456

REVIEW

Social capital: a key factor in child health inequalities

T Waterston, G Alperstein, S Stewart Brown
...............................................................................................................................

Arch Dis Child 2004;89:456–459. doi: 10.1136/adc.2003.024422

The widening gap in health outcomes between rich and types of networks are described—bonding, brid-
ging, and linking. Bonding occurs in relation-
poor is particularly evident among children and social ships that people have with people like
inequalities in health are therefore of great concern to themselves. Bridging relationships are those with
readers of this journal. Reducing inequalities in health is an people who are not like themselves—for exam-
ple, those from a different socioeconomic group
important component of UK health policy. or ethnic group. Linking refers to relationships
........................................................................... people have with those who have power or
influence.
As Coleman9 points out in his description of

A
lthough there is some research to suggest
that we would all be happier and healthier the diamond broking community in New York,
if we lived in a society with narrower close knit communities can be very excluding of
income differentials,1 2 the obvious solution to others, so one person’s perspective of how a
this problem—redistributive fiscal policy—is still community functions may be very different from
resisted by many Western governments. In these another’s. Social norms can bind a community so
circumstances health policy tends to focus on strongly that individuals who do not conform
ways of protecting the poor from the health are, or consider themselves to be, ostracised,
damaging consequences of poverty. The possibi- harassed, or marginalised. Furthermore, social
lity that local factors in communities may have a support which is offered grudgingly or patron-
protective effect on health, providing a measure isingly may do more harm than good. Intimate
of immunity to poverty,3–5 makes social capital an relationships characterised by domestic violence,
important concept for paediatricians. Health for although often close are clearly not beneficial. It
all Children 46 proposes that that social capital is is reasonable to assume that to enhance health,
not just protective against the impact of poverty, relationships need to support the development
but that it may be ‘‘as important a predictor of and maintenance of trust, respect, reciprocity,
child health … as absolute levels of wealth or cooperation, empathy, and acceptance of diver-
poverty’’. sity.

UNDERSTANDING SOCIAL CAPITAL HOW DOES ALL THIS RELATE TO CHILD

Health for all Children 46 defines social capital as HEALTH?
‘‘the social cohesion of a community, and the Let us consider two types of neighbourhoods. In
sense of belonging and the level of involvement the first, a socially cohesive neighbourhood,
in community affairs’’. This definition adopts people talk to and trust their neighbours. They
both a community focused and an individual have plenty of local friends whom they can turn
perspective on social capital. The former is that to when in difficulties and they often volunteer
preferred by Putnam,7 one of the originators of to babysit for each other. Local people are
the concept: concerned about the local environment and its
governance. They participate in voluntary and
community activities and vote in local elections.
‘‘Social capital refers to the institutions,
Relationships even between strangers are warm
relationships and norms that shape the and friendly. Children can play out quite safely
quality and quantity of a society’s social and there is a feeling of safety in the area. Crime
interactions. Increasing evidence shows that and vandalism rates are low. In the second, a
social cohesion is critical for societies to socially non-cohesive neighbourhood, parents
prosper economically and for development are afraid to let their children play out because
to be sustainable. Social capital is not just the of fear of strangers or crime, and there is a lack of
sum of the institutions which underpin a friendship and trust between neighbours as
See end of article for society—it is the glue that holds them people do not talk to one another. There are
authors’ affiliations
....................... together.’’ few community activities and racism and intol-
erance are rife. People feel powerless to influence
Correspondence to: The Organisation of Economic Cooperation their environment and do not vote in elections.
Dr T Waterston, Dept of
Community Paediatrics, and Development (OECD)8 defines social capital Intuitively most would agree that the former
Newcastle General as ‘‘networks, together with shared norms, neighbourhood would be more ‘‘healthy’’ for
Hospital, Newcastle upon values and understandings which facilitate children than the latter.
Tyne NE4 6BE, UK; a.j.r. cooperation within or among groups’’. Social Morrow10 has examined children’s experiences
[email protected]
networks are considered to be a central element of their community to determine the relevance of
Accepted of social capital in both perspectives. Networks the concept of social capital to them. In a
1 September 2003 are dynamic and are maintained and expanded qualitative study of secondary school pupils she
....................... by social interactions and transactions. Three found that friendships are critical to wellbeing.

