RHP&EO is the electronic journal of the International Union for Health Promotion and Education

•  Retour en français
• Apoye en el español

Developing evidence based health promotion for older people: A systematic review and survey of health promotion interventions targeting social isolation and loneliness among older people

Mima Cattan BSc (Hons), MSc,
Dr Martin White MB ChB, MSc, FFPHM

Health Promotion Research Group
School of Health Sciences
University of Newcastle

2nd Nordic Health Promotion Research Conference
Stockholm, September 9-11, 1998

Publication Date: 29 December 1998

Cattan M, Developing Evidence Based Health Promotion for Older People: a systematic review and survey of health promotion interventions targeting social isolation and loneliness among older people. Internet Journal of Health Promotion, 1998. URL: ijhp-articles/1998/13/index.htm.

Introduction

Social isolation and loneliness are likely to become an increasingly widespread problem among older people in Britain as a result of people living longer, changing family structures, and greater mobility in the working population. By the year 2025, 20% of the population in industrial countries will be aged 65 and over1, and over 60% of women over 75 will be living alone2.

People aged 65 and over make up 23% of the population in Newcastle. Newcastle has the third highest proportion of people over retirement age and the highest number of pensioners living alone compared to other Metropolitan districts3. As a consequence, there has been increased interest and numerous activities in Newcastle upon Tyne over the past three years concerned with the health of older people.

In the green paper for England and Wales, "Our Healthier Nation"4, the government suggested that local strategies, based on best evidence, should be developed to maintain and improve the health and well-being of older people.

In 1996 an inter-disciplinary working group, facilitated by the Institute for the Health of the Elderly in Newcastle upon Tyne, proposed a programme of research to establish the evidence base for a health promotion strategy for older people. Initial research, undertaken over two years, has comprised a systematic review of the current evidence of effectiveness of interventions to combat social isolation and loneliness, and a regional survey to identify current practice and establish the extent to which it is evidence based. The group acknowledged at once that the subject area "health promotion interventions targeting older people" would be too great to undertake in one review. The initial stage of the study, therefore, comprised informal interviews and group discussions with older people in Newcastle and North Tyneside, and discussions with professionals working with older people or conducting research in the field, to identify their perceptions of health priorities for older people. Social isolation and loneliness was identified as a major concern for older people by the majority of those contacted.

Aims

The study aimed to:

1. review critically the evidence of effectiveness of health promotion interventions targeting social isolation and loneliness among older people
2. identify local patterns of health promotion practice in the field and assess the extent to which they are evidence based and as the following stage
3. establish a strategic framework for planning health promotion programmes for older people.

Methods

The study was divided into two parts: a systematic review and a regional survey of health promotion practice, conducted over two years.

Systematic review

The systematic review was undertaken in the first year, using conventional methods for systematic literature reviews5, 6. The review inclusion criteria were kept broad and inclusive rather than exclusive:

• studies relating in full, or in part, to older people;
• studies where the intervention was intended to target social isolation and/or loneliness in full, or in part;
• articles describing health promoting interventions that have or intend to achieve health gain;
• studies recording some form of outcome measures with or without process measures;
• articles published in any language, between 1970 and 1997.

The following topic definitions were applied:

• older person was defined by the criteria used in any of the studies included in the review, regardless of race, gender, physical disability or ability
• health promotion intervention was regarded as any measure taken to improve people’s quality of life and achieve health gain
• social isolation was seen as the absence of an accessible social network and lack of interaction with other individuals
• loneliness referred to emotional isolation, i.e. absence of any close emotional attachment or lack of intimacy7.

In order to define relevant search terms, the causes of social isolation and loneliness were identified through the literature, by questioning experts in the field and by asking older people themselves how they viewed social isolation and loneliness and what the main areas for concern might be. These were grouped under five broad headings:

• physical disability (e.g. loss of mobility, deterioration of eyesight, loss of hearing)
• social (e.g. bereavement, loss of home, loss of family, break-up of community)
• environmental (e.g. poor housing, lack of transport, lack of local services)
• psychological (e.g. fear of crime, fear of traffic, motivation, locus of control)
• financial (i.e. poverty).

