This document provides an overview of how social networks are studied in relation to health. It discusses how health has traditionally been studied separately from a physical and mental perspective, focusing on individual characteristics. However, it notes that social networks may influence health through spreading behaviors like obesity. The document then discusses different methods for studying social networks, including whole network and ego-centric approaches. Specific topics covered in more depth include how social networks relate to mental health, managing long-term health conditions, and the role of pets and social relationships in chronic illness management. It concludes by outlining the author's PhD research on how social networks influence mental health service usage among Pakistani women.

1. Applications of Social Network
Analysis
Lecture 4: Health Networks
Dharmi Kapadia
PhD Student, Social Statistics

2. How is health studied in the social sciences?
• Broad topic: medical sciences, psychology, sociology,
• Physical and mental health – generally studied separately.
• What affects your health – individual
physiology, characteristics (age, gender), neighbourhood
characteristics
• Poor health and relationship with poverty, health
inequalities (Tudor-Hart, 1971; Marmot &
Wilkinson, 2005)
– Theories of how poor health might spread e.g. obesity
(Christakis & Fowler, 2007) [Elisa will over this next week – see
Week 5 outline]
• Separate question of how people go about managing their
health and navigating health services and what might be
at play here.
• Individualistic and structural aspects – but what about
the social network approach?

3. Methods
• Whole network approach
• Ego-centric approach – tend to be more studies
that use this way of collecting data but we have
examples of both
• WORKSHOP: What are the advantages and
disadvantages of each approach?

4. Social Ties and Mental Health
• Kawachi and Berkman (2001):
– “smaller social networks, fewer social relationships, and
lower perceived adequacy of social support have all been
linked to depressive symptoms” (p.458)
– Protective effects of networks (Cohen & Wills, 1985)
– BUT networks are not always supportive, they can also be
a source of stress, perhaps especially for women.
– What comes first – type of network or the mental health
problems? Cross sectional data does often allow us to
test for this adequately.
– But can the network be looked at in isolation when
considering the impact of it on mental health? What
about societal structures?

5. Romantic and Sexual networks
• Large study in one US high school in midwest, in
mid-sized town (n=832). Predominantly White.
Taken from Wave I of National Longitudinal Study of
Adolescent Health (Add Health). See
http://www.cpc.unc.edu/projects/addhealth for
details.

6. Data Collection
• Adolescents asked to describe their 3 most
current relationships (and 3 individuals with
whom they had a nonromantic sexual
relationship). Collected via CASI (Computer
Assisted Self-Interviewing) – audio version for
sensitive data.
• Most of the relationships were within school
(90% +), asked to identify partners from school
roster
• This is a whole network approach (as opposed to
a sample of ego-centric reports). What is the
benefit of this?

8. Implications/ Applications of Research
• One large component (spanning tree) is highly conducive
to disease transmission. Potential for transmitting disease
– note: the researchers did not collect data on the
infection in the sample.
• Series of tests to check if the structure that is observed in
Jefferson High is different than what would be expected
by chance i.e. if there was random mixing.
• Highly dense, centralised, large mean geodesic length
• So if it not due to random chance, what is going on in
these networks which gives rise to the observed
structure? Through use of simulated networks
– Homophily: students select partners with similar SES, college
plans, attachment to school, drinking behaviour, IQ, sexual
experience
– NON-homophilous on sex and age.

9. Implications/ Applications of Research
• One main finding: absence of cycles of 4
• Seems to be taboo to date your old partner’s current partner’s old partner.
Would the researchers have been able to ascertain this simply by asking the
students?
• Generalisability of findings – might not hold in adult populations where
relationships are not carried out in such closed networks and under scrutiny.
• What do we do with this research? Does it matter that the sexual health of
these adolescents was not collected?
• Sexual health education is not simply about targeting the students with the
greatest number of partners. For a tree spanning network, targeting
individuals who have fewer links within a larger component could make
substantial breaks in the network.

10. Self-management of health conditions
• Previous example did not focus on how people who are actually ill might
utilise/ mobilise/ capitalise on their networks
• Came out of a need to redress balance between individualistic approaches to
patients’ management of poor health and wider societal influences.
• Using theoretical construct of illness work (Corbin & Strauss, 1985)
• Patients with long term health conditions (diabetes, chronic heart disease
(CHD)) recruited from Greater Manchester between April 2010 - January 2011
(n=300)
• Specific interest in their social network in relation to question:
“Who do you think is most important to you in relation to managing your health
condition?”
• Mixed methods: qualitative interview nested within a survey

