Appendix C: The evidence

This appendix lists the evidence statements from four reviews, a cost-effectiveness review and a primary research study provided by external contractors (see appendix A) and links them to the relevant recommendations. (See appendix B for the key to quality assessments.)

The evidence statements are presented here without references – these can be found in the full reviews (see appendix E for details).

The appendix also sets out a brief summary of findings from the economic analysis.

The four reviews (reviews 1–4), the primary research study (review 5) and the cost-effectiveness review (review 6) are:

  • Evidence reviews:

    • Review 1: 'Prevention of cardiovascular disease at population level (Question 1; phase 1)'

    • Review 2: 'Prevention of cardiovascular disease at population level (Question 1; phase 2)'

    • Review 3: 'Prevention of cardiovascular disease at population level (Question 1; phase 3)'

    • Review 4: 'Barriers to, and facilitators for, multiple risk factor programmes aimed at reducing cardiovascular disease within a given population: a systematic review of qualitative research'

  • Primary research:

    • Review 5: 'Population and community programmes addressing multiple risk factors to prevent cardiovascular disease: A qualitative study into how and why some programmes are more successful than others'

  • Cost-effectiveness review:

    • Review 6: 'Prevention of cardiovascular disease at population level (Question 1; cost-effectiveness)'

Evidence statement R3.E1a indicates that the linked statement is numbered E1a in review 3. Evidence statement R5.12 indicates that the linked statement is numbered 12 in the primary research study (review 5). Evidence statement ER1 indicates that the linked statement is from expert report 1 (see additional evidence). Evidence statement CE1 indicates that the linked statement is numbered 1 in the cost-effectiveness review (review 6). '

The reviews, economic analysis and additional evidence are available online where a recommendation is not directly taken from the evidence statements, but is inferred from the evidence, this is indicated by IDE (inference derived from the evidence).

Where the PDG has considered other evidence, it is linked to the appropriate recommendation below. It is also listed in the additional evidence section of this appendix. This includes evidence used to develop other NICE guidelines and guidance.

Recommendation 1: evidence statement R3.E1m; additional evidence ER3

Recommendation 2: evidence statements R3.E1b; R3E1h; R3.E1i; R3.E1j; R3.E1k; additional evidence ER10

Recommendation 3: additional evidence ER9; ER10

Recommendation 4: additional evidence ER4; ER6; CG43 (1)

Recommendation 5: additional evidence ER4

Recommendation 6: additional evidence ER3; ER4; ER6; CG43 (2)

Recommendation 7: additional evidence ER1; ER10; ER13

Recommendation 8: additional evidence ER10; ER13

Recommendation 9: additional evidence ER1; CG43 (3)

Recommendation 10: additional evidence CG43 (4)

Recommendation 11: evidence statements R4.18a; R4.18b

Recommendation 12: IDE

Recommendation 13: evidence statements R3.E2a–f, R4.2a–c, R4.17a–d, R4.18a–c, R4.19a, R4.20a, R4.21a–c, R4.22a, R4.23a–b, R4.24a–d, R4.25a–d, R4.26a–d, R5.2

Recommendation 14: evidence statements R3.E1a–o, R3.E2a–f, R3.E3a–c, R4.14a–b, R5.5, R5.7, R5.10, CE 1–9; additional evidence ER 7, ER 8, ER 13

Recommendation 15: evidence statements R4.3a, R4.3b, R4.10a–d, R4.11a, R4. 11b, R4.12a, R4.13a–d, R4.15a, R4.15b, R5.2, R5.4, R5.6, R5.7

Recommendation 16: evidence statements R4.1b, R4.7c, R4.10d, R5.2, R5.3

Recommendation 17: evidence statements R4.16a–c, R5.12

Recommendation 18: evidence statements R3.E1n, R3.E3c, R4.4a, R4.9f, R4.18a, R4.18c; additional evidence ER 1

Recommendation 19: evidence statements R3.E5c, R4.18a–c; additional evidence ER 4

Recommendation 20: evidence statements R3.E1a–m, R3.E3c, R4.4a, R4.9e, R4.18a, R4.18b, R4.25e; additional evidence ER 3, ER 9, ER 10

Recommendation 21: evidence statements R4.1b, R4.3a, R4.3b, R4.7a–c, R4.8a–d, R4.16a–c, R4.17a–d, R5.3, R5.7, R5.8, R5.9, R5.11, R5.12; additional evidence ER 5, ER 7, ER 8

Recommendation 22: evidence statements R4.3a, R4.18a, R4.18c, R5.10; additional evidence ER 1, ER 5, ER 7, ER 8, ER 12, IDE

Recommendation 23: evidence statements R3.E3a, R3.E3c, R4.18a, R4.18b; additional evidence ER 3, ER 4, ER 9, ER 10

Recommendation 24: evidence statements R4.9e; additional evidence ER 3, ER 9, ER 10

In addition, the approach taken in recommendations 1 to 12 is supported by the following evidence statements: R4.3a, R4.3b, R4.4a, R4.18a, R418b, R418c, R5.9 and additional evidence ER10, ER11, ER12, ER13.

Evidence statements

Please note that the wording of some evidence statements has been altered slightly from those in the evidence reviews to make them more consistent with each other and NICE's standard house style.

