Subscriber access provided by CORNELL UNIVERSITY LIBRARY
Article
Occupational health impacts due to exposure to organic chemicals over an entire product life cycle Gaël Kijko, Olivier Jolliet, and Manuele Margni Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b04434 • Publication Date (Web): 30 Oct 2016 Downloaded from http://pubs.acs.org on October 31, 2016
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 27
Environmental Science & Technology
Occupational health impacts due to exposure to organic chemicals over an entire product life cycle Gaël Kijko†*, Olivier Jolliet‡, Manuele Margni,† †
CIRAIG, Polytechnique Montréal, Chemical Engineering Department, 3333 Chemin QueenMary, Suite 310, P.O. Box 6079, Station Centre-ville, Montréal, QC, Canada, H3C 3A7 ‡
Department of Environmental Health Sciences, School of Public Health, University of Michigan, Ann Arbor, MI 48109, USA
1
* Corresponding author: Phone: +1 (514)340-4711 ext. 4122. E-mail:
[email protected] 2
Abstract
3
This article presents an innovative approach to include occupational exposures to organic chemicals in
4
life cycle impact assessment (LCIA) by building on the characterization factors set out in Kijko et al.
5
(2015) to calculate the potential impact of occupational exposure over the entire supply chain of product
6
or service. Based on an economic input-output model and labor and economic data, the total impacts
7
per dollar of production are provided for 430 commodity categories and range from 0.025 to 6.6
8
Disability-Adjusted Life Years (DALY) per million dollar of final economic demand. The approach is
9
applied on a case study assessing human health impacts over the life cycle of a piece of office furniture.
ACS Paragon Plus Environment
Environmental Science & Technology
10
It illustrates how to combine monitoring data collected at the manufacturing facility and averaged sector
11
specific data to model the entire supply chain. This paper makes the inclusion of occupational exposure
12
to chemicals fully compatible with the LCA framework by including the supply chain of a given
13
production process and will help industries focus on the leading causes of human health impacts and
14
prevent impact shifting.
15
Introduction
16
Life cycle assessment (LCA) aims to provide decision makers with meaningful information on the
17
potential environmental impacts of different product systems in a life cycle perspective1 with the aim to
18
avoid impact shifting from one activity to another or from one life cycle stage to another2. Human health
19
(HH) impact is among the indicators considered in life cycle impact assessment (LCIA)2,with a primary
20
focus on assessing the potential impacts associated with chemicals or particulate matter emitted
21
outdoors3-5. Despite recent research efforts to integrate the occupational environment into the LCIA
22
framework
23
there is growing interest in LCIA to better account for the potentially high near-field exposures to
24
chemicals emitted in the direct vicinity of the exposed population13. Efforts to develop methods
25
integrating near-field exposures in LCA have been based on developing LCA-adapted indoor
26
compartmental models to assess the impacts of exposure to chemicals based on indoor emission data
27
during the production phase11,
28
Other methods quantify the dermally-mediated impacts of chemicals in personal care products22, 23.
29
Recently, the USEtox model, a scientific consensus model used for characterizing environmental
30
dispersion and impact of chemicals24, published its 2.0 version including a one-box indoor model for
31
near-field including household and occupational settings.
6-12
, current LCA studies have paid little attention to occupational exposure. Nevertheless,
14-19
and the product use (e.g. during cooking20, for wood products21).
ACS Paragon Plus Environment
Page 2 of 27
Page 3 of 27
Environmental Science & Technology
32
However, intake fraction predictions and emission-based impact modeling in occupational settings
33
remain a challenge to perform over multiple industry sectors since a) workplace chemical emissions
34
during the manufacturing stage of a product are not always known or the data are not publicly available
35
as are the characteristics of the emission location; b) impacts will occur throughout the entire
36
manufacturing supply chain (unlike use phase near-field exposures, which are often clearly delineated
37
for the LCA of a given product) and c) worker populations vary by industry, worksite and workplace
38
practice. In the absence of a way to systematically apply these methods in LCA, another operational
39
method must be developed to cover the entire supply chain.
