Personal exposure to inhalable cement dust among construction workers

Abstract

Objective — A case study was carried out to assess cement dust exposure and its determinants among construction workers and for comparison among workers in cement and concrete production.

Methods — Full-shift personal exposure measurements were performed and samples were analysed for inhalable dust and its cement content. Exposure variability was modelled with linear mixed models.

Results — Inhalable dust concentrations at the construction site ranged from 0.05 to 34 mg/m³, with a mean of 1.0 mg/m³. Average concentration for inhalable cement dust was 0.3 mg/m³ (GM; range 0.02–17 mg/m³). Levels in the ready-mix and pre-cast concrete plants were on average 0.5 mg/m³ (GM) for inhalable dust and 0.2 mg/m³ (GM) for inhalable cement dust. Highest concentrations were measured in cement production, particularly during cleaning tasks (inhalable dust GM = 55 mg/m³; inhalable cement dust GM = 33 mg/m³) at which point the workers wore personal protective equipment. Elemental measurements showed highest but very variable cement percentages in the cement plant and very low percentages during reinforcement work and pouring. Most likely other sources were contributing to dust concentrations, particularly at the construction site. Within job groups, temporal variability in exposure concentrations generally outweighed differences in average concentrations between workers. ‘Using a broom’, ‘outdoor wind speed’ and ‘presence of rain’ were overall the most influential factors affecting inhalable (cement) dust exposure.

Conclusion — Job type appeared to be the main predictor of exposure to inhalable (cement) dust at the construction site. Inhalable dust concentrations in cement production plants, especially during cleaning tasks, are usually considerably higher than at the construction site.

---

Introduction

Cement dust exposure occurs commonly in the cement production and construction industry. Most studies carried out in these industries, however, focus on silica exposure. There is no measurement data published about cement dust at construction sites at all. On cement dust exposures in cement plants only limited data is available. For instance in a Tanzanian study, the exposure levels turned out to be very high with a considerable variability.¹ Woskie and colleagues (2002) reported a large variability in inhalable dust exposure levels during construction operations as well.²

The Health and Safety Executive (HSE) in the UK concluded in 1994 that there was no convincing evidence for an increased incidence of any site-specific cancer resulting from cement exposure, but acknowledged that the data available at that time was not consistently and reassuringly negative.³ In 2005 the HSE reviewed findings in more recently published studies. Among these and earlier studies increased incidences of cancers at several sites (stomach, lungs, colon, head and neck) were noted. A causal association between Portland cement exposure and cancer has however not been established and therefore the uncertainty concerning a possible cancer risk maintains.⁴ Few studies performed after the HSE review also show some evidence for pharyngeal carcinoma,⁵ oesophageal adenocarcinoma,⁶ squamous cell carcinoma of the skin⁷ and gastric cancer.⁸ Besides a possible cancer risk, occupational exposure to cement dust has been associated with reduced respiratory function. At present evidence is insufficient to establish exposure-response relationships for these effects.⁴

The present study was conducted in the Rhein-Neckar area in Germany, where in a recent epidemiological hospital-based case-control study on head- and neck cancer an increased risk of laryngeal carcinoma for construction workers was suggested with cement dust exposure as possible cause.⁹ In this context, a case study was carried out in 2006–2007 to assess current cement dust exposure of construction workers involved in a full-scale construction project. Job tasks in all stages were included. For comparison, we also incorporated workers involved in the various stages of cement and concrete production. This method is rather unique as we studied the entire process; both the production chain and the total building process. Aim of this study was to assess the actual cement dust exposure among construction workers involved in a full-scale construction project and as a comparison among workers in various stages of cement and concrete production. Determinants of exposure to inhalable (cement) dust were evaluated.

