This site is intended for health professionals only

At the heart of general practice since 1960

Read the latest issue online

Gold, incentives and meh

NICE recommends introduction of 10% statin threshold into QOF despite GP opposition

NICE has given the green light to a new QOF indicator rewarding practices for how many patients they treat with statins at the 10% primary prevention threshold, despite unease among GP members on the independent advisory panel.

The introduction of the 10% threshold for prescribing statins to patients newly diagnosed diabetes and hypertension was discussed this afternoon at the first meeting of the newly convened QOF committee.

The indicator will now be put forward for GPC and NHS Employers to negotiate its introduction into the contract for 2015/16.

It comes after the RCGP, the GPC and the former chair of the QOF advisory panel all strongly spoke out against the introduction of the indicator.

Several GP members of the panel voiced concerns about the proposed indicator rewarding practices only for treatment with statins and not lifestyle advice - which they said goes against the NICE lipid modification guidelines which emphasise lifestyle advice should be offered before statin therapy.

They argued for the indicator to be reworded to reward offering lifestyle advice and other interventions as appropriate, including statins, to allow GPs room to discuss the option of statins with patients. They also proposed the term ‘offer’ a statin be used rather than ‘treated with’ a statin to allow more room for GPs to give patients the option of a statin rather than be paid only if the patient ended up with a prescription.

However, other committee members, including one GP, argued that NICE guidelines were clear that GPs should be prescribing statins at the 10% threshold, and that GPs would be able to exception report patients who chose not to take the drug option.

The panel agreed instead to introduce a new indicator incentivising lifestyle advice in these groups, as well as a new indicator incentivising statin treatment at the 10% threshold - and to introduce new business codes to allow for the exception reporting.

NICE QOF Committee chair Professor Danny Keenan said patients in these risk groups at the 10% risk threshold ‘should be on a statin’.

He added: ‘We have very clear guidelines, they couldn’t be clearer - and we’ve been over and over this. We’ve introduced the lifestyle indicator and allowed for exception reporting and we should go ahead with this.’

Dr Andrew Green, chair of the GPC prescribing committee, said it was ‘obviously disappointing that NICE have chosen to ignore the views of both the body that represents GPs as well as our royal college’.

However, he added, ‘whether this eventually becomes part of QOF remains subject to negotiation’.

Dr Green said: ‘I have no doubt the proposed indicator 11a [the proposal to measure the prescription of statins] will be a measure of prescribing activity but not of the quality of patient care, which depends on many more factors than a simple tablet-count. As general practice becomes more complex it is vital that measures of performance are sophisticated enough maintain validity, and have the confidence of GPs; this proposal meets neither of these requirements.’

Pulse revealed today, the former chair of the committee - Dr Colin Hunter, a GP in Westhill, Aberdeenshire - said that NICE had ‘lost the plot’ to consider introducing the new indicators.

He said: ‘I personally am completely against the 10%. I don’t think there is enough evidence to support it and I think it’s a societal question.

‘I think it is where NICE has lost the plot – when you end up with the majority of people over 65 needing a statin because the economics tell you that. The economics are far from a good science.’

Dr Hunter’s intervention followed those of the RCGP and the GPC, who both strongly opposed the proposals in their consultation responses.

The GPC said that it was ‘vital for the credibility of QOF’ that indicators have a robust evidence base, make significant difference to patients and are backed for the profession, adding that these proposals ‘fail on all these counts’.

The RCGP warned that the proposals risked ‘the loss of professional confidence in the healthcare targets they are being asked to meet’.

Proposed wording of new indicators

  • IND-11: % of patients aged 25-84, with a new diagnosis of hypertension or type 2 diabetes* who have a recorded cardiovascular risk assessment score of 10% or greater who have been given lifestyle advice;
  • IND-11a: % of patients aged 25-84, with a new diagnosis of hypertension or diabetes* who have a recorded cardiovascular risk assessment score of 10% or greater who are currently treated with statins.

* this may include ‘three months either side of diagnosis’


Readers' comments (44)

  • Dear Samuel,

    You’re mistaken. I’m talking primarily about RCT data. And none of the RCTs on statins have been anything like 10 years in duration. They have all been around 5 years or less. We have a couple of follow-up studies covering about 11/12 years total duration, however, these are not RCTs. For example there was a follow-up study done on the WOSCOPS trial, however, for the follow-up period some people who were in the placebo group now started on a statin, and some people in the statin group had now stopped their statin. Hence, the follow-up study was not very meaningful and of course no longer has a control arm.

    The 4S study you mentioned was of course a secondary prevention trial on super-high risk population - it doesn’t really fit into our discussion of low risk populations, and I’m not questioning the role of statins in secondary prevention.

    You said that RCT evidence shows reductions in all-cause death by unto 20% and CVD events 30%. As you know, these are relative percentages which are very misleading. Relative percentages are a very good tool for spinning the results. Absolute percentages are the most important.

    For example, in the JUPITER trial the risk of heart attack was reduced by 0.41% in absolute terms (0.76% placebo vs 0.35% statin) however, the press release issued by the sponsor of the trial quoted a 54% reduction in relative terms - which was unfortunately copied by the world’s media. And incidentally, CVD deaths were the same in both groups.

    By using relative percentages to exaggerate the benefits and then just saying that the risk of diabetes and other adverse effects is small compared to this is a gross misrepresentation of the data. In the JUPITER trial the risk of diabetes was about 0.6%, which is greater than the reduction in heart attach risk. Some people might say, well having diabetes is better than having a heart attack, but this is the illusion of certainty principle

    Combining the data from RCTs on all-cause deaths shows that 1 person in 1000 could live longer - a number that is consistent with the play of chance.

    I have checked this out - as I said previously the numbers do not add up for primary prevention and especially not at increasingly lower levels of risk. I hope others will also consider this.

    Unsuitable or offensive? Report this comment

  • Samuel Lewis

    dear justin,

    you confirm my suspicions as to your mindset.

    you are of course quite right to decry the use of relative risk reduction in isolation, even at 30% across the entire risk ranges studied.

    but everyone here is talking about ABSOLUTE risks of 10 to 20%, from which we can infer an Absolute Risk reduction of circa 15% x 30% ( AR x RRR)..

    of course the absolute benefit falls as AR falls. But the cheaper the statin becomes, the lower the affordability threshold for cost-benefit savings, as NICE argues.

    NICE is saying you should give more patients the chance and the choice, but you seem to want to make the decision for them.

    At what level of risk would you prescribe, and why ?

    Unsuitable or offensive? Report this comment

  • Samuel Lewis

    oh, and yes you are right that none of the RCTSs per se continued to 10 years, although lots of follow-up does continue, confirming that benefit is sustained. 4S follow-up is now 20+ years.

    The reason is of course because the trials all achieved significant benefits within about 5 years, and hence MUST be stopped according to the Helsinki ethic, so that the control group can be offered the chance of treatment. Just can't win, can we ?

    Unsuitable or offensive? Report this comment

  • Samuel Lewis

    for a thorough analysis of the Numbers Expected To be Treated, and the estimated costs and benefits downsized to one practice, see

    Unsuitable or offensive? Report this comment

View results 10 results per page20 results per page50 results per page

Have your say