This snapshot taken on 16/07/2006, shows web content selected for preservation by The National Archives. External links, forms and search boxes may not work in archived websites.

Department for Environment, Food & Rural Affairs

Expert Panel on Air Quality Standards
Airborne Particles: What is the appropriate measurement
on which to base a standard ?
A Discusson Document

Personal Exposure to Particles

Summary

Almost all studies of the health effects of particles have relied upon urban background measurements of particles as a surrogate for the exposure of all individuals in the population being studied. However, it is the actual exposure of the individual that determines any adverse effect, and better means of differentiating the exposures of the individuals studied would increase the ability of these studies to characterise effects and to detect thresholds.

A number of research studies have now been carried out and have shown that, in general, individual exposures to PM10 are higher than those recorded at central monitoring sites. It is known that small particles can penetrate effectively into buildings and in the absence of indoor sources the average particle concentration inside and outside are likely to be similar. However, individuals in their daily activities stir up a personal cloud of coarser particles, and some activities, such as cooking, housework and smoking, may also lead to generation of particles in the immediate environment of the individual. The static background sampler is not affected by such local activity. Personal measurements of PM2.5, which excludes the coarser particles, would be less influenced by the personal cloud.

Further studies of personal exposure are needed in order to develop methods for estimating indirectly the exposures of large numbers of people in populations and for differentiating the effects of particles derived from different sources.

Introduction

34. Epidemiological studies, discussed in paragraphs 88-129, have shown relationships between various indices of ill health and measurements of particle concentrations. These measurements have almost exclusively been made at one or more points in a city under study, and have been taken to represent the exposures of all individuals in that community. This may in part be responsible for the inability of such studies to demonstrate threshold concentrations below which health effects are undetectable; moreover it is probable that in any population certain individuals, because of their activities and proximity to sources, are more highly exposed to particles than are others (Watt et al 1995). There are several reasons for studying the exposure of individuals within populations. First, it is likely that harmful effects occur in those most highly exposed, and it is therefore important to define who these individuals are likely to be. Secondly, since there are sources of exposure in the home as well as in the outside ambient air, it is important to define the major sources of individual exposure and the relative toxicity of particles derived from these sources. Thirdly, individual exposure measurements (or estimates thereof) improve the ability of epidemiological studies to detect subtle effects and give greater power to studies targeted at specific vulnerable groups. In this chapter we review the published studies of personal exposure to particles.

Sources of Personal Exposure to Particles

35. An important source of personal exposure to particles is outdoor pollution from motor vehicles, chimneys, and other combustion sources, either while the person is outdoors or after the aerosol has penetrated into the indoor environment. Indoor sources, such as cigarette smoking, cooking and cleaning, are also important contributors to exposure. For some people exposure at work will provide the greatest contribution to personal exposure to airborne particles. Larger airborne particles (greater than about 10 µm in diameter) are lost from the air within minutes and are therefore likely to contribute more to exposure if the person is relatively close to the source of the aerosol, i.e, within two or three metres. This may occur during cleaning, moving materials or other activities that resuspend settled dust. Particles smaller than about 5 µm are likely to remain suspended in the air for a significant period of time and so may be carried further from the source and still contribute to exposure. Such particles may be produced indoors by smoking, cooking and other combustion sources. Smaller particles are predominantly lost from the air by deposition after diffusion towards surfaces and to a lesser extent by electrostatic attraction.

Measuring Personal Exposure

36. The procedures for monitoring personal exposure to airborne particles have been adapted from those used in the workplace. The equipment generally comprises a small battery-operated pump connected to a size-selective sampling head containing a filter paper or some other particle collection system. The main problems associated with using these techniques arise because the low airborne mass concentrations require sampling over relatively long time periods, typically 24-hours, in order to collect sufficient material to be weighed accurately. Because the pump should be worn by the individual throughout the sampling period, except during sleep, it must be lightweight and quiet and this limits the maximum airflow. As a consequence, the minimum detectable mass concentration is generally about 10 µg/m3 over 24-hours (Janssen et al 1998a).

