Abstract
Increased focus on renovating and maintaining the existing building stock is an integral part of the circular economy, however this might pose challenges to workers health. The aim of this study was to assess the renovation workers’ exposure to inhalable dust, thoracic dust, respirable dust, and respirable crystalline silica (RCS). Personal aerosol samples were collected as full shift samples from 92 workers to a total of 407 samples. Fourteen locations around Oslo, Norway was visited for multiple days with repeated measurements of the same individual. Particulate matter from 3 aerosol fractions, respirable, thoracic, and inhalable, were analyzed gravimetrically, and the respirable fraction was analyzed for RCS by NIOSH 7500 method for X-ray diffraction (XRD) with low temperature plasma ashing sample preparation. The total measured concentrations of respirable dust (n = 192) had a geometric mean (GM) of 0.88 mg/m3, RCS concentrations (n = 182) had a GM of 0.040 mg/m3, thoracic dust (n = 131) had GM 2.4 mg/m3, and inhalable dust (n = 84) had a GM of 8.5 mg/m3. The maximum measured concentrations were 29 mg/m3, 3.2 mg/m3, 65 mg/m3, and 163 mg/m3, respectively. Workdays involving tasks such as mechanical demolition and clearing out demolished materials led to the highest exposure levels of both dust and RCS. However, other workers at the renovation sites were indirectly exposed to a considerable amount of RCS. This study revealed substantial exposure to both RCS and dust during renovation, and protective measures are warranted to reduce exposure levels in the industry.
Keywords: quartz, respirable silica, respirable dust, construction industry, exposure assesment, particulate matter
What’s Important About This Paper?
This study determined exposure to dust and respirable crystalline silica in the renovation industry. Substantial personal exposure levels were uncovered for demolition workers, as well as other workers at indoor renovation sites. Preventative measures are needed to ensure workers health, as number of renovation projects increases with a sustainable construction focus.
Introduction
Renovating and maintaining the current building stock is crucial to lowering energy and material consumption, according to the 2020 Circularity Gap Report (Circle Economy 2020). In 2021, 35% of EU buildings were above 50 yr old and 75% were energy inefficient (Filippiduo and Jiménez Navarro, 2019). Furthermore, 40% of total material resources worldwide consumed by the construction industry (Fufa et al. 2020). However, renovation work can pose increased health risk in comparison to new construction, as renovation work has been found to yield higher exposure to respirable crystalline silica (RCS) (Sauvé et al. 2013).
Exposure to RCS becomes a concern particularly when renovating concrete or masonry buildings. Crystalline silica is present in most igneous and metamorphic rocks, and sediments. Stone can be used directly as building blocks or crushed into aggregates for composite materials such as concrete, bricks, and mortar(Callister and Rethwish, 2007). Depending on the source of the added aggregates, concrete and mortar may have a crystalline silica content ranging from 12% to 70% (w/w) (Lumens and Spee 2001; HSE 2022). Demolition of building materials may generate fine dust. Occupational exposure to RCS is associated with adverse health effects such as silicosis, chronic obstructive pulmonary disease, and autoimmune disorders such as rheumatoid arthritis as well as being a classified carcinogen by The International Agency for Research on Cancer (IARC) (Hnizdo and Vallyathan 2003; IARC 2012).
Construction workers and carpenters have among the highest self-reported exposure to mineral dust in the Norwegian workforce (Bakke et al. 2021). Work operations such as milling, grinding, sawing, demolition, and cleaning up debris from cement and brick are associated with a considerable dust and RCS generation. Exposure to RCS has often been found to exceed 0.1 mg/m3 during these tasks, according to a report from the Swedish Environmental Research Institute (Antonsson and Sahlberg 2019). Previous studies from other countries have shown that the current Norwegian occupational exposure limit (OEL) for RCS of 0.05 mg/m3 was exceeded more often than the corresponding OEL for respirable dust in this industry (5 mg/m3) (Tjoe Nij et al. 2004; Flanagan et al. 2006; Kirkeskov et al. 2016; Grahn and Lewné 2017). A study comparing manual demolition, mechanical demolition, and waste management, found that the exposure to RCS was highest when manually demolishing with a geometric mean (GM) of 0.69 mg/m3, whereas respirable dust levels were highest when removing waste (5.06 mg/m3 GM). Mechanical demolition had RCS levels from 0.02 to 0.45 mg/m3 with a GM at 0.09 mg/m3(Kirkeskov et al. 2016).
The objective of this study is to determine renovation workers’ exposure to RCS, respirable, thoracic, and inhalable dust, as well as the influence of work tasks, materials, and direct or indirect exposure. The study is not aimed at complete demolition of buildings, but demolition done in conjunction with renovation, e.g. dismantling walls and floors with handheld tools. Particulate matter has been collected through personal sampling and analyzed by means of gravimetry, while X-ray diffraction (XRD) was used for the determination of RCS.
Method
Aerosol sampling
Personal inhalable, thoracic, and respirable samples were collected from the breathing zone of workers, within 30 cm radius from nose and mouth. Each worker carried one personal sampler for each size fraction throughout a workday (202 to 530 min). The respirable and thoracic cyclones were carried in parallel on the right side of the chest, while the inhalable dust sampler was placed on the left side. However, it was not always possible for the workers to carry multiple pumps and samplers. The respirable dust fraction was then prioritized.