40
In addition to emissions-based impact modeling in occupational environments, other methods
41
developed to address occupational health in LCA can be classified into two categories. The first one
42
includes methods that use reported fatal and nonfatal occupational injury and illness data to calculate
43
generic characterization factors for each economic sector6-10, 25, 26. These methods are limited by the
44
data: nonfatal injury and illness data are contingent on both declaration and acknowledgement and
45
therefore induce a significant risk of undercounting6. Moreover, these methods do not provide chemical-
46
specific characterization factors, but only aggregated sector-specific industry values for all reported
47
injuries and illnesses. The second category regroups methods combining measured occupational
48
concentrations and the number of hours worked in the environment to directly determine intake, thus
49
overcoming the need to gather sector-specific emissions parameters14. Several articles in the
50
occupational health and safety and LCIA fields use this method, which generally focuses on a single
51
process or industry17, 19, 27-29. Relying on measured occupational concentrations in the workplace, Kijko et
52
al. proposed a method that provides, for each U.S. industrial sector, the chemical-specific impacts per
53
blue collar worker30 hour (BCWH) in the sector31. Blue collar workers are “manual industrial workers”
54
while white collar workers that are “persons whose job is professional or clerical and usually salaried”32.
55
The characterization factors (in DALY/BCWH), or hour-based impact intensities, developed by Kijko et al.
ACS Paragon Plus Environment
Environmental Science & Technology
56
account for 235 organic chemicals and their occupational impacts in a given manufacturing sector.
57
However, supply chain occupational impacts attributable to chemical exposures across the product life
58
cycle are not yet characterized. In this article, occupational impacts due to exposure to organic chemicals
59
occurring at the manufacturing facility producing a given product or commodity are considered to be
60
generated by emissions occurring in the sector itself, while supply chain impacts are generated by the
61
activities along the supply chains upstream the manufacturing sector (i.e. all activities of the product
62
supply chain excluding the manufacturing facility). To calculate these supply chain impacts for a given
63
product, the inventory of BCWH worked in each sector per considered functional unit (FU) must first be
64
determined. Major life cycle inventory databases such as ecoinvent33 are based on physical amounts and
65
do not provide the number of hours worked. Input-Output (IO) databases use currencies as base units34
66
but could be extended to calculate hours worked in an industry supply chain. Social LCA, and working
67
hours have, for example, been considered and integrated into the Social Hotspot Database35. Labor
68
hours (calculated using IO model) have been used to calculate to characterize international labor and
69
wage flows36. This research uses BCWH worked across the entire supply chain to inform both the
70
inventory in LCA and the calculation of impacts per hour worked.
71
This article aims to expand the scope of previous studies, in particular Kijko et al. (2015)31 and make it
72
possible to characterize potential human health impacts attributable to occupational exposures to
73
chemicals across the entire supply chain. More specifically, we aim to: (a) develop an approach to
74
compute BCWH worked throughout the supply chain per functional unit; (b) provide recommendations
75
to account for available measured data and generic data generated with IO for the supply chain; (b)
76
illustrate the application of the approach in an LCA case study that integrates human health impacts
77
from occupational exposure to organic chemicals throughout the entire life cycle and compares them to
78
other sources of human health impacts; and (d) calculate and compare manufacturing facility and supply
ACS Paragon Plus Environment
Page 4 of 27
Page 5 of 27
Environmental Science & Technology
79
chain occupational impacts of organic chemicals per US dollars for each commodity manufactured in the
80
US;.
81
Methodology
82
By convention, in the following equations, characters with an upper arrow represent vectors and bold
83
characters represent matrices. The notation correspond to the diagonal matrix based on vector .
84
Roman characters represent scalars. The $ symbol refers to US dollars. DALY is a unit developed by
85
Murray et al37 that serves to compare human health impacts that would be impractical to compare
86
otherwise (such as cancer and non-cancer impacts). For practical reasons, in this paper we only provide
87
aggregated results (cancer DALY and non-cancer DALY). Characterization factors (CF) for cancer cases,
88
non-cancer cases and aggregated DALY are provided in supporting information (see SI.xlsx).
