National Institute for Occupational Safety and Health

Summary of Dose Reconstruction Feasibility for Subcontractor Construction Trades Workers (CTWs) at the Savannah River Site (SRS) – SEC00103

Timothy D. Taulbee, PhD, CHPAssociate Director for ScienceJohn J. Cardarelli II, PhD, CHP, CIH, PEResearch Health Physicist

Advisory Board on Radiation and Worker Health MeetingApril 14, 2021

Key Dose Reconstruction Documents for Unmonitored Subcontractor Construction Trades Workers (sCTWs) DCAS-IG-006: Criteria for the Evaluation and Use of Co-Exposure

Datasets (March 2020)

– Section 2.0 Data Adequacy and Completeness• Completeness: “determine if there are sufficient measurements to

ensure that the data are either bounding or representative of the exposure potential for each job/exposure category at the facility”

• Guidance also indicates that there should be consideration for temporal gaps in the data and provides an example with respect to completeness from the Nevada Test Site (NTS)

ORAUT-OTIB-0081: Internal Dosimetry Co-Exposure Data for the Savannah River Site (September 2020)– 9 Radionuclide models for both CTWs and non CTWs

4/14/2021 2

Job Specific vs. Routine Bioassay Samples “The purpose of the job-specific bioassay sampling program is to collect bioassay

samples from workers whose routine bioassay program does not include some or all of the radionuclides present at the work site or who are not on a routine program.” (SRDB# 167757)

Most workers, including subcontractor Construction Trades Workers (sCTW), were on a routine bioassay schedule.

1997 DOE Notice of Violation (NOV) indicated only 21% (68/324) compliance of submitting job-specific bioassays. The workers who did not submit job-specific bioassays (79% or 256/324 workers) were followed-up and none indicated an internal exposure. (SRDB# 167497)

At SRS in 1997 there are over 6,000 routine non-tritium bioassays. This indicates that job-specific bioassays comprises a relatively small fraction (≈5%) of the overall internal monitoring program and likely has an insignificant impact on co-exposure models.

4/14/2021 3

Documentation Evaluating Subcontractor CTW Monitoring for Completeness and Representativeness RWP Analysis - ORAUT-RPRT-0092: Evaluation of Bioassay Data for

Subcontracted Construction Trade Workers at the Savannah River Site NOCTS Data Evaluation – ORAUT-RPRT-0094: Bioassay for Subcontractor

Construction Trade Workers at the Savannah River Site from 1972 to 1997 Plutonium Bioassay Logbook Analysis: 11,316 bioassay samples from

7,028 subcontractors CTWs between 1972-1990 Response to SCA Finding #3

Savannah River Site Plutonium Construction Trade Worker Stratification Refinement 2019 White Paper

Analysis of Uncertainty in Co-Exposure Models 2021 Bootstrap White Paper

Practical Implications of the Bootstrap Uncertainty Analysis on Co-Exposure Models 2021 Practical Implications White Paper

4/14/2021 4

Conclusion - Weight of the Evidence Weight of evidence from evaluations, stratification, & uncertainty analysis

– Robust subcontractor CTWs monitoring in the 1990s– Acceptable subcontractor CTW monitoring (>50%) in the 1980s– Limited monitoring data in the 1970s (DuPont CTWs are bounding)

We do not see any evidence where subcontractor construction trades workers were not monitored to a degree that would bias the current co-exposure models

Based on the weight of the evidence, NIOSH believes that the co-exposure models are bounding and representative of the exposures that would be received by an unmonitored subcontractor construction trades worker

NIOSH concludes that dose reconstruction is feasible

4/14/2021 5

Questions?

NIOSH December 2020 Presentation on Dose Reconstruction Feasibility

6

For more information, contact CDC1-800-CDC-INFO (232-4636)TTY: 1-888-232-6348 www.cdc.gov

The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.

Extra SlidesThese slides are for clarification purposes to assist in answering potential questions

Subcontractor Monitoring Limited assessment of 3200

bioassay requirements –33% compliance on Job-specific bioassay

Full assessment –“about21% compliance” on Job-specific bioassay ≈ 324 Job-specific ≈ 6481 total bioassay

1997 Total # of samples NOT

s

3200 Samples95%

5%

3.35%

1.65%

95%

107 Samples

67%, 79%

33%, 21%

s96.65%

3092 Sample

received was 256

100% follow-up of 256workers indicated no intake

SRDB# 16775712/9/2020 8

Subcontractor CTW monitoring in NOCTS Only NOCTS data (Claimant Data)

– 6097 Total SRS Claimants– 886 (15%) Subcontractor CTWs– Most, if not all, subCTW job titles are

Electrician26%

Pipefitter22%

Laborer9%

Other8%

Painterrepresented

NIOSH Evaluation– External Monitoring– Internal Monitoring

• Tritium bioassay• non-tritium bioassay (actinides)• Whole Body Counting (fission products)

3/23/2021 9

Why the simplistic internal monitoring approach?