Respirable dust was collected onto 37 mm 5.0 µm pore-size polyvinyl chloride (PVC) filters (Merck Millipore Corp., Billerica, MA, USA) placed inside Higgings-Dewell cyclones (JS Holdings, Stevenage, UK) held at an air flow of 2.2 L/min. Thoracic dust was collected onto 37 mm 5.0 µm pore-size PVC filters mounted in a 37-mm plastic cassette (Merck Millipore Corporation, Corp., Billerica, MA, USA) placed in a thoracic cyclone sampler (Mesa Labs, Lakewood, CO, USA). The airflow for thoracic sampling was 1.6 L/min. Inhalable dust was collected onto 25 mm 5 µm pore-size PVC filters mounted in a stainless-steel cassette within a plastic IOM multi dust sampler (Institute of Occupational Medicine, Edinburgh UK, manufactured by SKC Ltd) operated at an airflow of 2.0 L/min. Casella Apex2 (Casella, Bedford, UK) pumps were used and the airflow were checked with a calibrated rotameter (Aalborg Instruments & Controls, New York, USA)
Study design
A total of 92 workers at 14 different renovation sites were monitored for exposure of dust and RCS. Each location was visited for multiple shifts and repeated measurements made from the same individual when possible. In this study, full shift samples were collected to assess the total exposure of dust and RCS for workers in the indoor renovation. The number of samples per individual varied from 1 to 5 full shift samples. Information was collected through a personal sampling form on the different work tasks, exposure scenarios, exposure management, respiratory protection, and material demolished. The measurements were always performed outside of the mask.
The measurements were divided into primary, secondary, and indirect exposures scenarios based on whether the dust was produced or resuspended by the worker (Table 1). This was defined by the dominating work tasks performed. Primary exposure scenarios included tasks like mechanical demolition, cutting, milling, and core drilling of concrete and brick, where the task is a primary source of exposure. Secondary exposure scenarios included removing demolished material and cleaning, and thereby resuspending dust. Indirect exposure scenarios covered workers that were conducting tasks unrelated to demolition but were exposed to dust because they were working in the same room or building as the demolition. In the case of workers preforming different tasks throughout the day, the measurement was placed in the scenario which took most of the time (>4 h).
Table 1.
Descriptions of exposure scenarios and work tasks.
| Exposure scenario | Work task | Description |
|---|---|---|
| Primary—Workers directly producing and suspending RCS through mechanical demolition. | Core drilling | Drilling or cutting out holes in walls or floors. Water cooled |
| Milling and grinding | Strips of the top layer of concrete by a rotating cutting tool or an abrasive tool. | |
| Demolition with jackhammer/breaker | Demolition of walls or floors by handheld or robotic breakers | |
| Cutting | Cutting concrete into pieces or cutting slits in the concrete. Water cooled | |
| Secondary—Workers resuspending dust | Clearing out | Clearing out demolished materials. |
| Indirect—Workers performing other tasks, who do not produce or resuspend dust themselves. | Soft strip demolition | Removal of plaster, wood, electrical installations, floor, and wall coverings. |
| Bas/Foreman | In charge of work crew, often ambulating tasks. | |
| Bricklayer | Bricklaying and plastering | |
| Formworkers | Building and removing metal and wood framework to cast concrete in | |
| Construction worker | Construction tasks like building in wood and installation of doors. |
Gravimetric analysis
All filters were conditioned in a controlled environment, more specifically at a relative humidity of 40 ± 2% and temperature of 20 ± 1 °C, for at least 48 h before weighing. The balance used was Sartorius semi-micro model MC5 balance (Sartorius AG, Göttingen, Germany) with a 210Polonium source to reduce static effect. Limit of detection (LOD) was calculated as 3 times the standard deviation of blank filters and found to be 0.01 mg per filter.
Respirable crystalline silica
Respirable crystalline silica was determined by X-ray diffraction, using the NIOSH method 7500 (NIOSH 2003), with standard reference material NIST 2950a (NIST, Gaithersburg, USA). Quantification control was done using the SRM-1878b standard for respirable α-quartz (NIST, Gaithersburg, USA) and SRM-1879a for respirable cristobalite (NIST, Gaithersburg, USA). Both calibration standards and quality control were treated according to NIOSH 7500. Samples were prepared according to NIOSH 7500 using low temperature plasma ashing (Diener Electronic, Ebhausen, Germany), and redeposited onto a pre-weighted silver membrane filter (Millipore Corp., Billerica, MA, USA) for absorption correction. The XRD instrument was Malvern Panalytical X’Pert3 Powder diffractometer, equipped with a PIXcel1D detector and an Empyrean X-ray tube (Malvern Panalytical B.V., Eindhoven, Netherlands). The LOD for α-quartz was 2 µg per filter.
Statistical analysis
Linear mixed models (LMM) were used to determine proportion of between-worker, within-worker, between location, and between-date-within-location variance, as well as the impact of different materials, exposure scenarios, and work tasks on the exposure to dust and RCS. LMMs were chosen instead of linear regression and ANOVA due to the dependency arising from repeated measurements and clustering within locations and dates. Mixed models also allow for a varying number of samples per individual, that as a characteristic of our data. Measurements of dust and RCS were log-transformed prior to statistical analysis to ensure approximately normally distributed response variables. The LMMs were used to assess the effect of work tasks on exposure to dust and RCS. Log transformed dust concentrations of thoracic dust, respirable dust, and RCS was used as response variables, with work task as fixed effects. Location and date were included as random effects, with date nested within location. A separate random effect (i.e. crossed with those of location and date) was added for worker. For the LMM for RCS, the percentage of quartz in the dust was also included as a fixed effect. Non-demolition work was set as the reference task as these workers were expected to have the lowest exposure. Worker was not nested within a given location as one worker could be assigned to multiple different locations throughout the study. Samples of inhalable dust were not analyzed using LMM as the number of samples were low compared with the number of variables. Seven samples (4%) analyzed for RCS were below the LOD of 0.002 mg on filter. As a substitution, individual values based on machine readings were used. A P-value less than 0.05 was considered statistically significant for all statistical analyses. Data analysis was performed in Rstudio v4.2.2 (R Core Team 2022) using the packages lme4 v1.1-31 (Bates et al., 2015). Figures were made using the ggplot2 package (Wickham, 2016).
