry , Jérôme Lavoué , Jean-François Sauvé , Denis Bégin , Mounia Senhaji Rhazi , Guy
Perrault , Chantal Dion & Michel Gérin (2013): Occupational Exposure to Silica in Construction Workers: A Literature-Based
Exposure Database, Journal of Occupational and Environmental Hygiene, 10:2, 71-77
To link to this article: http://dx.doi.org/10.1080/15459624.2012.747399
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://oeh.tandfonline.com/page/terms-and-conditions
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to
anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should
be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims,
proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in
connection with or arising out of the use of this material.
Journal of Occupational and Environmental Hygiene, 10: 71–77
ISSN: 1545-9624 print / 1545-9632 online
c 2013 JOEH, LLC
Copyright DOI: 10.1080/15459624.2012.747399
Occupational Exposure to Silica in Construction Workers:
A Literature-Based Exposure Database
Downloaded by [Bibliothèques de l'Université de Montréal] at 06:41 12 March 2013
Charles Beaudry,1 Jérôme Lavoué,1,2 Jean-François Sauvé,1 Denis Bégin,1
Mounia Senhaji Rhazi,3 Guy Perrault,4 Chantal Dion,1,5 and Michel Gérin1,6
1
Université of Montréal, Department of Environmental and Occupational Health, Montreal, Quebec,
Canada
2
Université de Montréal Hospital Research Center (CRCHUM), Montreal, Quebec, Canada
3
INRS-Institut Armand-Frappier, Unit of Epidemiology and Biostatistics, Laval, Quebec, Canada
4
Consultant, Laval, Quebec, Canada
5
Institut de recherche Robert-Sauvé en santé et en sécurité du travail, Montreal, Quebec, Canada
6
Université de Montréal, Institut de recherche en santé publique de l’Université de Montréal (IRSPUM),
Montreal, Quebec, Canada
We created an exposure database of respirable crystalline
silica levels in the construction industry from the literature.
We extracted silica and dust exposure levels in publications
reporting silica exposure levels or quantitative evaluations of
control effectiveness published in or after 1990. The database
contains 6118 records (2858 of respirable crystalline silica)
extracted from 115 sources, summarizing 11,845 measurements. Four hundred and eighty-eight records represent summarized exposure levels instead of individual values. For these
records, the reported summary parameters were standardized
into a geometric mean and a geometric standard deviation.
Each record is associated with 80 characteristics, including
information on trade, task, materials, tools, sampling strategy, analytical methods, and control measures. Although the
database was constructed in French, 38 essential variables
were standardized and translated into English. The data span
the period 1974–2009, with 92% of the records corresponding
to personal measurements. Thirteen standardized trades and
25 different standardized tasks are associated with at least
five individual silica measurements. Trade-specific respirable
crystalline silica geometric means vary from 0.01 (plumber)
to 0.30 mg/m3 (tunnel construction skilled labor), while tasks
vary from 0.01 (six categories, including sanding and electrical maintenance) to 1.59 mg/m3 (abrasive blasting). Despite
limitations associated with the use of literature data, this
database can be analyzed using meta-analytical and multivariate techniques and currently represents the most important
source of exposure information about silica exposure in the
construction industry. It is available on request to the research
community.
[Supplementary materials are available for this article. Go
to the publisher’s online edition of Journal of Occupational and
Environmental Hygiene for the following free supplementary
resource: appendices containing a list of data sources and
detailed descriptions of each parameter.]
Keywords
exposure database, crystalline silica, construction
Correspondence to: Michel Gérin, Department of Environmental
and Occupational Health, Université de Montréal, C.P. 6128 Succursale Centre-ville, Montreal, Quebec, H3C 3J7 Canada; e-mail:
[email protected].
