|
  
Cox Companies:
|
WHAT IS THE ECONOMIC VALUE OF INFORMATION FROM
TRACKING CANADIAN CATTLE IMPORTED TO THE US?
Amid growing concern about the potential emergence of BSE in North America since
the discovery of a BSE-positive cow of Canadian origin in Seattle in 2003, there
has been debate and litigation over whether, when and how fully the US should
re-open its borders to the import of Canadian cattle and beef products. In 2004,
Cox Associates led a study for R-CALF (Ranchers-Cattlemen Action Legal Fund) to
identify and apply decision and risk analysis frameworks to help inform the
debate. One of the results is a Value-of-Information (VoI) framework that
suggests that tracking Canadian-origin cattle in the US may have a high economic
value if additional BSE cases are discovered. This framework and results can be
downloaded here.
WHICH
FOOD-BORNE MICROBIAL HAZARDS MATTER MOST?
In June of 2003, Cox Associates
delivered to the Animal Health Institute (AHI) a new approach to quickly and
accurately rate food-borne microbial hazards and risk management proposals.
The new approach, called the Rapid Risk Rating Technique (RRRT), rates hazards
and risk management interventions based on their probable human health impacts.
These are estimated by multiplying several risk factors, including relative
values for expected contamination exposure frequency, infectivity, dose-response,
and health consequence. Data values are provided in the methodology.
The new rating methodology
is intended to build upon and refine FDA's proposed Draft Guidance Document
#152 while seeking to overcome some of #152's potential limitations (http://www.fda.gov/ohrms/dockets/ac/03/slides/3919OPH2_01_Cox.ppt
http://www.fda.gov/ohrms/dockets/ac/03/transcripts/3919T2.htm).
Simulation-optimization is used to optimize and validate the performance of
the proposed rating system in identifying the worst problems, taking into account
realistic uncertainties.
The RRRT method can consider
human health benefits as well as risks, account for projected changes in future
antimicrobial resistance rates among multiple bacterial pathogens, and assess
risks for specific human sub-populations. Applications to tetracyclines, macrolides,
ionophores, and other classes of animal antimicrobials are ongoing.
ARE ANIMAL ANTIBIOTICS
TIME BOMBS?
Concern over evolving resistance
to antibiotics among human bacterial pathogens and commensals due to the use
of similar drugs in food animals has been growing in the US, Japan, and the
European Union. Biomathematical modeling of the evolution of resistance has
added to this concern by suggesting that continued use even of drugs (such as
virginiamycin) with long histories of safe use in animals and without apparent
significant adverse effects on human resistance might lead to a sudden increase
in human resistance in the future. An April 2003 draft report by Cox Associates,
undertaken with partial support from Phibro Animal Health to understand and
quantify the potential human health risks, has found that these mathematical
models predict a very low probability of future resistance outbreaks and a vanishingly
small probability of resistance epidemics triggered by continued use of animal
antibiotics and antimicrobials if real, rather than hypothetical, parameters
values are used. Using virginiamycin as a case study, the report applies techniques
of Bayesian uncertainty analysis to a discrete-event dynamic simulation model
with parameters estimated from data to show that the human health risks are
small (much less than one excess statistical mortality in the whole US population
expected over many years due to emerging resistance caused by animal drug use)
with very high statistical confidence. This analysis based on dynamic simulation
produces quantitative risk estimates similar to those previously obtained by
simpler calculations (www.cox-associates.com/VIRGINIAMYCIN.ppt.)
WHEN
ARE LARGER EXPOSURES LESS RISKY?
