Abstract
This should be the last time.
After three-and-a-half years (not long by previous standards) of editing the Scottish Medical Journal, I’m hanging up my blue pencil.
I am not entirely sure why. Certainly, it has taken up more time than I expected. And after forty years of clinical medicine, with the mixed-metaphor monkey forever looking over your shoulder, it’s a surprise to find something as gentle as journal-editing produces a similar feeling of unease, an inability to be at peace, a persisting niggle at the back of the brain that there is something needing to be done.
To be fair, it’s a monkey I like dealing with. I like the great variety of submissions that come my way (130+ in the first half of 2019) from all corners of the globe (might make mixed-metaphors a theme …). Assessing them, spotting possibilities, working on them with authors; all rather enjoyable. But there are frustrations. Access to the journal for “subscribers” seems uncertain. The mechanics of the submission procedures are not always forgiving – for authors or “editorial”. Reviewers are increasingly scarce in the current NHS climate where the specialist’s week is programmed to the last nano-second. And causing delays for authors, when I feel a paper could eventually be published as long as there is no specialist flaw, certainly feeds the monkey.
However.
If I’m honest, I think the main problem is that I don’t understand the current world of medical publication., of dissemination of medical information as a whole. Or, at least, I don’t fit in with it.
Since beginning the editorship, I’ve been receiving – maybe you all do – emails from medical journals such as this:
Dear Dr Larkin. We get to know your published article with the title of Brave new world in SCOTTISH MEDICAL JOURNAL and the topic is quite interesting.
So we want to invite you to contribute other articles of related fields to the journal.
The SMJ received an odd submission a few months back; a paper that had already been published in one of the “Sub-Natures”. My first diagnosis was … naïve plagiarism in the extreme, but the author’s name was the same. USA-based. Some mistake, surely.
“Why have you sent me this?”
“It’s a submission for your journal.”
I checked the author’s previous publications. Many were about the “dubious” nature of some medical journals. It seemed likely we were being “checked out”, perhaps as part of a new study. I’ll never know if we passed the test; my polite “bouncing” of the paper, with attempted follow-up for clarification of the author’s intentions, met with silence.
And my worry is; now that the SMJ is a largely on-line journal (print run of approximately 96 copies), are these potentially fly-by-night publications what we are competing against? Does this explain the occasional author’s surprise, even irritation, at the questions or stipulations coming from myself or a reviewer?
On seeing a biologically unlikely correlation of >0.95 (between a blood test and a dynamic physiological measurement) in one paper, I requested a dot-plot. I received a graph with a different r-value, which was incorrect for the actual dots. Declining to pursue publication, I was met with major umbrage, including an assertion that I had “no right” to request data not present in the original submission. They would not be using us again, I was abruptly informed.
Nearer home, a huge survey had one or two very counter-intuitive results. When I queried the way the questionnaire had been set up, the senior author told me that whilst they were happy to entertain questions where I had contrary evidence, they took a dim view of reservations simply because I did not agree with their results … though they concurrently admitted there was indeed a glitch in their survey which caused the counterintuitive findings, and would affect other major results …
The authors still expected publication. As one co-author informed me, I could be assured that the survey had been performed by “leaders in the field.”
Which brings me to another worry. The power wielded by leaders in fields.
I remember attending one “guideline-generating” symposium in a rheumatology sub-specialty a few years back. Eight different topics, eight different groups doing a carousel of workshops over two days. Each workshop (not group) with a stationary “leader”. My first workshop had a surprising statistical no-no as its basis. A voice in the group pointed it out, to ensure it would not be the basis for the next seven groups. When we all came together at the end, it was clear this anomaly had been perpetuated, perhaps so the final “consensus” guideline would come out as planned by the powers-that-be.
All sub-groups have such gurus. There was one conference where the undisputed guru showed that their new management technique reduced the incidence of a particular complication from 30% to 15%. At question time, a voice in the audience pointed out that starting off with a very poor “baseline” is one way to get good improvement, and pleaded with anyone in the room to speak up if they had experienced an incidence of this complication of anything even approaching 30%. Silence. But an awkward silence. Faces to the floor. And the guru carried on to the next question as if nothing had happened.
