Dear Editor

The letter [1] by Professor Cherrie in this issue addresses the long-standing question as to whether static (area, stationary) samples can be used to estimate exposure to people in lieu of personal measurements for epidemiological investigations. In my previous letter [2] the question was ask “are personal and static samples related?” and Cherrie [1] answered this question as “yes”. As mentioned in my previous letter [2] and recently by Cherrie [3] there have been numerous studies [4-10] that evaluated the “relationship” between static and personal air measurements. The vast majority of studies [7] have suggested that there is no similarity in comparison of personal and static samples, although a few [8] have reported an association. However, these investigations directly compared concentration reported [7] by the two measurement methods and do not consider the effects of dilution in determining a ratio of concentrations. A study [3] by Cherrie of personal and static samples in which he evaluated investigations published in the Annuals of Occupational Hygiene during a ten-year period found a personal to static concentration ratio of 1.5. Another investigator [11] reported that when ventilation condition was applied as good, fair and poor, ratios became 3.2, 3.0 and 1.5, respectively. Cherrie’s study [3] along with others [4,8] that reported no statistical difference in concentration between the two-measurement methods support the concept of some relationship.

It must be noted that most reports comparing personal and static samples are in relationship to occupational exposures.[5] Occupational comparisons will be different from environmental exposures in that occupational measurements are usually collected at the source of contaminates (near field) and are commonly contained within a structured environment. However, there are exceptions in that occupational activities do occur in the general environment, such as on streets (e.g. parking enforcement officers).

It is generally agreed [1,7] that personal samples exhibit a higher concentration than static samples in the occupational environment. One explanation for a lack of concentration relationship between personal and static samples is the proximity to the emission source.[1,5] In general, workers are near the emission source with their movement around the work location. Static samples do not move from their placed location so they will not exhibit differences in “exposure” concentrations as would be associated with movement of a worker. Location of a static sampler in comparison to movement and activity of a worker has been suggested to greatly influence any relationship.[5] Other factors such as work practices, episodic releases, air streams, and ventilation will also have a major impact on the concentration collected by static samples and their subsequent comparison with personal sample concentrations.[1,5] If static samples are placed closer to the emission source they will exhibit a higher concentration than personal samples, which does not represent actual exposure to the worker.[12] Thus, placing of the static sample is possibly of greater importance than ventilation and size of the room. However, most static samples are placed outside the workers’ workstation so as not to interfere with their activities and will exhibit a lower concentration as a result of dilution caused by distance and ventilation as well as effect from room size and configuration.

A major factor to consider in comparison of these sampling methods that has not been well discussed is sample variability (geometric standard deviation - GSD). When examining variability (table) in comparison studies, static samples generally have a greater variability than personal samples, although this is not universally consistant.[10] Why static samples generally exhibit the larger variability is not clear; however, may be related to stratified zone samples having a smaller variability.[12] Greater variability provides less predictability of samples for future exposure. However, the large variability observed for occupational samples makes the increased variability in static samples somewhat moot in that there is a large day-to-day variation.[13]

Table Comparison of variability in studies reporting a relationship and no relationship for personal and static samples.

Key
x – asbestos samples, in f/cc
y – mg/m3 of respirable dust
a – with outliers
b - without outliers
c – no outliers

When comparing environmental and occupational exposure sampling these two differ in that environmental exposures are those which exhibit contaminates to the entire community or within a home type setting while occupational is strictly associated with employment activities in an occupation. However, for some conditions there will be a considerable overlap such as associated with persons working outdoors or locations that are frequented by the public (e.g. shopping malls).

Environmental exposure sampling from near sources will likely result in a similar relationship as seen with occupational sampling. Exposure from far sources will on the other hand likely be similar since the pollutant concentration will be more evenly dispersed due to mixing and distance. However, this assumes that there is no trapping or magnification of the contaminate in a building or location. These explanations provide some suggestions as to the variations observed between personal and static sampling and by including microenvironment monitoring and modeling from diary information as suggested by Harrison et al,[14] will allow establishment of a relationship. Some have pointed out that use of static samples to predict exposure for epidemiological evaluation is so flawed that environmental samples even after modeling are not suitable for predicting personal exposures.[15] These two publications [14,15] demonstrate the uncertainly associated with relating personal and static samples by occupational hygienists and subsequent use in epidemiological studies.