Search terms were categorised under:

• population/target group (e.g. older, elder, ageing, geriatric, senior)
• problem area (e.g. social isolation, loneliness, depression, psycho-social)
• prevention/promotion topic (e.g. physical disability, bereavement, loss, social support, transport, activity)
• intervention/method (e.g. promotion, prevention, self help, education, community programme, policy, screening)
• type of article (e.g. review, overview, evaluation, intervention study, demonstration project, discussion paper)

These search terms were combined to find additional keywords, and to conduct electronic and manual searches

Judgement of effectiveness was made on the basis of:

1. there being a change in social isolation and/or loneliness among older people
2. the outcomes reported by authors, whether quantitative or qualitative, taking into account aims, study design, methodological quality and appropriateness for the intervention and stage of research

The intervention studies were divided into 4 categories based on the programme or method type: group activity; one to one intervention; service delivery; whole community approach.

The methodological quality of these studies was judged on the basis of the quality of the paper, (i.e. was appropriate information provided?); on the quality of the study, (i.e. was it replicable, were the conclusions "reasonable", were the findings generalisable?); and on our judgement about the effectiveness of the intervention, (i.e. did we agree with the authors?).

Survey of current practice

The survey was conducted across Northern and Yorkshire (NHS) Region in England in the second year of the research. A postal questionnaire was sent to 278 local authority departments, health service departments, non government voluntary and community organisations. The receiving organisations were asked to fill in one form for each project they organised or managed that dealt with social isolation and loneliness among older people. In addition they were encouraged to pass the questionnaire on to other appropriate organisations. The same inclusion criteria and broad categories were used in the survey as in the systematic review.

Finally, structured interviews with staff from selected projects that had responded to the postal questionnaire were undertaken to complement and add further detail to the survey.

Key results

Systematic review

Twenty one intervention studies were identified, of which almost half (10/21) were randomised control trials. The largest number of studies had been conducted in the USA, and a smaller number had been conducted in the UK, the Nordic countries and the rest of Europe (table 1).

In addition 9 surveys of various designs, and 25 purely descriptive articles were identified. A letter to health authorities and health promotion departments across the United Kingdom brought in a further 21 projects where reports, leaflets or other written were available.

Almost half (10/21) of the studies fell into the group activity category and 6 studies were categorised as one to one interventions (table 2). One study compared group and one to one interventions.

In addition we found 4 studies that investigated the effect of existing services. Two were from Denmark: one investigating the co-ordination of hospital discharge (judged as ineffective regarding the alleviation of social isolation), and the other evaluating home visiting combined with screening (judged as ineffective); one was from Germany which evaluated the effect of hearing aids being fitted (judged as inconclusive); and one from the USA which evaluated transportation and recreational activities (judged as inconclusive). The only study which evaluated a whole community approach or community development was judged as inconclusive, mainly because of the poor quality of the paper.

In summary, we found that an effective intervention to combat social isolation and loneliness among older people, tended to be a long-term group activity aimed at a specific target group, with an element of participant control using a multi-faceted approach (see list below). The characteristics of an ineffective intervention were less clear.

The second stage: survey of current practice in the field

Of the 278 questionnaires that were sent out, 200 were returned, identifying 136 projects. Ninety three replied that they were not aware of any projects targeting social isolation and loneliness among older people, and 13 had been sent on to more appropriate agencies.

Activities

Just under 40% projects listed up to 3 different activities that they used when working with socially isolated and lonely older people (table 4). It should also be noted that a number of organisations ran several projects to offer a broad range of activities to meet the differing needs of socially isolated older people.

The main activities listed were: social activities which included going on outings, playing cards or bingo, and making crafts together; befriending and social support which included both volunteer home visiting as well as small group support; having meals together, which in some cases was the "traditional" luncheon club, in other projects involved education about healthy diets, preparing and eating a meal together.

Between 30% and 40% listed 3 activities and several organisations ran more than one project.

Methods

Similarly, the majority of projects used several methods to deliver activities aimed at alleviating social isolation and loneliness. On average, projects listed 3,5 methods used within each activity. Of the 470 methods listed by the projects, 287 methods could be categorised as group activities. However, as a single item, one to support, such as befriending, companionship and carer support, was listed the greatest number of times (76/470). In addition counselling, which in most cases tended to be one to one counselling, was mentioned by 31 projects (table 5).

When these survey results are compared with published evidence of effectiveness we find an apparent conflict between practice and research evidence. One to one support is provided widely, despite the relative lack of published evidence to demonstrate that it is effective in combating social isolation and loneliness among older people.

