Search tips
Search criteria 


Logo of brjgenpracRCGP homepageJ R Coll Gen Pract at PubMed CentralBJGP at RCGPBJGP at RCGP
Br J Gen Pract. 2007 April 1; 57(537): 277–282.
PMCID: PMC2043334

Health risk appraisal in older people 2: the implications for clinicians and commissioners of social isolation risk in older people

Steve Iliffe, FRCGP, Reader in General Practice and Kalpa Kharicha, MHSc, Senior Research Fellow
Department of Primary Care and Population Sciences, University College London
Danielle Harari, FRCP, Consultant Geriatrician and Senior Lecturer (Hon)
Department of Ageing and Health, St Thomas' Hospital, London
Cameron Swift, PhD, FRCP, FRCPI, Professor of Health Care of the Elderly
King's College London, Clinical Age Research Unit, Kings College Hospital, London
Gerhard Gillmann, MSc, Statistician
Department of Social and Preventative Medicine, University of Bern, Switzerland
Andreas E Stuck, MD, Professor of Geriatrics



Social isolation is associated with poorer health, and is seen by the World Health Organisation (WHO) as one of the major issues facing the industrialised world.


To explore the significance of social isolation in the older population for GPs and for service commissioners.

Design of study

Secondary analysis of baseline data from a randomised controlled trial of health risk appraisal.


A total of 2641 community-dwelling, non-disabled people aged 65 years and over in suburban London.


Demographic details, social network and risk for social isolation based on the 6-item Lubben Social Network Scale, measures of depressed mood, memory problems, numbers of chronic conditions, medication use, functional ability, self-reported use of medical services.


More than 15% of the older age group were at risk of social isolation, and this risk increased with advancing age. In bivariate analyses risk of social isolation was associated with older age, education up to 16 years only, depressed mood and impaired memory, perceived fair or poor health, perceived difficulty with both basic and instrumental activities of daily living, diminishing functional ability, and fear of falling. Despite poorer health status, those at risk of social isolation did not appear to make greater use of medical services, nor were they at greater risk of hospital admission. Half of those who scored as at risk of social isolation lived with others. Multivariate analysis showed significant independent associations between risk of social isolation and depressed mood and living alone, and weak associations with male sex, impaired memory and perceived poor health.


The risk of social isolation is elevated in older men, older persons who live alone, persons with mood or cognitive problems, but is not associated with greater use of services. These findings would not support population screening for individuals at risk of social isolation with a view to averting service use by timely intervention. Awareness of social isolation should trigger further assessment, and consideration of interventions to alleviate social isolation, treat depression or ameliorate cognitive impairment.

Keywords: depression, elderly, screening, risk, service use, social isolation


Social isolation is one of the major issues facing the industrialised world,1 and its prevention has been an aim of the World Health Organisation (WHO) for 30 years.2 Social interaction is beneficial for both health and wellbeing,35 while social isolation is associated with elevated blood pressure,6 poor physical health and increased mortality,6 and mental ill-health including depression,7,8 suicide,9 and dementia.10 Supportive social ties enhance physical and mental health among older adults whereas social isolation, loneliness and stressful social ties contribute to a higher risk of disability, poor recovery from illness, and early death.1114 The magnitude of health risk associated with social isolation is seen as comparable with that of cigarette smoking and other major risk factors.15

These associations suggest that isolated older people will need, and use, more health and social services than their less isolated peers,16 a possibility that will interest commissioners as well as service providers. The numbers involved appear to be large. Around 10% of those aged 65 years and over appear to be socially isolated when an objective measure of social interaction is used, while about 12% express dissatisfaction with their social contacts, leading to loneliness and emotional isolation.1 Targeting attention (and services) may be possible because of the concentration of isolation among the poor. Social network studies suggest that older people with low socioeconomic status (indicated by level of education or income) have smaller networks with a larger proportion of kin relationships, compared with those of higher socioeconomic status,17 thus making them more vulnerable to isolation. While the higher levels of education and increasing affluence of current older cohorts may protect them from social isolation and its associated morbidity, the isolation of the poorest old should remain the centre of policymakers' and practitioners' attention.17