Methods

Exposure measurements

Full-shift personal inhalable dust measurements in the breathing zones were performed during several tasks in the construction project (i.e. reinforcement work, concrete pouring, concrete repair, bricklaying, floor screed laying, installation work and tile setting) and among workers in a cement plant, in five ready-mix concrete plants and in a pre-cast concrete plant. The sampling equipment was in conformity with European standard EN 1232.¹⁰ A portable pump operating at a flow rate of 3.5 l/min was used, together with a GSP (CIS) sampling head,¹¹ using PVC filters with a 37 mm diameter and a 5-µm pore-diameter. Measurements were all performed after clear instructions of the workers by a trained industrial hygienist. Daily activities and relevant situations were documented on extensive field forms. Several field blank GSP filters were collected without passing air through the filter on each sample location. Based on these blanks, the limit of detection (LOD) was estimated to be 0.065 mg/m³ for inhalable dust. The LOD for the cement-content was based on the analytical limit of detection for calcium (0.015 mg) and was estimated to be 0.035 mg/m³.

At each location and each measurement day, local weather conditions were monitored using an automated weather station. Thirty minute averages (including minimum and maximum values) were determined for the components wind speed and direction, temperature, and amount of rain and sunshine during the whole 8-hour sampling period. The weather station was always placed on ground level where data collection was not disturbed by nearby buildings or other high obstacles.

Sampling strategy

At the construction site it was planned to include five workers per job and collect four samples from each worker. As reinforcement work and pouring were always done on the same day by the same workers, ten workers were sampled for these jobs together. Moreover, there were not enough bricklayers and concrete repairers present at the construction site. We therefore measured the maximum number of workers possible, resulting in two concrete repairers and four bricklayers. One bricklayer was only measured once, so five samples were taken from the other three bricklayers. Two bricklayers performed concrete repair work as well. Even though five floor screed layers were included, only sixteen samples could be taken before the work was finished. Installation workers were measured according to the protocol. Also the planned twenty measurements for the tile setters were accomplished, although the number of samples per worker varied.

At the non-construction sites, five workers per location were randomly selected for a first measurement period. Two samples were obtained per worker. From the same workers another sample was collected in a second period, which was within a few weeks after the first. In this second period, two samples were also collected from an additional group of five workers. Due to the small number of workers in the ready-mix plants, only five workers (one worker per plant) were measured here in total. In the second measurement period in the cement plant, one worker could not be measured twice; hence the second sample was taken from another worker.

Gravimetrical measurements

Gravimetrical and element measurements were done under continuous temperature and air humidity control. The laboratory balance used was a Sartorius MC Balance, max capacity 5.1 g, stability <1 µg. The mass detection limit using an in house 37 mm PVC filter kept under the above conditions was 0.011 mg (3 times standard deviation). Provisions to discharge the filters prior to weighing were used at all occasions.

Elemental composition

After weighing, each filter was placed in a Teflon autoclave and 1 ml of aqua regia, 1 ml of hydrofluoric acid and a known quantity of beryllium chloride solution were added as an internal spectroscopic standard. The Teflon autoclave was heated in a microwave unit (MLS 1200, Teflon container SV140, 10 Bar, Milestone, Sorisole, Italy); after cooling, to the digest was added 1 ml of saturated boric acid and reheated before final dilution to 15 ml. The Ca, Cr and Si concentrations in the solutions were measured by inductively coupled plasma optical emission spectrometry (ICP-OES) using a Perkin-Elmer Optima 3000 spectrometer. Certified NIST reference Portland cement materials were used for quality assurance of the chemical measurements. The cement content of the inhalable dust was based on the Ca-content of the collected samples. Cement coded CEM II/B-S 32.5 R was used as reference. The composition of this cement was as follows: SiO₂ (23.6%), Al₂O₃ (6.67%), Fe₂O₃ (2.54%), CaO (56.8%), MgO (4.23%), MnO (0.14%), K₂O (1.25%), SO₃ (3.05%), Cl (0.078%), S (0.22%), Cr 58.2 ppm.