37. During personal sampling the entry to the sampling head is normally located close to the individual's nose and mouth, typically on the outside of their clothing on a lapel. Sampling in this way gives an estimate of the concentration of particles inhaled, although for larger particles or where the person is very close to the source of the particles this may be biased towards over-estimation of the concentration. During the hours of sleep the sampler is generally located in the individual's bedroom. To ensure that the noise of the sampler does not disturb the person during sleep some investigators have devised sound insulated boxes to muffle the pump noise (e.g. Janssen et al 1998a). These systems may also be used at other times when the noise would be distracting, for example while watching television or reading.

38. A number of sampling heads may be used to measure personal exposure to airborne particles, the inlet determining the size fraction of particles sampled. Measurements made more than 15 years ago may have been made as "total" dust, although this is a misnomer since sampling heads which measured total dust generally had an undefined sampling characteristic which would have included some, but not all, of the larger particles suspended in the air. More recently measurements have been made in ways that are designed to be relevant to deposition of particles in the human respiratory tract, i.e. size-selective sampling heads. Soutar et al (1999) have used a PM10 sampling head where the entry into the head selects the inhalable fraction of dust and then a foam prefilter removes the remaining oversize dust so that the PM10 aerosol is collected on the sampling filter. Janssen et al (1998a) used an alternative design where the oversize dust is removed by an impaction system, again leaving the PM10 fraction of the aerosol to be collected on the sampling filter.

39. Personal samplers have recently become available for PM2.5 (Kenny, personal communication). There are also personal samplers available which can measure particle number concentration by means of light scattering (Frosig and Sherson 1997). However, they are less widely used because they do not directly measure the mass concentration of airborne particles and may possibly overestimate the true concentration. Finally, the UK Health and Safety Laboratory has developed a passive dust sampler which comprises a polymer layer with a permanent electric charge mounted in an earthed electrically conducting cage (Brown et al 1995). The sampler does not require a pump but attracts charged dust particles because of the high electrical field. It is not clear whether this sampler approximates to any of the sampling criteria defined above, although preliminary data suggest that there is a correlation between its measurement of airborne dust and those made using conventional inhalable samplers.

Exposure to Airborne Particles in Different Research Studies

40. A few studies, mostly carried out over the last ten years, have provided measurements of personal exposure to airborne particles. Many of these studies have been in North America, although more recently studies from European populations have been published, including two undertaken in the UK. Almost all of the data refer to urban or suburban populations and the studies generally exclude workers in known "dusty" trades; most measure exposure over a 24-hour period. From the principal studies in the literature there are, at the time of writing, just over 1800 measurements from either volunteers, children, elderly people, patients with chronic obstructive pulmonary disease (COPD) or patients with asthma. These data are summarised in Table 3.

41. In these studies the mean concentration of PM10 ranged from 42 µg/m3 in a study of patients with severe COPD (Linn et al 1996) to 114 µg/m3 for residents of Riverside, California (Pellizzari et al 1993). The second highest average exposure was obtained from a study of children (Janssen et al 1997). Almost all of these measurements were made over 24-hour periods and where this was not the case (Mark et al 1997) the data have been adjusted to estimate 24-hour averages by using the average fixed-point monitoring data to represent the unmeasured time. Other studies have shown that during the night indoor, outdoor and personal concentrations agree quite well (Pellizzari et al 1993) and so this adjustment seems justified. The maximum individual PM10 concentration was 971 µg/m3 (24-hour average), measured by Lioy et al (1990) in Philipsburg NJ.

42. In contrast, the mean PM2.5 or PM3.5 concentrations ranged from 24 µg/m3 in a Swedish study (Philips et al 1996) to 67 µg/m3 in the study undertaken by Philips et al (1997) in Italy. Figures 10 and 11 summarise the average personal exposure concentrations measured in these studies. Concentrations are approximately half of those reported for the PM10 fraction, reflecting the fact that some of the airborne dust measured as PM10 is excluded from the measurement as PM3.5 or PM2.5. In the study by Linn et al (1996), where there were data for both size fractions, the ratio of the average PM10 to PM2.5 personal exposure concentration was 1.5.

Table 3. Summary of studies where personal exposure to airborne particles has been measured.