How this fits in

Social isolation appears to be a common problem in later life, and is associated with higher levels of morbidity and service use. Policies and interventions that reduce such social isolation could reduce the illness burden in the most vulnerable stratum of the older population and have implications for service utilisation. The single assessment process now being introduced may be a vehicle for assessing social isolation among older people. However, a number of questions remain unanswered. Do those identified as socially isolated have a significantly higher burden of morbidity and make significantly greater use of services than their more socially-embedded peers? This study's results suggest that being at risk of social isolation is not necessarily associated with greater service use. Those commissioning services should not assume that interventions to relieve isolation will reduce service use. GPs who become aware of social isolation in older patients should investigate this, with particular attention to depression and memory problems.

Given the ageing of the population, policies and interventions that reduce social isolation could reduce the illness burden in the most vulnerable stratum of the older population and have implications for service utilisation. In the UK, the single assessment process of the National Service Framework for Older People19 expects GPs and community nurses to make initial assessments of the needs of older people, leading, where necessary, to more detailed assessments by social workers, nurses or geriatricians. There is evidence that group-based educational and social activity interventions can alleviate social isolation, although individual interventions (like befriending schemes) do not appear to be effective.20 Assuming that such group interventions were available, practical tools for identifying social isolation would seem useful additions to the clinical and social care repertoire, in brief forms suitable for routine use in clinical encounters and fuller forms for an in-depth, ‘diagnostic’ interview.

However, before such tools are deployed across general practice and social work, the conclusions reached in social gerontology must be verified before they are actually applied in clinical practice. Is social isolation present in a significant proportion of community-living older persons? What health characteristics are associated with risk for social isolation in older people? Do those identified as socially isolated make significantly greater use of services than their more socially-embedded peers?

This paper reports on a cross-sectional study of older people conducted in general practice, and has two hypotheses: the risk of social isolation (measured using a validated and standardised scale) occurs in approximately 10% of the community-dwelling, non-disabled population aged 65 years and over; and, being at risk of social isolation in this population is associated with a higher likelihood of having depression symptoms, self-reported memory impairment (independent of depression), more chronic diseases, more medicines, poorer self-reported functional ability, living alone status, and higher service utilisation.


Four large group practices in outer London were invited to participate in a randomised controlled trial of health risk appraisal in older people, the methodology of which is reported elsewhere.2123

A purposive sampling strategy was used to recruit practices, its criteria being known interest in primary care for older people and extensive use of electronic medical records for clinical data capture. Patient recruitment and eligibility criteria are described in the accompanying paper.24

Eligible responders were asked to complete a detailed questionnaire on their health and health risk behaviours, the Health Risk Appraisal Older (HRA–O) people instrument.21 The HRA–O is a multidimensional, self-completion questionnaire that collects information on health, functional status, health behaviours, preventive care and psychosocial factors in older people. This included a short (6-item) version25 of the Lubben Social Network Scale,26 which was developed specifically for use among older adult populations and which has been widely used in both research and clinical settings.2732 A scale of 0–30 captures the extent of social contact with family and friends, and being at risk of social isolation is defined as having a score of less than 12. The 5-item Mental Health Inventory Screening test33 was also used to identify depressed mood, a scale for self-reported memory loss,34 and the basic activities of daily living35 (BADL) and instrumental activities of daily living36 (IADL) scales. Self-reported functional change (changed and decreased functioning in the last 12 months);37 and falls; (multiple falls in the last 12 months; and activity limitation due to the fear of falling)38 were measured, as were demographic details and questions about education and sources of income; and self-reported use of health services over the previous 12 months (hospital admission and primary care or outpatient appointments). Non-responders to the HRA–O mailing were sent a postal reminder.

