Background

Behaviour is recognised as a key determinant of health, with modifiable lifestyle behaviours, such as smoking, physical activity, unhealthy eating and excessive alcohol use, resulting in significant morbidity and mortality.¹ There is a substantial knowledge base with respect to effective lifestyle intervention approaches. However, the successful translation of this into practice is a continuing challenge.² The consequent individual and societal costs are considerable. These major health risks tend to be more prevalent among lower socioeconomic groups and, consequently, large sociodemographic differences exist in both experiences and expectations of health.^3–6 With respect to the UK context, the Public Health White Paper Choosing health: making healthy choices easier sought to address this issue by taking action to encourage and enable individuals to make healthier choices, with a particular focus on those living in disadvantaged communities.⁷ It recognises the central importance of changing behaviour to improve population health and also builds on the vision of a ‘fully engaged scenario’, in which people take control of their own health and the wider determinants of ill health are addressed.⁸

Approaches to health-care provision are therefore changing in recognition that clinical and curative foci are unsustainable, inappropriate or insufficiently effective.^9,10 Many Western health-care systems are currently undergoing a shift from paternalistic to partnership models of care, with policy-makers, clinicians and consumers all seeking ways to promote increased involvement of patients and the wider public.¹¹ There is therefore a movement in public health approaches ‘from advice from on high to support from next door’ (p. 13).⁷ These shifts in policy require an expanded portfolio of public health interventions, including an expanded workforce continuum, in order to effectively address the health needs of both the general population and the most vulnerable groups in society.

The introduction of new roles or the expansion of existing roles to deliver HRLA or training represents one response to these developments. In the UK, NHS health trainers were introduced in the Public Health White Paper Choosing health,⁷ as one element of a wider workforce, offering a range of approaches to helping people change their behaviour in relation to their health. They are described as ‘people who are in touch with the realities of the lives of the people with whom they work and connected through a shared stake in improving the health of the communities that they live in’(p. 106)⁷ and ‘Offering practical support instead of preaching, and good connections into the advice and support available locally’ (p. 106).⁷ It is also recognised that a one-size-fits-all approach will not be appropriate, noting that ‘different neighbourhoods will need different types of health trainers’ (p. 106) and that different models of provision will be required to achieve best outcomes for different individuals and communities.

Versions of the health-related lifestyle advisor (LA) role represent a strategy that has been widely used to promote behaviour change and self-care across diverse conditions and population groups.^12,13 It is becoming increasingly important in health-care environments that are challenged by limited financial and human resources, enduring inequalities issues and expanding populations with chronic diseases.^8,14,15 Much of the formal literature describing peer-based models comes from North America, where health promotion and disease prevention programmes that rely on LHAs have proliferated since the 1970s.¹⁶ Research has shown that people are more likely to hear and personalise messages, and thus to change their attitudes and behaviours, if they believe the messenger is similar to them.¹⁷ In addition, peer-based interventions can often be implemented economically, allow for direct involvement of clients and can result in long-term benefits for the peer educators themselves.¹⁸ Preliminary work conducted in relation to the implementation of health trainers in the NHS identified a range of models varying by degree of targeting and mode of delivery.^19,20 However, it is not currently known what the effects of these various models are on health outcomes. It is therefore timely to bring together the available data on the impacts of HRLA or training to determine how effective the various approaches are. Drawing on both qualitative and quantitative research, this report synthesises the evidence on the component intervention techniques, effectiveness, cost-effectiveness, equity and acceptability of different versions of the health-related LA role in improving health and well-being in the UK.