INTRODUCTION
C
rystalline silica is an important component of many construction materials, such as sand, concrete, brick, block,
mortar, or stone. Construction workers may be exposed to
respirable crystalline silica (RCS) while involved in tasks as
demolition or maintenance of concrete structures, crushing,
drilling, grinding, or sawing concrete or similar material.(1)
Chronic exposure to respirable dusts containing crystalline
silica is known to lead to silicosis, a progressive fibrosis of
the lungs.(2) The International Agency for Research on Cancer
(IARC) has designated crystalline silica as a human carcinogen
(Group 1) when inhaled by workers in its quartz or cristobalite
forms.(3) The National Toxicology Program (NTP) has labeled
this agent a human carcinogen,(4) and ACGIH has labeled it
a suspected human carcinogen.(5)
Construction is a multi-faceted industry. The multiplicity
of tasks within many trades, work force mobility, the transient nature of some construction sites, and a wide range
of determinants of exposure make it difficult to adequately
portray worker exposure. Perhaps because of these particular
challenges, overexposure to RCS has remained frequent in
the building trades.(6) Several reports on RCS exposure in the
construction industry are available in the published literature
(for example, References 6–9), a notable effort being the study
of Flanagan et al.,(7) who assembled 1374 respirable quartz
Journal of Occupational and Environmental Hygiene
R
February 2013
71
Downloaded by [Bibliothèques de l'Université de Montréal] at 06:41 12 March 2013
personal measurements. However, several of these authors
concluded that the data they assembled were too limited for
an accurate characterization, given the complexity of this milieu.(8,9)
Similar to meta-analyses in epidemiology, which aim to
integrate results of several individual studies to obtain a clearer
view of risks,(10) several authors have performed exposure
meta-analyses by gathering available data in published or
database format in order to support an exposure assessment
effort.(11–16) Amassing data from multiple studies, in addition
to augmenting sample size, permits integrating data covering multiple situations and avoiding the pitfall of drawing
conclusions from a limited set of potentially idiosyncratic
observations. Lavoué et al.(14) recently proposed a framework
that allows exposure meta-analysis by reconstructing exposure levels from summary parameters through Monte Carlo
simulation and combining them with single measurements.
Following a request by the Quebec Workers’ Compensation
Board (CSST) for a detailed literature review on silica exposure in the construction sector, we created an occupational
exposure database (OEDB) of RCS exposure in the construction industry based on data from multiple sources. This article
describes the development and structure of this database and
provides descriptive statistics on silica exposure levels derived
from it.
METHODS
Identification of Relevant Literature—Published,
Peer-Reviewed, and Unpublished
We performed searches in Medline, Toxline, Current Contents, HSELINE, NIOSHTIC, EMBASE, Chemical Abstracts,
CISILO, and BIOSIS. They were limited to documents published or produced in or after 1990, as our goal was to describe
recent exposures. In addition to the literature search, we established contact with the National Institute for Occupational
Safety and Health (NIOSH) to obtain their bibliography related
to silica and gain access to documents that were not readily
available through a public literature search. Two additional
sources were solicited: Mary Ellen Flanagan, formerly of the
University of Washington (personal communication, Seattle,
Washington, Sept. 2, 2009) and the Institut de Veille Sanitaire (InVS) of France (personal communication, Corinne
Pilorget, InVS, Lyon, France, April 21, 2009). Mary Ellen
Flanagan is first author of a previous report for the ACGIH
Construction Committee about silica exposure in the construction industry based on measurements from various U.S.
organizations, including two state Occupational Safety and
Health Administrations (OSHA).(7) The InVS assembled a
database of quartz measurements on construction sites, including original data from Paris area worksites, for the purpose of creating a French job exposure matrix.(17) The U.S.
OSHA Integrated Management Information System (IMIS)
database was not considered as a source, as the number of
available descriptors in it was deemed insufficient for our
purposes.(7)
72
After an initial list of publications was created, the following selection process was used to identify those relevant to
our objectives: only publications in English or French with
validated sampling and analytical methods and with either
measurements of construction workers’ RCS exposure or measurements of silica and dust performed to assess control measure efficiency were retained.