11. Self-management of health conditions

12. Methods & Results
• Visual method to map ego-network (Kahn & Antonucci, 1980; Pahl and
Spencer, 2004).
• Name generator: collect attribute data about each alter as well as the amount
of work (illness, practical and emotional) that the alter is perceived to do by
ego. Gives indication of the types of people that are involved in chronic illness
management.
• Goes beyond relying on measurement of individual characteristics to explain
how people manage their health – deeper explanation of who is involved and
how.
• Statistical modelling: multilevel model using network members (alters) nested
within ego networks (if interested in this method, see de Miguel Luken &
Tranmer, 2010)
• Partners/ spouses, female alters, children who lived nearby, those in frequent
contact, living close by – provided more support to patients
• Some structural aspects of networks were predictive of the amount of work
that alters did:
– More diverse networks provided more emotional support
– Networks with higher density provided more illness work
• WORKSHOP: Possible to feed these findings into governmental health
policy? What are the advantages/ disadvantages of highlighting that social
networks are helpful in managing long term health conditions?

13. Pets in chronic illness management
• Something a little bit different! But based on the same sample of patients.
• Sub analysis of networks where patients had named a pet as being helpful in
the management of their condition.
• Mixed methods but focus on the qualitative aspects of the paper
• 19% of sample identified a pet as part of their illness management network.

14. Main themes
• Pets, relationships and relatedness
– Patients spoke of pets anthropomorphically
– No need to “pay back” pets, normal terms of reciprocity
suspended
• The nature of work undertaken by pets
– Emotional support most evident in those networks where
human relationships were absent or viewed negatively
• Pets as mediators of social relationships
– Providing opportunity to forge new relationships and
maintaining existing relationships. Go-betweens facilitating
interactions with “familiar strangers”
• If data had not been collected within a network
perspective overall, would not have been able to examine
the role of pets (this was not asked as a survey question,
independently of the network map).

15. The role of social networks in accessing
mental health services
• Network Episode Model - Pescosolido (1992, 1998a, 1998b)
– “…social interaction is the basis of social life, and social networks provide the
mechanism (interaction) through which individuals learn about, come to
understand, and attempt to handle difficulties. This approach shifts the focus from
individual ‘choice’ to socially constructed patterns of decisions, including
consultation with others” (1992: p1096)
– “interaction in social networks form the principal mechanism through which
individuals recognize health problems, contact health facilities, and comply with
medical advice” (1998a: p1057)
• Choices about seeking mental health care come from the social world in which
the individual is embedded – resorting to a member of one’s social network
can be a decision or action in itself. Further, people do not turn to their
network just for advice on help-seeking, they are carers and advisors
themselves.
• Gourash (1983): 4 ways in which a network could work in relation to help
seeking for mental health services
1. Buffer stress, thereby reducing the need to use MH services
2. Provide emotional support when unwell
3. Refer to services
4. Transmit values and attitudes relating to services

16. Research Findings
• In this field there have been mixed findings: smaller networks AND
larger networks relate to being more likely to access mental health
services. Much of this research has been with samples of patients with
more severe mental illnesses (schizophrenia, bipolar disorder)
• Where the advice comes from also might have a role in shaping
whether someone accesses mental health services (Horwitz, 1978):
Friends are more likely than relatives to suggest seeking help from
professionals. But perhaps relative are providing the required support
(cf. Albizu-Garcia et al., 2001- people who had more relatives to rely
on/ speak to, less likely to use mental health services.
• Considerable stigma attached to seeking help for mental health
problems (cf. Thornicroft, 2006)– probably more than for most physical
health problems (but not all e.g. HIV).
– WORKSHOP: Across different ethnic groups, cultures, ages, sexes, stigma
could be worse – where does this stigma come from? The social
network? Another example of the negative effect of networks?

17. • Research Questions
My PhD
– How do social networks influence usage of mental health
services for Pakistani women?
– How does this association differ between Pakistani women
and women from other ethnic groups?
• English survey data: Ethnic Minority Psychiatric
Illness Rates in the Community (EMPIRIC, 2000)
• N=4281 (2340 females, 387 Pakistani ethnic
group)
• 6 ethnic groups: White, Irish, Black Caribbean,
Bangladeshi, Indian and Pakistani

18. Data: Social Network Variables
1. Number of close people
2. Relatives
i.
ii.
iii.
iv.
Regular contact (Yes/No)
Frequency of contact (5 options)
Frequency of face to face contact (5 options)
Number seen once a month or more
3. Friends
i.
4 categories as above
4. Relationship type of 2 nominated closest people (3
categories: partner/spouse, friend and relative)
(cont…)