Data entry methods and statistical tests used are described in the accompanying paper.24 The data was analysed in two stages. Firstly, in a bivariate analysis, χ2 tests were used to compare the proportions of those at risk of social isolation with those not at risk by health behaviours, health risk factors, diagnosed chronic conditions and use of health services. This descriptive data provides a clinical impression of social isolation risk in older people. Secondly, the epidemiology of social isolation risk was explored, using multiple logistic regression to examine its relationships with health behaviour, health risk, chronic conditions and use of health services, when controlling for the confounders of increasing age (65–74 years versus 75 years and over), female sex, educational level (basic education up to the age of 15 or 16 years versus more than basic education) and income (receipt of the state pension alone versus receipt of additional income from other sources), the latter two being used as proxies for socioeconomic status. All associations that were found to be statistically significant in the bivariate analyses were included in the regression model.


Complete data from the short Lubben Social Network Scale was available for 2598 (82.8%) of the total sample of 3139 responders. Those who did not complete the scale were significantly more likely to be women with depressed mood and a history of two or more hospital admissions in the previous 12 months. After controlling for age, sex, and two proxies for socioeconomic status (state pension only and educational attainment) only depressed mood remained significantly associated with non-completion of the Lubben scale.

The factors associated with risk of social isolation are shown in Table 1. Analysis of the factors associated with risk of social isolation, confirms our first hypothesis. The risk of social isolation occurs in more than 15% of the older age group, and increases with advancing age, affecting 12.3%in the 65–74 years age band, 15.2% of those aged 75–79 years, 20.2% of the 80–84 years age band, and 31.8% of those aged 85 years and over. The mean age of those scoring in the at risk of social isolation range on the Lubben scale was 76.6 years (standard deviation [SD] = 6.85) compared with 74.2 years (SD = 6.04) for those with wider networks. This difference was statistically significant (t = −6.54, df [degree of freedom] = 513, P < 0.001) taking into account that the variances were statistically significant (F = 18.434, P < 0.001).

Table 1
Characteristics associated with risk of social isolation.

Bivariate analysis showed that the risk of social isolation is associated with older age, living alone, depressed mood, poor memory, fear of falling, worse functional ability (both BADLs and IADLs) and self-rated fair or poor health. Those who were found to be at risk of social isolation reported more chronic conditions (mean = 2.15, SD = 1.61) than those who were not (mean = 2.01, SD = 1.52), but this difference did not reach significance. Using Levene's test (F = 4.315, suggesting a significant difference of variances with P = 0.04), the t-test was non-significant P = 0.09 (t = −1.72, df = 2554). There may be a trend for those at risk of social isolation to have slightly more comorbidity than those with larger social networks. There was no difference between groups in the use of more than four prescribed medications.

Approximately half of those scoring in the at risk of social isolation range of the Lubben Social Network Scale were not living alone (Table 2); 190 of the 397 potentially socially-isolated older people were living alone, compared with 197 people living with others (data was not available for 10 individuals). Those living with others who were also potentially socially isolated on the Lubben scale were significantly more likely to be male (P < 0.001), to have only the state pension as income (P = 0.007), to be less impaired in IADLs (P = 0.007) and less fearful of falling (P < 0.001), but to have a higher risk of hospital admission (P = 0.002), as compared with those who were living alone and at risk for social isolation. There was also a weak association between the risk of social isolation and being aged 75 years and over (P = 0.026).

Table 2
Characteristics of those at risk of social isolation (n = 397) by living alone versus not living alone.

Multivariate analysis adjusting for the effect of age, sex and two proxies for socioeconomic status (state pension only and educational attainment) plus all significant associations from the bivariate analysis showed a different pattern of associations, as shown in Table 3. Sex shows a weak association with social isolation risk, with men being more likely to be at risk of social isolation, as do memory impairment and perceived poor to fair health. Living alone and low mood remain significantly associated with social-isolation risk, while fear of falling and functional loss have lost their significance. The association of risk for social isolation and difficulties with basic activities of daily living was not statistically significant (odds ratio = 1.56, 95% confidence interval 0.98 to 2.48, P = 0.058).