Project evaluation

Just under half (59) projects had conducted some form of evaluation or monitoring, but only 7 of these had been evaluated by independent researchers. Internal evaluation was seen as monitoring and keeping records, by conducting user surveys and presenting annual reviews. When evaluation was cross-tabulated with funding source, we found, not unexpectedly, that those projects which mainly relied on donations and their own small scale fund raising were the least likely to have any form of monitoring or evaluation in place. Surprisingly, one fifth of those receiving local authority funding and one quarter of those receiving health authority funding stated that they did not conduct any form of evaluation or monitoring.

Discussion

The results from this study show that:

• there are a small number of published outcome studies, evaluating a narrow range of intervention methods
• the small number of outcome studies have judged group activities to be effective, while one to one interventions have been judged mainly to be ineffective or inconclusive
• there is a wide range of local activities in practice, using a large variety of methods
• the evidence base for local activity is long term experience and short term pilot work
• there is an apparent conflict between "hard evidence" and practitioner experience of one to one interventions, such as home visiting schemes.

The study has highlighted that there are many interventions targeting social isolation and loneliness among older people which are considered effective by practitioners or lay people, but for which there is no research evidence at present. Many of these interventions are concerned with quite complex behavioural, psycho-social and environmental changes, and make use of a variety of methods to combat social isolation and loneliness and encourage social support. Most evaluation studies on the other hand investigate the effects of fairly simplistic interventions, involving one or two methods.

This may be one of the reasons for the apparent conflict between practice and research evidence. While practitioners base their activities on long-term experience of their target group and the environment they operate in, evaluation studies are often short-term and based on outcome measures. Only a limited number of studies include process measures, while on the other hand practitioners often emphasise the importance of understanding the process of the intervention.

The results from the survey and literature review have indicated some major gaps in evaluation research investigating the effectiveness of interventions aimed at alleviating social isolation and loneliness among older people. We found no evaluation studies exploring the impact of : drop-centres and the provision of low cost meals; exercise and physical activity groups; rural advice and information services; or different modes of transport on social isolation. Anecdotal evidence suggests that churches and other religious establishments, activities relating to the arts, and companion animals are effective in providing social support and alleviating loneliness. Our study has been unable to corroborate this. The majority of intervention studies have focused on individual behaviour change, and the impact of legislation or policy development in, for example, health and social services, housing or transport on social isolation and loneliness has not been investigated.

Conclusion

This study has demonstrated that local projects intended to alleviate social isolation and loneliness among older people, are characterised by a wide range of activities and methods used, while outcome studies, on the other hand, have focused on a small number of intervention types. Acceptable and user friendly evaluation tools are required to enable projects to make judgements regarding the effectiveness of their activities. The lack of published evidence to demonstrate the effectiveness of one to one interventions in combating social isolation and loneliness highlights the need for future research in this field. Further qualitative research is also needed to assess older peoples' perceptions and needs regarding social support and the alleviation of social isolation and loneliness, to inform the development of effective interventions.

An effective and acceptable health promotion strategy for older people can only be developed on the basis of matched published evidence of effectiveness, practice in the field and older peoples' own perceptions of their needs.

References

1. Kinsella, K., Demographic Aspects, in Epidemiology in Old Age, S. Ebrahim and A. Kalache, Editors. 1996, BMJ Publ.: London. p. 32-40.
2. Allen, I. and E. Perkins, eds. The Future of Family Care for Older People. . 1995, HMSO: London. 234.
3. OPCS, General Household Survey, . 1991, HMSO: London.
4. Department, o.H., Our Healthier Nation. 1998, London: HMSO.
5. Chalmers, I. and D.G. Altman, eds. Systematic Reviews. . 1995, BMJ Publ.: London. 119.
6. CRD, Undertaking Systematic Reviews of Research on Effectiveness, . 1996, NHS Centre for Reviews and Dissemination, The University of York: York.
7. Weiss, R.S., Issues in the Study of Loneliness, in Loneliness: A Source Book of Current Theory, Research and Therapy, L. Peplau and D. Perlman, Editors. 1982, John Wiley and Sons: New York.

[ Home ] [ Up ] [ IUHPE ] [ Our Mission ] [ Editorial Board ] [ Reviews ] [ IJHP Articles ]

Copyright © 1999-2001 Reviews of Health Promotion and Education Online,
Last modified: June 21, 2001

Internet Explorer 5.0 or later version gives the optimal visual effect of this website.