The between-worker and within-worker variance from LMMs that included worker as random effect and exposure scenario as fixed effect was used to determine the degree of overexposure. Overexposure was used to describe the probability that the long-term mean exposure of an individual exceeds the given occupational exposure limit. It was calculated using the method described by Tornero-Velez et al. (1997).
Results
The total measured concentrations of respirable dust had a geometric mean (GM) of 0.88 mg/m3, and a range from 0.028 mg/m3 to 29 mg/m3, Table 2. The samples analyzed for RCS contained a GM of 0.040 mg/m3 of RCS, with concentrations ranging from below limit of detection to 3.2 mg/m3. The total measured concentrations of thoracic dust ranged from 0.17 mg/m3 to 65 mg/m3, with a GM of 2.5 mg/m3. Inhalable dust concentrations ranged from 0.98 mg/m3 to 163 mg/m3 and had a GM of 8.5 mg/m3. There was collected 81 complete sets of all dust fractions from the same worker during one shift.
Table 2.
Air concentrations (mg/m3) of inhalable dust, thoracic dust, respirable dust, and respirable crystalline silica (RCS), and the probability of overexposure based on current occupational exposure limits (OEL) stratified by exposure scenarios.
| Total | Indirect | Primary | Secondary | |
|---|---|---|---|---|
| Inhalable dust | N = 84 | N = 34 | N = 30 | N = 20 |
| Median | 6.2 | 3.8 | 22 | 16 |
| GM | 8.5 | 3.8 | 14 | 15 |
| AM ± SD | 20 ± 29 | 4 ± 3.3 | 30 ± 33 | 30 ± 36 |
| Min | 0.98 | 1.3 | 0.99 | 0.98 |
| Max | 163 | 18 | 135 | 163 |
| Thoracic dust | N = 131 | N = 58 | N = 51 | N = 22 |
| Median | 1.9 | 1.1 | 6.0 | 3.6 |
| GM | 2.5 | 1.2 | 4.7 | 3.8 |
| AM ± SD | 6 ± 11 | 1 ± 2.4 | 10 ± 15 | 8 ± 12 |
| Min | 0.17 | 0.17 | 0.36 | 0.32 |
| Max | 65 | 15 | 65 | 51 |
| Respirable dust | N = 192 | N = 81 | N = 80 | N = 31 |
| Median | 0.79 | 0.41 | 1.9 | 1.3 |
| GM | 0.88 | 0.43 | 1.7 | 1.6 |
| AM ± SD | 2 ± 5.3 | 0.7 ± 0.95 | 4 ± 6.5 | 4 ± 6.9 |
| Min | 0.028 | 0.028 | 0.11 | 0.13 |
| Max | 29 | 6.4 | 29 | 27 |
| Overexposurea (%) | 10% | 2.4% | 14% | 15% |
| Respirable crystalline silica | N = 182 | N = 71 | N = 80 | N = 31 |
| Median | 0.044 | 0.018 | 0.096 | 0.046 |
| GM | 0.040 | 0.020 | 0.11 | 0.050 |
| AM ± SD | 0.2 ± 0.42 | 0.06 ± 0.15 | 0.3 ± 0.54 | 0.2 ± 0.43 |
| Min | <LODb | <LODb | 0.0085 | 0.0076 |
| Max | 3.2 | 0.96 | 3.2 | 2.2 |
| Overexposurec (%) | 52% | 31% | 69% | 62% |
GM = Geometric mean, AM = Arithmetic mean, SD = standard deviation, N = number of measurements.
aOverexposure probability based on Norwegian OEL of 5 mg/m3.
bLimit of detection for RCS was 2 µg per filter.
cOverexposure probability based on Norwegian OEL of 0.05 mg/m3.
The distribution of respirable, thoracic, and inhalable dust concentrations was similar across all 3 exposure scenarios as shown in Fig. 1. The lowest dust concentrations were found for the indirect exposed group. All 3 exposure scenarios had individual measurements with dust concentrations that appear as outliers (1.5*range between 25 percentile and 75 percentile). However, the measurements were not regarded or removed as outliers, as there are no known disturbances in the measurements.
Fig. 1.
Air dust concentrations (mg/m3) for the inhalable dust, thoracic dust, respirable dust, and respirable crystalline silica based on exposure scenarios (primary, secondary, and indirect) for the complete sample sets (n = 81). Significance was tested using Wilcoxon test with P < 0.001 marked “***,” and P < 0.05 marked with “*.”
When comparing the 3 different exposure scenarios, the indirect exposed group were found to be exposed to lower concentrations of both dust and RCS than the primary and secondary exposed groups (Fig. 1). There was no significant difference between primary and secondary exposed groups for inhalable, thoracic, and respirable dust. However, the exposure to RCS was significantly lower for secondary exposed compared with primary exposed. The 81 complete sample sets of respirable, thoracic, and inhalable dust were used for these comparisons.
Seven samples had RCS concentrations below LOD (2 µg on filter) and all were from indirectly exposed workers. An earlier study points out that some bricks contain cristobalite, depending on the duration and temperature of firing when the brick was made (Chisholm 1999). None of the 182 respirable dust samples analyzed for RCS contained cristobalite, despite 90 samples being from locations with brick demolition or bricklaying.
