Reporting the future

One thing that really irritates me is the new habit of the media of reporting things that haven’t happened yet.  This applies particularly to speeches made by politicians.

The report will read something like “Mr Cameron is giving a speech tomorrow on terrorism/the economy/the EU.  He will say that…”

What is wrong with reporting the speech after it has been given? After all, the following day there will appear another report which says more or less the same thing. Except of course it will now be in the past tense.

I suppose one might argue that if a draft of the speech is issued beforehand then whoever is giving it has a chance to make some prior modifications depending on the reaction.  However, in the recent case of Ed Miliband, giving his leader’s speech at the party conference in 2014, all that became apparent was that he forgot to include some of the most important things.

I cannot believe that we need news so instantly that it appears before the event has occurred.  It is a form of hysteria.

Whatever the rewards, you can only bang your head against a brick wall for so long

My NHS hospital, Queen Mary’s, was built in 1974 to replace an old WW1 hospital (of considerable importance, but that’s another story) which had been run up in 1917 with the standard construction of asbestos sheet. When it was demolished the spoil was removed from the site.

Or so it was thought. When the old site was being surveyed (probably, though I am not certain, when it was thought ripe for development that would fund a growing deficit – though not for long) it became clear that something was up, as the entire area was suddenly fenced off and notices appeared warning that the site was contaminated.

I enquired further and discovered that test borings had revealed the existence of substantial quantities of asbestos which had obviously not been removed, but had been ploughed in.

The hospital archives contain a wealth of material about the redevelopment and without much trouble I found the original contract for demolition. It was written in black and white that any hazardous material was to be taken away. Clearly this had not happened.

I wrote to the Chief Executive pointing this out. I had established that the contractors were still in existence and suggested that they be sued for the sum now required (several million pounds) to clear the site properly.

Nothing happened.

I wrote again, and to the legal department suggesting that as the contract failure had only just come to light the statute of limitations would not apply.

Nothing happened.

Notwithstanding the land being possibly Green Belt, and thus unusable anyway, the opportunity to clear up properly was passed up. I gave up.

When threatened, respond with facts

When I was President of the British Society for Rheumatology I was telephoned by a colleague at another hospital who told me that their department was to be closed, as the local PCT had decided not to fund patient referrals.

I paid a visit. I asked for some departmental statistics, like numbers of new patients, follow-ups, casemix and so forth. They had none. Surprised, I asked how many patients were on biologic agents (at that time all such patients were being entered on the BSR Biologics register). They didn’t know. By now appalled I suggested that they needed to get this information, and pretty damn fast, for otherwise they had no way of defending themselves.

I myself have kept some sort of database since I started as a consultant, beginning with an exercise book noting names of new patients, their diagnoses and GP. As a result I was able to provide myself with some simple statistics, and indeed a collaborative paper came out of it (1) . I stopped recording when it became apparent after 10 years that the casemix was not changing much (although that itself was to change later) and by now sensible computers had come along, so I ditched my BBC B and migrated a follow-up database onto a machine in the hospital’s Respiratory lab. It was huge. It had a gigantic hard drive of 20Mb! This was 1985… Later still when Microsoft released Excel and desktop machines came to outpatients I developed a follow-up database listing patients by name, birth date, date first seen and date most recently seen (to calculate follow-up duration), and drugs used and date to next appointment (to calculate follow-up interval). As part of this I was able to check the mean methotrexate dose, biologic exposure and all sorts of other things. There were separate sheets for little projects such as reviews of new patients and their outcomes (useful for assessing who was to be followed up or who was “one stop”). Three further presentations followed (2,3,4) .

But the bottom line was that when any managers came by asking me to change what I did, or asking – usually with a furtive grin that suggested they knew I wouldn’t be able to answer their query – I had all the information at my fingertips and could provide better and more useful statistics than the hospital database. Mostly they slunk away with tails between legs to lick their wounds.