Statistical analyses

Exposures were log normally distributed and analyses were consequently performed on log transformed data. Values below limit of detection (LOD) were replaced by LOD/√2 (n = 1 (0.6%) for inhalable dust and n = 10 (6%) for inhalable cement dust). Air concentrations of inhalable dust and cement dust were summarised as arithmetic means (AM), geometric means (GM) and geometric standard deviations (GSD). Exposure variability was modelled with linear mixed models, divided into a personal (between-worker variability in average concentrations) and temporal (within-worker variability from day-to-day) component. ‘Worker’ was included as random effect, ‘job type’ and other assumed influential factors as fixed effects. To express the variability, the R.₉₅ was calculated, which is the ratio of the 97.5^th and 2.5^th percentiles of the log-normal distribution of the between (₉₅R_bw) and within (₉₅R_ww) worker distribution.¹² The exponent of the estimated β-coefficient (e^ß) was used to estimate the percentage increase or decrease in exposure levels. All analyses were performed using SAS version 9.1 software (SAS Institute Inc.).

Results

A total of 180 personal measurements were performed, with an average sampling time of 7 hours and 43 minutes. Eleven measurements lasted less than 6 hours, of which the shortest covered only 2 hours. Concentrations were not standardized to an 8-hour working day.

As shown in Table 1, lowest exposures were observed for reinforcement and pouring workers, with a geometric mean (GM) of 0.33 mg/m³. Personal exposure levels in ready-mix and pre-cast concrete plants were very similar, with GMs of 0.37 and 0.39 mg/m³, respectively. The highest exposures at the construction site were seen for concrete repairers and tile setters, who had GMs of 2.4 and 3.9 mg/m³. A similar picture was seen for inhalable cement dust exposure. Cement plant workers had by far the highest inhalable dust exposure. Especially while performing cleaning tasks it was extremely high. The maximum value of 194 mg/m³ (in an 8-hour sampling) was measured when a worker performed cleaning tasks, most of the time using a broom.

Table 1 Inhalable dust and inhalable cement dust (mg/m³) with variance components, by job type^a

	D	K	N	AM	GM	GSD	Range	95Rbw	95Rww
a D = number of measurement days; K = number of workers; N = number of samples; AM = arithmetic mean; GM = geometric mean; GSD = geometric standard deviation; 95Rbw=fold-ratio of average concentration between workers; 95Rww=fold-ratio of variability within worker (from day to day). b These workers wore face pieces as breathing protection while cleaning the majority of time.
Inhalable dust
Total group	54	55	180	5.4	1.04	4.13	0.05 –194	122	24
All construction workers	35	29	120	2.0	0.97	3.20	0.05 –34	48	14
Reinforcement & pouring	4	10	40	0.46	0.33	2.34	0.05–2.7	15	8
Concrete repair	4	2	8	3.0	2.4	1.96	0.96–6.2	1	14
Bricklaying	5	4	16	0.91	0.76	1.71	0.38–3.7	1	8
Floor screed laying	8	5	16	3.4	1.6	2.36	0.77–34	1	29
Installation work	6	5	20	1.5	1.2	1.90	0.44–6.5	3	11
Tile setting	8	5	20	4.7	3.9	1.85	1.2–15	3	9
Cement plant	6	11	19	7.0	3.0	3.53	0.35–52	38	27
Cement plant (Cleaning)^b	2	3	6	98	55	3.77	8.8–194	1	182
Ready-mix plants	10	5	10	0.47	0.37	2.17	0.10–1.1	18	4
Pre-cast plant	5	10	25	0.43	0.39	1.61	0.16–1.1	2	6
Inhalable cement dust
Total group	54	55	180	3.6	0.38	5.53	0.02 –194	431	25
All construction workers	35	29	120	0.86	0.31	4.28	0.02 –17	185	13
Reinforcement & pouring	4	10	40	0.09	0.06	2.25	0.02–0.51	12	8
Concrete repair	4	2	8	1.5	1.2	1.96	0.44–3.3	1	14
Bricklaying	5	4	16	0.34	0.32	1.47	0.19–0.62	1	5
Floor screed laying	8	5	16	1.7	0.75	2.45	0.36–17	3	28
Installation work	6	5	20	0.47	0.34	1.94	0.15–3.0	3	11
Tile setting	8	5	20	2.3	1.9	1.80	0.58–7.5	3	8
Cement plant	6	11	19	5.2	1.7	4.72	0.10–49	88	45
Cement plant (Cleaning)^b	2	3	6	72	33	4.73	3.9–194	1	443
Ready-mix plants	10	5	10	0.19	0.11	3.22	0.02–0.58	69	10
Pre-cast plant	5	10	25	0.21	0.17	1.81	0.07–0.67	3	7