In addition to exposure scenarios, each worker was divided into groups by work task, Table 1, to evaluate the effect on dust levels. Concentrations of dust and RCS was stratified by work tasks, shown in Table 3, with the non-demolition tasks foreman, bricklayer, formworker, and construction worker combined as “Non-demolition.” The lowest exposed tasks were non-demolition work, core drilling and soft strip demolition for all dust fractions and RCS. The highest exposure levels to RCS, respirable, and thoracic dust were from cutting concrete with GM values of 0.57 mg/m3, 5.5 mg/m3, and 39 mg/m3, respectively. The greatest variation in exposure levels for the 2 largest size fractions, inhalable and thoracic, defined as difference between GM and maximum measurement, was found in the tasks of clearing out, clearing out and some demolition, and grinding. There was not found a significant difference between exposure from grinding with or without local exhaust.
Table 3.
Concentrations (mg/m3) of inhalable dust, thoracic dust, respirable dust and respirable crystalline silica (RCS) stratified by work task.
| Total | Non-demolition | Cuttinga | Core drillinga | Soft strip demolition | Jackhammeringb | Clearing out | Clearing out and demolition | Grindingc | |
|---|---|---|---|---|---|---|---|---|---|
| Inhalable dust | N = 84 | N = 15 | N = 2 | N = 19 | N = 9 | N = 18 | N = 12 | N = 9 | |
| Median | 6.2 | 3.4 | 4.4 | 4.1 | 22 | 14 | 30 | 5.5 | |
| GM | 8.5 | 3.7 | 4.3 | 3.9 | 15 | 14 | 30 | 7.8 | |
| AM ± SD | 20 ± 29 | 4 ± 4.7 | 4 ± 1.2 | 4 ± 1.7 | 20 ± 18 | 30 ± 38 | 40 ± 31 | 30 ± 44 | |
| Min | 0.98 | 1.4 | 3.5 | 1.3 | 2.9 | 0.98 | 8.4 | 0.99 | |
| Max | 163 | 17 | 5.2 | 7.6 | 50 | 163 | 112 | 135 | |
| Thoracic dust | N = 131 | N = 28 | N = 4 | N = 6 | N = 30 | N = 12 | N = 20 | N = 23 | N = 10 |
| Median | 1.9 | 1.4 | 47 | 1.1 | 1.0 | 5.3 | 3.4 | 7.4 | 2.2 |
| GM | 2.5 | 1.4 | 39 | 1.3 | 1.0 | 2.4 | 3.5 | 7.0 | 2.3 |
| AM ± SD | 6 ± 11 | 2 ± 3.1 | 40 ± 22 | 1 ± 1.2 | 1 ± 1.3 | 6 ± 7.9 | 8 ± 13 | 8 ± 6.0 | 8 ± 18 |
| Min | 0.17 | 0.40 | 18 | 0.66 | 0.17 | 0.0050 | 0.32 | 1.0 | 0.36 |
| Max | 65 | 15 | 65 | 3.8 | 6.0 | 29 | 51 | 22 | 60 |
| Respirable dust | N = 192 | N = 50 | N = 8 | N = 8 | N = 31 | N = 16 | N = 29 | N = 38 | N = 12 |
| Median | 0.79 | 0.48 | 6.4 | 0.26 | 0.34 | 2.0 | 1.3 | 2.9 | 0.35 |
| GM | 0.88 | 0.49 | 5.5 | 0.33 | 0.34 | 1.7 | 1.5 | 2.6 | 0.68 |
| AM ± SD | 2 ± 5.3 | 0.7 ± 1.1 | 10 ± 11 | 0.4 ± 0.32 | 0.6 ± 0.77 | 2 ± 1.3 | 4 ± 7.1 | 4 ± 6.3 | 2 ± 6.0 |
| Min | 0.028 | 0.060 | 1.1 | 0.17 | 0.028 | 0.41 | 0.13 | 0.12 | 0.11 |
| Max | 29 | 6.4 | 29 | 1.1 | 3.2 | 4.7 | 27 | 27 | 21 |
| Respirable crystalline silica | N = 182 | N = 40 | N = 8 | N = 8 | N = 31 | N = 16 | N = 29 | N = 38 | N = 12 |
| Median | 0.044 | 0.025 | 0.93 | 0.027 | 0.0039 | 0.095 | 0.046 | 0.12 | 0.041 |
| GM | 0.040 | 0.030 | 0.57 | 0.030 | 0.010 | 0.13 | 0.060 | 0.10 | 0.070 |
| AM ± SD | 0.1 ± 0.42 | 0.08 ± 0.19 | 1.0 ± 0.93 | 0.04 ± 0.020 | 0.02 ± 0.040 | 0.2 ± 0.23 | 0.2 ± 0.45 | 0.2 ± 0.25 | 0.3 ± 0.89 |
| Min | <LODd | 0.0022 | 0.093 | 0.017 | <LODd | 0.020 | 0.0076 | 0.012 | 0.0085 |
| Max | 3.2 | 0.96 | 2.6 | 0.077 | 0.15 | 0.65 | 2.2 | 1.2 | 3.2 |
GM = Geometric mean, AM = Arithmetic mean, SD = standard deviation, N = number of measurements.
aWater cooling acting as exposure management.
bJackhammering was often done in combination with clearing out of materials. If most of the workday (>4 h) was spent demolishing the sample was regarded “Jackhammering” although some clearing out was done. If <4 h were spent demolishing and the rest clearing out the sample category “clearing and demolition” was used.
cLocal exhaust used in 6 of the measurements.
dLimit of detection for RCS was 2 µg per filter.
