Banner Project

Banner Project Upstream took part in a collaborative exhibition at the Royal Albert Museum Exeter. The 12' banner involved stencilling, fabric painting, embroidery and stuffing bubble wrap fish with tissue paper.

Tiverton Canal Barge trip
Tiverton Canal Barge trip. This summer project inspired our participants to draw, note or remember what they experienced on the trip to work up later into diaries and sketch pads


to keep up to date with all the latest developments

Summary of PMS Report



Upstream Healthy Living Centre: Research on the processes and outcomes of an intervention to address social isolation in the elderly

Dr Colin Greaves

R&D Manager, Mid-Devon Research Group; Research Fellow, Peninsula Medical School (Primary Care)

Dr Lou Farbus

Research Fellow, Peninsula Medical School (Primary Care)


May 2005 (updated with Oct 2005 data)


The UK's population is ageing. In the next 25 years the number of people over the age of 80 will treble, and those over 90 will double. The potential future impact on healthcare resources has led to new government and local policies concerning service provision and strategies for promoting ‘quality of ageing' for older people.1;2

As the proportion of older people increases, more are living alone. A recent UK survey indicated that 12% of people over 65 feel socially isolated.3 Social isolation and loneliness are consistently associated with reduced well being, health and quality of life in older people.3;4 Depression in particular is associated with social isolation and affects 1 in 7 over 65s.4 A careful review of the literature (App. A of main report) suggests that interventions which promote active rather than passive social contact, and encourage stimulating creative activity, with support and guidance from a mentor, are likely to impact positively on the health and quality of life of older people.

Between June 2003 and January 2005 research was conducted around the Upstream Healthy Living Centre, aiming to develop and evaluate just such an intervention.

Research Aims

  • To examine in detail the processes of intervention, and to provide feedback  with a view to developing and optimising the intervention (action research)
  • To describe and develop the theoretical basis underpinning the interventioN
  • To qualitatively identify the range and nature of impacts on participants
  • To quantitatively assess the scale of likely impact in terms of particular measures of participants' physical and mental health

The Upstream Intervention

Upstream is a community-based intervention operating on an outreach (without walls) basis with mentors working closely with socially isolated people over the age of 50 to re-kindle their passion and interest in life by engaging in self-determined programmes of social, exercise and/or cultural activities. In other words it aims to promote social participation, well being and health through activity and creativity, and to work with older people before they become substantial ‘downstream' users of health and social services. The intervention is individually tailored to suit each participant's own interests and passions. An activity-based intervention is provided, with visits from mentors initially on a weekly basis, gradually diminishing as participants become more confident and able.

Theoretical basis

The creative and individually tailored aspect of the intervention aims to maximise the stimulation of higher cognitive functioning, which has been related to successful ageing in terms of both quality of life, and health outcomes. Encouraging social interaction and creative activity may also indirectly increase physical activity, and it is known that positive effects on health are related to even small changes in activity.5-7 Tailoring activities to individual abilities and interests is intended to maximise the level of engagement and personal meaningfulness, thereby making activities more likely to be sustained. Sustainability is also enhanced by the explicit acknowledgement of the need to build self-efficacy (confidence about ability to conduct activities and to maintain them as mentoring is withdrawn). It was expected that the theoretical basis might be adapted as feedback accrued.


To date, a wide range of activities have been provided ranging from painting to tai chi. Each participant chooses the activity or activities they wish to do, and therefore receives a unique programme of intervention, and there are several alternative models of provision:

  • Group activities provided by Upstream (69%)
  • One-to-one activities provided by Upstream (5%)
  • Signposting to (often subsidised) existing community-based activities (26%)

Activity data are available for the 317 participants who had received at least one mentor visit by the end of February 2005. Of these, 255(80%) had been engaged in some kind of activity. The main reasons for non-engagement were ill-health, or deciding that Upstream, once explained, was not for them.

Inclusion /Exclusion Criteria

The current definition of Upstream participants is that they are "people from their late 50s onwards, who's lives may have changed or are about to change in some way (perhaps through retirement, moving home, age or illness), or people with time on their hands, or who might, for whatever reason, find it difficult to keep in touch with the local community and would enjoy the opportunity to share their interests, skills and enthusiasm with others" In addition, participation is restricted to people over the age of 50, who live in the catchment area of Mid-Devon PCT, and have no mental or physical health problems that are sufficient to make them a danger to others or to require special nursing care when attending group activities.