---

Table 1 also shows that in most cases the temporal variability (within a worker from day-to-day) dominates the total variability in both inhalable dust and cement concentrations. The ₉₅R_ww for inhalable dust ranges from 4 in the ready-mix plants to 182 for cleaning in the cement plant. Temporal variability among construction workers was highest for floor screed layers with a ₉₅R_ww of 29, where it ranged from 8 to 14 for the other tasks. In the cement plant, the ready-mix concrete plants, and during reinforcement work and concrete pouring considerable differences in average concentrations between individual workers existed, with ₉₅R_bw′s of 38, 18 and 15, respectively. The personal differences were quite low for all other workers. The structure of the variability was similar for inhalable cement concentrations.

The distribution of inhalable dust and inhalable cement concentration is graphically presented in figure 1 (log-scale). The actual correlation between inhalable dust concentrations and amount of cement in the dust samples based on Ca-content was rather high (Pearson R = 0.91). The correlation was highest (Pearson R>0.94) for samples taken in the cement plant, the ready-mix concrete plants and during concrete repair, floor screed laying, installation work, and tile setting, while for cleaning in the cement factory it was somewhat lower (0.82). The correlation was also lower (Pearson R<0.78) at the pre-cast concrete plants and when reinforcement work and pouring at the construction site were sampled. It was by far the lowest (Pearson R<0.18) for bricklayers. From Fig. 2 - showing cement percentages of inhalable dust - it becomes clear that the actual cement percentage varied extensively from sample to sample, especially in the cement-producing factory where the average percentages in the collected dust lay above 60%. The remaining job types showed much lower percentages. On samples from the ready-mix concrete plants only 33% of the inhalable dust was analysed as cement. Dust in reinforcement and installation work contained even less than 30% cement. On 94% of the filters, both silicon (Si) and calcium (Ca) were detectable. Fig. 3 shows the ratios between Ca and Si found for the different jobs. The highest ratios (4–5) were observed in the cement factory and lowest during reinforcement and pouring (0.7).

Fig. 1 Inhalable (cement) dust concentrations by job type (Each box shows the median, 25^th and 75^th percentiles; extreme values are marked with a square).

Fig. 2 Percentage cement in inhalable dust by job type.

Fig. 3 Calcium-Silicon ratio by job type.

‘Job performed’ turned out to be the most important determinant of exposure variability. This variable explained over 80% of the differences in average concentration between monitored workers. All other determinants together (meteorological conditions, tools used, ventilation characteristics and smoking) only explained 55% of the between-worker variability. This picture was similar for cement concentrations, with 83% and 53% respectively. Restricted to construction workers, ‘job performed’ explained 80% of inhalable dust exposure variability and 89% of the cement dust exposure. A large amount of variability for both inhalable and cement dust exposure (71% and 64%) could however also be explained by the other determinants together (data not shown).

Table 2 illustrates how several determinants affected the exposure concentrations. In the total group of subjects, wind speed significantly decreased both inhalable dust and inhalable cement dust concentrations. For the construction workers only, this effect was somewhat weaker. Rain had a negative effect on the observed concentrations, which was strongest for inhalable cement dust. Temperature as such had no influence when wind speed and rain were taken into account.