Evidence statement R3.E1a

CVD mortality and morbidity: Limited evidence from 3 out of 38 programme evaluations using different summary effect measures demonstrate a mixed effect of multiple risk factor interventions (MRFI) on CVD mortality (the majority of programmes were beneficial) with two controlled before-and-after (CBA) studies demonstrating a net decrease in CVD mortality and one randomised controlled trial (RCT) demonstrating no net change. Limited evidence from 4 out of 38 programme evaluations, using different summary effect measures, demonstrate a mixed effect of MRFI on CVD morbidity (the majority disbeneficial) with one CBA study and one RCT demonstrating a net increase in morbidity and one RCT demonstrating no net change in morbidity. The effect of one programme on morbidity and mortality is unclear.

Evidence statement R3.E1b

Blood cholesterol: A large body of evidence from 15 CBA studies and 5 RCTs demonstrates a mixed direction of effect (majority of programmes beneficial) of MRFI programmes on blood cholesterol. Fourteen studies (nine CBA and five RCTs) demonstrate a beneficial net effect. Four CBA studies demonstrate no net effect or inconclusive net effects and two CBA studies demonstrate a disbeneficial net effect. The most optimistic result was from a CBA study, reporting a 0.7mmol/l net reduction in blood cholesterol. The least optimistic result was from a CBA study, reporting a +0.5mmol/l net increase in blood cholesterol.

Evidence statement R3.E1c

Diastolic and systolic blood pressure: A large body of evidence demonstrates a mixed direction of effect (majority of programmes beneficial) in favour of MRFI programmes on diastolic and systolic blood pressure. Fourteen CBA studies and five RCTs demonstrate a mixed direction of effect (majority of programmes beneficial) on diastolic blood pressure. Twelve studies (seven CBA studies and five RCTs) demonstrate a beneficial net effect. Five CBA studies demonstrate no net effect or inconclusive net effects and two CBA studies demonstrate a disbeneficial net effect. The most optimistic result was from a CBA study, reporting a 5.5mmHg net reduction in diastolic blood pressure. The least optimistic result was from a CBA study, reporting a 6mmHg net increase in diastolic blood pressure. Fourteen CBA studies and five RCTs demonstrate a mixed effect (majority of programmes beneficial) on systolic blood pressure. Ten studies (five CBA studies and five RCTs) demonstrate a beneficial net effect. Five CBA studies demonstrate no net effect or inconclusive net effects and four CBA studies demonstrate a disbeneficial net effect. The most optimistic result was from a CBA study, reporting an 11.8 mmHg net reduction in systolic blood pressure. The least optimistic result was from a CBA study, reporting a 5mmHg net increase in systolic blood pressure.

Evidence statement R3.E1d

Smoking: A large body of evidence from twenty CBA studies and four RCTs demonstrate a mixed effect of MRFI on smoking prevalence (the majority of programmes beneficial). Twelve studies (nine CBA studies and three RCTs) demonstrate a beneficial net effect. Seven studies (six CBA studies and one RCT) demonstrate no net effect or inconclusive net effects and five CBA studies demonstrate a disbeneficial net effect. The most optimistic result was from a CBA study, reporting an 18.6% net reduction in smoking prevalence. The least optimistic result was from a CBA study, reporting a 12.8% net increase in smoking prevalence.

Evidence statement R3.E1e

BMI: A large body of evidence from fourteen CBA studies and three RCTs demonstrate a mixed effect of MRFI programmes on body mass index (BMI) (the majority of programmes beneficial). Ten studies (seven CBA studies and three RCTs) demonstrate a beneficial net effect. Four CBA studies demonstrate no net effect or inconclusive net effects and three CBA studies demonstrate a disbeneficial net effect. The most optimistic result was from a CBA study, reporting a 1.3kg/m2 net reduction in BMI. The least optimistic result was from a CBA study, reporting a 0.7kg/m2 net increase in BMI.

Evidence statement R3.E1f

Blood glucose: Limited evidence from 3 out of 38 programme evaluations, using different summary effect measures, demonstrate a mixed effect of MRFI on blood glucose. One RCT and one CBA study report mixed results: net decreases in men and net increases in women, whilst one CBA study demonstrates no net effect.

Evidence statement R3.E1g

Triglyceride levels, high-density lipoprotein/low-density lipoprotein (HDL/LDL) ratio or lipid levels: No evidence has been identified on the effects of MRFI programmes on triglyceride levels, HDL/LDL ratio or lipid levels.

Evidence statement R3.E1h

Dietary change – low versus high fat spreads: Five CBA studies and one RCT (+) demonstrate a mixed effect of MRFI programmes on consumption or low versus high fat spreads (the majority of programmes beneficial). Four studies (three CBA studies and one RCT) demonstrate a beneficial net effect. One CBA study demonstrates an inconclusive net effect and one CBA study demonstrates an unfavourable net effect. The most optimistic result was from a CBA study, reporting a 24% net reduction in the number of people with high consumption of fat spread on bread. The least optimistic result was from a CBA study, reporting a 3.3% net decrease in the use of unsaturated spreading fats.

Evidence statement R3.E1i

Dietary change – vegetable versus animal fats for cooking: Four CBA studies demonstrate a mixed effect of MRFI programmes on the use of vegetable versus animal fat for cooking (the majority of programmes beneficial). Three CBA studies demonstrate a beneficial net effect and one CBA study demonstrates an inconclusive net effect. The most optimistic result was from a CBA study, reporting a 6% net increase in the use of unsaturated fats for cooking. The least optimistic result was from a CBA study, reporting a 2% net decrease in the use of vegetable fats for cooking.