Selection and Validation of Database Parameters
Rajan et al.(18) and the Joint ACGIH-AIHA Task Group
on Occupational Exposure Databases(19) each proposed a list
of ancillary information that should accompany quantitative
measurements in occupational exposure databases to enable
their proper interpretation. Unfortunately, measurements reported in the literature have traditionally been accompanied
by only a small subset of the recommended parameters and
limited auxiliary information for the exploration of exposure
determinants.(20) Both the probability of having most of the
cells of the envisioned exposure database show “Not availableN/A” and the scope of such an endeavor drove the authors
to design a conservative list of parameters. A list was built
from a subset of the parameters of Rajan et al.(18) and the
Joint ACGIH-AIHA Task Group(19) from some determinants
of exposure used in the data compilation project from Flanagan
et al.(7) and some others from the InVS database.
While limiting this list to a minimum, the objective was
to gather sufficient information to (1) enable uniform coding
of occupation, tasks, tools, materials, and exposure control
methods; (2) properly characterize the construction site and the
sampling methodology; and (3) make this database available
in a readily interpretable form to the scientific community.
A standard codification of occupations within the construction industry was built based on regulations and trade union
collective agreement definitions in Quebec.
R
Data Gathering, Transformation, and Analysis
Each RCS exposure measurement was usually accompanied by a respirable dust measurement (73% in our database),
as it was part of the analytical process. Other respirable dust
measurements in our database (17%) come from studies of the
efficiency of control measures where the before-after methodology does not require the precision of crystalline silica measurements offered by X-ray or infrared analysis.
Each line of the OEDB (spreadsheet format) relates to either
a single exposure measurement or a group of “n” individual
measurements, the distribution of which is described by a set of
statistical parameters such as a mean and standard deviation
(geometric or arithmetic), median, 95th percentile, or range
(Min – Max). The non-documented quantitative parameters
cells were left blank or filled by the authors based on other
clearly identifiable information in the document, such as sample duration deduced from machine time of operation. The nondocumented qualitative parameters were always coded. When
no information was available from the source for a specific
parameter, the “Not available” classification was used. The
only exception to this rule concerns the “Sampling objective”
Journal of Occupational and Environmental Hygiene
February 2013
Downloaded by [Bibliothèques de l'Université de Montréal] at 06:41 12 March 2013
parameter: for a measurement made by a regulatory agency
such as federal or state OSHAs, the “Regulatory compliance”
code was used by default unless information was available to
modify that choice. Duplicates (same measurements presented
in more than one document or database) were identified and
eliminated.
After initial data entry, if not provided in the document, a geometric mean (GM) and geometric standard deviation (GSD)
were estimated for each set of measurements summarized by
other statistical parameters using equations reported in Lavoué
et al.(14) When the measurements with values lower than the
limit of detection (LOD) of the analytical method were not
assigned a calculated value in the document, the concentration
value equivalent to LOD/2 was assigned by the authors if the
sampling flow rate was available or otherwise the value was
excluded.(21)
Although the purpose of this article is not to report on
any detailed analysis of the database, we performed simple univariate analyses for trades and tasks to illustrate its
potential.(22) First, the data was restricted to breathing zone
RCS samples (including samples identified as quartz, cristobalite, and tridymite). Measurements for which the value of
the “Sampling objective” parameter was either “8-hr TWA”
or “Regulatory compliance” were used for the analysis by
trades. The “8-hr TWA” code is made of measurements done
by non-regulatory organizations where the authors identified
their measurement objective as an assessment to be compared
with an 8-hr time-weighted average (TWA) standard. The
“Regulatory compliance” code corresponds to either federal or
state OSHA measurements. The database contains two sets of
measurement values: (1) the raw data of the source represented
by the “Raw measurement” parameter and/or the statistical
parameters, and (2) the “Measurement value interpreted as
a function of the sampling objective.” The latter was the
parameter used for the univariate analysis by trade. For “8-hr
TWA” data, it can differ from the raw value when hypotheses
on unsampled time were made or when raw samples are part
of a series covering the full shift.