www.archdischild.com
Social capital 457

For these children the local ‘‘community’’ was the school. The perceived health16 and mortality19 in adults. Other studies
wider community seems of less direct relevance to children, have shown individuals’ reports of social networks and social
though issues such as transport, play facilities, and attitudes support to predict health outcomes such as coronary heart
of adults towards teenagers are influential. More research is disease,20 cancer,21 and mortality.22 Although most of the
needed on children’s own views of their communities. studies on social capital and health relate to adults,12 13 16 19–22
there are some showing health effects in children. For
CRITICISM OF THE CONCEPT OF SOCIAL CAPITAL example, in a study of a number of equally poor and
Social capital is an American concept and its relation with disadvantaged neighbourhoods in Chicago between 1980 and
health has mainly been studied using US data on adult 1986, the rates of child abuse were found to be two to three
health. Recent research by the Health Education Authority times higher in socially disorganised neighbourhoods and
and its successor the Health Development Agency3 10 has with a lack of social coherence, than socially cohesive
elucidated its relevance in the UK and has also provided a neighbourhoods.4 Similar findings were reported from the
critique of some of the early writing by Putnam and others. western suburbs of Sydney, Australia.5 Rates of child abuse
The main elements of this critique are as follows: ranged from 8.1/1000 to 53/1000, and lower rates correlated
with variables such as neighbourhood cohesion, high levels
N There is no firm agreement among researchers as to the
exact components of social capital; UK data indicate that
of trust, and high value placed on local friendships. Mortality
from a number of causes, including infant mortality, have
trust, perceived citizen power, and civic engagement may been shown to be strongly associated with perceptions of lack
be more important than reciprocal help and support in of helpfulness, lack of fairness, and social mistrust in a
enhancing health. community.19 High levels of social capital predict develop-
N In UK studies, the influence of social capital and social
support on health, stress, and health behaviour appears
mental and behavioural scores in children in low income
areas,23 and low social capital increases the chances of
weaker than the influence of socioeconomic factors. dropping out of high school.7 Low social capital is associated
N Gender and age issues have been overlooked, but UK data
show them to be critical, with stress having a higher effect
with poverty and social deprivation, but most of these studies
have adjusted for a variety of indicators of socioeconomic
on women’s than men’s health. circumstances and shown an independent impact of social
capital.
N There may be political benefits for anti-welfare politicians
in playing up the role of local as opposed to governmental
It has been suggested that high social capital is associated
with good mental health,24 but the evidence so far is
solutions to ill health in poor communities.
controversial, though means are now available to test this
These criticisms are important. They indicate that the social hypothesis.25
capital story still has to fully unfold. They make the point that
initiatives to promote social capital are not a substitute for POSSIBLE MECHANISMS TO EXPLAIN THE
initiatives to redress income inequality. However they do not INFLUENCE OF SOCIAL CAPITAL ON HEALTH
suggest that the intriguing phenomenon of social capital is of Some have proposed that the health effects of social capital
no value to those interested in child health. could be explained by the promotion of the economy through
networking and collaborative ventures (economic capital),
MEASURING SOCIAL CAPITAL and the development of skills and competencies (human
Methods to measure social capital continue to evolve and capital) in the community or group with the effect of
include directly observing social interactions in community reducing poverty. Social capital may also represent a resource
settings, counting civic associations and their membership,11 for further development of the community, in that new
and gathering statistics relating to participation in local networks may be built on older networks using the social
elections and crime rates.11–13 Probably the commonest way in relationships, norms and values, trust, and information
which social capital is measured, however, is by community developed in them. At its simplest level however social
questionnaire surveys. The social capital module of the UK capital can be seen as a description of supportive, respectful
General Household Survey,14 for example, includes: respon- relationships between community members resulting in a
dents’ views of the area in which they live, civic engagement, civil society. Such relationships could have a direct effect on
reciprocity and trust, social networks, and social support. The health by enhancing emotional wellbeing and by reducing
Australian Centre for Community Organisations and the stress generated by day to day life events. We all feel
Management questionnaire,15 although using slightly differ- better when people around us are nice to us. Destructive
ent questions, covers very similar information: civic engage- relationships—those characterised by misuse of power,
ment (for example, participation in sports, unions, school suspiciousness, exclusion, and fear—appear to have a direct
groups, societies, church, elections, etc), trust and perceived detrimental effect on a range of physiological processes,
safety (including people looking out for one another, leading to susceptibility to a range of diseases.26
community fairness, and social mistrust (people taking Studies of UK data have shown that material living
advantage of one another)), reciprocity (helpfulness to conditions and socioeconomic position were stronger pre-
others), tolerance of diversity, and connections with family, dictors of adverse health than measures of social capital.3 The
friends, neighbourhood, and work. Individuals’ responses to authors suggest that the association between poor general
questions can be aggregated to produce a measure of health and low social capital can largely be explained by the
community social capital.16 The WHO Health and Behaviour greater amount of stress experienced by women living in poor
survey17 of 11–15 year olds, includes questions about the quality neighbourhoods; women’s chances of smoking
ethos of the school and found that children who are socially consistently increases as social capital decreases.
integrated report better health than those who do not feel
part of school life.18 CAN HEALTH SERVICES INFLUENCE SOCIAL
CAPITAL?
WHAT IS THE RELATION BETWEEN SOCIAL CAPITAL Social capital impact is potentially important for all of society,
AND HEALTH? but programmes to improve social capital have concentrated
Aggregated individual data have been used to show that on poor communities. The belief that social capital can be
aspects of community social capital are determinants of both improved by outside influences rests largely on experiential

