Cardiovascular disease risk communication in NHS Health Checks using QRISK®2 and JBS3 risk calculators: the RICO qualitative and quantitative study

General practices

General practices were recruited if they met the following criteria:

• were delivering NHSHCs

• were already using the QRISK2 percentage risk score during health checks

• were already delivering (or were willing to deliver) health checks in specific clinics to facilitate data collection

• had signed up to the ‘incentive scheme’ implemented by the Clinical Research Network (CRN) to ensure that the practice is ‘research ready’

• were willing to participate.

Postcodes were used to stratify general practices into the bottom or top 50% based on national deprivation rankings,62 as a proxy for the typical socioeconomic status of the local population (Table 1).

Patients

The patient population were those eligible for NHSHCs based on national criteria. These criteria excluded people who:

• were outside the target age range of 40–74 years

• had existing diagnoses for certain cardiovascular-related chronic conditions

• were taking statins

• had attended an NHSHC in the last 5 years

• were known to be at high risk of CVD (i.e. had a 10-year CVD risk score of ≥ 20%). 17

Practitioners

Participating practitioners were health-care professionals who usually delivered health checks in participating practices and who were willing to participate (usually one or two PNs or HCAs per practice).

Practice sampling

The CRN facilitated practice sampling. Briefly, this involved an initial e-mail inviting expressions of interest, followed up with telephone calls and subsequent practice visits. Practice participation was incentivised through financial reimbursement of service support costs and additional remuneration for completing all parts of the study. Following practice-level consent, practice pairs that were matched on level of deprivation (bottom vs. top 50% of national deprivation rankings), were randomly assigned to the QRISK2 or the JBS3 group using a random number generator in Microsoft Excel^® (Microsoft Corporation, Redmond, WA, USA). After randomisation, the research team completed initiation meetings at practices to provide further information and basic training for staff involved.

Patient and practitioner sampling

There were three levels of patient sampling:

1. Total sample (target, n = 240) – to achieve the 144 recorded consultations suitable for qualitative analysis (12 recorded consultations per practice allowing for non-attendances and consultations with no/minimal discussion of CVD risk), the aim was that health check clinics were recorded until 20 recordings per practice (240 total) were complete. In each practice, the patient database was searched to identify eligible patients, who were then stratified by age, sex and ethnicity to ensure that there was representation from different demographic groups (see Table 1).

   Invitation letters, information sheets and consent forms were sent out to eligible patients (up to 400 per practice depending on the size of the eligible cohort), asking them to contact the CRN for more information or to arrange their video-recorded health check. Those who did not respond were contacted by telephone by the general practice staff.

2. Qualitative analysis (target, n = 144) – video-recordings were quantitatively coded soon after the health check to identify those for qualitative analysis (target of 12 video-recordings per practice) and VSR interview (target of four VSR interviews per practice). Where risk was not discussed by patient or practitioner, the patient’s data were not used for either.

3. VSR interviews (target, n = 48) – VSR interviews were conducted with patients (target, n = 24 VSR interviews per group) sampled from the 144 recorded health checks. The aim was to stratify by sex, age and CVD risk (Table 2), although issues with recruitment meant that this could not be strictly adhered to and, in some under-recruiting practices, all those who consented were asked to take part.

The target of 144 recorded consultations (12 per practice) was comparable to other studies using audio-recordings of similar consultations to explore CVD risk communication in patients with psoriasis (e.g. n = 130 in 10 practices),63 while the targets of 48 patient VSR interviews and 18 practitioner VSR interviews were in alignment with the number of interviews carried out in other VSR studies (n = 9–39). 52,64

All practitioners who delivered the video-recorded health check were asked to participate in VSR interviews.

Groups

General practices were randomly assigned to one of two groups:

1. QRISK2 group (usual practice) – practitioners delivered health checks using the QRISK2 risk calculator as per usual practice.

2. JBS3 group (intervention) – practitioners delivered health checks using the JBS3 risk calculator. 40 An introductory session with practitioners established the minimum requirements to use:

   • the first two output screens, which show heart age and healthy years (event-free survival age)

   • risk score manipulation to show the effects of intervention through modifying one of those risk factors (e.g. lowering blood pressure, smoking cessation) (see Figure 1).

Practitioners were also asked to practise using JBS3 in at least two health checks prior to the video-recorded clinics and could seek further clarification from the research team should they wish. During these introductory sessions, practitioners were provided with a verbal explanation of the tool, given written materials to support data entry and given a digital versatile disc (DVD) and link to an online training video (www.youtube.com/watch?v=idecGzlwIc4%26feature=youtu.be; accessed June 2021), which was also played to them during the introductory session. As a requirement of the NHSHC programme, patients in the JBS3 group were also told their QRISK2 10-year risk.

Video-recorded health checks

Digital camcorders were positioned in health check clinic rooms to provide an audio-visual record of consultations. Informed by patient and public involvement (PPI) and pilot work, cameras were positioned to capture both the patient and the practitioner, but prioritising the view of the patient. Video-recordings were screened soon after the health check. If there was no discussion of CVD risk, this was noted and the file retained. In the case of consultations that involved discussion of CVD risk and were eligible for qualitative analysis, the audio-recording was separated from the visual recording [using Adobe^® Premiere Pro (Adobe Systems Incorporated, San Jose, CA, USA)] for transcription and qualitative analysis.

Semistructured video-stimulated recall interviews with patients and practitioners

Semistructured one-to-one VSR interviews with patients were conducted at the patients’ home or their general practice (depending on patient preference) within 4 weeks of their health check. VSR interviews with practitioners were conducted at the general practice within 2 weeks of their final recorded health check. No others (i.e. non-participants) were present in the room during interviews. After each clinic, the recorded heath checks were watched to identify sections of the consultation to use in VSR interviews that related to discussion of the CVD risk score, modification of the risk score (in the JBS3 group) and practitioner advice, recommendations and interventions. During their VSR interviews, practitioners were shown excerpts of video consultations with some of their patients who had subsequently been selected for VSR interview. Semistructured VSR interviews followed a pre-piloted process and topic guide, with variation depending on whether the patient/practitioner was in the QRISK2 group or the JBS3 group. All VSR interviews were audio-recorded and transcribed verbatim for analysis.

Patient medical record review

Data from the 12 weeks following the health check were extracted from patient medical records. Data were processed by the principal investigator (CJG), with verification by a GP co-investigator (EC), to identify any relevant recorded activity or prescriptions that occurred as a result of the health check. Classifications were data driven, as detailed in Appendix 1, Tables 11 and 12. Where possible, these were based on Read codes/terms, but also included uncoded additional text.

Qualitative data: recorded health check consultations

Qualitative data were analysed using deductive thematic analysis,65 following the six-stage process described by Braun and Clarke66 (Table 3). A coding template was developed around PMT. 56 Use of this framework (see Figure 2) was considered appropriate to investigate the use of the two CVD risk assessment tools given the complexity of patient–practitioner interactions52,53 and the translation of risk information into health-protective behaviour. 54

Each transcript was uploaded to NVivo version 12 (QSR International, Warrington, UK) qualitative data analysis software. 67 This allowed the interpretation of how QRISK2 and JBS3 were used to communicate risk in the context of PMT components (e.g. verbal persuasion, influencing patient prior beliefs and priorities, and how patients respond, which will reflect the nature of their appraisal within the consultation).

Initially, 14 transcripts were inductively coded independently by Lisa Cowap and Victoria Riley. This was to check the application of PMT to health check consultations and agree coding between the researchers. Following inductive coding, 13 new codes were added to the framework (e.g. medical history, clarification of results). The final version of the coding template, including examples for each code from health check consultations, is provided in Appendices 3 and 4. The remaining 114 transcripts were individually coded by Lisa Cowap and Victoria Riley; 2 out of every 20 transcripts were independently dual coded to check their reliability using kappa coefficients for each node in the PMT framework. Reliability ranged from 0.48 to 0.71 over the five reliability checks conducted, indicating fair to good reliability. 68 Data saturation was considered to have been reached at the point of completion of coding.

Subsequent analysis of codes was led by Sophia Fedorowicz (qualitative researcher and doctoral student, Staffordshire University) (supported by SG, CJG, NE and VR) to identify codes for key elements of the PMT model, splitting health checks into two groups (QRISK2 and JBS3). Specific parts of transcripts that illustrated the practitioner communicating CVD risk to the patient, and the patient responses, were identified. These related to cognitive appraisal (threat appraisal and coping appraisal) and coping modes (adaptive and maladaptive).

Qualitative data: semistructured video-stimulated recall interviews with patients and practitioners

Patient VSR interview transcripts were analysed using inductive thematic analysis, with codes and themes generated from data based on individual reflections, perceptions and experiences. In the case of patient VSR interviews, line-by-line coding and preliminary theme development were undertaken by Lisa Cowap. Themes were discussed with Victoria Riley and Sarah Grogan, before reviewing and agreeing final themes.

Practitioner VSR interviews were line-by-line coded by Naomi J Ellis and Sarah Grogan (see Acknowledgements). Sian Calvert led theme development and was supported by Victoria Riley and Christopher J Gidlow, who reviewed and agreed the final themes.

Quantitative: content of health check consultations (process evaluation)

Recorded health checks were viewed by two authors (LC and VR) and the content of the consultations was characterised using a second-by-second coding framework developed specifically for this study. The framework comprised 36 items grouped into six categories: patient–practitioner communication, health check general (e.g. collecting and inputting data), risk dialogue (e.g. overall discussion of risk, 10-year risk reference, heart age, patient question on CVD risk), causal CVD risk factors (e.g. medical, lifestyle), risk management (lifestyle intervention or medical intervention) (see Appendix 2). This allowed derivation of aggregate indicators for each consultation to allow between-group comparisons of:

• length of health check

• time (absolute and proportion of consultation) discussing CVD risk, CVD risk factors (overall, lifestyle, medical) and risk management (lifestyle, medical)

• practitioner–patient communication balance (proportion of health check time for which practitioners and patients spoke, ratio of practitioner to patient speaking time)

• number and proportion of patients asking questions about CVD risk

• use of heart age, healthy years (event-free survival age) and risk score manipulation (as fidelity check in the JBS3 group).

As noted in a previous paper,1 the coding process and framework development was iterative, using four mock health checks that were video-recorded as part of PPI. To reach consistency in approach, Naomi J Ellis and Lisa Cowap coded mock health checks by consensus. Victoria Riley then coded the same four consultations independently and intraclass correlation coefficients (ICCs) demonstrated excellent inter-rater reliability (ICCs from 0.968 to 0.995). Once data collection had started, health checks were coded by authors Lisa Cowap and Victoria Riley, with verification of 10% (2 out of every 20 independently coded) to mitigate the risk of coder drift. ICCs ranged from 0.992 to 0.999, indicating excellent inter-rater reliability.

Data on patient sex, age, ethnic background (classified as white British or ethnic minority) and lower-layer super output area (which was used to derive deprivation decile)62 were also extracted from patient medical records.

Following checks for normal distributions, QRISK2 and JBS3 groups were compared using key variables. To take into account the nature of the sampling, which was in clusters, 95% confidence intervals (CIs) were calculated; as usual, where the CIs of the two groups do not overlap then the groups can be considered to differ significantly. Data processing and analysis were performed in SPSS version 26 (IBM Corporation, Armonk, NY, USA).

Quantitative: patient medical record review (process evaluation)

Activities extracted from the patient medical records were classified as relating to further assessment of CVD risk (or QRISK), weight, blood pressure, cholesterol level, diabetes, other tests (e.g. liver function test, kidney function test, echocardiography), lifestyle, lifestyle referral and new diagnoses. Relevant medications that were prescribed following the health check were grouped as statin/lipid lowering, anticoagulants, cardiovascular or diabetes. The specific composition of each category is shown in Appendix 1, Tables 11 and 12.

Between-group comparisons were explored, but were somewhat limited by the relatively small number of patients with relevant recorded activity or prescriptions in their medical records.

Case studies

A subsample of 10 patients were selected for within-case analysis. Selection was on the basis of evidence of positive patient intentions and/or behaviours to reduce their CVD risk following the health check, and to provide coverage across general practices. The aim was to further explore apparent mechanisms by which the risk calculators may lead to changes in patient or practitioner behaviour. A data extraction template was created to bring together data from all sources (see Appendix 6, Table 13).

Length of NHS Health Check consultations

Table 7 summarises the characteristics of health check consultations by study group and overall. Consultation duration varied widely (range 6.8 to > 38 minutes), but the majority of consultations lasted between 15 and 20 minutes, with most consultations (60%) lasting < 20 minutes (Figure 3). Consultations were only slightly shorter, on average, in the QRISK2 group than in the JBS3 group (with a relatively small effect size of 0.13).

FIGURE 3.

Duration of health check consultation by CVD risk calculator group.

Discussion of cardiovascular disease risk

Overall, < 10% (9.1% ± 4.3%) of consultation time was devoted to CVD risk discussion, which equated to 1.7 ± 0.83 minutes. A higher proportion of consultation time was spent discussing CVD risk when using JBS3 (equivalent to 2.1 ± 0.82 minutes) than when using QRISK2 (equivalent to 1.31 ± 0.63 minutes), with a medium effect size. Nearly all health checks in both groups included reference to the 10-year percentage CVD risk score (94% vs. 94.5%; r = 0.01). The proportion of patients asking questions about CVD risk was higher in the JBS3 group than in the QRISK2 group (32.0% vs. 12.3%; r = 0.23).

In the JBS3 group, nearly all health checks included discussion of heart age (100%) and event-free survival age (97%), and manipulation of the risk score(s) to show the potential effect of intervention on risk (92%). This showed fidelity to the requested minimum use of JBS3 outputs. Use of heart age and risk manipulation was also evident in 52.1% and 21.9% of QRISK2 consultations, respectively. This is a result of two general practices in the QRISK2 group using Informatica (a software addition that offers some JBS3 functionalities), and because heart age and risk manipulation are possible (but not main features) in QRISK2.

Discussion of cardiovascular disease risk factors and risk management

Over one-third of total health check time was spent discussing causal CVD risk factors. This was slightly higher in health checks using QRISK2 than in those using JBS3 (small to medium effect size), but with wide variation within groups (see Table 7).

Interventions to manage risk were discussed for approximately one-fifth of total consultation time and predominantly related to lifestyle rather than medical intervention, which was not discussed at all in over 30% of QRISK2-informed and 42% of JBS3-informed health checks (r = –0.12).

Verbal dominance

Practitioners spoke for just over half of the total time in QRISK2 consultations and just under half in JBS3 consultations (compared with ≈ 23% for patients) (Figure 4). There was an indication of higher practitioner verbal dominance in health checks using QRISK2 than in those using JBS3 (r = 0.27).

FIGURE 4.

Mean percentage of total health check time with speaking by practitioner or patient. Error bars: 95% CI.

Threat appraisal

Threat appraisal focuses on the source of a threat (CVD risk) and factors that increase or decrease the probability of maladaptive responses (i.e. behaviours that inhibit patients’ ability to adjust to the threat). This theme is central to the health check. It focuses on the discussion of risk as it relates to patients’ perceived severity of CVD risk, the consequences of CVD, the perceived vulnerability to future CVD, and the intrinsic and extrinsic rewards for not addressing CVD risk [i.e. perceived benefits of not acting to manage or reduce risk (maladaptive response)]. Threat appraisal was the most commonly identified element of the PMT model. It was observed in all health checks, although less frequently in those using JBS3 (coded 584 times; average 9 times per consultation) than those using QRISK2 (coded 634 times; average 10 times per consultation).

Patients, when presented with their 10-year risk, generally acknowledged it, but the extent of their understanding was often unclear. For example, one asked ‘Is that percentage of risk alright?’. Most often, the risk score was acknowledged with a single-word response, such as ‘yeah’ or ‘OK’, limiting practitioners’ ability to gauge the patient’s understanding and classification of response for this analysis. Heart age in JBS3 aided patient understanding of CVD risk, resulting in questions such as ‘. . . so really what can I do about that? I mean I know it is all estimated’. Such questions reflected a level of understanding of the score and intention towards risk-reducing behaviour. Several patients expressed surprise at their risk. The patient in the extract below appeared to question how the score was calculated because he perceived himself to be healthier than the outcome suggested, leading to some mistrust. They also made two references to being ‘fitter’ than the risk score indicated, which was not addressed by the practitioner:

I thought I was fitter than that though.

[Laughter] You are doing good exercises.

But I was fitter than that though . . .

OK, so the health years, so on average expect to survive is 80 for yourself without a heart attack or a stroke, yeah? And then your risk of a heart attack or stroke in the next 10 years is 15%, so you do need to look after yourself, because we would say that is a medium risk.

Yes.

So wouldn’t say it is too high or low, but a medium to high.

OK.

OK, and then that’s what it looks like so from now until there, that’s the last one the chance of surviving without a heart attack.

That’s estimated?

This is estimated, we don’t know what’s going to happen you might be even longer.

So about 94 I might snuff it?

JBS3, 11_028, male, 58

Because the practitioner appeared to ignore the patient’s surprise, continuing to focus on the process of the health check, the patient switched off briefly, until presented with his event-free survival age. Moreover, there was evidence of misunderstanding among some patients and practitioners who interpreted event-free survival age as estimated age of survival (i.e. age of death).

Practitioners provided little follow-up explanation of risk scores when using QRISK2 or JBS3:

Right, this is the screening I was telling you about. I will just print that out for you. So your risk of any heart disease is 15%.

Yeah, which is not very high.