Table 2 Determinants of inhalable dust and inhalable cement dust exposure

	Total group (k = 55, n = 180)			Construction workers (k = 29, n = 120)
	β (SE)^a	Effect	P-value	β (SE)^a	Effect	P-value
a β-coefficient and standard error of estimate. b There was no local ventilation at the construction site; bwS^2 = between worker variance; wwS^2 = within worker variance.
Inhalable dust
Intercept	0.21(0.41)		0.601	−0.45(0.40)		0.268
Temperature (2–30 °C)	−0.01(0.02)	−1%	0.535	0.01(0.02)	+1%	0.491
Rain (0 –1)	−0.63 (0.28)	−47%	0.024	−0.34(0.28)	−29%	0.218
Wind (0.05 –5)	−0.34 (0.07)	−29%	<.0001	−0.23 (0.08)	−21%	0.005
Broom (yes vs no)	0.88 (0.19)	+141%	<.0001	0.05(0.20)	+5%	0.786
Machine saw (yes vs no)	0.13(0.26)	+14%	0.609	0.42 (0.24)	+52%	0.083
Local ventilation vs outdoor^b	0.41(0.67)	+51%	0.538	—	—	—
Natural ventilation vs outdoor	0.68 (0.24)	+97%	0.005	1.37 (0.25)	+294%	<.0001
Smoking (yes vs no)	0.19(0.26)	+21%	0.473	0.28(0.24)	+32%	0.251
Random effects
bwS^2 (naïve model)	0.67(1.50)			0.28(0.98)
wwS^2 (naïve model)	0.54(0.65)			0.38(0.45)
Inhalable cement dust
Intercept	−1.03 (0.47)		0.031	−1.33 (0.43)		0.005
Temperature (2–30 °C)	−0.01(0.02)	−1%	0.460	−0.01(0.02)	−1%	0.541
Rain (0 –1)	−0.75 (0.29)	−53%	0.010	−0.53 (0.27)	−41%	0.056
Wind (0.05 –5)	−0.31 (0.08)	−27%	0.0001	−0.17 (0.08)	−16%	0.034
Broom (yes vs no)	0.83 (0.20)	+129%	<.0001	0.05(0.21)	+5%	0.815
Machine saw (yes vs no)	−0.23(0.28)	−21%	0.413	−0.05(0.24)	−5%	0.843
Local ventilation vs outdoor^b	0.76(0.75)	+114%	0.310	—	-	-
Natural ventilation vs outdoor	0.96 (0.27)	+161%	0.0006	1.43 (0.28)	+318%	<.0001
Smoking (yes vs no)	−0.01(0.32)	−1%	0.975	0.27(0.34)	+31%	0.431
Random effects
bwS^2 (naïve model)	1.13(2.39)			0.64(1.77)
wwS^2 (naïve model)	0.57(0.68)			0.36(0.43)

---

‘Working with a broom’ increased concentrations of both inhalable dust and inhalable cement dust significantly by more than 100% in the total group, although this was not observed amongst the construction workers solely. Using a machine saw resulted in higher inhalable dust concentrations among construction workers, but cement concentrations were not increased. Subjects working indoors (with local exhaust ventilation or natural ventilation) seemed to experience much higher dust concentrations compared to workers working outdoors. Smoking during sampling had no noticeable effect on measured concentrations. Overall, working with a broom and outdoor wind speed were most prominent factors affecting inhalable dust and cement concentrations in the absence of job type.

Discussion

Personally measured concentrations of inhalable dust and its cement content varied over more than four orders of magnitude. Inhalable dust concentrations at the construction site ranged from 0.05 to 34 mg/m³, with a geometric mean (GM) concentration of 1.0 mg/m³. For inhalable cement dust the GM concentration was 0.3 mg/m³ (range 0.02–17 mg/m³). Here the highest levels were observed for concrete repairers and tile setters, whereas reinforcement and pouring work had the lowest. By far the highest concentrations were observed in cement production, particularly during cleaning operations. Personal inhalable dust concentrations in the ready-mix and pre-cast concrete plants were similar to concentrations during reinforcement and pouring work and were on average below 0.5 mg/m³ for inhalable dust and below 0.2 mg/m³ for cement. Elemental measurements showed highest but very variable cement percentages in the cement factory and very low percentages during reinforcement and pouring work. Analysis of exposure variability revealed that as expected temporal variability outweighed differences in average concentrations between workers in the majority of the job groups. In the cement plant, the ready-mix plants and during reinforcement and pouring work, workers appeared not to be uniformly exposed. This is probably a result of differences in tasks and work locations.