www.archdischild.com
458 Waterston, Alperstein, Stewart Brown

Ways in which services may be successful in What can paediatricians do to improve social
increasing social capital in a community capital?
(adapted from Productivity Commission, 3 4 box
4.1, p. 55) N Become conversant with the literature on social capital
and health and social outcomes
N Provide services with opportunities for networking, e.g. N Use that information to inform and advise on the
through a health centre providing rooms for local self- development of programmes to enhance social capital
help groups to meet
N Engage in multi-agency work in their local community,
N Create opportunities (e.g. through local events or e.g. through Sure Start programmes
projects) for community networking and participation
N Advocate for programmes to improve social capital
N Enable local people to take control of projects and and particularly for those that benefit children
community functions
N Set up or support programmes which enhance parent-
N Traffic management and urban design changes to child relationships
create safe places for children to play, to encourage
neighbourliness, and to facilitate ‘‘mingling’’ between
N Participate or play a leading role in the monitoring and
evaluation of strategies to improve social capital and
social groups the health and social outcomes
N School size, governance, and curriculum changes;
increases in the availability and scope of extra-
curricular activities; and mandating community ser-
broader multi-agency teams. In this context they can
vices for students
participate in developing strategies to improve social capital
and assist in evaluating their efficacy. They can work with
their patients and their parents in an empowering way35 and
reporting from community development projects. Some of they can support the development of parenting programmes.
the latter evidence is, however, compelling27 and has under- Well facilitated group based programmes have the effect of
pinned the development and implementation of a number of developing mutual support and connectedness between
government initiatives, including Sure Start.28 Similar initia- families in local communities.36–40 They also have a beneficial
tives in other parts of the world such as the Families First impact on antisocial behaviour and criminality,41–43 and
Initiative29 and Schools as Community programmes in New relationships with peers.44 Antisocial behaviour and crime
South Wales (NSW), Australia,30 31 and Community Access to are important features of low social capital communities and
Child Health (CATCH)32 in the USA have all aimed to improve children who have difficulty making relationships with peers
social capital through community development. Factors in schools, tend to grow up to have problems with relation-
dictating success in the NSW initiatives have included a ships,45 and therefore be dogged by low social capital
focus on improving community trust through creating wherever they end up living. There are an increasing number
opportunities for community networking and participation of interventions that are effective in enhancing parent-child
in local events, involvement of, and eventual control by local relationships. These initiatives need to be pursued in the
people of specific initiatives. Creating opportunities for people context of antipoverty strategies.
to meet and share experiences in a safe environment can Initiatives to enhance social capital are likely to fail if they
break down barriers of mistrust. Enabling people who would are presented as an alternative to anti-poverty measures, but
not otherwise do so to participate in community affairs and as an adjunct to the latter they have the potential to be
discover that they are able to make a difference is beneficial. Indeed Richard Wilkinson has suggested1 that
empowering. Strategies have been developed from the
equity in the distribution of income may go hand in hand
ground up, supported from the top by all local agencies and
with social capital. We propose that local community
organisations, and have been based on community repre-
involvement be seen as a necessary and valuable area for
sentation in planning and implementation. A study con-
clinical work and study by paediatricians.
ducted in South Australia estimated that a 10% increase in
community participation would result in a 2.4% decrease in .....................
violent and property crime.33 The American Academy of Authors’ affiliations
Pediatrics’ CATCH programme32 is designed to increase local T Waterston, Dept of Community Paediatrics, Newcastle General
involvement by paediatricians in their communities. The Hospital, Newcastle upon Tyne, UK
basis of CATCH is that paediatricians are highly respected G Alperstein, Central Sydney Area Health Service, Sydney, Australia
and good leaders, and hence in a good position to initiate or S Stewart Brown, Division of Health in the Community, Warwick
support community health work. Medical School, University of Warwick, UK
Although there is still much research to be done in
determining the extent to which social capital affects
REFERENCES
children’s health, the best methods to improve social capital,
1 Wilkinson R. Unhealthy societies: the afflictions of inequality. Routledge,
and the mechanisms through which social capital improves 1996.
health, these studies suggest that social capital needs to be on 2 Vagreo D, Lundeburg O. Health inequalities in Britain and Sweden. Lancet
the agenda of programmes to improve children’s health in 1989;ii:35–6.
3 Cooper H, Arber S, Fee L, et al. The influence of social support and social
disadvantaged communities. capital on health. A review and analysis of British data. Health Education
Authority, 1999.
4 Garbarino J, Kostelny K. Child maltreatment as a community problem. Child
ROLE OF PAEDIATRICIANS, HEALTH VISITORS, AND Abuse Negl 1992;16:455–64.
COMMUNITY HEALTH WORKERS IN IMPROVING 5 Vinson V, Baldry E. The spatial clustering of child maltreatment: are micro-
SOCIAL CAPITAL social environments involved? Trends and issues in crime and criminal justice.
Canberra: Australian Institute of Criminology, June 1999:1–6.
Paediatricians, GPs, and health visitors all have a potential 6 Elliman D, Hall D. Health for all children, 4th edition. Oxford: Oxford
role to play in the development of social capital, but if they University Press, 2003.
are to engage in this way they will need to do so as a part of 7 Putnam RD. Bowling alone. New York: Simon and Schuster, 2000.