Linear mixed model analyses of the log-transformed concentrations of thoracic dust, respirable dust, and RCS, showed significantly higher exposure concentrations when doing the demolition tasks cutting, jackhammering, clearing out, and clearing out in combination with demolition (P < 0.05) (Table 4). The majority of the total variance was explained by location (44% to 58%), followed by day-to-day variance (11% to 27%), between worker variance (11% to 19%), and lastly within worker variance from 11% to 12%) across thoracic, respirable, and RCS dust concentrations. Including work tasks in the LMM models explained 24% and 25% of the total variability in thoracic and respirable dust concentrations respectively, primarily decreasing the variability between locations and within workers.
Table 4.
Linear mixed model of work tasks and quartz percentage in association with log-transformed concentrations of thoracic dust, respirable dust, and respirable crystalline silica (RCS). The non-demolition work tasks were used as reference.
| Thoracic dust | Respirable dust | Respirable crystalline silica | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Parameter | Exp(β) | 95 % CI | P | Exp(β) | 95% CI | P | Exp(β) | 95% CI | P |
| (Intercept) | 1.982 | 0.914–4.299 | 0.083 | 0.671 | 0.371–1.212 | 0.185 | 0.012 | 0.006–0.025 | <0.001 |
| Fixed effects | |||||||||
| Non demolition (REF) | 1 | 1 | 1 | ||||||
| Cutting | 4.185 | 1.469–11.921 | 0.008 | 3.656 | 1.960–6.820 | <0.001 | 4.459 | 2.191–9.073 | <0.001 |
| Core drilling | 0.837 | 0.191–3.667 | 0.812 | 0.689 | 0.191–2.483 | 0.568 | 1.086 | 0.292–4.044 | 0.901 |
| Soft strip demolition | 0.997 | 0.406–2.444 | 0.994 | 1.014 | 0.487–2.110 | 0.970 | 0.861 | 0.386–1.919 | 0.713 |
| Jackhammeringa | 2.218 | 1.014–4.852 | 0.046 | 2.955 | 1.776–4.917 | <0.001 | 2.971 | 1.667–5.294 | <0.001 |
| Clearing out | 2.237 | 1.106–4.522 | 0.025 | 2.199 | 1.380–3.502 | 0.001 | 2.447 | 1.445–4.143 | 0.001 |
| Clearing out and demolition | 3.506 | 1.691–7.267 | 0.001 | 2.733 | 1.660–4.500 | <0.001 | 3.001 | 1.742–5.171 | <0.001 |
| Grinding | 1.439 | 0.556–3.724 | 0.450 | 1.346 | 0.637–2.843 | 0.434 | 1.457 | 0.654–3.246 | 0.355 |
| % quartz in dust | - | - | - | - | - | - | 1.154 | 1.104–1.207 | <0.001 |
| Random effects | |||||||||
| Between location variance (naïve estimateb) | 0.72 (1.16) | 0.45 (0.90) | 0.64 (1.73) | ||||||
| Day-to-day variance within location (naïve estimate) | 0.29 (0.21) | 0.65 (0.55) | 0.52 (0.71) | ||||||
| Between worker variance (naive estimate) | 0.28 (0.39) | 0.22 (0.37) | 0.22 (0.36) | ||||||
| Within worker variance (naive estimate) | 0.23 (0.25) | 0.23 (0.24) | 0.25 (0.34) | ||||||
| Total variance (naïve estimate) | 1.52 (2.01) | 1.55 (2.06) | 1.63 (3.14) | ||||||
| % explained variance by the model | |||||||||
| Between location variance | 38% | 50% | 63% | ||||||
| Day-to-day variance within location | −38% | −18% | 27% | ||||||
| Between worker variance | 28% | 41% | 39% | ||||||
| Within worker variance | 8% | 4% | 26% | ||||||
| Total variance | 24% | 25% | 48% |
Exp (β) = inverse log transformation of β-coefficients for the exposure concentrations.
aJackhammering was often done in combination with clearing out of materials. If most of the workday (>4 h) was spent demolishing the sample was regarded “Jackhammering” although some clearing out was done. If <4 h were spent demolishing and the rest clearing out the sample category “clearing out and demolition” was used.
bNaïve estimates derived from a model without fixed factors.
A significant effect of quartz percentage was found with an increase of 15% in quartz percentage for each percent increase in quartz content in respirable dust (Table 4). Including both task and quartz percentage as fixed effects explained 48% of the total variability in RCS concentrations (Table 4). Using only quartz percentage or work task as fixed effect explained 33% and 16% of the total variation in RCS concentration, respectively (Supplementary Table S1). There was no significant difference in thoracic dust, respirable dust, and RCS exposure levels between workers handling brick materials and those handling concrete when using a LMM with both task and material as fixed effects.
Discussion
The total dust concentrations in indoor renovation showed great variation between measurements (Table 2). Both the geometric mean and median for respirable dust and RCS were below the current Norwegian OELs of 5 mg/m3 and 0.05 mg/m3. However, the median RCS exposure level was 0.044 mg/m3 and the probability of overexposure was 52%. The total exposure levels of RCS may thereby be underestimated if one only considers the exposure to respirable dust. Swedish and Danish studies from demolition and mechanical work of concrete and brick found similar results, indicating that the risk of exposure levels exceeding the RCS OELs (0.1 mg/m3 or 0.05 mg/m3, depending on the country), often was higher than exceeding the corresponding OEL for respirable dust (Kirkeskov et al. 2016; Grahn and Lewné 2017; Antonsson and Sahlberg 2019).
