On May 2, 2002, HHS published its final rule on the guidelines to be used by the Department of Labor (DOL) to determine the probability that a cancer included in a claim under The Act was caused by workers' exposure to radiation during nuclear weapons production. DOL will make such a determination, determining "probability of causation," for each claim for which NIOSH is required to complete a radiation dose reconstruction. Generally, with some exceptions, these claims are for workers who are not a member of the Special Exposure Cohort.
Probability of Causation Final Rule
Final Rule: Guidelines for Determining the Probability of Causation--42 CFR 81
PDF 184 KB (20 pages)
Top of Page
NIOSH-IREP
To determine probability of causation for a cancer claim under The Act, DOL uses a computer software application NIOSH has developed in collaboration with the National Cancer Institute, called NIOSH-IREP. This computer software is a science-based tool that allows DOL to determine the probability a cancer was caused by a person's radiation dose from nuclear weapons production work. The actual outcome of a claim depends on a number of important factors such as the dose estimates of past exposures and the type of cancer that has been diagnosed. OCAS is responsible for maintaining and updating this software and the scientific elements (cancer risk models) that it uses.
NIOSH-IREP (on-line, interactive software program)
External Link:
http://www.niosh-irep.com/irep_niosh
User's Guide for the Interactive RadioEpidemiological Program (NIOSH-IREP)
- User's Guide
Version 5.5.3
January 2009 PDF 1 MB (22 pages)
NIOSH-IREP was upgraded to v.5.5.3 on January 5, 2009. Version 5.5.3 features
two minor changes: (1) Change of the title labeled "Claimant SSN"
on the input of the Excel template file, the main IREP input page, and the
summary page into "DOL Case No"; and (2) Rounding algorithm for dose
input in the dose summary table of the Summary Report. Now the dose input is
rounded to 3 significant digits and the exponential format is used when needed.
(Note: v5.5.3 includes no changes to cancer risk models and has no effect upon
PC results.)
- Attachments: Example input files that may be uploaded
into NIOSH-IREP. Page 7 of the User's Guide provides instructions for uploading
input files; summaries of the example files are provided on pages 20-22. To work
properly, the files must first be saved to a disk or to your computer. From there,
they can be uploaded directly into the IREP program.
Example 1a
(Microsoft Excel File; 169 KB)
Example 1b
(Microsoft Excel File; 169 KB)
Example 2a
(Microsoft Excel File; 169 KB)
Example 2b
(Microsoft Excel File; 169 KB)
Example 3
(Microsoft Excel File; 169 KB)
Example 4
(Microsoft Excel File; 169 KB)
Information regarding the development and use of the NIOSH-IREP program can be found in:
Verification of the NIOSH-IREP Computer Code Version 5.5.3 Report
This report presents the results of a formal verification effort organized to ensure that the most current version of NIOSH-IREP software (version 5.5.3)
calculates risk and probability of causation (PC) according to the methodology agreed upon by NCI and NIOSH. This verification process was performed by
individuals of SENES Oak Ridge, Inc. who were not involved in the initial development of the code. It is important to note that this effort has not attempted
to "validate" or question the models and procedures developed by NCI and NIOSH; this effort "verifies" that each part of the NIOSH-IREP code
operates according to its intended use described in its technical documentation.
Top of Page
Compensation Results by NIOSH-IREP Cancer Model
Compensation Results by NIOSH-IREP Cancer Model
PDF 944 KB (15 pages)
The compensation rates are based on DOL decision notices received by NIOSH through August 14, 2008.
Top of Page
Graphs Showing POC Values for Various Levels of Acute Radiation Exposure
Graphs Showing POC Values for Various Levels of Acute Radiation Exposure
PDF 96 KB (22 pages)
- What do these graphs illustrate?
These graphs are intended to illustrate the relationship between the probability of causation (PC) and acute exposure to ionizing radiation for the cancer models employed in IREP. Although these graphs are not in themselves useful for predicting the PC for an individual case, they can be used to examine the relative differences in radiation sensitivity among organs for two selected ages at diagnosis.
- What characteristics were used to generate these graphs?