The measured concentrations for all cleaners at the cement plant exceeded the current German exposure limit for inhalable dust (10 mg/m³), except for one sample (8.8 mg/m³). These workers wore however face pieces as breathing protection while cleaning the majority of time. Among the other cement plant workers, concentrations of three out of 19 samples were above the limit. At the construction site only two workers exceeded the limit value: one floor screed layer using a grinding machine and a tile setter. For inhalable cement dust, the same persons and two additional cement plant workers exceeded the German limit value for Portland cement dust exposure (5 mg/m³).

Inhalable dust exposure has been monitored previously for several tasks in the US construction industry.² Exposures during concrete repair (GM 3.6 mg/m³; with a maximum of 12.9) were slightly higher than levels we found (GM 2.4 mg/m³; max. 6.2). Inhalable dust concentrations for concrete pouring were much higher than exposures we observed (2.5 mg/m³ versus 0.33 mg/m³). The subjects performing concrete pouring in our study were reinforcement workers as well, which is likely to have decreased the overall exposure level in this group. This indicates that it is hard to compare exposure levels of specific jobs between studies, as actual tasks performed might be quite different. Highest concentrations were however more similar for concrete pouring (2.6 mg/m³ versus 2.7 mg/m³). Recess milling is one of the tasks in the construction industry which is known to occur with high dust exposure levels.^13,14 In the current case, most of that type of work was taken care of in the pre-cast concrete plant. There was however some recess milling done in this project, performed by installation workers, but only for a limited amount of total work time. Installation workers were more involved in other tasks with low cement exposure. Recess milling is generally more frequently done in renovation projects. Demolition workers are known to be exposed to high dust levels as well.¹⁵ Measurements of these workers were not conducted in this study though, as no demolition work was done in this building project.

Observed variability of total dust concentrations in the cement plant (ranging from 0.35 to 52 mg/m³ (GM 3.0) for production tasks and even up to 194 mg/m³ for cleaners) is in line with previous findings among cement plant workers. In the USA a range of 0.01 to 79 mg/m³ with a GM of 2.9 was reported,¹⁶ and in a Norwegian study an exposure range of 0.4–54 mg/m³ and AM of 7.4 was found.¹⁷ Concentrations in the production areas of a Tanzanian cement factory were somewhat higher, but also showed considerable variability (GM 10.6 mg/m³, with a range of 0.21 to 229).¹⁸

In our study, temporal variability was larger than variability in personal average exposure for most jobs. For cleaners in the cement factory temporal variability was so extreme that differences in average concentration could not be detected. Workers were supposedly working together all the time, resulting in similar exposure levels which varied considerably from day to day. The temporal variability may be explained by different cleaning tasks and locations in the plant on various days. Other tasks in the cement plant also showed a rather high temporal variability in concentrations, which might be due to task circulation. The personal variability is however somewhat higher in this group. Working on different stages of the production process may lead to different average exposure concentrations. In the ready-mix concrete plants there was only limited variability for especially cement dust. Differences in ordered products could have caused this variability in concentrations from day to day. The large between-worker variability in average concentrations is not surprising, given that these subjects worked at five locations (plants). At the pre-cast concrete plant, working activities will not be exactly the same every day, due to special orders to be delivered. Variation was however not likely to be large since the (indoor) work environment remained relatively stationary, resulting in ₉₅R_ww′s of only 6 for inhalable dust and only 7 for inhalable cement dust. On the construction site, temporal variability was highest for floor screed layers (₉₅R_ww = 29), which may be explained by the intermittent use of tools. Using a grinding machine will lead to a different exposure than if sitting on the knees spreading cement. The highest personal variability was observed for reinforcement work and concrete pouring together, as these jobs were taken as one group. Tasks were however not equally distributed among the workers. One person may have spent more of his time on reinforcement work than the other, but also during the reinforcement work special tasks existed. As a consequence of this variety in tasks performed, materials used, distance to exposure sources and work style, differences up to 8-fold exist in average concentration. For example wood sawing may have led to high dust exposure, or standing next to a delivery truck may have caused diesel motor exhaust exposure. This might explain why between-worker variability was somewhat higher for inhalable dust than for cement content. When taking into account all construction workers, our findings are in concordance with observations in the US construction industry, where the personal component for silica exposures tended to be much larger than the temporal component.¹⁹ The overall variability in average concentrations between construction workers in our study was high (₉₅R_bw = 185): a logical consequence of the large differences between tasks. ‘Job type’ appeared to be the main determinant for (cement) dust exposure levels in this population, explaining over 80% of the variability in average concentrations. This is consistent with a previous study to quartz exposure among construction workers.²⁰