There was a significant difference in dust air concentrations between the indirect exposure scenario and the primary and secondary exposure scenarios (P < 0.05) (Fig. 1). This study shows that in an indoor renovation environment, exposure to RCS can exceed recommended levels, even for indirectly exposed workers staying in the same building during structural demolition. Demolition within an enclosed building with limited ventilation may result in significant background exposure to suspended dust, independent of exposure scenario. If the demolition dust is not actively removed from the enclosed building or otherwise controlled, it may remain suspended through internal airflow or be resuspended by workers throughout the workday. Visual observations and data from the sampling form indicated that indirect exposed workers tended to not use personal dust protection. Primary and secondary exposed workers, in comparison, tended to use personal dust protection during activities that generated visible amounts of dust like jackhammering, clearing out and cutting.
The findings on indirect exposure are supported by the findings of Tuomi et al. (2022) and show the importance of reducing the general background dust concentrations. An underestimation of exposure levels may occur if assessments in the renovation industry are solely based on task measurements, where the worker is believed to be unexposed when not performing a specific demolition task. Full shift measurements will, on the other hand, underestimate short-term peak exposure. However, the workers are rarely exposed solely to the pollutant from the task they work with, but rather to all the different work operations or tasks being done within the enclosed space (Guo et al. 2022). If the average air concentration is not reduced for the remainder of the working day, the worker may still be exposed to a substantial amount of RCS (Tuomi et al. 2022). Although multiple dust control measures are on the market, with varying degrees of dust control effectiveness, the use of these does not rule out the risk of overexposure to RCS (Tjoe Nij et al. 2003; Akbar-Khanzadeh et al. 2010; Anlimah et al. 2023).
There was significant variation in the exposure levels of all dust health fractions and RCS during the tasks of cutting and grinding. Equipment used for cutting and grinding often had integrated dust control measures such as water or local exhaust ventilation and exposure levels may depend on the type and effect of these measures. Similar variance in exposure levels may be found in literature where the GM of exposure levels from concrete cutting vary from GM 0.72 mg/m3 to 1.9 mg/m3 for respirable dust and 0.08 mg/m3 to 0.42 mg/m3 RCS (Tjoe Nij et al. 2004; Flanagan et al. 2006). Grinding had a lower overall exposure level compared to cutting but had the highest maximum value measured for RCS of all samples in the study (3.2 mg/m3). No local exhaust was used for the maximum measurement.
The exposure levels found when jackhammering were similar to earlier studies like Flanagan et al. (2006), Sauve et al. (2013), and Tjoe et al. (2004), while exposure levels during clearing out varied in literature. Some studies reported lower exposure levels than those found in this study during cleaning/clearing out, 0.58 to 0.66 mg/m3 for respirable dust and RCS from 0.017 to 0.05 mg/m3 (Tjoe Nij et al. 2004; Flanagan et al. 2006). A Dutch study with demolition workers that both demolished and cleared out showed similar results with a GM for respirable dust of 1.17 mg/m3, with maximum measurements up to 34 mg/m3, and RCS levels with GM at 0.12 mg/m3 (van Deurssen et al. 2014). Cleaning and clearing out may be an unspecific category that may include waste management, clearing out and cleaning of newly constructed buildings. Clearing out materials may be a work task that is either overlooked as described by Chisholm (1999) or included as part of demolition as mentioned by Tjoe Nij et al. (2004) (Chisholm 1999; Tjoe Nij et al. 2004). Clearing out is both time consuming and a physically straining task where dust management and use of personal protective equipment can be challenging. The high variability of exposure levels for clearing out can be due to difference in ventilation, where dust is resuspended and stay suspended if ventilation is not sufficient.
Despite up to 5 times higher respirable dust concentrations when clearing out than during jackhammering, the RCS concentrations were only 2 to 3 times higher for the same samples, indicating a lower weight percentage of RCS in dust from clearing out. In a study by Boudigaard et al., the use of power tools as well as the quartz content in materials were found to be main determinants for exposure to respirable quartz exposure (Boudigaard et al. 2021). The dust that is resuspended during clearing out may originate from many different sources, leading to a reduction in RCS content compared to dust from the demolished material. Thereby, the percentage of crystalline silica in the respirable dust is an important factor when evaluating the exposure to RCS, as well as the total amount of respirable dust.
Water was utilized during core drilling and cutting to cool down the blade and drill bits in this study and thereby indirectly reducing the workers’ exposure to dust. If the slurry produced is not removed in the wet stage, the water will dry up, and the dust may be resuspended and contribute to the background concentration within the building. Water mist suppressions were not used as dust control at the locations in this study in order to not damage the buildings or to save time to dry the buildings before continuing renovation. An intervention study from the Dutch construction industry showed an substantial overall reduction in RCS exposure for demolition workers, concrete drillers and tuck pointers with an increased use of technical control measures, especially water suppression (van Deurssen et al. 2015). However, the magnitude of intervention was hard to assess due to other factors changing over time that were not an element of the intervention, showing the complexity of exposure in the construction industry.
The LMMs were used to determine the effect of work task, worker, date, site of sampling, and quartz concentration upon exposure levels. The between-location variance was greater than the between-day variance within the same location. This may be explained by differences between locations in the materials being demolished since some materials differ in ability to form dust and silica content, although the difference in exposure concentration between brick and concrete was insignificant. The work tasks are heavily influenced by the tool used, with significant increases in exposure concentrations during cutting and jackhammering, both of which involve power tools. However, a similar increase in exposure concentrations was seen for clearing out, indicating that work task can be a good alternative determinant to tools to include this type of work. Quartz percentage explained 33% of total variability in RCS concentrations and in combinations with task it explained almost 50%. The expected quartz content in the demolished materials and thereby dust can be an indication of RCS exposure.