Terminology

The term ‘LHA’ belongs to a group of roles that have been given, over time, a range of titles, but which have some common principle of recruitment, purpose or operation. These include ‘natural helpers’,²¹ ‘peer educators’,^22,23 ‘lay health advisors’,¹⁶ ‘lay volunteers’,²⁴ ‘community health advisor’,² ‘community health aides’,²⁵ ‘peer counsellors’,²⁶ ‘lay health volunteers’,²⁷ ‘navigators’,²⁴ ‘community health workers’,¹² ‘health trainers’,⁷ ‘community guides’,²⁸ ‘indigenous encouragers’,²⁹ ‘buddy’³⁰ and ‘telecarer’.⁹ There does appear to be a consistent term ‘promotora’, used in Spanish-speaking communities. Summarising this diversity, Devilly et al.²³ suggest that peer education constitutes an umbrella term covering a ‘range of different approaches including peer training, peer facilitation, peer counselling, peer modelling or peer helping’ (p. 221).²³

The variety of language has been highlighted as an issue of note.¹² There are, for example, consequences for clarity of role and expectations of impact. Opportunities for comparability of impact and effectiveness are also inhibited. With respect to this review the search strategy had to be particularly broad, utilising complex search strings. For clarity in this report, the intervention will be referred to as HRLA and the person delivering it as an LA.

History of the LA role

Accepting the LHA as an umbrella term, the role has a considerable history, more so in other parts of the world than in the UK, and with particular focus on certain health needs. For example, Earp and Flax¹⁶ report a 30-year history of the development and increased utilisation of the role in the USA with respect to health promotion and disease prevention programmes. Similarly, Bishop et al.²¹ report an increase during the 1990s in the development of links between communities and service providers through the training of indigenous community members.

The role of LA is more established in some fields, for example breastfeeding, sexual health, screening, chronic conditions/Expert Patients Programmes. Emerging roles are appearing, for example as exemplified in the development of the role for health improvement activities with offenders.²³

The development of the role has not been unproblematic and reference to the World Health Organization (WHO) report on community health workers (CHWs)¹² provides an eight-item list of areas of potential weakness:

1. minimal policy and organisational commitment – vertical programmes, implemented with little professional interest, structural, political and economic factors neglected, lessons not learned from other sectors
2. poorly defined functions
3. poor selection
4. deficiencies in training and continuing education
5. lack of support and supervision
6. uncertain working conditions
7. undetermined cost and sources of finance
8. lack of monitoring and evaluation.

It seems reasonable to assume these issues may potentially apply to the LA role.

This brief review highlights that the LA role has had a precarious history and diversity of development that has not always benefited from rigorous evaluation.

Definitions and distinguishing features

As with role titles, there is also ambiguity with respect to role definitions. Significant debate has been devoted to attempting to clarify the role and what distinguishes it from other intervention approaches. Some definitions are offered here to both assist the process of distinguishing the role boundaries and characteristics and highlight the inherent challenges: ‘community members who work almost exclusively in community settings and who serve as connectors between health care consumers and providers to promote health among groups that have traditionally lacked access to adequate care’ (p. 1055);³¹ ‘members of the communities where they work, should be selected by the communities, should be answerable to the communities for their activities, should be supported by the health system but not necessarily a part of its organisation, and have a shorter training than professionals’ (p. 6);³² ‘CHW must be of the people they serve. They must live with them, work with them, rejoice with them, suffer with them, grieve with them and decide with them’ (p. 6).³²

Being ‘of the community’ is recognised, however, as a complex issue. The WHO Study Group on strengthening CHW performance recognises that:¹² ‘Community is not a homogeneous group – its members can have strong conflicts of interest. In this report, the word community is therefore used in the geographical sense of the population potentially served by a CHW; there is no assumption that such social groupings cooperate harmoniously in everyday affairs’ (p. 16).

Walt³³ identifies that traditional definitions of the role are being challenged as new derivations emerge. She reports that until the 1980s CHWs were ‘people who were selected by the community, resident in the community and from the community’ (p. 3). These foundations are seen to be challenged, for example, when the degree of the relationship and the affiliations of the worker with the health-care system are strengthened or formalised, by issues of volunteer or financially remunerated worker, selection by the service provider rather than the service recipients. Transition from ‘community’ member to paid employee, as is the case for some LHAs, is an issue worthy of scrutiny. If, and how, this changes the individual's and /or the communities' perception of and relationship with the individual is open to debate.³⁴ Braithwaite et al.,³⁵ when exploring the experiences of community members who were involved in action research, found the transition from community member or voluntary worker to a paid researcher to change the way that LHAs were perceived by community members.

