In order to bring you the best possible user experience, this site uses Javascript. If you are seeing this message, it is likely that the Javascript option in your browser is disabled. For optimal viewing of this site, please ensure that Javascript is enabled for your browser.
Did you know that your browser is out of date? To get the best experience using our website we recommend that you upgrade to a newer version. Learn more.

How to use risk predictions tools in daily practice

Use risk prediction tools to identify and motivate high risk patients and to assist your clinical decisions-making process.



Background

Coronary artery disease (CAD) remains the major cause of death in Europe (1). It is well known that timely preventive interventions could have a major impact on the CAD burden due to progressive development of CAD over multiple decades (2,3). Recent and ongoing innovations such as digital health, precise diagnostics and new therapies created the goal to deliver more tailored preventive care. Risk prediction tools can be used to identify and even motivate high risk patients (and their doctors) to start a tailored prevention program by enhancing healthier lifestyles, pharmacological and other healthcare interventions and reducing risk factor prevalence (e.g. smoking) (4,5). Risk prediction tools use multivariable algorithms to give physicians a validated way to estimate risk and prognosis based on the integration of patient’s characteristics, clinical signs and laboratory test (6). The main purpose of risk prediction is to prevent cardiovascular disease by timely giving the right interventions to the correct patients.

Risk prediction tools are developed to identify patients at risk and to facilitate physician decision making. The result of the prediction models can be used to decide the most appropriate/recommended course of action. In the ideal world, the decision is been made in the form of shared decision making to improve patient’s motivation and adherence (6). A study of Kappen et al. (6) demonstrated that the implementation of a risk prediction tool facilitated physicians in becoming more aware of the outcomes, in becoming more informed on risk factors and to have a more positive attitude toward preemptive management.

Choosing the right tool

It is easy to get confused in the tangle of available risk algorithms. Many healthcare professionals are overwhelmed by the vast amount of risk scores and therefore it is important to give an overview of the guideline recommended risk algorithms. The EAPC advises the use of the U-Prevent tool (www.U-Prevent.com) in clinical practice. This website is an interactive website that helps to select the right risk prediction tool for every individual patient (5). The goal of the U-prevent tool is to provide individualized cardiovascular risk management based on estimated individual risk and treatment effects.

How to interpret and communicate risk

A study of Müller-Riemenschneider et al. (8) in Germany demonstrated that general practitioners rarely use risk prediction tools in primary prevention of healthy patients. One of the main reasons for the underuse is the lack of knowledge on how to use and interpret the risk scores. Some health professionals have the feeling that simplicity of algorithms seems to disrespect clinical complexity (6). In the past, physicians did this based on their clinical judgement, previous experience and personal beliefs. However, this can lead to wrong interpretation of data and therefore might generate a mismatch between the risk profile of the patient and the type of care required (5). Moreover, a study of Jaspers et al. (7) demonstrated that opinions about what "meaningful" treatment is vary widely among patients and doctors. This can result in different treatment thresholds among patients.

Another problem is the fact that guidelines also don’t provide clear advice about treatment consequences of a certain level of risk. This also means that caregivers are not so well aware what the advantage is of using prediction tools.

Not only risk interpretation but also risk communication is very important because it has to improve patient’s risk perception and understanding. Risk communication should cover the probability of the risk occurring, the importance of the adverse event, and the effect of the event and treatment on the patient (9). But in cardiovascular medicine, it is also important to try to let the risk communication be a trigger for stimulating changes in health behavior and shared decision making (9).

The increasing number of (online) prediction tools and decisions aids can also help during risk communications. Research suggested that clinicians using these aids could increase patients' knowledge and understanding. Furthermore, patients who use decision aids are consistently more ready to make a decision than those receiving usual care (10).

Why use risk prediction tools in daily practice

Cardiovascular risk assessment is a promising tool for daily clinical practice because it can reassure low-risk patients and motivate high risk patients to a lifestyle change. Furthermore, the risk prediction tools can also be used to monitor individual’s progress as risk factors come under control.

Risk prediction tools provide objective risk estimates that can assist the decision making of health professionals. Good risk prediction models should improve individual clinical outcomes and also resources allocation by avoiding both under- and overtreatment (5) Risk prediction tools will also be becoming increasingly important as clinical practice guidelines continue to move toward personalized preventive care. The integration of clinical risk prediction equations will be essential for guiding absolute risk assessment (11).