In May of 2003, as part
of a research initiative supported by the American Chemistry Council, Cox Associates
completed an initial version of an integrated pharmacokinetics-and-pharmacodynamics
(PK-PD) simulation model of volatile organic chemical (VOC) interactions with
hematopoietic (blood-forming) stem cell populations in the bone marrow. Building
on an earlier biomathematical model of hematotoxicity (www.alceon.com/cp99pa2.pdf,
http://ehpnet1.niehs.nih.gov/docs/1996/Suppl-6/cox.html),
the new model integrates the homeostatically regulated responses of normal hematopoietic
stem cell populations into a two-stage stochastic model of carcinogenesis and
cytogenetic damage. It is able to quantitatively simulate, for the first time,
the theoretical cancer-causing potential of different exposure patterns. Initial
results show that cumulative exposures (ppm-years), the most widely used exposure
metric in most epidemiological analyses and statistical risk models of VOC cancer
risks, do not yield accurate risk predictions if exposure patterns are highly
variable. Instead, in addition to the cumulative exposure, the average concentration
within a day, hours-per-day of exposure, days between consecutive exposures,
and length of post-exposure follow-up can all strongly affect predicted hematotoxic
and leukemogenic risks. These factors combine non-linearly, and their interactions
are so strong that they can easily lead to larger cumulative exposures posing
smaller risks (essentially by allowing more time for the hematopoietic system
to adapt to the stress of sustained exposure, reducing the extent of compensating
proliferation and the number of initiated stem cells formed.) These initial
results were presented at the 2003 BELLE conference on Non-Linear Dose-Response
Relationships in Biology, Toxicology, and Medicine: An International Conference.
(University of Massachusetts, Amherst, MA, May 28-30th, 2003 www.belleonline.com/abstracts/session6.html).
They appear to be robust to many scientific and modeling and uncertainties.
They suggest a new class of multivariate exposure metrics that can better explain
past animal and human data and better describe the aspects of exposures that
are most useful for predicting occupational and public health risks from exposures
to hematotoxic VOCs. Slides describing these preliminary results can be downloaded
here.
ANIMAL
ANTIMICROBIAL RISK ASSESSMENT DO'S AND DONT'S
In February of 2003, Cox
Associates in partnership with the Animal Health Institute prepared a report
offering suggested technical guidance on how to assess human health risks from
antimicrobial drugs used in animal feed additives. The report, entitled Animal
Antimicrobial Feed Additives and Human Health: A Guide to Risk Analysis, reviews
methods and principles for sound and practical risk analysis in this area, as
well identifying practices that should be avoided. It can be downloaded here.
WHAT
DO VOLATILE ORGANIC CHEMICAL (VOC) CARCINOGENS HAVE IN COMMON?
In January of 2003, Cox
Associates completed a research report on data sources available to support
integrated pharmacokinetics-and-pharmacodynamics (PK-PD) simulation modeling
of cancer risks from mixtures of volatile organic chemicals (VOCs). This work
suggests a framework for exploiting the common features of many VOC-induced
carcinogenic responses - including Phase 1 and Phase 2 metabolism leading to
reactive products causing cytotoxicity and regenerative hyperplasia, or other
mechanisms of stem cell proliferation - to create a common template for streamlining
the analysis of cancer risks from exposure to the most common VOCs and their
mixtures. Data sources and technical modeling literature for PK and PD components
available to support such an ambitious attempt at integrated risk modeling are
surveyed and hyperlinks to relevant documents are provided. The report can be
downloaded here.
SOCIETY
FOR RISK ANALYSIS BEST PAPER AWARD, 2002
In December of 2002, a
paper on "Quantifying human health impacts of antimicrobial risk management
alternatives for enrofloxacin", by Drs. Tony Cox and Doug Popken of Cox
Associates, received a Society for Risk Analysis Best Paper Award at the Society
for Risk Analysis Annual Conference, New Orleans, LA. December 9-11, 2002 (www.sra.org/news0203.pdf,
page 5). This paper presented the results of mathematical model validation and
extension efforts originally recommended by Dr. Cox to the FDA in 1999 (http://www.fda.gov/cvm/antimicrobial/vm120999.htm#Anchor-MATHEMATICA-43816).
