Beryllium oxide

Epidemiological data

Administrative data

Endpoint:

epidemiological data

Type of information:

calculation (if not (Q)SAR)

Remarks:

Migrated phrase: estimated by calculation

Adequacy of study:

weight of evidence

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: Exposure estimates for early period 1935-1970 based on low number of samples.

Cross-reference

Reason / purpose for cross-reference:

reference to other study

Data source

Reference

Reference Type:

publication

Title:

Unnamed

Year:

2001

Materials and methods

Study type:

other: Estimate of exposure to Be

Principles of method if other than guideline:

Based on air measurements from different periods and with different methods a standardized exposure matrix was estalished for a Beryllium plant.

Test material

Test material information

Constituent 1

Reference substance name:

Beryllium oxide

EC Number:

215-133-1

EC Name:

Beryllium oxide

Cas Number:

1304-56-9

Molecular formula:

BeO

IUPAC Name:

oxoberyllium

Details on test material:

- Name of test material (as cited in study report): Beryllium (the authors don't discriminate between different chemical forms of beryllium in the text although they give a brief synopsis of the production process and the different beryllium forms arising in the plant)
- Substance type: metal
- Physical state: solid

Method

Details on study design:

METHOD OF DATA COLLECTION
Airborne beryllium concentrations had been measured using all-glass impingers, high volume air filters, and personals respirable and total dust samplers. The available samples are listed in Table 1. To provide consistency in exposure estimates over time, measurements collected by other monitoring methods were converted to approximate the most frequently used high-volume, time-weighted average measurements. Because exposure data were not collected in every year for all jobs throughout the duration of the case-control study (see Sanderson et al 2001 b), exposure estimates had to be extrapolated from existing measurements over time and across jobs.
A total of 7347 analytical results were available. Since 1971 quarterly DWA measurements are available, the geometric mean of year is taken for the job and year
Exposure data were assigned to 325 different jobs in 23 departments (based on records from 1953 to 1971).
- Details: see Table 1
The data then were used to establish a job matrix. A validity check was done by an experienced review panel and adjustment were made where needed.

STUDY PERIOD: (extrapolated 1935), 1947 - 1992


OTHER DESCRIPTIVE INFORMATION ABOUT STUDY: The study covers the beryllium facility at Reading (PA), USA

Exposure assessment:

estimated

Results and discussion

Results:

EXPOSURE
- Number of measurements: 7347
- Average concentrations: only reported for selected jobs, See Table II, ( DWA Exposure estimates, Types and form of Beryllium and potential concounders).

- Date(s) of measurement(s): 1947 - 1992, Quarterly monitoring since 1971.


FINDINGS
Exposures were variable from job to job, but regularly declining by period of observation (1953-1960, 1961-1970, 1971-1980, 1981-1992) at statistically significant levels.

Calculation of exposure over time for each case and corresponding control as in Table III (see attached document) with
- Cumulative exposure to beryllium compounds while employed at Reading (Sum of number of days x exp. DWA)
- Average exposure (cumulative exp. / days employed)
- Maximum or highest DWA held during tenure

The cumulative exposure correlates with tenure (r = 0.84)
Average and maximum exposure correlate (r = 0.90)
Average and maximum exposure do not correlate with duration of employment or cumulative exposure
Tentatively increasing exposures 2 or 10 times for periods before 1946 did not substantially change the case-control comparisons

The number of days of exposure, cumulative exposure, average exposure and maximum exposure to each of the types of beryllium (beryl ore, beryllium fluoride, beryllium hydroxide, beryllium oxide, beryllium-copper alloy- and each form of beryllium aerosol-dust, fume, or mixed-were calculated for each case and control. These exposure metrics were calculated using the same methods as used to calculate total beryllium exposures, except that job exposures only contributed to the exposure metrics for the various types and forms of beryllium associated with that job.
The number of days that the cases and controls were exposed to chemical agents which might confound or modify the observation of an associationbetween beryllium and lung cancer were also calculated. Because no measurement data were available for these agents, only the number of days the worker had been employed in jobs which mmight encounter exposure to the potential confounder.


INCIDENCE / CASES
not reported in this publication, but in Sanderson 2001b

OTHER OBSERVATIONS:
Housekeeping improved after 1960, OSHA implemented 1970

Confounding factors:

Type and form of Beryllium (ore, BeOH, BeF, BeO, BeCu) (fume or dust) mentioned but not taken into consideration, also other chemicals such as F, Cryolyte, Cu, Al, Cr, Ni, nitric acid and machine oil. No exposure data are available for these. .

Strengths and weaknesses:

There are relatively few measurements for the periods before 1970. The analytical methodology is not uniform and results had to be standardized.

Applicant's summary and conclusion

Conclusions:

Quantitative exposure data to different forms of beryllium ( including beryllium oxide) are available for 325 jobs in one plant (Reading, PA, USA). Results show that exposure levels were variable from job to job, but regularly declining by period of observation (1953-1960, 1961-1970, 1971-1980, 1981-1992) at statistically significant levels.

Executive summary:

For the Reading (PA, USA) plant and based on a total of 7347 analyses performed between 1947 and 1992 (6943 analyses between 1971 and 1992) with different methodologies the Daily Weighted Average (DWA) Exposure estimates, types and form of beryllium and potential concounders were calculated or extrapolated for 4 different periods (1953 -1960, 1961 -1970, 1971 -1980, 1981 -1992). For 325 different job descriptions exposure was established for cumulative beryllium exposure while employed at Reading. The number of days of exposure, cumulative exposure, average exposure and maximum exposure to each of the types of beryllium (beryl ore, beryllium fluoride, beryllium hydroxide, beryllium oxide, beryllium-copper alloy- and each form of beryllium aerosol-dust, fume, or mixed) were calculated for each case and control but ot stated in the manuscript. These exposure metrics were calculated using the same methods as used to calculate total beryllium exposures, except that job exposures only contributed to the exposure metrics for the various types and forms of beryllium associated with that job.

The number of days that the cases and controls were exposed to chemical agents which might confound or modify the observation of an association between beryllium compounds and lung cancer were also calculated. Because no measurement data were available for these agents, only the number of days the worker had been employed in jobs which might encounter exposure to the potential confounder.

Exposures were variable from job to job, but regularly declining by period of observation (1953-1960, 1961-1970, 1971-1980, 1981-1992) at statistically significant levels.

The cumulative exposure correlates with tenure (r = 0.84), average and maximum exposure correlate (r = 0.90).

Average and maximum exposure do not correlate with duration of employment or cumulative exposure.

The exposure data (not reported) were used in the companion study Sanderson et al. (2001b), a nested case-control study.