|
Reprint
from Epidemiology Wit and
Wisdom—The Best of the Epidemiology Monitor
Outgoing SER
President Addresses Group on Faith, Evidence, and the Epidemiologist
[Ed: In honor of colleague Paul Stolley who died August
4, 2017, we reprint here excerpts from a talk he gave to SER members
at the Winnipeg Manitoba meeting in 1983. His remarks continue to have
relevance today.]
Published July 1983
Outgoing
SER president and University of Pennsylvania epidemiologist Paul
Stolley delivered the traditional departing address to more than 600
epidemiologists assembled in Winnipeg, Manitoba for the 16th annual
meeting of the Society for Epidemiologic Research. Dr. Stolley argued
for greater reliance on the whole body of actual scientific evidence
in settling controversies and was critical of epidemiologists who
criticize or dismiss important epidemiologic findings because of real
or imagined minor flaws. Excerpts from his talk are presented below.
“...My brief talk today will be an attempt to sustain the modest
proposal that epidemiologists should persist in their efforts to
substitute evidence for faith in scientific controversy, to whatever
extent possible...
Pseudo-Science
A
curious phenomenon has been introduced into scientific controversy
involving epidemiologists during the last decade... I should call this
phenomenon a variant of pseudo-science; it is characterized by an
inability or unwillingness to synthesize available data coming from
all fields that bear on the problem at hand and instead placing
extraordinary importance upon small defects in study designs. Thus a
convincing group of studies might relate the toxic shock syndrome to
the introduction and the use of the highly absorbent tampons...
Nevertheless, a group of investigators, either acting independently or
sometimes hired by the company at risk, begin a kind of ‘witch-hunt’
for alleged bias and confounding in order to challenge the findings.
Biases that may be only postulated are somehow given a reality before
their actual existence is even demonstrated.
Social Responsibility
The
charitable view of some of the activities of so-called epidemiologists
in this regard would be to say that they are perhaps excessively
iconoclastic...
That is not to say that all findings should not be scrutinized and
challenged, but this should be done with a sense of social
responsibility. There is a decided conceptual difference between
posing a test to challenge a hypothesis and applying the test. A
hypothesis does not fail a test just because it is speculated that it
will. A group of case-control studies, for example, are not invalid
because certain biases that might have occurred are postulated. It may
well that some severe biases misled the investigator; but merely
raising these possibilities does not destroy the validity of the
study...
The
Scientific Spirit
So it is
clear that life is becoming increasingly complex for epidemiologists.
We will increasingly be engaged in public controversy, will be working
with industry, and will be asked to participate in heated scientific
debates about risk or benefit.
It is hoped that the scientific spirit and a reliance on evidence will
guide us through the turbulent times we face. To quote the philosopher
Bertrand Russell, ‘the scientific state of mind is neither skeptical
or dogmatic. The skeptic holds that the truth is undiscoverable; while
the dogmatist holds that the truth is already discovered. The
scientist holds that the truth is discoverable though not yet
discovered (at any rate, not in the matters which are under
investigation). But even to say that the truth is discoverable is to
say rather more than the scientist believes, since he does not
conceive his discoveries as final and absolute. Absence of finality is
of the essence of the scientific spirit. In the welter of conflicting
fanaticisms, one of the few unifying forces is scientific
truthfulness, by which I mean the habit of basing our beliefs upon
observations and inferences as impersonal and as much divested of
local and temperamental bias as is possible for human beings!’
Postscript in 2000
In 1983,
I was concerned with the problem of the “epidemiologist-for-hire” who
thought it professional and ethical to create the “best defense” for
an indicted product or risk exposure in the manner of a lawyer. This
problem has probably not improved as many recent examples can attest.
My current concern with our field is derived from the drastic changes
we have seen in university life so that sharing of data, openness and
free flow of scientific information is threatened by
university/private sector financial arrangements and the desire of the
universities to gain patent rights. This spills over onto epidemiology
where creating private businesses as a result of scientific discovery
is glorified by the euphemistic term “technology transfer” or
“translational research.”
The net result of this lamentable application of the business model to
the university is a stifling of data sharing and collaboration, free
exchange and an exaltation of priority of discovery and patentability.
The elevation of much of the scientifically deficient “alternative
medicine” has everything to do with money and little to do with
actually improving the health of the public. We now have medical
schools practicing homeopathy, endowing named chairs to
self-proclaimed guru/healers who use their own rules of evidence while
the school catalog prattles about its curriculum in “evidence-based
medicine.” I await with trepidation the first Division of Astrological
Healing. Sadly, much of this was predicted by the
sociologist/economist Thorstein Veblen in his 1919 book, The Higher
Learning in America. ■
|