Original Paper 207-

Effectiveness of Engineering Controls for Airborne Lead Exposure during Renovation/Demolition of a Commercial

Building John H. Langea Kenneth W. Thomulkab Training and Consultants, Pittsburgh, Pa., and Ub niversity of the Sciences in Philadelphia, Philadelphia, Pa., USA

Envirosafe a

Key Words Lead abatement . Occupational lead exposure. Lead regulations ; Industrial hygiene ; Source exposure variation ; Engineering/work practice controls

Abstract Airborne lead exposure was evaluated during renovation/demolition of a building that contained lead-based paint (LBP). During this work, different engineering and work practice controls were utilized for prevention and reduction of worker exposure to lead. These results suggest that the highest exposure occurred when no engineering or work practice controls were employed. Summary exposure levels under these conditions exceeded the Occupational Safety and Health Administration (OSHA) Permissible Exposure Level (PEL). When engineering/work practice controls, in the form of negative air machine filtration (NAM) and wet methods were used, airborne concentrations of lead were well below the OSHA PEL (50 μg.m ) and action level (30 μg.m -3 ). -3 of NAM without wet methods reduced Employment arithmetic mean exposure below the OSHA PEL and geometric mean below the OSHA action level. Occupational exposure concentration with NAM/wet methods was statistically lower when compared to no NAM/no wet methods and NAM/no wet methods. Exposure concentrations

exhibited a nonnormal distribution. Comparison of within and between-worker variability suggests that process/ environmental conditions (engineering/work practice controls) govern exposure. This study suggests employment of engineering and work practice controls for lead abatement and that process or environmental controls as compared to individual work practices/tasks are most important in controlling exposure. Study results support employment of NAM/wet method controls during renovation or demolition of building materials that contain LBP. Copyright© 2001

S.

Karger AG, Basel

Introduction

The hazardous nature of lead has been realized for thousands of years [1, 2]. Recently regulatory agencies begun establishing regulations to protect those considered most sensitive to effects of lead [3-5]. In the United States, these criteria have been extended to those in the occupational environment [5-7]. For many years in the United States the Occupational Safety and Health Administration (OSHA) had regulations for the nonconstruction lead industries and these were generally developed to prevent obvious signs and symptoms of lead poisoning [8]. However, more recently, OSHA and some

John H. Lange Envirosafe Training and Consultants PO Box 114022, Pittsburgh, PA 15239 (USA) +1 724 325 3360, Fax +1 724 325 3375 E-Mail [email protected] Tel.

states have

published regulations to protect occupationally exposed persons in the construction industry from low levels (biochemical hazards) of lead [5, 6, 9-11]. These requirements have been extended to the construction and demolition industries where exposure to lead is common, often occurs at high airborne concentrations and is currently not a well-recognized risk [3, 10, 12, 13]. Many buildings in the US and other countries contain lead in pipes, solder, and paint [3, 6, 10, 13]. During renovation and demolition of buildings, many of the components containing lead, including lead-based paint (LBP), are disturbed [4, 6, 10, 12, 14]. Such disturbances generate lead dust and debris, which can be taken up by workers through inhalation and ingestion (pica) [10]. Little information currently exists on the airborne exposure to workers conducting renovation or demolition activities

[4, 6, 10, 12, 13]. OSHA has set standards for exposure to airborne lead [6]. These standards include a permissible exposure limit (PEL) of 50 J-lg.m-3 and an action level (AL) of 30 J-lg.m-33 [6]. When these exposure limits are exceeded, various requirements for medical monitoring and employment of personal protective equipment must be initiated [6]. As the hazards of lead become better known to those in the building industries, information on exposure will become

of greater importance [3, 4, 10, 12, 13]. The present study was undertaken to determine

personal exposure levels to airborne lead associated with various engineering and work practice controls during demolition and renovation (including abatement of hazardous materials) of a multi-story building located in the eastern region of the United States. The building was constructed at the turn of the 20th century, and LBP and building components could be classified by visual observation to be in good condition [10]. Based on the demolition/renovation plan, as compared to a complete demolition, this building was also considered to be structurally sound and consisted of about ten floors. Area for the work was about 75,000 square feet in size. Part of the renovation activities was demolition of internal components in the basement area of the building that contained pipes and walls that had been previously painted with LBP. The components being demolished were mostly mechanical systems, walls and nonsupport structures. Other activities, including asbestos and lead abatement, were being undertaken on other floors of this building, but due to the structure of the building were isolated from work in the basement area.

