Appendix C: The evidence

This appendix lists evidence statements provided by two reviews and links them to the relevant recommendations (see appendix B for the key to study types and quality assessments). The evidence statements are presented here without references – these can be found in the full review (see appendix E for details). It also sets out a brief summary of findings from the economic appraisal.

The two reviews of effectiveness are:

  • 'The effectiveness of smoking cessation interventions to reduce the rates of premature death in disadvantaged areas through proactive case finding, retention and access to services.'

  • 'The use of statins: proactive case finding, retention and improving access to services in disadvantaged areas'.

Evidence statement 1SM indicates that the linked statement is numbered 1 in the review 'The effectiveness of smoking cessation interventions to reduce the rates of premature death in disadvantaged areas through proactive case finding, retention and access to services'. Evidence statement 1ST indicates that the linked statement is numbered 1 in the review 'The use of statins: proactive case finding, retention and improving access to services in disadvantaged areas'.MR is used to indicate that supporting evidence on current practice can be found in the mapping review.

As noted in appendix B, study quality provides an overall indication of how well a study was conducted to minimise the likelihood of bias. For example, a quality rating of '++' indicates minimal likelihood of bias, whereas a rating of '-' indicates a significant likelihood of bias. Some of the studies that informed the evidence statements below were rated '-', due to poor methodology. However, this quality rating does not always apply to the way the studies actually identified, supported and improved individuals' access to services – the areas under investigation for this guidance.

The reviews and economic appraisal are available on the NICE website. 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) below.

Where PHIAC has considered other evidence, it is linked to the appropriate recommendation below. It is also listed in the additional evidence section of this appendix.

Recommendation 1: evidence statements 1SM, 2SM, 6SM, 10SM, 13SM, 1ST, 2ST, 5ST, 7ST, 9ST, 10ST, 11ST, 12ST; MR

Recommendation 2: evidence statements 2SM, 3SM, 4SM, 5SM, 6SM, 7SM, 10SM, 11SM, 13SM, 14SM, 3ST, 4ST, 12ST, 13ST, 14ST, 15ST, 16ST, 18ST, 19ST, 20ST, 22ST, 23ST, 24ST; MR

Recommendation 3: evidence statement 1SM; MR

Recommendation 4: evidence statements 4SM, 6SM, 13SM, 14SM, 4ST, 11ST, 12ST, 13ST, 19ST, 20ST, 22ST; MR

Recommendation 5: evidence statements 8SM, 9SM, 14SM, 4ST; MR; IDE

Evidence statements

Evidence statement 1SM

Evidence from one UK observational study (++) suggests that the QOF component of the 2004 GP contract may have continued, rather than reversed, differences in the quality of care delivered between primary care practices in deprived and less deprived areas.

Evidence from another UK observational study (++) suggests that the new GP contract has resulted in an improvement in the recording of smoking status and the recording of the delivery of brief cessation advice in primary care, but not the prescribing of smoking cessation medication.

As these studies took place within UK primary care, they are directly relevant to the review.

Evidence statement 2SM

One cluster RCT in the UK (++) found that proactively identifying smokers through primary care records was feasible, and providing these smokers with brief advice and referral to NHS Stop Smoking Services increased contact with services and quit attempts but did not increase rates of cessation.

One observational study (-), one descriptive study (-), one cluster-controlled trial (+) and one RCT (+) conducted in the USA demonstrate that proactively identifying smokers in a number of ways, for example, through primary care, using a screening tool, or through cold calling, is possible and that these provide effective ways of recruiting smokers to cessation interventions. One observational study in Sweden (+) demonstrates that direct mailing to smoking mothers can be successful in increasing both participation in smoking cessation programmes and quit rates. One study took place within English primary care and it is directly applicable to the review. The remainder took place in the USA and may have limited applicability. Only one (American) study focused upon disadvantaged individuals and therefore the applicability of this evidence to target populations for this review may be limited.

Evidence statement 3SM

Two observational studies (both [++]) demonstrate that the NHS Stop Smoking Services have been effective in reaching smokers living in disadvantaged areas of England. As both took place in England and are focused on disadvantaged individuals, they are directly applicable to the review.