Since the source of contaminate is the same in occupational studies and static samples are collected from the same population of pollutants except experiencing a dilutant effect from the source there must be some relationship.[8] This relationship is likely related to the individual situations because of location of the source, work environment, worker activity, and static sampling conditions. Since epidemiological studies are interested in population effects and generally cumulative exposure, at least for chronic toxicants, a general relationship between sampling measurements is viable.[8] To better apply any relationship between these sample methods, inclusion of a ventilation condition [11] or other summary predictor of the sampling location would appear to be appropriate.

Cherrie [3] set forth a simple model for evaluating the relationship between personal and static samples for uniformly mixed contaminates, while Pundham et al,[11] incorporated ordinal conditions for establishing a ratio. Cherrie’s ratio for personal to static samples in occupational settings does not allow adjustment for variable conditions; thus, not permitting adaptation of his model to complex occupational epidemiology studies. The model proposed by Cherrie [1] should be modified for various occupational settings where different conditions of the workplace environment can be imputed. Thus, there needs to be a combination of the model proposed by Cherrie [1] and conditions of the work environment as suggested by Purdham et al.[11] However, development and future implementation of Cherrie’s model [1,3] signify a new era in the old concept of employing static samples for estimating exposure to affected populations.

For environmental epidemiology scenarios I agree that a relationship will exist as indicated by Harrison et al,[11] especially when evaluating far field contaminates. This is especially true for contaminates that have been well mixed in the environment. However, environmental static samples will suffer from the same problems as occupational sampling in that there will be cases of near sources (e.g. emissions from cooking devices in homes – a point source)[17] Relationship of personal and static samples in an environmental setting from a point source would be the same as that seen in occupational studies. Thus, Cherrie’s [1] model would be useful for these scenario’s as well.

As suggested by Cherrie [1] employment of personal samples has provided great insight to exposure, disease and evaluating controls in occupational health. His simple model is an excellent start to better understand the relationship between the two forms of sampling. As mentioned by Cherrie [1] there needs to be included a multitude of factors in predicting the relationship of these forms of sampling. Most have considered that there is no relationship and consider the subject matter not worth pursuing. For this I disagree. A better understanding of a relationship between personal and static sample concentrations will add insight to environmental and occupational epidemiology and will allow use of additional data in many epidemiological studies. As mentioned by Professor Sarin,[17] statistical modelling will be required, as started by Cherrie,[1,3] in establishing acceptable factor(s) of applying static samples as a representative measure of exposure to a study population. Clearly, additional research is needed to better understand the use of static samples in determining exposure. Hopefully, this letter and the one by Cherrie [1] will ignite such investigation.

References

(1) Cherrie JW. Personal and static samples are related. [electronic response to Lange JH Are personal and static samples related?] occenvmed.com 2003 http://oem.bmjjournals.com/cgi/eletters/60/3/224#66

(2) Lange JH. Are personal and static samples related? (Letter). Occup Environ Med 2003 60:224-225.

(3) Cherrie JW. The beginning of the science underpinning occupational hygiene. Ann Occup Hyg 2003;47:179-183.

(4) Lange JH, Thomulka KW. Airborne exposure concentration during asbestos abatement of ceiling and wall plaster. Bull Environ Cont Tox 2002;69:712-718.

(5) Lange JH, Kuhn BD, Thomulka KW, Sites SLM. A study of matched area and personal airborne asbestos samples: evaluation for relationship and distribution. Indoor-Built Environ 2000;9:192-200.

(6) Linch AL, Weist EG, Carter MD. Evaluation of tetraethyl lead exposure by personal monitoring surveys. Am Ind Hyg Assoc J 1970;31:170-179.

(7) Lange JH. A statistical evaluation of asbestos air concentrations. Indoor-Built Environ 1999;8:293-303.

(8) Lange JH, Lange PR, Reinhard TK, Thomulka KW. A study of personal and area airborne asbestos concentrations during asbestos abatement: a statistical evaluation of fibre concentration data. Ann Occup Hyg 1996;40:449-466.

(9) Sherwood RJ. On the interpretation of air sampling for radioactive particles. Am J Ind Hyg Assoc 1966;32:840-6.