...... ’ .

208

,,

~

.,

Air samples were collected from the breathing zone of workers performing renovation and demolition work. These data provided information on airborne exposure to lead during building activity under the different and varying engineering and work practice controls that were used at different time periods during the work. Such data will add to the current basic information for exposure to airborne lead during demolition or renovation activities [4, 10, 13, 15, 16]. Also included in this paper is information on use of engineering and work practice controls for lead abatement during demolition and/or renovation of buildings [4]. This report may possibly be the first that has investigated directly different types of engineering and work practice controls related to exposure during lead abatement work.

Materials and Methods All work was performed during the winter time period of 19992000. The contract for work described in this study involved only the basement area of a building. Work undertaken in other parts of the building were under separate contracts with other parties. Workers performing renovation received awareness training on the hazards of lead and had biological monitoring performed after starting this pro-

ject. Renovation practices employed included burning and cutting of pipes and removal (demolition) of walls that were painted with LBP. In the initial phase of this work, there were no engineering or work practice controls employed for reduction of dust and debris in the work area [4, 12]. After results of initial air monitoring were known, negative air machines using HEPA filtration (NAM) were incorporated into the work area with exhaust going to the outside environment (out of work area) [6]. It was estimated that NAM provided about 1-2 air exchanges per hour. During most of the project, no wet methods were employed and when pipes or other building materials/ structures were burned or cut, LBP was not

preremoved from the surface. Demolition of pipe-work was either through burning using a torch or cutting with a saw. Pieces of pipe were then disposed of as construction debris along with other building materials, such as dry wall. Demolition of walls was undertaken at the same time as both burning and cutting of pipes. The pipes were metal and walls were constructed of plaster, dry wall material and particleboard. Dry sweeping also occurred during this work activity. In the latter part of this project, wet methods (using tap water to mist/hose the area of work) [12] were employed on dust, debris and surfaces being removed. Analysis of paint chip samples using atomic absorption spectroscopy revealed that pipes and walls had a lead concentration of about 0.4 and 49%, respectively. The standard by US Department of Housing and Urban Development (HUD) for defining a material as LBP is 0.5 % or greater lead [ 10]. A single barrier consisting of a one-stage decontamination station, without a shower, was set up at the entrance of the work location. This decontamination facility was constructed from 6-mil polyethylene plastic and was primarily used to impede entrance by other trades into the area. Workers employed disposable ‘Tyvek’®( or simi-

lar clothing during their work activities and used respirators after about the first week of work. Respiratory protection employed was a negative-pressure half mask with HEPA filters (half mask) [6]. No precleaning was performed for any stage of the work. Debris, dust and residue were placed in dumpsters that were located outside the work area. Materials being disposed were not bagged and wet methods were not generally employed in association with disposal [10, 12]. Clean-up was performed mostly through dry sweeping. The exception was when wet methods were utilized toward the end of the project. This work was not identified by contract requirements, at least initially, as lead abatement and was undertaken by a demolition/renovation contractor. However, after initial personal exposure monitoring and paint chip/bulk sampling for lead, occupational concerns became apparent. Previously, the contractor performing this work did not have experience in LBP activities. The number of workers on-site ranged from between 5 and 12. Air samples were collected from the breathing zone of those workers (one-foot radius of the nose and mouth - personal samples) that were expected to have the highest exposure levels [6]. This generally included, individually or in combination, those cutting and burning pipes, demolishing walls and cleaning up. One or two air samples were collected per day. Each working day consisted of one shift of about 8 h. The work time period was about 50 days and samples were collected during approximately the last 40 days of work. Sampling data were segregated into activity time periods (groups/categories): no NAM and no wet methods employed for cutting (group A), NAM and no wet methods for cutting (group B), NAM/no wet methods for burning/torching activities on pipes (group C), NAM and wet methods for cutting (group D), and NAM no wet methods for demolition/ clean-up of debris (group E). Each group, except demolition/cleanup, also included in their schedule demolition and abatement of walls. Total (exposure for all workers) for all samples is listed as group F in the results. These categorizations (groups) are based on records collected by the on-site sampling technician, which included details as to what activities occurred for any given day. Summary data for all samples combined are also presented. An on-site technician, using low flow personal pumps, collected samples for the full shift, which was about 8 h [6]. Samples were collected with a closed face on a 37-mm mixed cellulose ester filter cassette (0.8 pm nominal pore size) [12] and analyzed by flame atomic absorption spectroscopy for lead by the NIOSH 7082 method [ 17]. Flow rate for sample collection was (nominal) 2 liters. min -1 and was determined using a calibrated rotameter [4, 12]. Detection limit for samples was established as 1.0 Jlg.m-3. Exposure results for airborne lead were reported as a time-weighted average (TWA) in Jlg.m-3 [6]. Control samples (blanks) were analyzed and were below the detection limit. Exposure data were reported as summary statistics (arithmetic mean - AM, geometric mean - GM, median, standard deviation SD, and geometric standard deviation - GSD) [4, 12, 13, 18-21]. Summary statistical analyses were determined using nontransformed data as previously described [4, 12, 19, 21 ] except as noted. GSD was calculated using natural logarithms (In) [4, 12]. Distribution, as nontransformed data, was evaluating using the Shapiro-Wilk W test for samples with 50 or less values and D’Agostino test for samples with greater than 50 measurements [22-24]. Total samples were also evaluated for distribution after being transformed (In) [ 12]. Outliers were evaluated using data in both a non-transformed and transformed form [4, 12] and employed the Grubbs test [25, 26]. Confidence intervals (CI), at 95%, for the AM, using nontransformed data, were