Risk prediction can also impact the patient’s behavior and treatment decisions by improving insight into their cardiovascular prognosis and anticipating the potential impact of some therapies. Secondly, by giving the patient more insight in his individual risk, the patient will be empowered to take part in the decision making (5,12) Shared decision making can improve motivation for therapy adherence and lifestyle change, including changes in nutrition, physical activity, relaxation training, weight management and participation in smoking cessation programs for resistant smokers (5).

Furthermore, risk prediction will also become important in the development of new medication for intensive risk factor reduction such as PCSK9i, SGLT2i, and GLP1-RA. These therapies are generally more expensive and riskier and therefore not applicable to large groups of patients. Risk scores are needed to identify the patients with the highest risk and therefore the most benefit. In elderly, patient-specific risk scores could be used to determine which elderly preventive medication is no longer useful and can be stopped.

Future directions

Personalized care will become more and more present in cardiovascular prevention. Therefore, in the future risk prediction tools will be needed to identify the individuals at risk and to determine the personalized prevention program. In the future, risk prediction tools will become more accurate by moving away from population‐based cohort studies toward contemporary and real‐world populations from electronic health records (EHRs) that reflect current trends in racial diversity, risk factor prevalence, preventive medication use, and disease incidence (11). Furthermore, big data analysis and machine-learning may help to improve accuracy of cardiovascular risk prediction, increasing the number of patients identified who could benefit from preventive treatment, while avoiding unnecessary treatment of others (13).

References


  1. Timmis A, Townsend N, Gale C, et al. Atlas Writing Group. European Society of Cardiology: Cardiovascular Disease Statistics 2017. Eur Heart J 2018;39(7):508-579.
  2. Landmesser, U., & Catapano, A. (2018). Coronary disease prevention: towards a more personalised approach. European Journal of Preventive Cardiology, 25(17), 1884–1886. https://doi.org/10.1177/2047487318805578
  3. Piepoli MF, Hoes AW, Agewall S, et al. 2016 European Guidelines on cardiovascular disease prevention in clinical practice. Eur Heart J 2016; 37: 2315–2381.
  4. Cooney MT, Dudina A, D'Agostino R, Graham IM. Cardiovascular risk-estimation systems in primary prevention: do they differ? Do they make a difference? Can we see the future? Circulation. 2010; 122:300-310.
  5. Rosselo X, Dorresteijn JAN, Janssen A, et al. Risk prediction tools in cardiovascular disease prevention . European Journal of Preventive Cardiology 2019, Epub.
  6. Kappen TH, Van Loon K, Kappen MA, et al. Barriers and facilitators perceived by physicians when using prediction models in practice. J. Clin. Epidemiol., 70 (2016), pp. 136-145
  7. Jaspers NEM, Visseren FLJ, Numans ME, et al. Variation in minimum desired cardiovascular disease-free longevity benefit from statin and antihypertensive medications: a cross-sectional study of patient and primary care physician perspectives. BMJ Open. 2018;8:e021309.
  8. Müller-Riemenschneider F, Holmberg C, Rieckmann N, et al. Barriers to Routine Risk-Score Use for Healthy Primary Care Patients: Survey and Qualitative Study. Arch Intern Med. 2010;170(8):719–724. doi:10.1001/archinternmed.2010.66
  9. Ahmed H, Naik G, Willoughby H, Edwards AGK. Communicating risk. BMJ : British Medical Journal. 2012;344:40-44.
  10. Stacey D, Bennett CL, Barry Ml, Col NF, Eden KB, Holmes-Rovner M, et al. Decision aids for people facing health treatment or screening decisions. Cochrane Database Syst Rev 2011; 1 :CD001431.
  11. Karmali KN, Lloyd-Jones DM. Implementing Cardiovascular Risk Prediction in Clinical Practice: The Future Is Now. J Am Heart Assoc. 2017;6(4): e006019. Published 2017 Apr 24. doi:10.1161/JAHA.117.006019
  12. Hedberg B, Malm D, Karlsson J-E, et al. Factors associated with confidence in decision making and satisfaction with risk communication among patients with atrial fibrillation. Eur J Cardiovasc Nurs 2018; 17: 446–455
  13. Weng SF, Reps J, Kai J, Garibaldi JM, et al. Can machine-learning improve cardiovascular risk prediction using routine clinical data?. PLOS ONE 2017. 12(4): e0174944. https://doi.org/10.1371/journal.pone.017

 

The content of this article reflects the personal opinion of the author/s and is not necessarily the official position of the European Society of Cardiology.