Background: Cardiovascular disease (CVD) risk communication is a challenge for clinical practice, where physicians find it difficult to explain the absolute risk of a CVD event to patients with varying health literacy. Converting the probability to heart age is increasingly used to promote lifestyle change, but a rapid review of biological age interventions found no clear evidence that they motivate behavior change.
Objective: In this review, we aim to identify the content and effects of heart age interventions.
Methods: We conducted a systematic review of studies presenting heart age interventions to adults for CVD risk communication in April 2020 (later updated in March 2021). The Johanna Briggs risk of bias assessment tool was applied to randomized studies. Behavior change techniques described in the intervention methods were coded.
Results: From a total of 7926 results, 16 eligible studies were identified; these included 5 randomized web-based experiments, 5 randomized clinical trials, 2 mixed methods studies with quantitative outcomes, and 4 studies with qualitative analysis. Direct comparisons between heart age and absolute risk in the 5 web-based experiments, comprising 5514 consumers, found that heart age increased positive or negative emotional responses (4/5 studies), increased risk perception (4/5 studies; but not necessarily more accurate) and recall (4/4 studies), reduced credibility (2/3 studies), and generally had no effect on lifestyle intentions (4/5 studies). One study compared heart age and absolute risk to fitness age and found reduced lifestyle intentions for fitness age. Heart age combined with additional strategies (eg, in-person or phone counseling) in applied settings for 9582 patients improved risk control (eg, reduced cholesterol levels and absolute risk) compared with usual care in most trials (4/5 studies) up to 1 year. However, clinical outcomes were no different when directly compared with absolute risk (1/1 study). Mixed methods studies identified consultation time and content as important outcomes in actual consultations using heart age tools. There were differences between people receiving an older heart age result and those receiving a younger or equal to current heart age result. The heart age interventions included a wide range of behavior change techniques, and conclusions were sometimes biased in favor of heart age with insufficient supporting evidence. The risk of bias assessment indicated issues with all randomized clinical trials.
Conclusions: The findings of this review provide little evidence that heart age motivates lifestyle behavior change more than absolute risk, but either format can improve clinical outcomes when combined with other behavior change strategies. The label for the heart age concept can affect outcomes and should be pretested with the intended audience. Future research should consider consultation time and differentiate between results of older and younger heart age.
International Registered Report Identifier (IRRID): NPRR2-10.1101/2020.05.03.20089938
Cardiovascular disease (CVD) risk communication is a challenge for clinical practice, where physicians find it difficult to explain the absolute percentage risk of a CVD event to patients with varying health literacy [, ]. Absolute risk calculators are recommended in clinical guidelines around the world as the best way to predict the risk of a CVD event over a relatively short period, by incorporating both modifiable (eg, smoking, blood pressure, and cholesterol levels) and nonmodifiable (eg, age and gender) risk factors [ - ]. These calculators are designed to determine whether preventive medication should commence, which is generally recommended for high risk and low risk categories. They are not designed to motivate or determine when to commence lifestyle changes, as this is recommended for all risk categories. For example, smoking cessation should always be recommended for a person who smokes, regardless of the calculated absolute risk. However, the implementation of absolute risk calculators has been poor, and communication barriers have been identified as one reason for this [ , - ]. One proposed solution is to use more intuitive and motivating risk communication concepts rather than abstract probabilities [ , ]. Heart age has been suggested as an alternative to absolute percentage risk of a CVD event and can be calculated by comparing an individual’s absolute risk over 5 or 10 years with ideal risk factors (eg, blood pressure of 120/80 mm Hg) or the average for their age or gender category [ ].
Heart age tools are increasingly used to promote behavior change around the world, including clinical contexts and web-based consumer resources. They have not generally been used to guide decisions to commence medication in the same way that absolute risk calculators have, although the Joint British Societies (JBS)-3 guidelines in the United Kingdom do suggest that older heart age may be considered as a reason to initiate medication . Following clinical trials of the lung age and heart age concepts, the World Heart Federation collaborated with Unilever for an international promotion to at least 2.7 million consumers in 2009 [ ]. Since then, heart age tools have also been promoted to support clinical guidelines in the United Kingdom, New Zealand, and Australia, reaching millions more through nonprofits and health services [ , ]. However, despite this mass appeal, there is no clear evidence that such tools actually motivate behavior change. A review of CVD risk communication in 2011 identified heart age as a potentially useful concept that requires further research [ ]. A 2016 review of vascular age concepts in clinical applications found limited trials testing the effects of communicating this concept [ ], but a recent rapid review in 2020 highlighted the increasing number of studies on age-based risk formats, suggesting there may now be more evidence available [ ].
Heart age tools vary widely in terms of their underlying risk models and results, with the possibility of receiving an older heart age on one calculator but a younger result on another . This is because the tools have different underlying algorithms to assess CVD event risk (eg, Framingham vs QRISK), and different thresholds for comparing an individual’s result with ideal or average risk factors (eg, systolic blood pressure may be compared with 120 for ideal vs 125 for average). Another factor is that many people do not know their blood pressure or cholesterol levels, and different techniques are used to estimate this (eg, BMI or population average) [ ].
In addition to the underlying algorithms, the way heart age results are explained can come in many forms. The Australian heart age calculator is relatively simple with a single heart age result and prompts an individual to see a physician for a more accurate risk assessment . UK-based heart age calculators include numerous risk communication formats, including the percentage chance of a heart attack or stroke over 10 years or a lifetime, estimated life expectancy, and graphical displays [ ]. Heart age tools are often linked to further lifestyle change messages [ ], but research on heart age interventions has not differentiated well between these components, despite the fact that they represent different behavior change techniques—the active ingredients of behavioral interventions [ ]. Heart age interventions may range from a simple one-off message frame (eg, communicating risk as older heart age without any further information) to complex programs involving health professional counseling and goal-setting or monitoring of heart age over time.
