Beyond the “I” in the Obesity Epidemic: A Review of Social Relational and Network Interventions on Obesity

Abstract

Background:
Recent research has shown the importance of networks in the spread of obesity. Yet, the translation of research on social networks and obesity into health promotion practice has been slow.

Objectives:
To review the types of obesity interventions targeting social relational factors.

Methods:
Six databases were searched in January 2013. A Boolean search was employed with the following sets of terms: (1) social dimensions: social capital, cohesion, collective efficacy, support, social networks, or trust; (2) intervention type: intervention, experiment, program, trial, or policy; and (3) obesity in the title or abstract. Titles and abstracts were reviewed. Articles were included if they described an obesity intervention with the social relational component central. Articles were assessed on the social relational factor(s) addressed, social ecological level(s) targeted, the intervention's theoretical approach, and the conceptual placement of the social relational component in the intervention.

Results:
Database searches and final article screening yielded 30 articles. Findings suggested that (1) social support was most often targeted; (2) few interventions were beyond the individual level; (3) most interventions were framed on behaviour change theories; and (4) the social relational component tended to be conceptually ancillary to the intervention.

Conclusions:
Theoretically and practically, social networks remain marginal to current interventions addressing obesity.

Full text

Hindawi Publishing Corporation
Journal of Obesity
Volume , Article ID ,  pages
http://dx.doi.org/.//
Review Article
Beyond the ‘‘I’’ in the Obesity Epidemic: A Review of
Social Relational and Network Interventions on Obesity
Janette S. Leroux,1Spencer Moore,1,2 and Laurette Dubé3
1School of Kinesiology and Health Studies, Queen’s University, 28 Division Street Kingston, ON, Canada K7L 3N6
2Department of Community Health and Epidemiology, Queen’s University, Kingston, ON, Canada K7L 2N8
3DesautelsFacultyofManagement,McGillUniversity,1001rueSherbrookeOuest,Montreal,QC,CanadaH3A1G5
Correspondence should be addressed to Spencer Moore; mooresp@queensu.ca
Received  April ; Accepted  July 
Academic Editor: Terry Huang
Copyright ©  Janette S. Leroux et al. is is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
cited.
Background.Recentresearchhasshowntheimportanceofnetworksinthespreadofobesity.Yet,thetranslationofresearchonsocial
networks and obesity into health promotion practice has been slow. Objectives.Toreviewthetypesofobesityinterventionstargeting
social relational factors. Methods. Six databases were searched in January . A Boolean search was employed with the following
sets of terms: () social dimensions: social capital, cohesion, collective ecacy, support, social networks, or trust; () intervention
type: intervention, experiment, program, trial, or policy; and () obesity in the title or abstract. Titles and abstracts were reviewed.
Articles were included if they described an obesity intervention with the social relational component central. Articles were assessed
on the social relational factor(s) addressed, social ecological level(s) targeted, the intervention’s theoretical approach, and the
conceptual placement of the social relational component in the intervention. Results. Database searches and nal article screening
yielded  articles. Findings suggested that () social support was most oen targeted; () few interventions were beyond the
individual level; () most interventions were framed on behaviour change theories; and () the social relational component tended
to be conceptually ancillary to the intervention. Conclusions. eoretically and practically, social networks remain marginal to
current interventions addressing obesity.
1. Introduction
Obesityisrecognizedasoneofthegravestthreatstopublic
health of our time []. Current intervention strategies meant
tocurbthespreadofobesityhavebeenineective[,].
Addressing the magnitude of the obesity epidemic requires
the development of multilevel and cross-sectoral interven-
tions []. Genetic, biological, and psychological factors inter-
act with obesogenic environmental conditions to promote
inactivity, poor nutrition, and, resultantly, widespread weight
gain [–]. Social epidemiological research has highlighted
the importance of social determinants, such as gender, age,
socioeconomic status and ethnicity, on health. Interventions
on individual behaviors and choices fail, however, to account
for the social relational conditions that inuence personal
choices and behaviors and limit the eectiveness and impact
of obesity interventions [,]. ere is growing consensus on
the need to shi the paradigm for addressing the prevalence
ofobesitytosocialdomainsbeyondtheindividual[,,].
e degree to which social relational constructs have been
integrated into obesity interventions remains unclear.
Anumberofsocialrelationalconstructshavegained
prominence in recent social epidemiological research on
obesity. ese constructs included social cohesion, collec-
tiveecacy,trust,socialcapital,socialsupport,andsocial
networks. Social cohesion describes the trust, respect, and
participation within a community and has been conceptu-
alized as a social-structural, cultural condition that impacts
health through community integration []. Collective e-
cacy may refer to the norms and networks of relationships
that enable collective action and a culture of informal social
control and social cohesion, whereby people are united
and willing to act for the good of the community [].
Collective ecacy has been proposed as a constraint on