In these days of “evidence-based medicine”, we are still being told what is “evidence-based” by a small number of gurus. Only, instead of hanging on every word coming from the lips of our mentor – whose clinics and ward-rounds we have attended, and whose qualities we have been able to assess – we now make our clinical decisions based on the prognostications of a total stranger who has risen to the top in his/her field almost certainly not because he/she is the finest clinician on the planet. We are told to follow guidelines written by a small group of practitioners who have time to do that sort of thing before jetting off to give another talk outlining their juniors’ most recent research findings. We are to believe a “meta-analysis”, of 167 papers, whittled down by the meta-analysers to 7 that fulfil their criteria and will give the definitive answer. Meantime, another meta-analysis of the same question will use other criteria and come up with 15 overlapping papers that give a slightly different definitive answer. And we ourselves don’t really see any of the numbers. The actual data. They are filtered by the middle-man because we don’t have the time. So we trust the middle-man to do it for us.
I like to see the raw data. Dotplots or similar. So we can see whether it works. I attended a cardiology conference once (don’t ask) where one therapy was compared with another. Raw results were that A was better than B, but once “we did the regression analysis for confounding factors”… B was actually better than A! Which was what the authors had originally expected. Everybody seemed happy to accept this. A voice from the audience asked what the factors were that differed between the groups to cause this anomaly. “I don’t know. I’m not a statistician.”
Well bring one along! Don’t expect us to accept blindly that your bottom line is justified by all of the counter-intuitive top and middle lines. Let them explain to us. They should be here. If you’re not interested in your own stats, and they aren’t interested in the clinical nuances of the outcome, any little quirk in your computer programme/input will be noticed by neither of you.
I once attended a mock Scottish Medicines Consortium (at that time, the SMC decided which drugs would be accepted for use in NHS Scotland) meeting at the behest of a drug company, to assess their proposed presentation. One “slide” of their separate computer-run Markov Model results (1,000 simulations of 1,000 imaginary patients’ progress using research data on drug effectiveness, side-effects, mortality, costs etc.) didn’t look right. I mentioned this. Carry on. Three slides later, I had the information to prove to the satisfaction of the economists present that the computer-run simulation of 1,000 … etc. had made a mistake. Which it had.
So I have an inherent suspicion of “sophisticated” statistical methods that you don’t actually see in action. This may have resulted in many recent SMJ original articles consisting of simple ideas that can be investigated locally. And any reviews hopefully taking some controversial (anti-guru?) stance – though the submission rate for “In My Opinion” has been disappointing. Hopefully, the SMJ will remain an outlet for such papers, as well as for case reports – a medium now beneath the dignity of many of the more celebrated journals. There’s a place for quirky ideas. And it’s nice for that place to be governed by peer review without the author seeking sponsorship to pay $1500 for the privilege of publication.
But that will be for others to provide.
It has all become a bit much. The voice in the audience (yep … it’s always me) feels like a voice in the wilderness, full of sound and fury, signifying nothing, wasting its sharpness on the desert air (New theme. Mixed quotations).
But I’ll miss it.
… back at the wrench, this quarter’s issue kicks off with an assessment of potentially unnecessary LFT measurements perioperatively in cancer patients. The second “original Article” will catch the eye of many of us who have ever been called upon to decide whether a patient should be “detained” in hospital because of their mental state, only to find ourselves totally out of our depth. The last comes from GP-land, and involves a double-take on whether an adopted policy, “successful” in reducing benzodiazepine prescribing, actually made any difference.
Case reports include an unsurprising – to us physician-types – example of surgeons being replaced by a “robot”, and a patient with a well-documented diagnosis of having Von Willebrands’ Disease -which they don’t; followed by a patient admitted because they have nephrotic syndrome – which they don’t. A nice one, this, where sensible self-treatment hides the diagnosis from your doctors.
And … to finish … a book review!!
As we go towards press, the most recent “Impact Factors” for medical journals have been announced. The SMJ has gone up from 0.475 to 0.678 – the highest it has been in the ten years list that is available to me. Maybe somebody is listening.
I should here also thank those who have helped me in my short sojourn in the world of medical publishing. High amongst these would be “my” deputy editors. Both Jill Murie and Colin Geddes stayed on to help me through the earliest days, while Tom Pullar stepped manfully into the breach on their departure, giving welcome support on pretty much every paper that came through the portals of the SMJ. The difficulty sourcing reviewers has meantime served to highlight the major contribution of those who did help. And my thanks go to all of them. Perhaps too many to mention, but they know who they are.