Risk scoring in primary prevention of atherosclerotic cardiovascular disease: Strengths and limitations

Prevention of atherosclerotic cardiovascular disease (ASCVD) is possible through lifestyle adaptations and an optimal management of various bioclinical risk factors.

Significant improvements have been achieved in this respect over the past decades but ASCVD remain one of the leading causes of premature mortality and of disability adjusted life years in many countries of the world.¹

A majority of new cases occur in the apparently healthy population; therefore primary prevention of ASCVD remains a major public health challenge.

Primary prevention of ASCVD is possible but should be developed in a cost-efficient manner. This requires the stratification of the adult population in various groups of total ASCVD risk, allowing the application of preventive actions in accordance with the total risk, aiming at an acceptable number needed to treat for high-intensity preventive actions.

Risk scoring in primary prevention of ASCVD involves an estimation of the likelihood of a person developing ASCVD over a defined period of time. It is an estimate, not a precise calculation, of the absolute risk. It allows the stratification of the asymptomatic adult population in subgroups of total ASCVD risk; the application of the limited resources for prevention can then be guided in accordance with the total ASCVD risk level: those at higher risk should be targeted for intensive lifestyle interventions and, when appropriate, for drug therapies. Cost-efficient applications of limited resources and considerations of cost–benefit are essential in this respect.

The use of risk scoring may also stimulate greater equity in the distribution of effective therapies.

It has been shown that a purely clinical estimation of total ASCVD risk by health professionals may under- or over-estimate the real situation in numerous subjects.²

Indeed, risk factors interact synergistically in building up total ASCVD risk. The management of risk factors in isolation may introduce overconsumption and high-risk subjects may be overlooked when their elevated total ASCVD risk is based on a combination of several seemingly modest risk factors.

Total ASCVD risk is a continuum. In most risk scoring tools cut-offs are used to categorize the population into ‘low’, ‘moderate’, ‘high’ or ‘ very-high’ total ASCVD risk. These cut-offs are mainly based on practical considerations in relation to the health care system, to health insurance plans, to economic determinants but not to strong scientific evidence. The choice of the cut-offs reflects the ability of the health system to care for the persons at risk; they depend on the risk/benefit ratio and on the resources available.

‘High’ in preventive terms is that level of total ASCVD risk above which the likelihood of developing ASCVD is increased AND above which a reduction of the total ASCVD risk is more effective than harmful and cost-efficient. This requires that the risk scoring tool includes modifiable risk factors for which it has been demonstrated that a change is associated with disease prevention. These risk scoring tools should not just be used to identify subjects at high risk; risk scoring tools should not be interpreted in a dichotomous way, not end up in high-risk people versus all the other. This could lead to a replacement of the old dichotomous approach for single risk factors – hypertension versus normotension, diabetes versus no-diabetes – by another dichotomous classification of total ASCVD risk in high versus low risk. Preventive actions are also needed to keep subjects at low risk in that category for as long as possible and to prevent in those at moderate risk further increase of their risk, using lifestyle adaptations in the first instance.

In this issue of the European Journal of Preventive Cardiology an educational paper is published on ‘Risk prediction tools in cardiovascular disease prevention’.³ It has been developed within the framework of the ESC Prevention of Cardiovascular Disease Prevention Programme. In this paper currently available risk prediction tools are reviewed and additional support is provided to health professionals on when, with whom and how to use these tools.

Regarding the choice of which tool to use the clinician is referred to the U-prevent.com website; for asymptomatic subjects free of ASCVD antecedents, not known with diabetes and <70 years old the SCORE risk estimation tool is recommended with models for low- and high-risk European countries.

From a theoretical viewpoint the most suitable model to use would be one that is based on observations in a recent cohort of people that is recognizably comparable to the patient population with whom the health professional is working. In reality one has to realize that most of the existing risk scoring models are based on observations in prospective studies that were carried out many years ago. When applied to populations where the incidence of ASCVD has declined these tools will overestimate total ASCVD risk and the reverse is true when applied in populations where the disease is on the increase. One should realize that the epidemic of ASCVD has been and still is very dynamic.^1,4 This limitation of risk scoring systems based on historic cohorts can be dealt with by re-calibration, a technique which adjusts for such time trends.

The SCORE model has been recalibrated in numerous European countries; in some countries the recalibrated tool has been validated and the recalibrated versions perform better.^5,6 Other risk scoring models can also be recalibrated but the process is easier for models using only fatal ASCVD events as outcomes; indeed, non-fatal events are dependent on definitions, change in diagnostic tests, and methods of ascertainment, all of which may vary over time and between studies.

New up-to-date recalibrated versions of the SCORE model will become available for most European countries later this year.

Besides the use of risk scoring for priority setting, risk estimation tools may also be useful for communication with patients; it may be helpful to inform them about their risk of developing ASCVD; it may increase their motivation to adhere to lifestyle changes and to adopt optimal compliance with drug therapies. In recent guidelines it is strongly recommended to involve patients more actively in decision-making in order to improve persistence with all the pieces of advice on the long run.⁷ This is consistent with reviews on the effects of health risk appraisal tools, suggesting that such information is rather a tool to be used in conjunction with other tools to promote adherence and compliance.

In communication with the patient the health professional should include not only the results of the risk scoring model but also other factors that are considered as risk modifiers but that are for various reasons not included in the risk estimation tools, such as physical activity, overweight/obesity, family history, other lipid fractions such as triglycerides, high-density lipoprotein cholesterol and lipoprotein(a), psychosocial stress, ethnicity and co-morbidities. The assessment of these risk modifiers can be useful particularly in subjects at intermediate or moderate total ASCVD risk; they may move the total risk estimate upwards or downwards and by doing this influence management decisions.

Providing more global ASCVD risk information to adults may improve accuracy of risk perception and increase the intent to initiate preventive actions.⁸

There are also observations from prospective studies demonstrating that the total ASCVD risk in asymptomatic subjects can be higher than the risk scoring tool predicts in the presence of markers of subclinical atherosclerosis, detected by coronary artery calcium (CAC), ankle-brachial index, pulse wave velocity or carotid ultrasonography. In a study comparing these markers, CAC had the best reclassification ability.⁹ The use of these markers may be of interest in subjects who belong to the intermediate total ASCVD risk category.

However, it should be emphasized that the ASCVD risk estimation of a given person, expressed in terms of absolute risk, remains hazardous, very approximate and therefore, at present, elusive. The problem is that health professionals and the public tend to look at these risk scoring tools on an individual basis while the strategies that are proposed are based on a group approach. At the individual level the result of the absolute risk estimate should not be the most important point of discussion but other measures that are related to the risk scoring model may be of interest in communication with the patient, such as estimates of relative risk, risk age or lifetime risk.

Finally, although the use of risk scoring tools is recommended in virtually all recent guidelines on prevention of ASCVD, the benefit of a total ASCVD risk approach has only been tested in a few studies. In one trial it was concluded that it was worth developing clinical support systems that calculate cardiovascular risk before the consultation.¹⁰ In a focus group study the assessment of global coronary risk was perceived by the participants as useful and motivating.¹¹ In the CHECK-UP study it was demonstrated that providing patients and their physicians with a coronary risk profile increased the likelihood of appropriate treatment for dyslipidaemia and arterial hypertension.^12,13

One should also realize that these risk estimation tools should not be used in isolation but as part of a comprehensive strategy to implement guidelines on the primary prevention of ASCVD. The challenge nowadays is not the need for a more personalized prevention but the failure to act in those who have the potential to benefit.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References