The morning was devoted to the four-weekly checks by those lovely people at OCDEM, whilst the afternoon (or part of it) was taken up with a six monthly diabetes review by the practice nurse.
It was all change at the Churchill, as the previous study nurses, Lynne and Judy, had retired to be replaced by Clare and Nicky. There was the usual problem with getting blood which was tackled by the application of warming pads (basically mini-electric blankets) to both arms. Breakfast, which had previously been served on a tray at my bedside, was now to be taken in the patients’ kitchen with a couple of Biobank volunteers under a poster urging us to get to know our duodenums.
So far. So good.
The appointment with the practice nurse started well. My blood results were wonderful; my cholesterol level was exemplary; my liver and kidneys were up for a couple of awards and my HbA1c was significantly down. HbA1c is something boring-but-important that sits on your haemoglobin and is used to work out your average blood glucose over the last 6 months.
Practice Nurse: “Your Hb1Ac is down to 6.7 – that’s really good”
Me: “That would be the exenatide”
PN: “What exenatide?”
Me: “The exenatide that I’m injecting for the Sustain study”
There only thing on my record was the briefest of notes from one of the GPs giving the name of the study and the two drugs involved. I brought her up to speed and she updated the record.
Just as I was about to leave, I noticed the worrying word “insulin” on the screen (worrying, because I don’t use insulin). Confusingly, this was there to indicate that I was injecting something, not necessarily insulin.
As far as I can see, this was a problem with the data entry interface. There was a drop down list of common values for this particular field (entitled something helpful like “Additional Information”), creating the impression that these were the only permitted values and “insulin” was deemed to be the best option from the selection on offer. A bit of experimentation established that you could actually type in free text and it was duly amended.
(Personally I would always have an “other” option with a separate field to add detail and then use this to highlight any gaps or changes in the standard options (or problems with recording) – though why “insulin” or, indeed, any medication being supplied by a third party isn’t listed explicitly under “medication” or “other medication” is beyond me. Just saying)
Which brings us back to data sharing. Never mind CareData, what about the NHS Summary Care Record (SCR) or, heaven preserve us, the Oxfordshire Care Summary (OCS). The SCR is for use by emergency or out of hours staff and contains basic information about medicines and allergies and is clearly a good thing, as long as the information is accurate (and complete). The OCS allows healthcare professionals to read standardized information drawn from multiple databases (but only ones in Oxfordshire).
The assumptions behind all these NHS data sharing schemes is that all the relevant information is on the databases in the first place, that it’s accurate and that all the data selected for sharing will include all the relevant information. That’s an awful lot of assumptions, especially with an organization as large and complex as the NHS. Add to that the fact that many treatments and medical conditions are excluded and the whole thing starts to look even flakier.
Nothing that happened today was a failure by any of the individual (and highly conscientious) professionals involved, but it did highlight how changes to one part of a system can cause other parts of the system to destabilize. It’s a bit like a giant game of Jenga – except that the point of failure is when you’re whipped unconscious into A&E after tripping over next door’s cat.