Sources of Referral

Referrals have come from a mixture of sources including statutory bodies (54%), word of mouth (31%) and self-referral (13%). The highest single sources of referral were GP surgeries (19%), existing local groups (15%), the residential care /assisted accommodation sector (14%) and self-referral from participants after seeing posters or leaflets (13%). Social services (8%) and reablement teams (10%) are also significant sources of referral.

Research Design and Measures used

Qualitative (interview based) research, incorporating an action research methodology was conducted alongside a longitudinal observational study (administering pre and post intervention questionnaires).

Interviews were conducted with a wide range of stakeholders in the Upstream service, including participants, carers of participants Upstream staff, and strategic partners (e.g. health and social services, mental health services, local council, arts organisation, other voluntary sector organisations).

Questionnaire measures assessed quality of life (SF-12 questionnaire) in terms of both physical and mental well-being, depression (Geriatric Depression Scale (GDS)) and perceived social support (MOS Social Support Survey). Questions on health behaviours (physical activity and alcohol use) were also given.

Two cycles of feedback were provided to Upstream staff, who derived action points or ‘aspirations' for further developing the service.

Qualitative Results (feedback from interviews)

1. Service Delivery

Mentoring: Participant feedback about the mentors was overwhelmingly positive. No negative comments were recorded, and the vast majority took the opportunity to commend them. Referrers also spoke highly of the mentors' work and acknowledged the value of the skills learned about working with this vulnerable group.

Strategic Partnerships: Upstream seems to have been successful at communicating its ethos and identifying how it can work alongside partners to the mutual benefit of all. Recommendation for future development included providing brief feedback to referrers on participants (with permission) to encourage partners to make further referrals. Upstream were recently selected (from the pool of 257 HLCs in England) by the Big Lottery Fund as one of 5 ‘Pathfinder' HLCs. As a result the organisation is well positioned to become an example of good practice for other HLCs around the country.

Activities: Participant enjoyment and satisfaction: The vast majority of interviewees spoke highly of the quality and appropriateness of activities and their enjoyment of them, Participants reported increased confidence in engaging in activities, and in interacting socially with others.

2. Impact of Upstream

Psychological, Health and Social Benefits: Examples of psychological, social and physical health benefits were reported by carers and health professionals, as well as by participants themselves. Noticeable benefits were reported for 15 of 18 interviewees who had received the intervention. The range of benefits reported is summarised below:

  • Psychological and social benefits
  • Reductions in depression and loneliness
  • Increased alertness /cognitive awareness
  • Increased well being and optimism
  • Less dwelling on concerns /worries, better sleep
  • Increased social interaction and community involvement
  • Increased sense of self-worth and willingness to engage in life
  • Collateral benefits for carers and family (both in seeing loved ones enjoying life more, and in the respite opportunities provided)
  • Health behaviour and health benefits
  • Increased physical activity, more energy
  • Healthy changes in diet and less heavy drinking
  • Less health visits, reduced medication use
  • Potentially reduced risk of falls (due to alertness effect)
  • Facilitated rehabilitation of co-ordination /mobility post-stroke

Four of the 18 interviewees provided striking testimonies of stronger ‘transformational' change affecting multiple aspects of their' lives. These were consistent with the notion of recovery from depression, involving increased sense of meaning in life, increased social and physical activity, and more attention to self-care.

SG3-001: I still have osteoporosis, I still have collapsed vertebrae, I've still got fibrosing alveolitis [...] I can't walk more than 40 yards [...] But I'm better now than I was in 1999 before I was ill [...] I don't get agitated about things. I think I'm much more able to sort loads and loads of different things [...] Now I've woken up from the fog it's like I'm really enjoying my life as though this is what I'm meant to be doing now.

SG4-003: She said, ‘I never go out except to the shops every day because all my friends are dead.  My husband's dead, I don't have any children' [...] She doesn't read and she doesn't have hobbies [...], and so her life was absolutely a barren desert. So then I got Upstream involved and they did some home visits and  gradually introduced her to this little art group and [...] she's made friends and she's a new woman.  She's not depressed and withdrawn as she was.  She's got confidence and I think that's terrific [...] She's cheerful, she has got a brighter step and [...] it's just opened up new horizons for her and made her life better. It gives them confidence.  It gives them value.