It does increase with age. If it is above 10% we then pass it on for them to have a look at it and they will be able to decide when to have your next health check, which should be 3 years or 1 year. Obviously next time you come in any results you’ve got in the red tend to up your risk and they tend to up your heart age as well. So when you come in next time if your blood pressure is back down, and obviously it could be less so . . . Your heart age has come up as 66.

Well I am 66 this year.

Yes, yes, so it is quite near isn’t it? Yes. So, for example, if you were a smoker and that was in the red that would put your heart age at 75. So the only one we have got in the red really is that one cholesterol . . .

It’s only marginal though isn’t it?

QRISK2+, 2_016, male, 65

The patient quoted above was identified as medium–high risk, but the practitioner did not elaborate on the severity or implications, leaving the patient’s interpretation of his risk score as ‘not very high’. This was compounded when the patient received his heart age. The practitioner did not address the patient’s misinterpretation of the severity of his risk nor explain why his results were conflicting (i.e. percentage risk score is age and sex dependent, heart age is not), again perhaps focusing on the consultation process more than the patient. This led the patient to dismiss his elevated cholesterol as ‘only marginal’. The absence of active listening skills was recurrent across both groups, making it difficult to gauge patient understanding, and lends support to the apparent imbalance between patient and practitioner contributions (see Chapter 3).

There was evidence, albeit still limited, that patient engagement in conversations about the threat of CVD was greater in the JBS3 group, prompted by risk score manipulation (e.g. practitioners visually showed patients that a reduction in blood pressure could lower their heart age):

. . . so obviously your blood pressure is not too bad, that is fine where it is at 128, but your cholesterol, so ideally we like that to be below 5. So if you could get it below 5, so let’s put it down to 4.8, you can see that automatically that it brings your risk down to 1.8%.

Oh, I see, yes.

. . . improves your life expectancy slightly, and probably brings your heart age down a year. So it is just, you know, showing that it can and, obviously, the lower you can keep these factors that you influence, for longer, the better quality of life and life expectancy there is . . . your risk is going to increase slightly with age. So it is about trying to moderate those other factors.

So what impact does exercise have on that?

It has quite a significant impact on your cholesterol, it does help your cholesterol a lot. We know that it helps because that increases your good cholesterol, which can help increase the balance so, that can help with it as well.

So what’s the normal range that is seen for HDL cholesterol?

HDL can be anything from sort of 1.1 to about 2.5, you don’t get much over, I can’t say I have seen many, I have seen a few. But your cholesterol could be anything down to you know 3.5.

OK and really bad would be?

6 or 7s, so would be sort of . . .

Oh, OK – so 5.6 is yeah it is edging up, isn’t it?

JBS3, 7_020, male, 45

The patient evaluated the threat and sought information to facilitate their appraisal. Although positive, this exchange again demonstrated misunderstanding of CVD event-free survival age as expected survival age, this time on the part of the practitioner. The visual impact of demonstrating how CVD risk can be reduced through risk factor modification (e.g. cholesterol, smoking status) appeared to aid patient understanding and realistic threat appraisal. There were fewer examples of active engagement during discussion of the CVD risk score during QRISK2 consultations, which may be because of the inability to show risk factor modification when using the calculator.

Coping appraisal

Coping appraisal focuses on the coping responses available to the individual to deal with a health threat (i.e. evaluation of ways to reduce CVD risk). This included patients’ perceptions of self-efficacy to engage in adaptive coping, practitioners’ promotion of self-efficacy through individualisation, the perceived response efficacy of adaptive coping and the response cost of adaptive coping (see Figure 2). References to coping appraisal were more common in JBS3 (n = 60, 94%) than QRISK2 health checks (n = 55, 86%). The communication of risk in JBS3 consultations was not observed in the same way as in those using QRISK2; most such consultations focused on facilitators of adaptive coping (i.e. risk-reducing changes that patients could make):

Erm and then this gives you your healthy years outlook [event-free survival age]. So based on your current lifestyle your risk of a heart attack or a stroke in the next 10 years is coming out at 2.4%. We aim for people’s risk to be below 10% so that’s . . .

Yeah.

. . . absolutely fine and on average you expected to survive to an age of 84 without a heart attack or stroke, so brilliant. So as I say, your blood pressure pretty good as it is, you not going get that much lower.

No.

Diet-wise would you say you got a pretty good diet do you know the sorts of . . .

We sort of grow our own vegetables and fruit and stuff like that . . .

Yeah.

. . . so, erm, I mean we eat reasonably healthy.

JBS3, 7_044, female, 54

Following communication of the risk score, the practitioner moved on to ways that the patient could maintain a low risk through dietary behaviours. This suggested that, although practitioners (from both groups) spent little time talking about the CVD risk score, the additional risk information in JBS3 may have helped to facilitate more risk factor discussion between the patient and the practitioner (than if using QRISK2).

Discussions around response costs for adaptive coping (i.e. perceived costs associated with a recommended behaviour) related to the use of statins or blood pressure medication, and were only observed in seven JBS3 consultations (11%) and none of the QRISK2 consultations. Data are limited, but may offer some evidence to suggest that JBS3 was more likely than QRISK2 to promote discussion around adaptive coping:

Obviously we’ve tried them, and they haven’t agreed with you.

I tried the [medication name] statin.

Yeah, and there are other statins we can discuss and obviously benefits of those they can reduce your cholesterol obviously and we can reduce your risk of cardiovascular disease so it might be worth having a think about and if you want to just discuss that further or a different type of statin . . .

All they did was it affected my reflux and it made the reflux worse.

Yeah.

So.

Yeah.

I was on that and an aspirin – I did the aspirin first and then . . .

Yeah, but it was affecting you. I mean it might be worth a having another . . . err you know a think about whether you wanted to erm take that, because obviously it would lower your cholesterol, obviously add to a healthier heart erm and reduce that risk of cardiovascular disease, but then obviously we’ll not gonna push that onto you, err it is something you can talk to myself, one of the doctor’s once you have had time to think erm and they can advise or [name] the prescribing nurse, because they can prescribe, you know talk about you know what’s best, which statin would be best, and not all statins agree with everybody but there might be one out there that actually has a better erm compatibility with yourself OK?

Yeah.

How do you feel about what I have told you today?

I would consider it.

JBS3, 8_177, male, 71

Here, the patient’s prior engagement with statins as a response cost was discussed, leading to a re-evaluation of the medical intervention by the patient. However, rather than addressing the patient’s concern regarding their previous experience of statins, the practitioner appeared to interrupt them to repeat the benefits of statins. There were examples where the perceived cost of taking statins provided motivation to adopt risk-reducing lifestyle behaviours:

But well done!

I am pleased about that, yes.

That’s really good, no I am very pleased with you because that’s really good. And where you were at 10% just before, it is now 5%, so you have halved the risk in that time. So that’s really good. So it shows it can be done.

Yeah, yeah, and that’s what I would rather do than taking tablets.

Of course.

I would rather think, no I know what’s wrong, I will deal with it in time.

QRISK2+, 12_055, female, 64

However, opportunities to discuss facilitators of adaptive coping were sometimes missed by practitioners who communicated risk using QRISK2:

I look at your [total : HDL cholesterol] ratio and your ratio is good. But just to keep a little eye on it, maybe they will test it again in a year’s time. You probably won’t be due this Health Check, because your risk is only 3%, which is low. It will increase as you age, so your Health Check wouldn’t be due again for 5 years, but you could probably have your cholesterol done in about a year, with you know normal bloods taken out of your arm. Erm your heart age, because you got such results in the green, your heart age has come up less than your actual age, but that’s with the 2 years added on from being an ex-smoker.

So is it possible that I could get that even lower? If my cholesterol came down a lot.

Well we will have a look now, I will play about with it. So if you had never smoked at all, your heart age would be 45. If you were still smoking, it could be 51. So being an ex-smoker tends to add 2 years, so with your cholesterol, it could be brought down to 46.

Massively, yeah.

QRISK2+, 2_077, male, 48

The patient above attempted to understand how their risk could be reduced. The practitioner did not engage with this to encourage the risk-reducing behaviour or discuss ways to reduce cholesterol. Rather, they proceeded to talk about the impact of previous smoking status (which is unmodifiable) on CVD risk. Although references to coping appraisal were more common in JBS3 consultations, practitioners in both groups appeared to focus on the consultation process more than the patient.

Maladaptive coping

Maladaptive coping is classified as the patient negatively engaging in risk management discussion and being dismissive of suggestions (e.g. the patient believes that they have a sufficiently healthy lifestyle and dismisses discussion about change). As noted, patient responses to risk information were often limited to a single word. Where context allowed, apparent lack of engagement and minimal verbal responses from patients were also interpreted as a maladaptive coping response if the risk information communicated by the practitioner did not provoke a response by the patient. Risk management interventions were classified as medical [appointment referral, medication (e.g. blood pressure or cholesterol lowering), re-testing and managing blood pressure or retesting cholesterol] or lifestyle (weight reduction and management, increasing physical activity, diet modification, smoking cessation and reducing alcohol consumption). Maladaptive coping was identified in 49 (77%) QRISK2 consultations (coded 139 times; average 2.8 times per consultation), compared with 40 (62.5%) JBS3 consultations (coded 110 times; average 2.8 times per consultation). The practitioner in the exchange below briefly communicates QRISK2 before moving on to heart age (using Informatica):

Yeah this is the screening I was telling you about. So, your risk is 9%.

Right.

Which is your key risk for you over the heart disease and diabetes and stroke risk.

And heart disease.

As you, as you age your risk does seem to increase, erm any results that you’ve got in the red tend to push up your heart age slightly.

Aha.

So if we can get the results out of the red and back into the green, that can reduce that one down.

Right, OK.

So, for example, being an ex-smoker actually puts 2 years onto your heart age there.

Yeah.

So would be its 66 and it would be 66 if you never smoked at all.

Right.

Erm, if you were still smoking it would be 73.

Oh my gosh!

Your heart age has come up as 71 – you are 66. Any results you have got in the red do tend to increase your heart age. It is just that one cholesterol one that was in the red.

QRISK2+, 2_001, female, 66

Sometimes, maladaptive responses to the 10-year percentage risk score could be shifted towards a more positive response through communication of heart age. The brief exchange prior to the communication of heart age (above) may have also suggested that the practitioner was less confident discussing QRISK2, a recurrent observation (see Chapter 6). If practitioners cannot clearly explain the meaning of a patient’s percentage risk score to confer understanding, subsequent discussion/actions regarding risk management may be undermined, and the link between CVD risk and lifestyle may be unclear.

Minimal engagement following communication of the risk score was also identified in JBS3 consultations:

OK. And your blood pressure being under 82 but that’s fine everything is OK with that. Now, on average what they’re saying is that your risk of a heart attack or stroke in the next 10 years is 15%, again, that is down to the fact that you smoke.

Hm.

OK.

[Sigh.]

And to expect to survive till the age of 78 without a heart attack or a stroke OK. And if we have a look at the next, this one, just reiterates its this, but if I changed it to . . . say if you didn’t smoke OK and we went to the next your heart age would then become equal with your age.

Hm hm.

And your risks in . . . of a heart attack or stroke in the next 10 years comes down to 9.6% and your actual survival to the age of 83 without a heart attack or a stroke OK and that reiterates it in that as well.

Hm hm.

OK so that’s the difference.

Hm hm.

Hm hm.

OK. Erm.

[Cough.]

So it gives you food for thought.

Hmm hmm. You haven’t told me anything I didn’t already know.

JBS3, 1_181, male, 65

The patient’s dismissal of discussion around smoking cessation may reflect the influence of environmental and social factors on their smoking behaviour (and related intentions), preventing a meaningful exchange. The practitioner did not encourage the patient to quit smoking or explore previous experience of attempted smoking cessation and, therefore, was unlikely to promote intentions to change behaviour. There is an added pressure of time during health check consultations. Appointment times allocated to health checks ranged from 15 to 30 minutes (see Table 4), during which time practitioners need to assess CVD risk, communicate the risk score(s) and discuss risk management. Adherence to the process of completing the health check may result in patients being passive recipients of information. As shown above (and throughout), the practitioner delivered the information presented on the screen without asking questions to check understanding or provide context. This prompted little response from the patient, which may be indicative of deference to the practitioner’s health knowledge, which would indicate a power imbalance.

Negative engagement in discussion of risk factor management was also evident following the suggestion of statin use:

What we do tend to say if you risk is above 10%, obviously I don’t know whether the doctors have ever discussed a statin with you?

I don’t see the point, I mean if I am going to live to 83, I am quite happy to live to 83.

So it’s just about being aware that we know that taking a statin can help reduce your overall risk, so it’s one that sort of we usually advise that . . .

If we do this next time and I don’t know, it was ’04 [last cholesterol check], and we are now in 2018, so what does that mean, it could be another 12 to 14 years [for the next health check]?

Well I do normally try and do these every 5 years, so yeah.

So yes, if it is hugely worse.

Yeah.

. . . in 5 years, I will consider it.

JBS3, 7_012, male, 70

Again, the patient misinterpreted event-free survival age and suggested that their risk was not severe enough to consider medical intervention in the short term, only if it was ‘hugely worse’ at the next health check. This was another example of a missed opportunity for the practitioner to question the patient’s understanding of their risk and potential false reassurance provided by the 10-year percentage risk score (which was 15%).

Adaptive coping

Adaptive coping focused on the patient’s positive engagement with discussion of interventions to manage CVD risk, apparently listening and engaging in the consultation and accepting what is being said/suggested. Adaptive coping was identified in 58 QRISK2 (91%) consultations (coded 310 times; average 5 times per consultation) and 55 JBS3 (86%) consultations (coded 328 times; average 6 times per consultation). The frequency of occurrences overall and per consultation was similar between the two groups for adaptive coping in medical interventions [39 QRISK2 (61%) consultations and coded 116 times (average 2.9 per consultation); 42 JBS3 (66%) consultations and coded 142 times (average 3.4 per consultation)] and lifestyle change [11 QRISK2 (17%) consultations and coded 15 times (average 1.4 per consultation); 20 JBS3 (31%) consultations and coded 32 times (average 1.6 per consultation)]. A number of patients showed intentions to change their behaviour as a result of their CVD risk:

So your ratio is 3.5. So this is the screening I was telling you about. So your risk is 3%. That will increase as you age.

Yeah.

And obviously if we can, perhaps with your smoking, it has pushed your heart age up to 48, and your age is 41. Because that is the only result you have got in the red. Because all your other results are really good, they are in the green.

They are really good, so I need to . . .

Yeah, so if you had never smoked at all, your heart age would be aged 40.

I think I need to do something about that don’t I?

QRISK2+, 2_122, male, 41

This was another example of how heart age changed the way that the patient responded to the information presented. Although a positive response was evident, the practitioner did not allow much time before moving on. By giving the patient time to confirm his understanding, the practitioner gave the patient the opportunity to increase his confidence in engaging in coping behaviours. Another example of positive engagement during the discussion of risk was identified from another practice:

OK that’s good. Err let’s see your key risk.

If I know what weight so I can just try to change my life.

Yeah, yeah, it would be good if you can cut down and, and lose a bit of the weight err what was it 13.8. So it’s only a little higher, it should ideally be below 10% is what we want so 13.8 is a bit high but it is because of, because of your weight. OK, you don’t smoke, you don’t drink alcohol, so that’ all good, but your waist is a bit big as well.

Yeah

Your waist is erm it’s 112, let’s have a look.

Around my tummy around here.

Yeah let’s have a look. So your waist is 44 inches.

And that’s this bit here.

QRISK2, 3_259, male, 57

The patient above engaged with the information presented about their risk and suggested a need for weight management, somewhat reinforced by the practitioner. However, the interaction was disjointed, possibly because the practitioner needed to complete all other elements of the health check and created a barrier for adaptive coping through not truly attending to what the patient was saying. Although scarce, a successful strategy for supporting adaptive coping used by one practitioner was to ask the patient to reflect on the risk information that they had received, prompting consideration of action needed:

So average survival free of heart attack or stroke is 84.1 years OK? So how do you feel about that?

Oh I will make more of an effort to lose some weight.

JBS3, 1_154, female, 70

The approach adopted by the practitioner encouraged the patient to express their immediate reaction to their CVD risk, which gave the patient an opportunity to evaluate their action and show intention to change their behaviour. This was a rare example of PMT in action, showing connection between the risk information and the patient’s intention to change her behaviour, helping to redress the power imbalance evident in most consultations across both groups. It also demonstrated the significant role that the practitioner plays in ensuring that risk communication is delivered effectively regardless of the risk calculator.

Not concerned by cardiovascular disease risk

For most patients, perceptions were largely focused on a lack of concern, regardless of the way in which CVD risk information was delivered. One patient described how being told her 10-year risk ‘didn’t alarm me . . . I thought 6% wasn’t too bad at all really . . . I thought that was a pretty good, average’ (QRISK2+, 12_131, female, 57). Other patients described how they were ‘quite relaxed about that really’ (JBS3, 7_117, male, 49) after being told their heart age. A small number of patients spoke of their worries after finding out their CVD risk information. For example, on learning that his heart age was 6 years older than his chronological age, one patient explained: ‘that shocked me. No, it concerned me’ (JBS3, 7_105, male, 59).

Receiving CVD risk results also appeared to reassure patients, regardless of the way in which risk was presented. For example, one patient talked about the relief of being given a 10-year risk that she perceived to be low:

So saying to be 6.0 – well under 10, so yeah, yeah, yeah, it is reassuring. Which when she said it was under 10 that’s why I forgot the number, but I was just relieved.

