This site is intended for health professionals only

At the heart of general practice since 1960

Read the latest issue online

CAMHS won't see you now

GPs told to review patients at risk as IT error miscalculates CV score in thousands

Exclusive Hundreds of thousands of patients have potentially been given an incorrect cardiovascular risk estimation after a major IT system error, Pulse can reveal.

The MHRA has told GPs they will have to contact patients who have been affected by a bug in the SystmOne clinical IT software since 2009.

The regulator says that means that ‘a limited number’ of patients may be affected, and the risk to patients is ‘low’.

But Pulse has learnt that the 2,500 practices using SystmOne are having lists sent to them of around 20 patients per partner who may need to be taken off statins, or be put on them, after their risk is recalculated.

And this number could increase if a practice provides NHS Health Checks routinely. In addition, the lists being sent to practices only go back to October 2015, but practices will be sent further lists potentially dating back to 2009 over the next few weeks.

A statement from MHRA to Pulse said: 'An investigation has been launched into a digital calculator used by some GPs to assess the potential risk of cardiovascular disease (CVD) in patients.

'We are working closely with the company responsible for the software to establish the problem and address any issues identified.

‘Clinical advice is that the risk to patients is low and only a limited number of patients are potentially affected. GPs have been informed and they will contact individual patients should any further action be necessary.’

TPP told Pulse they were working to address the ‘Clinical Safety Incident’ and that the QRISK calculator was provided as an advisory tool to support decision making. They added they were working to ensure the issues were addressed and GPs are informed of affected patients ‘as soon as possible’.

Deputy chair of the GPC’s IT subcommittee Dr Grant Ingrams told Pulse it would be ’loads of work’ to sort out.

He said: ‘It affects everyone who has had a QRISK, and SystmOne are sending out messages to say “look at these patients”. But then you have to see if the change is significant, and whether you would have made a different decision at the time, or put them on a different treatment’

Dr Ingrams said: ‘There’s potential harm both ways…What happens when a patient who had been of a high risk and this hadn’t been identified and they’ve now had a stroke or heart attack?

‘Similarly if someone had a low risk and they’ve been put on a statin and had a side-effect who’s responsible? That’s the clinical risk.’

Dr William Beeby deputy chair of the GPC’s clinical and prescribing subcommittee, said the bug ‘certainly had the potential to impact on patient confidence’ and this could create even more work.

He told Pulse: ‘If you do hear that an assessment tool your doctor uses is incorrect then of course people are going to either stop their statins or they’re going to contact you and ask what their real risk is?

‘It’s the tool we’ve been told to use. So if the tool is inaccurate, then you start to lose confidence and the doctors will then lose confidence as well.’

A TPP spokesperson told Pulse: 'TPP is dealing with the Clinical Safety Incident involving the QRISK2 Calculator in SystmOne. The tool is intended to support GPs in assessing patients at risk of developing cardiovascular disease and in developing treatment plans. The QRISK2 Calculator is presented within SystmOne as an advisory tool.

’We are actively working to ensure the issues identified are addressed and to ensure that clinicians are informed of any patients that may have been affected as soon as possible.’ 

Readers' comments (19)

  • But i thought we could all be replaced by computers?

    Unsuitable or offensive? Report this comment

  • Peter Swinyard

    I am not programmed for that response. I am receiving a higher than expected volume of calls.

    Unsuitable or offensive? Report this comment

  • who's using these scores against their professional judgement anyway?
    if you trust the score anyone over 75 needs a statin but a 50 y o with cholesterol of 9 (and not familial hyperlipidaemia cos its not meant to be covered by scores)- doesn't

    Unsuitable or offensive? Report this comment

  • "Should have gone to EMIS"

    Unsuitable or offensive? Report this comment

  • Has anyone checked all the other risk assesment tools in TPP? After all, this seems to be an implementation problem rather than a problem with QRisk itself.
    And what is the problem? is it with the calculator or the data entered into the system or the Codes selected for use in QRisk (in TPP)?

    Unsuitable or offensive? Report this comment

  • I want to know how much out the calculator is, is it reading too high or too low?