Evidence statement R3.E1j

Dietary change – low versus high fat milk: Five CBA studies and one RCT (+) demonstrate a mixed effect of MRFI programmes on the consumption of low- versus high-fat milk (the majority of programmes beneficial). Three CBA studies and one RCT demonstrate a beneficial net effect and two CBA studies demonstrate an inconclusive net effect. The most optimistic result was from a CBA study, reporting a 9% net increase in the use of low fat milk in men. The least optimistic result was from a CBA study, reporting a 1% net decrease in the use of low fat milk in women.

Evidence statement R3.E1k

Dietary change – consumption high fat foods: Six CBA studies demonstrate a mixed effect of MRFI programmes on the percentage of high-fat foods in the diet (the majority of programmes beneficial). Three CBA studies demonstrate a beneficial net effect, two CBA studies demonstrate no net effect or inconclusive net effects and one CBA study demonstrates a disbeneficial net effect. The most optimistic result was from a CBA study, reporting a 24% net decrease in saturated fat intake. The least optimistic result was from a CBA study, reporting a 3.4% net increase in high-fat/junk food consumption.

Evidence statement R3.E1l

Dietary change – consumption of fruit and vegetables and wholemeal bread: Limited evidence is available on the effects of MRFI programmes on the consumption of fruit and vegetables and wholemeal bread (the majority of programmes beneficial). Three CBA studies demonstrate a mixed effect of MRFI programmes on the consumption of fruit and vegetables. Two CBA studies demonstrate a beneficial net effect and one CBA study demonstrates an inconclusive net effect. The most optimistic result is from a CBA study, reporting a 9% net increase in the number of people consuming five portions of fruit and vegetables per day. The least optimistic result is from a CBA study, reporting a 0.2% net decrease in fruit consumption. Two CBA studies demonstrate a mixed effect on the consumption of wholemeal bread. One CBA study demonstrates a beneficial net effect and one CBA study demonstrates an inconclusive effect. The most optimistic result is from a CBA study, reporting a 3% increase in children. The least optimistic result is from the same CBA study, reporting a 0.3% net decrease in adults.

Evidence statement R3.E1m

Dietary change – salt intake: Two CBA studies (one [+] and one [-]) provide mixed results for the effects of MRFI programmes on salt intake. One CBA study demonstrates a beneficial net treatment effect and one CBA demonstrates an inconclusive net treatment effect.

Evidence statement R3.E1n

Physical activity: Evidence from 11 CBA studies and one RCT (+) provide a mixed pattern for the effect of MRFI programmes on physical activity (the majority of studies are disbeneficial). Three CBA studies and two RCTs demonstrate a favourable net effect. Three CBA studies demonstrate inconclusive net effects and four CBA studies demonstrate a disbeneficial net effect. The most optimistic result is from a CBA study, reporting an 11.5% net increase in the number of people doing strenuous physical activity more than three times per week. The least optimistic result is from a CBA study, reporting a 6% net decrease in the number of people who were physically active.

Evidence statement R3.E1o

Attitudes, knowledge and intentions relating to CVD risk factors: Limited evidence is available on the effects of MRFI programmes on CVD risk factor attitudes, knowledge and intention to change. One CBA study and one uncontrolled before-and-after study suggest beneficial changes in CVD knowledge following MRFI programmes. One of these studies showed a net increase in the number of individuals intending to lose weight. No evidence has been identified on the effects of MRFI programmes on CVD risk factor attitudes.

Evidence statement R3.E2a

General: Evidence for variation in effectiveness in subgroups of the population is limited and inconsistently reported across included programmes. There is no clear pattern with respect to gender, age, ethnicity or measures of deprivation which may be the result of the limited information available, confounding and selective reporting.

Evidence statement R3.E2b

Ethnicity: Three programmes report the results of subgroup analysis of effectiveness according to ethnicity. One uncontrolled before-and-after study reports lower effectiveness in ethnic minorities in acquisition of CVD knowledge. One CBA study reports lower effectiveness in ethnic minority groups for reducing smoking prevalence, reducing BMI and increasing fruit and vegetable intake and one CBA study reports no difference in effectiveness according to ethnic group.

Evidence statement R3.E2c

Age: Six programmes report results of subgroup analysis according to age. Two uncontrolled before-and-after studies report a reduction in effectiveness in acquisition of CVD knowledge in younger participants and one uncontrolled before-and-after study reports a reduction in effectiveness in reducing salt intake in younger participants. One CBA study reports a reduction in effectiveness in promoting CVD awareness in older participants. Two CBA studies report no differences in effectiveness according to age.

Evidence statement R3.E2d

Gender: Seven programmes report results of subgroup analysis according to gender. Four programmes report a reduction in effectiveness in women compared to men. One RCT reports a reduction in effectiveness in increasing physical activity in women compared to men. One uncontrolled before-and- after study and two CBA studies report a reduction in effectiveness in reducing smoking prevalence in women compared to men. One CBA study reports a reduction in effectiveness in reducing cholesterol in women compared to men. One CBA study reports a reduction in effectiveness in drinking low-fat compared to high-fat milk in women compared to men. Two programmes report a reduction in effectiveness in men compared to women. Two CBA studies report a reduction in effectiveness in promoting CVD awareness and acquisition of CVD knowledge in men compared to women and one CBA study reports a reduction in effectiveness in reducing CVD morbidity and mortality in men compared to women. One CBA study reports no differences in effectiveness according to gender.

