|
Project ImpACT was initiated in 2003 at
Johns Hopkins University after much preliminary discussion
with colleagues and other experts around the world.
The first goal of Project ImpACT will be to develop an edited
book describing the most important randomized controlled
trials (RCTs) performed in all disciplines of medicine and
public health since 1948, the year of the first large-scale
RCT in medicine, the British Streptomycin trial. The RCT
is universally recognized as the most valid scientific tool
medical researchers have to assess the relative benefits
of virtually all therapeutic or preventive interventions,
yet there is currently no comprehensive source that documents
the contributions of landmark clinical trials in all domains
of medicine. The book will aim to fill that gap.
Longer-term goals include developing a web-based resource
and community for the continual monitoring and documentation
of important new clinical trials in medicine as they appear,
and conducting oral-histories of persons who participated
in landmark trials in the 20th century.
Project ImpACT will publish products that not only summarize
design and results of the ImpACT trials but will also describe
what cannot typically be found in published trial reports;
the state of knowledge or practice before the trial was reported,
the political and scientific context in which the trial was
developed, the various design choices that the investigators
considered and the battles that may have been fought to get
the trial designed the way it was, the reasons it had an
impact where previous trials may not have, and the nature
of the immediate and longer term reactions to the trial conduct
or results. The ultimate goal is to learn from these trials
in ways that will teach us lessons that are relevant to present
struggles and tough questions, and be informative and interesting
to clinicians, medical researchers, methodologists, policy
makers, and accessible to the scientifically literate lay
public.
Return to top of page.
Project ImpACT comprises a coordinating
group at Johns Hopkins University, a network of clinical
experts and methodologists within each field of medicine
to help nominate the studies for inclusion and identify
contributors, and an international, interdisciplinary advisory
board that will help select trials from among those
nominated.
Return to top of page.
Nominations for trials will be solicited through several
means including via this website.
Nominations are also actively sought from an international
network of field experts. As nominations are received, these
experts will be acknowledged on the Project ImpACT website.
Review of nominated trials will be handled in a manner
similar to a medical journal. The Coordinating Group will
review nominated trials and choose the most promising for
project staff to prepare review packages. If there is some
doubt, the trial will be sent to the original nominator or
Field Experts for further input. The review packages will
include the main publication from the trial, trial details
regarding participants and outcomes, categorization of the
types of impact and the evidence for this impact. Trials
will be identified as to types of impact; namely, impact
on:
- Practice of medicine or public health
- Methods of designing, conducting or analyzing trials
- The course of subsequent research
- Understanding of pathophysiology or biology
- Regulations, law or policy
- Ethical understanding of trials
The evidence to be collected will depend on the types of
impact made by the clinical trial. Evidence to be sought
could include:
- documentation of changes in medical or clinical trial
practice or in policies;
- information on prescribing patterns (for drugs);
- insurance claims information;
- number and type of citations, including editorials;
- follow-on trials (or absence of follow-on trials);
- roles in meta-analyses;
- books about the trial;
- references in major textbooks or review articles;
- prominence of the trial in guidelines or government report
recommendations;
- coverage in the media.
Return to top of page.
|
