The committee's discussion
See evidence reviews for details of the evidence.
The committee agreed that many older people are already involved in activities that keep them independent and maintain and improve their mental wellbeing. Members also agreed that many such activities are not always seen as contributing to mental wellbeing or keeping someone independent.
The committee agreed that ageing is an individual experience and that not all approaches may be right for everyone – or certainly not at the same point in their lives. Members discussed how risk factors build up and then result in a decline in independence and mental wellbeing. But they also acknowledged that people have different levels of resilience. The committee was aware that not everyone who could be assessed as being at 'high risk' will experience poor mental wellbeing.
The committee took an 'assets‑based' approach (Ageing well: an asset based approach Local Government Association) when developing this guideline. This involves taking a broad view of factors or resources that help people, communities and populations to maintain and sustain health and wellbeing.
The committee noted that many older people who are at risk of a decline in their independence and mental wellbeing may not identify themselves as such.
No evidence was identified on people aged 55 and over who are at risk of the same health conditions as people aged 65 or older. The committee agreed to make a research recommendation for mid‑life interventions.
The committee was concerned that if there was not enough choice in terms of activities – and if people were not given the opportunity to say what they would like to do – this could be detrimental to people's mental wellbeing. For example, it may lead to some people being excluded, or it may lead to conflict over the choice of activities.
The committee was aware of the risk of widening inequalities if activities only reach people who already use services. The committee agreed that inequalities could be avoided by recommending a variety of interventions and providing help to access services, including reducing the barriers to access, as described elsewhere.
The committee recognised that promoting social activities outside the home and involving a range of people working in the community in those activities could make older people more vulnerable to crime. For example, theft from an unoccupied home or fraud through 'bogus callers'. The same is true of training to help older people use information and communication technologies – they could then be susceptible to internet‑based scams.
The committee agreed that this potential problem could be overcome by using governance arrangements and by providing advice, training and support.
The types of effective intervention identified by the committee link to the Foundations of Mental Wellbeing model (see expert paper 1). According to the model, 4 key 'pillars' contribute to positive mental health and wellbeing: functional ability, psychological attributes, power and resources and 'social connectedness'. These change throughout life.
The committee noted that other similar models are available (for example, Lay perspectives of successful ageing: a systematic review and meta-ethnography Cosco et al.).
The committee discussed different approaches to identifying vulnerable older people. Members noted that risk assessments of individual circumstances and needs are useful. However, they agreed that this guideline should focus on a general assessment of local need using routine data, such as information available from the census or Public Health England, or drawing on the knowledge of people working in the local community. It could also involve taking account of 'key life events' such as bereavement or divorce.
The committee identified groups that may need specific approaches and activities. These included: men, people older than 85, carers and people with a sensory impairment. But there is a lack of evidence on effective activities for these subgroups. Members also noted that people's needs and interests vary within any group.
The committee was aware that the evidence reviews, economic evaluation and expert testimony on the effects of specific interventions did not represent all relevant activities. In addition, it was not possible to identify specific interventions or features (such as the ideal length of an intervention) to allow 1 activity to be prioritised over another. However, the committee agreed that broad types of intervention do appear to be effective in the current UK context.
The committee agreed that people would only be able to choose which services or activities to get involved with if they were given enough information about what was on offer. Members discussed the idea of a local 'repository' of information (based on set criteria) using the internet. They also acknowledged the difficulties involved in maintaining such a resource – and that using online information is a significant challenge for some people. So there is not a specific recommendation on this.
The committee noted that there was a lack of evidence on the differences between activities provided in urban and rural settings. Members acknowledged that it may be harder to provide some activities in a rural setting where populations are more dispersed.
When developing recommendations, the committee considered relevance to the scope of this guideline, strength of evidence, applicability, evidence on cost effectiveness, expert knowledge of practice and broader social value judgements.
It agreed which activities described in the review evidence should be developed into recommendations. Members also agreed which activities should not be included in this guideline. For example, mentoring of older people, health education, a national arts programme, attitudes of health and care professionals toward older people, and self‑management ability (evidence statements 1.1.3, 1.1.4, 1.1.6, 1.2.4 and 1.5.1 respectively). The committee agreed there is considerable uncertainty as to how effective these interventions would be across different settings.
The committee used the evidence to prioritise areas for implementation (see implementation: getting started).
Recommendations in this section are linked to evidence statements: 1.1.1, 1.1.2; review 3; review 4; EP1, EP3, EP4, EP6; IDE.
The committee agreed that the following are important principles:
Improving the types and variety of provision of interventions, based on the population's needs and interests.
Targeting interventions at those at greatest risk of decline.