QRISK2, 3_125, female, 56

This was also observed for heart age: ‘I was quite pleased because it came out with my heart age at being my age, so I thought phew [laughter]’ (JBS3, 4_263, female, 61).

It is possible that patients’ lack of concern about their level of CVD risk was related to a widespread perception in the sample of a low level of personal CVD risk (as 70% of the 128 were classified as low risk). However, this perception that personal CVD risk was ‘small’, ‘low’ or ‘quite low’ or ‘wasn’t that high’ was observed among patients with different levels of risk. For example, one patient whose 10-year risk was estimated at just 1% (which was heavily influenced by his relatively young age):

. . . so do you remember what you were told by the practitioner about your level of developing cardiovascular disease, or your risk?

I was very low 1% something like that.

QRISK2+, 12_064, male, 40

And another patient who had a 10-year risk just under the 10% threshold at which intervention might be recommended:

OK so she said that your risk of heart attack or stroke was 8.1%.

Yes so quite low.

JBS3, 8_162, female, 67

However, a patient considered at high risk (≥ 20%) described how they felt when informed of their 10-year risk score, indicating that even those with a high risk of developing CVD did not perceive or understand the severity of this information:

Err, well it didn’t really err, how can I put it, well err, err, well it . . . I thought about it but, 25% that’s yeah I thought well err most people walking round now are at, what is it 25%? . . . You know erm, I, I thought the odds were pretty good myself to be honest with you [laughter].

JBS3, 5_132, male, 74

Patients who were told that they had a low risk of CVD were understandably reassured (see Positive impact of Health Checks). However, there is evidence of false reassurance for those with a high risk of developing CVD who do not recognise the risk severity from the information provided or the related discussion. This can be linked to a lack of understanding (see Mixed levels of understanding), particularly surrounding the notion of percentage risk.

In addition, several patients commented on CVD risk being an estimate or a prediction, and this appeared to influence the perceived severity of threat to their health:

So getting your actual cholesterol result, so rather than having kind of your 10-year risk or your heart age, or your survival age, it would be your cholesterol . . . ?

Yeah very much so because that is the now, you know all those other things are projections.

OK, that’s interesting.

Like a forecast as such. This is the forecast, but that forecast is only determined by the here and now. If there is a lifestyle change tomorrow that information that was given yesterday is null and void to a degree.

JBS3, 4_080, male, 57

This appeared quite prominent in relation to event-free survival age, which this patient considered less credible than heart age:

Cause, yes that [lifetime risk] . . . I think I’d take that as a pinch of salt, to be honest . . . You can’t predict that . . . You know, that’s pretty ridiculous . . . I can listen to myself and think err . . . I’ve got a 51-year heart or you know . . . they’re gonna know that . . . But to predict how long I’m gonna, live really that’s errr . . . science fiction ain’t it really?

QRISK2, 9_087, male, 48

Patients viewing their CVD risk results as a prediction or estimate might facilitate beliefs that CVD risk can be reduced, in turn, enhancing self-efficacy towards positive behaviour change (see Facilitating change). However, for other patients this could be detrimental to the appraisal of their risk and lead to a fatalistic view, rather than considering their CVD risk as being in their control.

I would be more worried if cardiovascular disease risk was higher

Patients dismissed their CVD risk if they perceived it to be low (regardless of risk level). For example, one woman, whose heart age was 5 years older than her chronological age, said:

No I just thinking well it doesn’t bother me ’cos, 5 years OK 5 years, 5 years, but it’s not you know about 20 years, 20 years be thinking ‘Oh my God’, you know . . . horrified, but I think 5 years is not too bad and there’s things that I can work on to try and reduce that really so I wasn’t horrified by it, no.

QRISK2+, 12_131, female, 57

This suggests that this patient, because her heart age was similar to their actual age, regarded the threat as not severe, and perceived that she had self-efficacy over the changes needed to reduce it. Similar thinking was observed in relation to 10-year risk. In the example below, the practitioner had suggested that a 10-year risk of 1.6% was slightly raised:

You tend to look at it and then tend to be kind of, I don’t know, kind of brush it away . . . although she said it was quite high, I think because I think it is such a low percent, it’s kind of not at the front of my mind and I am not concentrating on it.

QRISK2, 9_083, female, 39

Therefore, it appeared that patients’ perceptions of the severity of their CVD risk and vulnerability to it are important in fostering intentions towards risk-reducing behaviour (in line with PMT).

It was common for patients to explain that, if their CVD risk had been higher (whether this was higher 10-year risk, older heart age, or older event-free survival age), they would have been more concerned, and this would have been more likely to prompt change. For example, one patient believed that if he had been given a 10-year risk that was double his current risk (15%), it would have prompted more concern and increased steps to lower it:

You know, I mean if you’d said to me it was 30% . . .

Hm.

. . . a risk . . .

Yeah.

. . . obviously I’d be concerned about it.

Yeah, cos it would have been high . . .

High, it’d be double really, and higher, and obviously I’d have to take erm . . . or listened more, or look at taking steps, to . . .

Hm.

. . . to rectify it or do something about it.

JBS3, 11_028, male, 58

Similarly, patients frequently noted, that if their heart age had been a lot higher than their actual age, it would have been perceived as a more severe threat, prompting behavioural action (in this example, heart age was 51 years vs. chronological age of 55 years):

If it [heart age] had of been a lot higher then I would have thought ‘oh I will definitely go and get a bike and I will go swimming as well’ [laughs].

QRISK2+, 12_083, female, 55

This suggests that an increased understanding surrounding CVD risk level is required for patients to appraise how their individual risk may affect them. If patients do not perceive their CVD risk as a threat, then they are perhaps less likely to make positive behaviour changes and follow practitioner recommendations.

Feedback on cardiovascular disease risk was unexpected

Throughout patient accounts, it became apparent that there were mismatches between expectation and reality of CVD risk. Some patients expected to be told that their overall CVD risk was higher than it was based on known risk factors:

I thought they [10-year risk] were going to be higher, because of saying with my blood pressure and the cholesterol, and so I thought ‘oh I am in trouble here’, yeah and I thought . . . I really did think it was going to be high.

QRISK2, 3_125, female, 56

This was particularly true for 10-year risk. One patient, for example, explains that he was expecting the score to be higher than it was, despite it placing him in a medium–high risk category: ‘I think it was a case of [name] understood that I understood the 17% . . . As I say, I expected it to be higher but thankfully it wasn’t’ (QRISK2 +, 2_084, male, 64).

Many patients had expectations of their heart age that did not match the estimates from the health check and expected a low heart age based on a perception of being in relatively good health:

That is a bit of a surprise really for that, because I don’t feel that you know, and I don’t know I still feel quite energetic and still play you know the sports I do, I am never tired, or feeling like I can’t go on any . . . you know, in fact I do the complete opposite. Yeah, if I feel oh right I am feeling a little bit – like if I am playing badminton for example, I can always push and push and push yeah, to you know go a little bit harder and that so you know I was a bit shocked at that to tell the truth. Not shocked, as in disappointed, but I thought you know . . .

You thought it would be lower rather than higher.

Lower than that yeah, because I don’t feel like I physically can’t do anything.

JBS3, 4_311, male, 57

This suggests that patient awareness or perceptions of their own health status are important. If a patient perceives themselves to be physically fit, they may underestimate their personal level of CVD risk, which might be elevated. This could stem from a lack of understanding of the CVD risk score itself and the information on which it is based, which is important if patients are to understand how lifestyle changes can be used to manage and reduce their risk.

Theme summary

It is likely to be insufficient to provide patients with their 10-year risk, or a heart age, in the absence of supporting information. Putting these results into context or presenting them in a different way to ensure that the severity of patient’s risk is understood is important to engage patients in discussion to address misunderstandings or misperceptions about what the score means, why it might be higher/lower than expected and the implications for their health.

Understanding versus lack of understanding

Throughout patient accounts, there was a distinct division between patients who seemed to understand the CVD risk information provided and those who did not. This was particularly true for 10-year risk and event-free survival age and, to a lesser extent, heart age. Some patients demonstrated a level of understanding of percentage 10-year risk (albeit with some confusion):

Yeah, OK, and do you know erm . . . kind of do you understand what [practitioner] meant when she gave you that percentage score?

Yeah that you’ve got 6% risk of, of getting heart disease in life really yeah, out of 100 people you know 1 in 6.

QRISK2+, 12_131, female, 57

For other patients, gaps in knowledge and understanding of the score were observed. Some could not recall what their score meant. For example, one patient struggled to ‘remember what the significance of 9.5 was’ (JBS3, 4_233, female, 69), when talking about her 10-year risk. Other patients found it difficult to put the percentage score into context in terms of a ‘good’ and ‘bad’ outcome. This was evident across all risk categories (low, medium and high). For example, a patient with a relatively low percentage demonstrated confusion about her 3% 10-year risk:

It depends, like I said, what it actually means. If 3% is not good, then obviously that information wasn’t useful to me, because I have gone away thinking that’s fantastic. Even though I am a bit sceptical, erm but if you are saying 3% actually is good, then yeah it was accurate what she was saying to me. But it almost seems as if, because she did say like slightly higher, I think if 3% is a high risk of you know heart disease, then 3% is not good, because 3% makes you think it’s good, because 3 out of 100 is good.

QRISK2, 9_295, female, 51

This lack of clarity was also observed where risk was higher:

I can’t quite understand what like 25% is, what’s, what’s good and what’s bad with 25%?

JBS3, 5_132, male, 74

This suggests that, regardless of the actual score, patients may find it difficult to understand the implications and severity of a 10-year risk. Despite misunderstanding, some described their experience of being told 10-year risk positively: ‘I thought that bit of it was very good indeed, obviously I didn’t recall it, but I did think that was good how that was explained and showed’ (JBS3, 4_080, male, 57). Yet the potential benefit of such positive perceptions is likely to be undermined by the lack of recall or understanding of CVD risk information.

Lack of understanding was also common for JBS3 event-free survival age. Instead of viewing lifetime risk as the expected age of survival free from CVD events, some patients interpreted this as predicted age of survival. This appeared to be the case not only for patients who were given this lifetime risk estimate during the health check, but also for QRISK2 patients when asked their opinion on this alternative way of presenting CVD risk:

So this would be your age to expect to live to without a heart attack, or a stroke . . .

Hm.

. . . and then after that age, your risk would increase significantly.

Hm hm.

So it is not necessarily kind of when you are dying.

When you are expected to live to? Hm. Hm. Well you see now, what’s interesting about that, because now that I have read that, I have realised what it says . . .

Hm.

. . . but initially, what I thought, and I think that’s probably what a lot of people would think. Initially you read it quickly and it looks like it is saying that we expect you to live until this age.

OK, so life expectancy.

And when you see that, then you cut off from what that’s saying . . .

Right OK.

. . . so I think that’s . . . because that is exactly what I have just done.

Yeah

I have just dismissed that bit there, and only read to there, on average expect to survive to 73.

QRISK2, 9_295, female, 51

This was also the case for one patient on being informed of his event-free survival age in the health check, which was relatively close to his actual age. The confusion understandably caused a great deal of concern (until this was identified and resolved):

So did you, did it make you feel sort of what emotions was it sparking was it . . . ?

Erm deflated me a little bit.

Yeah.

Erm, because I’ve never, ever thought about it and err it was my fault, that, that was the bit I felt where it let itself down a little bit.

Yeah.

Erm, if she’d have said yeah, yeah we can guarantee, well not guarantee, but we can see you’ve got reasonable amount of having good health until you’re 80 you know 82.

Yeah.

And then things might deteriorate a little bit.

Yeah.

That would have been perfect I’d have . . .

Yeah.

But then as I say 82.7 mentioned and I thought to myself and I never, that was it, the end of the interview really, I think.

Yeah.

As I thought, then, then it gets in your, gets in your head then.

JBS3, 5_132, male, 74

Understanding of heart age was also mixed, but patients seemed to have a better understanding of this than on other risk scores, and the comparison with chronological age allowed patients to quickly appraise their individual risk as ‘good’ or ‘bad’:

Where it says ‘your heart age is’ and gives you a heart age, straight away you know whether that is good, or bad, because if your [chronological] heart age is lower than the reading [heart age estimate], then you know straight away that is not so good. Whereas if it is higher you know. So I think that one is a bit more clearer.

QRISK2, 9_295, female, 51

However, some patients found the concept of heart age challenging, and failed to understand the context or implications of a heart age different to their actual age:

. . . again it, it’s there was this lack of information you know, what do these numbers mean? . . . I can’t remember ever being told what the numbers actually mean . . . You know what is the, you’ve got a heart of a 72-year-old, or 73-year-old and you’re, you’re 62. What does that actually mean?

JBS3, 4_143, male, 62

Overall, lack of understanding was evident across all methods, especially QRISK2 and event-free survival age. This highlights the need for practitioners to check patient understanding during the consultation and, where necessary, tailor the information to the patient.

Importance of context

One way to improve understanding of CVD risk relates to the context in which the different risks are presented [i.e. the extent to which practitioners demonstrate the relevance to patients (its severity or their vulnerability to it)]. This was highlighted by some patients, who believed that the CVD risk information lacked context, making it difficult to understand or appreciate the severity:

. . . walked off and probably thought afterwards, I could have done with actually understanding is the heart age kind of thing, normal . . .

Hm.

. . . I was kind of thinking come out probably late 30s mid to late 30s something like that . . .

Yeah.

. . . but that’s without knowing anything about it . . .

Hm and was it useful to be told a heart age?

It’ it needs more context think . . .

Hm.

. . . on reflection as I walked away I don’t, didn’t really ask what that meant, it’s just stuck in me that it’s like hm . . . it’s not younger than me which, I imagined it should be.

QRISK2+, 12_064, male, 40

This idea was also discussed in relation to 10-year risk, in which additional context helped the patient to understand (in this example, as noted in Chapter 7, this quotation also reflected understanding conferred through a follow-up telephone call in which a specialised nurse practitioner provided further explanation):

Erm, and did you understand what she meant when she said a risk of 17%?

Erm, yes, I think . . . We did do kind of an analogy of you know like 100 people in a room – 17 can be expected . . . but you don’t know which of the 17 . . . which I thought was quite a nice graphic analogy to give you a visual picture.

Yes. Was that something that she showed you on a screen or was it . . . ?

No we just discussed it.

It was conversation?

Conversation, yes.

Brilliant. So that kind of helped you to put it into context a little bit about sort of the 17%?

Yes.

QRISK2+, 2_084, male, 64

Patients who were not given as much detail on the 10-year risk to put it into context appeared to have less understanding of the implications:

I think with the percentage unless you have been given the range it should be in for your age and for your, you know, capabilities, then it’s kind of a mismatch of information. I don’t know which to kind of . . . they are saying it’s high, but I think it’s quite low, but I don’t know what high is because I haven’t been given anything to compare it against.

QRISK2, 9_083, female, 39

Therefore, the context in which CVD risk information is communicated is important to aid understanding. Providing patients with more explanation on how to position their score relative to the population might facilitate understanding of the severity of their personal risk. One method of doing this is to provide some more information (e.g. use of analogies, relative risk) during the CVD risk discussion. Another method relates to the presentation of information, particularly given the positive perceptions of the ability to manipulate a patient’s CVD risk using JBS3 (see Importance of presentation).

Perceived changeability of cardiovascular disease risk

In addition to varied understanding of CVD risk in general, there were apparent differences in understanding of whether or not it could be changed by lifestyle factors. A small number of patients thought that it was fixed:

Yeah, you know, so err and having known the fact that my dad had a heart attack has increased my score . . . and she said herself on the video you can’t do anything about that. That’s just there you are there’s 10 points on the score.

QRISK2, 6_044, male, 61

In contrast, other patients talked about CVD risk as changeable: ‘Yeah, yeah, I remember thinking so I could do something about it’ (JBS3, 4_233, female, 69), or even acknowledged their own role in lowering the score: ‘But obviously it could come down, it could be better I guess if I change . . .’ (QRISK2 +, 2_001, female, 66). There was also an awareness that CVD risk could also increase over time, an important point for patients to understand from a prevention perspective:

. . . so, but again she did say that it wouldn’t stay the same it would change, so I do know that I, I do need to start looking after me self . . . a bit better.

QRISK2, 10_539, male, 51

The idea of CVD risk being a prediction or a ‘lottery’ was also evident in relation to discussion about the patient’s ability to make changes. This highlights an important separation between patient’s lifestyle or risk-reducing behaviours and what they see as a prediction of future disease risk, which might lack credibility:

Yeah, so was that information in particular helpful, or unhelpful, the 9%?

Oh!

To you?

To me it didn’t mean anything, because to me you know I can change my lifestyle and all that sort of thing, but at the end of the day it is a bit of a sort of like lottery really isn’t it? [Laughs].

Yeah, yeah, I think it is isn’t it really?

I would be quite happy if I had 9% chance of winning the lottery! [Laughing].

QRISK2+, 2_001, female, 66

This could be positive because of associations with change and adaptability, but negative in terms of fatalistic attitudes and not recognising CVD risk as the context or reason for making health-promoting changes.

Finally, despite the predominant belief that CVD risk (or risk factors) was modifiable, some patients lacked understanding about how to implement positive changes, for example to reduce cholesterol:

So it then led me to think ‘Well, where am I with it, can I do something about it?’. I don’t think I can do much more with my standard of living and exercise to be honest. So I think ‘well if there is an issue it might be genetic’, but I am now in the process of thinking ‘well you know what let’s get this tested again and see where we are’.