Evidence statement 4SM

Two studies provide evidence to suggest that barriers such as fear of being judged, fear of failure and lack of knowledge need to be tackled in order to motivate smokers from lower socioeconomic groups to access cessation services. Interventions need to be multi-dimensional in order to tackle social and psychological barriers to quitting as well as dealing with the physiological addiction. (Two UK-based studies, one involving focus groups [++] and one involving interviews [++]). As both these studies took place with disadvantaged smokers in the UK, they are directly relevant to this review.

Evidence statement 5SM

Evidence from four studies suggests that social marketing has a role to play in delivering client-centred approaches to smoking cessation to disadvantaged individuals. (One UK-based observational study [-], one international RCT [+], one international population-based study [+] and one international controlled-before-and-after study [-]). One of these studies took place with disadvantaged smokers in the UK and is directly relevant to the review. Three took place in the USA and may have limited applicability to this review.

Evidence statement 6SM

One UK-based (+) study suggests that including lay people or community members as advisers may form an important part of a successful smoking cessation intervention targeted at a specific group, in particular, if the service is tailored to their specific needs and allows them to explore smoking in the context of relevant issues in their lives. This study took place with smokers in the UK and is relevant to this review.

Evidence statement 7SM

Two American studies suggest the need to test existing cessation interventions to determine their suitability for the specific group, to receive feedback from that group and to make amendments to any aspects that are unsuitable. In order for the client group to benefit, the intervention must fit their level of need and understanding, and be suitably accessible. (One USA-based RCT [++], and one USA-based cohort study [-].)

Evidence statement 8SM

There is evidence from a number of studies that training pharmacists to deliver smoking cessation interventions is important and that pharmacies may be a valuable means of reaching disadvantaged individuals and increasing their smoking cessation rates (one UK systematic review comprising two RCTs and three non-randomised experimental studies [++], one UK observational study with interviews [++] and one international pilot study [+]). Two studies took place within the UK and are directly applicable to the review. One took place in the USA and so may have limited applicability to this review.

Evidence statement 9SM

There is evidence from three reviews that training dental professionals to deliver smoking cessation interventions is important, and that this setting has the potential to reach large numbers of smokers and increase cessation rates (one international systematic review comprising six RCTs [-], one UK review of mixed-study designs [-] and one international review of seven RCTs [+]). One study took place within the UK and is directly applicable to the review. Two studies took place in the USA and so may have limited applicability to this review. There is limited reference to disadvantaged individuals in any of the reviews and therefore the applicability of this evidence to target populations for this review may be limited.

Evidence Statement 10SM

Three studies provide some evidence of the potential benefit of drop-in or rolling, community-based sessions to reach smokers and increase cessation rates: two UK-based studies involving face-to-face interviews (both [-]) and one UK-based observational study (-). All studies took place within the UK and are directly applicable to the review.

Evidence Statement 11SM

One cohort study (+) provides evidence of the potential benefits of locating smoking cessation services in the workplace of manual groups to increase cessation rates. This study took place in the USA and so may have limited applicability to this review but does have potential implications for the UK population.

Evidence Statement 13SM

One RCT in the UK (++) with CHD patients randomised to nurse-run clinics or controls found little evidence of a change in smoking behaviour. Two RCTs in the UK (+) and (-) exploring smoking cessation interventions at routine cervical screening appointments found some evidence that brief interventions change the motivation or intention to quit smoking. One international RCT (+) examined the recruitment of women smokers attending a child's paediatric appointment into a smoking cessation intervention and found some evidence of an impact on quitting smoking. One international RCT (+) and one observational study using face-to-face interviews (+) investigated the use of cellular phones for smoking cessation in HIV-positive patients and showed a potential benefit for using this method of support. One US cohort study (+) provided preliminary evidence that offering a reduction programme could reach and influence more smokers than a programme just offering cessation. Three studies were carried out in the UK and are directly applicable to the target population, but they did not examine disadvantaged individuals separately. Four studies were carried out in the US and so may have limited applicability to this review.

Evidence Statement 14SM

Two UK surveys (one telephone [+] and one internet [+]) and one descriptive and audit survey (-) carried out in the UK provide evidence of pregnant smokers' perceptions of barriers to using smoking cessation support. Barriers include, among others: unsatisfactory information, lack of integration of cessation into routine antenatal care, lack of enthusiasm or empathy from health professionals and lack of short-term support. One RCT in the UK (+) of motivational interviewing with pregnant smokers and two international RCTs, one of a brief versus more intensive intervention (++) and one of proactive telephone support (-) provide little evidence of the effectiveness of these interventions. One US descriptive study (-) described the reach of a multifaceted pregnancy campaign but reported no outcomes. The UK studies are directly applicable to the target population, although only one of these focused on pregnant smokers in disadvantaged areas.