Evidence statement R3.E2e

Social class: Two programmes report results of subgroup analysis according to social class. One CBA study reports a reduction in effectiveness in reducing smoking in lower social classes compared to higher social classes. One CBA study reports no differences in effectiveness according to social class.

Evidence statement R3.E2f

Level of education: One programme reports results of subgroup analysis according to level of education. One CBA study reports a reduction in effectiveness in CVD awareness in those relatively more educated.

Evidence statement R3.E3a

Nature of the interventions: Thirty one programmes were concerned with the effectiveness of population programmes using education and mass media, and seven with screening programmes directed at large populations in the community or primary care. However, 16 of the education and mass-media programmes contained screening components. Counselling was a key process in many programmes, undertaken individually in 24 programmes and amongst groups in 16 programmes. The 38 programmes varied in many other ways. Programme length ranged from one to over 20 years. The size of the population addressed ranged from approximately 2500 to over 1 million. Fourteen of the programmes implemented changes to the environment. Health departments (n=23) [where n is the number of programmes in which the organisations indicated were involved], local health committees (n=12), voluntary organisations (n=11) and community volunteers (n=9) had roles in programme delivery. Programmes were delivered in a variety of settings including workplaces (n=12) and schools (n=18).

Evidence statement E.3b

Education and mass-media based programmes compared to screening based: As indicated this was the most marked contrast between the programmes. However comparing the effectiveness of the two groups is complicated:

  • Many of the education and mass-media based programmes contain elements of screening.

  • There are many CVD screening programmes, particularly focused on moderate or high-risk populations which are not included in this review.

  • The comparison between the two groups is likely to be confounded by other factors, a very important one of which is that CBA studies are used to evaluate most of the education and mass-media based programmes, and RCTs all the screening based programmes.

With these provisos (and reference to pages 127–31 in report 3), the pattern of results for the risk factors of cholesterol, blood pressure (BP), smoking and BMI in the two different groups of programmes are summarised in the table below:

Programme type (n=38)

Programme result, based on direction of effect

Beneficial

Inconclusive

Disbenefical

No data

Net change in mean total cholesterol in mmol/L

Educ & MM 9

4

2

16

Screening 5

0

0

2

Net change in systolic BP in mmHg

Educ & MM 6

4

4

17

Screening 5

0

0

2

Net change in diastolic BP in mmHg

Educ & MM 7

5

2

17

Screening 5

0

0

2

Net change in BMI in kg/m2

Educ & MM 8

3

3

17

Screening 3

0

0

4

Net change in smoking prevalence in %

Educ & MM 9

6

5

11

Screening 3

1

0

3

Although the results are similar, there does appear to be a more consistent pattern of benefit in the programmes focusing on screening. As well as the provisos mentioned above, the following also need to be borne in mind when taking this observation at face value:

  • Whether this difference could be accounted for by chance alone.

  • Whether the difference would persist if the size of the effects could be taken into account.

  • Vote counting as a method of summarising the results in a systematic review is recognised to be the weakest approach.

Evidence statement R3.E3c

Possible variations in effectiveness by other aspects of the nature of the intervention: Over the three reports, many other plausible reasons for the noted variation in effectiveness have been identified. These include:

  • duration of programme

  • intensity of programme

  • use of an underlying theoretical model to inform the design of the programme

  • pre-programme investigation of particular risk factors operating in a population

  • community involvement in planning and/or design of programme

  • adaptability of the programme as new challenges emerge

  • level of integration of the separate components of the programme

  • inclusion of environmental changes as part of the programme.

Whether any of these factors account for differences in effectiveness which could not arise by chance alone has not been fully explored, and their potential importance can neither be confirmed nor refuted. Unfortunately, the extent to which the differences could ever be satisfactorily explored using the results from these evaluations is debatable given noted limitations in the reporting of the precise differences in nature of the programmes and the amount of statistical information available.

Evidence statement R3.E5c

Age: Fifteen programmes report participation in programme interventions and/or programme evaluation surveys according to age. One uncontrolled before-and-after study and 13 CBA studies report lower participation in evaluation surveys or programme interventions by those of younger age whilst one CBA study reports no difference in participation according to age.

Evidence statement R4.1b

These suggest that factors influencing success include: time limitations for projects, leadership, (including difficulties engaging community members at strategic levels), and cooperation between partner organisations.

Evidence statement R4.2a

Four study reports show conflicting evidence about the degree and methods of community engagement important for success.

Evidence statement R4.2b

For programmes that were successful, one study reports that positive community expectations about the potential of the programme to effect wider change facilitated community engagement. It is suggested that insufficient community engagement did not significantly impact on another programme's success.

Evidence statement R4.2c

For programmes that were unsuccessful, one study reports that previous negative experiences of community programmes discouraged community engagement. Conversely, another study reports engagement in the programme increased willingness for future involvement.

Evidence statement R4.3a

There is evidence from five study reports that community programmes to address heart health can be affected by the broader political context.