For the task analysis, measurements for which the authors
associated the exposure value to a specific task or tasks were
used; this corresponds to the “Specific Task” code of the
“Sampling objective” parameter. Aggregation of single and
summarized exposure levels was performed by creating (for
each measurement set of size n) n individual values equal to
the GM. The created values were added to individual measurements before stratification by trade or task. This procedure,
while it would affect estimation of variance parameters, does
not bias the estimation of the GM.
RESULTS
Selection of Relevant Publication, Sources of Data
The literature search led to the evaluation of 539 documents,
115 of which contained relevant dust or RCS exposure data. Of
those 115, 44 were peer-reviewed publications, 69 were public
organization reports, and two were occupational exposure
databases, one from Mary Ellen Flanagan and the second from
the InVS. The complete list of data sources is available in
the online Appendix 1. Three important data sources were
excluded from our OEDB for methodological reasons: the
BGIA Report(23) because area measurements were mixed with
breathing zone measurements in an average value, the Draft
Final Report(24) to OSHA, and the Flanagan et al.(7) publication
because all their data tables contained data already compiled
from other sources.
Spreadsheet Content
The spreadsheet contains 80 parameters. They can be
grouped in nine classes of information, which are briefly
described in Table I. The full list is available in the online
Appendix 2.
The final database is available on request from the Institut
de recherche Robert-Sauvé en santé et en sécurité du travail
(IRSST) in spreadsheet format for scientific research purposes.
(Requests should be addressed to Chantal Dion of the IRSST,
[email protected].) In this file, all standardized variables are available in English. The original variables contain
both English and French depending on the descriptions in
the original documents. In collaboration with the IRSST, we
also created an ACCESS user-friendly French version that
allows searches of the exposure data by trade, task, tool,
and material. It can be downloaded from the IRSST Website
(http://www.irsst.qc.ca/-outil-bd-exposition-silice.html).
Descriptive Statistics of the Final Database
The spreadsheet contains 6118 records: 5630 records with
one measurement, and 488 records with summarized parameters for a total of 11,845 single measurements. There are
2767 records containing respirable dust exposure measurements, while 2858 contain RCS exposure measurements. The
remaining records containing measurements are for total dust
(225), total crystalline silica (CS) (145), inhalable dust (54),
inhalable CS (17), thoracic dust (17), thoracic CS (17), and
unspecified contaminant (18). Seventy-five percent of the data
(in terms of records) is associated with measurements from
the United States, 21% from European countries, 3% from the
Québec and Ontario provinces of Canada, and 1% from Asia.
The proportion of missing information in the principal determinants of exposure was lowest (2%) for trade categories
and highest (68%) for the wearing of respiratory protective
equipment during sampling (Table II).
Because some measurements were associated with multiple
tools, tasks, and materials in the original document, their codes
sometimes included titles such as “Multiple tasks - 1 (Sawing
of masonry and other tasks)” or “Multiple tools - 1 (Masonry
saw and . . .).” The proportion of those is relatively small (2,
3, and 12%, respectively, for tool, task, and material).
Regarding the RCS records (n = 2,858), the three major
record contributors were the databases provided by Ms. Flanagan (33%), the InVS (7%), and one NIOSH Health Hazard
Evaluation Report(25) (5%), while the median contribution
provided by each of the 106 sources of RCS data was 0.3%.