Journal of Obesity
unhealthy behaviors []andameansthroughwhicha
communityisabletooperateasaunittoprocuresocial
trust, security, and resources within society at large [].
Depending on the perspective, social capital can be consid-
ered as a communitarian- or network-driven phenomenon. A
communitarian denition would conceptualize social capital
as comprising elements of a sense of belonging, partici-
pation and civic engagement, reciprocity and cooperation,
and community trust. A network-based denition of social
capital would consider the availability and accessibility of
resources within an individual’s social network. Independent
of these dierences in denitions and measurements, both
approaches have yielded associations with health outcomes
[], including obesity []. Social networks can be dened as
a web of social relationships and are characterized by overall
structure, as well as the individual ties of which it is com-
prised. More recent sophisticated methods of social network
analysis have revealed a social patterning of a number of
health outcomes. Christakis and Fowler [] demonstrated
thespreadofobesityinsocialnetworksusinglongitudinal
data and validated old and new interest in harnessing the
potential of social networks in relation to population health.
Social support, which is categorized by instrumental and
nancial, informational, appraisal, and emotional forms of
support, is conceptualized as a psychosocial mechanism
which connects social relationships and individual health
through psychological, behavioral, and physiological path-
ways [].
Findings on the impact of social relationships on obesity
encourage the shi to interventions beyond the “I”, or
individual level and toward interpersonal dynamics by which
behaviours are shared, norms formed, and resources (e.g.,
information, support) exchanged. e objective of this review
is to examine the current state of social relational interven-
tions on obesity and characterize the degree to which these
interventions have addressed key social relational constructs
in intervention planning and implementation.
2. Methods
2.1. Search Strategy, Search Terms, and Search Criteria. To
identify the types of interventions targeting obesity from a
social inuence perspective, we conducted a systematic lit-
erature review on social relational interventions targeting
obesity. PubMed, Web of Knowledge, CINAHL, EMBASE,
TRoPHI, and OVID MEDLINE were all searched in January
. e searches were restricted to full-text, English-
language articles. A Boolean search strategy was employed
with the search designed to identify articles with the following
sets of terms in their title or abstract: () social dimensions:
social capital, social cohesion, collective ecacy, social sup-
port, social networks, or trust; () experimental conditions:
intervention, experiment, program, trial, or policy; and ()
obesity.
2.2. Inclusion Criteria, Review Methods, and Data Synthesis.
Duplicatearticleswereremovedfromthedatabaseofarticles.
From this pool, articles were included in the next stage if they
described an obesity-focused intervention among the general
population, and the social relational construct was central
enough to the intervention that it was included in the title
or abstract. Studies that were removed from the original pool
of articles included those that addressed eating disorders,
chronic diseases, or postpartum women. ese criteria were
applied independently by two researchers. Disagreements on
the inclusion of specic articles were discussed and resolved
by consensus.
e nal selection of studies was reviewed to assess
and characterize each study by () social relational con-
struct addressed, () social ecological level targeted, () the-
oretical approach used to guide the intervention, and ()
the placement of social relational construct on the inter-
vention’s conceptual pathway. e social relational con-
structs were social capital, social cohesion, collective e-
cacy,socialsupport,socialnetworks,andtrust.esocial
ecological model was used as a framework by which to
determine the social ecological level(s) targeted by the inter-
vention [] and included individual, interpersonal, orga-
nizational, community, and political levels. To distinguish
between interpersonal-level interventions and individual-
level interventions that included an interpersonal compo-
nent, the ensuing criteria were followed: a study was consid-
ered an interpersonal intervention if it involved one or more
members of a study participant’s existing social network. e
theoretical rationale for each intervention was garnered from
each study when provided.
A conceptual typology was developed based on the role
of the social relational construct in the intervention. e
typology identied three potential roles for social relational
constructs to play in an obesity intervention: intervention tar-
get, delivery channel, and ancillary resource. e intervention
target was dened as a modiable social relational construct
lying directly on the intervention pathway. e delivery
channel was dened as the functional or structural means of
delivering the intervention, or a vehicle meant to facilitate
the intervention. e ancillary resource was dened as a
reinforcing but noncentral dimension of the study. Ancillary
resources might contribute to the uptake or success of the
intervention but was not a critical component of the delivery
channel or intervention target. For example, an ancillary
resourcewouldbeonewheretheinterventionwasseekingto
change health behavior and delivers the program in a group
setting which facilitates group cohesion and social support
among study participants.
3. Results
Database searches using title criteria yielded  titles.
Application of the inclusion criteria narrowed results to 
studies. Interrater reliability of the  full-text articles to the
 nal studies was calculated as Cohen’s kappa coecient
(Kappa = ., SE: .) []. Table  provides a comprehen-
sive overview of each study, organized by social relational
construct (type, modality, and measurement), intervention
type, theoretical explanation or reference, social ecological
level the intervention was targeting, and type of social rela-
tional construct conceptual pathway placement.