Previous reviews relating to the heart age concept have been descriptive about the models [, ] or concepts but have not made a distinction between the comparison groups involved in trials (eg, heart age vs standard care or absolute risk [ , ]), have not clearly identified the behavior change techniques included within heart age interventions [ , , ], or have not included qualitative studies that may provide additional insights into why these tools are so widely used in spite of limited evidence for their effectiveness [ , , ]. The aim of this systematic review is to identify the content and effects of heart age interventions presented to patients or consumers for the purpose of CVD risk communication in detail, in order to shed light on mixed evidence of their effectiveness.
Our methods protocol was prepublished on the preprint server medRxiv  (not peer-reviewed).
Studies were considered eligible if they met the following criteria:
- Published from the inception of the database to April 2020 (this was later updated to March 2021) in peer-reviewed journals.
- Population: used an adult population (over 18 years of age) or, if not explicit regarding age, are clear that participants were not children.
- Intervention: present the concept of heart age to patients or consumers for the purpose of CVD risk communication, in any setting, including general practices, hospitals, health clinics or community centers, workplaces, or on the web. This included both simple message frame experiments and complex programs in applied settings.
- Comparators: we placed no restrictions on whether a comparison or control group was used.
- Outcomes: report qualitative themes or quantitative outcomes related to psychological or behavioral responses to heart age, including clinical outcomes.
Studies that were not peer-reviewed journal articles, such as conference proceedings, dissertations, or government reports, were excluded. Protocol papers, opinion papers, reviews, web-based user descriptions, and heart age algorithm development or validation were excluded. Some studies applied a heart age algorithm to population data or individual patients as an outcome, but the results were not conveyed to individuals, so these were excluded. Studies that presented heart age to adults but did not measure outcomes or collect qualitative data were also excluded.
The following databases were searched: the Cochrane Central Register of Controlled Trials (via OvidSP), MEDLINE (via OvidSP), Embase (via OvidSP), and PsycINFO (via OvidSP) up to March 2021. The search terms are based on an earlier vascular age review in 2016, with additional free text terms based on known relevant papers. The full list of terms is based on a previous review and includes (vascular, vessel, arterial, heart, cardiovascular, coronary, risk AND age, ages, ageing, or aging), OR heart forecast, and limited to human studies. We then searched the citations and references of the final included studies and 2 previous reviews of vascular age models and more general age-related risk concepts. We also included any papers mentioned on publicly accessible heart age websites.
We downloaded the references identified in the searches (electronic databases and additional searches) into Microsoft Excel. Duplicates were then removed, and 2 reviewers (SC and C Batcup) screened the titles and abstracts of each study to determine whether they should be included. Discrepancies were resolved by discussion with C Bonner.
The screening process was undertaken by 2 review authors (C Batcup and SC). Each reviewer independently assessed a study’s suitability to be included in the review by marking against each study on an Excel spreadsheet, which contained the title and abstract. Studies that did not meet the inclusion criteria were excluded. We obtained the full text of the remaining papers and then assessed the remaining papers against the full inclusion terms for the review to determine their eligibility for inclusion. Non–English language papers were translated into English using Google Translate and verified for inclusion or exclusion by speakers of the relevant language. The review authors resolved disagreements through a consensus-based decision-making process or, when necessary, discussion with a third review author (C Bonner). Two Japanese language studies were considered for inclusion but deemed ineligible by an author who could read Japanese (JD).
Two review authors (C Batcup and SC) completed web-based training to apply the behavior change technique taxonomy to published methods and used a predefined data extraction form to collect data from the included studies. Reviewers piloted the data extraction form with a sample of included papers; however, no amendments were made. An Excel database was used to extract quantitative and qualitative data from the included studies.
Quantitative data extracted included year, country, study design, study population (age, education, socioeconomic status, health literacy, race or culture or ethnicity), number of participants, intervention format (web or paper based), comparison groups (standard care or absolute risk alone), clinical measures (blood pressure, cholesterol levels, weight, BMI, waist circumference, and prescribed medications), behavioral measures (medication adherence, lifestyle intentions or self-report), psychological measures (probabilistic or evaluative risk perceptions, positive or negative emotional responses, credibility, and recall), and a summary of significant effects of heart age communication.
Qualitative data extracted included behavioral themes (eg, lifestyle change), psychological themes (eg, emotional reaction), stated benefits of heart age (eg, motivates people to take action), and stated problems with heart age (eg, reduced credibility).
Intervention content data included additional risk communication formats (eg, absolute risk, risk level, graphs), underlying model (eg, 5-year or 10-year CVD risk model), and behavior change techniques (eg, email prompts or action plans) coded based on the taxonomy of 93 techniques by Michie et al . This was done based on the methods published in the results paper and any referenced protocol papers.