JBS3, 4_080, male, 57

Or in terms of lowering CVD risk itself:

But that part wasn’t . . . I am not clear about, erm I think at that point you are just . . . I am not sure actually, I am not sure how . . . right now so it’s 4.8% over the next 10 years, but I don’t know how you would lessen that.

JBS3, 4_263, female, 61

This makes clear the important link between clarity and understanding of CVD risk information, and patients’ perceptions (and subsequent intentions) regarding their ability to reduce risk through lifestyle changes.

Theme summary

Patients often do not understand CVD risk information (particularly 10-year risk and event-free survival age) and the implications of the scores. They are often unable to determine if their CVD risk is higher than that of others of the same age and, therefore, they may not be sufficiently motivated to change their behaviour. Patients would benefit from practitioners providing further information to make clear the relevance of their CVD risk. Visual presentation and risk score manipulation offer tools that can accommodate the range of patients, and also to make the important link between lifestyle change and overall CVD risk.

Facilitating change (risk management)

The extent to which patients reported to have altered their intentions or behaviours as result of the health check varied. For some, the health check prompted them to contemplate behaviour change: ‘did make me think a bit more . . .’ (JBS3, 7_136, female, 53); ‘put it [CVD risk] on my mind’ (QRISK2 +, 12_083, female, 55). For some, it resulted in a positive change in attitude towards health: ‘Overall . . . I came out with the attitude I’ve got to be more health focused’ (JBS3, 7_105, male, 59). For others, it appeared to increase their awareness of lifestyle aspects that might need to be changed, and promoted contemplation to engage with this:

Erm . . . no, the good thing was like you know get your cholesterol checked I had that, had previously checked it was still little bit high, but I’d gone up again erm . . . so it’s made me think about changing my diet and . . . and erm . . . tweaking it a little bit here and there . . .

QRISK2+, 12_131, female, 57

Consequently, a number of patients had not made any changes based on the recommendations provided, but were considering change. Others had made positive changes to lifestyle and behaviour. Overall, the health check seemed to facilitate patients to ‘work on that [behaviour] a bit harder’ (JBS3, 4_311, male, 57). Positive changes in several health behaviours were seen, mostly those most discussed by practitioners (see Chapter 4), such as physical activity, healthy eating and blood pressure monitoring.

Many patients talked about increasing their levels of physical activity following the health check. Typical examples involved increasing walking ‘I’ve started to go out a few minutes earlier and walking up the hill [laughs] . . . it’s just a few minutes a day . . . might make a difference’ (QRISK2 +, 12_055, female, 65), or other home-based exercise:

I have, I mean like I have got my step machine . . . I have got a step machine that was in the garage, I have brought that inside, so I can do that while I am watching television.

How brilliant!

So like it does get me out of breath.

QRISK2, 9_202, female, 46

There were also several instances of dietary change. This could involve reducing intake of certain foods:

I’ve cut sugar down no end.

JBS3, 5_132, Male, 74

Brilliant. So has being told your level of cardiovascular risk, so either the percentage, the heart age, the survival age, has that affected you in any way at all, sort of you know how you are approaching kind of lifestyle and things like that?

Well, certainly, over diet really, over the cheese, because I was . . . she actually suggested cutting it down to a couple of times a week, which I have done already, so that definitely has had an impact.

JBS3, 4_263, female, 61

Others had substituted foods:

Do you mind me asking what change you made?

. . . erm . . . cut out some dairy . . . so I stopped using milk in porridge, and things like that and using water . . . got coconut milk as well, erm . . . and cut out bread . . . and erm . . . I’ve actually got a bit more, I’m not gonna go vegetarian, but I’ve gone a bit more vegetarian . . . I’m eating probably slightly less, but I’ve cut the crap out . . .

QRISK2+, 12_064, male, 40

When blood pressure was highlighted as a particular issue during the health check, some participants engaged in positive behaviour change related to further monitoring, either at the general practice or at home:

But the fact that I went out and ordered a small blood pressure monitor and one thing and another erm and presented my wife with those sheets and whatever. I’ve never done anything like that before.

JBS3, 7_105, male, 59

Despite the positive contemplation and actual behaviour change taking place, the changes made by patients were small: ‘I have not made any massive changes’ (JBS3, 7_105, male, 59). This could be explained by a number of the preceding themes, such as feeling reassured by their CVD risk, thinking CVD risk was low, considering behaviour change only if CVD risk is perceived to be high, believing that they were already doing the right thing, and perceived ability to change. For example, one patient talked about taking a ‘small step’ as a result of attending the health check:

And then erm, I forgot about the heart age, when she said about the heart age yeah. I just went back with the erm the life expectancy age. Erm, yeah so it has it, it gave me the momentum to do something, I know it’s only a small step but it’s a step.

JBS3, 5_185, male, 47

Overall, evidence related to contemplation of change and also the relatively modest changes made by many patients suggests that something else, in addition to what was included in the health check, may be needed to facilitate intentions for, and implementation of, more consistent and substantive behaviour change.

Impact of heart age

When asked about the CVD risk information they could remember, numerous patients could recall their heart age, more so than 10-year or event-free survival age. As a result, heart age seemed to stand out to patients as being more impactful:

I think, I think the thing that registered with me and that sort of really grabbed my attention, and I know I said this earlier, all I could see was that 65 on that screen.

Yeah.

And I couldn’t . . .

Yeah so the heart age really kind of . . .

. . . that was wallop.

Yeah.

Yeah. That heart age and I think perhaps I missed some of the, shall we say, the finer detail because I was focused on that.

Yeah.

I could see that 65 and I was thinking, bloody hell, I ain’t 65!

JBS3, 7_105, male, 59

Clearly, it would be problematic if patients, in trying to process heart age information, ‘switch off’ and are unable to engage with subsequent recommendations from the practitioner. However, given the lack of retention of 10-year risk (and confusion over event-free survival age), understanding the implications and retaining one of the metrics can be interpreted as a positive, as it could provide the context for a meaningful discussion of risk management.

The concept of heart age was also well received by patients in the QRISK2 group who were interviewed, who were shown alternative ways of presenting CVD risk information during their VSR interview:

Err I think that one has got the most positive impact really. Even if it wasn’t positive from point of view of your heart age, if it was higher than what your [chronological] heart age is. I think you would still straight away understand that more and know what you have got to do.

QRISK2, 9_295, female, 51

Heart age was perceived to increase understanding of CVD risk among patients in both the JBS3 group and the QRISK2 group:

But when it was very clearly written, you know so the model brings it up as your heart age, given the information that it has got, is 61 years and I thought ‘well that is so clear and understandable’, so I found that very helpful, because if it had been much higher, then that would have also been very understandable.

JBS3, 4_263, female, 61

In addition, heart age appeared more relatable and, perhaps, more credible. As reported earlier, one patient described event-free survival age as ‘science fiction’, but could believe the idea of heart age:

I can, I can listen to myself and think err . . . I’ve got a 51 year heart, or you know . . . they’re gonna know that . . .

QRISK2, 9_087, male, 48

This links in with the earlier idea that CVD risk is only a prediction. If heart age is perceived as more relatable and credible, patients may have an increased understanding of how their risk of developing CVD, be able to appraise its severity and their vulnerability to it and, therefore, be more likely to make behaviour changes to mitigate the risk. This was illustrated in the patients accounts around heart age prompting change as ‘it is that swift kick to say: “get out there and do something” ’ (QRISK2 +, 2_084, male, 64):

Don’t get me wrong, it was enough to make me say to my wife ‘I’ve got to do something about what I eat, and I’ve got to get more exercise and I’ve got to do this, and I’ve got to get that’, and I ordered a blood pressure monitor and all the other things. I even checked that the bathroom scales worked properly.

JBS3, 7_105, male, 59

The other methods of presenting CVD risk information (10-year risk, event-free survival age) were also commented on by patients in terms of their strengths, limitations, ability to increase understanding and also impact on behaviour change. However, conversations around heart age were much more common, and rarely in a negative context. Some patients were alarmed or even shocked to learn that their heart age was higher than their actual age and clearly did not expect to receive such information:

Well it made me, what she said is that this is the age group I would be . . . so obviously that’s alarming, because . . . it’s like adding another 10 years to your life. You know which you don’t want . . . you know you don’t mind erm . . . being at the level where you are at your age . . .

JBS3, 11_028, male, 58

Clearly, a CVD risk metric, such as heart age, that is understandable and impactful has the benefit of giving patients a better appreciation of their risk than does 10-year or event-free survival age.

Theme summary

There was evidence that attending a health check had a positive impact on many patients, but these often related to contemplating change or making relatively small changes to lifestyle. Heart age appeared to be preferred by patients, and beneficial for facilitating patients’ understanding and appraisal of CVD risk (compared with 10-year or event-free survival age).

Difficult to absorb information

Patients frequently highlighted the difficulties in absorbing information presented in the health check and commented that they ‘Didn’t really sort of take in you know what the numbers were’ (QRISK2, 3_125, female, 56). They felt that the volume of information provided to them in a short space of time limited their capacity to retain and accurately recall it:

As I say, when you are in somewhere like that you can’t take on too much either can you really? Because it all becomes a bit muddled together.

JBS3, 4_263, female, 61

This appeared particularly problematic when exclusively presenting information verbally, which is common practice during a health check:

You know what I mean, you, you could take the time to look at it whereas, when, when it’s when it’s spoken to yer, it’s spoken then it’s gone.

QRISK2, 10_539, male, 51

This supports evidence for the benefit of presenting risk information visually. 23,77 Patients who feel overloaded with information that they cannot effectively process will be limited in the extent to which they understand their own personal CVD risk and appraise the implications.

Positive impact of presenting cardiovascular disease risk visually

One key feature of the JBS3 risk calculator, aside from the various risk metrics (10-year risk, heart age, event-free survival age), is the visual presentation of risk information. The interactive nature of the tool facilitates a more collaborative consultation between patient and health-care practitioner. It was often observed in the video-recorded health checks that practitioners would share their computer screen with patients to communicate CVD risk information (see Chapter 4).

Patients in the JBS3 group typically commented on the visual presentation of information, which was perceived positively and overcame some of the barriers related to verbal communication of information:

It was good that you could see the screen and how she actually . . . how she worked it out as well rather than somebody just telling you . . . I thought it was a good to see . . .

JBS3, 7_044, female, 54

There was also a common feeling that seeing information written on the screen, and the accompanying visuals, strengthened the message and aided recall (more than simply hearing it from the practitioner):

. . . because it was on the screen, I think that is such an aid to memory . . . you know it’s just that sort of interactive ability really to be able to see something, rather than just being told information. Because in any situation that is new to you, if there’s a lot of things going on and you are not sure what’s going on, you don’t hear . . . But if you see it, it is actually much, much clearer to you.

JBS3, 4_263, female, 61

Patients also reported that they took more notice of the information when presented graphically, and that the presentation influenced intentions to engage in positive behaviour change:

I certainly got the gist of what [practitioner] was saying and it’s quite graphic seeing it there on screen erm, you know heart age 65 and I’m, I’m not quite 60 so you’re thinking yeah I ought to do something about that and yeah the intention is there.

JBS3, 7_105, male, 59

This did not appear limited to one type of CVD risk information provided. Rather, visual presentation appeared to increase understanding and impact of 10-year risk (‘Yeah she did give me a percentage score, she did something on the computer that showed you erm what it was and that was quite good actually’; JBS3, 7_044, female, 54) and heart age (‘But the 65 was a big 65 on the screen’; JBS3, 7_105, male, 59) although was not commented on in terms of event-free survival age.

Positive perception of risk manipulation

The JBS3 calculator function that allows the practitioner to manipulate a patient’s CVD risk by altering modifiable risk factors was perceived positively by patients, with one describing it as ‘quite impressive really’ (JBS3, 8_162, female, 67):

Yes I think it helps, rather than somebody talking to you and saying, ‘well it’s like this, it’s like that’, but actually when you can see it and then by altering it, you know and saying, ‘if we put this information in you can see how . . . so if you were much heavier say, for example, or if you smoke, or if you do these sorts of things’, so I found that really helpful. Really clear and understandable from my point of view anyway.

JBS3, 4_263, female, 61

This also appeared to have more of an impact on patients, potentially increasing motivation to engage in behaviour change:

You can see the differences . . . not just somebody saying or if you do this, you can, you’d be better . . . or better off . . . shall we say, not better, better off, you know but if you can, if you can read it and you can, you can compare the two at the same time . . . You can, you can it, it has better more, more of an impact.

QRISK2, 10_539, male, 51

It was also evident that some patients had made positive lifestyle changes and followed practitioner recommendations based on their risk manipulation in JBS3:

Erm do you remember being shown how your risk might change, you kind of mentioned it a little bit earlier how she said if you reduce your cholesterol erm it could have an effect on your risk.

Yes she did yeah bring the percentages down and all this, all the above which is as I say I took on board . . . Which err you come out, I came out thinking well yes my lifestyle needs to change . . . because it’s an easy thing to slip into . . . I have made the effort and through that, through this meeting you know so it . . . the benefits are there it’s definitely done something for me . . . In as much as made me decide well ‘come on enough pull your finger out get on with this’, you’ve been told, you know.

JBS3, 4_394, male, 65

Theme summary

Patients are given a lot of information in health checks, which they can struggle to absorb and retain. There appeared to be benefits of visual presentation of CVD risk information and risk score manipulation within JBS3 to foster patient understanding of CVD risk and motivation to follow recommendations for behaviour change.

Ten-year percentage risk (QRISK2)

QRISK2 is presented as percentage 10-year risk, and an equivalent score is also presented in JBS3. Practitioners generally reported feeling confident in communicating to patients their estimated 10-year risk (QRISK2): ‘I think I’m OK, I think I’m confident . . . I think I deliver it well . . .’ (9.1, QRISK2, HCA); ‘Quite confident’ (11.1, JBS3, HCA); and ‘I feel fairly confident’ (6.1, QRISK2, PN).

However, there was reported variation in its application. Some practitioners stated that they tend to deliver the CVD risk information to all patients in the same way: ‘I’m like a robot. I think I say the same thing to every patient . . .’ (6.1, QRISK2, PN). This is problematic given the evidence supporting tailored risk communication to accommodate varying patient needs and understanding. 24 Other practitioners said that they adapted their delivery of 10-year risk, recognising that ‘the way in which it’s delivered can sometimes [be] more than the actual QRISK . . .’ (10.1, QRISK2, PN):

I try and explain it for the level of the person that is sitting there and adapt it.

6.2, QRISK2, PN

Despite the communication of QRISK2 (or QRISK3) being mandated in NHSHC,17 there was variation in whether or not practitioners chose to communicate 10-year risk during the health check:

[If] you think the patient perhaps is not going to pay any attention to you, they are not going to take it in, then no.

6.2, QRISK2, PN

I personally always do it . . . because the whole point of the health check is that you reach that number . . . the QRISK is below the percentage that you want and that you know, and that is fine.

1.1, JBS3, PN

Some practitioners suggested that the decision of whether or not to tell patients their 10-year CVD risk ‘very much depends on the patient’ (6.1, QRISK2, PN) and can be informed by patients’ age, level of understanding and engagement.

Other factors that appeared to influence the extent to which practitioners used 10-year risk included, first, the perceived usefulness. Through identifying patients with elevated risk (i.e. ≥ 10%), 10-year risk was seen to have value in guiding recommendations and clinical decision-making: ‘QRISK is more the sort the diagnosis and the referrals and things’ (2.1, QRISK2+, HCA). But it was generally not considered beneficial for the patient:

. . . giving them a percentage, doesn’t inspire them, doesn’t motivate them really . . .

1.2, JBS3, HCA

I suppose when you look at it and you have got that 10%, it is quite clear isn’t it? . . . Because you see it and you think about it ‘well they are going to need a statin’ . . . So it is quite clear cut then. I don’t always know how helpful it is to patients.

6.2, QRISK2, PN

Practitioners descriptions of the QRISK2 suggested that it was not used to facilitate discussion of CVD risk with patients: ‘A quick go-to tool. . . it’s OK. It’s all that we have’ (10.1, QRISK2, PN). In addition, although practitioners expressed confidence in communicating 10-year risk, there was evidence that their understanding of and confidence in explaining the score was, in fact, limited and that this prevented them from engaging patients in a discussion of their risk:

Several things are taken into account, so age, sex, their BMI [body mass index] . . . which does make it, to me, very complicated. . . to try and explain it to [patients] is the hardest.

12.1, QRISK2+, HCA

There was also a perception among practitioners that patients’ ability to understand risk information is limited, and as a result it is challenging for practitioners to communicate information in a way that patients will understand:

What I find difficult is sort of putting it in a way that the patient will actually understand . . .

10.1, QRISK2, PN

This has implications for practitioner training. There is clearly a need to improve practitioner competency in communicating CVD risk communication in a way that enables a meaningful discussion around risk management, in accordance with the competency framework. 19

Event-free survival age (JBS3)

Practitioners who used JBS3 discussed issues with communicating lifetime risk to patients. For some, this was new, and it became easier with experience: ‘. . . it is new for me, it is new for the patients and yeah it just got easier every clinic that I did’ (11.1, JBS3, HCA). But event-free survival age appeared to be ‘. . . the hardest one to try and communicate . . .’ (7.1, JBS3, HCA). This difficulty was two-fold. First, unlike 10-year risk, there are no thresholds for event-free survival age to identify patients as low, medium or high risk, and, therefore, no prompts to initiate discussion or action (e.g. risk > 10% would be followed up with further tests and discussions around statins):

. . . That’s there’s no benchmark to give that comparison to be like ‘right you can live till 84 and so-and-so live till 82’, so that one was harder . . .