Journal of Occupational and Environmental Hygiene
February 2013
73
Downloaded by [Bibliothèques de l'Université de Montréal] at 06:41 12 March 2013
TABLE I. Parameters in the Database Template
ClassA
nB
Definition
1. Classification of the
source of information
4
2. Description of the
trade, task, tool,
material, and silica
content of the
material
3. Uniform coding for
trade, task, tool, and
material
4. Coding for the
construction site
5. Quantitative data
associated with the
exposure
measurements
6. Qualitative data
associated with the
exposure
measurements
12
Information enabling to link the data back to the original document
and its type (e.g., research report, journal article), the quality of the
reported information on the determinants and sampling
methodology
Description of the trade, task, tool, material as written in the
document. Information pertaining to the crystalline silica content of
the material in its different forms (i.e., bulk sample) and the assay
method
7. Coding of control
measures used during
sampling
8. Coding of the
respiratory protection
used during sampling
9. Miscellaneous
AA
4
Standardized codes for trade, task, tool, and material to facilitate data
handling
2
Standardized codes for class (e.g., residential, industrial) and type
(e.g., new construction, demolition) of the construction site
Number of measurements, “Raw measurement” in its individual or
statistical form, “Measurement value interpreted as a function of the
sampling objective,” sample duration and limit of detection of the
analytical method
Information such as “Sampling objective,” contaminant ID (e.g., total
dust, respirable dust, crystalline silica, quartz), sample type (e.g.,
area, personal), sampling train (cyclone, cassette and filter type),
analytical method type (e.g., gravimetric, X-ray, IR . . .), specific
analytical method (e.g., NIOSH 7500 or 7602), year when samples
were taken, and so on
Presence (yes/no) of control measures used during the measurement
and control measures type (exhaust on tool, water spray on tool,
wetting of material . . .)
Presence (yes/no) of respiratory protective equipment (RPE) used
during the measurement, type used and comments on RPE.
20
20
9
3
2
Photographs (if available) and general comments
detailed description of each parameter per class number can be found in online Appendix 2.
of parameters (i.e., columns of the spreadsheet) per class.
BNumber
Personal samples make up 86% of the records, area samples
12%, and source and undefined 2%. The median sampling year
was 1996, and the range was from 1974 to 2009. Standard nylon or aluminum cyclones were used in 92% of the records for
8-hr TWA samples; this number dropped to 57% in task-based
measurements, while, in those, 23% of the records showed
the use of higher-flow cyclones (4.2 L/min). Documented
X-ray analytical methods were used on 79% of the samples,
documented IR analytical methods on 19%, and description of
the method was not available for 2% of the records.
RCS Exposures by Trade and Task
The estimated geometric means for 13 trade categories are
presented in Figure 1, along with the sample size and number
of data sources used for each stratum. The mean exposure
was above the Quebec Occupational Exposure Limit (OEL)
(0.1 mg/m3) for eight trades, with the highest GMs found for
the three “Tunnel construction worker” categories (0.18 to
74
0.30 mg/m3), followed by “Cement mason/Concrete finisher”
(0.28 mg/m3), and “Bricklayer/Stone Mason” (0.17 mg/m3).
The only trade with a GM lower than the ACGIH threshold
limit value (TLV) of 0.025 mg/m3 was “Plumber/Steamfitter,”
R
TABLE II. Proportion of Unreported Factors by
Major Variable
Major Variable
Percent Unreported (%)
Trade
Task
Material
Tool
Control measures
RPE
2
9
25
29
46
68
Journal of Occupational and Environmental Hygiene
February 2013
Downloaded by [Bibliothèques de l'Université de Montréal] at 06:41 12 March 2013
FIGURE 1. Estimated geometric means of RCS exposure by construction trade, based on full-shift exposure data. ∗ Sample size per stratum;
number of publications in parentheses.
although the value of 0.01 mg/m3 should be taken as a gross
approximation, with all the samples reported as below the
LOD and sourced from a single publication. For the measurements used in this analysis, the median sample duration was
242 min (range 7 to 625) in 1458 records. For 64% of the “8-hr
TWA” data, the “raw” and “interpreted” values were the same,
meaning that the sampled period was deemed representative
of the whole shift. Their median sample duration was 430 min
(range 107 to 625).
The restriction of the database to personal task-based RCS
samples yielded 29 task categories associated with five or more
measurements. Figure 2 shows the 11 tasks with the highest
GMs, which ranged from “Pick and shovel work” (0.08 mg/m3)
to “Abrasive blasting” (1.59 mg/m3). Two “Multiple tasks”
categories, one involving masonry cutting (0.70 mg/m3) and
the other concrete grinding (0.56 mg/m3), were the second
and third tasks associated with the highest exposures. There
were six other tasks with a GM above 0.1 mg/m3, including “Breaking/Jack hammering concrete” (0.41 mg/m3) and
“Cutting tunnels” (0.39 mg/m3). The median sample duration for this analysis was 140 min (range 4 to 775) in 956
records.
