Solutions for project design

The problem:

a) Doctors, ie. MDs, or other project coordinators, often have brilliant ideas which have a sound theoretical basis which, if implemented correctly, could give a scientific breakthrough.  Sometimes, however, there are problems in setting up the project design for a clinical or experimental study correctly eg. to optimize scientific output by using correct control and numbers of patients.
b) This is complicated by the fact that the choices of design for clinical studies are far more complex than for experimental studies in other sciences, and these decisions are frequently made suboptimally ie. that the studies are done with suboptimal scientific output, eg. a “case-control” study is done when in fact a “double-blind” study was possible. Doctors get used to doing one type of study and continue with this even when a more advanced type of study is possible, resulting in sub-optimal science.
c) Two fundamental causes of the above are: (a) the mentality behind modern (ie. correctly controlled) research takes many years of experience to obtain, time which is not available to an MD, and (b) the Doctor is forced to make value judgements at the level of individual patients, whereas a scientist must make value judgements at the level of groups of patients and/or controls.

Ioannidis2005 – Why most published research findings are false.

Ioannidis2014 – How to make more published research true.

A solution – the goals of CMB.

BEFORE a project is submitted to other committees the MD (or other project coordinator) will be allowed to approach, and work with, a panel of scientists who will work with the project coordinator to achieve optimal design in two areas: (1) the design of the clinical research and (2) the statistical aspects of the clinical research.

We intend to facilitate the above using interactive internet questionnaires and, for example, tools for power studies, in order to allow efficient internal scientific audit – ie. BEFORE SUBMISSION OF GRANT PROPOSALS.

This will result in a quantum-leap in scientific standards at the University, will result in less design errors and will give more significant results to be published in journals with a higher impact factor, thus resulting in an increase in funding for the University.

Advertisements