Involving older people in the design and delivery of interventions. This is crucial to ensure interventions are acceptable to the target population (expert paper 1).
Based on their own experiences, committee members agreed this was likely to increase the number of people who use and benefit from the interventions. They also thought this approach would most effectively reduce risk and rates of decline and subsequent morbidity in the target populations.
The committee drew on expert papers 3, 4 and 6 and reviews 3 and 4 when formulating this recommendation, in addition to evidence from review 1. The evidence from review 1 included evidence statements:
1.1.1 on multi‑location activities, which was inconsistent
1.1.2 on single location activities, which was of poor quality but positive and consistent.
Overall, the committee thought that activities should have a regular location and be held at regular times to be most effective.
The committee recognised it did not have cost effectiveness evidence to support these recommendations. However, because they are general principles on which all interventions should be based, they were not considered to have a measurable cost impact or additional resource implications beyond the interventions themselves.
The committee was confident that using these principles would increase the likelihood that interventions would both benefit the target population and be cost effective.
Recommendations in this section are linked to: evidence statements 1.1.1, 1.1.2, 1.1.5, 1.1.7, 1.2.1, 1.2.2, 1.2.3, 1.4.1, 1.4.2, 1.6.1, 1.6.2, 1.6.3, 1.6.4; review 3; review 4; EP3, economic modelling report; IDE.
Recommendation 1.2.1 The evidence on multicomponent programmes included evidence statements:
1.1.1 on multi‑location activities, which was inconsistent
1.1.2 on single location activities, which was of poor quality but positive and consistent.
The committee was confident that, for most older people, the activities specified (see below) will prevent a decline in independence and support their mental wellbeing. There was strong, consistent evidence of effectiveness, including good to moderate quality studies with a low likelihood of bias.
Strong cost effectiveness evidence was also reported for singing, intergenerational, and information and communication technology activities. This included the economic modelling and an economic evaluation identified in review 5.
All the evidence was considered applicable and transferable to UK practice.
The committee discussed the evidence on singing and noted that it is unclear whether it is the singing itself that produces the benefit, the group‑based nature of the activity or something else. But members agreed that the evidence (evidence statement 1.1.5) demonstrated a clear benefit. This was further supported by the consistent direction of effect in the evidence. In addition the best quality evidence, from a randomised controlled trial (RCT), counterbalanced uncertainty from the poor quality before‑and‑after study.
The committee also noted that the evidence on singing was based on professionally‑led programmes but there was no strong evidence that singing programmes are only effective if there is a paid lead. The singing recommendation acknowledges the value of leadership and professional qualities of the lead.
The committee agreed that the evidence (evidence statement 1.1.7) demonstrated clear support for a range of art‑ and music‑based interventions. This was further supported by the consistent direction of effect in the evidence.
Some studies may have some inherent uncertainty because of their type or quality. But the committee was confident that the evidence base indicated a benefit and implied such activities would be a good use of limited resources. This was supported by members' own experiences of delivering such interventions.
The committee agreed that the group‑based activities in recommendations 2 and 3 in NICE's guideline on occupational therapy and physical activity interventions to promote the mental wellbeing of older people were relevant and should be referred to in this guideline.
The committee agreed that the evidence demonstrated a clear benefit: evidence statements 1.2.1, 1.2.2 and 1.2.3.
Only the RCT was of good quality, with a design to control bias. But the direction of effect of the evidence was consistent and the committee was confident that, on balance, these interventions are beneficial. The committee also noted that if the principles laid out in recommendation 1.1 are followed, then the likely benefits would outweigh any uncertainty regarding the evidence base.
In addition, committee members reflected on their own involvement in interventions of this type. They agreed that these interventions were a good use of resources and would provide benefits beyond those captured in the data for both the target population and for the younger people involved. They also agreed that this would mean the positive outcomes would be underestimated.
Recommendation 1.2.2 The committee noted that there was a substantial evidence base on activities relating to information and communication technologies, and some evidence on education and learning.
The evidence comprised evidence statements:
1.4.2 on the benefit of education via the internet and other electronic media, with weak but consistent evidence
1.6.1 on the benefit of training to use personal computers and the internet, with inconsistent evidence
1.6.2 on the positive effects of telephone‑ and internet‑based communication, with weak but consistent evidence
1.6.3 on the positive effects of information and communication technologies on older people who are carers, with weak but consistent evidence
1.6.4 on the improvement in mental health and positive mood among older people from using computer or console games, with weak but consistent evidence.
The committee also noted the results of an economic evaluation of a computer skills training course that led to increased internet usage by older people and cost approximately £564 per person. Members agreed this could be linked to a number of improvements in older people's wellbeing.