DISCUSSION
W
ith over 100 data sources totaling 11,845 individual
measurements, we have assembled, to our knowledge,
the largest database of silica/dust exposure in the construction
industry to date, resulting from a comprehensive search of
publicly available datasets. The number of parameters coded
is larger than those provided in the databases of Flanagan and
of the InVS, the two largest contributors in our database, and
the number of records is 2.5 times larger than the sum of their
content. The breadth of situations covered in our database
in terms of occupations, tasks, materials, and tools used will
help occupational health practitioners to better characterize
exposure to RCS in the construction industry.
Because our purpose was not to report on the detailed
analysis of the levels collected in the database but, rather, to
describe its contents, we have presented very limited exposurerelated results. Figures 1 and 2, nevertheless, provide a preliminary portrait of differences among trades and tasks. They
reveal that contrasts between tasks are more important than
between trades, which is explained by the very nature of
construction activities (i.e., different tasks performed in most
trades). Moreover, while not all tasks lead to high exposure
levels, all trades evaluated seem to involve overexposure to
RCS compared with the ACGIH TLV.
The current format of the database, comprising standardized information and summary parameters for “aggregated” results, renders it readily compatible with statistical multivariate
model. These models now represent a state of the art approach
for interpreting complex databases of exposure levels, notably
disentangling the distinct effects of several determinants.(26)
Lavoué et al.(14) proposed an innovative approach that permits applying multivariate methods to mixtures of single and
aggregated measurements.
Despite the sheer size of the database and the increase
in number of tasks and trades covered compared with previous notable efforts, such as the one by Flanagan et al.,(7)
Journal of Occupational and Environmental Hygiene
February 2013
75
Downloaded by [Bibliothèques de l'Université de Montréal] at 06:41 12 March 2013
FIGURE 2. Estimated geometric mean of RCS exposure by task category, based on task-based exposure measurements. ∗ Exposure data
from publications focusing solely on abrasive blasting were excluded from the database. ∗∗ Sample size per stratum; number of publications in
parentheses. (Note: Only the 11 tasks with the highest exposure levels are shown.)
there are several limitations to the representativeness of the
accumulated information on all potential exposure situations
encountered in the construction industry. First, as reported
by several authors,(7,14,27) the quality and quantity of ancillary information are highly variable across sources, varying
from comprehensive (e.g., NIOSH Health Hazard Evaluation
reports) to scant (state OSHA measurements). In our case,
the major determinants (trade and tasks) were documented
in most cases, while information on materials, tools, and
controls were documented ≈50% of the time, and at one
extreme, information on respiratory protective equipment was
rarely documented. This amount of missing data undeniably
constitutes a source of uncertainty for the interpretation of
exposure levels in our database.
The interpretation of sampling objective (TWA vs. task) illustrates a consequence of lack of documentation. Sometimes,
very short measurements were classified as representative of
full shift and were used in the univariate analysis by construction trade (as low as 7 min). As mentioned in the Methods
section, the classification as representative of full shift was
based on our best interpretation of the authors’ intentions,
irrespective of sample duration. The shortest sample durations
come from the regulatory agencies measurements in the Flanagan database where no measurement objective was specified
but with the assumption that regulatory measurements represent full-shift exposure. The availability of sampling duration
for most of the data permits potential users of this OEDB to
select ranges compatible with their objectives, as with any of
the documented variables.
Second, our database is based on the compilation of many
different sources of data, particularly diverse in terms of purpose of study (epidemiology, exposure surveillance), sampling
strategy (representative vs. worst case), and scope (focus on
one occupation in one survey vs. multi-occupation historical
76
data). It is probable that none of the individual sources of
information used in this work provide a representative portrait
of exposure to silica in the entire construction industry, and
we have included in our database variables associated with
potential biases (task based vs. full shift, compliance vs. representative). This multiplicity of purposes and objectives can
be regarded as a major hurdle to valid assessment of exposure.
