Engagement with advice to reduce cardiovascular risk following a health check programme: A qualitative study

Abstract Background The success of a cardiovascular health check programme depends not only on the identification of individuals at high risk of cardiovascular disease (CVD) but also on reducing CVD risk. We examined factors that might influence engagement and adherence to lifestyle change interventions and medication amongst people recently assessed at medium or high risk of CVD (>10% in the next 10 years). Method Qualitative study using individual semi‐structured interviews. Data were analysed using the Framework method. Results Twenty‐two participants (12 men, 10 women) were included in the study. Four broad themes are described: (a) the meaning of ‘risk’, (b) experiences with medication, (c) attempts at lifestyle change, and (d) perceived enablers to longer‐term change. The experience of having a health check was mostly positive and reassuring. Although participants may not have understood precisely what their CVD risk meant, many reported efforts to make lifestyle changes and take medications to reduce their risk. Individual’s experience with medications was influenced by family, friends and the media. Lifestyle change services and family and friends support facilitated longer‐term behaviour change. Conclusions People generally appear to respond positively to having a CVD health check and report being motivated towards behaviour change. Some individuals at higher risk may need clearer information about the health check and the implications of being at risk of CVD. Concerns over medication use may need to be addressed in order to improve adherence. Strategies are required to facilitate engagement and promote longer‐term maintenance with lifestyle changes amongst high‐risk individuals.

However, other commentators argue that attempts to increase risk perceptions will rarely result in behaviour change. 14 The way in which risk and other health information is presented to patients may influence adherence to medication and lifestyle change advice, and there is probably potential for improvements in delivery of information and advice to enable more informed decision making. 15 A meta-analysis of experimental studies suggested that the impact of risk information on behaviour change is moderated by the extent to which patients believe that behaviour change will reduce risk and their perceived ability to change. 16 Therefore, accompanying risk information with other behaviour change techniques may result in risk-reducing behaviour.
The health check programme provides a unique opportunity in terms of addressing multiple risk factors through risk communication, behaviour change interventions and/or medication provision that, if successful, might reduce the risk of several diseases.
Although previous studies have examined patients' experiences with the programme as a whole, focusing on higher-risk patients' experience is important for the programme's outcomes. The aim of the present study was to examine higher-risk patients' longer-term impressions of feedback given to them during the health check about their health, including risk factor levels, and to explore this and other factors associated with engagement in suggested risk-reducing interventions, including medication use. We aimed to interview patients up to six months following their health check in order to identify factors related to their engagement with risk-reducing interventions in the longer term.

| Study design
Qualitative study was conducted using semi-structured interviews with people who were assessed at medium to high risk of developing CVD in the next ten years during the health check. The study was part of a wider service evaluation of UK's NHS Health Check programme in South East London and was registered on the database of the Research Development Centre for South East London NHS Organisations at Southwark Public Health Department (RDLSL2047).

| Study participants and sampling
Consistent with the eligibility criteria for the NHS Heath Check, eligible participants were aged between 40 and 74 years. Participants were eligible if they had received a health check in the last six months and were assessed at medium to high risk (>10% risk) of developing CVD in next 10 years and were registered with general practices across two South East London boroughs: Lewisham and Lambeth. CVD risk assessment was done using QRISK2 score. 17 A convenience sample was employed, where potential participants were identified by general practice staff from the results of their health check and invited to take part in the study by their general practitioner (GP). Out of 353 patients who were invited, 26 agreed to participate in the study. Four patients were not recruited for logistic reasons.

| Data collection
Topics for the interview (Supplementary material 1) were drawn from the literature on uptake and adherence to lifestyle change interventions and medication use. Items in the schedule were influenced by the Theoretical Domains Framework (TDF) which is drawn from models to explain behaviour change to understand factors that influence implementation of interventions. 18 The TDF covers a set of domains comprising the main evidence-based factors influencing behaviour change. 18 As we were examining a range of behaviours, however, rather than focusing on one aspect of behaviour change, we drew also from other theories, including those concerning medication adherence (eg Clifford et al 19 ) and access to health care. 20 Interventions targeting multiple behaviours need to take account a wide variety of factors including beliefs, social influences and environmental context and resources.
Interview items were generated and agreed by members of the research team. The interview covered participants' experience of having a health check, their understanding of their personal risk of CVD, and their feelings and attitudes about the feedback and advice given to them about lifestyle change and medication. Interviewees were asked about any changes they had made to their lifestyle, their experience of medication, where relevant, and about influences on adherence to these changes. The interview schedule remained the same throughout the interview process although the prompts and probes differed for each individual depending on their responses.
Sociodemographic data were also collected at the time of interview, including age, sex, ethnic group and Index of Multiple Deprivation (IMD) rank. The English indices of Deprivation provide a measure for relative deprivation for small areas that are measured based on seven domains: income, employment, health deprivation and disability, education, housing, crime and living environment. 21 The interviews were all undertaken by one researcher experienced in qualitative interviewing and research (CB). Participants had been sent an information sheet about the study by their GP, explaining why they had been invited. Interviews were conducted face to face at home, at the general practice, or over the telephone, according to the participant's preference. Data collection stopped when no further participants could be recruited within the timeframe of the study.

| Data analysis
Interviews were digitally recorded with the participant's consent and fully transcribed. The analysis was conducted after completion of the interviews. A qualitative analysis based on the Framework method was used. 22 The Framework method can be adapted for use with deductive or inductive analyses, and it is not aligned with a particular theoretical approach. 23 This method adopts a 'case by theme' approach which involves several stages: transcription, familiarization, coding, sorting and charting the data according to important or dominant issues and themes and data interpretation. 22 Our themes and codes were not pre-selected in a deductive way; however, they were generated from the data using coding and refinement of themes. One member of the research team (BK) coded each transcript, giving a label to segments of the data that appeared significant or important and a random sample coded by another member of the team (CB). The two researchers met to compare coding for the first three transcripts and to agree on codes to be applied to subsequent transcripts. Categories were derived from grouping codes together to produce an analytic framework. The analytic framework was refined in an iterative way during the analysis via discussion with the research team. Finally, team meetings were conducted to ensure agreement about the themes and to assess whether the data were representative of the themes.