Journal of Obesity 
T : Comprehensive overview of intervention studies found pertaining to social relational constructs and obesity.
Article Social relational constructs eory Social ecological level
targeted
Conceptual pathway
placement of SRC
Type Modality Measure
Lubans et al.,  []Socialsupport Textmessages — Social cognitive
theory
Individual/
interpersonal Resource
Angelopoulos
et al.,  []Social support
Student workbook and teacher
manual including motivational
method and strategies (change
social inuence through modeling,
mobilizing social support)
—eory of planned
behavior
Individual/
interpersonal/
organizational
Resource
Peterson and
Ward-Smith ,     [ ]Social support Community-based obesity support
group Social support questionnaire
Tra n s t heore t i c al
model, social
comparison theory,
and social support
Individual Channel
Gellertetal.,[] Social support Support group — Stage-of-change
model Individual Resource
Kushneretal.,[]Socialsupport Cohesiveness between owners and
pets
Social support and readiness
questionnaire Social learning theory Individual/
interpersonal Tar g e t
Rimmer et al.,  []Socialsupport
Professional advice, brochure
information, PA device, telephone
consultation, and monthly exercise
support group
— — Individual Channel
Stolleyetal.,[]Socialsupport
Activities to promote group
cohesion (ice breaker, potluck
dinners, outside activities, and
inclusion of friends and family)
Social support for eating and
exercise questionnaire
Social cognitive
theory, health belief
model
Individual Resource
Hemmingsson
et al.,  []Social support PA behavior change booklet, care at
obesity unit, and group sessions —Tra n s t heore t i c al
model Individual Channel
Gold et al.,  []Socialsupport
Behaviour therapy with social
support lesson; group support Perceived social support scale — Individual Channel
Anderson
et al.,  []Social support Proposed mediator for changes in
HRQOL Family social support (Sallis) Social cognitive
theory Individual Resource
Gallagher et al.,
 []Social support Group sessions (behavioral
strategies to elicit social support) Behavioural processes subscale Social cognitive
theory Individual Resource
Pettman et al.,  []Socialsupport
Peer group setting incorporating
self-management programs,
establishing peer support networks;
information, shared experiences,
and outside interaction
—eory of planned
behaviour Individual Channel
Kiernan et al.,  []Socialsupport Friend and family support for
healthyeatingandPA
Support subscales and sabotage
subscales; general supportive and
strained interactions with family
and friends subscales; qualitative
question on social support
Social support
measurement Individual Resource
Kalodner and DeLucia,
 []Social support Classmate interaction to facilitate
social cohesion and support — Behaviour change Individual Resource