Limitations
One of the challenges with the study was the unpredictability of sampling. Although the goal was to include tasks and have repeated measurement performed on each individual, it was not always possible to achieve this. Task categories such as clearing out with demolition might appear unspecific, but one worker rarely performed one single task throughout the workday. One example being restrictions on noise levels throughout the working day, which resulted in 3 h of active demolition and 5 h of clearing out materials. It would have provided an inaccurate image of the overall dust concentrations if the measurements had only been made during active demolition with a jackhammer. Although the full shift approach resulted in fewer, more general task categories, the overall exposure picture for workers in this industry may be better understood.
Although information on personal protection was collected through personal sampling forms, it was proven difficult to collect exact information on correct usage without monitoring the workers continuously throughout the workday.
There were also challenges when comparing the obtained exposure levels to other findings as previous studies were often based on either few measurements per task, shorter task-based measurements, or aggregated studies with a bigger sample pool, but less specific information about each category. Comparing the exposure of a group to a single or limited measurement of a work task may give a wrong impression of the total exposure within the industry. Although this study only included a small number of task specific measurements for cutting, core drilling and grinding, these measurements demonstrate the variation in dust concentrations in renovation projects.
One limitation with the study is the NIOSH 7500 method which is validated up to 2 mg respirable dust and/or quartz on filter. Thus, for the respirable samples above this limit, the accuracy of the method may be reduced. For the time being, there are no methods validated for masses above 2 mg on filter.
Conclusion
The study revealed substantial exposure to dust and RCS in the renovation industry. Mechanical tasks and clearing out, yielded the highest exposure levels, and were significantly higher than soft stripping, core drilling with water, and non-demolition work. No difference was found between exposure levels of dust and RCS between workers renovating brick and concrete buildings. In addition, indirectly exposed workers should be included in assessments due to individually high exposure levels of RCS. Preventive actions to remove dust at the source and reduce resuspension of dust are required to effectively protect workers in the renovation industry from RCS and respirable dust.
Supplementary material
Supplementary material is available at Annals of Work Exposures and Health online.
Acknowledgments
We are greatly appreciative of the funding support from the Norwegian Agreement for a More Inclusive Working Life (IA Agreement) in the Construction Industry. The authors would like to thank Stine Eriksen Hammer, currently at the Norwegian Coastal Administration, for help with planning in the early stages of the project. Thanks to Kari Dahl at STAMI for support with XRD analysis and to Øivind Skare at STAMI for statistical support. Lastly, a special thanks to all the contractors and their workers for participating in the project.
Contributor Information
Johanne Ø Halvorsen, STAMI, National Institute of Occupational Health, Gydas Vei 8, 0363 Oslo, Norway; Faculty of Environmental Sciences and Natural Resource Management, Norwegian University of Life Sciences, P.O. Box 5003, NO-1432 Aas, Norway.
Pål Graff, STAMI, National Institute of Occupational Health, Gydas Vei 8, 0363 Oslo, Norway; Faculty of Environmental Sciences and Natural Resource Management, Norwegian University of Life Sciences, P.O. Box 5003, NO-1432 Aas, Norway.
Elin Lovise Folven Gjengedal, Faculty of Environmental Sciences and Natural Resource Management, Norwegian University of Life Sciences, P.O. Box 5003, NO-1432 Aas, Norway.
Torunn K Ervik, STAMI, National Institute of Occupational Health, Gydas Vei 8, 0363 Oslo, Norway.
Funding
This research was supported by funding from the Norwegian Agreement for a More Inclusive Working Life (IA Agreement) in the Construction Industry.
Conflict of interest
The authors declare no conflict of interest relating to the material presented in this article. Its contents, including any opinions and/or conclusions expressed, are solely those of the authors.
Data availability
The data underlying this article will be shared on reasonable request to the corresponding author.
References
- Akbar-Khanzadeh F, Milz SA, Wagner CD, Bisesi MS, Ames AL, Khuder S, Susi P, Akbar-Khanzadeh M.. 2010. Effectiveness of dust control methods for crystalline silica and respirable suspended particulate matter exposure during manual concrete surface grinding. J Occup Environ Hyg. 7(12):700–711. https://doi.org/ 10.1080/15459624.2010.527552 [DOI] [PubMed] [Google Scholar]
- Anlimah F, Gopaldasani V, MacPhail C, Davies B.. 2023. A systematic review of the effectiveness of dust control measures adopted to reduce workplace exposure. Environ Sci Pollut Res. 30(19):54407–54428. https://doi.org/ 10.1007/s11356-023-26321-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- Antonsson A-B, Sahlberg B.. 2019. Referensmätningar för kvartsexponering vid rot-arbeten inom byggindustri. Stockholm, Sweden: IVL Svenska Miljöinstitutet. [Google Scholar]
- Bakke B, Degerud E, Gravseth HM, Hanvold TN, Løvseth EK, Mjaaland BB, Sterud T, Øygardslia H.. 2021. Faktabok om arbeidsmiljø og helse. Oslo, Norway: The National Institute of Occupational Health in Norway. [Google Scholar]