7.1, JBS3, HCA

Second, as the above example illustrates, there was some confusion around the meaning of event-free survival age. Despite providing brief introductory training and a training video in the use of JBS3 and its outputs (see Chapter 2), some practitioners described lifetime risk in terms of expected age of survival (rather than CVD event-free survival). This is clearly an issue, resulting in the communication of incorrect or misleading information.

One practitioner also questioned whether event-free survival age would motivate a patient to make a change if the predicted age the patient is expected to live without a CVD event was high:

Again, because you have told them that they are going [to] live until they’re 82 without any heart attacks or strokes . . . Is that motivating them, or not motivating them? I would say not really motivating them.

Why?

Because . . . they are going get to 82 without anything happening to them.

1.2, JBS3, HCA

Overall, event-free survival age appeared problematic in terms of practitioner understanding and application. There was a perception that this kind of information could shock patients into changing their lifestyle to reduce long-term risk: ‘. . . if that’s telling them younger, than they’d expect it to be then that’s quite a shock tactic isn’t it . . .’ (9.1, QRISK2, HCA). But this did not reflect the majority view or experience.

Heart age (JBS3)

The heart age output from JBS3 was perceived positively. Practitioners expressed confidence in communicating heart age, ‘ ’cos it’s just the easier one to . . . sort of explain’ (8.2, JBS3, PN). Heart age has inherent benchmarking through comparison with chronological age, which overcomes a perceived limitation of event-free survival age: ‘. . . they know how old they are and then they are exactly . . . “oh it’s the same age as me”, so I think they understood that more’ (11.1, JBS3, HCA). Practitioners also suggested that patients liked and responded strongly to the ‘visual of heart age’:

. . . because it is an actual [heart] when you go onto that screen of heart age. The heart is there, you can’t escape that and then you have got your age right by it.

8.1, JBS3, HCA

. . . they liked the heart age . . . I don’t know whether it was because there was a picture of the heart . . .

8.2, JBS3, PN

It appeared that practitioners were using heart age as a motivational tool to highlight to patients the need to make lifestyle changes rather than simply a risk score metric to be relayed to patients, as was the case for 10-year risk.

Risk score manipulation and visual displays (JBS3)

The risk factor manipulation function and visual displays in JBS3 were discussed positively. Those using JBS3 reported that visual features provided an alternative method to deliver key messages to patients: ‘. . . [patients] have said it is quite nice to see visually, [as] opposed to me talking’ (8.1, JBS3, PN). Some practitioners perceived that patients were more engaged when presented with visual information, based on non-verbal cues:

Because, like I said, it is visualised. They are seeing, they actually are interested. They have come closer to me, their body language was good, they were looking at that, they did ask questions and they were happy, because I think it was visualised.

11.1, JBS3, HCA

The visual stimuli in JBS3 were thought to ‘make a far better impact’ by being ‘able to show them visually, rather than just verbally tell them’ (5.1, JBS3, PN), promoting interaction and discussion with patients:

So, this is the point [going through JBS3 outputs] then they [are] starting to ask questions, or looking like they [are] actually wanting to be here . . .

8.2, JBS3, PN

There was some conflicting evidence from the video-recorded health checks that most patients give limited responses when presented with risk information, even when JBS3 was used; often single-word acknowledgements, infrequent questions and scant non-verbal cues (see Chapters 4 and 7).

Theme summary

Ten-year risk is viewed as a guide for clinical decision-making, not for facilitating discussion of CVD risk, and the evidence suggests that practitioners lack the understanding and confidence to explain the score. Event-free survival age appeared problematic in terms of practitioner understanding and application; training would be needed before this type of metric was recommended for use in practice (in addition to general training in discussion of risk with a preventative focus). Heart age was favoured for being easier to explain and well-received by patients, who were thought to appreciate the visual presentation. Practitioners thought that patients responded well to the risk manipulation and visual displays in JBS3.

Practitioner understanding

There was a strong link between practitioners’ understanding and communication of CVD risk. The majority described 10-year risk in a similar way, with reference to current lifestyle and risk factors:

. . . so the score that we are giving them is their risk of heart attack or stroke in the next 10 years, based on their . . . current lifestyle as it is as the moment, so it’s not sort of you know if their lifestyle changes or any risk . . .

7.1. JBS3, HCA

They also referred to the thresholds used to identify those with elevated risk: ‘So, we are looking at patients over the 10% mark, so their risk factors of developing strokes and heart disease is increased’ (6.2, QRISK2, HCA). Despite apparent confidence in communicating 10-year risk expressed by most, one practitioner doubted their understanding of what the score meant (and cited a preference for heart age):

I feel confident in the way that I give it . . . but then you are only reading off a piece of paper . . . I don’t feel that I understand . . . what the percentage is really . . . I prefer the heart age.

1.2, JBS3, HCA

Conversely, some felt that they understood the risk score, but questioned their ability to communicate it effectively:

Putting it into sentences which they can think well OK, or they might say, ‘well it just doesn’t make sense’ . . . I can probably babble sometimes and think, even I didn’t understand that.

12.1, QRISK2+, HCA

In some cases, watching the video-recordings of themselves delivering health checks (during the VSR interview) prompted practitioners to appraise their risk communication, with comments such as ‘It was just sort of abrupt wasn’t it? . . . There didn’t seem to be much of a consultation around it’ (1.2, JBS3, HCA) and ‘There is definitely room for improvement I can see myself there’ (6.1, QRISK2, PN).

Several practitioners acknowledged a need for additional training to improve risk communication with health check patients:

No, I think we do need more training in that.

1.1, JBS3, PN

We need more training to understand the risks better to deliver it to the patients.

1.2, JBS3, HCA

When asked about any training they had received, practitioners discussed this in terms of ‘can’t even really call it on-the-job training, can you, really?’ (1.1, JBS3, PN) and ‘. . . it’s been on how to use the [PoC testing] machine . . . rather than how to talk . . . and understand the risks’ (6.2, QRISK2, PN). The latter point offers support to findings from the video-recorded consultations (see Chapter 4), which suggested that the process often takes precedence over quality; the need to populate all parts of the NHSHC template might be at the expense of the quality of interaction.

Perceived patient understanding

It was suggested that patients who understand risk information are more likely to acknowledge their personal CVD risk and understand the options that could help them modify their risk factors: ‘Because if they understand that, they make changes to their lifestyle’ (4.1, JBS3, HCA). Perceived barriers to understanding the risk information conveyed within the health check, included patients’ level of education and possible ‘language barrier[s]’ (11.1, JBS3, HCA), but also a willingness to engage:

Why do you think some don’t understand [% risk score]?

Maybe poor education . . . maybe they do understand, but they don’t care . . . so they don’t want to know, they don’t want to discuss it . . .

3.1, QRISK2, HCA

Similar to findings from patient VSR interviews (see Chapter 5), there was also the suggestion that the consultation involved a lot of information for patients to process, which could contribute to lack of understanding and engagement:

. . . So, it’s alright giving them all the information [10-year risk], but whether they’ve taken any of it in . . .

8.2, JBS3, PN

Practitioners reported using patients’ verbal and non-verbal reaction to risk information to gauge understanding: ‘Nodding their head, so I think they all understood what I was trying to say’ (11.1, JBS3, HCA). They also appeared to recognise that this was limited: ‘It was based on their reactions, but you did wonder sometimes whether they actually fully understood’ (5.1, JBS3, PN). The challenge of using patient response to judge understanding was evident from analysis of video-recordings of their health check consultations, which rarely demonstrated clear patient responses, positive or negative (see Chapter 4). Here, there was a common feeling that even if patients ‘say they understand the percentage, some will, and some won’t’ (12.1, QRISK2 +, HCA) (i.e. that patients might not tell practitioners when they do not understand the CVD risk information presented):

Yes, they seemed to be [understanding % risk score], that’s the impression that I had got, and they didn’t say otherwise, but maybe they wouldn’t, I don’t know.

6.2, QRISK2, PN

Moreover, there was little evidence that practitioners routinely explored patient understanding through asking them, which again confirms findings from video-recorded consultations that health checks tend to be practitioner dominated (see Chapter 4).

Heart age and risk manipulation functions of JBS3 were again discussed favourably. Heart age was considered easier for patients to understand (than 10-year percentage risk): ‘. . . they can get their head around that concept of their [heart] age a lot better than risk score . . .’ (7.1, JBS3, HCA). This could be due to the inherent comparison/benchmark for heart age. The visual demonstration of manipulating risk was thought to promote patient understanding; it provided a tool to help facilitate discussion of risk-reducing behaviours, not just a risk score. This was discussed in terms of showing patients the positive effects of their risk-reducing lifestyle choices; this is an example of gain framing (i.e. benefits of reducing risk), which has been linked with increased self-efficacy to prevent heart disease:78

No, I think they understand more, because then they have an idea how . . . what they need to do to change to find the results so . . . When you show them that if you bring your blood pressure down, your cholesterol down, your weight down . . . how that can affect the results . . . And they start to think about exercise and lifestyle . . . it’s like an eye opener to them . . .

4.1, JBS3, HCA

There was also a suggestion that this was used to present risk with a loss frame (i.e. consequences of high CVD risk), which might be more appropriate for those with a family history of CVD:78

So when I was increasing the blood pressure, or the cholesterol, or if they were a smoker, they actually saw the difference and then they were like, ‘no I won’t, and thank God I am not [a smoker]’, you know so it was a difference on them as well.

11.1, JBS3, HCA

Theme summary

Practitioners acknowledged gaps in their understanding or communication of 10-year risk, often relying on the thresholds (low, medium or high risk). There was recognition that understanding of CVD risk varied between patients, and a perception that heart age was understood more easily. Practitioners associated patient understanding with lifestyle changes. They relied on patient reactions to information to gauge understanding, rather than asking patients, although the VSR interviews did prompt a reflection that they should involve patients more.

Promoting risk management

Practitioners described some routine medical follow-up for patients with elevated CVD risk: ‘raised risk score over 10%, we’d always get them to come back have a complete workup of blood test done and book in to see a doctor’ (7.1, JBS3, HCA). Practitioners reported that, during the consultation, they try to promote risk management, primarily through educating and advising patients on how to modify lifestyle (i.e. improving diet, increasing exercise, reducing alcohol consumption) to lower cholesterol, blood pressure and body mass index:

I sometimes go into portion sizes and carbohydrates and things like that.

2.1, QRISK2+, HCA

Get a little bit more movement back and to get their heart pumping and use up some of the fats and calories that they are eating and storing.

12.1, QRISK2+, HCA

Alongside this, some practitioners said that they use written materials, which they felt helped patients to understand lifestyle and recommended changes: ‘. . . and then when we give that diet sheet, then they look at it and they read and they go “oh, yeah, this is where I am going wrong” ’ (11.1, JBS3, HCA).

Referral to services to support risk reduction, such as exercise referral, smoking cessation and weight management, were discussed. Some practitioners suggested that not all those referred to such services will attend, ‘they don’t very often like to take you up on them’ (12.1, QRISK2+, HCA), and, for those who do, attendance might not be sustained: ‘we have a big drop-out . . .’ (10.1, QRISK2, PN). Practitioners suggested that a variety of contributing factors could explain uptake and retention of intervention services, such as the cost of continued attendance (e.g. gym membership after period of subsidised exercise referral); the absence of local services; lack of time to attend (working or lack of childcare); or preferring to make lifestyle changes unaided: ‘Oh well I know what changes I need to make, I’ll go away and make them’ (7.1, JBS3, HCS).

There were a small number of reports by practitioners about patients making lifestyle adjustments following a health check:

I see patients in general around the practice . . . say 6 weeks later and somebody will catch me and say, ‘oh, I really took in what you said, I’ve made some changes and my diet’s changed and this that and the other’.

9.1, QRISK2, HCA

A lack of communication between the general practice and external services to which patients could be referred meant that practitioners were uncertain if patients attended such programmes and, therefore, how effective programmes were. Even when patients are referred to the GP following the health check, practitioners were sometimes unsure about subsequent actions and outcomes.

Barriers to risk reduction

Practitioners identified several patient factors related to environment and intrapersonal factors in the PMT (see Figure 2) that could be barriers to patients forming risk-reducing intentions or behaviours. Some practitioners inferred that patients’ culture and personal beliefs could be barriers to risk reduction. It was suggested that cultural beliefs may affect current lifestyle choices and patients may not recognise that current lifestyle choices pose a health risk: ‘Yeah, because if you do say to them, they just say, “oh, but we have parties, we have this that, and we have a lot of sweet food” ’ (4.1, JBS3, HCA). Further, personal health beliefs may affect intentions towards behaviour change. Practitioners suggested that patients who do not feel unwell may have less incentive to make a change: ‘You know think they’re fine they think they’re invincible and you know nothing wrong, ’cos they don’t feel poorly’ (9.1, QRISK2, HCA). Practitioners also indicated that some patients believe that their risk of ill health is related to family history, rather than their lifestyle:

Sometimes, no matter how much you say to patients ‘I think it’s your diet and lifestyle’, they are still adamant that, because their mum and dad have got it, that that’s why they have got it.

6.2, QRISK2, HCA

Moreover, some practitioners reported finding it difficult to refer patients to an intervention service because local lifestyle support services were no longer operating or were not considered to be effective:

So, they have taken the lifestyle programme off us haven’t they, so we can’t offer that any more.

6.1, QRISK2, HCA

In addition, some practitioners expressed a lack of confidence in intervention services’ ability to support patients to change, despite referring patients to such services: ‘I don’t think the Help to Quit is very useful’ (2.1, QRISK2+, HCA). This experience and perception could explain the relatively low number of referrals observed in patient medical records following health checks (see Chapter 3).

There was a common inference that patients’ willingness or apparent intentions to make risk-reducing changes was largely determined by their intrinsic motivation, that although practitioners could provide information, the impact that it had depended on patient motivation:

You can give them all the options in the world and then I think a lot of it is down to the patient and their motivation to change.

6.2, QRISK2 HCA

It just depends how important they think it is for themselves, erm we can only do what we can do here. Once they are out that door we don’t know, yeah, so hopefully some people do take the recommendations and do what they can do.

11.1, JBS3, HCA

In support of this idea that patients are recipients of information and advice, rather than active participants in a discussion of their risk and its management, one practitioner described how they might respond to a patient who does not understand CVD risk by moving on to risk factors and modification:

That they just might not have understood, like the complexity of how you explain the QRISK and that maybe where I would then adapt it, and maybe there are times when I might explain it, but not explain the QRISK and the percentage as much, like I did with those patients you know?

So how would you explain it with those types of patients then?

Just basically saying your risks of strokes and heart disease have increased . . . You have got a lot of risk factors, you know, we need to give you a statin to bring you down. I need you to stop smoking, I need you to eat better. You know and keep it a bit more basic.

6.2, QRISK2, PN

With this approach, the practitioner would be telling patients what to do, rather than engaging in patient-centred dialogue. This confirms the indications that many health check consultations are practitioner dominated (see Chapters 3 and 4) and do not include elements of motivational interviewing, as recommended in the best practice guidance. 17

Finally, even if patients appear to understand and respond positively to the risk information and advice in the consultation, there can be uncertainty about the sustainability of the changes implemented:

Well I think it will have quite a good impact. Whether it would be sustained is another thing, but initially it did seem as though ‘oh yeah I will do something about that’, but it is like human nature isn’t it with things . . .

5.1, JBS3, PN

Theme summary

Most practitioners described risk management as lifestyle advice, routine medical follow-up triggered by raised 10-year CVD risk (≥ 10%) or risk factors, and referral to support services (e.g. exercise referral). Lifestyle advice was the most common intervention, sometimes supplemented by written information. Perceived barriers to risk management included patients feeling healthy and, therefore, not recognising the need for risk management (risk severity or vulnerability), lack of patient motivation, and limitations of the services to which patients could be referred.

Summary

This case study describes Abbie, who was in relatively good health, with no clear CVD risk factors and who did not require any post-health check follow-up. CVD risk was communicated using heart age and risk manipulation prompted Abbie to make dietary changes to address her elevated cholesterol, which was identified in the health check.

Patient

Abbie is a 61-year-old white British female living in a relatively deprived area (IMD decile 2). She does not drink alcohol or smoke, walks at least 10,000 steps a day, and regularly undertakes gardening and housework. She has no family history of CVD and has lost 8 kg since her weight was last recorded (8 years previously). She feels that having this type of health check is important and expresses gratitude to the practitioner. Her total cholesterol is slightly elevated, but not enough to warrant follow-up.

Practitioner

This health check is delivered by a British Asian female HCA with 2 years’ experience of delivering health checks and who has received ‘on the job’ training (no formal training). The HCA is positive about NHSHCs in general and likes the JBS3 manipulation options. She feels confident in communicating the risk score as the ‘GP has explained it’ to her and thinks that, in general, patients understand the information presented to them.

The NHS Health Check

The health check is shorter than average (just over 17 minutes). However, the HCA talks for a smaller than average proportion of time (46.0%) and Abbie speaks a little more than average for the sample. Abbie’s QRISK2 score is calculated as 4.8% and her heart age is estimated as 61 (equal to her actual age).

What worked well

Abbie reacted positively to heart age:

I was pleased to hear that [heart age was 61] and encouraged really . . . I found the whole thing encouraging, it encourages me to make some changes and to look after myself, because I think that is my duty.