(10) Seixas NS, Heyer NJ, Welp HAE, Checkoway H. Quantification of historical dust exposures in the diatomaceous earth industry. Ann Occup Hyg 1997;41:591-604.

(11) Purdham JT, Bozek PR, Sass-Korsak A. The evaluation of exposure of printing trade employees to polycyclic aromatic hydrocarbons. Ann Occup Hyg 1993;37:35-44.

(12) Corn M. Assessment and control of environmental exposure. J Allergy Clin Immunol 1983;72:231-241.

(13) Leidel NA, Busch KA, Lynch JR Occupational exposure sampling strategy manual. DEHW (NIOSH) Publication Number 77-173, National Technical Information Service Number PB-274-792. Cincinnati, Ohio: National Institute for Occupational Safety and Health, 1977.

(14) Harrison RM, Thornton CA, Lawrence RG, Mark D, Kinneisley RP, Ayres JG. Personal exposure monitoring of particulate matter, nitrogen dioxide, and carbon monoxide, including susceptible groups. Occup Environ Med 2002;59:671-9.

(15) Kromhout H, van Tongeren M. How important is personal exposure assessment in the epidemiology of air pollutants? Occup Environ Med 2003;60:143-144.

(16) Lange JH, Thomulka KW. Air sampling during asbestos abatement of floor tile and mastic. Bull Environ Cont Tox 2000;64:497-501.

(17) Sarin PS. Use of personal exposure modeling in risk assessment of air pollutants [electronic response to Kromhout H and van Tongeren M How important is personal exposure assessment in the epidemiology of air pollutants?] occenvmed.com 2003http://oem.bmjjournals.com/cgi/eletters/60/2/143-a#68

Dear Editor

Kromhout and van Tangeren [1] raise important issues regarding the papers by Cherrie [2] and Harrison et al.[3] The major shortcomings of the paper by Harrison et al.[3] are the small size of the sample (6 subjects each) used in the extrapolation of results. The three groups studied were the children, elderly and subjects with preexisting disease. The sample size in the disease category used only 2 subjects each with chronic obstructive pulmonary disease (COPD), left ventricular failure (LVF) and severe asthma. These sample sizes are rather inadequate to draw any correlation. Thus the paper by Harrison et al.[3] does not adequately represent generalized level of exposure of the individuals to carbon monoxide (CO), nitrogen dioxide (NO₂) and particulate matter (PM₁₀).

In risk assessment of ill-health association with air contaminants, uniform sampling of the pollutants at home, school, work and outdoor activities is important. In a recent article [4] detailing the effect of World Trade Center collapse, the authors emphasize that environmental monitoring for exposure assessment is a complex technical task that involves selection of pollutants for monitoring, location of monitors, sample collection methods, risk assessment standards, sampling results and data analysis. For example, the initial approach to the environmental sampling at the World Trade Center [4] was to locate monitors at the perimeter of the site, at locations where emergency and debris carrying vehicles were leaving the site, on the debris pile, and at locations in the surrounding community. Additional monitors were set up in the community to ensure a safer environment for workers and students returning to workplaces, homes and schools. Approximately, 66,000 results were entered in a database collected between September 11 to November 13 for subsequent analysis. Substances monitored included asbestos, particulate matter (PM_2.5 and PM₁₀), dioxins, PCBs, CO, heavy metals and volatile organic compounds (VOCs).

As pointed out by Schneider [5] deterministic models are developed from equations based on mathematical principles, while statistical models are developed by fitting to observed data.[6] In the statistical modeling of inhalation exposure, mixed effect models have been useful. Linear Mixed Effects Model with AR-1 autoregressive correlation structures has recently been used by Levy et al.[7,8] in their studies.

For example, due to the difficulties in the measurement of personal exposure, data on air pollution patterns in homogenous microenvironments linked with activity data are often used as surrogates (7). In these studies [7] PM_2.5 indoor-outdoor ratios were found to be greater than1 in settings with high levels of human activity. Cooking activities contributed significantly to elevation in levels of PM_2.5 along with other pollutants. Using Linear Mixed Effects Models with AR-1 autoregressive correlation structures, 10 minute average outdoor concentrations were generally weak predictors of indoor levels.