In summary, the distinguishing features of belonging to a community are highly complex.³⁶ At any one point in time, one LA individual may belong to several ‘communities’, such as gender, age, geography, religion and occupation, and the challenge arises with respect to which affiliation to prioritise, or which results in the most effective health improvement intervention.

As an alternative to a community affiliation as a distinguisher, it may helpful to refer to Ungar et al.'s³⁷ discussion on the drivers for such role development in social care, which they identify as increased recognition of the value of indigenous knowledge (IK).

Indigenous knowledge can be broadly defined as the knowledge that an indigenous (local) community accumulates over generations of living in a particular environment.³⁸ This definition encompasses all forms of knowledge – technologies, know-how skills, practices and beliefs – that enable the community to achieve stable livelihoods in their environment.³⁸ A number of terms are used interchangeably to refer to the concept of IK, including ‘traditional knowledge’, ‘indigenous technical knowledge’, ‘local knowledge’ and ‘indigenous knowledge system’.³⁸

Indigenous knowledge is unique to every culture and society, and it is embedded in community practices, institutions, relationships and rituals.³⁸

Indigenous knowledge is based on, and is deeply embedded in, local experience and historic reality, and is therefore unique to that specific culture; it also plays an important role in defining the identity of the community.³⁸

Rationale for role/intervention

The LA role is generally used to achieve three broad aims: (1) access to communities or individuals who are in some way marginalised from the mainstream; (2) access from marginalised communities into the health and social care systems; and (3) alternative delivery mechanisms to professional provider. Varying degrees of detail and distinction on each of these aims have been reported, with the level of sophistication developing over time and role history.

Referring to the role of ‘indigenous helpers’, Reiff and Reissman³⁹ identify two distinct role intentions: one they describe as ‘expediters or service agents’ and the other as ‘care aides or therapeutic agents’. Witmer et al.³¹ differentiate role rationale under four headings: increasing access to health care; improving quality of care; reducing costs of care; and broader social contributions.

With respect to the use of peer education as a health promotion intervention, Turner and Shepherd⁴⁰ provide a list of 10 rationales:

1. More cost-effective.
2. Peers are credible.
3. Peer education is empowering.
4. Uses already established means of communication/information transfer.
5. People identify with peers and so peers are more successful than professionals.
6. Can act as positive role model.
7. Beneficial to those involved in providing it.
8. May be more acceptable than other education provider.
9. Reaches those hard to reach through conventional methods.
10. Reinforcement of learning through ongoing contact.

In summary, the purpose and aims of the LA role are broad and varied.

Theoretical basis of health-related lifestyle advisor

The theoretical basis of LA interventions is another debated issue, and one for which there is inconsistent reference in the literature on the topic. Although potentially only a dimension of HRLA, Turner and Shepherd⁴⁰ describe peer education as ‘a method in search of a theory rather than the application of theory to practice … Although located broadly within the field of social psychology, peer education does not appear to have its roots within a particular school of thought’ (p. 235).

Drawing on Turner and Shepherd's⁴⁰ work and a general review of the LA-type role, this report highlights a range of possible theoretical underpinnings: social network theory, social learning theory, self-efficacy theory, social inoculation theory, role theory, differential association theory, subcultures theories and communication of innovations theory.

Mechanism and models of intervention

As distinct from a theoretical basis, most reports of LA activity do make reference, even minimally, to mechanisms of intervention.

With respect to the lay health worker as a distinct intervention provider, the mechanisms may be grouped into three broad categories. One category is mechanisms that address embellishment of standard care, such as the provision of a ‘bridge’ between communities and service providers,³³ a ‘complement’ to formal systems,¹⁶ a ‘link’ between communities and organisations.⁴¹ Another is the provision of social network support for behaviour change messages and activity.⁴² Schulz et al.⁴⁰ further differentiate support into affective support (caring, trust, love), informational support (advice, suggestions, information) and instrumental support (tangible aid and services). The third mechanism is style of information transmission, which can range from repeated message provision in several social contexts to individual one-to-one tailored message giving.⁴²

With respect to models of provision, working alone or in partnership with another provider are two clear distinctions. For example, Nunez et al.⁴³ report an approach that combines the knowledge of a nurse with an advocate's understanding of the social reality of the community as a ‘package’ of provision.