Abbie liked the visual element of heart age: ‘because it was on the screen, I think that is just such an aid to memory’ and ‘an understandable way of presenting it’. The HCA also recognised Abbie’s positive reactions. She felt that being able to show Abbie her heart age on screen had helped Abbie to understand the concept, and that it was helpful for the two of them to compare Abbie’s heart with her true age, to see whether it was ‘good or bad’.

During risk score manipulation, there was discussion around good and bad cholesterol, and how to reduce Abbie’s total cholesterol, which was elevated. Abbie and the HCA appeared to enter into a mutual discussion, which concluded with Abbie thanking the practitioner. Abbie later reflected:

I thought that was very well done actually, and because the nurse showed me . . . by changing some of the data . . . that was very clear and understandable to me, so I thought that was quite a good way of doing it.

This information was remembered by Abbie, who had implemented some changes: ‘[the practitioner] actually suggested cutting [cheese] down to a couple of times a week, which I have done already, so that definitely has had an impact’. Abbie said during her VSR interview that when ‘information was given to me in the recommendation . . . I have implemented that immediately’. The HCA also liked using the risk manipulation function within JBS3, describing it as an ‘eye opener’ for patients.

Areas to improve

Abbie highlighted the challenge for practitioners in reading patients’ reactions to CVD risk information, and the need to ask patients if they understand and tailor appropriately. Abbie could not remember her 10-year risk: ‘I am not sure I fully understood the importance of the cardiovascular part of it’. However, the HCA perception of Abbie’s response, when she watched the video clip of their consultation (‘they just nodded and quite calm’), and our ratings of engagement based on non-verbal cues identified Abbie as being generally positively engaged.

The HCA liked JBS3 and being able to ‘show [patients] on the screen, they can see what is going on’, but referred to event-free survival age risk as ‘a good way to show patients if you can control things, your life expectancy will increase’. This showed that the HCA misunderstood the score and, as a result, did not describe it clearly to Abbie during the health check: ‘So on average it’s [the system] picked inside 80–85 about a heart attack or stroke’.

In turn, Abbie did not understand or ask for clarification. Comparing it with heart age, she said she did not understand it ‘as clearly I don’t think . . . I don’t think I took it on board quite in the same way’. This lack of understanding was not obvious to the HCA, who attributed Abbie’s lack of reaction to the fact that she was reading the screen and taking in the information, rather than to a lack of understanding.

Risk score manipulation was performed using blood pressure, which was normal, rather than cholesterol, which was elevated. This highlighted the need for tailoring. Abbie left with unmet information needs. She was keen to learn more about how to implement a change, but did not follow-up:

Until you know what to do you can’t implement it and particularly over blood pressure. I would like to learn [more about] the blood pressure, but I don’t know how you go about doing that.

Had the HCA asked Abbie if she understood or required further clarification, this could easily have been addressed.

Summary

This case study describes Barry, a patient who presented with a number of CVD risk factors including increased alcohol consumption, poor diet and slightly elevated cholesterol. Barry was recommended to make a specific lifestyle change that he was already considering. Despite other elements of CVD risk communication being suboptimal, he made a positive change, demonstrating that the health check can serve as a prompt to initiate risk-reducing behaviour when patients are on the cusp of acting.

Patient

Barry is a 47-year-old white British male living in a deprived area (IMD decile 1) who drinks 40 units of alcohol per week, has never smoked and reports doing ‘a bit of running every now and again’. His diet is the main area of discussion, as he admits to eating ‘a whole pack of chocolate biscuits every single day’ and excessive consumption of dairy products. Barry has a family history of heart problems (maternal grandmother) and diabetes ‘runs in my family’.

Practitioner

This health check is delivered by a white British female PN who has 8 years’ experience of delivering health checks and had previously attended two NHSHC training courses with the last being ‘at least 2 years ago’. The PN felt that she did not ‘have any problems’ with communicating CVD risk to patients and preferred JBS3 over QRISK2 because ‘you can actually communicate far better, and it’s got visual . . . everything about it, the patient can actually understand it better than just me talking’.

The NHS Health Check

This health check lasts just over 22 minutes, approximately average for the JBS3 sample, but is relatively dominated by the practitioner, who speaks for 56% of consultation time, and Barry speaks for 18.4% of the time. Barry’s 10-year risk is 3.1%. Only total cholesterol was considered ‘a little bit high’, at 5.3 mmol/l (ideally < 5 mmol/l). Barry’s heart age was estimated at 54, which was 7 years older than his actual age, which the PN attributes to his family history and elevated cholesterol. Based on a discussion of lifestyle, the main recommendations are to reduce consumption of chocolate biscuits and increase fruit and vegetable intake.

What worked well

About 3 weeks post health check, Barry had implemented dietary changes and felt that this was sustainable. Although he had ‘wanted to reduce my biscuit intake anyway’, the health check ‘prompted’ this change, and he had not ‘eaten chocolate biscuits now for over a fortnight’. The discussion around alcohol had also moved Barry to consider moderating his alcohol, but he saw this as a next step. A blood test 3 weeks post health check showed that cholesterol had fallen to within the normal range and, after an appointment with Barry’s GP, no further action was taken.

Areas to improve

Barry said that he was ‘expecting more’ from the health check, ‘something more in depth’, with more tests.

Although the PN was confident in delivering 10-year CVD risk and believed that ‘the majority of them [patients]’ understand, when watching the video clips of her interview she ‘did wonder sometimes whether they [patients] actually fully understood’. Barry could not remember his 10-year risk when asked by the researcher, even when prompted with a video clip. Although he interpreted the score as ‘quite low’, he said that he ‘didn’t remember it no, I don’t understand it’.

In line with current findings from the VSR patient interviews (see Chapter 5), the PN felt that being told heart age would be shocking, but Barry did not react in this way. His interview demonstrated that he had not fully understood how this information had been derived. The consultation would have been improved by the PN picking up on the trigger of this apparently contradictory result (of old heart age, but low 10-year risk).

There was misinterpretation of the discussion of event-free survival age, from which Barry took away that he had survival age of 73 (rather than event-free survival). Barry’s response was to ‘live my life to the full’, but such misunderstandings could have significant negative consequences including a sense of futility in relation to lifestyle that negatively affects health behaviour. The practitioner’s VSR interview revealed that the PN had similarly misinterpreted this risk score.

Finally, the PN altered the blood pressure as an example of how making small changes can be effective, but this point was missed by Barry, who was confused by the different numbers being entered: ‘I thought my blood pressure was low?’.

Summary

This case study illustrates a patient, Carl, who presented with a number of CVD risk factors, and in whom presentation of heart age had a significant impact. This resulted in Carl making a number of small lifestyle changes following his health check.

Patient

Carl is a 59-year-old white British male who lives in a relatively affluent area (IMD decile 9). He reports rarely drinking and that he stopped cigarette smoking in 2015, after ‘40 odd years’. He does little exercise because of ‘spinal problems’ and had a paternal grandfather who ‘had a stroke . . . he was late 60s, 70s’. Carl reveals that his friends and family ‘think I’ve got type 2 diabetes’, which was one of the reasons for attending the health check.

Practitioner

This health check is delivered by a white British female HCA with 5 years’ experience of delivering health checks. She received training in health checks prior to delivering them, but this primarily focused on PoC testing, and she cannot otherwise recall the training content. The HCA is ‘quite happy communicating’ CVD risk to patients’ and ‘primarily do[es] it in the same way’ for all patients she sees. She prefers ‘JBS3 because it has got the heart age on it’.

The NHS Health Check

This health check lasts longer than average (> 27 minutes) and involves more patient speaking than most (32.9% of total time), and CVD risk is discussed for > 2 minutes. Carl’s QRISK2 is 11% (medium risk), described as ‘a little bit higher than we’d like it to be’, and his total cholesterol has ‘gone up a little bit’, to 4.9 mmol/l from 4.5 mmol/l (in 2016). It is suggested that Carl considers improving his good cholesterol. His blood pressure is ‘just sort of on the higher end of where we’d want it to be’ and his body mass index (BMI) is 28.4 kg/m².

What worked well

Carl said the health check was ‘what I expected and I’m very happy, yeah’ and was ‘now more mindful of, of what I am or I’m not doing’.

The visual presentation of heart age had the most impact on Carl. He reported that it ‘shocked me, no it concerned me’, and made him think ‘yeah, I ought to do something about that’. He remembered ‘the picture on that screen, the 65 . . . being in the middle’ and ‘thinking bloody hell I ain’t 65’.

The HCA manipulated Carl’s information within JBS3 to illustrate that reducing his total cholesterol and increasing HDL cholesterol through simple changes to his diet and physical activity levels would reduce his CVD risk to 9.3% and lower his heart age to 62. The HCA thought ‘it’s really good having that option’ because ‘they can actually visually see the changes’. Carl did ‘remember that’, but had ‘already got the message’ by that point.

Carl found a diet sheet provided by the HCA ‘helpful’. He reported several positive behaviour- and health-related outcomes:

• ‘ordered a small blood pressure monitor’, which he admitted he had ‘never done anything like that before’

• felt ‘more mindful’, and was trying to ‘do a bit more exercise’ and ‘now looking at certain foods’

• had ‘been researching because of my concerns about diabetes’

• diabetes was later diagnosed following a full blood check at 6 weeks post health check and Carl was prescribed metformin.

Areas to improve

Ten-year risk was not explained by the HCA and, subsequently, was not recalled or understood by Carl. On watching the corresponding video clip, he said that it ‘didn’t make a significant impact on me’. In the absence of an explanation from the HCA, Carl questioned ‘exactly how she arrived at that percentage’, only remembering that ‘they’d like me to be 10% or less and I’m only 11’.

The HCA did not understand and, therefore, miscommunicated event-free survival age: ‘currently you can sort of expect to survive the age of 79 based on your lifestyle factors’. The HCA found it ‘the hardest one to try and communicate’. In this health check, it did not create a serious issue as Carl reported that seemed ‘a long way off, you’re looking like 20 years in the future’ and recalled thinking ‘well that’s better than what my dad or what my grandparents had’.

Summary

This case study presents Deborah, a patient in good health, for whom the health check provided reassurance. In this example, Deborah attended with a positive attitude to her health. She tried to lead a healthy lifestyle and was grateful of the health check as an opportunity for reassurance.

Patient

Deborah is a 63-year-old white British female living in a relatively affluent area (IMD decile 10) who has never smoked, reports modest alcohol consumption (‘7–10 units per week’), ‘excellent . . . varied diet, nice portion sizes’ and is a ‘Pilates teacher’ who teaches aerobics, and practises yoga for 2 hours each day. She has no known family history of CVD, but is aware of high cholesterol in the family (including herself).

Practitioner

The health check is delivered by a white British, female, newly qualified PN. She has 9 months’ experience of conducting health checks and has ‘had a lot of shadowing’, but no formal health check training. She felt confident ‘explaining what the QRISK is . . . yeah I’m fine with that’.

The NHS Health Check

This health check is relatively short (17 minutes), but includes > 3 minutes of CVD risk discussion, and Deborah speaks for > 28% of the health check (above average). Deborah’s 10-year risk is 4.7% and her heart age is 58 years. Total cholesterol is slightly elevated (5.8 mmol/l) but, in the context of LDL and HDL figures, her cholesterol level is described as ‘within normal range’.

What worked well

Overall, Deborah reported that the health check ‘was as I expected’ and that it was ‘good to actually have some reassurance that everything is OK’. She said, ‘I have always been interested in my health, so, erm, for me it’s a positive thing’ and was ‘grateful that [they] had the opportunity’ to have the health check.

Heart age was the CVD risk information that affected Deborah most positively: ‘when she told me that my heart age was 5 years less than my actual chronological age’ it served as ‘a positive reinforcement both for me and within my job’. The PN reported that heart age came ‘more naturally the more I used it’ and felt that patients ‘generally responded really well’.

The PN correctly described event-free survival age and, in turn, Deborah later interpreted it to mean that ‘I should be alright until I am in my 80s, yeah, before anything might start to develop . . .’. This emphasises the importance of practitioners having accurate knowledge of CVD risk scores in order to convey this to patients and mitigate against misinterpretation.

Using JBS3 had led to a change in practice. The PN suggested that they ‘much prefer using [JBS3]’ as they ‘think it gives the patients a better understanding’ and ‘still use it now’ following the conclusion of the study.

Areas to improve

First, information presented without a visual aid was forgotten. Deborah was unable to recall her 10-year risk and, when watching the corresponding video clip, reflected that she ‘understood’ it and ‘wasn’t confused at all’. The PN also thought that ‘they [patients] all understood fully the 10-year risk’, but that QRISK2 is ‘lacking . . . the visual aids’.

Second, event-free survival age was communicated accurately by the PN in the health check, perhaps as they read the information from the screen, during the VSR interview, their discussion of this risk score indicated some misunderstanding: ‘really good you know . . . [patients can think that] “I am going to survive to 82” ’.

Third, risk manipulation could have been better tailored to Deborah by showing potential CVD risk reductions through lowering cholesterol (which was raised), rather than how risk would increase should they start smoking. Deborah had never smoked and confirmed the irrelevance of this scenario: ‘would never happen’. In contrast, the PN thought that manipulating patients’ risk scores ‘had a positive effect’ and that patients ‘really responded to that’. This highlights a mismatch between practitioner perceptions of how patients received CVD risk information and the reality.

Fourth, there was some inaccuracy in how the PN relayed patient HbA_1c information in relation to a result in the normal range, which ‘put a big red flag up in my head when I heard that’. Despite describing it as ‘up at the top end’, the PN did not engage in a discussion around this, but rather spoke about what would happen if it was in the pre-diabetes range. This resulted in Deborah deciding to ‘work on that although my solution to that is not to buy [sugar] and not have it in the house’. This could be an unnecessary burden resulting from inaccurate information.

Finally, Deborah was positive about the overall experience. However, her body language during CVD risk discussion was rated as either passive or slightly positively engaged. If picked up by the practitioner, this could have triggered a discussion to address some of the above issues with how CVD risk information was provided and received.

Summary

This case study presents, Eid, a patient with a generally healthy lifestyle in whom the health check identified mild hypertension. Although this prompted lifestyle changes, follow-up tests and an appointment, heart age was the only CVD risk information that he could recall and no lifestyle advice was offered.

Patient

Eid is a 58-year-old Asian British male who lives in a relatively deprived area (IMD decile 2). He has never smoked, does not drink alcohol and ‘play[s] a bit of sport’ including football and table tennis. Eid reports ‘eat[ing] a lot of fruit’ and has no known family history of CVD. He felt having a health check ‘would[n’t] do any harm’ as there is ‘nothing wrong with making sure everything is OK’.

Practitioner

The health check is conducted by an Asian British female HCA who has been delivering health checks ‘ever since the NHS programme started’ and received training ‘when it started’ and ‘. . . last year as well, and then it’s online as well’. The HCA feels ‘OK’ about communicating risk to patients and as ‘I know what I am saying, I am more confident and happy to speak to the patient’. She thought JBS3 was ‘really good, because it tells you the heart age, erm, it tells you the lifespan’ (reflecting a misunderstanding of event-free survival age observed among other practitioners). The HCA ‘enjoyed using it [JBS3]’, preferring it to QRISK2 ‘because it’s visualised, the patients you know they are seeing what their results are. It is not just more score’.

The NHS Health Check

The health check lasts just over 15 minutes. The HCA speaks for only 36% of the time, but Eid’s contribution is low (20%) and a relatively high proportion of the health check passes without either participant speaking. Eid’s 10-year risk of 15% is described as ‘medium’ and his heart age is estimated at 59, approximately 6 months older than his actual age. As Eid’s blood pressure is ‘high’, the HCA tells him that she will speak with the GP, who will ‘get back to you [Eid]’, and advises Eid that he is likely be called ‘in for a 24-hour blood pressure’ check.

What worked well

Overall, Eid felt that the health check was ‘helpful for me because, obviously, it gives me an idea of how things are’, as he rarely attended the surgery. Ultimately, the success of this health check was in identifying and following up the identified hypertension. Eid remembered that his blood pressure ‘was a bit on the high side’ and was ‘surprised’ as he had not ‘had any issues with it before’.

The most impactful element of CVD risk communication was heart age, with which Eid was ‘quite happy’ and found it ‘very useful to know’. The result ‘reassured what I already knew, that I was quite healthy’. The HCA suggested that patients ‘understood [heart age] a lot more . . . because obviously . . . they know how old they are’ and believed that the visual presentation was helpful (‘them seeing it, I think it’s a big difference’) compared with relaying the percentage risk score, as is usual practice.

Areas to improve

Eid did not recall the HCA telling him ‘what I needed to do’. Indeed, the HCA did not offer advice aside from clinical follow-up (re-test/GP appointment). Consequently, Eid had received ‘a few tips on what to do’ to reduce his blood pressure from his sister. Eid assumed that this was because the HCA ‘wouldn’t know until’ she had completed further blood pressure monitoring.

There were several issues with CVD risk communication. First, Eid had a medium–high 10-year risk, yet could not recall it, nor did he ‘put any importance on it’ (when reminded in his VSR interview) because he was ‘well within the percentage of being healthy’ and ‘thought the risk out of 100 was minimal’. The HCA reported using only the risk categories (‘high/medium and low’) because patients ‘don’t really understand the percentage of the number’. This seemed to be an assumption and was not checked by enquiring about Eid’s understanding or feelings.