These graphs show the changes in the probability of causation at various doses for two scenarios. The red line represents a person whose cancer was diagnosed at 40 years of age, and the blue line represents a person who was diagnosed at 50 years of age. In addition, the following parameters were used in the generation of these graphs:
- Male (except in the case of breast or ovarian cancer);
- One acute exposure to high energy (i.e. >250 keV) photons;
- The latency period (the time from the first occupational exposure to radiation to the time when the cancer was diagnosed) was 20 years for most of the solid cancers, but was 10 years for leukemia, bone or thyroid cancer;
- In case of lung cancer, the smoking history.
- What limitations do these graphs have?
As described above, these graphs depict the probability of causation for a specific scenario. Because of this, they can not be used to interpret the outcome of a specific case.
Top of Page
New Procedure for Resolving Cases Close to 50% Probability of Causation
New procedure for resolving cases in which the upper 99th percentile credibility limit of probability of causation is equal to or greater than 45% but less than 52% using the default simulation sample size of 2000 and default random number seed of "99"
(Effective June 6, 2006)
Previously, each case with an initial probability of causation (PC) value falling between 45% and 50% at the upper 99th percentile credibility limit (C.L.) was processed by increasing the simulation sample size to 10,000, choosing a new random number seed, and rerunning the case in NIOSH-IREP. The resulting upper 99% C.L. of PC obtained with a sample size of 10,000 determined the case outcome, supplanting the initial PC value that had been obtained with a sample size of 2000. This procedure was adopted in order to provide better statistical precision for cases approaching the compensation threshold of 50%.
To achieve even greater statistical precision for cases close to the compensation threshold, the following new procedure was adopted on June 6, 2006 and replaces the procedure described above.
For cases in which the initial PC is equal to or greater than 45% but less than 52% using the default sample size of 2000:
- The simulation sample size will be increased to 10,000.
- 30 additional IREP runs will be performed, using a new random number seed for each run.
- The average value (arithmetic mean) of the upper 99% C.L. of PC of the 30 runs will determine the case outcome.
- For cases with more than one primary cancer in which the initial PC calculated from the "multiple primary" equation is equal to or greater than 45% but less than 52%, 30 runs will be performed for each primary cancer per steps 1 and 2 above. The arithmetic mean of the upper 99% C.L. of PC of the 30 runs for each cancer will then be entered into the multiple primary equation. The newly calculated PC, based upon the arithmetic mean PC value of each cancer as entered into the multiple primary equation, will determine the case outcome.
The NIOSH-IREP User's Guide is currently being revised to reflect this procedural change.
Top of Page
Changes to the NIOSH-IREP Lung Cancer Risk Model
- Program Evaluation Plan: Modification of NIOSH-IREP Lung Cancer Risk Model: Impact of
"Combined" Lung Model on Non-Compensable Lung Cancer Claims
PDF 132 KB (5 pages)
Document Number: OCAS-PEP-008 Rev-00
About this Document: New document to evaluate the modification
of the NIOSH-IREP lung cancer model on
previously competed cases.
Approved: December 7, 2006
- NIOSH-IREP Version 5.5 was installed on February 28, 2006, replacing NIOSH-IREP v5.4. This updated version of NIOSH-IREP incorporates modifications to the lung cancer risk model for calculating probability of causation for cancers of the lung, trachea, or bronchus. Specifically, NIOSH-IREP v5.5 is programmed with two alternative lung cancer risk models. NIOSH-IREP v5.5 calculates separately the probability of causation produced under each risk model for each cancer of the lung, trachea, or bronchus, and reports the higher probability of causation of the two models as the probability of causation of record for the case. This risk model change was endorsed by the Advisory Board on Radiation and Worker Health, with the provision that NIOSH revisit the issue in approximately one year to determine if new evidence might warrant consideration of a single lung cancer risk model. Further details regarding this change, including PDF copies of all relevant documents provided to the Advisory Board on Radiation and Worker Health, can be accessed below.
NIOSH-IREP v5.5 also incorporates a bias correction factor for random errors in dosimetry for "never smokers" who were exposed to radon. Due to a programming oversight, this correction had been inadvertently omitted for "never smokers" and was applied only to smokers in earlier versions of NIOSH-IREP. NIOSH-IREP v5.5 corrects this error.
