Ethical Considerations for Innovations and
Clinical Trials
Randi Zlotnik-Shaul and Martin F. McKneally
The ethical imperative to improve practice through innovation and research finds justification in the
requirements of physicians to help patients, minimize the harms of treatment and disease, and to bring
the benefits of scientific medicine to those suffering from illnesses for which satisfactory treatment has
not yet been developed. This article discusses the values and principles that underlie clinical trials; the
need for evidence-based knowledge; the difference between research and therapy; the ethical duty to
improve care through research; the challenge of conflict of interest; the relevance of clinical equipoise;
the need for research review and postapproval monitoring; and the value of disseminating results to
research subjects, colleagues, and the public. Public confidence and willingness to participate in clinical
trials will continue if the ethical standards set by professional, government, and international bodies
are met.
© 2003 Elsevier Inc. All rights reserved.
Key words: Ethics, research, surgery, innovation, clinical trials.
A
trial is defined as “the act of trying, testing,
or putting to the proof, a tentative or exper-
imental action to ascertain results.”
1
In a sense,
every treatment is a trial or quasi-experiment
aimed at learning whether the treatment will
work in a particular patient. When Hippocrates
and his colleagues attempted by trial and error to
drain empyema, they concluded that it was safer
to place the drain above or through the rib, avoid-
ing the intercostal artery, and to cover the wound
with a wet dressing to prevent pneumothorax.
When the Hippocratic school repeated, taught,
and wrote about its experience with this treat-
ment, it was conducting and reporting the results
of its clinical trials. Hippocrates wisely reminded
us that “experience is delusive and judgment
difficult.”
2
In recent years, the formal structure
and regulation of clinical trials has increased
their reliability.
3-6
Using measurement as its car-
dinal procedure, clinical research “seeks to ascer-
tain the actual relationship between phenomena.
Uncritical reliance on initial observations, poten-
tially distorted by bias, can lead to a systematic
divergence from the truth.”
7
Few of the major advances in the treatment of
pulmonary disease were made through the appli-
cation of rigorous clinical trial methodology. Like
Hippocrates’ drainage procedure, Evarts Gra-
ham’s successful pneumonectomy for lung cancer
entered practice as a clinical innovation.
8
The
benefits of Cooper’s successful introduction of
lung transplantation for end-stage lung disease
were documented by objective testing of pulmo-
nary function in each patient before and after
transplantation.
9
Although never validated
through controlled clinical trials, this procedure
has become the standard of care on the basis of
positive outcomes.
Validation, the proof of causality and efficacy
of a treatment, can be established in several
ways. Before-and-after testing provides convinc-
ing evidence to validate clear and simple treat-
ments, like analgesics for pain or chest tubes for
pneumothorax. Randomized controlled trials,
highly valued by proponents of evidence based
medicine, are needed to validate the linkage of a
favorable outcome to a particular treatment
when other confounding influences, such as gen-
eral advances in care or spontaneous healing,
might account for the improvement observed.
Small randomized trials can lead to erroneous
conclusions 1% to 5% of the time. Our experience
with immunotherapy for lung cancer illustrates
From the Department of Bioethics, The Hospital for Sick Children,
Department of Pediatrics, and Department of Surgery, Joint Centre for
Bioethics, University of Toronto, Toronto, Ontario.
Address reprint requests to Martin McKneally MD, PhD, 77 Forest
Grove Drive, Toronto, Ontario, Canada M2K 1Z4. E-mail:
martin.mckneally@utoronto.ca
© 2003 Elsevier Inc. All rights reserved.
1043-0679/03/1504-0009$30.00/0
doi:10.1016/S1043-0679(03)00099-6
380
Seminars in Thoracic and Cardiovascular Surgery, Vol 15, No 4 (October), 2003: pp 380-385