There is some uncertainty because of the mixed quality of the studies. But the committee agreed that, on balance, and from their expert opinion and experience, the interventions are likely to be of benefit. As a result, it recommended that providers consider them.
The committee noted that there was some evidence on education and learning opportunities for older adults (both face‑to‑face at university and in community settings and via the internet). The evidence comes from expert paper 3 and evidence statements:
1.4.1 on an improvement in psychological measures and wellbeing, with weak but consistent evidence
1.4.2 on an improvement in wellbeing from education provided via the internet and other electronic media, with weak but consistent evidence.
The direction of effect in the evidence statements was consistent. But the overall quality of the studies leads to some uncertainty, because the weak methods used could produce a different result if they were re‑run. Based on the committee's expertise members agreed that, on balance, these interventions are likely to be of benefit and recommended that providers consider them.
No evidence was identified in the effectiveness review on hobbies (such as gardening) or interests (such as current affairs) but the committee noted that this does not necessarily mean they would not be effective. Members noted from review 3 that varied programmes involving hobbies, interests, physical activity and learning were being used in practice in the UK to improve older people's independence and mental wellbeing.
The committee thought it likely that all the interventions cited will benefit some people and may be cost effective. But the interventions listed in recommendation 1.2.1 are more likely to be cost effective. That is because they are based on more consistent evidence of effect from higher quality studies and are supported by evidence of cost effectiveness.
The choice of intervention, and whether or not to get involved, is likely to depend more on the person's values and preferences than whether an intervention has been shown to be effective. The committee therefore believes providers should spend time considering wider options to ensure the interests of all older people can be covered. That is because it is likely that involvement in any kind of group activity aimed at maintaining their independence and mental wellbeing will have some measurable benefit.
The recommendation in this section is linked to: evidence statements 1.3.1, 2.6; review 3; IDE.
The effectiveness evidence was of mixed quality but consistent, and the committee was confident about the overall benefit of this type of intervention. The effectiveness evidence on interventions to build friendships demonstrated a clear benefit (evidence statement 1.3.1). The following also support a strong recommendation:
economic modelling data on friendship programmes
evidence statement 2.6 on the barriers to, and facilitators of, 'social connectedness'
examples in review 3.
Recommendations in this section are linked to: evidence statements 1.2.1, 1.2.3, 2.2; review 3; review 4; EP4; IDE.
The committee discussed evidence that 'reciprocity' (exchanging things with others for mutual benefit) and 'payback' (benefit resulting from a previous action) both have a positive impact on the mental wellbeing of people involved in voluntary work. Members also discussed the fact that many older people want to contribute to their community.
The effectiveness evidence on volunteering demonstrated a clear benefit. It comprised evidence statements:
1.2.1 on school‑based intergenerational activities (including older people reading to children or offering life‑skills training) improving older people's mental wellbeing, with moderate, consistent evidence
1.2.3 on intergenerational activities involving volunteering having a positive effect on older people (this included older people providing mentoring and other support and activities) with weak but consistent, positive evidence.
The studies have potential for bias. But, the positive direction of the evidence was consistent and, in their expert opinion, the committee was confident that volunteering would have a positive impact. The additional qualitative study supported this view by providing an understanding of the mental wellbeing benefits as perceived by older people involved in volunteering.
The favourable economic modelling evidence (using 1 RCT from Japan) also supported a strong recommendation. The committee acknowledged that volunteering programmes do have an associated cost and should not be 'done on the cheap'. But members also agreed that they provide benefits to the local economy in terms of the time, skills and support that volunteers offer.
The committee noted evidence (evidence statement 2.2) on barriers to, and facilitators of, older people volunteering. It discussed the factors presented in review 2 (based on 8 studies that used survey and interview methods) and members' expert knowledge of practice. It was agreed that awareness‑raising and a range of recruitment methods were key.
Recommendations in this section are linked to: review 3; EP1, EP2, EP3, EP5, EP6; IDE.
The committee considered it important to take a preventive approach to all those at risk. But members agreed to focus on those at greatest risk due to limited local authority funds and because this would be a way of addressing health inequalities.
The committee noted that some of the data needed to identify people most at risk may need to be collected locally because it is unlikely to be available via routine national data collection sources. (National data sources, for example, are unlikely to include people who have recently experienced or developed a health problem.) The committee also recognised that the type of data collected, and how complete it is, may be inconsistent across areas.
Members noted that although extra data might need to be collected locally, the fact that it could lead to earlier identification of, and support for, people at greatest risk would outweigh any extra resource implications.