Evidence statement R4.3b

This can effect diverse organisational elements such as: the availability of project funding, the development of partnerships between organisations and a sense of shared purpose at different administrative levels. Individual responses may also be affected through legislation incentives to healthier behaviours.

Evidence statement R4.4a

There is evidence from four study reports that high pricing can impact on people's ability and willingness to adopt healthy eating behaviours and to participate in organised physical activity.

Evidence statement R4.7a

Six study reports discuss factors relating to organisational and strategic issues.

Evidence statement R4.7b

There is evidence from four studies that short time frames limit the ability to plan and develop the programme, engage the community, develop partnerships and communication, meet targets and leave a positive legacy.

Evidence statement R4.7c

Leadership was identified as a key organisational benefit of programmes by three studies. It is required to develop partnerships and collaborations within communities, and is important at all levels, from volunteers to with senior administrators. One study failed to see the desired shift in leadership to the community itself.

Evidence statement R4.8a

Evidence from six study reports is related to the organisational culture and partnerships of those involved in CVD programme services.

Evidence statement R4.8b

Three studies note differences in culture between partner organisations including frames of reference, terminology and programme expectations although this didn't always lead to conflict.

Evidence statement R4.8c

Four studies suggest that CVD programmes are enhanced where partner organisations have aligned values, priorities, focus and goals between organisations.

Evidence statement R4.8d

Partnerships may have positive effects through interagency learning, increasing the visibility of smaller organisations and enhanced funding opportunities.

Evidence statement R4.9e

Two studies note that to sustain the provision of healthier food options, communities need to take them up and so they need to be made attractive and clear.

Evidence statement R4.9f

One study found that community projects were largely unwilling address smoking, preferring to promote physical activity.

Evidence statement R4.10a

Five studies reported on staffing successful programmes.

Evidence statement R4.10b

Three studies report difficulties in recruitment and retention.

Evidence statement R4.10c

Positive staff contributions were defined in three studies where successful networking allowed staff to use their time effectively because they were not duplicating activities; where they were assisted by structures that focus on heart health issues and where flexibility allowed them to spend significant periods of time with participants.

Evidence statement R4.10d

Positive staff characteristics included knowledge and interest in heart health and being upbeat and friendly. A range of specialist staff should be involved.

Evidence statement R4.11a

Six study reports make specific comments about funding and resource requirements.

Evidence statement R4 11.b

Five study reports note the need for those with existing roles and responsibilities to be given resources to take on additional CVD programme work, or for dedicated positions to be created. Limited time for school-based staff may be a particular problem.

Evidence statement R4.12a

Two studies identify effective communication between organisations, staff and the community, to be important, but use different mechanisms to achieve this: lay health advisers or having a full time project coordinator.

Evidence statement R4.13a

Four study reports discuss recruiting and retaining programme volunteers.

Evidence statement R4.13b

Volunteers need adequate resourcing and leadership, and may be motivated by witnessing positive changes in the community.

Evidence statement R4.13c

One study suggests that volunteers find health promotion less satisfying than traditional patient service roles.

Evidence statement R4.13d

Two study reports about the same programme note that lay health advisers, recruited from the target community, were key.

Evidence statement R4.14a

Two study reports mention the role of GPs in community CVD projects.

Evidence statement R4.14b

GP uptake was slow and it is suggested that GPs may be less comfortable in health promotion roles than their traditional role of secondary prevention.

Evidence statement R4.15a

Evidence relating to staff training were identified by six study reports.

Evidence statement R4.15b

While skills training is needed, involvement in the programme itself increases skills and knowledge through sharing information and implementing theoretical knowledge.

Evidence statement R4.16a

Five study reports relate to the evaluation of CVD prevention programmes.

Evidence statement R4.16b

Two study reports found that process evaluation and action research raised self-awareness among staff and promoted programme improvement. While a third reports that time limited projects limit the possibility of such learning.

Evidence statement R4.16c

Data management was a challenge in long-term projects and those with multiple strands across a number of organisations.

Evidence statement R4.17a

There is evidence from eight study reports about community engagement in CVD prevention programmes.

Evidence statement R4.17b

Two study reports suggest that successful community engagement requires multiple approaches across populations.

Evidence statement R4.17c

Two study reports suggest that successful programmes need to be sensitive to communities' habits and cultural patterns, while a further three describe the important of matching programme staff to the social and/or ethnic characteristics of the target communities.

Evidence statement R4.17d

Challenges to community engagement include engaging community representatives at strategic levels; building confidence in community leaders; difficulties breaking into existing networks; competing with other community events; reaching young people; reluctance due to the legacy of negative experiences with previous initiatives; lack of enthusiasm in the community.

Evidence statement R4.18a

Seven studies report that the local physical environment had important effects on the ability of community CVD risk-reduction projects to be successful.

Evidence statement R4.18b

Five studies reported that access to healthy food options was limited, while unhealthy food was more visible, both in the community and in school-based programmes.

Evidence statement R4.18c

Local barriers to physical activity including no sidewalks, unmetalled roads or loose dogs; lack of school provision to secure bikes or store kit which discourages extra-curricular exercise; and local availability of gyms or other facilities.

Evidence statement R4.19a

Community and familial norms: There is evidence from one study report among British Asians that stress from a variety of sources was a noted problem among both men and women and this might lead to inability to access essential services or to communicate with professionals.