Challenges of evaluating public health interventions

The WHO¹² endorses the effectiveness of CHWs by reporting: ‘They have achieved much in many countries at different times, but shortcomings of CHW programmes are often imputed to the CHWs themselves. However, this debate is a sterile one: there is no longer any question of whether CHWs can be key agents in improving health; the question is how their potential can be realized’ (p. 9).

These comments can potentially be applied to the more generic LA role. Evidence of effectiveness is not readily available and is hindered by acknowledged evaluation challenges for this type of service provision. Twenty years ago, Walt³³ described the methodological difficulties as ‘enormous, further hindered by the financial resources required to conduct rigour evaluation designs and a limited service provision history’. This comment still applies, as the practical difficulties in measuring the impact of public health interventions remain unchanged. Indeed, public health interventions may need to adapt to local circumstances and needs, preventing tight control of the intervention:⁴⁴ randomisation of community-based trials can be difficult⁴⁵ – possible contamination may preclude individual randomisation, and randomisation at community level may be beyond the resources of the trial⁴⁶ – and measurement of lifestyle changes inevitably relies on self-reported data, as observation of health improvements at community level is rarely feasible given the size and duration of a typical study.⁴⁷

The impact of lifestyle and behaviour changes in terms of health gains is often not manifest until old age.⁴⁸ A measure of effectiveness almost inevitably necessitates extrapolation of health benefits from surrogate markers and measures of lifestyle changes. These benefits are dependent on the maintenance of lifestyle changes.⁴⁹ Considerable literature is available in certain disease areas allowing estimates of the health gains from changes in behaviour. However, little evidence is available on the long-term maintenance of lifestyle/behaviour changes.⁵⁰ Most of the available evidence comes from the smoking cessation literature, which suggests that 65%–75% of quitters at 1 month will relapse at 12 months.^51,52 A further 35%–54% of those abstaining at 12 months will subsequently relapse.^53,54

Further challenges in evaluating public health interventions arise from the complex nature of these interventions. Interventions aimed at changing lifestyles inevitably interact with the social environment in which they are delivered.⁵⁵ The environment shapes and modifies the effect of the intervention.⁵⁶ Subtle differences in social environment may have a significant modifying effect on the impact of the intervention.⁵⁷ The intervention may also modify the social environment in terms of attitude towards health improvement and empowerment to make changes. While the impact of the intervention can be captured within the social environment studied, generalising the effects of the interventions to other contexts may not be possible. A thorough understanding of how the intervention works might be necessary before a judgement can be made on whether that intervention can be transferred to another context.⁵⁸

The impact of the intervention on the social environment brings additional and unique challenges to public health evaluations. It requires consideration of the possibility of benefits (or harms) that extend well beyond the recipients. The impact of any particular intervention on the social values may be too small to detect.⁵⁹ Nevertheless, it is clear that social norms and lifestyles can and do change. The decline in cardiovascular disease (CVD) across the Western world over the last 30–40 years⁶⁰ is only partly a product of medical intervention – significant changes in diet have taken place.^61,62 However, the impacts of health-promotion programmes on changes in attitudes to diet are difficult to quantify.

Unsurprisingly, then, ‘many LHA programmes are only minimally evaluated, if at all, and little published information is available about LHA evaluation strategies’ (p. 443).⁶³ This situation is a consequence of the need for evaluation activities to not disturb the spontaneous and informal processes of natural helping, the difficulty in intermediate outcome measurement of unstructured roles and the generally modest evaluation budgets available to such interventions. Despite recent methodological developments in the public health and health improvement fields, these comments remain highly pertinent.