Risk of bias assessment included randomized studies that were critically appraised independently by 2 review authors (MF, C Bonner, or C Batcup if C Bonner was an author) using the relevant Johanna Briggs Institute tools for the study design . Disagreement was resolved by discussion, with decisions applied consistently where there was a common methodology, for example, if the participants were randomized on a computer, was allocation to groups concealed was marked as not applicable (N/A). Similarly, as all participants in all the papers could see the risk communication format they were allocated to, all were marked as No for being blind to treatment assignment.
shows the search process and results. In April 2020, 5839 database results and 159 references from previous reviews were assessed for eligibility, and in March 2021, an additional 543 database results and 1385 citations and references were reviewed. From 7626 total results, 16 eligible studies were identified with heterogeneous study designs and outcomes, and the results are reported below by study type: 5 randomized web-based experiments comparing heart age with percentage risk, 5 randomized clinical trials with mixed interventions, 2 mixed methods studies with quantitative outcomes, and 4 qualitative studies. Study details and outcomes are summarized by study category in - . [ - ] provides details about the measures used in each study, and [ , - ] provides details of behavior change techniques included in the control versus intervention groups.
|Study details||Intervention format||Comparison groups||CVDa risk algorithm||Participants||Principal findings|
|Soureti et al (2010) ; 2-arm RCTb in the United Kingdom||Web-based questionnaire, postintervention outcomes||1. 10-year percentage risk|
2. Heart age
|Framingham||413; 209 in percentage risk, 204 in heart age; aged 30-60 years|
|Witteman et al (2014) ; 2-arm RCT in the United States||Web-based questionnaire, postintervention outcomes||1. 10-year percentage risk|
2. 10-year percentage risk and heart age
|Framingham||3630, numbers in each group not given; aged 35-74 years; mean 53 (SD 10) years|
|Bonner et al (2015) ; 2 × 3 factorial design RCT in Australia||Web-based questionnaire. Participants shown either 5-year absolute risk or heart age, and within that different text and visual formats. Postintervention outcomes and followed up on the web 2 weeks later||1. 5-year percentage risk: (a) text only; (b) text + bar graph; (c) text + line graph|
2. Heart age: (a) text only; (b) text + bar graph; (c) text + line graph
|Framingham||570; 281 in percentage risk and 289 in heart age; aged 45-64 years; mean 54 (SD 6) years|
|Damman et al (2018) ; 2×2 factorial design RCT in the Netherlands||Web-based questionnaire (hypothetical results). Postintervention outcomes||1. Infographics of 10-year risk information (a) alone or (b) with a risk percentage and icon array|
2. Text of risk information (a) alone or (b) with a risk percentage and icon array
3. Heart age with infographics
|Framingham||727; 151 in infographics alone, 145 in infographics plus risk percentage, 133 in risk text alone, 168 in risk text plus risk percentage, 130 in heart age; aged 45-65 years|
|Van Der Pol-Harney et al (2021) ; 2 × 3 factorial design RCT in Australia||Web-based questionnaire. Participants randomized to one of 3 risk communication formats and received low or high risk based on self-report lifestyle risk factors. Postintervention outcomes||1. Lifetime percentage risk|
2. Heart age
3. Fitness age
|Provided with either low (5%, 16 years) or high (69%, 35 years) lifetime risk. High risk=smoke or eat 1 or no servings of fruit per day||174; 53 in percentage risk, 50 in heart age, 71 in fitness age; mean age 19 (SD 2.3) years|
aCVD: cardiovascular disease.
bRCT: randomized clinical trial.
cDASH: Dietary Approaches to Stop Hypertension.
dGP: general practitioner.
|Study details||Intervention format||Comparison groups||CVDa risk model||Participants||Principal findings|
|Lowensteyn et al (1998) ; RCTb in Canada||Physicians enrolled their own patients who they thought would benefit from a risk profile. Followed up 3 months later||1. Control: usual care|
2. Intervention: paper-based risk profile, including their 8-year risk of developing coronary disease, and how this risk would reduce if they changed one or more risk factors. Cardiovascular age also shown.
|Framingham||958; 176 in control, 782 in risk profile; aged 30-74 years; mean age 51 (SD 11) years|
|Grover et al (2007) ; RCT in Canada||Physicians enrolled their patients. Baseline visit, and followed up at 3, 6, 9, and 12 months||1. Control: usual care|
2. Intervention: paper-based risk profile including cardiovascular age, ongoing feedback on risk after lifestyle changes or medication
|Framingham (or Cardiovascular Life Expectancy Model for patients with CVD)||3053 received initial intervention: 1510 in risk profile group and 1543 in control; mean age 64 (SD 8) years|
|Lopez-Gonzalez et al (2013) ; RCT in Spain||Participants interviewed by researchers and clinical assistants; measurements taken. Follow-up measurements taken 12 months later||1. Usual care|
2. Percentage risk
3. Heart age
|Framingham||2844: 975 in usual care, 955 in percentage risk, and 914 in heart age; mean age 46 (SD 7) years|
|Näslund et al (2019) ; RCT in Sweden||Participants meeting with their primary care physician, measurements taken. Follow-up measurements taken 12 months later||1. Control: completing a primary care health survey including CVD risk factor screening, pharmacological CVD prevention if required, and advice on healthy lifestyle, and an ultrasound|
2. Intervention: the above plus a pictorial representation of carotid ultrasound (including vascular age) plus a nurse phone call to confirm understanding 2-4 weeks later and information repeated after 6 months
|Framingham||3532; 1783 in control, 1749 in intervention; aged 40-60 years|
|Svendsen et al (2020) ; cluster RCT in Norway||Participants discussed risk with pharmacy staff. Follow-up after 4 weeks|| 1. Control: conventional risk communication, each risk factor categorized in 4 groups from good (green) to poor (red), and diet and lifestyle advice given verbally and in written form|
2. Intervention: the above plus heart age
|JBSe-3||257; 120 in control, 137 in intervention; mean age 60 (SD 13) years|
aCVD: cardiovascular disease.
bRCT: randomized clinical trial.
cLDL: low-density lipoprotein.
dHDL: high-density lipoprotein.
eJBS: Joint British Societies.