Evidence Statement 1ST

There is evidence from three case studies suggesting interventions inviting specific populations (South Asians, homeless people or patients with psychosis) to attend risk screening at their GP practice or primary care clinic may identify a number of people at risk of coronary heart disease (outcomes reported in two case studies [+], [-]). However, it is difficult to draw firm conclusions on how well such interventions are attended due to poor reporting of participation rates (outcomes reported in three case studies: two [+] and one [-]).

Evidence Statement 2ST

There is evidence from one small case study (+) that screening long-term psychiatric hospital patients can identify previously undetected CHD. Screening 64 patients identified one new case of established CHD and 22 previously undetected test abnormalities. Participation in the intervention was high (66%) but only a small proportion consented to having blood tests.

Evidence Statement 3ST

There is evidence from one RCT (+) that in an area of deprivation, postal prompts to patients and their GPs following an acute coronary event, improves monitoring of the patient's risk and the likelihood of the patient having at least one consultation with their GP or nurse.

Evidence Statement 4ST

There is evidence from one case study (+) to suggest that, in an area of deprivation, a project funding a nurse and exercise worker to develop practice nurse and GP skills in identifying and monitoring patients and facilitate the provision of exercise facilities for CHD patients, may lead to a small improvement in cholesterol testing of patients. 72.5% of control patients reported receiving cholesterol tests in the past year compared to 77.8% of the intervention group, p=0.002. No differences were seen in blood pressure measurement.

Evidence Statement 5ST

There is weak quality evidence from two case studies (both [-]) to suggest that offering cardiovascular risk assessment opportunistically to African-Caribbean general practice patients, or patients from a range of socioeconomic categories, may identify a number of people at risk of CHD. However, the interventions require further research using well-conducted studies before firm conclusions can be made.

Evidence Statement 7ST

There is evidence from three studies to suggest that workplace cardiovascular screening provided in schools or businesses in multi-ethnic, low-income areas (CBA [-], case study [-]), or for factory workers (case study [+]) is moderately well attended. Results suggest that a number of participants were identified for referral to a physician for follow-up (outcome reported in two studies: CBA [-], case study [-]). No firm conclusions can be made on patients' completion of follow-up as this was only reported in one poor quality study (case study
[-]).

Evidence Statement 9ST

Evidence from one UK case study (-) evaluating the establishment of a health screening clinic in a prison indicated a moderate 35% voluntary uptake by the inmates. There were active interventions following the screening for 87 (34%) inmates and 13 (32%) staff screened. These ranged from simple anti-smoking and dietary advice to more formal medical interventions to manage raised blood pressure and cholesterol. Uptake data should be viewed cautiously, as the number of potential participants was not reported.

Evidence Statement 10ST

Two case studies suggest that offering blood pressure measurements at community sites in areas of deprivation can identify a number of people with elevated blood pressure. No firm conclusion can be made on participation rates as these were not reported in the studies. One UK case study (+) found 221 people out of 758 first-time users of self-reading sphygmomanometers placed in public sites had elevated blood pressure measurements. No firm conclusions can be made regarding physician follow-up as the researchers were unable to contact all of these people. One US RCT (+) providing blood pressure measurements at a range of community sites identified 31.4% with elevated blood pressure and 10.7% with severely elevated blood pressure. Transferability and cost-effectiveness of such interventions requires further study.

Evidence Statement 11ST

There is evidence from two case studies evaluating phase one (+) and phase two (-) of the Well-Integrated Screening and Evaluation for Women Across the Nation (WISEWOMAN) programme to suggest that adding cardiovascular screening to state breast and cervical cancer screening programmes reaches financially disadvantaged and minority ethnic women and identifies a number at risk of CHD. No conclusions can be made on participation rates or physician referrals as these outcomes have not been reported. Applicability and transferability of these programmes to a UK setting requires further study.