The modifications incorporated in NIOSH-IREP v5.5 pertain only to the "Lung (162)" risk model and apply only to cancers of the lung, trachea, or bronchus. In conjunction with these revisions, NIOSH will review all relevant previously completed cases that have not been compensated to identify those for which the changes are applicable, and will re-evaluate the cases using the new guidelines. NIOSH will also apply the new guidelines to all currently active cases and any future cases. This may result in the Department of Labor calculating higher probability of causation determinations for select lung, trachea, or bronchus cases among previously decided and current EEOICPA cancer claims. (Note: It will not result in lower probability of causation determinations for any case.)
NIOSH invites public comments on this change to the NIOSH-IREP lung cancer risk model. NIOSH will fully consider comments received and, on such basis, may reconsider the change or consider further revisions to the lung cancer risk model, as appropriate. Comments may be submitted by mail or e-mail. Mail comments concerning this change to Larry Elliott, Director, Office of Compensation Analysis and Support, National Institute for Occupational Safety and Health, 4676 Columbia Parkway, Mailstop C-46, Cincinnati, OH 45226. Submit electronic comments, subject-titled "NIOSH-IREP Lung Cancer Model," to ocas@cdc.gov.
- Background Information: Proposed Modification of NIOSH-IREP Lung Cancer Risk Model (October 2005)
NIOSH proposes to install the NIH-IREP lung cancer risk model into NIOSH-IREP, but to retain the current NIOSH-IREP lung model as well. Under this proposal, EEOICPA lung cancer cases would be processed by running each case under both IREP models, separately, but NIOSH-IREP would report only the set of results associated with the higher probability of causation at the upper 99th percentile credibility limit. Thus, the higher of the two values at the upper 99th percentile would determine compensability. Should this NIOSH-IREP change be implemented, all previously completed non-compensable EEOICPA lung cancer cases would be reevaluated.
This proposal is based in part on recommendations obtained from four outside experts who agreed to review the IREP lung models. NIOSH concludes that the points raised by reviewers warrant erring on the side of EEOICPA claimants by adopting the methodology described above.
A letter announcing this proposed change, along with eight supporting documents including the four expert opinions, was delivered to ABRWH members in late September 2005. These documents, taken together, provide a detailed explanation and chronology of the factors leading to this proposal. Links to the nine documents are provided below:
NIOSH/OCAS. Notice of intent to change the NIOSH-IREP lung cancer risk model for estimating probability of causation under EEOICPA
(letter from NIOSH to ABRWH).
PDF 27 KB (3 pages)
Land CE and Pierce DA. Likelihood profile for parameter alpha used in computation
of statistical uncertainty for ERR/Sv in NIH-IREP lung
cancer model (identified in letter to ABRWH as Enclosure
#1).
PDF 36 KB (5 pages)
Apostoaei AI and Trabalka JR. Differences in the estimation of lung cancer
risk between NIOSH-IREP and NIH-IREP (identified in letter
to ABRWH as Enclosure #2).
PDF 228 KB (27 pages)
NIOSH/OCAS. Evaluation of NIH-IREP lung cancer risk model for application to
NIOSH-IREP (NIOSH instructions to outside reviewers, identified
in letter to ABRWH as Enclosure #3).
PDF 21 KB (4 pages)
Comments from Faith G. Davis, PhD, Professor, Division of Epidemiology and
Biostatistics, University of Illinois at Chicago School
of Public Health (identified in letter to ABRWH as Enclosure
#4).
PDF 16 KB (5 pages)
Comments from Jonathan M. Samet, MD, MS, Professor and
Chairman of the Department of Epidemiology, Johns Hopkins
University School of Public Health (identified in letter
to ABRWH as Enclosure #5).
PDF 114 KB (5 pages)
Comments from David B. Richardson, PhD, Assistant Professor of Epidemiology,
University of North Carolina School of Public Health (identified
in letter to ABRWH as Enclosure #6).
PDF 29 KB (7 pages)
Comments from David J. Brenner, PhD, Professor of Radiation Oncology and Public
Health, Columbia University School of Public Health (identified
in letter to ABRWH as Enclosure #7).
PDF 33 KB (4 pages)
SENES Oak Ridge Inc. How should NIOSH estimate risk of lung cancer in workers
covered under EEOICPA in the face of uncertainties in
the interaction between smoking and low-LET radiation.
(Summary and analysis of expert comments, identified in
letter to ABRWH as Enclosure #8).