|Study details||Intervention format||Comparison groups||CVDa risk model||Participants||Principal findings|
|Goldman et al (2006) ; focus groups in the United States (qualitative)||Responded to 3 visual representations of risk (all of a hypothetical man aged 42 years)||1. Icon chart risk vs ideal|
2. Bar chart risk vs ideal
3. Bar chart heart age vs ideal vs age
|Hypothetical person but based on Framingham||50 adults in 7 focus groups; aged 27-84 years|
|Bonner et al (2014) ; think-aloud process and interviews in Australia (qualitative)||Participants viewed 2 different heart age calculators||1. Heart age|
2. 10-year percentage risk and heart age
|Framingham||26 patients recruited from general practice; aged 39-67 years|
|Shefer et al (2016) ; interviews and focus groups in the United Kingdom (qualitative)||Patients randomized to different web-based questionnaires. Then either interviewed or took part in a focus group||1. Control|
2. Lifestyle advice only
3. Lifestyle advice plus 10-year percentage risk (phenotype)
4. Lifestyle advice plus 10-year percentage risk (phenotype + genetic) includes heart age and ideal risk
|Framingham||41 adults in interviews (22 in group 4, 15 in group 3, and 4 in group 2) and 13 adults in 2 focus groups (one with 6 patients and one with 7; 8 in group 4, 5 in group 3); aged 40-80 years|
|Riley et al (2020) . Recorded GPb consultations in the United Kingdom (qualitative)||GP consultations using either JBSc-3 or QRISK calculators were recorded and analyzed qualitatively||1. JBS-3 calculator|
2. QRISK calculator
|JBS-3 or QRISK||128 consultations analyzed; 64 in QRISK group and 64 in JBS-3; aged 40-74 years|
|Gidlow et al (2020) . Recorded GP consultations in the United Kingdom (quantitative)||GP consultations using either JBS-3 or QRISK calculators were recorded||1. JBS-3 calculator|
2. QRISK calculator
|JBS-3 or QRISK||173 general practice consultations; 73 QRISK and 100 JBS-3; aged 40-74 years|
|Bonner et al (2020) ; survey in Australia (quantitative outcomes and content analysis of open responses)||Web-based heart age calculator open to the public. Some participants elected to receive their results via email. A subgroup completed a survey about their results, 10 weeks after seeing them||1. Heart age||Framingham||361,044 heart age calculator users; 30,279 provided email to receive heart age report; 1303 survey respondents; Mean age of users 49; of those who requested a report 56; survey respondents 60|
aCVD: cardiovascular disease.
bGP: general practitioner.
cJBS: Joint British Societies.
Randomized Web-Based Experiments
Direct comparisons between heart age and absolute risk in 5 web-based experiments [- ], with no in-person computer lab experiments identified. The studies included 5514 consumers and found that heart age leads to more negative emotional responses (3/4, 75% of relevant studies show higher negative emotions or lower positive emotions); higher perceptions of CVD risk being higher probability, more serious or in a higher risk category (2/4, 50% of relevant studies); higher exact or verbal recall (2/2, 100% of relevant studies); lower perceived credibility (2/3, 67% of relevant studies); and generally had no effect on lifestyle intentions (1/5, 20% of relevant studies) or self-reported behavior (no study). One study compared heart age and absolute risk to fitness age among young adults and found that fitness age led to lower lifestyle change intentions compared with the other formats [ ]. All trials used the US Framingham model for risk except Van der Pol-Harney et al [ ], which used lifetime CVD risk estimates and hypothetical ages to indicate low risk (younger age) and high risk (older age). All trials measured self-reported outcomes immediately postintervention, and 1 study conducted a follow-up survey after 2 weeks [ ]. Soureti et al [ ] compared the calculated 10-year risk to heart age among 413 people and found no significant differences for smoking, diet or exercise intentions, risk perception, emotional response, or credibility. They found that younger people with higher risk were more likely to be worried and perceive the result as a wake-up call when receiving heart age. Witteman et al [ ] compared personalized 10-year risk to 10-year risk + heart age among 3630 people and found no significant differences in smoking, exercise, diet, weight, or physician visit intentions. Bonner et al [ ] compared personalized 5-year risk to heart age with varying graphical formats in a 2 × 3 design among 570 people and found no significant difference in diet, exercise, smoking, or physician visit intentions or behavior after 2 weeks, or information seeking. They found that heart age led to lower positive emotions and credibility, and higher risk perception (such that low-risk people thought they were high risk) and recall after 2 weeks. Damman et al [ ] compared hypothetical 10 year risk in various formats to heart age in a 2×2 design among 727 people and found mixed effects of heart age on intentions (higher intentions to visit a physician and exercise; no effect for diet or medication), higher risk perception, no difference in information perceptions relating to credibility (but heart age was perceived as more helpful), and higher recall for the verbal increased risk evaluative label. Van der Pol-Harney et al [ ] compared the hypothetical lifetime risk to heart age or fitness age among 174 younger adults with different low- versus high-risk values for those with and without lifestyle risk factors in a 2 × 3 design. They found that heart age and percentage risk were generally equivalent, but there was a detrimental effect of fitness age, including lower exercise and diet intentions, lower credibility when given a high-risk result, and lower risk perception for high-risk results. Receiving a high-risk result was associated with higher risk perception, higher worry especially for smokers, and lower credibility, especially if the results were worse than expected ( ).