determined using a technique for nonnormal populations [27]. Comparison between groups and within-workers (group B, workers A, B and C) was performed using the Mann-Whitney U test [27]. A p value was provided for each comparison group, except for within comparisons. Significance was defined as a level less than 5%. A computer program

was

used for calculation of summary statistics and U test

[28]. Within-worker exposure was reportable for 3 workers and between-worker exposure was evaluated for the three within measurements (group B) [4, 21, 29]. Summary results were reported for within (groups B, C and D) and between-worker (group D) measurements [21, 29]. Within- and between-worker exposure are more accurately semi-within- and semi-between-worker distributions [21, 23]. Workers with three or more within measurements were employed for within evaluations. Due to the small number of measurements for within and between evaluation for outliers was not conducted. Determination of between-worker exposure as whether it is

homogeneous/monomorphic or heterogeneous was performed as previously described [21, 30]. A monomorphic group was defined, as an exposure population being within a range of ’0.5-2 times the group’ AM [30]. The between-worker GSD was also evaluated for assisting determination and employed a level of < 1.4 (GSD) for defining a monomorphic group [30]. GSD of within- and between-workers was employed to determine whether process/environmental conditions or individual worker practices/tasks was the source of exposure variability in this renovation activity [21, 29, 30]. If the largest GSD is represented by between-worker measurement individual practices/tasks are considered to be of primary importance as the ’determinant’ of exposure [30]. However, if the GSD for within-worker measurements was larger than between-worker, measurement process/environmental conditions would be suggested as the ’determinant’ of exposure [30]. This information provides a suggestion as to whether process/environmental conditions or individual worker practices/tasks is the primary source for exposure to workers [21, 29, 30]. Risk of overexposure was evaluated using a probability for exceedance by a graphical method as previously described [21, 29, 31]. Probability methodology employed GSD and AM in relationship of a fraction of the standard plotted on a figure [31]. This provides a graphical estimation for a confidence coefficient (probability) that at least 5% (or greater) of the average exposure concentrations exceed the set standard. Some values had probability values and fraction of exposure to standard that required extrapolation on the figure and has been listed as estimated. Standards employed in evaluating the probability of exceeding a ’set’ exposure level were the OSHA AL, OSHA PEL, protection limit afforded by a half mask respirator with HEPA filter (500 Ilg.m-3 TWA), and loose-fitting hood or helmet powered-air purifying respirator (hood) in continuous flow mode with HEPA filter ( 1,250 gg-m-3 TWA) [6]. All standards for comparison

were

obtained from the OSHA Lead Standard

as

referenced for

respiratory protection (29 CFR 1926.62, table 1) [6]. Results

outlying values (either nontransformed for total samples and each engineering/ transformed) work practice control group/category (table 1). OSHA does not employ a standard, as does HUD, for defining a There