Second, in the health check, the HCA communicated event-free survival age by saying ‘on average expect to survive is 80 for yourself without a heart attack or a stroke’. The HCA later discussed how this was a challenging screen and her communication suffered as a result: ‘telling them that your life expectation, it could be . . . I think I find it a bit hard’. This highlights a mismatch between what the HCA read from the screen in the consultations (which was correct) and her apparent understanding of this score as expected survival age. Moreover, the HCA thought that Eid ‘knew exactly what I was saying’ on watching the video clips. In contrast, Eid could not ‘recall it, I think she said . . . something about 80 something’, again misinterpreting event-free survival age as expected age of mortality: ‘I’ve got a few more years to live’.

Finally, Eid had never smoked and presented with mild hypertension. Yet, for risk manipulation in JBS3, the HCA showed how Eid’s risk would increase if he were a smoker. He appreciated being ‘reassured . . . that what I’m doing with my health is good and I’ve got to keep it going’, but a more relevant and tailored manipulation of risk would have been preferable.

Summary

This case study describes Freddie, a male patient with a range of CVD risk factors, and a medium–high 10-year CVD risk. Freddie was aware of the need to change his lifestyle. He attended with an ‘open mind’, and received quick and multiple follow-ups after the health check. In this way, the health check served as a catalyst for change and Freddie subsequently made multiple positive lifestyle changes. Quickly, his efforts were rewarded with weight loss and a reduction in blood pressure.

Patient

Freddie is a 64-year-old white British male who lives in an area of approximately average affluence (IMD decile 6) and has not visited his GP for 13 years. He has no family history of CVD, but presents with a range of risk factors: BMI of 45 kg/m² (morbidly obese category), moderate hypertension (167/90 mmHg) and very high alcohol consumption. His main reason for attending is to achieve weight loss, but he also has concerns about the physical effects of his alcohol intake. Freddie is positive about the health check, which he recognises as being about ‘prevention rather than cure’.

Practitioner

The health check is delivered by a female white British HCA with 2.5 years’ experience of delivering health checks and who has attended generic NHSHC training and shadowed a colleague. Despite positive perceptions of the programme, she feels that it ‘needs tweaking’, and thinks that the primary purpose of QRISK2 is to identify patients to refer to the GP, rather than to facilitate discussion around CVD risk.

The NHS Health Check

This general practice has access to Informatica, which offers some of the JBS3 functionalities, including heart age and risk score manipulation, which were used in this health check. At 27 minutes, this health check is considerably longer than most, and Freddie speaks more than the sample average (31% of total time). Freddie’s 10-year risk is 17% (approaching high risk, 20%), which he interprets as being ‘. . . lucky that’s quite low really – I was expecting it to be higher’. The HCA is disappointed with her delivery of 10-year risk information when watching the corresponding video clip (‘like a parrot’; ‘maybe I don’t do enough on that’).

What worked well

Freddie appeared engaged throughout the health check and spoke more than many other patients in the sample. He was ready to make changes, and the suggestions made by the HCA resonated with the issues of which he was already aware. This meant that Freddie was later able to recall parts of the health check conversation with the HCA (e.g. ‘wine has got sugars in them – sugars go to the wrong places’). This appeared to make Freddie receptive to suggestions and meant that the HCA did not need to convince him of the importance of CVD risk or lifestyle.

Freddie received a telephone consultation with a specialist nurse practitioner within ‘a couple of hours’ of his health check to reinforce the messages given (telephone reinforcement). This appeared to facilitate understanding and retention of 10-year risk, as he was able to recall ‘an analogy of you know like 100 people in a room – 17 can be expected . . .’ (words not used in his health check). A barrier to exercising was identified (could not afford gym membership), and the health check and subsequent telephone consultation led to a series of outcomes:

• discussion of statins

• exercise referral for 12 weeks of subsidised gym sessions at a local authority leisure centre

• Freddie changed his diet and alcohol intake ‘. . . I have dropped all the things that I know I should not be eating and it seems to be having an effect . . . I seem to have dropped a few pounds . . . I have dropped the booze a lot – well totally for the minute, erm, which has helped me sleep a lot better’

• reduction in blood pressure to the normal range.

Areas to improve

Freddie’s 10-year risk put him close to the high risk category, yet he did not recognise the risk severity.

Freddie could not recall being provided with heart age or the HCA manipulating risk scores, which contrasted with the HCA’s perceptions that ‘heart age is most effective’. Heart age is likely to have been forgotten because the follow-up telephone consultation focused on 10-year risk. There was scope to better tailor CVD risk discussion; for risk manipulation, the HCA showed how heart age would increase if Freddie were a smoker, yet he had never smoked and subsequently forgot his heart age. It would have been more appropriate to show changes in a risk factor more relevant to Freddie, such as reducing blood pressure through modifying alcohol and diet: ‘Well that just reaffirms my thinking about smoking and I did think, erm, the alcohol side might come into it’.

Summary

This case study of Grace illustrates that, even when relatively short with minimal discussion of CVD risk, a health check can be valuable in identifying areas for change, if the patient is already actively trying to improve their health.

Patient

Grace is a 56-year-old white British female who lives in a relatively deprived area (IMD decile 2). Her parents died suddenly of heart failure in their 70s. She stopped smoking 14 months earlier, attends Slimming World^® (Alfreton, UK), has not gained weight since her last visit to the GP, and reports moderate alcohol intake. Grace is engaged with, and has a positive view of, preventative health care (‘usually have the well-woman check’) and describes the health check as a way ‘to find out if there’s any underlying problems which you may have and it’s just really good to keep on top, because you feel fine but you don’t know what is going on inside’.

Practitioner

The health check is delivered by a white British female HCA with 2.5 years’ experience of delivering health checks. After participating in the recorded health checks, the HCA completed training in NHSHC, after which she concluded ‘I don’t think you should give these health checks unless you have had the training’.

The NHS Health Check

The Health Check lasts just 16 minutes and Grace speaks for only 10% of the total time. CVD risk is discussed for just over 40 seconds. During the health check, Grace’s 10-year risk is calculated at 6.9% (low). Slightly raised total cholesterol (6.1 mmol/l) and blood pressure are flagged as initial concerns.

What worked well

This health check identified a rise in Grace’s cholesterol, which she attributed to less healthy dietary habits ‘creeping in’. Risk management discussion focused on this also; the interview data demonstrated that Grace was clear about her related goals:

I really am going to like get the cholesterol down that’s my main thing . . . So I am really, really saying right ‘no, we are not even having that in the house, we are not eating this, we are not having that any more,’ and we are definitely eating better . . .

Two in-clinic blood pressure measurements showed Grace’s blood pressure to be raised, but, at home readings were found to be normal, no further action was deemed to be necessary.

Areas to improve

The HCA spent very little time discussing 10-year risk:

I’m going to access your QRISK2 score, erm which is, which tells us your likelihood of contracting cardiovascular disease within the next 10 years . . . So it’s 6.85, which is still OK, it’s below 10.

Hmm.

So, erm so it’s not something that we really need to worry about, but getting your cholesterol down.

Grace could not recall her 10-year risk, only that it was ‘low’, and said she ‘just didn’t really absorb it . . . afterwards I thought “well she said it was under 10”, but what it was, I couldn’t quite put my finger on it’. Despite that, Grace was motivated to reduce her cholesterol. Encouraging Grace to maintain her good lifestyle while making small changes, particularly given her positive engagement with the health check, could prevent further decline.

Summary

Harry presented with multiple CVD risk factors and the PN spent > 3 minutes explaining his CVD risk. This seemed to convince him of the need to go back on to a low-dose statin, despite initial reluctance. However, following the apparent success of his health check, Harry could not recall his 10-year CVD risk, remained reluctant to take statins and had no intention of changing his lifestyle.

Patient

Harry is a 61-year-old white British male, who lives in an area of approximately average affluence (IMD decile 5). He has a family history of heart attack (father) and has previously taken statins, but managed to come off them with the support of his GP. His blood sugars have been in the pre-diabetic range for 3 years, which is Harry’s motivation for attending the health check: ‘for the last couple of years the doctors been saying I am on the edge of becoming diabetic . . . that was the thing that was pushing me just to make sure I got the health check’. However, when asked if he had heard of the NHSHC he commented ‘Yes, but I wasn’t particularly bothered’.

Practitioner

The health check is delivered by a female white British, PN. She has 2 years’ experience of delivering health checks, but has not received formal training: ‘when I first started, I perhaps would have shadowed somebody doing them’. Despite feeling ‘fairly confident’ in communicating 10-year risk, she conceded that ‘all I have ever been told is anything over 10% needs to be started on a statin. Anything below 10% are fairly low risk, you know, and obviously yeah I discuss that with the patients, but other than that I don’t know a huge amount about that with the QRISK score at all’.

The NHS Health Check

The health check is one of the longest (> 37 minutes). Harry speaks for 29% of the total time, and CVD risk is discussed for > 3 minutes. During the health check, Harry’s 10-year risk is calculated as 14.2%, and blood results showed slightly elevated total cholesterol (5.4 mmol/l) and blood sugars in the pre-diabetes range. Blood pressure was normal. Both the PN’s assessment and our analysis of patient engagement from non-verbal cues of the video-recorded consultation suggested that Harry was positively engaged throughout the CVD risk discussion.

What worked well

The PN explained some factors from which Harry’s 10-year risk was calculated to highlight the need for him to consider going back on to a low dose statin as a means of lowering risk:

So this is your risk of getting heart disease in the next 10 years . . . that puts it to a calculation of your weight, your smoking, or lack of smoking . . . and your cholesterol, that sort of thing . . . it’s come out with a risk score of 14.2%. So ordinarily anything over 10% we would offer a cholesterol-lowering tablet . . . we would recommend that you start back on your statin.

The PN then went on to explain that Harry’s risk was above average for a man of his age:

. . . let me just go back on to that and I explain in a bit more detail. So like I say, erm ,this is coming out that your risk score is 14.2% the average 61-year-old male is 10.1%. So you’re slightly above the average . . .

After additional information was given, Harry appeared to recognise the severity of his CVD risk (‘I don’t like the fact . . . about the risk score’) and moved towards action:

So how do you feel about starting back on it?

I didn’t want to.

OK.

I will if that what it takes.

Harry showed confidence and positive engagement in requesting a discussion of his pre-diabetic status, which was his primary concern, and sought specific details: ‘What does that mean? Does it mean tablets, does it mean injections?’. This allowed the PN to explain and again highlight the importance of lifestyle (as well as medication) to prevent progression to diabetes.

Areas to improve

Harry could not recall his 10-year risk and reported that being shown the video clip ‘brought the disappointment back to me because I was expecting to be better than that’. However, ‘it didn’t make me feel any different . . . it hasn’t made me change my lifestyle at all’. He revealed that he had not yet picked up his statins, but ‘I still got my old ones which I had stopped taking. So I have been taking half of one of them every day’.

Harry’s description of the health check highlighted two issues that prevented more engagement with the PN’s recommendations. First, the PN’s explanation of his above average 10-year risk did not empower him:

Mainly to do with the fact that you got family history of cardiovascular disease . . . So although everything else is pretty good . . . we can’t help what you’re already predisposed to unfortunately, which is your dad.

This left Harry feeling like ‘there was nothing in there that I am not already doing. There was nothing to do’.

Second, a more tailored approach was needed. Harry expressed a desire for specific lifestyle guidance, as he left feeling that ‘without any clear direction, I haven’t got anything to change . . . the nurse said “keep walking the dogs, keep doing the exercise . . . just try and do healthy things, eat your greens and you know have your five a day”. . . do those things which I am doing anyway’. The lack of specific advice was acknowledged by the PN on watching the corresponding video clip of this health check:

I did notice when the diet one came up, I did just put a piece of paper in front of them, and I didn’t read it. So actually in hindsight, perhaps I should have said, ‘you shouldn’t eat this, you shouldn’t eat this, you shouldn’t eat this’ . . . in hindsight, I could have explained a little bit better. Like the statins, he wasn’t overly keen on taking the statins was he?

Finally, Harry thought that it would have been useful to have the opportunity to reflect and then ask further questions ‘In some kind of follow-up, even if it’s a week later on or whatever’. This confirms the potential role of telephone reinforcement.

Summary

This case study describes Ian, a male who was very positive about the overall health check experience, the practitioner and the advice given, and had implemented some lifestyle changes. Ian had excessive alcohol intake and had made beneficial changes. The CVD risk discussion was brief and, ultimately, forgotten; however, in this case it did not matter.

Patient

Ian is a 48-year-old white British male who lives in a relatively affluent area (IMD decile 10) and is active through his job as a postman. He is an ex-smoker (of 4 years) and has been trying to improve his diet to manage his cholesterol, but reports excessive alcohol consumption (> 30 units per week). He has not heard of NHSHC.

Practitioner

This health check is delivered by a female white British HCA who has 6 years’ experience of delivering health checks and attended generic training several years earlier, but cannot remember the details. Although reporting confidence in communicating a 10-year risk, when asked about her understanding of the score, the HCA focuses only on the thresholds:

. . . anything over 10% is a higher risk, erm . . . so then other risk factors need to be taken in consideration like family history and things, erm . . . to see, like I say, if it’s worth going on a statin . . . those people are then would forward onto the doctor for further review or we’ve got a pharmacist.

The NHS Health Check

The health check lasts just under 19 minutes and Ian speaks for just 18% of the consultation. CVD risk is discussed for just over 1 minute. The HCA collects information on Ian’s lifestyle and offers advice before his 10-year risk is discussed. Ian’s 10-year risk is 4.5% (low risk), but he is noted to have a slightly elevated total–HDL cholesterol ratio and excessive alcohol intake. Discussions around dietary changes to manage cholesterol and reducing alcohol intake account for the majority of the health check.

What worked well

Ian thought it would be longer (‘I’d thought I’d been in there half an hour, at the most I were in there 10 minutes’), yet felt that the HCA ‘was excellent . . . she gave me some great advice’. The way in which the HCA communicated was clearly appropriate for him:

It weren’t sterile . . . we had a chat . . . it was like visiting a friend.

Ian’s alcohol consumption identified a risk of alcohol-related harm. The impact of the corresponding discussion was clear during Ian’s VSR interview: ‘I didn’t think it was a problem . . . and maybe it did become a problem’. Ian made immediate changes, ‘as soon as I talked to her I’d stopped drinking for a fortnight’. Although he had since had a drink, he had also made dietary changes to address the elevated cholesterol ratio through eating more fish (albeit tinned tuna, rather than the recommended oily fish) and vegetables: ‘I actually feel healthier’.

The health check ended with the HCA saying that ‘we are going to set some goals for you’, and Ian had responded. This was something the HCA specifically mentioned in her VSR: that she tends to set goals with the patient’s agreement:

I sometimes find if I ask the patient to set a goal they sort of sit there . . . they don’t really know what goals to set, so I do find that I do, sort of instigate . . . but they agree . . . It’s rare that I’ll instigate a goal and patients don’t agree with it.

In the light of the low 10-year risk, the HCA tried to emphasise the importance of prevention in general:

As we get older our percentage naturally goes up because we are older, so when we are a young age because that is a young age, that’s my age, so when we are of that age we have to look after yourselves and prevent, it’s all about preventative measures OK.

Areas to improve

When 10-year risk was conveyed, it was quickly given as a relative and then absolute percentage risk score, without much explanation or checking of Ian’s understanding:

OK, so, for your age group, the risk of developing heart disease in the next 10 years is 3.5%. Your risk is 4.5%, so it’s slightly up. Erm, is still a low risk percentage when no major concerns of that but we do need to say to you is that obviously you just need to make sure you keep active, which you are . . .

Communication of the risk score was brief, and Ian did not retain anything from the information given. During the interview he indicated that this information was not important to him and he had made changes without understanding this: ‘it don’t really bother me to be honest . . . don’t really register with me’.

The lifestyle goals could have been more tailored. They combined specific behavioural goals and general outcome goals, covering three lifestyle behaviours, and included an inappropriate exercise-related goal, which has not been implemented:

‘cos I’m a postman you see, so I was shocked when she says err, cos I walked 10–15 miles a day . . . with a big bag on me back and everything, and I was surprised when she said . . . ‘do you go home and do any gardening?’ . . . I just wanna go home and sit and chill.

Finally, Ian thought that it would have ‘helped turning the computer round and showing me the information . . . ’cos visual has always been better than audio for me for everything’.

Summary

This case study describes the health check for a patient, Jessica, who had already made lifestyle changes following an ‘episode . . . they thought might have been a mini stroke’ 12 months prior (this was not confirmed as this would preclude her from NHSHC). Jessica’s engagement with, and positive experience of, the health check appeared to result from the reassurance that it provided around the success of steps she had already implemented.

Patient

Jessica is a 64-year-old white British female who lives in an area of approximately average affluence (IMD decile 5). She experienced an episode the previous year, which was initially suspected to be a ‘mini stroke’ (but was not confirmed). She was prescribed statins at that time, but ‘thought, “I don’t want to do this” ’ and consequently made dietary changes, reducing her cholesterol and successfully losing weight. She had not heard of NHSHC before and assumed it was a general health check, and wondered why it was focused on CVD: ‘why can’t you go into everything at the same time?’.

Practitioner

The health check is delivered by a white British, female HCA, who has 4 years’ experience of delivering health checks and received some training when she started, which she describes as: ‘slides and screen presentations . . . talked through it erm and then it was moving on to how to use the [point-of-care] machine . . . it was pretty straightforward really’.