    What part of the algorithym is causing this issue? It would allow us to assess our patients quickly, with getting on for 700 Qrisk scores on patient records this is a massive piece of work, knowing how

    Unsuitable or offensive? Report this comment

  • Another issue - How do TPP know the patients involved to give the practice a list. This implies that they can access the full identified patient history. Is this a Data breach?
    All hell has broken loose in a local trust because of a few sets of paper records being found in an inappropriate place. If TPP has access to the full identified medical history of all its patients ... . . .

    Unsuitable or offensive? Report this comment

  • Who is paying for the extra work involved?

    Unsuitable or offensive? Report this comment

  • So when will QRISK be correct? Should we stop using it now?

    Unsuitable or offensive? Report this comment

  • Seeing that being on a statin makes about 3-4 days difference to your date of death I really wouldn't get my knickers in too much of a twist over this, just tell them all to stop smoking, get of their collective backsides and eat food that is real food and not stuff that pretends to be food.

    Unsuitable or offensive? Report this comment

  • It is not just the QRISK2 algorithm that is faulty. A recent study by Kaiser Permanente in the USA showed that the US risk tool also overpredicts, with a predicted risk of 5% translating to an actual risk of 1.85%. Thus there is a global overprediction of risk; if the real risk was used as a baseline no-one would be on statins at all.

    This of course is quite separate from the debate as to whether statins (a) actually have any significant effect on the risk of cardiovascular disease and (b) whether that effect has anything to do with lowering serum cholesterol. The evidence is mounting that it doesn't, and that the drop seen serves only to confirm that the subject is taking statins. I used to think we should be cautious with statins; I now think they should be banned, along with all the other cholesterol-lowering drugs. Just think how much money would be saved to be re-allocated th real need areas...

    Unsuitable or offensive? Report this comment

  • Andrew,
    The problem here isn't with the QRisk calculator (MHRA recommenda that SystmOne users should use the on-line calculator at )but with the erroneus results being produced in one GP system supplier.
    However, your view that all the CVD risk calculators give misleadingly high values (leading to the prescription of too many statins) is interesting: could you share the URL for the study you're quoting?

    Unsuitable or offensive? Report this comment

  • Martin Harris

    umm, really sorry to hear this for our GP colleagued who do use TPP SystmOne, since MHRA (Medicines and Healthcare products Regulatory Agency) concerned about the interaction between QRisk2 calculator (works out risk of having a heart attack or stroke over the next ten years by answering some simple questions. It is suitable for people who do not already have a diagnosis of heart disease or stroke).

    Unsuitable or offensive? Report this comment

  • Seriously tempted to invoice TPP for the extra time this will take to sort out. Hilariously we've now been informed that the original tasks sent out were incorrect. Still waiting for new tasks, with no Qrisk calculator available in Systmone. #unimpressed

    Unsuitable or offensive? Report this comment

  • So how do we sue TPP for the extra staff and workload costs consequent to their failure?

    Unsuitable or offensive? Report this comment

  • What does one expect from a reputed manipulated American system that was being forced upon general practice, not particularly user friendly and which mem beer of administration or confederation was within their pockets. Fired or sort forthwith.

    Unsuitable or offensive? Report this comment

  • This wouldn't be the system of computerised records being forced on practices by PCTs. An American system forced to fit the UK formula, presumably after a junket over the pond.

    Unsuitable or offensive? Report this comment

  • Like I said in my previous comment about the RCGP election, long live paid slavery.
    Being independent contractors we have to abide by the emperor's edicts. Extra payments- dream on. ( I am a retired GP) . Freed from slavery because of my age!

    Unsuitable or offensive? Report this comment

  • Almost no-one wants a statin for primary prevention when I show them the so called benefit of taking them for ten years. The more I do this the less I prescribe, the less cost spent on them and their monitoring. It's ironic that statins have more evidence of benefit than many of our treatments but hey ho everyone thinks it won't happen to them....Secondary prevention rates of adoption are so much higher.

    Unsuitable or offensive? Report this comment

Have your say