were no

or

209

material

’lead-containing’ or LBP, but rather evaluates against the AL and PEL [6]. Blood samples from workers were collected about 15 days after this work was started and the highest blood lead level was reported to be 25 gg. dl- 1. Respirators (half mask with HEPA filter) were employed sporadically during the first few days of work. After LBP was identified as a building component and receipt of initial exposure results, respirators were used consistently. Work was started initially using no engineering or work practice controls (group A) but later employed engineering controls (NAM) (groups B, C and E), and finally incorporated work practice controls (wet methods) (group D). Exposure was found to be reduced as these controls were applied (table 1). Summary data for exposure results are shown in exposure

table 1. These data suggest that total exposure as measured by AM exceeded the PEL for airborne lead. However, GM for total samples was below the action level. When exposure was divided into different engineering or work practice categories (groups) [group A (no negative air machine filtration and no wet methods for cutting), group B (NAM) and no wet methods for cutting, group C

and no wet methods for burning, NAM with wet methods for cutting), and group D (NAM with wet methods for cutting)] differences among groups were found to exist both descriptively and statistically (table 2). The highest exposure level for cutting was associated with group A. Groups B and C are similar for AM and GM, although a descriptive difference is apparent for median values. When NAM is combined with wet methods for cutting, the exposure level is greatly reduced as compared to any other group. Since group E (NAM and no wet methods) has only one value, it cannot be compared with the other groups; however, this single measurement does suggest exposure greater than the OSHA PEL for airborne lead. Total samples had two exposure measurements that involved roof abatement and jack-hammering of floor material. Airborne exposure levels for jackhammer of floor and roof material abatement were 4.7 and 5.0 yg. m-3, respectively. Both of these activities had work associated inside the work area where renovation/demolition was being undertaken. A few samples were collected while only demolition/renovation of walls was conducted. Although it is difficult to separate these data due to various

(NAM

Summary data for exposure, in flg.m-3, to airborne lead various engineering or work practice controls

Table 1.

210

during renovation

activities with and without

occurring in a similar area, it is suggested that above the detection level did occur. exposure Distribution for all samples and groups is shown in table 3. Total sample distribution (group F) was nonnormal for data as nontransformed and normal when transformed. Three exposure groups (groups A, C and D) were activities

normal for distribution and this observed distribution is likely related to the small sample size. Group C was nonnormal at the 1 % level. The three groups (groups A, C, and D) identified by the Shapiro-Wilk test at a 5% level for normality in distribution also had GSDs which suggest their distribution approaches normality, except for group

B [21, 31]. comparisons between practice conengineering

Table 2. Statistical

various trols

and work

Statistical comparisons for groups are shown in table 3. All comparisons were statistically different at 5 % or less except one, group B vs. group C. The largest statistical differences were for those comparisons with group A and

Table 3. Distribution for various groups that evaluated and work practice controls

engineering

NS is not significant at 5% or a p value of Group A - no NAM/no wet methods for cutting; group B - NAM/no wet methods for cutting; group C - NAM/no wet methods for burning; group D - NAM/wet methods for 0.5.

cutting.

Table 4.

I

2

I

Nonnormal at 1 % level, but normal at 5 % level.

Summary data for within measurements of individuals and between measurements of these individuals

Worker A in group B is statistically different when compared to workers B and C of this same group. Workers B and C, in group B, are not statistically different when compared to each other.

211

Probability value for at least 5 %, or greater, by which mean employee exposure exceeded the comparison standard

Table 5.

() Comparison for determination of probability; AL represents a standard for exposure of 30 ~g ~ m-3 TWA; PEL represents a standard for exposure of50~g’m’~ TWA; half mask represents a standard for exposure of 500 pg. m-3 TWA; hood represents a standard for exposure of 1,250 gg. m-3 TWA. I Probability represents the confidence coefficient. 2 Best estimation for values by extrapolation from graph.