Study Location Study Group Measurement Type Number of Subjects Number of Measurements
Sexton et al (1984) USA (Vermont) healthy non-smoking volunteers PM3.5 48 46
Spengler et al (1985) USA (Tennessee) healthy non-smoking adult volunteers PM3.5 101 249
Lioy et al (1990) USA (Philipsburg, NJ) healthy non-smoking adult volunteers PM10 14 189
Linn et al (1996) USA (Los Angeles) adult COPD patients PM10 and PM2.5 45 45
Silverman et al (1992) Canada (Toronto) adult asthmatics TSP and PM3.5 36 34
Pellizzari et al (1993) USA (Riverside, CA) representative selection of the non-smoking adult population PM10 178 171
Philips et al (1997) Italy (Turin) representative selection of the non-smoking adult population PM3.5 188 187
Philips et al (1996) Sweden (Stockholm) representative selection of the non-smoking adult population PM3.5 190 175
Janssen et al (1997) Netherlands (Amsterdam and Wageningen) healthy volunteer children PM10 45 301
Mark et al (1997) UK (Birmingham) healthy adult volunteers (30% were smokers) PM10 30 178
Janssen et al (1998b) Netherlands (Amsterdam) healthy non-smoking adult volunteers (aged 50 to 70 years) PM10 37 262
Seaton et al (1999) UK (Edinburgh and Belfast healthy non-smoking adult volunteers (aged over 60 years) PM10 111 111

Comparison of Personal Exposure with Fixed Location Monitoring Data

43. Many studies have compared personal exposure measurements with simultaneously collected data from outdoor fixed location monitors or samples collected in the subjects' homes. Ozkaynak and Spengler (1996) have discussed the theoretical relationship between outdoor and indoor particle concentrations and have shown that, in the absence of indoor sources, the concentrations will be lower inside buildings. In addition, the difference between indoor and outdoor concentrations will be greater for larger particles (10-2.5 mm) than for smaller particles (less than 2.5 mm). This difference is mostly due to the deposition of particles on surfaces within buildings rather than the failure to penetrate into buildings.

Figure 10. Mean PM10 exposure concentration in seven research studies.
Figure 10

 

Figure 11. Mean exposure concentration of PM2.5 or PM3.5 measured in six research studies.
Figure 11

44. The ratio of the mean personal exposure measurements to outdoor sample concentrations ranged from 0.8 to 3.7 for the PM10 data and from 0.8 to 2.4 for the PM3.5 data and for the one study which measured PM2.5 (Figure 12). In both cases the lowest value was obtained for a group of patients with severe COPD in Los Angeles, and they would have spent a considerable amount of time indoors and away from sources of airborne particles. The ratio of the mean personal exposure concentration to the indoor concentration ranged from 1.1 to 1.8 for PM10 and from 1 to 1.5 for PM2.5 or PM3.5.

45. There has been speculation as to why personal exposure levels are higher than indoor and outdoor concentrations. Pellizzari et al (1993) discussed four reasons for the discrepancies:

Figure 12. Ratio of mean personal exposure concentration to the corresponding mean indoor or outdoor concentration

Figure 12

46. The first possibility was tested by the authors and shown to be an unlikely explanation. Indeed this conclusion will apply to most studies, although if Tapered Element Oscillating Microbalance (TEOM) monitors are used to determine fixed site concentrations, the well known under-reading of the TEOM instrument may lead to discrepancies (e.g. Soutar et al 1999). Although Pellizzari et al (1993) observed skin flakes and fibres on sample filters they were judged to add less than 4 µg/m3 to the mass concentration and so the second alternative seems unimportant. The third possibility is plausible since persons may be exposed while in transit to and from work or during work. However, Pellizzari and his co-workers found lower exposures amongst those in their study who went to work compared with those who stayed at home. They concluded that the final alternative was most likely, cooking, cleaning and other activities producing high local concentrations of airborne particles, i.e. what has been described as the personal cloud. These authors noted that the elemental composition of the personal, indoor and outdoor samples were similar, which supports this contention. Moreover, in studies of people exposed to hazardous substances at work similar differences between personal and indoor samples are seen and these differences are generally attributed to the proximity of the worker to the source of the pollutant (Cherrie 1999).