Table 3
Multiple regression analysis: independent associations of risk of social isolation.


Summary of main findings

The risk of social isolation, as measured using the short Lubben Social Network questionnaire, is found in more than one in six community-dwelling older people, confirming this study's first hypothesis, and half of those at risk of isolation were living with others. The bivariate analyses shows the clinical picture encountered by the practitioner working with older people. The risk of social isolation is associated with older age, education up to 16 years only, depressed mood and impaired memory, perceived fair or poor health, perceived need for help with both basic and instrumental activities of daily living, diminishing functional ability and fear of falling. Social isolation risk appears to be unrelated to sex, low income, multiple chronic conditions or multiple medication use, multiple falls, higher use of medical services, or increased risk of hospital admission.

Multivariate analysis taking into account all statistically significant associations shows a different pattern. The risk of social isolation appears to be associated with depressed mood and living alone, while male sex, memory impairment and perceived poor health may be weakly associated. For the other factors listed in the second hypothesis, no significant associations in bivariate or multivariate analyses were found.

When social isolation risk is identified in older people, a set of circumstances with a potentially complex origin is observed. The rising prevalence of social isolation with advancing age is consistent with isolation being associated with depression, perhaps as an expression of it, or possibly as either a cause or a consequence. It may also be a function of cognitive impairment, which also increases in prevalence with advancing age, but it cannot be confirmed in this data because the self-report instrument used was not designed to capture more than perceived memory loss. Social-isolation risk may also be related to living alone, perhaps reflecting the loss that occurs when a partner dies. But it is clear from this data that living alone should not be used as a proxy for social isolation risk.

Limitations of the study

Due to the cross-sectional nature of the data, it is not possible to determine causality, particularly in the relationships between social isolation, depressed mood and perceived decreasing functional ability. The sample subjected to secondary data analysis was drawn from four, purposively selected general practices in outer London and subject to eligibility criteria and screening for recruitment into a trial of health promotion for older people. The prevalence of health problems identified within this sample may be lower than that in the general population of older primary care patients, partly because disabled older people were excluded and partly because the participants were a self-selecting sub-group who returned lengthy questionnaires. The extent of social isolation may be underestimated given that those who did not complete the short Lubben scale were significantly more likely to be depressed than those who did complete it. The observation that some factors did not show a statistically significant association with risk for social isolation does not exclude the possibility of relevant associations with these factors. In addition, healthcare use was measured using a self-reported questionnaire capturing information from 1 year before enrolment in the study, while social isolation risk was evaluated up to 1 year after enrolment.

Comparison with existing literature

Given these possibilities, perhaps it is not surprising that service providers experience significant levels of concern and frustration, and a sense of powerlessness in meeting the needs of isolated older people.37 Researchers studying isolation find it so disturbing that they feel that they must be helpful to those who experience it.38 The finding that social isolation is not associated with increased service use differs from previous findings,16 and may reflect differences in the way that social isolation is measured, the selection biases inherent in this type of study, or the potential inaccuracy of self-report of service use. It may also challenge the assumption that the associations between morbidity and social isolation must lead to higher service use, when isolation actually leads to reduced service use.

Implications for future practice and research

Primary care professionals should pay special attention to isolated older people, to relieve their distress, without expecting to reduce the demands that they might make on a wide range of services. The findings do not support the view that social isolation risk is associated with greater service use, but they do support the belief that isolation and depressed mood are connected. Awareness of social isolation risk could prompt further enquiry about social support and potentially associated factors such as depression, memory problems, and disability, with a view to therapeutic intervention where possible. The short version of the Lubben scale (validated in the ProAge study) could be useful for identifying this risk group in clinical practice.39

If these findings were typical of UK populations, those commissioning health services would be unlikely to reduce downstream use of hospital care by targeting interventions at potentially isolated older people identified using a risk scale. Those commissioning social care, on the other hand, may benefit from knowing the prevalence of social isolation, because it is likely to predict demand for social support during periods of illness.


