Re: the healthy worker effect

[js_dukelow@ccmail.pnl.gov]

On Thu, 21 Mar 1996 10:33:27 -0500 (EST)
Jim Muckerheide wrote on the 
Subject: Healthy Worker Effect

<< Snip >>

 [.. the "healthy worker" effect holds that .. ] Workers do not include 
people who can't work due to medical problems, and workers get better 
health care, and some less significant (unquantified!) factors. But, for 
carcinogenesis, for radiation effects, how does that 20 or 30 year-old 
worker get more or less cancer than the general population (that does 
not already have cancer). "Public health" advocates would otherwise have 
us believe that the industrial workplace is a cesspool of INCREASED 
cancer. These people can have it both ways in their political rhetoric 
justifying "concern" (and funds). 

Consider for radiation specifically:  Many studies with internal 
controls confirm no such effect exists. See eg, the Nuclear Shipyard 
Worker Study (NSWS), the best worker study in the history of the planet 
re dosimetry and confounding factors and the size of the population 
(publication suppressed; now at least ref'd in UNSCEAR 94). The non-
nuclear workers SMR vs aged adjusted general population is 1.00 (+- 
0.03). 

It is not plausible. The "healthy worker effect" comes from the general 
public that includes people not well enough to work. This works for 
general illness and disease. It can not apply to long-term cancer.  (Ie, 
no data or plausibility supports that rationale that selecting a healthy 
worker today, and even good health care, can affect comparisons to 
cancers that appear later, with a 5, 10 or 15 year lag.) 

Even more disingenuous is that the "public health establishment" in 
other forums considers the industrial workplace to be a SOURCE of 
cancer. If there were anything to that, there should be an "unhealthy 
worker effect" _for cancer_ in industrial occupations exposed to 
industrial chemicals and other physiological damage, in addition to 
radiation. 

Requests to produce data confirming the existence, and quantifying, the 
effect to the proponents of this rationalization (which is easy to do 
epidemiologically, as in the NSWS, and which can/should be done in many 
non-nuclear occupational populations) has never gotten anyone to 
demonstrate and confirm/quantify the effect. (From specific challenges 
over the years, the knowledgeable science defenders of the justification 
and consequences of the linear _hypothesis_ know this well.) 

Thanks.

Regards, Jim Muckerheide
jmuckerheide@delphi.com
Radiation, Science, and Health

<< End of Jim Muckerheide post on the "healthy worker" effect >>

<<  On  21 Mar 1996 09:26:09 MDT  "TED S. BOHN" <BST@inel.gov> wrote on 
the Subject: Healthy Worker Effect  >> 

I haven't followed the entire string of this subject, but are there 
studies that establish that an UNHEALTHY person is more likely to 
contract cancer due to radiation exposure than a HEALTHY person?  I can 
see this in bacterial and virial caused diseases but not cancer.  Maybe 
the "Healthy Worker" types are assumed not to be in contact with 
carcinogens as regularly as the "Unhealthy Worker" types. I thought that 
radiation caused cancer however,is not discriminatory and even the 
healthiest worker has an equal opportunity. 

<< End of Ted Bohn post on the "healthy worker" effect >>

<< My two bits >>

It is certainly plausible that a working population that does not 
include those members of the general population who are too chronically 
ill or disabled to work ought to have better mortality and morbidity 
statistics than the general population.  Sizable and numerous 
epidmiological investigations of DOE contractor populations at Hanford, 
Oak Ridge, and Savannah River mostly seem to exhibit a healthy worker 
effect.  The healthy worker effect is a characteristic of many other
occupational health studies.  Indeed, if I came across a working 
population that did not exhibit a healthy worker effect, I would 
consider that prima-facie evidence that something was happening to that 
population and would want to investigate further. 

It is also plausible that this effect would extend to cancer morbidity 
and mortality, since the immune system is one on the principal lines of 
defense against the growth of transformed cells into tumors or non-tumor 
proliferating cell populations.  Jim Muckerheide has several times cited 
the datum (see, for instance, his 17 March 1996 post in RADSAFE 861) 
that there are 8000 damage events per cell per hour from normal 
biological processes.  Accepting this number, for the sake of argument, 
what keeps this enormous number of damage events from overwhelming the 
body with cancer?  There are two main lines of defense.  The first line 
is a variety of intra-cellular damage-repair mechanisms.  The second 
line of defense is the immune system -- both antigen/antibody tagging of 
"non-self" cells in the body and the various types of immune system 
killer cells.  A robust immune system should tend to reduce both cancer 
and non-cancer disease. 

Finally, although I tend to agree with Jim on the substantive point that 
the linear no-threshold hypothesis is not scientifically correct for low 
doses and low dose rates (of radiation and other carcinogens), I find 
his relentlessly ad hominem argumentation to be tedious, off-putting, 
and bad from a public relations point of view.  

According to Jim, linear no-threshold is the regulatory paradigm because 
BAD people, acting in their own interest rather than society's, put it 
in place and defend it against all challenges.  I disagree.  I am 
personally acquainted with some vigorous defenders of linear no-
threshold and know them to be more-than-competent and absolutely 
sincere.  

I think the blame for the current situation (in the US, at least) lies 
directly with past Congresses and Administrations that have created 
regulatory agencies and given them broad regulatory scope, but 
no license to consider how regulation in their area of responsibility 
should be balanced against costs, risks, and benefits of regulation and 
investment in other parts of society.  The US Nuclear Regulatory 
Commission, for instance, was initially chartered to assure safe 
operation of nuclear power plants, without consideration of the costs of 
implementing the regulations.  Subsequently, the 10 CFR 50.59 backfit
regulations did impose a cost-benefit basis for retroactive changes to 
regulations, but I think it is fair to say implementation cost is still 
the poor stepchild of NRC regulation and that there is no effort to 
balance the costs and benefits of NRC regulations against the costs and 
benefits of regulating other sectors of the economy (or, for that 
matter, investing in other sectors of the economy and society).  Similar 
comments could be made about the FDA and the Delaney Clause and about 
EPA and Superfund/RCRA/CERCLA.  The only part of government charged with 
balancing risk and cost across all of society and the economy has been 
the part probably least competent to do that job -- Congress. 

At their best, actions in the current and the immediate previous 
Congress to increase the use of risk analysis and risk management for 
prioritization of regulation represent a potential way out of the 
morass.  At their worst, these bills represent pandering to special 
interests that are inalterably opposed to all regulation. 

My prescription:  Jim (and all of us) should focus on demanding good 
science (in this case, sound risk analysis), with less muttering about 
dark conspiracies and demonization of our opponents.  We should also 
support the creation of mechanisms (i.e., risk management and sound 
policy formulation) to utilize that good science to optimize the mix of 
investment and regulation in our society. 

Best regards.

Jim Dukelow
Pacific Northwest National Laboratory
Richland, WA

js_dukelow@pnl.gov

These thoughts are mine and have not been reviewed and/or approved by my 
management or by the US Department of Energy.