Randomized Clinical Trials
When heart age was combined with additional strategies (eg, in person or phone counseling) in 5 applied trials [- ] for 9582 patients, it improved risk control (eg, reduced cholesterol and absolute risk) compared with usual care in most trials (4/5, 80% of relevant studies) up to 1 year. However, the direction of outcomes (lifestyle, blood pressure, cholesterol, and weight) was the same for absolute risk and heart age groups in one trial that compared each group to usual care (1/1, 100% of relevant studies). Follow-up periods ranged from 4 weeks to 12 months, and all trials used the US Framingham model for risk except Svendsen et al [ ], which used the UK JBS-3 model. Lowensteyn et al [ ] compared a paper-based risk profile intervention (8-year risk, cardiovascular age, effect of reducing risk factors) to usual care among 958 patients over 3 months and found no difference in blood pressure but lower cholesterol, leading to lower absolute risk. Grover et al [ ] compared a paper-based risk profile including cardiovascular age plus 3-monthly feedback to usual care among 2687 patients over 1 year and found lower blood pressure and cholesterol, leading to lower absolute risk. Lopez-Gonzalez et al [ ] compared a web-based interactive heart age tool to verbal communication of percentage risk or usual care among 2844 employees over 1 year and found higher physical activity and lower smoking, blood pressure, cholesterol, weight, and waist circumference in intervention groups versus control, with greater benefits in the heart age group; however, analyses of the difference between heart age and risk groups were not reported. Näslund et al [ ] compared 2 complex interventions, including one with heart age (heart age intervention: risk assessment and advice plus a carotid ultrasound image including vascular age and a nurse phone call, with information repeated after 6 months; control: risk assessment and advice only) among 3532 patients. They found higher use of cholesterol medication, lower cholesterol levels, and lower absolute risk in the heart age intervention group than in the control group; however, there was no difference in blood pressure or weight. Svendsen et al [ ] compared conventional risk communication (risk categories, colors, evaluative labels, verbal and written advice) to the same information plus heart age among 257 patients visiting a pharmacy, and they found that although using heart age was popular among pharmacy staff, adding heart age was no more effective than conventional risk communication alone in changing physical activity or reducing cholesterol and blood pressure levels ( ).
Mixed Methods Studies With Quantitative Outcomes
Two studies used a mixed methods design to investigate heart age: a survey of 1303 users of the Australian heart age calculator included quantitative outcomes and thematic content analysis of open responses ; and a UK study of video consultations using QRISK2 and JBS-3 risk calculators coded the content for quantitative outcomes in 12 general practices [ ]. Because of their study design, it is not possible to attribute outcomes to heart age from these studies, but they do suggest additional possible outcomes of heart age tools: different effects for older versus younger or equal heart age results, and risk communication content and time within consultations. An Australian survey found that a subsample of heart age users who completed a 10-week email journey had high recall and varied emotional responses to heart age, including motivation, optimism, anxiety, and worry [ ]. Recall was lower for equal heart age, and anxious or worried responses were higher for older heart age. Most of the respondents reported improved diet and exercise, with many reporting weight loss, reduced stress, and reduced smoking. People with older or equal heart age reported higher rates of diet and weight loss than those with younger heart age. People with older heart age were more likely to visit a physician or have a heart health check compared with those with younger or equal heart age. Credibility issues were identified in open responses. A UK study of consultations using 2 different risk communication tools found that 10% of consultation time (<2 minutes) was devoted to CVD risk [ ]. Using JBS-3 increased the time spent discussing CVD risk, the proportion of patients asking questions about CVD risk, discussion of heart age, and medication, whereas using QRISK2 increased the time spent talking about risk factors. One-fifth of consultation time was spent on interventions, mostly lifestyle ( ).
Mixed Methods Studies With Qualitative Analysis
Four studies used qualitative methods, including focus groups [, ], interviews [ - ], think-aloud [ ] and video prompt [ ] methods. In general, the findings reflected trial outcomes in relation to recall, risk perception, emotional response, motivation, and credibility. Heart age was noted by participants across the studies as being more memorable and easier to understand than other risk formats. In a UK study where recorded consultation videos were used as a prompt for interviews, some practitioners did not fully understand risk percentages and preferred to use heart age or seemed more confident in discussing heart age as opposed to percentage risk [ ]. This study also indicated that additional behavior change techniques may be added by the health provider depending on how they use the heart age tool. As reflected in the quantitative data, heart age also resulted in stronger emotional reactions from participants, for example, “it was the heart age that really shook me” [ ]; however, it was noted that this could be either motivating or frightening [ ]. Heart age prompted some participants to consider changing their lifestyle (eg, losing 9 kg [ ]; “my weight is something I need to work on” [ ]), which was also reflected in those who saw their risk percentage, and many with low or moderate risk scores discussed lifestyle changes. There was a suggestion that any intervention discussing risk with participants acts as a “reminder” or ”trigger,” “the kick that [they] needed” [ ] to change behavior. Similarly, discussing risk with a practitioner resulted in the intention to change behavior [ ]. Some participants were skeptical of the calculations for heart age and questioned their credibility ( ) [ , , ].
Behavior Change Techniques
The interventions included in the heart age studies varied in terms of behavior change techniques. From the Michie et al taxonomy of 93 techniques , we identified 14 from methods section descriptions or published protocols. All studies included salience of consequences by definition (as we interpreted heart age as increasing the salience of the risk assessment) and information about health consequences (ie, CVD outcomes). Instructions on how to perform a behavior, credible source, and comparative imagining of future outcomes were common for both the intervention and control groups. It should be noted that the results sections of the qualitative studies indicated that additional behavior change techniques may be used by health providers using these tools in a consultation, and this may influence how effective they are in clinical practice ( [ , - ]; ).
Risk of Bias Assessment
Risk of bias was noted for all randomized studies, with some applied clinical trials being particularly concerning in terms of unclear or questionable methods for randomization and analyses, including contamination between groups and analysis not per original randomized group. All experimental trials used self-reported outcomes rather than objective methods (, [ - ]).