We thank the practices and patients involved in the study.


Funding body

European Commission project QLK6-CT-1999-02205 and the Federal Education Science Ministry, Berne, Switzerland (BBW 990311.1)

Ethics committee

Approval was obtained from Brent Medical Ethics Committee (BEC 745) and King's College Hospital Research Ethics Committee (01–010)

Competing interests

The authors have stated that there are none


1. Findlay RA. Interventions to reduce social isolation amongst older people: where is the evidence? Ageing Soc. 2003;23(5):647–658.
2. World Health Organisation. Psychogeriatric care in the community Public Health in Europe 10. Copenhagen: WHO Regional Office for Europe; 1979.
3. Wenger C, Davies R, Shatahmasebi S, Scott A. Social isolation and loneliness in old age: review and model refinement. Ageing Soc. 1996;16(3):333–358.
4. Victor C, Scambler S, Bond J, Bowling A. Being alone in later life: loneliness, social isolation and living alone. Rev Clin Gerontol. 2000;10:407–417.
5. Berkman L, Glass T, Brissette I, Seeman T. From social integration to health: Durkheim in the new millennium. Soc Sci Med. 2000;51:843–857. [PubMed]
6. Bower B. Social links may counter health risks: research on how social isolation affects mortality in older adults. Science News. 1997;152(9):135.
7. Prince M, Harwood R, Blizard R, et al. Social support deficits, loneliness and life events as risk factors for depression in old age. The Gospel Oak project VI. Psychol Med. 1997;27:323–332. [PubMed]
8. Warner JP. Quality of life and social issues in older depressed patients. Int Clin Psychopharmacol. 1998;13(S5):S19–S24. [PubMed]
9. Centres for Disease Control and Prevention. Suicide amongst older persons: United States 1980–1992. JAMA. 1996;275(7):509. [PubMed]
10. Fratiglioni L. Influence of social network on occurrence of dementia: a community-based longitudinal study. Lancet. 2000;355(9212):1315–20. [PubMed]
11. Seeman TE, Singer BH, Ryff CD, et al. Social relationships, gender, and allostatic load across two age cohorts. PsychosomMed. 2002;64(3):395–406.
12. Stuck AE, Egger M, Hammer A, et al. Home visits to prevent nursing home admission and functional decline in elderly people: systematic review and meta-regression analysis. JAMA. 2002;287:1022–28. [PubMed]
13. Berkman LF, Glass T. Social Integration. Social networks, social supports and health. In: Berkman, LF, Kawachi I, editors. Social epidemiology. New York: Oxford Press; 2000. pp. 137–173.
14. Stuck AE, Walthert JM, Nikolaus, Bula T, et al. Risk factors for the functional status decline in community-living elderly people: a systematic literature review. Soc Sci Med. 1999;48:445–69. [PubMed]
15. House JS. Social isolation kills, but how and why? Psychosom Med. 2001;63:273–274. [PubMed]
16. Ellaway A, Wood S, McIntyre S. Someone to talk to? The role of loneliness as a factor in the frequency of GP consultations. Br J Gen Pract. 1999;49:363–367. [PMC free article] [PubMed]
17. Broese van Groenou M, van Tilburg T. Network size and support in old age: differentials by socio-economic status in childhood and adulthood. Ageing Soc. 2003;23(5):625–645.
18. Joung I, Kunst A, van Imhoff E, Mackenbach J. Education, ageing and health: to to what extent can the rise in educational level relieve the future health (care) burden associated with population ageing in the Netherlands? J Clin Epidemiol. 2000;53:955–963. [PubMed]
19. Department of Health. National Service Framework for Older People. London: HMSO; 2001.