It also presents initial results of a new simulation model suggesting how withdrawing
an approved animal drug can increase human health risks by increasing the prevalence
of a condition in chickens called airsacculitis that increases the microbial
loads of campylobacter, salomonella, and other bacteria reaching humans. Finally,
it reviews the evidence for and against the widely accepted hypothesis that
chicken is a primary source of campylobacteriosis in humans - a piece of common
wisdom that may have to be revised in light of data. The presentation slides
can be downloaded here.
VIRGINIAMYCIN
USE IN CHICKENS POSES MINIMAL HUMAN HEALTH RISKS
In May
of 2002, Cox Associates completed a quantitative risk analysis of the human
health risks that might plausibly be attributed to the use of Virginiamycin
as a growth promoter in poultry. The risk analysis makes use of a simulation
model, based on extensive human and animal data, that quantifies the number
of human treatment failures, early mortalities, and excess illness-days that
could result from bacteria (vancomycin-resistant Enterococcus faecium, or VREFs)
that also acquire resistance to the human drug quinupristin-dalfopristin due
to selection pressures from use of Virginiamycin in poultry. The model uses
recent genogroup and genotype data from amplified fragment-length polymorphism
(Willems et al., 2000, 2001) to show that an extreme upper bound on the number
of human lives saved by an immediate ban on Virginiamycin use in poultry is
less that 1/3 of a statistical life for the entire US population over the next
5 years. The true number could be zero, if resistance gene transfer from poultry
to humans does not occur.
NEW
BOOK ON RISK ANALYSIS
Many theoretical advances
and practical applications of health risk analysis developed at Cox Associates
between 1986 and 2001 have been summarized in a recent book on Risk Analysis:
Foundations, Models and Methods (http://kapis.www.wkap.nl/prod/b/0-7923-7615-3).
This advanced text, published in the prestigious INTERNATIONAL SERIES IN OPERATIONS
RESEARCH AND MANAGEMENT SCIENCE, makes full use of contemporary operations research
methods in tackling some of the most difficult and rewarding problems of risk
model creation, validation, and application needed in current health risk analysis
applications. The attached presentation illustrates a few of the ideas explained
in detail in the monograph.
BALANCING
"REAL" VS. "THEORETICAL" RISKS TO CANADIAN BLOOD SUPPLY
Member of
different communities of blood recipients, such as hemophiliacs and immunocompromised
patients, often place different weights on the relative importance of availability
and extremely high safety of blood supplies. This leads to potential conflicts
over policies for addressing the potential risks from emerging threats, including
prion-borne (vCJD, or mad cow disease related) contamination. Proposals to defer
donations from blood donors who have visited the United Kingdom or who may otherwise
have been exposed to vCJD may increase the theoretical safety of donated blood,
but may do so by increasing the risks of shortages and/or contamination from
new first-time donors. In March, 2002, Dr. Cox presented a decision-analytic
approach to making risk-risk tradeoffs and optimizing decisions about blood
supply in the face of large uncertainties about the underlying science and the
reality and extent of threats from prion-related contamination. The approach
calls for the use of simulation models that can be fit to available data to
estimate the probable consequences of alternative risk management decisions.
It provides a framework for defining and choosing the "best" risk
management option even when there are realistically wide uncertainties about
the probable consequences of different choices.
IS
PERCHLORATE BAD FOR YOU?
Perchlorate, widely used
in rocket fuel, is found as a drinking water contaminant in California, Nevada,
and elsewhere. While high doses of perchlorate can cause thyroid tumors and
other effects in rats by inhibiting iodine uptake into the thyroid gland, ultimately
leading to compensating cell proliferation and increased cancer risk, there
is no evidence of a comparable carcinogenic effect in humans. In March of 2002,
Dr. Cox participated as an invited Expert Peer Reviewer at a two-day public
meeting in Sacramento to advise the US EPA on how to maximize the usefulness
and validity of its recent Draft Risk Assessment document for human health effects
of perchlorate. A key recommendation was that the EPA risk assessment needs
to address pharmacodynamic differences between human and rat sequelae of iodine
inhibition in order to realistically explain and predict human health impacts
of perchlorate in drinking water.
WHAT
REALLY CAUSES CAMPYLOBACTERIOSIS?