The data suggests that Upstream has had considerable success in socially re-integrating people who were previously isolated. Indeed, the most widely reported benefit was enjoyment of the social interactions around Upstream activities. This was facilitated by Upstream's ‘receptive context' - a forum in which people could be offered and mutually share social support in terms of material support (e.g. sharing transport, material, skills), emotional support (e.g. encouragement) and information support (e.g. sharing information about community activities, or the benefits system).

No significant negative outcomes were reported, although this information was actively sought.

Causal Attribution: In terms of causal attribution, the benefits reported were in the majority of cases attributed to Upstream. Some participants felt that other factors (e.g. recovery from illness) had also been present, but that Upstream had still acted as a ‘catalyst' for recovery (especially recovery from depression).

Transferability: Regarding the transferability of the above benefits to the wider population of Upstream participants, discussion with mentors confirmed that from their perspective, for all engaged  participants, around 80-90% derived noticeable positive benefits, and about 20-30% underwent dramatic transformations in mood and behaviour. Examination of individual change scores from the questionnaire data confirmed these estimates to be reasonable (see below).

3. Factors Mediating the Impact of Upstream

Issues of access and availability of appropriate activities are crucial pre-requisites for engagement in activities. Transport issues in particular remain a significant problem in some parts of Mid Devon and were a frequently reported barrier preventing people from being able to see each other or attend activities as much as they would like.

Enjoyment of and satisfaction with activities seems to be largely mediated by the extent to which activities (and choice of venue) are tailored to individual needs and abilities. Good tailoring includes paying attention to individual preferences, health status, and social skills as well as the level of confidence about engaging with activities (i.e. pitching it at an appropriate skill level). Individual tailoring therefore seems to be one of the main keys to Upstream's success in the initial engagement of this elderly, socially isolated population. The issue of health /ability on entry into Upstream also mediated the amount of effort /input needed, and there is a consideration that the ‘top of the pyramid' participants with more severe morbidities /disabilities may need a more intensive intervention model. This could include an element of ongoing support (i.e. complete withdrawal of mentoring may not be appropriate in some cases).

The main factors which mediate continuation of activities and the derivation of benefits seem to be building confidence /self-efficacy and the creation of positive social dynamics. The key factor seems to be the ability of mentors to empower participants (building confidence and self-determination) to a) try out and succeed at engaging in activities, and b) to be able to engage socially in their groups. Successful engagement then seems to lead to improvements in self-representations (social identities, self-concept and self-esteem), and the relevance of existing theories around this theme was noted (e.g. self-representational change is the basis of cognitive behavioural therapy). These self-representational changes may then, in theory, translate into changes in psychological well-being and changes in health and self-care behaviours. These observations have been built into a revised theoretical model, which is described in the Discussion section of the main research report.

4. Moving Towards Sustainable Activities

Upstream's efforts to provide activities that can be sustained by participants after its withdrawal are beginning to bear fruit, and there are increasing numbers of success stories both from people engaging more independently in groups (some of which are partly supported by Upstream and some of which are wholly independent), and people making their own plans to engage in existing community-based activities.

Factors Influencing Sustainability: Many of the factors identified as mediating potential benefits also related to the sustainability of activities. These include transport, the availability of suitable (and accessible) activities, building on initial engagement of interest /positive social dynamics, and ensuring confidence /self-efficacy has been sufficiently developed. Other important issues identified were matching individuals to the right groups (which can be existing community based activities /groups if the participant is ready to ‘go it alone') and ensuring there is a suitable mix of people /abilities in groups which Upstream is facilitating. This includes people with all the skills needed to run and maintain the group (e.g. administration, completing physical tasks, social skills, and management or strategic leadership abilities). The issue of overall group size, in terms of maintaining a ‘critical mass' of participants was also considered relevant by Upstream staff.

Managing the withdrawal of mentor support (or continuing to provide it until it is no longer needed) may require more input over a longer time period than was originally envisaged in some cases. Upstream is also finding ways to provide informational /pragmatic support to better facilitate ongoing group work. These include developing an information toolkit, and providing activity ideas and ‘activity boxes'.

Despite the examples of success achieved so far, the amount of work required to facilitate sustainable groups presents a potential capacity problem. However, newer ways of working (less home visits, more mentor input delivered at group sessions, more assistance with group development) seem to be addressing this at the current time. There is a sense of transition towards allowing mentor input over a longer time period, but balancing this against less home visiting, and more group-work with larger groups developing in some areas.

