Page 146 - 2014 Printable Abstract Book
P. 146
(PS2-17) Influence of additional data on BEIR VII estimates of DDREF. Benjamin Haley; Will Liu; Tatjana
Paunesku; and Gayle Woloschak, Northwestern University, Chicago, IL
The cancer risk of radiation exposure is primarily derived from rates observed in the survivors of
the atomic bombs from Hiroshima and Nagasaki divided by a dose and dose rate effectiveness factor
(DDREF) in order to estimate the rates that would be observed if this same dose had been protracted
across time or spread across a greater population as is the case with most modern exposures. The BEIR
VII committed estimated a value of DDREF of 1.5 (1.1-2.3) based on data from atomic bomb survivors and
two mouse studies conducted at Oak Ridge National Laboratory [1]. Our work extends this analysis by
applying the same methodology to 12 additional animal studies. We find that the BEIR VII method does
not generalize to additional datasets producing extremely confident and contradictory estimates of
DDREF. These contradictions can mostly be eliminated by updating the BEIR VII methodology to use a
random effects model. The resulting DDREF estimate is centered around 1, e.g. no effect. However this
estimate still shows some signs of residual heterogeneity and should be considered provisional. Also, the
authors have yet to integrate this result with data from atomic bomb survivors to produce a final modified
estimate of DDREF. Overall we find that the BEIR VII estimate of DDREF does not hold up to the condition
of repeatability when estimated from different datasets. We are actively seeking ways to develop better
estimates of DDREF. Our ongoing analysis is online and regularly updated [2]. References
[1]: National Research Council. Health Risks from Exposure to Low Levels of Ionizing Radiation: BEIR VII
Phase 2. Washington, DC: The National Academies Press, 2006. www.nap.edu/ openbook.php
?record_id=11340&page=246, [2]: http://rpubs.com/benjaminhaley/ddref
(PS2-18) Cerebrovascular Diseases in the extended cohort of Mayak workers first employed in 1948-
1
1
1
2
1982. Tamara V. Azizova, MD PHD ; Maria B. Moseeva ; Richard E. Haylock ; Maria V. Bannikova ; and
1
1
Evgenia Grigorieva Southern Ural Biophysics Institute, Ozyorsk, Russian Federation and Public Health
2
England, Chilton, United Kingdom
Cerebrovascular Diseases in the extended cohort of Mayak workers first employed in 1948-1982
Azizova TV, Moseeva MB, Haylock R, Bannikova MV, Grigorieva ES The objective of the current study is to
assess incidence and mortality risks of cerebrovascular diseases (CVD: ICD-9 codes 430-438) in the
extended cohort of Mayak workers. The extended cohort consists of 22377 workers first employed at one
of the main facilities of the Mayak PA (Reactors, Radiochemical, and Plutonium Production) in 1948-1982
followed-up to the end of 2008. Reactors workers were exposed to only external gamma-rays,
Radiochemical and Plutonium production workers were exposed to combined external gamma rays and
internal alpha-radiation from incorporated 239 Pu. Overall, the same approach was used as employed in
the two previous analyses of CVD [Azizova et al 2010a; Azizova et al 2011] but with some additions.
Analyses of non-radiation factors were first performed to confirm which factors influenced to CVD rates.
The main analysis comprised a categorical analysis in which relative risks (RR) were calculated for
categories of total external gamma-dose (<0.1, 0.1-, 0.2-, >0.5 Gy) and total internal dose from alpha-
radiation to liver (< 0.01, 0.01-, 0.025-, >0.1 Gy) (restricted to those workers who were monitored for
plutonium exposure); a dose response analysis in which linear excess relative risk estimates were
calculated. This analysis revealed statistically significant relation of CVD incidence and mortality to well-
known non-radiation factors, such as gender, age, calendar period, smoking status etc. However, risk
analysis has shown that external and internal exposure contributes only to CVD incidence. No sufficient
144 | P a g e
Paunesku; and Gayle Woloschak, Northwestern University, Chicago, IL
The cancer risk of radiation exposure is primarily derived from rates observed in the survivors of
the atomic bombs from Hiroshima and Nagasaki divided by a dose and dose rate effectiveness factor
(DDREF) in order to estimate the rates that would be observed if this same dose had been protracted
across time or spread across a greater population as is the case with most modern exposures. The BEIR
VII committed estimated a value of DDREF of 1.5 (1.1-2.3) based on data from atomic bomb survivors and
two mouse studies conducted at Oak Ridge National Laboratory [1]. Our work extends this analysis by
applying the same methodology to 12 additional animal studies. We find that the BEIR VII method does
not generalize to additional datasets producing extremely confident and contradictory estimates of
DDREF. These contradictions can mostly be eliminated by updating the BEIR VII methodology to use a
random effects model. The resulting DDREF estimate is centered around 1, e.g. no effect. However this
estimate still shows some signs of residual heterogeneity and should be considered provisional. Also, the
authors have yet to integrate this result with data from atomic bomb survivors to produce a final modified
estimate of DDREF. Overall we find that the BEIR VII estimate of DDREF does not hold up to the condition
of repeatability when estimated from different datasets. We are actively seeking ways to develop better
estimates of DDREF. Our ongoing analysis is online and regularly updated [2]. References
[1]: National Research Council. Health Risks from Exposure to Low Levels of Ionizing Radiation: BEIR VII
Phase 2. Washington, DC: The National Academies Press, 2006. www.nap.edu/ openbook.php
?record_id=11340&page=246, [2]: http://rpubs.com/benjaminhaley/ddref
(PS2-18) Cerebrovascular Diseases in the extended cohort of Mayak workers first employed in 1948-
1
1
1
2
1982. Tamara V. Azizova, MD PHD ; Maria B. Moseeva ; Richard E. Haylock ; Maria V. Bannikova ; and
1
1
Evgenia Grigorieva Southern Ural Biophysics Institute, Ozyorsk, Russian Federation and Public Health
2
England, Chilton, United Kingdom
Cerebrovascular Diseases in the extended cohort of Mayak workers first employed in 1948-1982
Azizova TV, Moseeva MB, Haylock R, Bannikova MV, Grigorieva ES The objective of the current study is to
assess incidence and mortality risks of cerebrovascular diseases (CVD: ICD-9 codes 430-438) in the
extended cohort of Mayak workers. The extended cohort consists of 22377 workers first employed at one
of the main facilities of the Mayak PA (Reactors, Radiochemical, and Plutonium Production) in 1948-1982
followed-up to the end of 2008. Reactors workers were exposed to only external gamma-rays,
Radiochemical and Plutonium production workers were exposed to combined external gamma rays and
internal alpha-radiation from incorporated 239 Pu. Overall, the same approach was used as employed in
the two previous analyses of CVD [Azizova et al 2010a; Azizova et al 2011] but with some additions.
Analyses of non-radiation factors were first performed to confirm which factors influenced to CVD rates.
The main analysis comprised a categorical analysis in which relative risks (RR) were calculated for
categories of total external gamma-dose (<0.1, 0.1-, 0.2-, >0.5 Gy) and total internal dose from alpha-
radiation to liver (< 0.01, 0.01-, 0.025-, >0.1 Gy) (restricted to those workers who were monitored for
plutonium exposure); a dose response analysis in which linear excess relative risk estimates were
calculated. This analysis revealed statistically significant relation of CVD incidence and mortality to well-
known non-radiation factors, such as gender, age, calendar period, smoking status etc. However, risk
analysis has shown that external and internal exposure contributes only to CVD incidence. No sufficient
144 | P a g e
