This site is intended for health professionals only

GP score ‘improves stroke prediction’

GPs can more accurately determine which patients with atrial fibrillation should be given anticoagulant therapy using a new score that could replace the current guideline-recommended CHADS2 tool, claim UK primary care researchers.

Their report in the BMJ concludes that the score, based on the database used for the QRISK2 score in cardiovascular disease, provides ‘some improvement over current risk scoring methods’, such as the CHADS2 and CHA2DS2VASc, for patients with atrial fibrillation for whom anticoagulation may be required. 

But experts said the QStroke score only had a ‘marginal improvement’ on existing scores, and would be difficult to perform in clinical practice.

The QScore algorithm includes systolic blood pressure, total: HDL cholesterol ratio, Townsend deprivation score, smoking status, ethnicity and medical and family history of cardiovascular disease and risk factors (see box).

The Nottingham University researchers looked at data from the EMIS QResearch general practice database for over 5.4 million patients, aged 25 to 84 years, with no history of stroke or transient ischaemic attack, and with no use of oral anticoagulants.

They reported that QStroke predicted 10-year stroke risk more accurately than the Framingham stroke equation in the full cohort studied, and more accurately stroke risk more accurately than CHADS2 and CHA2DS2VASc in the 7,689 patients who had atrial fibrillation at baseline.

For example, in men, the C-statistic was 0.71 for QStroke compared with 0.67 for CHA2DS2VASc and 0.63 for CHADS2. The authors calculated that 9% of patients with atrial fibrillation were at low risk based on CHADS2 but high risk under QStroke; the same proportion were high-risk on CHADS2 but low-risk on QStroke. Similarly, 4% of patients would be reclassified from low to high risk on QStroke compared with CHA2DS2VASc and vice versa.

The researchers concluded: ‘QStroke shows some improvement over current risk scoring methods, CHADS2 and CHA2DS2VASc, for patients with atrial fibrillation for whom anticoagulation may be required. 

‘QStroke provides a valid measure of absolute stroke risk in the general population of patients free of stroke or transient ischaemic attack as shown by its performance in a separate validation cohort.’

‘Although QStroke has been designed to be used in all patients without a history of stroke or transient ischaemic attack, we envisage its primary use will be in the subset of patients with atrial fibrillation for whom anticoagulation is considered,’ they added.

But stroke prevention experts told Pulse any difference was marginal, with sensitivity and specificity virtually the same for QStroke and CHA2DS2VASc.

Professor Gregory Lip, consultant cardiologist and director of the Haemostasis, Thrombosis and Vascular Biology unit at the University of Birmingham, said: ‘If you look at the detail, if anything it’s marginal improvement.

‘These kinds of weighted scores can improve on simple risk scores, but you have to balance the practicality and simplicity of using the score with the predictive value. And what is nice about [CHADS2 and CHA2DS2VASc] is that even in the middle of a busy ward or clinic you can calculate the risk with these off the top of your head without a prompt.’

Dr Andreas Wolff, a GP in Darlington and Cardiology GPSI, said: ‘I don’t think it will make much difference, it’s not really any more predictive. It means you will capture just as many patients but the next step is what is important, doing something about it – and that’s where many practitioners still fall down.’

He added: ‘If you just use CHADS2 to its maximum potential you’d do much more good than moving from CHA2DS2VASc to QStroke.’

Both Professor Lip and Dr Wolff emphasised that best practice according to latest European Society of Cardiology guidelines on the management of atrial fibrillation is now to first identify low-risk patients for whom no antithrombotic therapy is needed using CHA2DS2VASc; anyone with a score of 1 or more should then be considered for anticoagulation.

Dr Wolff said: ‘The big shift is getting away from thinking about who is at high risk to identifying those at low risk and then treating everyone else.

‘I would definitely recommend [CHA2DS2VASc] because is differentiates the lower risk categories much better and you can be confident the ones you don’t treat are very unlikely to have stroke.’

Dr Wolff said the QOF should be updated to include CHA2DS2VASc instead of CHADS2 soon, as he hoped forthcoming updated NICE guidance on atrial fibrillation will include CHA2DS2VASc.

The study comes after Pulse revealed GPs are increasingly prescribing the newer anticoagulant alternatives to warfarin for the prevention of stroke, although their uptake has been slower than expected due to cost concerns.

BMJ 2013; 346: f2573


QStroke algorithm risk factors

Systolic blood pressure (10 unit increase)

Total cholesterol:HDL cholesterol ratio (1 unit increase)

Townsend deprivation score (5 unit increase)

Smoking status:

- Non-smoker

- Former smoker

- Light smoker

- Moderate smoker

- Heavy smoker


- White or not recorded

- Indian

- Pakistani

- Bangladeshi

- Other Asian

- Caribbean

- Black African

- Chinese

- Other ethnic group

Medical and family history:

- Family history of coronary heart disease

- Atrial fibrillation

- Coronary heart disease

- Congestive cardiac failure

- Type 1 diabetes

- Type 2 diabetes

- Treated hypertension

- Rheumatoid arthritis

- Valvular heart disease


Visit Pulse Reference for details on 140 symptoms, including easily searchable symptoms and categories, offering you a free platform to check symptoms and receive potential diagnoses during consultations.