Evidence statement R4.20a

Attitudes to food and cooking: There is evidence from three study reports that specific foods and eating patterns may be regarded as important expressions of cultural identity. Cultural norms about food types and their preparation may not be the most healthy from a CVD prevention perspective.

Evidence statement R4.21a

There is evidence from three study reports about cultural attitudes to weight and exercise.

Evidence statement R4.21b

These suggest that, among some groups, understandings of greater weight as a sign of wealth and health may persist which may challenge successful adoption of CVD prevention activities.

Evidence statement R4.21c

Further, specific connotations of language used to describe weight and physical activities may exist, so shared understandings between clinical and community meanings should not be assumed.

Evidence statement R4.22a

Fatalism and health: There is evidence from three study reports, among three different ethnic groups, of fatalistic attitudes where one's state of health is the will of God.

Evidence statement R4.23a

There is evidence from six study reports to suggest that a benefit of community CVD programmes is in providing leadership that encourages local attitudes to change for the better.

Evidence statement R4.23b

As well as making personal changes, such 'social health' encouraged changes within the family, within the local community and within the wider social and political community. Despite this, one study suggests that men remain less likely to use health services.

Evidence statement R4.24a

Nine studies discuss community perceptions of CVD risk factors.

Evidence statement R4.24b

Six studies report high levels of understanding about CVD risk among the target population while two suggest limited understanding and two suggest challenges in turning knowledge into action.

Evidence statement R4.24c

One study suggests that different types of knowledge are at play (theoretical, practical, experiential and intuitive), and where there is a discrepancy between theoretical and experiential knowledge, the latter influences what participants do. These links might be challenged by cues to action (health belief model) – most significantly breakdown of self-image and social networks.

Evidence statement R4.24d

One study develops a typology of six 'ideal types' of functional and dysfunctional attitude among programme participants who see it as a blessing, an opportunity, a confirmation, a watchman, a disappointment or an insult. The latter two are negative or 'dysfunctional' in terms of positive health choices and more men have these attitudes.

Evidence statement R4.25a

Nine study reports discuss people's motivations for, or resistance to, adopting risk reduction behaviours.

Evidence statement R4.25b

Two studies report that health concerns, sometimes serious, were motivating factors to participate and two that feedback of physiological test results was motivating.

Evidence statement R4.25c

Women may be targeted to take heart health practices home, however, two studies report on difficulties initiating or maintaining family interest and that resistance from family members was a barrier to adopting healthier behaviour. It is difficult to maintain behaviour changes amidst the usual business of family commitments.

Evidence statement R4.25d

One study suggests that there is a need for ongoing support in order for behavioural changes to be made and maintained.

Evidence statement R4.25e

One study found that secondary school pupils enjoyed the freedom to make food choices not available at primary school – pupils' food choices, and those of the wider population, may reflect issues other than health.

Evidence statement R4.26a

Six studies report on participant perceptions of programmes in which they were involved.

Evidence statement R4.26b

In two studies participants reported improving heart health through weight loss, increased exercise, as well as increased awareness and use of services and programme activities. One study also suggests that networks providing community support was a benefit.

Evidence statement R4.26c

One study found that practical demonstrations were much more successful than information provision alone.

Evidence statement R4.26d

Two studies suggest that the participants may doubt the credibility of health messages, with so many sources of, sometimes contradictory, information available. Matching the characteristics of the community may be important.

Evidence statement R5.2

Community engagement: Positive community engagement requires trusting, respectful relationships to be built which motivate and support change. Community engagement should be an ongoing and dynamic partnership which responds to community needs.

As CVD may not be seen as an immediate concern within targeted communities, staff may first need to listen and respond to the existing concerns of the community. This may be done through participating in existing networks and forums, or creating forums that have more open agendas, at least to start with.

Sufficient time is needed to ensure that this is done appropriately and also to ensure that changes become adopted by the community so that they are empowered to continue, even if the project itself comes to an end.

Information and education is likely to be more effective if it relates to the experiences of the community, and if those that deliver it are seen as part of that community. Appropriately skilled staff are needed for effective community engagement.

Greater levels of participation, that involve community members as partners or devolve power to them, may have additional benefits – ensuring that programmes are truly responsive to community needs, involving local people in the complexities of planning and delivering such programmes and so facilitating understanding within the community.

Done well, community engagement may create a positive feedback loop which motivates change, improving health which produces greater motivation. However, care needs to be taken to ensure that those adopting behaviour change are not just those already motivated to change, thereby increasing, rather than lessening, health inequalities.

Evidence statement R5.3

Staffing – leadership: Strong, inspirational leadership may be important to initiate, coordinate and drive complex programmes and motivate and encourage cooperation among multiple staff across a number of agencies with a range of responsibilities.

To fulfil this, staff are needed whose role is dedicated to the programme and those with multiple roles need to have appropriate time freed up.

Leaders may be needed for the project over all, but also for specific elements of the project, for example, to encourage primary care participation or ensure local political or funding support. Leaders from within the community are also needed to champion the project and facilitate engagement.

Expectations of leadership roles should be matched by appropriate control and responsibility, and given the necessary training and support.

Evidence statement R5.4

Staffing – staff engagement: To ensure that staff are engaged with the aims of a CVD prevention progamme, they require appropriate training and resources, a good understanding of how their role fits into the programme overall and a clear understanding of the extent of their roles and responsibilities.