The moral of this is that if you want to win battles you need to prepare for them better than the opposition. I have suggested to medical students that two important books that will come in useful are Machiavelli’s “The Prince” and Clausewitz’s “On War” (5). Bernard Montgomery was a great proponent of the sand table when planning battles. Presentations should be well prepared and rehearsed. No data, and no background, and you are dead.

[1] Bamji AN, Dieppe PA, Haslock I and Shipley ME. What do Rheumatologists do? A Pilot Audit Study. Br J Rheumatol 1990, 24, 295-8

[2] Litwic A, Bamji AN. Follow-up or discharge? A new patient outcome analysis. BSR Annual Meeting 2008, Abstract 417

[3] Bamji AN, Lane J. Impact of a community-based rheumatology clinic on a hospital department. BSR Annual Meeting 2010, Abstract 96

[4] Bamji AN. New: follow-up ratios: Dogma or Design? BSR Annual Meeting 2011, Abstract 59

[5] I was accosted by a trainee while I was collecting some stuff from the Postgraduate Centre after I had retired. He asked if I was Dr Bamji, and on my confirmation told me how he had been much influenced by the after-dinner talk I had given in Cambridge four years earlier, and both books were now in his library. I was much flattered.

Planning clinics

This may not apply to every doctor but the statistical lesson is an interesting one.

Many folk in specialties set up subspecialist clinics to deal with particular conditions. Thus in rheumatology you can find early rheumatoid arthritis clinics, lupus clinics, combined clinics with orthopaedic surgeons or paediatricians etc.

There is a nice Powerpoint presentation which proves that this should be avoided if you wish to be efficient (1). Subdividing your work will increase your waiting lists and you will end up with some underbooked clinics. This is queue theory at work.

Do not be afraid to overbook clinics. There has been repeated blathering about the problem of patients who fail to attend their outpatient appointments. The Department of Health wants us all to stop patients not attending, because it is wasting the NHS £700m per annum. Actually a “Did Not Attend”, or DNA for short, costs nothing (2). People overbook their clinics to compensate for DNAs, just like the airlines do; what’s good enough for EasyJet is good enough for me. If everybody does turn up you have a hell of a clinic but it doesn’t happen often, and a no-show saves a lot of investigation, and no-one is actually charged for them (it would be different if they were).

Anyway – why do patients DNA? I audited this about 15 years ago and doubt much has changed. Many never got the appointment in the first place (sent to the wrong address, postal failure etc). Some cancel, but this does not get through to the clinic. A tiny number forget. Many have some other pressing commitment, such as a family funeral, a relative in hospital somewhere else, a childcare problem. Or the car broke down. I fell these are all quite reasonable excuses. Given the overbooking bit there is no need to spend vast sums on armies of clerks who are going to chase the patients up the day before, whether by phone, email or text message.

However there is a curious DNA issue which I call the endless loop syndrome. Let’s suppose a patient cancels and the clinic doesn’t know. What happens next is as follows:

• Let’s say the patient rings to cancel the day before the appointment (January 30th), and is rebooked for 5th March
• The patient doesn’t appear in clinic, so the clinician completes a DNA form, which generates another appointment booking. However this takes a day to process, and is for 12th March
• Patient receives the letter confirming their own rebooking
• Two days later they receive a second letter generated through the DNA, for 12th March. The letter says “Due to unforeseen circumstances your appointment with Dr Bamji has been changed to the 12th March”
• The patient thinks this change refers to the appointment they made for the 5th, but this remains on the system
• On 5th March the patient doesn’t appear in clinic, so the clinician completes a DNA form, which generates another appointment booking. However this takes a day to process, and is for 27th April
• The letter confirming this is received by the patient on the 11th March, and they assume it refers to the appointment the following day, so they don’t turn up
• On 12th March the patient doesn’t appear in clinic, so the clinician completes a DNA form, which generates another appointment booking. However this takes a day to process, and is for 9th May

And so on.

Moral: If a patient fails to attend make sure that there is not some later extant appointment. Eventually they will complain at being continually postponed and wonder what they have done to offend you.


1. See for some demand management presentations, including queue theory
2. Letter in “The Times” from Yours faithfully, 18th December 2010