Evidence Statement 12ST

Evidence from three studies (two case studies [+] and one uncontrolled before- and-after study [+]) suggests that culturally-sensitive education sessions that include an element of cardiovascular risk assessment may be effective in the identification of at-risk individuals. Two moderate-quality studies evaluated educational interventions in black and minority community groups (+) and Turkish immigrants at a mosque (+) offering blood pressure measurements. Participation with blood pressure measurements were high, and revealed a number of patients with uncontrolled hypertension or with elevated blood pressure readings. Evidence from one case study (-) in which health checks were conducted before and after a church-based educational intervention with predominantly black participants should be viewed more cautiously owing to concerns of transferability and applicability.

Evidence Statement 13ST

Evidence from one qualitative study (++) of service users with severe mental illness (SMI), and primary care staff and community mental health teams, indicate a range of perceived obstacles to CHD screening. These include: lack of appropriate resources in existing services; anticipation of low uptake rates by patients with SMI; perceived difficulty in making lifestyle changes among people with SMI; patients dislike having blood tests; and lack of funding for CHD screening services or it not being seen as a priority by trust management. There was some disagreement about the best way to deliver appropriate care, and the authors concluded that increased risk of CHD associated with SMI and antipsychotic medications requires flexible solutions with clear lines of responsibility for assessing, communicating and managing CHD risks.

Evidence Statement 14ST

There is a paucity of good quality research on the effectiveness of pharmacist interventions to improve compliance with lipid-lowering therapy, particularly in disadvantaged individuals. Results from the four studies identified (two RCTs [-, -] one UCBA [uncontrolled before and after study] [-] and one observational study [-]) should be viewed with caution owing to poor methodological quality and doubts about applicability to disadvantaged individuals.

Evidence Statement 15ST

Evidence from one low-quality RCT (-) suggests that telephone reminders and postcards to reinforce messages about coronary risk reduction does not produce significant improvements in short-term compliance in patients prescribed pravastatin treatment. Results should be viewed with caution as the poor quality study is likely to be highly biased and may not be applicable to disadvantaged individuals.

Evidence Statement 16ST

Well-conducted research examining patient education to improve compliance with lipid-lowering therapy is required before firm conclusions can be made regarding its effectiveness, particularly in disadvantaged individuals. Evidence from one uncontrolled before-and-after study (+) of nurse-led education in heart failure patients suggested there was no significant difference in self-reported compliance at one year. One RCT (-) of a pharmacy intervention including patient education for heart failure patients found a significant difference in compliance at 2 and 6 months, but not at 12 months. Applicability of the studies may be limited as the medication prescribed was not specified.

Evidence Statement 18ST

Well-conducted research is required examining the effectiveness of improving retention of patients at risk of or with CHD within services. Evidence from the one systematic review identified (+) highlights the dearth of literature reporting the evaluation of simple interventions aimed at improving adherence to cardiac rehabilitation for all patients or specific groups of patients. The systematic review identified few studies of sufficient quality to enable the recommendation of specific methods to improve adherence to outpatient cardiac rehabilitation. The most promising approach was the use of self-management techniques based around individualised assessment, problem solving, goal setting and follow up. This was most likely to be effective in improving specific aspects of rehabilitation, including diet and exercise.

Evidence Statement 19ST

Evidence from one systematic review (+) highlighted the need for trials of interventions applicable to all patients and targeting specific under-represented groups. The review revealed some evidence to support the use of approaches aimed at motivating patients, regular support and practice assistance from trained lay volunteers and a multifaceted approach for the coordination of transfer of care from hospital to general practice. Applicability and transferability of these programmes to disadvantaged populations requires further study.

Evidence Statement 20ST

Evidence from three studies indicated the importance of providing additional staff resources to encourage or support the uptake of services by people living in socially deprived areas. One US moderate-quality RCT (+) in a predominantly black population from a low income area found improved uptake of services with a tracking and outreach intervention, where community health workers supported patients in completing referral to their physician for high blood pressure. Evidence from one non-comparative UK case study (+) indicates that additional resources for tertiary cardiology may have reduced socioeconomic inequities in angiography without being specifically targeted at the needier, more deprived groups, but the impact on revascularisation equity is not yet clear. Evidence from one UK case study (-) suggested that a project funding one nurse and one exercise worker to support GP practices in a socially deprived area increased the practices' provision of cardiac rehabilitation services such as exercise programmes, psychological and social support and dietary advice. Project nurses worked directly with practice nurses and GPs to develop their skills in identifying and monitoring patients with CHD, giving lifestyle advice and ensuring optimum medication regimes. An exercise worker worked with practices and the community to identify and facilitate the provision of exercise resources suitable for CHD patients.

