NPC Archive Item: Estimating lifetime cardiovascular risk – we can, but should we?

NOTE – This is an archive post from the NPC and has not been updated since first publication. Therefore, some hyperlinks may no longer be working.
MeReC Rapid Review NPC Logo

9th March 2011

This large UK cohort study used data collected from general practice databases to develop and validate a new QRISK cardiovascular risk assessment tool to estimate lifetime risk of cardiovascular disease (CVD). For most patients, lifetime risk estimation was no more helpful than 10-year risk estimation of CVD in identifying patients at risk of CVD.

Action
Practitioners should continue to follow NICE guidance and use a validated CV risk assessment tool that is best suited to their requirements. When deciding whether or not to offer drug treatment, a 10-year risk of CVD should be estimated. Estimating lifetime CV risk may be useful in younger patients who have a number of modifiable risk factors for CVD, e.g. to illustrate the benefits of stopping smoking.

What is the background to this?
A number of risk assessment tools have been devised for estimating CVD risk in clinical practice, and these have been discussed previously in a MeReC bulletin, and MeReC rapid review number 68 and number 169. Those based on equations developed from the Framingham Heart Study have been most widely used. ASSIGN and QRISK are alternative computer-based risk scoring systems developed from UK cohorts.

These tools estimate the risk of CVD over a defined period, usually 10 years. NICE guidance recommends that statins should be offered for the primary prevention of CVD in adults who have a 20% or greater 10-year risk of developing CVD. Predicted 10-year risk also influences decisions about other treatments, such as management of hypertension or use of aspirin.

As age has such a dominant effect in calculating absolute CV risk, some younger patients may have a relatively low absolute 10 year risk but which is still higher than others of their age.  Lifetime CV risk assessment tools have been advocated to identify these relatively high risk younger patients, who may have more to gain during their lifetime if they modify their lifestyle at that point, or start other interventions earlier rather than waiting until they cross the 20% threshold currently recommended by NICE.

What does this study claim?
The QRISK authors developed a model to estimate the lifetime risk of cardiovascular disease. They did this by looking at the relationship between a number of risk factors and later development of CVD in a large ‘derivation’ cohort (n = 2,343,759) of people who did not have CVD and were not taking statins at entry into the study. This used data from a large number of GP practices in England and Wales.

The authors assessed the performance of the lifetime model in a separate ‘validation’ cohort of patients from the same GP practices (n = 1,267159). The two main measures by which a risk prediction tool should be judged are calibration and discrimination. Calibration relates to how close the predicted risk is to the observed risk. More importantly, discrimination is the ability of the tool to differentiate between people who will have an event and those who will not, over a defined period of time. The statistical methods used in the study showed that the lifetime QRISK model slightly underpredicted risk in people at lowest CV risk, but showed good calibration for those at highest CV risk. It also showed good discriminating ability (see Study details below).

So what?
So, can the QRISK lifetime CVD risk assessment tool identify those younger patients with a high CV risk compared to their peers, and importantly, would this lead to useful clinical intervention in this group?

The QRISK lifetime risk assessment tool is likely to identify some patients who are at higher risk of CVD than their peers, at a younger age than compared to the 10 year QRISK2 model (and this is probably true for the other 10 year assessment tools commonly used).

This may be helpful if it encourages such people to attend to their modifiable lifestyle risk factors for CVD. The graphs that the tool generates from calculating a patient’s lifetime CVD risk compared to their ‘what if’ risk (after reducing modifiable risk factors) may be useful in explaining risk and risk reduction to patients.

However, it is not clear when medical intervention should be initiated in this group; any gain from the early addition of long term drug intervention must be weighed against the increased exposure to possible harms, the ‘medicalisation’ of conditions that can be addressed by a change in lifestyle, and increased cost. There is little evidence for any benefit of treatment in this group, as most lifetime risk accrues after the age of 65 years. The authors themselves conclude that ‘research is needed to examine closely the cost effectiveness and acceptability of such an approach’.

Further information about communicating CVD risk can be found in MeReC bulletin 19 and a patient decision aid for use of statins has also been produced by the NPC.

Study details
Hippisley-Cox J; Coupland C; Robson J; Brindle P; Derivation, validation, and evaluation of a new QRISK model to estimate lifetime risk of cardiovascular disease: cohort study using QResearch database. BMJ 2010; 341:c6624 published online 9 December 2010

Objective To develop, validate, and evaluate a new QRISK model to estimate lifetime risk of cardiovascular disease.

Design Prospective cohort study with routinely collected data from general practice.

Patients 3,610918 patients aged 30–84 years, free of cardiovascular disease and not taking statins between 1 January 1994 and 30 April 2010. There were 2 ,343759 patients in the derivation dataset, and 1, 267159 in the validation dataset.

Outcomes and results
Individualised estimate of lifetime risk of cardiovascular disease accounting for smoking status (sub-categorised into five variables), ethnic group, systolic blood pressure, ratio of total cholesterol:HDL cholesterol, body mass index, family history of coronary heart disease in first degree relative aged <60 years, Townsend deprivation score, treated hypertension, rheumatoid arthritis, chronic renal disease, type 2 diabetes, and atrial fibrillation. Age-sex centile values for lifetime cardiovascular risk compared with 10 year risk estimated using QRISK2 algorithm. Across all the 1,267159 patients in the validation dataset, the 50th, 75th, 90th, and 95th centile values for lifetime risk were 31%, 39%, 50%, and 57% respectively. In terms of calibration, in the validation dataset, the ratios of predicted to observed risk for women varied between 0.8 and 1.0, and for men between 0.9 and 1.0. Discrimination was assessed using the ROC statistic (where a score of 0.5 indicates the model is no better than results generated by chance, and 1.0 indicates a perfect model). This was 0.842 in women (95% confidence interval 0.840 to 0.844), and 0.828 in men (95% confidence interval 0.826 to 0.830).

Of the 10% of patients in the validation cohort classified at highest risk with either the lifetime risk model (lifetime risk CVD ≥50%) or the 10 year risk model (10 year risk CVD ≥23.4%), only 18,385 (14.5%) were at high risk on both measures. The patients with the high lifetime risk were more likely to be younger, male, and have a positive family history of coronary heart disease than those with a high 10 year QRISK2 score. There were also more patients from South Asian ethnic groups and current smokers in the high lifetime risk group, and fewer patients with associated clinical conditions such as type 2 diabetes and atrial fibrillation.

Sponsorship This study received no external funding

More information on risk assessment in cardiovascular disease can be found on the cardiovascular section of NPC

Please comment on this blog in the NPC discussion rooms, or using our feedback form.

Make sure you are signed up to NPC Email updates — the free email alerting system that keeps you up to date with the NPC news and outputs relevant to you