Nevertheless, we believe that merging all the available studies,
followed by careful analysis based on meta-analytical and
multivariate techniques, represents a way to avoid major biases
and add considerable value to each individual contribution.
In conclusion, we have assembled a comprehensive database of silica exposure data in the construction industry. Despite the numerous sources consulted, we do not expect every
potential exposure situation to be covered. Our work, building
on previous efforts, such as by the ACGIH construction committee, represents a further step to improving knowledge about
occupational exposure to silica in the construction industry.
Hopefully, the compendium of exposure information in this
industry is going to continue growing and the comprehensiveness of the knowledge database will improve.
ACKNOWLEDGMENTS
T
his research was funded by the Institut de recherche
Robert-Sauvé en santé et en sécurité du travail (grant
no. 099–753). Thanks to Alan Echt, Matt Gillen, and Faye
Rice of NIOSH, and Laurène Delabre, Corinne Pilorget, and
Ellen Imbernon of the InVS for the information they shared
with us and for helping us refine our methodology. Special
thanks to Mary Ellen Flanagan for gallantly sharing her mass
of raw data that make up close to 20% of our inventory.
Journal of Occupational and Environmental Hygiene
February 2013
Downloaded by [Bibliothèques de l'Université de Montréal] at 06:41 12 March 2013
REFERENCES
1. Occupational Safety and Health Administration (OSHA): Controlling
Silica Exposures in Construction, OSHA 3362–05. Washington, D.C.:
OSHA, 2009.
2. National Institute for Occupational Safety and Health (NIOSH):
Health Effects of Occupational Exposure to Respirable Crystalline Silica.
DHHS (NIOSH) Pub. No. 2002–129. Cincinnati, Ohio: NIOSH, 2002.
3. International Agency for Research on Cancer (IARC): A Review of
Human Carcinogens. Part C: Arsenic, Metals, Fibres, and Dusts. Lyon,
France: IARC, 2012.
4. U.S. Department of Health and Human Services (DHHS), Public
Health Service (PHS), and the National Toxicology Program (NTP):
Silica, Crystalline (Respirable Size). In Report on Carcinogens. Research
Triangle Park, N.C.: DHHS, PHS, NTP, 2011. pp. 377–379.
5. ACGIH: 2010 TLVs and BEIs. Cincinnati, Ohio: ACGIH, 2010.
6. Rappaport, S.M., M. Goldberg, P. Susi, and R.F. Herrick: Excessive
exposure to silica in the U.S. construction industry. Ann. Occup. Hyg.
47(2):111–122 (2003).
7. Flanagan, M.E., N. Seixas, P. Becker, B. Takacs, and J. Camp: Silica
exposure on construction sites: Rsults of an exposure monitoring data
compilation project. J. Occup. Environ. Hyg. 3:144–152 (2006).
8. Tjoe Nij, E., D. Hohr, P. Borm, et al.: Variability in quartz exposure in
the construction industry: Implications for assessing exposure-response
relations. J. Occup. Environ. Hyg. 1:191–198 (2004).
9. Verma, D.K., L.A. Kurtz, D. Sahai, and M.M. Finkelstein: Current
chemical exposures among Ontario construction workers. Appl. Occup.
Environ. Hyg. 18:1031–1047 (2003).
10. Blettner, M., and P. Schlattmann: Meta-analysis in epidemiology. In
Handbook of Epidemiology, E.W. Ahrens, and I. Pigeot, editors. Berlin:
Springer-Verlag, 2007.
11. Burstyn, I., P. Boffetta, G.A. Burr, et al.: Validity of empirical models of
exposure in asphalt paving. Occup. Environ. Med. 59(9):620–624 (2002).