The recommendation includes only groups at high risk cited in the expert testimony (expert paper 5). The impact of factors such as not having children, or factors linked to gender and sexual orientation, is not clear. But it was clear, for example, that the death of a partner in the previous 2 years could have a significant impact on someone's mental wellbeing.
The committee also drew on evidence from expert papers 1, 2, 3 and 6 and review 3 for this recommendation because the data either support the identification of people at most risk or give examples of how this has been done locally.
This is a strong recommendation because the evidence was clearly demonstrated in the expert testimony. In addition, the committee thought this recommendation would have limited opportunity or other cost implications, because staff in contact with older people are already visiting those most at risk. It recognised that some training may be needed. But it believed that any cost impact would be outweighed by the early identification of, and support for, those at most risk.
Based on the evidence presented and members' expertise, the committee discussed how the recommended interventions could be implemented. These discussions, and the evidence on which they are based, are outlined below. The actions proposed are not formal recommendations. (For details of these actions see implementation: getting started.)
The committee noted evidence from: review 3 (the practice mapping review) and expert papers3, 4 and 6 (plus IDE). It agreed plans developed by a local partnership could lead to a number of benefits, including:
more widespread knowledge of what is available
reduced duplication or better economies of scale by sharing resources
more effective identification of those in greatest need.
The only resource implications for operating the partnership may be the opportunity costs for attendance at meetings. But the committee noted that the added benefit from a more joined up approach should outweigh any resource implications.
The committee agreed that effective partnerships may only contain some of the representatives suggested for implementation. But the list provides a basis for establishing such groups. It also agreed that effective partnerships would always include older people and their representatives.
The committee noted the importance of including home improvement agencies in the partnership because they are likely to come into contact with people at risk on a regular basis.
The committee discussed the need to raise awareness of the importance of older people's independence and mental wellbeing among:
older people themselves (including those who are carers)
local policy makers
Committee members noted from their own experience that identifying a lead person to review and update the joint strategic needs assessment and health and wellbeing strategy would help to ensure the task is completed.
The committee noted evidence from: review 3; review 4; EP3; IDE.
The committee discussed the importance of conducting a local needs assessment so that local authorities can consider their population needs and plan to meet those needs.
Evidence on the use of needs assessments was found in reviews 3 and 4. But no evidence on the benefits and harms was considered by the committee because it is a requirement of health and wellbeing boards under the Health and Social Care Act.
The committee agreed that, on balance, needs assessment offers many benefits, not least because it gives local areas the detail they need for effective decision making. In addition, the data and information generated can be used as a baseline for service evaluation and drive an outcome‑focused approach.
The committee agreed that it was important to collect data from appropriate sources to inform the needs assessment. Without this, the needs assessment will not be fit for purpose.
The committee noted there was limited evidence on the best way to identify needs and address barriers to older peoples' participation (from expert testimony and review 3). For example, there was a lack of evidence in terms of gender, sexuality, disability, income or ethnicity.
In the committee's expert opinion, needs assessment can most effectively be used to support service design if people are asked to express their own needs as part of the planning process.
The committee agreed that feeding the results into the joint strategic needs assessment can help influence plans for the wider determinants of health, such as housing. It may also bring economies of scale compared with individual needs analysis in separate commissioning streams. It can highlight areas of unmet need and aid decisions about allocation of resources.
The committee noted evidence from: review 3, review 4 (practice mapping reviews); EP3, EP4, EP6; IDE.
The committee heard from experts about several 'wellbeing coordinator' models. These community‑based coordinators identify older people who need support and then coordinate local services and activities for them, as well as acting as a local source of information. The committee noted there were no economic effectiveness data on this.
The committee agreed that local authorities may consider appointing someone if there is a high level of local need. Members recognised the resource implications of appointing a coordinator. But they also felt that this could be important in areas with a higher than average ageing population at particular risk (for example, in rural communities). In such cases, any resources used to coordinate local activities could be offset by the cost of providing other services, as older people lose their independence or mental wellbeing.
The committee acknowledged that some geographical areas may be too large for 1 coordinator. It also agreed that information sharing within the partnership would help to avoid duplication of effort and connect older people at risk with appropriate services and agencies.
The committee noted evidence from: review 3, review 4; EP2, EP3, EP4, EP5; IDE.
The committee felt this was an important action and would have limited resource implications, because local authorities have to consult with communities regularly and can use forums already established to support this.
The committee noted that the cost of participation can be a barrier for some people, as demonstrated in expert paper 5.It agreed that helping maximise older people's income so they can get involved would be a positive move. This could include help with the cost of a bus or taxi fare or benefits advice, all of which local government can provide.