| The meaning of 'risk'
Whilst most study participants recalled a general discussion of risk, many could not recall specific information about this or their individual risk factors. There was evidence that whilst people did not recall their precise risk score, they did understand that their CVD risk was higher than it should be: Well there were two nurses and they were pretty nice, they chatted a lot, we had quite a good laugh, but no I don't remember much about it. They said my cholesterol was somewhat elevated. And my weight was a little higher than it should be.
Did they give you any information about your risk of cardiovascular disease?
Yes they did.

Do you remember what that was?
No I don't. (ID17, Male, aged 54, CVD risk >10%, 3rd IMD quintile) One man understood the implications of the information given regardless of recalling his risk score: Well he basically said, I forget the exact words, 'you're not in the danger zone, but you're sort of heading that

| Experiences with medication
Those who had been prescribed statins for high cholesterol reported a variety of side-effects, and some were prescribed several different types of statin in an effort to overcome these. Some did not adhere to the medication or decided not to take it at the outset. Some felt it was preferable to try lifestyle changes first before 'resorting' to medication.
People talk a lot about statins. But it always seems to me, and I'm not a medical person, that you want to put off being on medication for as long as possible.
Frankly you can do all the other avenues first because I think once you're on it you can't really get off again.
(ID16, Male, aged 54, CVD risk >10%, unknown IMD quintile) Some of those who were on regular medication expressed acceptance of the necessity of this as well as some concern over long-term use.

| Attempts at behaviour change
Regardless of their appreciation of the significance of their CVD risk, the predominant message recalled by most of those interviewed related to making changes to diet or exercise to improve health.

| Perceived enablers to longer-term change
Free gym access and exercise classes were viewed very positively and appeared to increase the likelihood of attendance.  Although wanting to make lifestyle changes, some participants reported medical, physical and financial barriers to change, as has been found in previous studies. 10 Patients reported difficulty changing behaviour because of stressful circumstances or work-related constraints. Although previous studies suggested that older adults felt that changing their lifestyles at their age was unnecessary, 10 older participants in the current study were motivated to change for a healthy retirement. Family history of CVD acted as a double-edged sword in relation to behaviour change. Whilst for some a family history of CVD was a source of motivation to change, others felt that changing their behaviour will not alter their inevitable risk of CVD.
This result echoes with previous findings that individuals with family history of heart diseases do not think that they can reduce their personal risk of developing heart diseases by changing their behaviour. 34 Participants at high risk explained the importance of increased support and follow-up to sustain behaviour change, either by family and friends or by health-care professionals. Ismail and Atkin conducted a study to understand the experience of patients who went for the health check, regardless of their CVD risk level, and reported their perspectives on behaviour change. 35 Participants in their study expressed the need for a proactive approach in advocating for lifestyle change by their health-care providers. 35 Social support appears to be a key factor not only for the initiation of behaviour change, but also for facilitating long-term behaviour change maintenance. 36 Previous evidence has suggested that health-care professionals implementing the health check believed in the importance of providing continued support and follow-up to encourage and maintain behaviour change. 37 However, due to time constraints and workload pressures on primary care professionals this is not always feasible. 37 Therefore, making referrals to external lifestyle interventions is believed to provide high-risk patients with the support they need.
Lifestyle support services are in place to facilitate behaviour change and reduce barriers as part of the health check programme.
However, these services are not always accessible or reliable. 10,38 Many of those we interviewed were referred to lifestyle change interventions including exercise and weight loss groups. However, there were often barriers to joining these interventions such as long waiting lists, distance from home and the timing of classes. The current study was conducted in the early stages of the programme im-

| Strengths and limitations
This study is amongst the first to examine adherence to behaviour change interventions and medications amongst higher-risk individu- This may have deterred some who felt some stigma attached to their lifestyle choices. No other means of recruitment was available, however, and we did get a low response rate, although this level of recruitment is fairly standard in this area of South East London. We were unable to recruit further participants during the timeframe of the study. If we had been able to recruit for longer, we may have ended up with a more diverse sample and captured a wider range of views.
It is possible that data collection and interpretation were influenced by the research team's background in health psychology and public health. Analysts from different disciplines may have developed an alternative analytic framework. Finally, face-to-face interviews may be preferable as the interviewer can be aware of non-verbal cues.
However, face-to-face interviews to discuss behaviour change can produce social desirability bias. 42 The participants might have given responses that they perceive as socially acceptable. In our study, the quality of the telephone interviews was good and it may have been easier for some participants to discuss issues, such as weight, without concerns about being judged by the interviewer.

| CON CLUS IONS
This study suggests that people are generally motivated and willing to change their behaviour following the health check, regardless of their understanding of their personal risk of a cardiovascular event.
Easy access to services to support behaviour change, however, is variable. Concerns over the side-effects of medications need to be addressed in order to improve adherence. Strategies are required to facilitate long-term behaviour change maintenance, possibly through greater provision and support of lifestyle change services.

ACK N OWLED G EM ENTS
The authors would like to thank all participants for their participation in this study. This research project was supported by a grant from the 'Research Center of the Female Scientific and Medical Colleges', Deanship of Scientific Research, King Saud University.

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest regarding this research.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data are not publicly available due to privacy and ethical restrictions.