Evidence statement R5.5

Staffing – GPs: The role of primary care was complicated and sometimes contradictory. Some GPs may be more comfortable with a secondary, rather than primary, prevention role, which may explain why some participants found it difficult to engage them in CVD prevention programmes. Conversely, other participants viewed primary care as crucial partners in CVD prevention. Advocacy among other local organisations may be a key role.

Where primary care is involved in CVD prevention programmes, they need to receive appropriate resources to free-up staff time.

Engaging primary care and keeping them appropriately informed may require tailored approaches.

Evidence statement R5.6

Staffing – volunteers: Volunteers from within the community may be particularly effective at informing, motivating and engaging their peers in the community and enhance community empowerment.

Volunteer workers need to be properly trained and supported to ensure that they continue to be involved and don't get burnt out.

The issues of paying those involved should be considered carefully.

Evidence statement R5.7

Staffing – multi-agency, multi-disciplinary teams: Public health work to reduce CVD is likely to require the involvement of multiple agencies and disciplines.

Coordination and cooperation is required to build trust and a sense of shared purpose through aligning the goals and activities of different agencies involved, and assigning clear roles and responsibilities to participating organisations and staff within them. Joint appointments may facilitate this. Ongoing feedback and communication is vital.

Sufficient time is needed to successfully negotiate and accommodate different expectations and bureaucracies.

Evidence statement R5.8

Legacy: CVD reduction programmes may enhance their longer-term impact through ensuring that programme activities are embedded within organisations and the community.

Appropriate training and support for key staff, and community members, from project inception may help to ensure activities become 'institutionalised'. Ongoing sources of funding should also be identified.

Programme impacts should be regularly assessed and results fed back to staff and organisations so that successful activities are recognised and adopted. This will require the identification of appropriate resources.

Early and ongoing community engagement may ensure ongoing changes in healthy behaviours, empowering the community to maintain positive changes. Short-term projects often fail to leave lasting benefits to a community as their short-term goal setting may preclude the necessary engagement required.

Evidence statement R5.9

Short time frames: Short time frames for CVD prevention programmes may threaten success at a number of levels: implementation, staff engagement and training, community engagement, evaluation and legacy. It is difficult for such programmes to meet community needs, staff needs or to permit changes to become embedded in the community. This may lead communities and local agencies to lose faith in such interventions, further hampering the ability of future work to be successful in those areas.

Evidence statement R5.10

Structural barriers: At a macro-level, changes in the broader political environment can have dramatic effects on the adoption and continuation of prevention activities.

Support for CVD prevention programmes may be affected by changing political priorities around prevention and treatment of illness.

Evidence statement R5.11

Piloting and monitoring: Cyclical approaches to monitoring and evaluation, such as piloting, process evaluation and action research, allow project to be responsive to local needs, adapting or removing inappropriate projects and allowing successful projects to be rolled out.

Information from this process fed back to staff in a timely way can help develop a sense of ownership and cooperation and motivate good practice.

Organisations and individuals should also learn from the experiences of previous projects.

Evidence statement R5.12

Challenges of evaluation: Commissioners and funders may need to allow flexibility in programme evaluation designs to allow them to adapt to local needs, rather than requiring fixed plans prior to funding. In addition, programmes and evaluations should allow sufficient time for outcomes to be achieved.

Multiple methods may be needed to evaluate important aspects of CVD prevention programmes, such as community empowerment, that are not all easily captured through numerical outcome data.

Programmes that measure only population-level changes may not capture large impacts for some individuals, and this may be important, especially where health inequalities are addressed.

Evidence statement CE1

Three studies gave results in cost per life-year gained for population-based programmes compared to no intervention. The results ranged from cost-saving to £240,000 per life-year gained.

Evidence statement CE2

Two studies gave results in cost per QALY or DALY (disability-adjusted life years) for population-based programmes compared to no intervention. Results ranged from £10 per QALY to £96 per DALY.

Evidence statement CE3

Two studies gave results in cost per case prevented for population-based programmes compared to no intervention. Results ranged from cost saving to £22,000 per case prevented.

Evidence statement CE4

Five studies reported results in cost per life-year gained for some form of screening strategy compared to no intervention. Results ranged from cost saving to £140,000 per life-year gained.

Evidence statement CE5

Two studies gave results in cost per case prevented for screening compared to no intervention. Results ranged from £10,000 to £730,000 per case prevented.

Evidence statement CE6

Two studies gave results per 1% reduction in coronary risk for screening compared to no intervention. Results ranged from £2.25 to £5.30 per 1% reduction for one person.

Evidence statement CE7

One study gave a result of £0.80 per pound weight lost for a screening programme compared to no intervention.

Evidence statement CE8

One study gave results ranging from £12,000 to £120,000 per life-year gained and £100,000 to £230,000 per QALY for screening compared to a population-based approach.

Evidence statement CE9

One study gave results from cost saving to £39,000 per life-year gained for some form of exercise training.