Riala (1988) found concentrations of 1 to 117 mg/m³ (AM 32 mg/m³) total dust in measurements (up to 4 hours) of cleaning tasks at a construction site.²¹ Dry sweeping was the dustiest work, particularly when it took place after demolition work. These levels were however much higher than presented here, ranging from 0.05–34 mg/m³ (AM 2.0 mg/m³) for all construction workers. Sweeping was a significantly exposure increasing factor when analysed among all subjects, but not so when the analysis was restricted to construction workers. This may be due to the relative small amount of time a broom was used at the construction site. A recent study on respirable dust exposure in bricklaying education institutions pointed however also towards sweeping as one of the major causes of high exposure levels.²² When measuring a more specific exposure (e.g.quartz exposure), type of material appeared to be the strongest determinant of exposure.²³ Unfortunately, in our study no detailed information was collected on handled material. We could therefore not include material as a potential determinant into our statistical models.

Based on the general composition of cement (65 w% CaO and 23 w% SiO₂), the ratio between these elements should have been approximately 4 for pure cement. In the cement plant this was on average the case (with a cement percentage of above 60%), but the large variability in these ratios is most likely related to other products present in the cement factory environment (e.g. blast furnace slag and limestone). Cement percentage and Ca/Si ratio were much lower in early phases of the construction project, when most of the pouring and reinforcement was done. Floor screed layers and tile setters had dust exposures with somewhat higher calcium-silicon ratios than other construction workers. This is most likely due to the actual mortar preparing and mixing these workers carried out. The considerably variation in cement percentages again points towards other dust sources, like diesel motor emissions, wood dust, sand and general air pollution. The calculated Ca/Si ratio confirmed this. Other sources are likely to have contributed particularly to dust concentrations at the construction site.

Conclusions

The highest inhalable (cement) dust concentrations at the construction site were observed for concrete repairers, floor screed layers and tile setters. In comparison these exposures were considerably lower than those found in a cement plant, where concentrations were particularly high during cleaning tasks. The main determinant for personal inhalable (cement) dust exposure appeared to be ‘job type’. Using tools, especially a broom, resulted in higher inhalable dust concentrations, whereas working outdoors, in particular with rain and wind, reduces exposure levels considerably.

Acknowledgements

This study was funded by a grant from the European Cement Association (CEMBUREAU http://www.cembureau.eu).

