|
Epi Wit & Wisdom Letters
Re: Reye’s and
Ethical Conduct
Dear Sir:
The article in the January
1985 issue that featured Colin Soskolone on the topic
of ethical standards for epidemiologists fanned the flames of an issue
that has concerned me for a number of years. Having been the recipient
of a “witch hunt” type of attack, I feel that I can speak to this
issue of ethical standards for epidemiologists.
The incident in question
centered on the controversy surrounding the association of aspirin and
the etiology of Reye’s Syndrome. In 1982, we presented the findings of
our case-control study of Reye’s Syndrome in the literature. (1) As
early as 1980 we presented our data several times to members of the
pharmaceutical industry. In addition, we presented our results to two
different committees assembled by the CDC and the NIH. This was
appropriate considering the potential ramifications at the time.
Beyond my definition of appropriateness, however, were the court
actions and six hour depositions which several of us endured at the
hands of the pharmaceutical company attorneys. But most inappropriate
of all was the review and subsequent statements about our data by the
“epidemiologists for hire.”
With the single-minded
intent of discrediting the data that was presented, the
epidemiologists of the pharmaceutical companies and more importantly
the “epidemiologists for hire” made subsets of subsets of subsets of
data that were not intended for this type of analysis. They, in fact,
tortured the data to the point that it did confess to whatever charge
they wanted to prove. Even this was not the most inappropriate thing
that was done. The worst thing that occurred in my mind was the use to
which these extensive subset analyses were put.
These analyses and the
subsequent “what if” scenarios were played before a Cong-ressional
sub-committee, the American Academy of Pediatrics and some key members
of the Reagan Administration staff to the point that the Office of
Management and Budget squashed the Food and Drug Administration
attempt to warn consumers about the potential association by the use
of warning labels on aspirin products.
It is now several years
later, and the preliminary results of the Public Health Service Reye’s
Syndrome Task Force have strongly replicated the study that was first
presented nearly five years ago. This current study is probably one of
the most complex, complete and expensive case-control studies ever
undertaken in this country. Only time will tell whether this study was
really necessary. What is in immediate question here, however, is the
role played by the epidemiologists hired to review the data and more
importantly how the subset analyses were used when dealing with
non-epidemiologist decision makers.
In conclusion, I feel that
it is appropriate that epidemiologic data be subject to review; even
review to the point of performing analyses on the data which the data
were never meant to withstand. But, there should be some ethical
conscience that dictates how the presentation of the “above and
beyond” analyses are conducted and presented. Especially when the
intent is to discredit the analyses that have already undergone peer
review for publication in a major journal. Needless to say, I support
some form of a code of ethical conduct for epidemiologists. The
feelings expressed here do not necessarily reflect the opinions of my
co-authors.
Frank
Holtzhauer
1. Halpin,
T., Holtzhauer, F., Campbell, R., et al: Reye’s Syndrome and
Medication Use. JAMA 1982; 248: 687 - 91.
Pubished February
1985
|