47. Concentrations of airborne particles will be higher in cars, buses and other forms of transport when compared with urban background measurements. Bevan et al (1991) measured respirable particle concentrations (PM3.5) in Southampton for subjects commuting by bicycle. They compared their measurements with corresponding data obtained while cycling along a busy city centre street and while cycling around common parkland in the city (probably comparable to urban background levels). Exposure to respirable particles while commuting was approximately six times higher than obtained when cycling around the park and for cycling along the busy street the exposure was about nine times higher than that from the park location. Van Wijnen et al (1995) measured PM10 concentrations during a small number of car journeys in Amsterdam, although they do not provide comparable data from urban background monitoring sites. They found the 1-hour average concentration ranged from 14 to 194 µg/m3. Janssen et al (1998b) concluded from their studies of adults in the Netherlands that the PM10 concentration in a vehicle was approximately 130 µg/m3 higher than the urban background concentration.

48. Thatcher and Layton (1995) studied the likelihood of settled dust in indoor environments being resuspended. They showed that large particles were more likely to be resuspended than smaller particles and that particles less than 1 µm diameter could not be resuspended by normal activities, even by vigorous use of a vacuum cleaner. However, simply walking into a room and sitting down was observed to increase the concentration of 5-10 µm particles by more than threefold.

49. Personal exposure measurements are also generally more closely associated with indoor concentrations than with concentrations measured by outdoor fixed-site monitoring stations. The proportion of the variation in personal exposure from one individual to another explained by changes in the outside concentration measured at an urban background location has been investigated in eight of the studies cited above. In any group of people who were monitored, only a small part of the variation in personal exposure between individuals was generally explained by the outside concentrations; from zero to 45%, and in about half of the studies less than 10% of the variation was explained. The two highest values were found among children in the Netherlands and volunteers in Philipsburg NJ (Janssen et al 1997; Lioy et al 1990). Three studies investigated the proportion of variation in personal exposure levels between individuals explained by indoor home sample concentrations (Table 4). In each case, the indoor concentration measurements were a more reliable predictor of personal exposure than was the outdoor concentration. It is however plausible that the importance of indoor and outdoor concentrations would change with season, as we typically spend more time outdoors in summer than in winter.

50. Recent studies involving the use of continuous particle monitors both inside and outside dwellings have shown that, in the absence of indoor sources, indoor concentrations follow those out-of-doors quite closely, with a reduced concentration indoors and a variable lag time of the order of half an hour. In these circumstances indoor concentrations of PM10 tend to be around 50% of those measured out-of-doors, whilst higher indoor/outdoor ratios are found when moving to smaller particle sizes, i.e. PM2.5 and PM1. In occupied houses indoor concentrations derived from external particle sources are modifed by episodic increases in concentration due to activities such as cooking, smoking and physical activity in the home (Jones et al 2000). Nevertheless, outdoor monitors reflect changes in relative concentrations of those particles indoors which are derived from outdoor sources.

Table 4. Comparison of the proportion of variance in exposure explained by outside and indoor home concentrations.

Study Proportion of variance in personal exposure between people explained by
  outside concentrations indoor home concentrations
  (%) (%)
Janssen et al (1998b) 25 52
Sexton et al (1984) 0 29
Spengler et al (1985) 1 49

51. The poor observed correlation between personal exposure and outside concentrations measured at urban background locations on any specific day has led some scientists to suggest that the associations between ill-health and urban background concentrations may not be causal. It has been argued that the particles that are inhaled must be the trigger for any adverse effect and, as has been seen from the studies described above, typically less than 10% of the variation in personal exposure between individuals seems to be explained by the outside concentrations. Janssen et al (1997 and 1998b) investigated this issue by measuring the personal exposure of a number of children and then adults over several days. They found that the day-to-day changes in personal exposure to PM10 were much more closely related to the outdoor concentrations than were those of the group as a whole. For the children, the median proportion of the temporal variation in exposure to PM10 within individuals explained by the outdoor concentration was 40%, compared with 8% of the variation between individuals being explained by the outdoor concentration. The corresponding data for the adult study were 50% and 25% for the within and between individual variation, respectively. This may be because the differences in the sources of exposure for each person obscure the fact that as the outside particle concentration goes up or down the corresponding personal exposure will also increase or decrease. These authors argue that their finding supports the use of urban background measurements in time series epidemiological studies where it is the relative day-to-day changes in exposure of large urban populations that are important.