PDF 121 KB (18 pages)
- Changes to the NIOSH-IREP Lung Cancer Risk Model: Request for comments regarding a change to a scientific element underlying the determination of probability of causation under EEOICPA
Changes to the NIOSH-IREP Lung Cancer Risk Model
Under the Energy Employees Occupational Illness Compensation Program Act of 2000
(March 24, 2006)
PDF 52 KB (2 pages)
NIOSH has changed a guideline for determining the probability of causation under the EEOICPA for energy employees with cancers of the lung, trachea, or bronchus. The change affects only the NIOSH-IREP cancer risk model termed "Lung (162)." The new guideline, which became effective on February 28, 2006, with the introduction of NIOSH-IREP Version 5.5, requires the use of both a National Institutes of Health (NIH)-IREP lung model implemented by NIH in 2003 and the original NIOSH-IREP lung model implemented by NIOSH in 2002. NIOSH-IREP Version 5.5 calculates separately the probability of causation produced under each model for each cancer of the lung, trachea, or bronchus. The result from the model that produces the higher probability of causation at the upper 99th percentile credibility limit is reported as the probability of causation result of record for the case. NIOSH-IREP Version 5.5 also incorporates a bias correction factor for random errors in dosimetry for those energy workers who had not smoked cigarettes ("never smokers") and who were exposed to radon. This correction was previously applied to smokers, but had been inadvertently omitted for never smokers. These changes may result in DOL calculating higher probability of causation determinations for select cases of cancer of the lung, trachea, or bronchus among previously decided and current EEOICPA cancer claims. The changes cannot result in any lower probability of causation determinations. Although this change to the NIOSH-IREP lung cancer risk model took effect February 28, 2006, NIOSH will fully consider all comments received regarding this change and may reconsider this change or consider further revisions to the lung cancer risk model based on public comment.
NIOSH invites written comments on the above changes from interested parties. Any public written comments must be received on or before May 23, 2006. Comments can be submitted to Larry Elliott, Director, NIOSH OCAS, 4676 Columbia Parkway, Mailstop C-46, Cincinnati, Ohio 45226. Submit electronic comments, titled "NIOSH-IREP Lung Cancer Model," to ocas@cdc.gov.
Top of Page
Program Evaluation Reports (PERs) and Program Evaluation Plans (PEPs)
NIOSH is committed to applying the best available science in dose reconstructions. In keeping with this commitment, completed cases with probabilities of causation less than 50% are reviewed as relevant new information becomes available. The results of these reviews are described in a Program Evaluation Report (PER). The PER details the effect, if any, of the new information on the completed dose reconstruction. If it appears that the compensability of a completed dose reconstruction may be effected, NIOSH is committed to working with the Department of Labor to reopen and rework the dose reconstruction, as appropriate. A Program Evaluation Plan (PEP) describes plans for evaluating specific program details or issues.
- PER 16: Implementation of IREP Procedure
for Claims near 50% Probability of Causation
PDF 17 KB (4 pages)
Document Number: OCAS-PER-0016 Rev-00
About this Document: New document to evaluate the effect of implementing a new IREP procedure on previously completed claims.
Approved: September 25, 2007
Summary: A total of 109 previously non-compensable claims with PC values of 45% or greater were evaluated. The average PC value remained below the 50% compensation threshold for each of the 109 claims. An itemized list of claims was provided to DOL containing the final evaluation result for each of the 109 claims.
- Program Evaluation Plan 8: Modification of NIOSH-IREP Lung Cancer Risk Model: Impact of "Combined" Lung Model on Non-Compensable Lung Cancer Claims
PDF 132 KB (5 pages)
Document Number: OCAS-PEP-008 Rev-00
About this Document: New document to evaluate the modification of the NIOSH-IREP lung cancer model on previously competed cases.
Approved: December 7, 2006
- Program Evaluation Report 9: Target Organs for Lymphoma
PDF 17 KB (2 pages)
Document Number: OCAS-PER-009 Rev-00
About this Document: New document to change target organs for lymphoma.
Approved: March 8, 2007
Summary: In February, 2006, OCAS determined that the internal and external dosimetry target organs used for several forms of lymphoma should be changed. The detailed rationale for this decision is described in OCAS-TIB-012. The change resulted from a detailed investigation by OCAS of the etiology of lymphoma.
Additional PERs and PEPs have been developed on various topics and/or sites. To view all PERs and PEPs that have been developed, please refer to the Program Evaluation Reports (PERs) and Program Evaluation Plans (PEPs) page.