www.archdischild.com
Social capital 459

8 OECD. Human and social capital and sustained growth and development: 27 Wallerstein N. Powerlessness, empowerment, and health: implications for
reconciling new economies and societies: the role of human and social capital. health promotion programmes. Behaviour Change 1992;6:197–205.
Paris, France: Directorate for Education, Employment and Labour and Social 28 Eisenstadt N. Sure Start: key principles and ethos. Child Care Health Dev
Affairs, 2000. 2002;28:3–4.
9 Coleman J. Social capital in the creation of human capital. Am J Sociol 29 Alperstein G, Nossar V. Can the Families First Initiative contribute to reducing
1988;94(suppl):s95–s120. health inequalities? NSW Public Health Bulletin 2002;13:38–41.
10 Morrow V. Children’s experience of community. In: Swann C, Morgan A, eds. 30 Wraith C, Murphy E. Working together to support children and families in
Social capital for health: insights from qualitative research. London: Health disadvantaged communities. NSW Public Health Bulletin
Development Agency, 2002:11–30. 1998;9(11):131–3.
11 Raudenbush SW, Sampson RJ. Ecometrics: towards a science of ecological 31 Fletcher J, Alperstein G. Connect Redfern: connecting the community. New
settings, with application to the systemic social observation of Doctor Summer 1999/2000;72:11–15.
neighbourhoods. Sociological Methodology 1999;29:1–41. 32 Hutchins VL, ed. Paediatricians’ Involvement in Promoting Community Access
12 Bruhn JG, Wolf S. The Roseto story. Norman: University of Oklahoma Press, to Child Health (CATCH). Pediatrics 1999;103(suppl):1369–432.
1979. 33 Carcach C, Huntley C. Community participation and regional crime. Trends
13 Putnam RD, Leonardi R, Nanetti RY. Making democracy work. Princeton, NJ: and issues in crime and criminal justice. Canberra: Australian Institute of
Princeton University Press, 1995. Criminology, April 2002:1–6.
14 Coulthard M, Walker A, Moyan A. Assessing people’s perceptions of their 34 Productivity Commission. Social capital: reviewing the concept and its policy
neighbourhood and community involvement. London: HDA, 2001. implications, Research paper. Canberra: AusInfo, 2003.
15 Bullen P, Onxy J. Measuring social capital in five communities in NSW: an
35 Stewart-Brown S. Is the ethos of medical practice in community
analysis. Centre for Australian Community Organisations and Management.
paediatrics compatible with that in public health. Arch Dis Child
Working Paper Series (No. 41). Sydney: University of Technology, 1998.
2000;83:101–3.
16 Hyppa MT, Maki J. Individual-level relationships between social capital and
36 Barlow J, Stewart-Brown S. understanding parenting programmes: parents
self-rated health in a bilingual community. Prev Med 2001;32:148–55.
views. Primary Care Res Dev 2001;2:117–30.
17 World Health Organisation. The health of youth: a cross national survey.