Campylobacter jejuni is
the most frequent bacterial cause of gastroenteritis in the United States and
frequently causes traveler's diarrhea in both developed and developing countries.
Recently, the U.S. Food and Drug Administration (FDA) proposed to ban a therapeutic
drug, enrofloxacin, from veterinary use in chickens in case such use might increase
drug-resistant campylobacteriosis in humans. On March 1, 2002, Dr. Cox presented
new data analyses in Boston, Massachusetts, indicating that eating chicken per
se is not a risk factor for campylobacteriosis in the United States, contrary
to widespread assumptions. Instead, it appears that restaurant dining is a risk
factor, whether or not chicken is eaten, while chicken and other meats consumed
at home are protective against campylobacteriosis. This finding has dramatic
implications for how risks of campylobacteriosis should be managed. It suggests
that banning enrofloxacin will have little or no human health benefit, but that
better restaurant kitchen hygiene could be important.
Drs. Cox and Popken of
Cox Associates are currently modeling the risks, costs, and benefits associated
with different risk management strategies for campylobacter. Initial simulation
results suggest that a ban on enrofloxacin may inadvertently increase the microbial
load reaching consumers by reducing uniformity of chickens at the processing
plant. In March, this and other insights from "farm-to-fork" simulation
modeling of risks associated with chicken-borne campylobacter were submitted
to the World Health Organization (see attached link).
ARE
PLAYGROUNDS POISONING OUR CHILDREN?
In April, 2002, the Detroit
News ran a story in which Dr. Cox helped to debunk claims by the Environmental
Working Group (EWG) that pressure-treated wood is putting children at risk of
cancer due to the preservative chromated copper arsenate (CCA). The EWG's conclusions
were extrapolated from a statistically non-valid study in which EWG invited
environmental activists to collect samples without using any standard or reliable
sampling plan and then interpreted the results as representing realistic exposures.
In Dr. Cox's opinion, the EWG study did not fulfill the requirements for a scientifically
credible or useful risk assessment. http://www.detnews.com/2002/editorial/0204/09/a11-460152.htm
CAUSAL
MODELING FOR WORKSHOP ON ANTIMICROBIAL RESISTANCE
In October,
2001, Cox Associates presented initial findings at the Second OIE Workshop on
Antimicrobial Resistance (http://www.oie.int/eng/press/a_051001.htm)
from a new causal model and analysis of case-control data from the Centers for
Disease Control (CDC). Causal modeling indicates that improper restaurant food
preparation, rather than microbial load of Campylobacter in chickens per se,
as has previously been supposed, is likely to be the main driver of campylobacteriosis
risks in the US. Indeed, our independent analysis confirmed and extended previous
findings suggesting that chicken prepared and eaten at home protects against
campylobacter risk.
RISK
ANALYSIS FOR WORLD HEALTH ORGANIZATION
In July, 2001, Dr. Tony
Cox participated as an independent expert in the World Health Organization's
Consultation on human health risks
from campylobacter (http://www.fao.org/es/ESN/pagerisk/reportCV.pdf).
DO
ANIMAL ANTIBIOTICS INCREASE THE RISK OF HUMAN RESISTANCE?
In February
2001, Cox Associates delivered to the Animal Health Institute a risk analysis
report on the relation between use of fluoroquinolones (FQ) to treat
critically ill chicken flocks and potential excess illness-days in the
US population due to increased resistance of Campylobacter jejuni (CP), a
chicken-borne pathogen, to FQ treatments. This report breaks new ground
by applying a dynamic discrete-event simulation model to quantify the
probable health impacts of different risk management interventions. Its
main conclusion is that the potential health benefits from tighter control
of FQ use on farms is less than 1% of the potential health benefits from
relatively modest changes in processing of chickens. In addition,
the report finds little evidence of a causal relation between FQ use
on farms and excess illnesses or illness-days in humans.