References

1. J. Mwaiselage, M. Bratveit, B. Moen and M. Yost, Variability in dust exposure in a cement factory in Tanzania, Ann. Occup. Hyg, 2005, 49(6), 511–9.
2. S. R. Woskie, A. Kalil, D. Bello and M. A. Virji, Exposures to quartz, diesel, dust, and welding fumes during heavy and highway construction, AIHA. J. (Fairfax., Va), 2002, 63(4), 447–57.
3. HSE. Portland Cement Dust. Criteria document for an occupational exposure limit. Sudbury: 1994.
4. HSE. Portland Cement Dust. Hazard assessment document. 2005. Report No.: EH75/7.
5. M. P. Purdue, B. Jarvholm, I. A. Bergdahl, R. B. Hayes and D. Baris, Occupational exposures and head and neck cancers among Swedish construction workers, Scand. J. Work Environ. Health, 2006, 32(4), 270–5.
6. C. Jansson, A. L. Johansson, I. A. Bergdahl, P. W. Dickman, N. Plato, J. Adami, P. Boffetta and J. Lagergren, Occupational exposures and risk of esophageal and gastric cardia cancers among male Swedish construction workers, Cancer Causes Control, 2005, 16(6), 755–64.
7. P. Mitropoulos and R. Norman, Occupational nonsolar risk factors of squamous cell carcinoma of the skin: a population-based case-controlled study, Dermatol. Online. J, 2005, 11(2), 5.
8. K. Sjodahl, C. Jansson, I. A. Bergdahl, J. Adami, P. Boffetta and J. Lagergren, Airborne exposures and risk of gastric cancer: a prospective cohort study, Int. J. Cancer, 2007, 120(9), 2013–8.
9. A. Dietz, H. Ramroth, T. Urban, W. Ahrens and H. Becher, Exposure to cement dust, related occupational groups and laryngeal cancer risk: results of a population based case-control study, Int. J. Cancer, 2004, 108(6), 907–11.
10. EN1232. Workplace atmospheres. Pumps for personal sampling of chemical agents. Requirements and test methods. 1997.
11. L. C. Kenny, R. Aitken, C. Chalmers, J. F. Fabries, E. Gonzalez-Fernandez, H. Kromhout, G. Liden, D. Mark, G. Riediger and V. Prodi, A collaborative European study of personal inhalable aerosol sampler performance, Ann. Occup. Hyg, 1997, 41(2), 135–53.
12. S. M. Rappaport, Assessment of long-term exposures to toxic substances in air, Ann. Occup. Hyg, 1991, 35(1), 61–121.
13. M. E. Lumens and T. Spee, Determinants of exposure to respirable quartz dust in the construction industry, Ann. Occup. Hyg, 2001, 45(7), 585–95.
14. E. Tjoe Nij, S. Hilhorst, T. Spee, J. Spierings, F. Steffens, M. Lumens and D. Heederik, Dust control measures in the construction industry, Ann. Occup. Hyg, 2003, 47(3), 211–8.
15. M. E. Lumens and T. Spee, Determinants of exposure to respirable quartz dust in the construction industry, Ann. Occup. Hyg, 2001, 45(7), 585–95.
16. H. L. Abrons, M. R. Petersen, W. T. Sanderson, A. L. Engelberg and P. Harber, Symptoms, ventilatory function, and environmental exposures in Portland cement workers, Br. J. Ind. Med, 1988, 45(6), 368–75.
17. A. K. Fell, T. R. Thomassen, P. Kristensen, T. Egeland and J. Kongerud, Respiratory symptoms and ventilatory function in workers exposed to portland cement dust, J. Occup. Environ. Med, 2003, 45(9), 1008–14.
18. J. Mwaiselage, M. Bratveit, B. Moen and M. Yost, Variability in dust exposure in a cement factory in Tanzania, Ann. Occup. Hyg, 2005, 49(6), 511–9.
19. S. M. Rappaport, M. Goldberg, P. Susi and R. F. Herrick, Excessive exposure to silica in the US construction industry, Ann. Occup. Hyg, 2003, 47(2), 111–22.
20. E. Tjoe Nij, D. Hohr, P. Borm, I. Burstyn, J. Spierings, F. Steffens, M. Lumens, T. Spee and D. Heederik, Variability in quartz exposure in the construction industry: implications for assessing exposure-response relations, J. Occup. Environ. Hyg, 2004, 1(3), 191–8.
21. R. Riala, Dust and quartz exposure of Finnish construction site cleaners, Ann. Occup. Hyg., 1988, 32, 215–220.
22. D. Huizer, T. Spee, M. Lumens and H. Kromhout, Exposure to respirable dust and silica in bricklaying education at Dutch vocational training centers, American Journal of Industrial Medicine, 2008 submitted.
23. E. Tjoe Nij, S. Hilhorst, T. Spee, J. Spierings, F. Steffens, M. Lumens and D. Heederik, Dust control measures in the construction industry, Ann. Occup. Hyg, 2003, 47(3), 211–8.

---