When randomized trials are separated into direct comparisons between heart age and absolute risk versus complex interventions, there is limited evidence for the effectiveness of heart age over absolute risk expressed as a percentage risk over time in terms of lifestyle change. Heart age does appear to evoke a greater emotional response (both positive and negative), increases risk perception (although low-risk people may think they are high risk), and reduces credibility [- ]. Both percentage risk and heart age can be effective as risk communication formats when combined with other various behavior change techniques in applied settings, and have the potential to reduce blood pressure, cholesterol levels, and weight, which in turn can reduce absolute risk [ - ]. Only one study [ ] compared different labels for the heart age concept and found different psychological effects, indicating the importance of testing evaluative labels with the intended target audience. Qualitative and mixed methods studies generally reflected the outcomes measured in the randomized studies but tended to conclude that heart age was more motivating than percentage risk for lifestyle change, whereas randomized trial data did not support this assertion [ , - ].
Owing to the heterogeneity of both the intervention components and the outcomes, we were unable to synthesize the results as a meta-analysis. Therefore, our findings are descriptive across a range of outcomes and measurement methods. All randomized studies had some risk of bias, particularly some of the complex intervention studies in applied settings where randomization and analysis processes did not follow best practices, including contamination between groups and analyses not reflecting the initial randomization to groups. Outcomes in experimental studies were based on self-report rather than objective measurements.
Comparison With Previous Work
This is the first systematic review of the effects of heart age interventions. Previous reviews of CVD risk communication formats or biological age concepts have either been descriptive in relation to the models themselves [, ] or have not differentiated between the behavior change techniques used in intervention versus control groups, leading to mixed results overall [ , ]. This study shows the importance of identifying active ingredients in behavioral interventions to identify meaningful comparisons for future reviews [ ]. The design of applied clinical trials of heart age interventions did not differentiate between heart age as a risk communication format and supplementary behavior change techniques, with an insufficient description of the meaningful differences between the intervention and control groups. The finding in one study that the label for the heart age concept (heart age vs fitness age) affected outcomes echoes recent findings on different terms for elevated blood pressure [ , ].
The risk of bias could be improved in future heart age trials, but we note that blinding is not generally possible in a risk communication intervention. The mixed methods studies suggest additional outcomes should be included, such as overall consultation time, clinical communication content, and time spent on different aspects (eg, risk factors, risk communication formats, risk manipulation, lifestyle change, and medication [, ]). Several studies highlight the importance of differentiating among older, younger, and equal heart age results in analyses and considering expectations in relation to this [ , , ]. Authors of heart age studies need to take care to specify the components of their interventions, including the risk communication tool itself and the way that health care providers use and explain this in a consultation. Our findings suggest that the most appropriate outcomes to measure for heart age as a standalone risk communication format are emotional response, perceived credibility, and risk perception. Combining heart age with other behavior change techniques may be effective for behavior change if they are selected for a specific outcome in mind based on other evidence. Authors should avoid overstating the benefits of heart age as a standalone risk communication or behavior change tool by ensuring that all conclusions are supported by the data.
The appeal of heart age to consumers is suggested by its widespread use among millions of people worldwide [- ]. However, if lifestyle change is the intended outcome, additional support is needed using evidence-based behavior change techniques, such as action plans and goal-setting. This is in line with the behavior change literature, where many different health models that show risk communication is necessary but not sufficient for behavior change. Heart age is a risk communication format that can capture attention and provoke an emotional response, but it is not enough as a standalone intervention for behavior change [ ]. For example, in the Australian survey identified in this review, the initial heart age assessment was followed by a 10-week email journey where behavior change could be reinforced with prompts, activities and planning tools [ ]. In a recent review on how to present probabilities in patient decision aids, biological age was not recommended because it may undermine understanding of absolute risk, which is required for making informed, shared decisions about medication [ ]. General practitioners, nurses, and cardiologists engaged in CVD risk assessment need to consider what their communication aim is to determine whether heart age or absolute risk is most appropriate. If heart age is communicated with the aim of motivating lifestyle change, it needs to be supported by other behavior change techniques such as action plans.
This review found little evidence that heart age motivates lifestyle behavior change more than percentage risk, but either format can improve clinical outcomes when combined with other behavior change strategies. The label for the heart age concept can affect outcomes and should be pretested with the intended audience. Future research should more carefully specify the components of the intervention, avoid overstating the benefits of heart age as a standalone risk communication format, consider effects on physician-patient consultations, and differentiate between older and younger heart age results.
C Bonner, C Batcup, LT, and JD have been involved in developing and evaluating heart age tools for research projects, including papers cited in this review. AH works for the Heart Foundation, which developed a heart age tool in Australia.
Conflicts of Interest
C Bonner is supported by the National Health and Medical Research Council/Heart Foundation fellowship, and C Batcup is supported by the Heart Foundation grant. C Bonner had authored some papers included this study.