19. Data: Social Network Variables
“Thinking about the person that you are closest to, please say how you would rate the practical and
emotional support they have provided to you in the last 12 months. How much in the last 12 months…”
Options: 1: Not at all, 2: A little, 3: Quite a lot, 4: A great deal.
a
…did this person give you information, suggestions and guidance that you found helpful?
b
…could you rely on this person? Was this person there when you needed them?
c
…did this person make you feel good about yourself?
d
…did you share interests, hobbies and fun with this person?
e
…did this person give you worries, problems and stress?
f
…did you want to confide in, talk frankly or share feelings with this person?
g
…did you confide in this person?
h
…did you trust this person with your most personal worries and problems?
i
…would you have liked to have confided more in this person?
j
…did talking to this person make things worse?
k
…did he/she talk about his/her personal worries with you?
…did you need practical help from this person with major things, for example looking after you when ill,
help with finances, children?
l
m …did this person give you practical help with major things?
n
o
…would you have liked more practical help with major things from this person?
…did this person give you practical help with small things when you needed it, for example, chores,
shopping, watering plants etc. ?

20. .05
.1
.15
.2
Mental health service use by negative aspects of support
0
1
2
3
4
5
6
7
8
Negative aspects of support
White
Black Caribbean
Indian
9
10
Irish
Bangladeshi
Pakistani
11
12

21. Next week
• Remember to hand in 2 sided essay plan. Printed
hard copy and references should be included

22. References
•
•
•
•
•
•
•
•
•
Albizu-Garcia, C. E., Alegría, M., Freeman, D., & Vera, M. (2001). Gender and health
services use for a mental health problem. Social Science & Medicine, 53(7), 865–78.
Bearman, P. S., Moody, J., & Stovel, K. (2004). Chains of Affection: The Structure of
Adolescent Romantic & Sexual Networks. American Journal of Sociology, 110 (1), 44-91.
Brooks, H. L., Rogers, A., Kapadia, D., Pilgrim, J., Reeves, D., & Vassilev, I. (2012).
Creature comforts: personal communities, pets and the work of managing a long-term
condition. Chronic illness, 9(2), 87-102. doi:10.1177/1742395312452620
Cohen, S., & Wills, T. A. (1985). Stress, social support and the buffering hypothesis.
Psychological Bulletin, 98, 310-357.
Corbin, J. & Strauss, A. (1985). Managing chronic illness at home: three lines of work.
Qualitative Sociology, 8, 224-247.
de Miguel Luken, V. & Tranmer, M. (2010). Personal Support Networks of Immigrants to
Spain: a Multilevel Analysis. Social Networks, 32(4), 253-262.
Gourash, N. (1978). Help-Seeking: A Review of the Literature. American Journal of
Community Psychology, 6(5), 413-423.
Horwitz, A. (1978). Family, kin, and friend networks in psychiatric help-seeking. Social
Science & Medicine, 12, 297–304.
Kahn, R. L. & Antonucci, T. C. (1980). Convoys over the life course:
Attachment, roles, and social support. In P.B. Baltes & O. Brim (Eds.) Life-span
development and behavior (Vol. 3, pp. 253-268). New York: Academic Press

23. References (cont.)
•
•
•
•
•
•
•
•
•
Kawachi, I. & Berkman, L. F. (2001). Social Ties and Mental Health. Journal of Urban Health,
78(3), 458-467.
Marmot, M., & Wilkinson, R. (2005). Social determinants of health. Oxford: Oxford University
Press.
Pahl, R. & Spencer, L. (2004). Personal Communities: Not Simply Families of ‘Fate’ or ‘Choice’.
Current Sociology, 52(2), 199-221.
Pescosolido, B. A., Wright, E. R., Alegría, M., & Vera, M. (1998a). Social Networks and
Patterns of Use Among the Poor with Social Networks Mental Health Problems in Puerto
Rico. Medical Care, 36(7), 1057-1072.
Pescosolido, B. A., Gardner, C. B., & Lubell, K. M. (1998b). How people get into mental health
services: Stories of choice, coercion and “muddling through” from “first-timers.” Social
Science & Medicine, 46(2), 275-286.
Pescosolido, B. A. (1992). Beyond Rational Choice: The Social Dynamics of How People Seek
Help. American Journal of Sociology, 97(4), 1096-1138.
Thornicroft, G. (2006). Shunned: Discrimination against People with Mental Illness. Oxford:
Oxford University Press.
Tudor Hart, J. (1971). The Inverse Care Law. The Lancet, 297(7696), 405-412.
Vassilev, I., Rogers, A., Blickem, C., Brooks, H., Kapadia, D., Kennedy, A., Sanders, C., et al.
(2013). Social networks, the “work” and work force of chronic illness self-management: a
survey analysis of personal communities. PloS One, 8(4), e59723.
doi:10.1371/journal.pone.0059723