20. Cattan M, White M, Bond J, Learmouth A. Preventing social isolation and loneliness among older people: a systematic review of health promotion interventions. Ageing Soc. 2005;25:41–67.
21. Stuck AE, Elkuch P, Dapp U, et al. for the Pro-Age pilot study group Feasibility and yield of a self-administered questionnaire for health risk appraisal in older people in three European countries. Age Ageing. 2002;31:463–467. [PubMed]
22. Iliffe S, Kharicha K, Harari D, et al. Health risk appraisal for older people in general practice using an expert system: a pilot study. Health Soc Care Community. 2005;13(1):21–29. [PubMed]
23. Stuck AE, Kharicha K, Dapp U, et al. The PRO-AGE study: an international randomised controlled study of health risk appraisal for older persons based in general practice. BMC Med Res Methodol. 2007;7:2. [PMC free article] [PubMed]
24. Kharicha K, Iliffe S, Harari D, et al. Health risk appraisal in older people 1: Are older people living alone an at-risk group? Br J Gen Pract. 2007;57:271–276. [PMC free article] [PubMed]
25. Lubben JE, Gironda MW. Social Support Networks. In: Osterweil D, Brummel-Smith K, Beck JC, editors. Comprehensive geriatric assessment. McGraw Hill Publisher: New York; 2000. pp. 121–137.
26. Lubben JE. Assessing social networks among elderly populations. Fam Comm Health. 1988;11:42–52.
27. Martire LM, Schulz R, Mittelmark MB, Newsom JT. Stability and change in older adults' social contact and social support: the cardiovascular health study. J Gerontol B Psychol Sci Soc Sci. 1999;54(5):S302–11. [PubMed]
28. Steiner A, Raube K, Stuck AE, et al. Measuring psychosocial aspects of well-being in older community residents: performance of four short scales. Gerontologist. 1996;36(1):54–62. [PubMed]
29. Rubinstein RL, Lubben JE, Mintzer JE. Social isolation and social support: an applied perspective. Appl Gerontol. 1994;13(1):58–72.
30. Luggen AS, Rini AG. Assessment of social networks and isolation in community-based elderly men and women. Geriatric. 1995;16:179–181.
31. Stewart AL, Hays RD, Ware JE. The MOS short-form general health survey reliability and validity in a patient population. Med Care. 1988;26:724–732. [PubMed]
32. Riege WH. Self-report and tests of memory aging. Clin Gerontol. 1982;I:23–36.
33. Katz S, Ford AB, Moskowitz RW, et al. Studies of illness in the aged. The index of ADL: a standardized measure of biological and psychosocial function. JAMA. 1963;185:914–919. [PubMed]
34. Lawton MP, Brody EM. Assessment of older people: self-maintaining instrumental activities of daily living. Gerontologist. 1969;9:179–186. [PubMed]
35. Fried LP, Bandeen-Roche K, Chaves PHM, Johnson BA. Pre-clinical mobility predicts incident mobility disability in older women. J Gerontol Med Sci. 2000;55A:M43–M52.
36. Kelsey JL, Browner WS, Seeley DG, et al. Risk factors for fractures of the distal forearm and proximal humerus. Am J Epidemiol. 1992;135:477–489. [PubMed]
37. Russell C, Schofield T. Social isolation in old age: a qualitative exploration of service providers' perceptions. Ageing Soc. 1999;19:69–91.
38. Andersson L. Loneliness research and interventions: a review of the literature. Aging Mental Health. 1998;2(4):264–274.
39. Lubben J, Blozik E, Gillmann G, et al. Performance of an abbreviated version of the Lubben Social Network Scale among three European community-dwelling older adult populations. Gerontologist. 2006;46(4):503–513. [PubMed]

Articles from The British Journal of General Practice are provided here courtesy of Royal College of General Practitioners