Measures used in each study, and whether significant outcomes were found.XLSX File (Microsoft Excel File), 37 KB
Behavior change techniques by intervention group.XLSX File (Microsoft Excel File), 13 KB
- Bonner C, McKinn S, Lau A, Jansen J, Doust J, Trevena L, et al. Heuristics and biases in cardiovascular disease prevention: how can we improve communication about risk, benefits and harms? Patient Educ Couns 2018;101(5):843-853. [CrossRef] [Medline]
- Bonner C, Jansen J, McKinn S, Irwig L, Doust J, Glasziou P, et al. Communicating cardiovascular disease risk: an interview study of general practitioners' use of absolute risk within tailored communication strategies. BMC Fam Pract 2014;15(1):106 [FREE Full text] [CrossRef] [Medline]
- JBS3 Board. Joint British Societies' consensus recommendations for the prevention of cardiovascular disease (JBS3). Heart 2014;100 Suppl 2(5):1-67. [CrossRef] [Medline]
- Guidelines for the assessment of absolute cardiovascular disease risk: approved by the National Health and Medical Research Council. National Vascular Disease Prevention Alliance (NVDPA). 2009. URL: https://www.diabetesaustralia.com.au/for-health-professionals/best-practice-guidelines/absolute-cardiovascular-risk/ [accessed 2021-10-07]
- Hippisley-Cox J, Coupland C, Brindle P. Development and validation of qrisk3 risk prediction algorithms to estimate future risk of cardiovascular disease: prospective cohort study. Br Med J 2017;357:j2099 [FREE Full text] [CrossRef] [Medline]
- Whelton PK, Carey RM, Aronow WS, Casey DE, Collins KJ, Himmelfarb CD, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA guideline for the prevention, detection, evaluation, and management of high blood pressure in adults: a report of the American College of Cardiology/American Heart Association task force on clinical practice guidelines. J Am Coll Cardiol 2018;71(19):127-248 [FREE Full text] [CrossRef] [Medline]
- Sposito AC, Ramires JA, Jukema JW, Molina JC, da Silva PM, Ghadanfar MM, et al. Physicians' attitudes and adherence to use of risk scores for primary prevention of cardiovascular disease: cross-sectional survey in three world regions. Curr Med Res Opin 2009;25(5):1171-1178. [CrossRef] [Medline]
- Hobbs FD, Jukema JW, Da Silva PM, McCormack T, Catapano AL. Barriers to cardiovascular disease risk scoring and primary prevention in Europe. QJM 2010;103(10):727-739. [CrossRef] [Medline]
- Gupta M, Singh N, Tsigoulis M, Kajil M, Hirjikaka S, Quan A, et al. Perceptions of Canadian primary care physicians towards cardiovascular risk assessment and lipid management. Can J Cardiol 2012;28(1):14-19. [CrossRef] [Medline]
- Gupta R, Stocks N, Broadbent J. Cardiovascular risk assessment in Australian general practice. Aust Fam Physician 2009;38(5):364-368 [FREE Full text] [Medline]
- Wells S, Kerr A, Eadie S, Wiltshire C, Jackson R. Your heart forecast': a new approach for describing and communicating cardiovascular risk? Heart 2010;96(9):708-713. [CrossRef] [Medline]
- Waldron C, van der Weijden T, Ludt S, Gallacher J, Elwyn G. What are effective strategies to communicate cardiovascular risk information to patients? A systematic review. Patient Educ Couns 2011;82(2):169-181. [CrossRef] [Medline]
- Neufingerl N, Cobain MR, Newson RS. Web-based self-assessment health tools: who are the users and what is the impact of missing input information? J Med Internet Res 2014;16(9):e215 [FREE Full text] [CrossRef] [Medline]
- Bonner C, Raffoul N, Battaglia T, Mitchell JA, Batcup C, Stavreski B. Experiences of a national web-based heart age calculator for cardiovascular disease prevention: user characteristics, heart age results, and behavior change survey. J Med Internet Res 2020;22(8):e19028 [FREE Full text] [CrossRef] [Medline]
- Patel RS, Lagord C, Waterall J, Moth M, Knapton M, Deanfield JE. Online self-assessment of cardiovascular risk using the Joint British Societies (JBS3)-derived heart age tool: a descriptive study. BMJ Open 2016;6(9):e011511 [FREE Full text] [CrossRef] [Medline]
- Groenewegen K, den Ruijter H, Pasterkamp G, Polak J, Bots M, Peters SA. Vascular age to determine cardiovascular disease risk: a systematic review of its concepts, definitions, and clinical applications. Eur J Prev Cardiol 2016;23(3):264-274. [CrossRef] [Medline]
- Kulendrarajah B, Grey A, Nunan D. How effective are 'age' tools at changing patient behaviour? A rapid review. BMJ Evid Based Med 2020;25(2):1-2. [CrossRef] [Medline]
- Bonner C, Bell K, Jansen J, Glasziou P, Irwig L, Doust J, et al. Should heart age calculators be used alongside absolute cardiovascular disease risk assessment? BMC Cardiovasc Disord 2018;18(1):19 [FREE Full text] [CrossRef] [Medline]
- Michie S, Richardson M, Johnston M, Abraham C, Francis J, Hardeman W, et al. The behavior change technique taxonomy (v1) of 93 hierarchically clustered techniques: building an international consensus for the reporting of behavior change interventions. Ann Behav Med 2013;46(1):81-95. [CrossRef] [Medline]
- Spiegelhalter D. How old are you, really? Communicating chronic risk through 'effective age' of your body and organs. BMC Med Inform Decis Mak 2016;16(1):104 [FREE Full text] [CrossRef] [Medline]
- Bonner C, Batcup C, Cornell S. Protocol for a systematic review of qualitative and quantitative effects of cardiovascular disease risk communication using heart age concepts: Preprint. medRxiv 2020:20089938. [CrossRef]
- Critical appraisal tools. Joanna Briggs Institute. URL: https://jbi.global/critical-appraisal-tools [accessed 2021-06-08]