Additional evidence

  • Expert reports:

    • ER 1: 'The effectiveness of physical activity promotion interventions'

    • ER 3: 'Expert testimony on salt and cardiovascular disease'

    • ER 4: 'The relationship between commercial interests and risk of cardiovascular disease'

    • ER 5: 'Regional development of a population-based collaborative CVD prevention strategy: the experience of NHS West Midlands'

    • ER 6: 'NICE guidance on the prevention of CVD at population level: evidence from the Co-operative Group'

    • ER 7: 'Population and community programmes addressing multiple risk factors to prevent cardiovascular disease (CVD): addendum to qualitative study produced by Peninsula Technology Assessment Group for NICE: CVD programme – Heart of Mersey (HoM)'

    • ER 8: 'Expert testimony paper on the independent evaluation of 'have a heart Paisley' phase one (Scotland's national CHD prevention demonstration project)'

    • ER 9: 'Expert testimony on the public health harm caused by industrially produced trans fatty acids and actions to reduce and eliminate them from the food system in the UK'

    • ER 10: 'Prevention of cardiovascular disease at a population level: evidence on interventions to address dietary fats'

    • ER 11: 'CVD risk factors: paradigms and pathways'

    • ER 12: 'CVD prevention in populations: lessons from other countries'

    • ER 13: 'Will CVD prevention widen health inequalities?'

  • 'Obesity'. NICE clinical guideline 43 (2006):

Cost-effectiveness evidence

The economic analysis consisted of a review of economic evaluations and a cost-effectiveness analysis.

  • 'Prevention of cardiovascular disease at population level (question 1; cost-effectiveness)'

  • 'Prevention of cardiovascular disease at population level: modelling strategies for primary prevention of cardiovascular disease'.

Some primary prevention programmes involving education, mass media and screening with a general population were found to be effective and cost effective. They may reduce some of the risk factors for CVD, including changing behaviours which increase the risk. However, when the findings from all programmes were summarised, the overall effect on health outcomes was uncertain. In addition, as these programmes were conducted many years ago, the findings may not be generally applicable in the UK now.

The cost-effectiveness analysis strongly suggests that legislation likely to reduce the risk of CVD can be expected to produce a net cost saving to the public sector – as well as improving health. (Unless a very large sum of money needs to be spent on implementation.)

For example, implementing a CVD prevention programme based on the North Karelia project would result in an incremental cost-effectiveness ratio of approximately £7000 per quality-adjusted life year (QALY). For the Stanford Five City Project, the total healthcare cost savings almost equal the estimated cost of the project. The benefits of reducing the prevalence of smoking would also make the programme cost saving.

At the request of the Programme Development Group (PDG), the scope of the modelling was extended beyond programmes for which there was direct evidence of effectiveness. Interventions modelled included:

  • The North Karelia project – including the effect of a net percentage reduction in serum cholesterol of 3% for men and 1% for women, and a reduction in systolic blood pressure of 3% for men and 5% for women.

  • The Stanford Five City Project – the effect of a 4% reduction in systolic blood pressure and a 2% decrease in serum cholesterol among the general population.

  • Legislation to ban trans fats and so reduce trans fatty acid (TFA) levels in the population so that it only accounts for approximately 0.7% of total energy intake.

  • Legislation to reduce the population's salt intake by 3 g and 6 g per day.

The modelling made a number of conservative assumptions. It found that halving CVD events across the entire England and Wales population of 50 million would result in discounted savings of approximately £14 billion per year. Reducing mean population cholesterol or blood pressure levels by 5% would result in discounted annual savings of approximately £0.7 billion and £0.9 billion respectively. Reducing population cardiovascular risk by even 1% would generate discounted savings of approximately £260 million per year.

Additional benefits to existing CVD patients, and reductions in other diseases, were not quantified.

As the model is based on a series of conservative assumptions, it probably seriously underestimates the true health benefits to be gained from the recommendations.

Fieldwork findings

Fieldwork aimed to test the relevance, usefulness and feasibility of putting the recommendations into practice. The PDG considered the findings when developing the final recommendations. For details, go to the fieldwork section in appendix B and online.

Feedback to the recommendations varied.

Food industry recommendations (1, 2, 3, 4, 5 and 10 in the consultation document)

It was indicated, particularly by some food industry stakeholders, that:

  • the recommendations were 'out of date' and did not reflect the current situation within the food industry

  • it was not feasible to implement parts of the recommendations and their impact would be minimal

  • some of the advice was already covered by other government agencies, as well as at European level

  • the role of these recommendations was questionable, particularly where they were quoting differences in target values from those agreed with the FSA.

  • It was felt that NICE has 'a lot less weight' in the food and planning sectors and that we would need to work much more closely with them to gain support for the recommendations.

  • Food industry representatives generally (but not unanimously) indicated that trans fats were 'no longer an issue' and should, therefore, not be included in the recommendations. They also questioned the levels of saturated fat and salt recommended, as they were unaware of the evidence to support a reduction to these levels.

  • Catering industry recommendations (6, 14 and 15 in the consultation document)

In general, stakeholders felt that the catering recommendations demonstrated how to follow good practice.

  • Local authority planning recommendations (7, 8, 9, 12, 13, 16, 17 and 18 in the consultation document)

Local planning and policy representatives stated that the recommendations needed to become part of national planning policy or law. Otherwise, it would not be possible to implement them.

Communications recommendation (number 11 in the consultation document)

There was limited feedback from the industry as some of the target organisations disputed that it was relevant to them and declined the opportunity to be interviewed.

CVD programme recommendations (19 to 24 in the consultation document)

In general, representatives from the NHS and PCTs indicated that the regional CVD prevention recommendations were appropriate and that they supported current work on CVD prevention.

  • National Institute for Health and Care Excellence (NICE)