The NHS Health Check

This health check lasts < 18 minutes, during which the HCA speaks for over 70% of the time. PoC testing confirms that Jessica has reduced her total cholesterol (from 6.7 mmol/l to 6.1 mmol/l) and total–HDL cholesterol ratio (from 3 to 2.8). CVD risk is discussed for about 1 minute, which shows a reduction from a measurement (recorded several years earlier): ‘you were at 10% just before, it is now 5%, so you have halved the risk in that time’. The HCA congratulates and encourages Jessica to keep going with changes she has made to lower her cholesterol and recommends that she increase her activity levels (as she is classified moderately inactive):

The one thing that will help you with that, is being a bit more physical, get doing something that is a little bit more cardiovascular, even if it’s upping your walking and doing it as a power walk, or doing hills, on an incline, it gets you out of breath.

What worked well

Overall, Jessica thought that the HCA was ‘very good’. At the start of the health check, the HCA made clear to the patient that it was ‘all to do with cardiovascular disease, so what we are looking for is how at risk you are of developing cardiovascular disease within the next 5 to 10 years’. This provided useful context for the subsequent discussion of risk factors and management.

Jessica appreciated feedback and reassurance regarding changes already made, and was pleased to see that ‘the work had been paying off’. She remembered clearly the main point to address and had made further small changes to her commute to work to incorporate moderate-intensity activity into their day: ‘I go to [work] on the school bus . . . it picks me up from outside the house, but I’ve started to go out a few minutes earlier and walking up the hill [laughs] . . . and meeting it up the hill [laughs] . . . it’s just a few minutes a day . . . might make a difference’.

She also appreciated the written report, as ‘you can’t take everything in straight away, so it’s nice to have that printout’ to ‘go back and have a look’.

Areas to improve

The 10-year risk score was mentioned briefly, but not explained. Jessica could partially recall the information as ‘. . . 5% over the next 5 years’. She had retained the correct percentage, but the timescale was incorrect (reflecting the HCA’s incorrect description, above) and she could not demonstrate a clear understanding of what it meant for her.

Patients attending already motivated to make changes

Health checks with patients who were already motivated to make lifestyle changes or, as in some cases, had started to implement changes had positive outcomes regardless of how CVD risk was communicated. These patients attended with a specific focus (e.g. behaviour change in mind), which provided the practitioner with an obvious focus to tailor risk management.

Cardiovascular disease risk communication

Risk management

Post-health check telephone reinforcement

Telephoning patients soon after their health check seemed to have a range of benefits: allowing patients time to reflect on the health check discussion and identify questions or areas of uncertainty; further explaining CVD risk, answering questions and confirming patient understanding; confirming recommendations for risk management; and making specific plans where appropriate (e.g. referrals, follow-up appointments, further measurement).

QRISK2

Communicating CVD risk is central to the NHSHC,17,19 but formed a small part of consultations (< 2 minutes on average). When using QRISK2, practitioners often perceived 10-year risk as a way to identify patients for routine medical follow-up (if risk ≥ 10%), rather than to initiate discussion around patients’ risks and risk management. Consequently, they tended to relay the percentage 10-year risk score and whether it was low (< 10%), medium (10–19%) or high (≥ 20%), without expanding to make clear the relevance to the patient. Minimal patient responses to risk information, particularly for 10-year risk, with a single-word response (e.g. ‘yeah’, ‘OK’) made it difficult for practitioners to gauge patient understanding, which would then allow them to address misunderstandings or concerns. This is important given the evidence that engagement in primary care consultations is linked with patients’ understanding of the consultation and confidence, which has subsequent benefits for recall of information and adherence to recommended treatments. 79

Practitioners reported being confident in communicating 10-year risk, but mainly in application of thresholds (low/medium/high risk). Understanding and confidence to engage in a discussion with patients around what risk means for them was less evident. Perhaps as a result, patients often did not understand CVD risk information, particularly 10-year risk. Again, this is in line with evidence that practitioners who lack confidence are less likely to invite patient engagement in primary care consultations. 79 Our findings confirm the findings of qualitative studies, which have also indicated that practitioner and patient understanding of percentage CVD risk is limited,21 and that practitioners find it difficult to explain CVD risk,26,36–38 resulting in patients leaving health checks unable to recall or with limited understanding of their 10-year risk. 21

JBS3

Practitioners in the JBS3 group were free to use all features of JBS3. However, the minimum requirements in this study were that practitioners communicate heart age and event-free survival age (on the Healthy Years Screen), and manipulate at least one of these risk scores to demonstrate the effect of risk factor changes. QRISK2’s 10-year risk was also delivered as a requirement of the national programme.

Heart age

Practitioners showed patients the on-screen output for heart age, and this was generally preferred to other risk scores. Analysis of health checks and practitioner interviews revealed greater confidence with communicating heart age than with other risk metrics (10-year or event-free survival age). The finding that patients had better understanding and recollection of heart age (than of 10-year risk or event-free survival age) is consistent with the findings from studies of online heart age80 and heart age provided to NHSHC patients (in the form of a risk report). 41

Interview data suggested that the comparison inherent in the concept of heart age (chronological age compared with predicted heart age) allowed patients to quickly appraise whether or not to be concerned by their level of CVD risk. In turn, we observed more pronounced patient responses than for other risk scores – some were ‘shocked’, others were ‘pleased’ or ‘reassured’. Mixed responses depending on results are understandable and have been observed elsewhere. 41,81 Moreover, some patients questioned the credibility of heart age. There is evidence that heart age can prompt consideration of lifestyle change, regardless of the results (younger or older than chronological age) and its perceived credibility. 42,81 This was not necessarily supported by our data, which suggested a dismissal of heart age where it was not seen as credible. This would need further exploration. In any case, variation in preference is in line with the literature24 and highlights the importance of practitioners having different types of risk information (score, visual, analogies, risk manipulation), which they can tailor to patients’ needs and preferences.

In addition to the heart age score, the visual presentation of heart age appeared important for impact and recollection. QRISK2 and 3 also generate heart age, but do not offer a similar visual output. Such visual representations of risk were perceived favourably among our sample, and have been preferred for promoting risk-reducing behaviour. 44

Finally, it is important to be clear about the role of heart age as a form of lifetime risk. The 2014 JBS3 report34 stated that ‘lifetime risk measurement is an adjunct to the estimation of 10-year absolute risk levels. It is intended not primarily as a guide to decisions about drug initiation, but rather as a way of allowing an individual to understand the lifetime risk’. Heart age was not designed to guide clinical decisions,42,82 but our data support its use in health check as a means of promoting patient understanding of CVD risk and a useful prompt for patients to consider lifestyle changes. 42,81

Event-free survival age

Practitioners’ description of the average age to which patients could be expected to survive free from a CVD event was often unclear and sometimes incorrect, which illustrated a lack of understanding of event-free survival. There was evidence that this score was misinterpreted by some practitioners and patients as expected survival age (i.e. age of death). Sometimes the average event-free survival age was relayed to patients correctly during the health check through reading the score from the screen, but subsequent VSR interviews (in the same practitioner) revealed a lack of understanding.

There is a lack of evidence exploring this form of lifetime CVD risk score with which to compare our findings. It appeared from our data that the lack of thresholds (such as those in 10-year risk) or inherent comparison (such as that in heart age) made it difficult for practitioners and patients to interpret event-free survival age as low/medium/high or good/bad. In turn, this risk score generally did not engender much response from patients, who often could not remember it. The graphical representation of event-free survival age was in the form of a visual analogue scale (see Figure 1) with text showing the age. This was not the problem. Patients appeared able to read the result from the display, but often misinterpreted what the figures represented. To be more widely understood, practitioners need training to enable them to better explain event-free survival and communicate its significance unambiguously.

The use of, and response to event-free survival age underline the aforementioned importance of clarity of information such that patients understand and have the confidence to engage in consultations,79 which can then lead to a patient-centred experience. 83 Our data suggest that event-free survival age should not be recommended for use without specific and more substantive training than that provided to practitioners in the RICO study. We provided a verbal explanation during practice initiation visits, written materials and a short video tutorial on how to use JBS3, asked practitioners to practise using JBS3 in health checks in advance of data collection, and were available to answer practitioner questions throughout (in the practice during data collection or by telephone at any time). The decision not to include more intensive training was to allow us to understand how practitioners would use this tool if it were made available for use in NHSHC and has identified an obvious training need.

Cardiovascular disease risk score manipulation

Practitioners liked the risk score manipulation function in JBS3 and thought that it was impactful. Heart age was often the chosen metric for risk manipulation, reiterating practitioner preference for heart age. However, often, it was not used to its potential. The importance of user competence with interactive graphical displays of risk has been noted elsewhere. 23 In the RICO study, practitioners tended not to select the most appropriate risk factor to exemplify the potential risk-reducing benefits of intervention; for example, in non-smoking patients with raised cholesterol, practitioners defaulted to showing increases in risk if they did smoke, which lacked relevance to the patient. Therefore, despite the positive perceptions, it is likely that limited user competence undermined the potential impact of this function in JBS3.

Heart age and risk score manipulation were perceived positively by patients and, if used appropriately, can be used to confer patient understanding of the risk and its severity/relevance. Visual presentation and risk score manipulation offer tools that can accommodate the range of patients, and also to make the important link between lifestyle change and overall CVD risk.

Both groups

Regardless of CVD risk calculator, practitioners rarely asked if patients understood their risk. Patient questions about CVD risk were also rare, but more commonly observed in the JBS3 group (32.0% vs. 12.3%). CVD risk communication and the consultations overall were practitioner dominated, with patients speaking for less than one-quarter of health check time. Processes relating to measurement (CVD risk and lifestyle) and relaying information appeared to take precedence over creating dialogue with patients, who said little in the majority of health checks.

Practitioners usually relied on patient reactions to risk information to gauge understanding. This was limited by a general lack of response, which, unless followed up by the practitioner, can lead to patients failing to understand or recall their risk information. Even in the case studies of patients with positive intentions or outcomes, practitioners rarely explained what CVD risk meant for patients, or checked their understanding (but it mattered less for those patients who were already aware of their need for lifestyle changes, or were motivated to change). Not knowing how the patient is receiving the information limits practitioners’ opportunities to tailor risk information to instigate change.

In turn, there was a lack of tailoring in CVD risk information. Some practitioners admitted to delivering the same information to all patients, recognising that some would understand and some would not. Others said that they adapted the information to the patient. But as patients were not asked if they understood the information presented, there was little opportunity for such adaptation. This is an important limitation. There is no single ‘correct’ way to communicate risk. It depends on individual preferences and understanding, which differ with education, numeracy and personality traits, such as optimism. 24 Tailoring is therefore critical, but impossible unless practitioners know how patients receive the information.

Overall, patients in both groups were given too much information within a short space of time, some of which was irrelevant, limiting their capacity to absorb or retain it. Some expressed an appreciation of written information to take away and consider in their own time. Hawking et al. 41 interviewed patients who were given a personalised risk report to take away from their health check and found that this was mostly well received, often for reassurance, but with potential to motivate lifestyle and discussions and changes with family and friends.

As recognised in the NHSHC competence framework,19 our data confirm that providing patients with a CVD risk score alone is insufficient. There is a challenge for practitioners to facilitate understanding so that patients can appraise their CVD risk information in terms of the severity of the risk (patients with elevated risk recognise it) and the vulnerability to risk (patients not currently at high risk recognise the need for preventative action).

JBS3’s visual risk presentation and heart age

The visual presentations of risk, such as those in JBS3, are recommended for use in health check. In particular, the heart age output was widely liked and considered impactful. But having a range is useful to accommodate patient preference.

If applied appropriately, manipulation of CVD risk should also be considered as a way to demonstrate to patients the link between their behaviour (and any related recommendations) and their overall CVD risk, but this would need further practitioner training, and observation of application in practice, before effectiveness testing.

Ask patients and tailor

When delivering CVD risk information, practitioners should ask patients about their understanding and how they feel about the information. Knowing how the information is being received would enable the practitioner to provide follow-up information or present risk in different ways to foster patient understanding and appraisal of risk and coping responses (i.e. risk management). This, in turn, can lead to risk-reducing intentions.

One practical approach to operationalise such a change would be for practitioners to record that they have checked patient understanding within the NHSHC template (within the practice system). However, adding to the already substantial list of administrative and clinical tasks could lead to this becoming another process for which there is insufficient time (see Time). Rather, this should be fundamental to the consultation approach, moving away from information provision, and towards shared decision-making. 51 We can consider this in terms of the characteristics of shared decision-making identified by Cipher et al. :83 (1) ‘accurate, objective and comprehensible information’ (i.e. CVD risk); (2) ‘a practitioner who is present and proficient in communication and able to individualise data to a particular situation’; and (3) that the ‘individual’s values, goals, informed preferences, and concerns are incorporated into the communications’. These recommendations identify the need for understandable and tailored CVD risk information and a practitioner who engages the patient in dialogue, which can then lead to discussion of risk management, as part of a two-way interaction.

Time

As discussed in Chapter 3, the time allocated to health checks and how that time is spent should be revisited in the context of what practitioners are expected to achieve.

Patients in our sample reported being provided with too much information, in too short a time. The length of appointments that practices allocated to health checks ranged from 15 to 30 minutes, yet the actual duration of practitioner–patient interaction ranged from 6.8 to 38 minutes (60% lasted < 20 minutes). Health check practitioners have a range of administrative and clinical tasks to complete within this appointment time: CVD risk assessment involving measurement of (and entry of data for) weight, blood pressure and, sometimes, cholesterol through PoC testing; lifestyle assessment (physical activity, alcohol, diet); explaining to patients their CVD risk score(s) and what it means; and patient-centred discussion of risk management to prompt risk-reducing behaviours.

Our data indicated that the health checks were process driven, rather than content focused, and some of the observed features of delivery reflected practitioners working through a series of administrative or clinical processes (e.g. interrupting the conversation, presenting information too quickly, or using inappropriate language). 84,85 Such behaviours reduce the likelihood that information will be understood and retained. 85 This suggests a need to revisit the health check to focus on content and the quality of interactions.

Consultations that involve shared decision-making take time. 85 NICE has endorsed ‘brief’ behaviour change interventions that take several minutes. The NHSHC practitioner competence framework19 refers to the ‘Brief intervention/signposting/referral’ part of the health check (i.e. risk management) as ‘more than a simple communication of information; the person signposting must be able to engage the client in the choice and communicate in a manner that will maximise the potential that the client will take up the agreed action and sustain’ (p. 23; contains public sector information licensed under the Open Government Licence v3.0). Therefore, even if the risk management discussion is brief, it should be delivered by practitioners with appropriate training, who are able to engage patients in designing their own risk management plan, thus tailoring it to patient needs. 86 However, NICE also recommends ‘extended brief interventions’, lasting ≥ 30 minutes, and potentially lasting several sessions;86 and motivational interviewing, which also features in the NHSHC competence framework, can require ≥ 15 minutes to be effective.

A short, practitioner-dominated health check might be sufficient when patients attend with motivation and ideas/actions about changes they need or want to make. If this appraisal has not taken place in advance, it is the role of, and challenge for, the practitioner to present CVD risk information that facilitates this patient appraisal, helping them towards risk-reducing intentions and behaviour. In this case, a short, practitioner-dominated health check in which CVD was risk neither well explained nor understood, is unlikely to be effective.

Therefore, our data suggest that:

• some health checks appear too short to accommodate all health check tasks in a patient-centred manner

• more time should be allocated to explaining CVD risk, allowing practitioners to check patient understanding and to tailor risk information using the appropriate tool for that patient

• sufficient time should be allowed for practitioners and patients to co-produce a specific, relevant and meaningful risk management strategy.

This could be achieved through increasing the overall time available to health check or by reducing the number of administrative and clinical tasks required, and practitioners should be supported with appropriate training (see below).

Telephone reinforcement

Follow-up telephone calls to patients soon after the health check, by the practitioners who delivered the health check or other clinical staff with the required knowledge and skills, should confer a range of benefits:

• patients have time to reflect on the health check and identify questions or areas of uncertainty

• practitioners can provide further explanation of CVD risk, answer questions and reaffirm patient understanding

• practitioners can confirm recommendations for risk management, making specific plans where appropriate (e.g. referrals, follow-up appointments, further measurement).

Such telephone reinforcement has been used in physical activity interventions to improve participant perceptions of the target behaviour87 and promote maintenance of that behaviour if used as part of an ongoing negotiated follow-up. 88 This has resource implications that are likely to be justified, but would need to be explored. For patients who do not require it, the call could be short and the resource implications modest. Others will need more time to revisit health check content and agree risk management.

Practitioner training

An overarching theme that cuts across all of the above implications for practice is the need to support practitioners (HCAs and PNs) with appropriate and specific training to ensure that practitioners know what they need to do in health checks, that they are confident and have the necessary knowledge and skills, and that they are supported with ongoing training to promote continued improvements in delivery. The Best Practice Guidance17 and practitioner competence framework19 set out high expectations. As the aim is to have consultations that are patient centred, for which shared decision-making has been described as the ‘pinnacle’,89 patients should be involved in the co-design of practitioner training. 90 Therefore, findings from this study highlight two training needs. First, there is a need for training to address the apparent knowledge deficit around different types of CVD risk. Here and elsewhere,21,91 it is clear that those delivering health checks either do not receive any training specific to conduct CVD risk measurement and how to communicate risk scores to patients or receive generic training that is focused on health check processes (e.g. completing the template, operating the PoC testing machine). Second, PNs and HCAs do not receive training in communication skills, such as those included in medical training. Our data have uncovered a previous ‘unknown unknown’ about a lack of communication skills in the context of real-world patient-centred risk communication and risk management discussion. The communication skills content of undergraduate and postgraduate nurse training and training of HCAs should, therefore, be evaluated in relation to this need.