Evidence Statement 22ST

A number of barriers and enablers to accessing services were identified in five qualitative studies involving people from socially deprived areas ([++], [+, +, +] [-]). Common themes were a lack of understanding of services and treatments and the need for flexible services; the inconvenient timing of appointments and the lack of transport were both cited as barriers; with the latter overcome by the provision of home visits. Personal factors, such as the need to minimise the severity of their illness, taking a 'cope and don't fuss' approach and fear of blame were also reported as barriers. The absence of cardiac rehabilitation services and long waiting lists was also noted and, for some patients, a reluctance to attend group care ([++], [+, +], [-]). Healthcare providers agreed on the need to expand cardiac rehabilitation services to reach out into communities and that the expansion would need to take place in the community (+).

Evidence Statement 23ST

A number of barriers and enablers to accessing services were identified in five qualitative studies involving Asian populations ([++], [+, +, +]) and African-Caribbean populations (+). Among Asian populations, a range of religious and cultural issues were identified including female inhibitions, religious practices, family commitments and influence and 'inappropriate' topics. The need for flexibility in the timing of services was highlighted and sensitivity in planning activities around religious events was viewed positively. Patients' lack of understanding of services and treatment was suggested as a barrier to access, including low levels of education and misunderstanding of western medicine, and lack of knowledge on what services were available and how to apply. Communication and language barriers were also perceived. A 'cope and don't fuss' approach among African-Caribbean hypertensive patients was a reported barrier to accessing services (+).

Evidence Statement 24ST

One qualitative study of cardiac rehabilitation coordinators in Scotland (+) found that age was widely perceived to influence access to services, both during initial assessment and in assessments for exercise components. Focus groups revealed that staff appeared to have knowledge of the benefits for older people but that scarcity of resources prevented them offering more accessible and appropriate services.

Cost-effectiveness evidence

Smoking cessation

The cost per quality-adjusted life year (QALY) of smoking cessation interventions for disadvantaged groups is low or very low. It is rarely likely to exceed £6000.

Statins

Secondary prevention of CVD (that is, after a CVD event) among a disadvantaged population costs an estimated £4000 per QALY gained (£3100 per QALY for finding the person and £900 per QALY for treating them with statins). Therefore, it is cost effective.

Whether or not it is cost effective to provide statins to prevent a first occurrence of CVD among a disadvantaged population depends on the number of people at risk in the baseline population. Data from a USA study of financially disadvantaged women aged 40–64 who enrolled in the National Breast and Cervical Cancer Early Detection Program was analysed. The analysis found that it is cost effective if more than 14% of the population is at risk. For example, when 40% were at risk of CVD, primary prevention was estimated to cost £8500 per QALY gained (£4900 per QALY for finding the person and £3600 per QALY for treating them). This compared with about £125,600 when only 1.6% were at risk (£122,000 per QALY for finding them and £3600 per QALY for treating them).

Fieldwork findings

Fieldwork aimed to test the relevance, usefulness and the feasibility of implementing the recommendations and the findings were considered by PHIAC in developing the final recommendations. For details, go to the fieldwork section in appendix B and 'Reducing the rate of premature deaths from CVD and other smoking-related diseases: finding and supporting those most at risk and improving access to services'.

Fieldwork participants who work with adults who are disadvantaged (in particular, those who smoke and/or are eligible for statins and/or are at high risk of CVD due to other factors) were very positive about the recommendations. Some said they will support work already being carried out in this area.

Participants felt that incentives had a role to play in helping to encourage people who are disadvantaged to attend NHS services and complete treatment. However, they felt that the use of incentives should be driven by national policy.

Overall, the lists of 'target populations' and 'who should take action' were seen as appropriate, although participants believed it would be helpful to include commissioners in the latter. Highlighting who should have overall responsibility for a recommendation would also aid implementation, they said.

Participants highlighted training, long-term funding, partnership working and cultural sensitivity as key issues that needed addressing for successful implementation of the recommendations.