37 Plugge E, Stewart-Brown S. Parenting programmes and mental health
WHO Regional Publications, European Series, No. 69. Geneva: WHO,
promotion: a qualitative study of the impact of the Family Caring Trust
1996.
18 Eder A. Risk factor loneliness. On the interrelation between social integration, parenting programme. In Press.
happiness and health in 11, 13 and 15 yr olds in 9 European countries. 38 Wolfe RB. Listening to children: three studies toward developing evaluating
Health Promotion 1990;5:19–33. and replicating a new approach to parent education, support and
19 Kawachi I, Kennedy BP, Lochner K, et al. Social capital, income equality and empowerment. Dissertation. USA: Northwestern University, 1997.
mortality. Am J Public Health 1997;87:1491–9. 39 Yuen TC. Filial therapy with immigrant Chinese parents in Canada.
20 Orth Gomer K, Rosengren A, Wilhelmsen L. Lack of social support and Dissertation. USA: University of North Texas, 1997.
incidence of coronary heart disease in middle-aged Swedish men. Psychosom 40 Thomas EM. Parental perceptions of the effective black parenting programme.
Med 1993;55:37–43. Dissertation. USA: Waldern University, 2000.
21 Reynolds P, Kaplan GA. Social connections and risk for cancer: prospective 41 Barlow J, Stewart-Brown S. Behaviour problems and group based parent
evidence from the Alameda County study. Behav Med 1990;16:101–10. education programmes. Dev Behav Pediatr 2000;21:356–70.
22 Hemingway H, Marmot M. Evidence based cardiology: psychosocial factors 42 Farrington DP, Welsh BC. Family-based prevention of offending: a meta-
in the aetiology and prognosis of coronary heart disease. Systematic review of analysis. Aust N Z J Criminol. Forthcoming.
prospective cohort studies. BMJ 1999;318:1460–7. 43 Dishion TJ, French DC, Patterson GR. The development and ecology of
23 Runyan DK, Hunter WM, Socolar RS, et al. Children who prosper in antisocial behaviour. In Ciccchetti D, Cohen DJ, eds. Developmental
unfavourable environments. Pediatrics 1998;101:12–18. psychopathology. Risk disorder and adaption. New York: Wiley,
24 Kawachi I, Berkman F. Social ties and mental health. J Urban Studies 1995;2:421–71.
2001;78:458–67. 44 Stoolmiller M, Eddy JM, Reid JB. Detecting and describing preventative
25 Henderson S, Whiteford H. Social capital and mental health. Lancet intervention effects in a universally school based randomised controlled trial.
2003;362:505–6. J Consult Clin Psychol 2000;2:269–306.
26 Brunner E, Marmot M. Social organization and stress. In: Marmot M, 45 Caspi A, Silva PA. Temperamental qualities at age 3 predict personality traits
Wilkinson RG, eds. Social determinants of health. Oxford: Oxford University in young adulthood: longitudinal evidence from a birth cohort. Child Dev
Press, 2000. 1995;66:486–98.

www.archdischild.com