MOLECULAR
BIOLOGY OF CHEMICAL LEUKEMOGENESIS FOR BENZENE
In February, 1999, Cox
Associates delivered to Exxon Biomedical Sciences, Inc., a review of the detailed
biological mechanisms by which benzene can cause hematotoxic damage and induce
secondary Acute Myeloid Leukemia in humans. A key finding from this study is
that many of the mechanisms of benzene-induced health effects are sub-linear
at low doses, helping to explain why recent epidemiological studies do not find
the excess leukemias and health damage at low exposure concentrations (e.g.,
1 ppm or less) predicted from older benzene health risk assessment models.
Top
CLINICAL
APPLICATIONS OF CYCLOPHOSPHAMIDE MODEL?
On December 15th, 1998
a Hematoxicity Modeling Workshop was held at the University of Ottawa Institute
of Population Health (Professor Dan Krewski, host), in conjunction with the
Benzene State of the Science Workshop, 1998. The topic of the one-day Hematotoxicity
Modeling Workshop was a review of the simulation model of
cyclophosphamide hematotoxicity developed by Cox Associates for the American
Petroleum Institute. Reviewers included government scientists from EPA and NIEHS,
biomathematical modelers from the Fred Hutchison Cancer Research Center and
the University of Ottawa, and experts in clinical hematology and research from
the University of Ottawa and the University of Colorado. Workshop papers will
be submitted for publication in 1999. The main finding, reported to the Benzene
State of the Science Workshop, was that the model appeared to offer useful,
apparently realistic predictions. An unexpected finding was that several of
the expert hematologists felt that the model might be useful in clinical practice,
in refining dose regimens for chemotherapy patients.
Top
Cox Associates' ongoing
applied research in data mining technologies has recently led to a promising
new application: using data mining algorithms to combine predictions from different
expert systems for predicting which chemicals are carcinogenic without undertaking
expensive experiments to find out. We discovered that the new prediction-combination
technique, which is based on classification trees, can produce "hybrid"
predictions than are more accurate than any individual expert system's prediction
and also than the predictions from previous combination methods. Click here to download a technical paper describing the new
approach and results.
Top
In September of 1998, Cox
Associates completed a new approach to finding the probable locations of greatest
soil concentrations of contaminants based on soil samples. Traditional geostatistical
and sampling and estimation methods such as kriging do not fully address the
problem of efficient search for buried hazards when the spatial distribution
of yield is highly uncertain. Our new approach adaptively optimizes the search
process so as to continually maximize its expected yield. This can dramatically
cut the sampling costs needed to show that spatial hazards have been adequately
reduced by sampling and remediation efforts. Details may be found in the attached
technical paper. Click here to download the paper.
Top
Since 1986, Cox Associates
has developed biologically-based risk assessment (BBRA) simulation models and
methods for understanding the human health risks from chemical exposures. A
program developed for the American Petroleum Institute (API) to predict the
effects on blood cells of dosing with the immunosuppressive drug cyclophosphamide
is now available to other researchers from our web site. Click here
to download the required modeling software and documentation.
Top
CAUSAL
ANALYSIS AND PUBLIC POLICY: THE CASE OF DIESEL EXHAUST
In early June 1998, Cox
Associates delivered written comments on behalf of the Engine Manufacturer's
Association to the Clean Air Science Advisory Committee (CASAC) on EPA's recent
(2-98) Draft Risk Assessment for diesel exhaust health risks. The comments address
causation (as opposed to statistical associations) in epidemiological literature
on health risks of diesel exhaust. It concludes that no causal link between
diesel exhaust and human lung cancer has been found.
METHODS: The document provides
references and web links to statistical resources that help untangle causation
from statistical association. This is a novel application of insights and methods
from our data mining and causal forecasting practice areas.
IMPACT: CASAC has directed
EPA to rework its draft risk assessment to address issues identified in public
comments by Cox Associates and others. The comments document identifies current
methods for overcoming modeling challenges in this area.
For information, please
see the attached report.
CLIENT CONTACT: Glenn Keller,
Engine Manufacturer's Association, 312-644-6610
Top
|