The committee noted evidence from: evidence statement 1.1.8; review 3; EP2; IDE.
The effectiveness evidence (evidence statement 1.1.8) was of mixed quality because of the varied study types and the potential for bias. But it did show consistent positive effects in terms of support for older carers.
The committee also drew on evidence from expert paper 2 and review 3 (highlighting the decline in mental wellbeing of older carers and examples of how they can be supported locally).
In addition, members recognised that carers may value emotional support from other carers, so activities for groups of carers may be beneficial. Members also noted that the Care and Support Act 2014 specifies that care practitioners should assess and provide the support carers need. This includes emotional support.
As a result, the committee believes these activities should be offered to older carers to reduce their risk of decline and to provide a broad range of support to help them participate in local activities to reduce their risk of isolation. Because local authorities are already required to support carers, the committee noted that it might be a matter of reallocating – rather than finding new – resources to implement these activities.
The committee noted evidence from: review 3, review 4: EP3, EP4, EP6; IDE.
The committee wanted to make a recommendation because of the importance of evaluation in sustaining and improving services and activities. But the evidence base was very limited (unpublished examples of evaluation in reviews 3 and 4 and support for, and further examples of, evaluation in expert papers 3, 4 and 6).
The committee noted that providing community organisations with support for evaluation may have some resource implications, but that any short‑term set‑up costs would be more than offset by improved outcomes in the community. This includes developing sustainable community assets and support.
Members noted that local solutions, based on expressed community needs, are preferable to continuing with an approach simply because it is the way projects have always been run. They also noted that there may be new ways to use existing resources; for example, by using social prescribing by GPs or self‑managed budgets. Members agreed that it would be useful to find out more about funding arrangements as part of any evaluation.
The committee considered that the cost effectiveness evidence identified in the literature review was limited, and with limited applicability to England. Therefore a new economic evaluation was developed using a cost–consequence analysis and a cost–utility analysis.
The committee felt that a cost–consequence analysis was the most suitable type of economic analysis, given the wide range of outcomes that are relevant to interventions to maintain and improve older people's independence and mental wellbeing. Where data permitted, the committee agreed a cost–utility analysis would be useful (albeit limited in scope) for comparing the cost effectiveness of different types of interventions using a common outcome.
The committee highlighted the complex nature of the evidence, in particular the inter‑relationship between independence, mental wellbeing and other health and non‑health outcomes. The fact that independence and mental wellbeing are also reported as outcomes in their own right was noted as a further complication.
In addition, there is a lack of published studies demonstrating a causal relationship or direction of any causality between the range of measures and outcomes. Members agreed that this meant the economic analysis would be an oversimplification of the scope of activities and outcomes.
The evidence reviews and expert testimony identified a vast array of different activities and interventions. The interventions selected for economic analysis represented the different types of interventions identified in the effectiveness reviews.
As with any economic analysis undertaken during guideline development, the results are subject to uncertainty and numerous assumptions. Nevertheless, based on the examples used in the present analysis, the committee considered that the types of interventions tested can be cost effective or even cost saving, and thus represent a good use of public money.
The committee noted that there may be a difference between the sector or organisation that pays for some of the proposed interventions and the sector or organisation that apparently benefits. For example, if a social care or other local authority budget is used to fund activities that primarily achieve a health benefit. Members acknowledged the difficulty for commissioners in such cases.
Details of the evidence discussed are in evidence reviews, reports and papers from experts in the area.
The evidence statements are short summaries of evidence. Each statement has a short code indicating which document the evidence has come from.
Evidence statement number 1.1 indicates that the linked statement is numbered 1 in review 1. Evidence statement number 2.1 indicates that the linked statement is numbered 1 in review 2. Evidence statement EP1 indicates that expert paper 1 'Theoretical model relating to independence and mental wellbeing of older people' is linked to a recommendation. EP2 that expert paper 2 'Carers' is linked. EP3 that expert paper 3 'Practice in England – local town council approach' is linked. EP4 that expert paper 4 'Age friendly cities' is linked. EP5 that expert paper 5 'Emotional wellbeing in later life: patterning, correlates, inequalities and resilience' is linked. EP6 that expert paper 6 'Living well in Cornwall and the Isles of Scilly' is linked.
If 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).
The Committee's assessment of the evidence, stakeholder and expert comment on older people and mental wellbeing identified a number of gaps. One of these gaps is set out below. The others are the subject of research recommendations
1. The needs of different populations as they age. In particular how interventions can be tailored for different stages of someone's life.
(source: review 1; review 2; IDE)