As mentioned by Levy et al.[8] although ambient particulate matter has been associated with ill-health, the health risks for individuals depend in part on their daily activities. Information about levels of PM size distributions in indoor and outdoor microenvironments can help identify high-risk individuals. The authors used Linear Mixed Effects Models with an AR-1 autoregressive correlation structure to evaluate statistical significance of differences between microenvironments. Levels of larger particles were generally higher near significant human activity, and levels of smaller particles were higher near combustion sources. The indoor PM₁₀ concentrations were reported to be significantly higher than the outdoors on buses and trolleys. Statistical models showed significant variability among some indoor microenvironments.

As pointed out by Chang et al.[9] simulation of activities performed by 65+ year olds indicate substantial variability in personal exposures of PM_2.5, O₃, CO and VOCs over a 12-hour period. For example, one hour personal CO exposures measured in vehicles were significantly higher than those measured in other microenvironments and the correlation between personal PM_2.5 exposures and ambient concentrations was lowest in the winter months for indoor non-residential microenvironment and highest in vehicle microenvironments.

Thus the conclusions drawn by Harrison et al. [3] from the data on a relatively small sample of subjects utilizing microenvironment modeling of CO, NO₂ and PM may not adequately reflect the overall exposure patterns. The variance observed between measured and modeled values for PM₁₀ among elderly, COPD and children could be minimized by taking measurements in a larger sample both indoors and outdoors and during summer and winter months.

References

(1) Kromhout H, van Tongeren M. How important is personal exposure assessment in the epidemilogy of air pollutants? Occup Environ Med 2003; 60: 143-144.

(2) Cherrie JW. How important is personal exposure assessment in the epidemiology of air pollutants? Occup Environ Med 2002; 59: 653-654.

(3) Harrison RM, Thornton CA, Lawrence RG, Mark D, Kinnersley RP, Ayres JG. Personal exposure monitoring of particulate matter, nitrogen dioxide, and carbon monoxide, including susceptible groups. Occup Environ Med 2002; 59: 671-679.

(4) Holtz TH, Leighton J, Balter S, Weiss D, Blank S, Weisfuse I, in Levy BS, Sidel VW (Eds). Terrorism and Public Health. A balanced approach to strengthening systems and protecting people Oxford University Press, Inc., Oxford, 2003, pp.19-48.

(5) Schneider T. Improving exposure assessment requires measurements and modeling. Scand J Work Environ Health 2002; 28: 367-369.

(6) Hornung RW, Griefe AL, Stayner LT, Steenland NK, Herrick RF, Elliott LJ et al. Statistical model for prediction of retrospective exposure to ethylene oxide in occupational mortality study. Am J Ind Med 1994; 25:825-836.

(7) Levy JI, Dumyahn T, Spengler JD. Particulate matter and polycyclic aromatic hydrocarbon concentrations in indoor and outdoor microenvironments in Boston, Massachusetts. J Expo Anal Environ Epidemiol 2002; 12: 104-114.

(8) Levy JI, Houseman EA, Ryan L, Richardson D, Spengler JD. Particle concentrations in urban microenvironments. Environ Health Perspect 2000; 108:1051-1057

(9) Chang LT, Koutrakis P, Catalano PJ, Suh HH. Hourly personal exposures to fine particles and gaseous pollutants-results from Baltimore, Maryland. J Air Waste Manag Assoc 2000; 50: 1223-1235.

Dear Editor

The scientific literature is full of of papers and ideas giving hypothesis. This was the objective of our letter. As it is stated at the end, we do not know if the finding is consistent or due to chance. We think that there are reasons pointing that our findings could have a solid fundament;

1) There were no musicians among the controls,
2) the musicians among the cases were wind musicians and,
3) the proportion of musicians in our region is extremely low.

It should be noted that we do not give any numeric estimation because it is not practical. As Dr Geis says, precisely the fact that there were no wind musicians among the controls reinforce our hypothesis, although confusion by smoking status can not be excluded. Since tobacco is responsible of around 85-90% of all lung cancers, only studies including never-smokers are completely free of confusion and then only this type of study, following the point of view of Dr Geis, could prove our hypothesis.

Dear Editor

The update of predictions of mortality from pleural mesothelioma in the Netherlands [1] provides welcome news that the peak number and the total during 2000-2028 is now predicted to be only a little more than half of the figures predicted only four years earlie.[2] This marked change in prediction has occurred because the known decrease in asbestos use after 1984 and a ban in 1993 were taken into account in the modelling, and there were five extra years of data (1994-98).