Information on the Chemical Composition of Personal Exposures

52. Two investigations have measured personal exposure to some of the aerosol chemical constituents, including acid aerosol, sulphate and ammonium (Brauer et al 1989) and sulphate and nitrate (Linn et al 1996). In both these studies the average personal exposure levels were less than the concentrations measured outside. The ratio of average personal to outdoor concentrations ranged from 0.4 (nitrate) to 0.9 (sulphate). These results are consistent with the absence of significant indoor sources for these chemical components and the deposition of aerosol indoors.

Modelling Personal Exposures

53. To date only one study has attempted to estimate individual exposures to particles indirectly, using a mathematical model based on measurements of urban background PM10, diary cards and multiple measurements of microenvironments (Seaton et al 1999). This study showed that it was possible to obtain valid estimates of exposure to PM10 in individuals taking part in an epidemiological study and to relate these to health end-points. This type of approach should enable estimates of exposure that are more reliable than those based solely on measurements at a central monitoring station. It is anticipated that similar estimates will prove possible for other fractions of airborne particles. No studies to date have investigated the relative toxicity of particles derived from different sources, such as cooking and passive smoking, included in the personal cloud to which individuals are exposed.

54. Exposure models might help identify appropriate strategies for individuals to reduce their exposure to airborne particles. These might include staying indoors with the windows and doors closed when outside concentrations are high, while at the same time eliminating or minimising indoor sources, such as cigarette smoking, cleaning and cooking using an oven or stove.

Conclusions

55. Personal exposure measurements averaged over 24-hours are clearly dependent on the activities undertaken by the individuals studied, their proximity to sources of aerosol and the characteristics of the indoor spaces they inhabit. In the absence of internal sources the concentration indoors will be lower than that found outside because of deposition of particles on internal surfaces. Indoor sources will produce higher exposure levels, typically increasing exposures by up to two or three times, averaged over 24-hours. Physical activity and cleaning will increase the exposure to particles with diameter greater than 1 µm but not to smaller particles.

56. The studies discussed above indicate that only a relatively small number of studies have measured personal exposure to ambient airborne particles, mostly either the PM10 (thoracic) or PM3.5 (respirable) fractions. The majority of these studies have surveyed non-smoking volunteers who do not work in dusty environments and it is possible that this may have biased the measurements towards the lower end of the distribution within the population. In the UK there are only two studies that provide relevant data and these are not easily compared because of differences in sampling strategy and the characteristics of the population recorded at the time of sampling. Some standardisation of methodology and data recording would be advantageous for future studies. Also, a central repository of non-occupational exposure measurements would provide a valuable resource for the future.

57. Day-to-day changes in the airborne PM10 concentration measured at an urban background site appear to correlate reasonably well with corresponding changes in personal exposure. This provides reassurance that epidemiological studies that have used measurements made at a central site as a surrogate for personal exposure are an appropriate way of investigating the association between particulate air pollution and health. However, the results should be interpreted with care since a 1 µg/m3 rise in concentration at an outdoor site would not necessarily correspond to the same increase in average personal exposure.

58. Clearly, reducing the concentration of airborne particles in the air of our cities will have an impact on the exposure of the whole population, but the magnitude of the reduction for individuals may not be great because of the differences in lifestyle and proximity to sources. Development of exposure models may provide a more reliable way of evaluating the impact of policy changes on people's exposure, and hence on the health of susceptible groups within society.

References

Bevan MA, Proctor CJ, Baker-Rogers J, Warren ND. Exposure to carbon monoxide, respirable suspended particulates, and volatile organic compounds while commuting by bicycle. Environ Sci Technol 1991; 25: 788-791.

Brauer M, Koutrakis P, Spengler JD. Personal exposure to acid aerosols and gases. Environ Sci Technol 1989; 23: 1408-1412.

Brown RC, Hemingway MA, Wake D, Thompson J. Field trials of an electret-based passive dust sampler in metal-processing industries. Ann Occup Hyg 1995; 39: 603-622.

Cherrie JW. The effect of room size and general ventilation on the relationship between near and far-field concentrations. App Occup Environ Hyg 1999; 14: 539-546.

Frosig A, Sherson D. Subject indoor and outdoor PM2.5 exposure profiles. Ann Occup Hyg 1997; 41 (supplement 1): 707-713.

Janssen NAH, Hoek G, Brunekreef B, Harssema H, Mensink I, Zuidhof A. Personal sampling of particles in adults: Relation among personal, indoor and outdoor concentrations. Am J Epidemiol 1998b; 147: 537-547.