with group D. All summary exposure groups employing no engineering controls and/or with or without wet methods had statistically higher occupational smallest

were

exposure when

compared to group D. Within- and between-worker measurement variations are shown in table 4. GSDs for within-workers ranged from 4.0 to 1.4 and between-workers was 1.6. These data suggest that the larger variability is associated with within-worker exposure, although GSD for workers A and B in group D were 1.8 and 1.4, respectively. Comparison of within- and between-measurements suggest that exposure, as evaluated by variation, is more ’governed by the process or environmental conditions shared by all’ as opposed to individual practices/tasks [21, 30]. GSDs for within- and between-worker measurements are suggestive of a non-normal distribution. When wet methods were incorporated into the engineering/work practice controls, as shown for workers A and B, GSD was dramatically reduced. Group D exposure variability is similar to variability for between measurements. When this similar variability is considered, neither process/environmental conditions nor individual practices/tasks appear to be identifiable as the primary source for exposure variability [30]. Within-worker exposure groups that did not include wet methods are suggested to be more highly variable than 212

those employing wet methods [21, 30]. Between-worker variation suggests homogeneous/monomorphic grouping based on arithmetic mean in comparison to concentration range and GSD. Measurement for probability of exposure to the faction of a standard (either with AL, PEL, half mask or hood respirator) for each group and total samples is shown in table 5. These data suggest that the probability of exposure at a level of 5 % or greater of the comparison standard varies greatly. The largest exposure probabilities are for groups A, B and C. For group A this elevated probability for half mask is a result of the AM exposure concentration, and for group B and C are due to variation (GSD).

Discussion

This study reports on personal air samples collected during renovation/demolition of internal components of a commercial building (table 1). Sampling was conducted to evaluate effectiveness of engineering and work practice controls employed during various renovation/demolition activities [4, 12, 18] associated with disturbance of leadcontaining materials, particularly LBP. A previous investigation relating to asbestos-containing materials suggested that evaluation of hazardous material releases might not be best performed using only personal air sampling [32]. This is based on the premise that personal air samples are for ’compliance purposes and not part of a protocol in a scientific project’ [32]. Such an assumption is fallacious based on the concept that controls employed in abatement industries are primarily for occupational protection [4, 6, 12, 33, 34]. Although it is true that these controls also provide information for protection to the environment and public (public health) [5]. Previous studies have suggested that occupational exposure (personal) sampling and area sampling are not related [21, 31, 33, 35-43] and provide statistically different exposure concentrations [21, 33, 35]. Comparison studies have further shown that personal (occupational) measurements are often statistically greater in concentration than area samples [21, 33, 35]. Since the purpose of engineering/ work practice controls is for occupation exposure protection [4, 33], the best measure of this effectiveness is personal sampling [4, 12, 19, 21, 41]. Other suggested requirements [32] such as condition of building and/or material, its age, type of material and concentration of hazardous substance of interest, sampling information, filter collection description (e.g. pore size of collecting filter) and analysis protocol are of importance and necessary

to understand conditions and activities undertaken

[4, 12,

19, 21, 32, 33].. Distribution for airborne lead samples was generally nonnormal and appear to best fit a logarithmic form (table 2). These findings are similar to those reported for airborne lead [4, 12, 20], asbestos [19, 21, 29] and other occupational contaminants [ 18, 23]. Exposure results suggest that various concentrations occurred with different engineering and work practice controls (table 1). Employment of engineering or work practice controls in any form reduced occupational exposure levels, importantly to levels below the PEL. Previous studies that did not employ engineering or work practice controls have also reported exposures above both the OSHA PEL and AL during demolition activities for a building containing LBP [4]. Since there were no outliers detected, extreme values would not influence or require consideration in evaluation of these data. The importance of engineering and work practice controls in maintaining a low-exposure environment can be observed in the study data (tables 1, 2). NAMs have been identified in the HUD Guidelines as an optional engineering control method for lead abatement [ 10], although this guideline primarily references residential abatement. Such conclusion, at least for renovation- and demolitionrelated lead abatement, is not supported in this study and previous comparison investigations [4, 12], as related to exposure. Based on comparison data (table 2), when NAMs were employed as compared to no NAMs, there was a significant decrease in the summary airborne lead exposure concentration. A similar finding is also observed for employment of wet methods as shown in the comparison of groups B and D. Statistical differences exist as well for comparisons involving both burning abatement methods that employed no NAM and NAM/no wet methods (groups A and C) and wet methods with NAM (groups C and D). However, these comparisons must be considered with caution since they evaluate different work practices (cutting vs. burning). As NAM and wet methods are incor-

porated into engineering/work practice controls, exposure levels are statistically reduced in all cases compared. The only comparison not being statistically different was for groups B and C. These results support a previous finding [4, 12] that NAM and wet methods are applicable engineering/work practice controls that should be incorporated into lead abatement practices [ 13], particularly in renovation, demolition