Radionuclide specific internal monitoring depends on where the worker conducted their work (SRS is a Very Large Site)– Subcontractors in reactor areas likely didn’t need plutonium

monitoring, but may have needed tritium or fission products– Subcontractors in plutonium areas likely didn’t need tritium

monitoring– Subcontractors in tritium areas likely didn’t need plutonium or

fission product monitoring

Fundamental question: Are subcontractors sufficiently represented or bounded in the co-exposure models?

3/29/2021 10

Comparison RWP Analysis and NOCTS Claimant DataNOCTS Data from Table 5-4

12/9/2020 11

Effect of limited monitoring on Co-exposure models

Based on NOCTS data analysis (RPRT-0094) there are six years (1974-1979) where the percentage of internal monitoring via non-tritium and WBC is less than 50%

To bias the co-exposure models, exposures to unmonitoredsubcontractor CTWs would have to be significantly higher than the monitored subcontractor CTWs– Considering the zero-intake policy and defense in depth approach to

radiological protection, coupled with the Health Physics coverage of the construction jobs, we don’t believe this is plausible especially considering that for most years the monitored subcontractors outnumber the unmonitored subcontractors

12/9/2020 12

Subcontractor CTWs identified from SRS Pu bioassay logbooks

Decrease in Pu monitoring

Significant increasein Pu monitoring

260 198 159 136 129 71 56217

583 652 637538 507 517

1042

740

1860

1289

283

0

500

1000

1500

2000

1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990

7,028 unique subCTWs11,316 records b/w 1972-1990

12/9/2020 13

Approximate # Pu bioassay samples forsubCTWs by Area (1972-1990)

N=11,316

C area likely contains CS samples due its

location adjacent to Central Shops

1

250

438

932

62

1835

170

2822

528

1877

338

620

83224

13

668

32 983

322

0

500

1000

1500

2000

2500

3000

21 TC-1 A C CMX CS D F G H K L M P R S Tech TNX Z Missing

Central Shops (CS) area

signifies CTWs

Main Pu potential exposure area

12/9/2020 14

Plutonium Sub-stratification analysis

Over 95% of the plutonium bioassay data is below the reportable level of 0.1 dpm/day

Bioassay data from DuPont or prime CTWs (Roll 2) appear to be slightly greater than subcontractor CTWs (Roll 4, 5, and 6)

When co-exposure models are developed, the Pu intakes for the two populations are quite similar

12/9/2020 15

Sub-stratification Type-S Plutonium Intake Results

Yearprime-CTW

50th %(dpm/d)

subCTW50th %

(dpm/d)

prime-CTW95th %

(dpm/d)

subCTW95th %

(dpm/d)

1973-1978 15.71 6.97 268.7 169.4

1979-1987 26.38 22.65 279.2 326.1

Bold denote higher 50th% IntakesUnderline denote higher 95th% Intakes

12/9/2020 16

Analysis of Uncertainty (Bootstrap analysis of 1986 tritium co-exposure model confidence bands and density plots)

50

5

0.5

0.1-3 -2 -1 0 1 2 3

Standard Normal Quantiles

Dose

(mre

m)

sub-CTWs, and DuPont CTWs Confidence Bands for all CTWs,

2 4 6 8 10 12 1450th percentile (mrem)

500

400

300

200

100

0

Dens

ity

Density Plots of the 50th percentile

3/23/2021 17

Uncertainty Analysis of Tritium Co-exposure Models

Subcontractor CTWs exposures were generally lower than DuPont CTWs between 1972 and 1990.

There is no practical difference between the two groups and the currentcombined CTW model.

Geometric Mean with confidence intervals

1975 1980 1985 1990Year

70

60

50

40

30

20

10

0

Triti

um D

ose

(mre

m)

183/23/2021

Analysis of Uncertainty: Downward trend in tritium dose

Downward trend in

tritium doses between

1972 and 1990.

Typical of improved

radiological controls and

decreased exposure

potential over time.

95th percentile with confidence intervals

1975 1980 1985 1990Year

350

300

250

200

150

100

50

0

Triti

um D

ose

(mre

m)

3/23/2021 19

1972 1975 1980 1985 1990Year

80

60

40

20

0

Num

ber o

f wor

kers

(Ann

ual D

oses

)

Analysis of Uncertainty: Representativeness

Smaller number of subCTWs than DuPont CTWs in the 1970s

Larger number of subCTWs than DuPont CTWs in the 1980s

3/23/2021 20

NOCTS Percent subCTWs with dosimetry data and monitored for tritium exposures (adapted from Table 5-3)

0%

20%

40%

60%

80%

100%

72 74 76 78 80 82 84 86 88 90 92 94 96Year (1972 - 1997)

3/23/2021 21