Science of practising medicine

DOCTORS have always been men and women of science. While no definitive timeline exists, it is possible to point to two distinct eras in medicine.

The first is the “expert-based medicine” era whereby doctors learned by observing their own patients and applying that knowledge to treat future patients. They also passed on their knowledge to their juniors. However, their juniors had no way of verifying that knowledge. Medical education was by apprenticeship and not by inquiry. Knowledge had to be accepted as is.

By about the 1950s things began to change. Doctors realised that they could pool their knowledge (multi-institutional studies) into databases. They could then query these databases by asking specifically crafted questions (scientific method/research methodology) which could then be answered by statistical analysis of the database (statistics).

This development gave rise to the modern “evidence-based medicine” (EBM) era. The engine that powered this revolution is what we call “research methods and statistics” (RM&S).

The benefits of EBM quickly became obvious. Suppose one wanted to try Drug A to treat baldness. One would assemble two similar groups of 100 patients each. Give one group Drug A and the other group a placebo.

In the Drug A group, hair grew in 30 patients and in the placebo group, hair grew in 10 patients. So, now we have three distinct categories of patients – 10 patients in whom hair grew without treatment, 20 patients in whom hair grew with Drug A and 70 patients who did not respond to Drug A.

Now, this information by itself is not helpful. Enter EBM. We can “profile” the three categories of patients so that when a new patient sits in front of us we can decide which category he “fits” into.

If he fits the first category, he likely needs no treatment as his hair will grow spontaneously. If he fits into the second category, he will likely benefit from Drug A and if he fits into the third category, his hair will likely neither grow spontaneously nor will he benefit from Drug A.

The benefits of this knowledge are enormous. Firstly, we will be able to advise against Drug A for those not likely to need or benefit from it and spare them the injury from adverse effects of Drug A. The patient now has some scientific basis to accept or reject our recommendation. Secondly, there will be large savings in terms of the financial cost.

One would think the medical community would have rushed to embrace this new development. Unfortunately, this was not the case. The RM&S component was a sticking point. Older doctors who were powerful within the medical hierarchy were reluctant to relinquish their “eminence” to a database which could disprove their opinions,

Further, mastering RM&S skills was a difficult task. They were used to the apprentice-based research publications system which has sometimes been uncharitably referred to as “Mickey Mouse” research to “decorate curriculum vitae”.

In the new EBM era one had to learn a wide range of study designs and statistical methods in order to be able to decide which answered a clinical question correctly.

Medical schools started teaching RM&S around the 1960s. It wasn’t until the 1990s that EBM became a formal part of the specialist curriculum and an inalienable part of medicine.

In many developed countries, clinical specialist training (CST) is entrusted to standalone dedicated institutions like the Royal Colleges in the United Kingdom. These are bodies with statutory licensing and regulatory powers. They were allowed to confer “degree-like” qualifications. Because they were publicly visible and hence held accountable, the Royal Colleges had to accept EBM and RM&S training is explicitly included in the specialist curriculum.

In Malaysia, for some reason, the CST programme was entrusted to universities and referred to as the Clinical Masters programme. For yet other reasons, probably because the course is nestled within large multi-disciplinary universities and shielded from public view, RM&S training has not been an explicit and/or consistent part of the curriculum.

There is no uniform requirement for clinical specialist trainees (who need to be able to read research) to be taught and tested on basic RM&S and there is no requirement for their lecturers (who need to be able to do and supervise research) to be taught and tested on advanced RM&S skills.

Without these qualifications being a routine part of the medical ecosystem, we will not have a “critical mass” of specialists to perform vital tasks like doing research, peer reviewing research, sitting on medical research ethics committees, teaching research or writing clinical practice guidelines (CPG) amongst other things.

The public will view our opinions, CPGs and scientific comments with suspicion as it is not backed by any prior formal training.


Histopathologist and statistician

science , letters , opinion