or abatement activities. Within- and between-worker variation suggest that these measurements are nonnormally distributed based on GSD (table 4). Comparison of within- and between-

provide information on the source of exposure variability, as due either to process or environmental conditions (engineering/work practice controls) or individual practices or tasks [30]. From this information, the ’most’ important source of exposure can be identified [21, 29, 30]. Comparison of GSD for within- and be-

worker GSD

can

tween-workers suggests that process or environmental conditions are the ’dominant’ source of exposure variability. Unfortunately, direct evaluation can only be made for group B. Other studies of abatement for asbestos-containing materials [21, 29, 33] have reported similar findings that the process or environmental condition is the most important ’factor’ of the two possibilities for exposure variation. A lack of either process or environmental conditions or individual work practices/tasks being identified as the ’primary exposure variability’ for group D is likely a result of the lower exposure level and implementation of good exposure controls. Exposure data associated with group D also occurred after workers had gained experience in performing job requirements at this site. Reduction of exposure through employment in the work area of more stringent engineering or work practice controls (e.g. NAM/wet methods) support process or environmental conditions as a critical factor in exposure. Employment of engineering and work practice controls is strongly supported through comparison of group D with other groups.

Overall, exposure variability for between-workers ap-

proaches a monomorphic characteristic [30, 44-46], indicating a uniformity of exposure, although it does not exhibit a GSD of < 1.4. Variability for within-workers suggests a similar day-to-day variation of exposure for individuals [45]. Such findings suggest that exposure from one individual will not be ’highly’ dissimilar to another and can be used to estimate exposure for all members of the group or population. Although this finding, as in other abatement studies [21, 29], suggests a somewhat monomorphic (homogeneous) grouping, other studies involving abatement of hazardous materials, including lead, have not reported similar conclusions [4, 21 ]. Caution must be applied to this interpretation of these study data based on statistical differences of worker A (in group B) when compared to workers B and C. Evaluation of between- and within-worker distributions is important since it provides information on sampling strategies [30, 45, 47, 48]. However, the small number of samples and evaluation of only three groups in determining between-worker variation must be considered and caution applied in interpreting these data.

213

Probability values for at least 5 %, or greater, of workers being exposed above either the AL or PEL for lead in different exposure groups were highly variable (table 5). As control methods became more stringent, requirements for respiratory protection were reduced. This alone provided a strong basis for use of engineering and work practice controls [4, 12, 21, 29]. Previous studies [21, 29, 33] involving asbestos abatement support these findings of the importance of controls in reducing exposure and requirements of respirators. It has been suggested that respirator use at low occupational exposure concentrations I

may have a more detrimental effect on the worker than benefit [21, 29, 49-51]. This study along with others [4, 12, 14, 21, 29, 52] in abatement of hazardous building materials demonstrates the importance of engineering and work practice controls. A recent study [53] has also suggested that contaminant dispersion is also important

when

considering occupational exposure.

Evaluation of these results must be examined from an occupational exposure perspective and the relationship of this with current regulatory interpretation [21, 54]. Exposure data from this study, in part, suggest that lead levels of concern may exist during demolition, renovation, or abatement activities involving building components that have paint with lead levels below the HUD guideline criterion of 0.5%. This supports the basis of OSHA regulations requiring precautions [6, 13], mainly related to personal air sampling, for projects where any concentration of lead exists, not just for those over the HUD criterion. Final clearance was conducted by a visual method and no

surface samples were collected. Therefore, no conclusions can be made as to the effectiveness of controls relating to re-occupancy as identified by HUD. Since the building concerned in the present study is not a residential dwelling and is unlikely to be occupied to any extent by children, surface criteria as established by HUD may not be

particularly applicable. ’

Conclusions

The investigation described in this paper provides information on the importance of engineering and work practice (whether process or environmental) controls during renovation and demolition of building components containing lead [4, 12, 13, 52]. The results suggest that if appropriate controls are implemented, occupational exposure levels can be maintained below the OSHA action level [4, 12]. Comparison of within- and between-worker variation suggests that process is the most import source for exposure variation and this finding is supported by statistical evaluation of different control methods. These study data suggest the importance of applicable engineering or work practice controls for exposure reduction and would appear to be the most important factor associated with exposure during renovation and/or demolition of lead-containing building materials. Additional studies are warranted for evaluation of lead abatement, exposure and engineering and work practice controls related to this

industry.

