Any lingering hope that Parliament’s Health Committee could hold the GMC to account at its annual accountability hearing earlier this week over doctor suicides while under GMC FtP investigation has now disappeared as sand in the desert. Though delayed for a few days, the transcript of the hearing is now available online. It makes grim and depressing reading. On the matter of FtP doctor suicides, the Chair of the committee – a doctor – came to the proceedings with a regrettably light grasp, if that is not too strong a term, on the facts. Stilton, the GMC pongo who answered the questions on these deaths, turned out to have an even lighter grasp. Indeed, so poor were these grasps that Dr No has no hesitation in saying that they were, and are, an insult to democratic process.
There are two key documents: the ‘internal review’ carried out by the GMC’s ‘independent consultant’ on FtP deaths, and the transcript of the ‘accountability’ hearing, notably the section marked Q40 to Q47: the former is here, and the latter here.
The Chair opened gamely enough, but loosely, getting the numbers wrong. Taken together, Q40 to 42 make it clear she believes there are a total of 142 (114 other deaths + 28 suicides) deaths in the report. The actual number is 114 in total, of which 28 were suicides (24) or suspected suicides (4). Stilton, for some bizarre reason – less deaths would after all work in his favour, did nothing to correct this error – so it persists. Perhaps Stilton isn’t as au fait with the facts as he might be. We shall have to wait and see.
The other thrust of questions 40-42 is the problem of under-reporting of suicides. Again, Stilton gets the numbers wrong, and then compounds the matter by saying “A small number of [the non-suicide] doctors may have been suspected suicides [wrong – the suspected suicides (4) have already been counted and are included in the 28 figure] but the vast majority [the actual numbers are 114 – 28 =86, or 75.4% – a majority, certainly, but vast majority?] of the others were, as far as we know, not suicides.’ He then adds: ‘ We did not have any evidence that any of those doctors had committed suicide”. Sorry, buster, absence of evidence (of X) is not the same as evidence of absence (of X) – basic science, dear boy. Some of the 86 may or may not be suicides – we simply don’t know. We don’t know how many fatal single vehicle accidents, drug overdoses and alcohol deaths, to name but a few of the causes that may be ‘masked’ suicides – real suicides masked to appear otherwise – occurred among the 86 deaths.
The Chair then turns, through Questions 43-46, to suicide rates (as opposed to numbers, which between them they have managed to mix up), and how they compare between FtP doctors and ‘comparable groups’. This is not necessarily that complicated. Back in 2013, Dr No was able, using standard statistical techniques, to show that all cause mortality among FtP doctors was, depending on comparator group, between 15 (compared to all working age adults) to 30 (compared to socio-economic class one) times more than expected. He could not, however, do any assessment of suicide deaths simply because the numbers were not publicly available back in 2013 (the GMC knew – “we do a significant event review every time a doctor commits suicide” – but it wasn’t letting on).
Now we do know how many FtP deaths there were, we might expect Stilton to come up with a straightforward answer to the Chair’s comparable rates question. Instead, he came up with Obfuscation Max. “If you compared a period of time running up to now for doctors who died within the GMC procedures with the wider population of doctors who commit suicide – I do not have the exact figure – around 19 would be in our procedures and about 120 had committed suicide, as it were, not in our procedures, or not that we were aware of.” Dr No is still trying to work out – he does not have the exact figure – what on earth that actually means.
The Chair to her credit spotted that Stilton had swapped counts for rates, and persisted with her comparable rates question. Stilton responded by going into Obfuscation Turbo Max. Avoiding answering the rates question directly, he opined that, well (to summarise) it’s sickos that get caught up in FtP procedures, so why are we surprised if they top themselves? Instead of rates, there was a lot of ‘absolutely’ in the field of play, as in “Absolutely, certainly’ and ‘Absolutely, it would be’. Another variant was ‘We absolutely will do that’, but throughout one thing remained absolutely absent: any specific mention of actual rates.
Putting aside the nonetheless crucial duty of care question for now, as it happens Dr No absolutely can, and indeed begs to, help Stilton and the Health Committee with the FtP/comparable suicide rates question. We know from the ‘internal review’ there were (at least) 28 suicides between 2005 and 2013 (inclusive), so an average of 3.1 per year (the actual numbers are very volatile, which is why Dr No uses the average). We also know (from GMC annual reports) there are something just under 300 new FtP investigations a year. If we assume investigations last one year on average (Dr No is keeping things simple) then that is near enough a suicide mortality of 1%, or 1000 per 100,000.
We also, as it happens, have some idea of the suicide rate per 100,000 in doctors at large, so to speak. Stilton is right in saying that rates appear to be declining – traditionally, doctors were counted as high risk for suicide, with knowledge of, and access to, the means at least the partial explanation, but the most recent figures (which are still not that recent) suggest this is nowadays less the case. The most recent report Dr No could find – see bottom part of Table I on page 1235 here – based on ONS England and Wales data, shows an all doctor suicide rate of 11.4 per 100,000 worker-years (worker-years are a way of rolling together denominator numbers and time: one doctor for ten years counts the same as ten doctors for one year) for the period 2001-2005; for comparison, the general adult suicide rate at that time was around 8.5 per 100,000. Other earlier doctor suicide rate data do exist, including some breakdowns by sex, but are less helpful given that rates have changed over time.
So, in summary: the general doctor suicide rate (albeit from 2001 to 2005) is 11.4 per 100,000; the FtP doctor suicide rate from 2005 to 2013 is 1000 per 100,000: almost one hundred times the general all doctor suicide rate.
Ah, but what about Stilton’s well, thet’re are all sickos anyway line? Sorry, no dice. Although a bit rough and ready, consider this. We know – see paras 10 and 17 here – that roughly one quarter of all suicides have had contact with mental health services in the twelve months preceding death. If – and it is a big if – a similar proportion applies to FtP suicides (and we accept contact with mental health services is a marker for mental illness) then that leaves – to borrow one of Stilton’s choice phrases – a ‘vast majority’ of the FtP suicides, three quarters, are not related to or explained by pre-existing mental illness.
Readers should note that no age/sex adjustment has been done (and may be, on the current data collected, impossible to do): these figures are the crude rates. It should also be noted that (in statistical terms) the actual numbers are small, and so we should expect, and indeed see, considerable year on year variation. But, even with these and other caveats we are still faced with an FtP suicide rate that is a hundred-fold larger than the all doctor suicide rate. Time to wake up and smell the poison, guys.