I must have written more than 500 expert witness reports over the last 30 years; on a score of occasions I have given expert evidence in a family, civil, or criminal court. I have read thousands of reports by other “experts”, some supporting and some opposing what I have stated. In court, I have heard some of them give their evidence orally. As a member and chairman of GMC Fitness to Practise panels (and certain of their forerunner committees) I have had, with colleagues, to decide if evidence from experts helps us determine the facts of a case. I deliberately used those inverted commas around the word expert.

Especially in earlier decades, I sometimes wondered what entitled certain individuals to describe themselves thus. Over the years I have questioned what has entitled me to do the same. The basic problem is never having been provided with a precise definition. Even when I was a paediatric SHO, I found myself giving evidence in family courts regarding alleged child abuse. I now see it as astonishing that my seniors let such a callow youth large in the judicial bullring and can only hope it no longer happens.

In the very first case in which I appeared there was an allegation that lesions to an infant’s feet had been caused by immersion in hot water. The defence was that tight shoes were responsible. I thought I had gone to court to provide information about my history and examination and the child’s progress in hospital. Instead, I found myself questioned about the temperature and time needed to cause different thickness burns – something about which I realised represented a far away place of which I knew little.

At the end of the proceedings, the court usher asked if I was a professional witness or an expert. I had no idea of the difference between the two or even why she had asked. She explained that the fee would be different. I said I would go for whichever was the greater.

Although I gained more knowledge and experience over the years (and learned that the more we know the less we can be sure) that question haunted me on numerous occasions through the seventies and eighties.

Often I didn’t know which I was and nor did the court. Sometimes I protested mildly that I preferred not to give an opinion but just to state the facts as I had observed them. Not infrequently, counsel, and occasionally the judge, tried to persuade me to be more forthcoming.

Fortunately, my forays into child protection diminished with the years – although as the process became more formalised and fairer I was sometimes asked to offer an independent opinion on the diagnoses of others. Once I went back into the witness box in order to change my opinion after having completed my evidence, because I was convinced by the defence expert that he was right and I was wrong.

That proved to be as profound a lesson as when I was one of three experts in a criminal hearing: a GP agreed he could not provide a strong opinion either way; I stated that, while the injuries in question were highly suspicious, I had grave doubts about their provenance and was not prepared to assert that they must have been the result of a parental assault.

A senior colleague then told the court he had never seen such an open and shut case of child abuse in his career. I could almost feel the judge’s sigh of relief that someone had given him a hook on which he could hang his decision.

The courts, a cynical judge once told me, are not there to decide between right and wrong, they are there to decide.

So I was grateful when the Civil Procedure Rules came along and I began to know where I stood. Or mostly: it is still the case that I have never been cross-examined on my qualification as an expert except on the one occasion when I gave evidence at a US Army court martial.

When providing a report for a civil court, in a clinical negligence case, say, I always provide a copy of my cv. Sometimes it is briefly alluded to when I start my evidence-in-chief, but it has never been subject to cross-examination. I have decided that I am an expert because my qualifications and experience show that I am likely to know more about the matter than the judge or jury.

My US experience was presumably based on that country’s Daubert Rules. This arose from a Supreme Court decision in a disputed case between the parents of a handicapped child and a pharmaceutical company over whether or not a particular drug might have harmed him. So far as I was made to understand, a witness in the US qualifies as an expert by reason of his or her knowledge, skill, experience, training, or education.

Their testimony must be scientific in nature and grounded in “‘knowledge”. That knowledge had to be arrived at by the scientific method. The judge would make a preliminary assessment as to whether certain scientific knowledge would assist in determining the facts, on the basis of whether the reasoning or methodology underlying the testimony is scientifically valid. The judge’s tests might include whether any technique or knowledge relied upon had been tested; whether it had been subject to peer review and published in a scientific journal; and whether the rate of error was known.

Having walked away from child protection litigation, largely as a result of what I had seen happen to paediatric colleagues, as well as being subject to some moderate harassment, I have stuck to accepting instructions from either side in clinical negligence cases.

It did not take me long to realise that, as a general paediatrician, I was entitled to offer an opinion on the standard of care to be expected from its practitioners. But I was sometimes on dangerous ground offering my opinion on causation where that area of knowledge had long been hijacked by subspecialists or academics. This rankled occasionally, given that as a journal editor for several years I had probably read more scientific papers than 95% of doctors.

I thought my day had come when a case involved alleged failure to diagnose a very rare, crippling, but surprisingly easily treatable condition. It had been described first in a single case report in a non-English language journal, then as a small case series in the same language; eventually it reached obscure specialty journals followed by odd case reports in more widely read general journals (including my own – although I couldn’t remember it).

After about 12 years it reached paediatric textbooks and ultimately the gene splitters blew the whole condition apart into its constituent ribosomes. Among the questions to be decided were when would all reasonably competent paediatric neurologists have known about the condition and when would all reasonably competent paediatricians in general hospitals have done the same.

As an editor I thought that was definitely part of my expertise, although I sensed the legal team was doubtful. The hearing degenerated into a goalless draw until being settled after a few inconclusive days, so the questions were never answered. In any case, the decision would have been quickly superseded as knowledge appeared increasingly online.

The Civil Procedure Rules (and since 2006 their criminal counterpart) seem to me to protect both experts and litigants. One’s duty is to the court, rather than to those who instruct you. You guarantee not to go beyond your area of expertise and accept you may be subject to public criticism (let alone a challenge to your registration) if you do. You are mandated to present the range of opinion that exists on a subject rather than your own take on it. Meetings between experts instructed by both sides sometimes lead to having to concede in advance, rather than under cross-examination, a much more comfortable stance.

Last week’s GMC guidance on expert witnesses is clearly based solidly on both the Civil Procedure Rules and its own publication Good Medical Practice. It could have wider application than the GMC might think. For example it occurs to me that authors of scientific papers would do well to study Paragraph 10: “you must take reasonable steps to verify any information you provide and not deliberately leave out relevant information,” and paragraph 12: “Wherever it is possible to do so without being misleading, you should use language and terminology that will be readily understood…”

Gornall’s BMJ feature raises issues about expert witnesses appearing before GMC panels. Over the years I have come to the conclusion that some experts providing reports or giving evidence in court are highly knowledgeable, easily understood, and scrupulously fair.

Others do not necessarily share all these criteria in spades. That is unlikely to be any different for those appearing before GMC panels, whether called by the GMC itself or by the defendant doctor. Counsel who is on top of his or her case and skilled in cross-examination should be able to avoid a jury (or a panel) being misled by an inadequate expert. As an added safeguard, GMC panels have the advantage of at least one, and often more, medically qualified person being present.

I infer that, as a matter of course, GMC Legal, or solicitors whom they appoint to prosecute a case, will forward the new guidance to any expert whom they seek to instruct.

Competing interest: Harvey Marcovitch is an associate of the GMC, chairing Fitness to Practise panels. The views expressed are his own and do not represent those of the GMC. A requirement of his contract with the GMC is that he provides a pre-publication copy of material he writes which refers to the GMC to that body for its approval for publication.