Top of Page
Submitting Comments on NIOSH-IREP
The public may comment on NIOSH-IREP at any time. Comments can be sent electronically by e-mail to ocas@cdc.gov. The preferred formats for electronic documents sent by e-mail are Word or WordPerfect. Comments can be mailed to:
NIOSH-IREP Comments
National Institute for Occupational Safety and Health
Office of Compensation Analysis and Support
4676 Columbia Parkway
Mailstop C-45
Cincinnati, Ohio 45226
Top of Page
Subject Matter Expert Review of NIOSH-IREP
- Comments from David G. Hoel, Ph.D., University Professor, Medical University of South Carolina.
PDF 157 KB (6 pages)
- Comments from Richard Hornung, University of Cincinnati.
PDF 157 KB (4 pages)
- Comments from David Richardson, Department of Epidemiology, School of Public Health, University of North Carolina. (revised on 2/11/02)
PDF 52 KB (6 pages)
- Comments from Roy E. Shore, Ph.D., Dr. P.H., Professor, Department of Environmental Medicine, Director, Division of Epidemiology and Biostatistics, New York University, School of Medicine.
PDF 164 KB (4 pages)
- Comments from Daniel O. Stram, Ph.D., Associate Professor, Keck School of Medicine, Department of Preventive Medicine, Division of Biostatistics, University of Southern California.
PDF 439 KB (5 pages)
- Response to Subject Matter Expert Review Comments on NIOSH-IREP (May 2002)
PDF 917 KB (27 pages)
Top of Page
Other Related Materials
- NIOSH's Evaluation of the Radiogenicity of Non-Presumptive Cancers
Report to the Senate Appropriations Committee on the Radiogenicity of Specific Cancers under the Energy Employees Occupational Illness Compensation Program Act of 2000 as Amended (June 2007) PDF 30 KB (11 pages)
-
NIOSH-IREP Version 5.5.1 Upgrade (05/16/06)
NIOSH-IREP v.5.5.1, introduced on May 16, 2006, displays the name of the uploaded template file on the summary results page. In addition, v5.5.1 incorporates memory and speed enhancements which allow a larger number of exposures for a given case. Probability of causation (PC) results from Version 5.5.1 are identical to those calculated using Version 5.5, the previous version of NIOSH-IREP. See below for more details:
Filename displayed on summary page: Based on a recommendation by a Department of Labor claims examiner, the name of the Excel template file uploaded for a given IREP run is now displayed on the IREP Summary (PC results) page. This makes it easier to identify printed summary pages for claims with multiple cancers.
Memory and speed improvements: Due to the need to adjudicate claims with large numbers of exposures, the underlying NIOSH-IREP source code has been streamlined for efficiency. As a result, NIOSH-IREP v5.5.1 requires 30-40% less memory than v5.5 on the server that houses it and the processing speed has been increased by 10-20%. These modifications have absolutely no effect on PC results. The PC results for a case processed using NIOSH-IREP v5.5.1 are exactly the same as if the case were processed using NIOSH-IREP v5.5.
-
CIRRPC
Science Panel Report No. 6
Use of Probability of Causation by the Veterans Administration in the Adjudication of Claims of Injury Due to Exposure to Ionizing Radiation
Complete document PDF 1.71 MB (43 pages)
Pages i-15 PDF 1.05 MB (21 pages)
Pages 16-B9 PDF 682 KB (22 pages)
- Report of the NCI-CDC Working Group to Revise the 1985 NIH Radioepidemiological Tables
External Link: http://dceg.cancer.gov/files/NIH_No_03-5387.pdf
NIH Publication No. 03-5387
This updated report, published in September 2003, uses epidemiological dose-response data and uncertainty analyses in providing a scientific basis for quantifying radiation-related cancer risks. The tabular version is an interactive radio-epidemiological computer program known as "NIH-IREP" and corresponds to the report. NIH-IREP can be accessed online at http://www.irep.nci.nih.gov/.
Both the Report and NIH-IREP may also be accessed on the National Cancer Institute's Interactive
Radioepidemiological Computer Program Web page.
Top of Page
Page last updated: July 16, 2009
Page last reviewed: June 9, 2009
Content Source: National Institute for Occupational Safety and Health (NIOSH)
|