Outcomes Survey Results

Response Rates: At 30th September 2005, Upstream had 373 referrals. Of these, 229 were eligible for the survey (those yet to receive a mentor visit, those declining further intervention at 1st visit, and those who referred to non-Upstream activities were excluded). Of the eligible participants, 171(75.0%) provided data.  Of these, 136 were eligible for 6-month follow-up questionnaires at 20th October, and 72(53%) provided data. For 12 month follow-up, 93 participants were eligible, and 51(55%) provided data.

Sample Characteristics (N=172): Participants were 76% female, average age 77 (52 to 96), 25% with financial pressures. Participants had poor physical and psychological health at baseline, commensurate with high levels of loneliness and social isolation

  • 74% with at least one longstanding health impairment which limits activities
  • SF12 health quality of life scores were significantly lower than norms for UK over-75s and the general UK population (both mental and physical health)
  • 53% had clinical levels of depression (45% mild; 8% severe) based on Geriatric Depression Scale scores[1]
  • 33% reported no physical activity lasting at least 15 minutes in the last month

Further demographic data were collected for the first 45 participants only. These showed high levels of usage in the last 3 months of NHS (73% used the service, with an average 2.5 GP visits) and Social Services (53% used some form of service). Of this early sample, 64% were widowed, 73% lived alone, and 93% were retired.

6 Month Follow Up (N =71): Measures were taken at entry into Upstream and 5.5 (mean) months later. The data showed:-

  • A statistically significant increase in SF12 mental health component (MCS) (Mean Diff: 3.0 points, p < 0.005).
  • This changeeffect size 0.30 SD) is clinically meaningful (2-3 points is considered a clinically meaningful change for SF12 scores). A change of 3 MCS points represents an increase of around 12 centiles in the UK population.
  • 60% experienced clinically meaningful benefit (2 points) in SF12 MCS scores.
  • Taking a 6 point SF12 MCS score to represent a high degree of change (equivalent to a 25 centile shift in population ranking), 30% experienced this level of positive change.
  • No significant change was found in SF12 physical health scores. The lack of a  decline may be a positive outcome in this high-morbidity group, however.
  • A statistically significant reduction in depressive mood (Diff=0.60, p<0.02)
  • The above change (effect size 0.31SD) is also clinically meaningful. The number with clinical levels of depression fell from 32(45%) to 25(35%).
  • A statistically significant increase in the proportion engaging in physical activity (from 63% to 77%, p<0.05, N=43).

These results were consistent with the qualitative data on outcomes, with psychological benefit (and depressed mood) being the most widely reported benefit. Around two-thirds of participants experienced clinically meaningful levels of psychological benefit, with around a quarter experiencing high levels of benefit.

12 Month Follow Up (N =51): Measures were taken at entry into Upstream and 12.0 (mean) months later. The data showed:-

  • Improved depression scores were maintained (Mean Diff=0.57, p<0.05)
  • The size of the difference in SF12 mental component scores decreased (Mean improvement=0.71 points, n.s)
  • SF12 physical component scores now showed a trend towards improvement (Mean Diff=1.57, p=0.06)
  • The overall health utility index (which combines SF12 mental and physical components) improved significantly (Mean Diff=0.027, p<0.05)
  • Social support scores (MOS Social Support Scale) also improved significantly (Mean Diff=0.20, p<0.05, effect size =0.34)

The picture is broadly consistent with a sustained increase in health quality of life, and particularly depressed mood, with additional benefits in terms of perceived social support and physical health emerging over time.


Alternative possible explanations of the data were considered, these being 1) Upstream causes benefits, or 2) regression to the mean for a high morbidity baseline population (people improving on their own). The latter hypothesis is not consistent with the qualitative data. Although there is no control group here, the research literature suggests that the improvement found is greater than that which would normally be expected for untreated depression in elderly populations. Furthermore, two major longitudinal studies of ageing show that the expected trend in both physical and mental health for this age group in the UK is downward, with a more dramatic decline in women. Hence, although the case for a causal effect of the Upstream intervention cannot be categorically proven by this observational methodology, on balance, the idea that Upstream is responsible for at least some of the change identified seems the most likely explanation.

Action Research Summary: The research process was received very well by Upstream, and staff participated fully in deriving action points from the ongoing  feedback, and in implementing recommendations which arose. The qualitative feedback about the processes of delivering Upstream provided a number of insights into the mechanisms involved, potential pitfalls, barriers to success, and ideas about how to optimize and further develop the intervention. At the time of writing, most of these aspirations have been actioned in some way.