References Lange JH, Kaiser G, Thomulka JW: Environmental site assessments and audits: Building inspection requirements. Environ Manage

1

1994;18:151-160. MO, Doll J, Klassen CD: Cassarett and Doulls’s Toxicology: The Basic Science of Poisons. New York, Pergamon Press, 1991. 3 Lange JH, Bruce KM, Johnson RL, Phillips MJ, Smith DM, Weidenboerner KM: Survey of lead-based paint abatement projects in public buildings of Erie and Crawford Counties, Pennsylvania, USA. Regul Toxicol Pharmacol 1998;28:73-78. 4 Lange JH, Sites SLM, Thomulka KW: Airborne lead exposure to workers during activity as part of a building demolition project. Indoor 2 Amdur

Built Environ 1998:7:210-216. 5 Ohio Department of Health: Lead abatement regulations. Ohio Administrative Code. Columbus, 1995, chapt 3742, 3701-32 and 370182.

214

6 US Occupational Safety and Health Administration : Lead exposure in construction: Interim final rule, 29 CFR 1926.62, Federal Register 58:26590-26649, Washington, DC, 1993. 7 Environmental Protection Agency: Lead; requirements for lead-based paint activities; proposed rule. 40 CFR 745. Federal Register 59,

Sept 2, 1994. Occupational Safety and Health Administration : Lead. 29 CFR 1910.1025, Washington, DC, 1987. 9 Rhode Island Department of Health: Rules and regulation for lead poisoning prevention (amended, May, 1993), Providence, 1993. 10 US Department of Housing and Urban Devel8

opment : Guidelines for the evaluation and control of lead-based paint hazards in housing,

Washington, 1995. 11 Mollar L, Kristensen TS: Blood lead as diovascular risk factor. Am J Epidemiol 136:1091-1100.

a car-

1992;

12

Lange JH, Sites SLM, Thomulka KW: Personal sample measurements of airborne lead during abatement procedures. Bull Environ Contam

Toxicol 1997;58:659-666. 13 Jacobs DE: Occupational exposures to leadbased paint in structural steel demolition and residential renovations. Intl J Environ Pollution 1998;9:1-14. 14 Lange JH, Gretz JB, Wade JW, Thomulka KW: A pilot study of a small scale lead-based paint abatement project that evaluated a blast machine which utilizes a soft recyclable plastic media: A case report. Fresenius Environ Bull

1993;2:671-676. 15 Zedd HC, Walker Y, Hernandez JE, Thomas RJ: Lead exposures during shipboard chipping and grinding paint-removal operations. Am Ind Hyg AssocJ 1993;54:392-396.

16

Conroy LM, Menezes-Lindsay RM, Sullivan PM, Cali S, Forst L: Lead, chromium, and cadmium exposure during abrasive blasting. Arch Environ Health 1996;51:95-99. 17 National Institute of Occupational Safety and Health: Manual of analytical methods, ed 3. DHHS (NIOSH), Publication number 84-100, Cincinnati, Ohio, 1984. 18 Burstyn I, Teschke K: Studying the determinants

29

30

31

of exposure: A review of methods. Am J

Ind Hyg Assoc 1999;60:57-72. 19 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;40:449466. 20 Lange JH, Thomulka KW: Occupational cxposure during lead abatement of steel surfaces by needle gun methodology. Bull Environ Contam Toxicol 2000;64:463-466. 21 Lange, JH Thomulka, KW: An evaluation of personal airborne asbestos exposure measurements during abatement of drywall and floor tile/mastic. Int J Environ Health Res 2000; 10: 5-19. 22 Shapiro SS, Wilk MB: An analysis of variance test for normality. Biometrika 1965 ;52 :591611. 23 Kumagai S, Kusaka Y, Goto S: log-normality of distribution of occupational exposure concentration to cobalt. Ann Occup Hyg 1997;41: 281-286. 24 Environmental Protection Agency: Statistical analysis of ground-water monitoring data at RCRA facilities (draft). Office of Solid Waste, Permits and State Programs Division, Washington, 1992. 25 Grubbs FE: Sample criteria for testing outlying observations. Ann Math Stat 1950;21:27-58. 26 Taylor JK: Statistical techniques for data analysis. Boca Raton, Lewis Publishers, 1990. 27 Daniel WW: Biostatistics: A foundation for analysis in the health sciences. New York, Wi-

ley, 1991. 28 McMillan College Software: ASP student copy: A statistical package for business, economics and social sciences. New York, Dellen/McMillan College Publishing, 1994.

