This site is intended for health professionals only

To B12 or not to B12


Surely I cannot be the only GP who finds him/her/whatever-self sitting at work late into the evening poring over blood tests I didn’t order, done for Gawd-knows what, and which the rules of when-the-music-stops mean are my responsibility to sort out? And that these blood tests, these days, almost invariably involve borderline low B12 levels?

So it was with a feeling of utter despair that I discovered that MHRA is advocating more stringent testing of B12 in patients on metformin. Yes, I realise metformin can cause B12 deficiency, and yes I realise that can have serious consequences, but most of all I realise that such serious consequences from B12 deficiency are vanishingly rare, especially in the asymptomatic, and that all these borderline readings are driving me mad. And no, that doesn’t mean I need my own levels checked.

Basically, B12 is the new vitamin D: increasing publicity, alleged deficiency causing no or vague symptoms, indiscriminate testing, borderline levels common and treatment making sod all difference – except, with B12, rather than treatment being a pat on the head and some OTC advice as per Vit D, it often involves injections and demands from patients that they should have higher doses or more frequent jabs on account of, guess what, not feeling better.

As a very intelligent and rational neurologist said to me about the current B12 obsession – and if anyone knows the relative likelihood of functional tingliness versus subacute combined degeneration of the cord, it’s him – ‘It’s the sort of thing that makes you want to retire.’

It’s not as if the labs can even agree on the normal range. And even if they could, no one can agree on the accuracy of the test, when treatment is needed, and what they’re trying to achieve.

And now, according to the guidance, we should continue ‘periodic’ monitoring even in the absence of symptoms in metformin-poppers who are at risk. Which includes, apparently, the elderly or those on a PPI, aka ‘everyone’.

So if your mouse is hovering over a haematology path lab form and you’re thinking, ‘To B12 or not to B12’ – please don’t.

Dr Tony Copperfield is a GP in Essex. Read more of Copperfield’s blogs at


Patrufini Duffy 24 June, 2022 11:18 pm

I heard there’s a nearby PCN social prescriber, that directs patients to best local supermarket and gives coupon offers on:
1. Animal liver and kidneys · 2. Clams · 3. Sardines · 4. Beef · 5. Fortified cereal · 6. Tuna · 7. Fortified nutritional yeast · 8. Trout.

Mark Howson 27 June, 2022 2:09 pm

It is an inaccurate test so any borderline results can be ignored in an asymptomatic patient . I was also taught never to do it in asymptomatic patients unless there was a clear anaemia with macrocytosis. I guess if we are forced into it then the treatment for persistently low B12 without symptoms is dietary advice

David Church 28 June, 2022 11:15 am

Let’s skip the GP part altogether, avoid overmedicalisation, and the CMO give everyone apporpriate dietary advice?
It would save a lot of NHS time and appointment availability, and the gOvernment does after all control the TV media (BBC anyway) and certain tabloid nespapers, so it is just ‘programme content’.

Justin Copitch 29 June, 2022 2:59 pm

I had a patient with a borderline B12 that I mentioned to her as a quick comment at the end of a longer consultation and she presumably quickly forgot about. She layer endured months of debilitating symptoms resulting in bullying by her employer and her now very low B12 was not picked up for some time due to her coexisting new diagnosis of hypothyroid months earlier. I’m still not sure what caused it but I wish I had arranged a follow up B12 in 6 weeks.

The question is are there better criteria that we can stick to, to make our screening blood tests a valid and useful tool.