Environmental tobacco smoke and tobacco related mortality in a prospective study of Californians, 1960-98

I read with absolute fascination not only the original Enstrom &
Kabat article on ETS and mortality, but the numerous responses to it,
published in BMJ, and ancillary articles about this phenomenon (which it
truly has become). I salute Dr Enstrom's decision to defend his research
with clarity and honesty. I agree with many of his conclusions, both
small and large and extend to him my support.

I ran into a similar but smaller scale problem when I was doing my
epidemiology dissertation many years ago. It concerned a new method of
determining the work-relatedness of ischemic heart disease (by examining
mortality data at young ages when the background rate IHD is low). I did
not have data on smoking in my dataset, so I was reduced to comparing
occupations and industries with similar levels of smoking, as ascertained
over time by different researchers. I left as a more or less open
question whether smoking was even a confounder that needed adjustment in
this study, given that there are no data demonstrating early deaths in
otherwise healthy workers in their 20s.

In any event, I was castigated (by the faculty) from the start:
accused of having analysed the data and in a post-hoc manner selected the
occupations; making up power formulae; having a political agenda! (I was a
graduate student supported by NCI and doing part-time work for a local
occupational medicine physician - and my thesis made no mention of cancer
whatsoever) - and this was only during the protocol development stage.
When I defended, I was told that smoking explained all the findings (a
patently idiotic suggestion), that I should change or delete some
references and quotations which questioned some of the 'basic tenets of
public health' (ie, that smoking is the cause of all [public health]
evil).

I have worked for the past 15 years for or in the pharmaceutical
industry where anything we produce is, by the same aberrant logic that so
harshly traduces Dr Enstrom, tainted - as is any work we support. There
are certainly many people who are genuinely concerned with the safety of
the drugs and vaccines we put into people, enough to suffer the
marginalization within industry (nobody likes people in Drug Safety
departments very much, not even within a drug company) as well as the
deprecations suffered from without. What I have learned over the years is
to evaluate a study based on its methods (the section I read first) rather
than the source of its funding; if the methods are poor, I read no
further, despite the authors' extensive or even non-existent conflicts of
interests.

I am shocked (somewhat) and disappointed (gravely) that this nonsense
continues today. It's not just the integrity of epidemiology that we have
to be concerned about, but science in general Albert Einstein wrote
that, 'The right to search for truth implies also a duty; one must not
conceal any part of what one has recognized to be true.' I'm glad that Dr
Enstrom and others like him understand their duty - they are not alone!

Competing interests:
I am a full-time employee of a vaccine manufacturing company.