32

33

34

35

Lange JH: A statistical evaluation of asbestos air concentrations. Indoor Built Environ 1999; 8:293-303. Gardiner K: Sampling strategies; in Harrington JM, Gardiner K (eds): Occupational Hygiene. Cambridge, Blackwell Science, 1995. 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. National Institute for Occupational Safety and Health, Cincinnati, 1977. Crossman RN, Williams MG: Quantification of fiber releases for various floor tile removal methods. Appl Occup Environ Hyg 1996; 11: 1113-1124. Lange JH, Kuhn BD, Thomulka KW, Sites SLM: A study of area and personal airborne asbestos samples during abatement in a crawl space. Indoor Built Environment, in press. Corn M: Assessment and control of environmental exposure. J Allergy Clin Immunol 1983; 72:231-241. Niven RMcL, Fishwick D, Pickering CAC, Fletcher AM, Warburton CJ, Crank P: A study of the performance and comparability of the sampling response to cotton dust of work area and personal sampling techniques. Ann Occup Hyg 1992;36:349-362.

36 Seixas NS, Heyer NJ, Welp EAE, Checkoway H: Quantification of historical dust exposures in the diatomaceous earth industry. Ann Occup

Hyg 1997;41:591-604. 37 Sherwood RJ: On the interpretation of air sampling for radioactive particles. Am Ind Hyg Assoc J 1966;32:840-846. 38 Stevens DC: The particle size and mean concentration of radioactive aerosols measured by personal and static air samples. Ann Occup

Hyg 1969;12:33-40. 39 Linch AL, Weist EG, Carter MD: Evaluation of tetraethyl lead exposure by personal monitoring surveys. Am Ind Hyg AssocJ 1970;31:170179. 40 Donaldson HM, Stringer WT: Beryllium sampling methods. Am Ind Hyg Assoc J 1980;41: 85-90. 41 Tebbens BD: Personal dosimetry versus environmental monitoring. J Occup Med 1973; 15: 639-641. 42 Scheeper B, Kromhout H, Boleij SM: Wooddust exposure during wood-working processes. Ann

43

Lange JH, Thomulka, KW:

Area and

personal

airborne exposure

during abatement of asbestos-containing roofing material. Bull Environ 44

Contam Toxicol 2000;64:463-466. Rappaport SM: Assessment of long-term exposures to toxic substances in air. Ann Occup Hyg

1991;35:61-121. 45 Kromhout H, Symanski E, Rappaport SM: A comparison evaluation of within- and betweenworker components of occupational exposure to chemical agents. Ann Occup Hyg 1993;17: 253-270. 46 Rappaport SM, Kromhout H, Symasski E: Variation of exposure between workers in homogeneous exposure groups. Am Ind Hyg AssocJ 1993;54:654-661. 47 Peretz C, Goldberg P, Kahan E, Goren A: The variability of exposure over time: A prospective longitudinal study. Ann Occup Hyg 1997; 41:485-500. 48 Tielemans E, Kupper LL, Kromhout H, Heederik D, Houba R: Individual-based and groupbased occupational exposure assessment: Some equations to evaluate different strategies. Ann

Occup Hyg 1998;42:115-119. 49 Raven PB, Dodson AT, Davis TO: The physiological consequences of wearing industrial respirators : A review. Am Ind Hyg Assoc J 1979; 40:517-534. 50 Louhevaara VA: Physiological effects associated with the use of respiratory protective devices. Scand J Work Environ Health 1984; 10: 275-281. 51 Lange JH: Health effects of respirator use at low airborne concentrations. Med Hypotheses

52

2000;54:1005-1007. Lange JH, Thomulka, KW: Occupational exposure

during lead abatement of steel surfaces by methodology. Bull Environ Contam

needle gun

Toxicol 2000;64:463-466. Welling I, Andersson I, Rosen G, Raisanen J, Mielo T, Martinen K, Niemela R: Contaminant dispersion in the vicinity of a worker in a uniform velocity field. Ann Occup Hyg 2000; 44:219-225. 54 Letter to the Editor: A log-normal distributionbased exposure assessment method for unbalanced bias. Ann Occup Hyg 1998;42:413-422. 53

Occup Hyg 1995;39:141-154.

215