- Soureti A, Hurling R, Murray P, van Mechelen W, Cobain M. Evaluation of a cardiovascular disease risk assessment tool for the promotion of healthier lifestyles. Eur J Prev Cardiol 2010;17(5):519-523. [CrossRef]
- Witteman HO, Fuhrel-Forbis A, Wijeysundera HC, Exe N, Dickson M, Holtzman L, et al. Animated randomness, avatars, movement, and personalization in risk graphics. J Med Internet Res 2014;16(3):e80 [FREE Full text] [CrossRef] [Medline]
- Bonner C, Jansen J, Newell BR, Irwig L, Teixeira-Pinto A, Glasziou P, et al. Is the “heart age” concept helpful or harmful compared to absolute cardiovascular disease risk? An experimental study. Med Decis Making 2015;35(8):967-978. [CrossRef]
- Damman OC, Vonk SI, van den Haak MJ, van Hooijdonk CM, Timmermans DR. The effects of infographics and several quantitative versus qualitative formats for cardiovascular disease risk, including heart age, on people's risk understanding. Patient Educ Couns 2018;101(8):1410-1418. [CrossRef] [Medline]
- Van Der Pol-Harney E, Turner R, McCaffery K, Bonner C. The effects of communicating cardiovascular disease risk as 'fitness age' on behavioral intentions and psychological outcomes. Patient Educ Couns 2021;104(7):1704-1711. [CrossRef] [Medline]
- Lowensteyn I, Joseph L, Levinton C, Abrahamowicz M, Steinert Y, Grover S. Can computerized risk profiles help patients improve their coronary risk? The results of the coronary health assessment study (CHAS). Prev Med 1998;27(5 Pt 1):730-737. [CrossRef] [Medline]
- Grover SA, Lowensteyn I, Joseph L, Kaouache M, Marchand S, Coupal L, et al. Patient knowledge of coronary risk profile improves the effectiveness of dyslipidemia therapy: the CHECK-UP study: a randomized controlled trial. Arch Intern Med 2007;167(21):2296-2303. [CrossRef] [Medline]
- Lopez-Gonzalez AA, Aguilo A, Frontera M, Bennasar-Veny M, Campos I, Vicente-Herrero T, et al. Effectiveness of the heart age tool for improving modifiable cardiovascular risk factors in a Southern European population: a randomized trial. Eur J Prev Cardiol 2015;22(3):389-396. [CrossRef] [Medline]
- Näslund U, Ng N, Lundgren A, Fhärm E, Grönlund C, Johansson H, et al. Visualization of asymptomatic atherosclerotic disease for optimum cardiovascular prevention (VIPVIZA): a pragmatic, open-label, randomised controlled trial. Lancet 2019;393(10167):133-142. [CrossRef]
- Svendsen K, Jacobs D, Mørch-Reiersen LT, Garstad KW, Henriksen HB, Telle-Hansen VH, et al. Evaluating the use of the heart age tool in community pharmacies: a 4-week cluster-randomized controlled trial. Eur J Public Health 2020;30(6):1139-1145. [CrossRef] [Medline]
- Gidlow CJ, Ellis NJ, Cowap L, Riley VA, Crone D, Cottrell E, et al. Quantitative examination of video-recorded NHS health checks: comparison of the use of QRISK2 versus JBS3 cardiovascular risk calculators. BMJ Open 2020;10(9):e037790 [FREE Full text] [CrossRef] [Medline]
- Goldman RE, Parker DR, Eaton CB, Borkan JM, Gramling R, Cover RT, et al. Patients' perceptions of cholesterol, cardiovascular disease risk, and risk communication strategies. Ann Fam Med 2006;4(3):205-212 [FREE Full text] [CrossRef] [Medline]
- Shefer G, Silarova B, Usher-Smith J, Griffin S. The response to receiving phenotypic and genetic coronary heart disease risk scores and lifestyle advice - a qualitative study. BMC Public Health 2016;16(1):1221 [FREE Full text] [CrossRef] [Medline]
- Bonner C, Jansen J, Newell BR, Irwig L, Glasziou P, Doust J, et al. I don't believe it, but I'd better do something about it: patient experiences of online heart age risk calculators. J Med Internet Res 2014;16(5):e120 [FREE Full text] [CrossRef] [Medline]
- Riley V, Ellis NJ, Cowap L, Grogan S, Cottrell E, Crone D, et al. A qualitative exploration of two risk calculators using video-recorded NHS health check consultations. BMC Fam Pract 2020;21(1):250 [FREE Full text] [CrossRef] [Medline]
- Sheridan SL, Crespo E. Does the routine use of global coronary heart disease risk scores translate into clinical benefits or harms? A systematic review of the literature. BMC Health Serv Res 2008;8(1):60 [FREE Full text] [CrossRef] [Medline]
- Muscat DM, Morris GM, Bell K, Cvejic E, Smith J, Jansen J, et al. Benefits and harms of hypertension and high-normal labels: a randomized experiment. Circ Cardiovasc Qual Outcomes 2021;14(4):e007160. [CrossRef] [Medline]
- Michie S, van Stralen MM, West R. The behaviour change wheel: a new method for characterising and designing behaviour change interventions. Implement Sci 2011;6:42 [FREE Full text] [CrossRef] [Medline]
- Bonner C, Trevena LJ, Gaissmaier W, Han PK, Okan Y, Ozanne E, et al. Current best practice for presenting probabilities in patient decision aids: fundamental principles. Med Decis Making 2021;41(7):821-833. [CrossRef]
|CVD: cardiovascular disease|
|JBS: Joint British Societies|
Edited by G Eysenbach; submitted 08.06.21; peer-reviewed by A Pfammatter, N Reynoso-Noverón, M Salimi; comments to author 17.07.21; revised version received 23.08.21; accepted 13.09.21; published 05.11.21Copyright
©Carissa Bonner, Carys Batcup, Samuel Cornell, Michael Anthony Fajardo, Anna L Hawkes, Lyndal Trevena, Jenny Doust. Originally published in JMIR Cardio (https://cardio.jmir.org), 05.11.2021.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Cardio, is properly cited. The complete bibliographic information, a link to the original publication on https://cardio.jmir.org, as well as this copyright and license information must be included.