89
Expanding the framework: worker health impacts over the entire supply chain
90
The starting point for this paper is the framework developed by Kijko et al.31, where human health
91
impact for effect e due to occupational exposure to chemical c in sector s (Decs, in DALY/FU) is calculated
92
as the product of the characterization factor per BCWH worked, for effect e (carcinogens or non-
93
carcinogen) due to exposure to chemical c in sector s ( , in DALY/h in the North American Industry
94
Classification System - NAICS38) and the number of hours worked in this sector s (ℎ , h):
95
= × ℎ
96
The and corresponding uncertainty ranges were calculated based on measured occupational
97
concentrations from the US Department of Labor Occupational Safety and Health Administration (OSHA),
98
which covers 235 chemicals in 430 industrial sectors. The total potential impact in the US manufacturing
99
industry amounts to 775 000 DALY. The global burden of disease (GBD) evaluates many risk factors
(1)
ACS Paragon Plus Environment
Environmental Science & Technology
Page 6 of 27
100
including occupational risks39,
40
101
tetrachloroethylene and trichloroethylene to be coherent with the GBD methodology, the total potential
102
impact in the US manufacturing industry due to exposure to organic chemicals is 300 000 DALY with
103
3 376 DALY attributed to occupational exposure to benzene. The 2005 GBD evaluates the impact of
104
occupational exposure to benzene at 5 762 DALY and occupational exposure to all carcinogens at
105
397 000 DALY39. For a more detailed analysis of the characterization factors for manufacturing facility
106
exposures, refer to Kijko et al. 201531.
107
We now extend this framework to calculate the BCWH worked in each sector over the entire supply
108
chain including the manufacturing sector. Based on Leontief’s work, the IO analysis is widely used in LCIA
109
to model the supply chain41-43 (e.g. CEDA model34) or coupled with process modeling in hybrid methods44,
110
yielding the total commodity production needed (vector , in $/FU, in the IO tables classification)45.
111
= ( − ) ×
112
, as per the IO framework46. ( − ) The vector may be calculated as the product of ( − ) and
113
is composed of columns representing the total commodity production needed to produce one dollar of
114
is the final demand vector per FU at the value of the column commodity ($/$ in IO classification).
115
manufacturing stage ($/FU in IO classification).
116
The total commodity production may be combined with an intervention matrix : a matrix providing the
117
dollar-based work intensity (number of BCWH worked in each industry, NAICS classification) per $ of
118
commodity-specific production values (IO classification) to obtain the total BCWH worked in each NAICS
119
sector.
120
ℎ = ×
. When excluding potential cancer impacts from styrene,
(2)
(3)
ACS Paragon Plus Environment
Page 7 of 27
Environmental Science & Technology
121
, we obtain the impact for effect e By combining equation 3 with equation 1 and 2, for final demand
122
due to the occupational exposure of workers to chemical c over the entire supply chain of commodity,
123
including the manufacturing sector ( , in DALY/FU):
124
× ℎ = × × = × × ( − ) × × = 2
=
125
is the vector of characterization factor for each IO commodity (in DALY/$ or cases/$) Where 2
126
provided in supporting information (see SI.txt for the detailed factors by chemical-commodity couple or
127
SI.xlsx for the aggregated factors by IO sector). A method to calculate the impact in each IO commodity
128
or NAICS sector is described in supporting information (see Equation S1 to S16 in SI.1 and SI.2.).
129
To build the matrix, we started with a binary matrix of elements !"# mapping the BCWH
130
from the NAICS classification (as rows, m sectors) into the IO classification (as columns, n sectors) using
131
conversion data from the US Bureau of Economic Analysis47. Seeing as several NAICS sectors must be
132
mapped to multiple IO sectors due to different definitions of sector boundaries, BCWH were
133
proportionally split to the total production ($, class_IO) of the IO sectors to which they are mapped. We
134
obtained a normalized conversion matrix $%&' of elements !()*+ "# :
135
!()*+ "# = ∑2
136
# is the total production of the IO sector j ($).
137
Then, using the same industry as technology construct used in the CEDA IO model34 we convert the IO
138
sectors into IO commodities:
139
× × :9 8 × $%&' × 9 = 567
)(,-. ×/.
(4)
(5)
134()(,-1 ×/1 )
ACS Paragon Plus Environment
(6)
Environmental Science & Technology
140
the diagonal matrix of the inverse of IO primary industry total output (1/$). is the make With 9
141
matrix with IO sectors as rows and IO commodities as columns. Each element !"# provides the total
142
is the diagonal matrix based on the inverse of total primary output of commodity ; per sector