12. Hein, M.J., M.A. Waters, A.M. Ruder, M.R. Stenzel, A. Blair, and
P.A. Stewart: Statistical modeling of occupational chlorinated solvent
exposures for case-control studies using a literature-based database. Ann.
Occup. Hyg. 54(4):459–472 (2010).
13. Hein, M.J., M.A. Waters, E. van Wijngaarden, J.A. Deddens, and P.A.
Stewart: Issues when modeling benzene, toluene, and xylene exposures
using a literature database. J. Occup. Environ. Hyg. 5:36–47 (2008).
14. Lavoué, J., D. Bégin, C. Beaudry, and M. Gérin: Monte Carlo
simulation to reconstruct formaldehyde exposure levels from summary
parameters reported in the literature. Ann. Occup. Hyg. 51(2):161–172
(2007).
15. Liu, S., S.K. Hammond, and S.M. Rappaport: Statistical modeling
to determine sources of variability in exposures to welding fumes. Ann.
Occup. Hyg. 55(3):305–318 (2011).
16. Park, D., P.A. Stewart, and J.B. Coble: Determinants of exposure to
metalworking fluid aerosols: A literature review and analysis of reported
measurements. Ann. Occup. Hyg. 53(3):271–288 (2009).
17. InVS: “Matrice emplois-expositions aux poussières alvéolaires de
silice cristalline libre” [Job-exposure matrix for respirable crystalline silica]. Available at http://www.invs.sante.fr/surveillance/matgene/
silice cristalline.htm (accessed December 16, 2011).
18. Rajan, B., R. Alesbury, B. Carton, et al.: European proposal for
core information for the storage and exchange of workplace exposure
measurements on chemical agents. Appl. Occup. Environ. Hyg. 12:31–39
(1997).
19. Joint ACGIH-AIHA Task Group on Occupational Exposure
Databases: Special report: Data elements for occupational exposure
databases: Guidelines and recommendations for airborne hazards and
noise. Appl. Occup. Environ. Hyg. 11:1294–1311 (1996).
20. Lavoué, J., C. Beaudry, N. Goyer, G. Perrault, and M. Gérin: Investigation of determinants of past and current exposures to formaldehyde
in the reconstituted wood panel industry in Quebec. Ann. Occup. Hyg.
49(7):587–602 (2005).
21. Hornung, R., and L.D. Reed: Estimation of average concentration in the
presence of nondetectable values. Appl. Occup. Environ. Hyg. 5:46–51
(1990).
22. Beaudry, C., C. Dion, M. Gérin, G. Perrault, D. Bégin, and
J. Lavoué: Bilan et analyse de la littérature sur l’exposition des
travailleurs de la construction à la silice cristalline [Literature review of
construction worker exposure to respirable crystalline silica]. Montréal:
Institut de recherche Robert-Sauvé en santé et en sécurité du travail,
2011.
23. Bagschik, U., M. Böckler, W. Chromy, et al.: BGIA—Report 8/2006e:
Exposure to Quartz at the Workplace. Berlin: German Social Accident
Insurance (DGUV), 2008.
24. Occupational Safety and Health Administration (OSHA), Department of Labor (DOL): “Draft Final Report: Technological Feasibility
Study and Cost and Impact Analysis of the Draft Crystalline Silica
Standard for Construction.” Prepared by Eastern Research Group Inc.,
August 19, 2003.
25. Cornwell, R.J., M.S. Filios, and C. Piacitelli: NIOSH Health Hazard
Evaluation Report: Norfolk Southern Railway Company. HETA 90–3412288. Cincinnati, Ohio: National Institute of Occupational Safety and
Health, 1993.
26. Burdorf, A., and M.V. Tongeren: Commentary: Variability in workplace
exposures and the design of efficient measurement and control strategies.
Ann. Occup. Hyg. 47(2):95–99 (2003).
27. Money, C.D., and S.A. Margary: Improved use of workplace exposure
data in the regulatory risk assessment of chemicals within Europe. Ann.
Occup. Hyg. 46(3):279–285 (2002).
Journal of Occupational and Environmental Hygiene
February 2013
77