Development of Theoretical Basis: The action research also proved useful in deriving a detailed theoretical model that is consistent with the research findings, and which outlines mechanisms by which Upstream is believed to work. This theory may be helpful in informing future research in this area, and to those seeking to replicate the intervention elsewhere. The theory builds on and integrates Social Identity Theory (SIT) and Self Concept Theory (SCT). The central premise of SIT is that the more an individual identifies with a group, the more self-representations become intertwined with group definitions. SCT tells us that social self-representations are a crucial component of human self-concept, which when evaluated (positively or negatively) is the basis of self-esteem.8 The promotion of new, positive social /group identities (social participation) is therefore likely to affect self-esteem in important ways. Self-esteem is a central pillar of mental health and psychological well-being, and self-representations also have a moderating role in emotional health. Having new socially meaningful roles therefore increases the opportunities for self-esteem to be enhanced, resulting in increased emotional health and well being (Fig.1).





Socio-contextual moderators

- Appropriateness and quality of activities

- Impressions of activities, venue, mentors

- Transport & access barriers

- Social dynamics within the group (perception of mutual social support)

Individual background moderators:

- Self-efficacy

- Levels of health & social isolation prior to intervention

Social identity development

Change in mental wellbeing:

- Self-esteem

- Quality of life

- Social participation

Affective valence of new social identity i.e. positive or negative


Key:                   Social Identity Theory                      Additions to the conceptual model


Fig.1: The Upstream Intervention Model

Increases in physical activity associated with attending activity sessions and groups (some of which entail physical activity as a major component) may also be relevant as a mechanism for health gain. Even mild increases in physical activity can result in strong health benefits in elderly population, with well-established effects on dementia (1 hour per week reduces risk by 20%), depression, and cardiovascular risk.5-7


Overall, Upstream seems to have been highly successful in identifying and engaging with their target socially isolated elderly population. The majority of participants at entry to Upstream were living alone, and substantial psychological and physiological morbidity was evident, including a 53% prevalence of depression and high rates of chronic illness and disability. Overall quality of life for both mental and physical well-being was well below the expected normative, age-matched population values.

The evaluation has increased our understanding of the processes by which health change may be delivered by engaging people in social and creative activities. This has helped to develop the underlying theory of Upstream's work, but more importantly has helped to develop the practice of delivering this type of intervention to socially isolated elderly people.

Taking the qualitative and quantitative findings together, the results suggest that engaging socially isolated elderly people in social and creative activities, using an individually tailored, mentoring approach, which focuses on building efficacy /confidence and self-determination can enhance their overall health-related quality of life, with a particularly impact on depression, as well as increasing perceived social support. Additional physical health benefits may be possible, particularly in the longer term if activities are sustained. The extent of the possible health gains is clinically meaningful in terms of the average changes in outcome scores reported. Furthermore, a substantial number of individuals were reported to experience quite radical transformations, including enhanced psychological well-being and lifestyle changes, as well as physical health benefits.

Reference List

   1.   Department of Health (2001). National Service Framework for Older People.  2001. London, HMSO.

   2.   Department of Health. Modernising Social Services.  1998. London, HMSO.

   3.   Owen T. The high cost of isolation. Working with Older People. 2001;5:21-3.

   4.   Philip I. Ageing and Health. Community Practitioner 2005;78:9-11.

   5.   Bassey EJ. The benefits of exercise for the health of older people. Reviews in Clinical Gerontology 2000;10:17-31.

   6.   Tuomilehto J, Lindstrom J, Eriksson JG, Valle TT, Hamalainen H, Ilanne-Parikka P et al. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N.Engl.J.Med. 2001;344:1343-50.

   7.   Weuve J, Kang JH, Manson JE, Breteler MMB, Ware JH, Grodstein F. Physical Activity, Including Walking and Cognitive Function in Older Women. JAMA 2004;292:1454-61.

   8.   Harter S. The Construction of the Self. New York: Guildford, 1999.

[1] The GDS is a validated tool for screening clinical depression with 92% sensitivity, 87% specificity


© 2018 Upstream. Upstream HLC is a registered charity no. 1087185 Company limited by guarantee no. 4114401
T: 01363 778366 E:

Site browsing information

Your current browser and operating system is:

  • Operating system: Unknown
  • Browser: Unknown
  • Version: Unknown

End of Summary of PMS Report - Evaluation page - Return to Page Content Navigation