Since most mesotheliomas are caused by asbestos the pattern of use during different periods of time has a marked influence on the risk in cohorts whose working lives covered different periods. The marked effect of the discontinuation of crocidolite importation by 1970 into the United Kingdom on the amount of crocidolite found in the lungs of men, born in 1943 or later who developed a mesothelioma between 1990 and 1996 when aged 36 to 52 years,[3] was clear when compared with the lung contents of mesotheliomas in 1976 and 1977.[4] The average amount of crocidolite in the lungs of the men with mesothelioma in 1990-96 was about a tenth of that in the lungs of the 1976-77 cases. This difference was a consequence of the shorter period of exposure and the longer time since exposure during which elimination of fibres occurred in the absence of new exposure.

The importance of the effect of elimination is illustrated by results from former workers at the Wittenoom crocidolite mine and mill in Western Australia. Exposure of this group ceased in 1966 when the mine and mill were closed. Based on an analysis of the number of deaths with mesothelioma in men to the end of 1986, predictions were made of the number occurring up to 2020.[5,6] The model of the mesothelioma death rate used was that the rate increased with a power of time since exposure, moderated by a factor representing elimination of crocidolite fibres over time since exposure. Rates of elimination from zero to 15% per year were considered. Assuming no elimination, which was the usual model used at that time, predicted more than twice as many mesotheliomas by 2020 than an elimination rate of 15% per year. Preliminary results to 1999 were given by Musk et al,[7] and it is now known that the number of mesothelioma deaths in men in the period 1987 to 2000 was similar to the lowest predictions made based on the number up to 1986. This result is evidence that models of mesothelioma incidence that take account of a gradual elimination of crocidolite from the lungs after exposure are more realistic. There is strong evidence from other sources that such elimination does occur and that for crocidolite the rate of elimination is in the range of 10 to 15% a year.[8]

There is a high continuing toll from the use of asbestos in Europe and from the mining and milling of crocidolite at Wittenoom in Australia, but fortunately recent evidence strongly suggests that the number of mesotheliomas will not be as high as earlier predictions.

G. Berry 1, A.W. Musk 2,3, N.H. de Klerk 3,4, A. Johnson 5 and D.H. Yates 5

Affiliations

1 School of Public Health, University of Sydney, New South Wales 2006, Australia
2 Sir Charles Gairdner Hospital, Nedlands, Western Australia
3 Department of Public Health, University of Western Australia
4 Centre for Child Health Research, University of Western Australia
5 Dust Diseases Board (NSW) Research and Education Unit, Sydney

References

(1) Segura O, Burdorf A, Looman C. Update of predictions of mortality from pleural mesothelioma in the Netherlands. Occup Environ Med 2003; 60: 50-5.

(2) Peto J, Decarli A, La Vecchia C, Levi F, Negri E. The European mesothelioma epidemic. Br J Cancer 1999; 79: 666-72.

(3) McDonald JC, Armstrong BG, Edwards CW, Gibbs AR, Lloyd HM, Pooley FD, Ross DJ, Rudd RM. Case-referent survey of young adults with mesothelioma: I. Lung fibre analyses. Annals of Occupational Hygiene 2001; 45(7): 513-8.

(4) Berry G. Asbestos lung fibre analysis in the United Kingdom. Annals of Occupational Hygiene 2002; 46: 523-6.

(5) Berry G. Prediction of mesothelioma, lung cancer, and asbestosis in former Wittenoom asbestos workers. Br J Ind Med 1991; 48: 793-802.

(6) de Klerk NH, Armstrong BK, Musk W, Hobbs MST. Predictions of future cases of asbestos-related disease among former miners and millers of crocidolite in Western Australia. Med J Aust 1989; 151: 616-20.

(7) Musk AW, de Klerk NH, Olsen NJ, Hansen J, Ambrosini GL, Fritschi L, Merler E, Hobbs MST. Mortality in miners and millers of crocidolite in Western Australia: follow-up to 1999. Ann occup Hyg 2002; 46 (Suppl 1): 90-2.

(8) Berry G. Models for mesothelioma incidence following exposure to fibers in terms of timing and duration of exposure and the biopersistence of the fibers. Inhal Toxicol 1999;11:111-30.