- Bates D, Mächler M, Bolker B, Walker S.. 2015. Fitting linear mixed-effects models using lme4. Journal of Statistical Software. 67(1):1–48. https://doi.org/ 10.18637/jss.v067.i01. [DOI] [Google Scholar]
- Boudigaard S, Hansen K, Kolstad H, Kromhout H, Schlünssen V.. 2021. Determinants of respirable quartz exposure concentrations across occupations in denmark, 2018. Ann. Work Expo. Health. 66(4):472–480. https://doi.org/ 10.1093/annweh/wxab116. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Callister WD, Rethwish DG.. 2007. Callister’s materials science and engineering. 7th. New York, US: John Wiley & sons. [Google Scholar]
- Chisholm J. 1999. Respirable dust and respirable silica concentrations from construction activities. Indoor Built Environ. 8(2):94–106. https://doi.org/ 10.1159/000024619 [DOI] [Google Scholar]
- Circle Economy. 2020. The Circularity Gap Report. Amsterdam, The Netherlands: Circle Economy. p. 69. [Google Scholar]
- Filippidou F, Jiménez Navarro JP. 2019. Achieving the cost-effective energy transformation of Europe’s buildings. Publications Office of the European Union: Luxembourg. https://doi.org/ 10.2760/278207 [DOI] [Google Scholar]
- Flanagan ME, Seixas N, Becker P, Takacs B, Camp J.. 2006. Silica exposure on construction sites: results of an exposure monitoring data compilation project. J Occup Environ Hyg. 3(3):144–152. https://doi.org/ 10.1080/15459620500526552 [DOI] [PubMed] [Google Scholar]
- Fufa SM, Flyen C, Venås C.. 2020. Grønt er ikke bare en farge: Bærekraftige bygninger eksisterer allerede. Oslo, Norway: SINTEF:105. [Google Scholar]
- Grahn K, Lewné. 2017. Hur höga är kvartsnivåerna inom byggbranschen i stockholm - en orienterende kartläggning 2016-2017 Stockholm, Sweden: Centrum för arbets-och miljömedicin. [Google Scholar]
- Guo P, Tian W, Li H.. 2022. Dynamic health risk assessment model for construction dust hazards in the reuse of industrial buildings. Build Environ. 210:108736. https://doi.org/ 10.1016/j.buildenv.2021.108736 [DOI] [Google Scholar]
- Hnizdo E, Vallyathan V.. 2003. Chronic obstructive pulmonary disease due to occupational exposure to silica dust: a review of epidemiological and pathological evidence. Occup Environ Med. 60(4):237–243. https://doi.org/ 10.1136/oem.60.4.237 [DOI] [PMC free article] [PubMed] [Google Scholar]
- HSE. 2022. Controlling exposure to stone dust. London, UK: Health and Safety Executive (HSE). [Google Scholar]
- IARC. 2012. Silica dust, crystalline, in the form of quartz or crystobalite. Lyon France: International Agency for Research on Cancer. https://www.ncbi.nlm.nih.gov/books/NBK304370/. Accsessed October 26 2023. [Google Scholar]
- Kirkeskov L, Hanskov DJA, Brauer C.. 2016. Total and respirable dust exposures among carpenters and demolition workers during indoor work in denmark. J Occup Med Toxicol. 11(1):45. https://doi.org/ 10.1186/s12995-016-0134-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lumens MEGL, Spee T.. 2001. Determinants of exposure to respirable quartz dust in the construction industry. Ann Occup Hyg. 45(7):585–595. https://doi.org/ 10.1016/S0003-4878(01)00025-4. [DOI] [PubMed] [Google Scholar]
- NIOSH. 2003. Silica crystalline, by xrd, method 7500 (filter redeposition). Niosh manual of analytical methods (nmam). 4th ed. Washington, USA: NIOSH. p. 9. [Google Scholar]
- R Core Team. 2022. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. [Google Scholar]
- Sauvé JF, Beaudry C, Bégin D, Dion C, Gérin M, Lavoué J.. 2013. Silica exposure during construction activities: statistical modeling of task-based measurements from the literature. Ann Occup Hyg. 57(4):432–443. https://doi.org/ 10.1093/annhyg/mes089 [DOI] [PubMed] [Google Scholar]
- Tjoe Nij E, Hilhorst S, Spee T, Spierings J, Steffens F, Lumens M, Heederik D.. 2003. Dust control measures in the construction industry. Ann Occup Hyg. 47(3):211–218. https://doi.org/ 10.1093/annhyg/meg023 [DOI] [PubMed] [Google Scholar]
- Tjoe Nij E, Höhr D, Borm P, Burstyn I, Spierings J, Steffens F, Lumens M, Spee T, Heederik D.. 2004. Variability in quartz exposure in the construction industry: implications for assessing exposure-response relations. J Occup Environ Hyg. 1(3):191–198. https://doi.org/ 10.1080/15459620490424528 [DOI] [PubMed] [Google Scholar]
- Tornero-Velez R, Symanski E, Kromhout H, Yu RC, Rappaport SM.. 1997. Compliance versus risk in assessing occupational exposures. Risk Anal. 17(3):279–292. https://doi.org/ 10.1111/j.1539-6924.1997.tb00866.x [DOI] [PubMed] [Google Scholar]
- Tuomi T, Johnsson T, Heino A, Lainejoki A, Salmi K, Poikkimäki M, Kanerva T, Säämänen A, Räsänen T.. 2022. Managing quartz exposure in the construction industry. Helsinki, Finland: Finnish institute of occupational health. [DOI] [PMC free article] [PubMed] [Google Scholar]
- van Deurssen E, Meijster T, Hengel K, Boessen R, Spaan S, Tielemans E, Heederik D, Pronk A.. 2015. Effectiveness of a multidimensional randomized control intervention to reduce quartz exposure among construction workers. Ann Occup Hyg. 59(8):959–971. https://doi.org/ 10.1093/annhyg/mev037. [DOI] [PubMed] [Google Scholar]
- van Deurssen E, Pronk A, Spaan S, Goede H, Tielemans E, Heederik D, Meijster T.. 2014. Quartz and respirable dust in the dutch construction industry: a baseline exposure assessment as part of a multidimensional intervention approach. Ann Occup Hyg. 58(6):724–738. https://doi.org/ 10.1093/annhyg/meu021 [DOI] [PubMed] [Google Scholar]
- Wickham H. 2016. ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York, 2016.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data underlying this article will be shared on reasonable request to the corresponding author.