Janssen NAH, Hoek G, Harssema H, Brunekreef B. Childhood exposure to PM10: relation between personal, classroom, and outdoor concentrations. Occup Environ Med 1997; 54: 888-894.

Janssen NAH, Hoek G, Harssema H, Brunekreef B. Personal sampling of airborne particles: method performance and data quality. J Exp Anal Environ Epidemiol 1998a; 8: 37-49.

Jones NC, Thornton CA, Mark D, Harrison RM. Indoor/outdoor relationships of particulate matter in domestic homes with roadside, urban and rural locations. Atmos Environ 2000; 34: 2603-2612.

Linn WS, Clark KW, Anderson KR, Gong H. Short-term particulate exposures and health changes in Los Angeles residents with chronic obstructive pulmonary disease (COPD). Proceedings of the second Colloquium on particulate air pollution and human health, Park City, Utah. 1996: 3.15.

Lioy PJ, Waldman JM, Buckley T, Butler J, Pierarinen C. The personal, indoor and outdoor concentrations of PM10 measured in an industrial community during winter. Atmos Environ 1990; 24B: 57-66.

Mark D, Upton SL, Lyons CP, Appleby R, Dyment EJ, Griffith WD, Fox AA. Personal exposure measurements of the general public to atmospheric particles. Ann Occup Hyg 1997; 41 (supplement 1): 700-706.

Ozkaynak H, Spengler J. The role of outdoor particulate matter in assessing total human exposure. In Particles in our air: Concentrations and health effects. Eds. Wilson R, Spengler JD. Boston: Harvard University Press, 1996.

Pellizzari ED, Thomas KW, Clayton CA, Witmore RW, Shores RC, Zelon HS, Peritt RL. Particle total exposure assesment methodology (PTEAM): Riverside, California Pilot Study - Volume 1. EPA/600/SR-93/050. Research Triangle Park: EPA, 1993.

Philips K, Bentley MC, Howard DA, Alvan G. Assessment of air quality in Stockholm by personal monitoring of non-smokers for respirable suspended particles and environmental tobacco smoke. Scan J Work Environ Health 1996; 22 (Supplement 1): 1-24.

Philips K, Howard DA, Bentley MC, Alvan G. Assessment of air quality in Turin by personal monitoring of non-smokers for respirable suspended particles and environmental tobacco smoke. Environ Int 1997; 23: 851-871.

Seaton A, Soutar A, Crawford V, Elton R, McNerlan S, Cherrie JW, Watt M, Agius R, Stout R. Particulate air pollution and the blood. Thorax 1999; 54: 1027-1032.

Sexton K, Spengler JD, Treitman RD. Personal exposure to respirable particles: A case study in Waterbury, Vermont. Atmos Environ 1984; 18: 268-280.

Silverman F, Hosein HR, Corey P, Holton S, Tarlo SM. Effects of particulate matter exposure and medication use on asthmatics. Arch Environ Health 1992; 47: 51-56.

Soutar A, Watt M, Cherrie JW, Seaton A. Comparison between a personal
PM10 sampling head and the tapered element oscillating microbalance (TEOM) system. Atmos Environ 1999; 33: 4373-4377.

Spengler JD, Treltman RD, Tosteson TD, Mage DT, Soczek ML. Personal exposures to respirable particulates and implications for air pollution epidemiology. Environ Sci Technol 1985; 19: 700-707.

Thatcher TL, Layton DW. Deposition, resuspension and penetration of particles within a residence. Atmos Environ 1995; 29: 1487-1497.

Van Wijnen JH, Verhoeff AP, Jans HWA, Van Bruggen. The exposure of cyclists, car drivers and pedestrians to traffic-related air pollutants. Int Arch Occup Environ Health 1995; 67: 187-193.

Watt M, Godden D, Cherrie J, Seaton A. Individual exposure to particulate air pollution and its relevance to thresholds for health effects: a study of traffic wardens. Occup Environ Med 1995; 52: 790-792.

[ Previous ] [ Contents ] [ Next ]

Published 17 May 2001
Expert Panel on Air Quality Standards
Air and Envirnonmental Quality
Environmental Protection Index
Defra Home Page