Journal of Obesity
T : Continu ed.
Article Social relational constructs eory Social ecological level
targeted
Conceptual pathway
placement of SRC
Type Modality Measure
Casazza and Ciccazzo,
 []Social support Computer-based education;
in-person lecture and pamphlets Social support survey — Individual Resource
Hajeketal.,[]Socialsupport
Interactive group sessions (group
support components, cohesive and
productive environment)
Client feedback questionnaire
includes “group support” as
potential component participants
found most useful
— Individual Channel
Yancey et al.,  [] Socialsupport Inclusionofclosefriendorrelative — Social ecological
models
Individual/
interpersonal Resource
Cousins et al.,  []Socialsupport
Emphasized family-oriented
approach to health behaviors;
manual, inclusion of spouses, and
group support
——
Individual/
interpersonal Resource
Williamson
et al.,  []Social support Classroom/internet program Children’s dietary social support
scale —Individual/
organizational Resource
Leblanc et al.,  []Socialsupport Structural social support provided by
group used in control arm of trial —
Health at every size
(health-centered
approach)
Individual Channel
Bjelland et al.,  []
(i) Social
support;
(ii) social capital
Classmate interaction to facilitate
social cohesion and support
(i) Perceived social support from
parents, friends, and teachers;
(ii) related to people in my
area/neighborhood: quality of
relationship with peers at school (in
+ out of classroom)
Social-ecological
model
Individual/
organizational Resource
Lee et al.,  []Group cohesion,
social support
Intervention group (shared goal,
working on team activities,
assigning team roles, encouraged to
contact each other outside of
intervention sessions)
Physical activity group environment
questionnaire; social support for
eating habits survey
Behavior change;
self-ecacy,
stage-of-change, and
social support
Individual Channel
De Niet et al.,  [] Family cohesion
Treatment team (psychologist,
dietician, pediatrician, and
physiotherapist) led information
sessions for parents
Family adaptability and cohesion
evaluation scales (FACES) III Social learning theory Individual/
interpersonal Resource
Leahey et al.,  [] Social cohesion Participation in group contingent
on weight loss Perceived cohesion scale — Individual Resource
Kimetal.,[]Socialnetworks
Participant recruitment through lay
health advisors social networks —Community-based
participatory research Individual Channel
Leahey et al.,  []Socialnetworks
Media, newsletters, motivational
and educational activities, online
log, and encouraged team support
Reported social inuence for weight
loss
Social inuence,
social learning theory,
and social modeling
Individual Resource
Gorin et al.,  []Socialnetworks
Instructional sessions to enhance
social support for weight loss eorts — — Individual Channel

Journal of Obesity 
T : Continu ed.
Article Social relational constructs eory Social ecological level
targeted
Conceptual pathway
placement of SRC
Type Modality Measure
Ashida et al.,  []Socialnetworks
Indentied encouragers for dietary
behavior
Social inuence (enumerated social
network members who “played
signicant role in life during past
year” and “have encouraged you to
eat more FVs/do PA”)
Social inuence Individual/
interpersonal Resource
Shaw-Perry et al.,
 []Social networks
Organized health programming
sessions transmitted to children
through social structures (home,
health class, school cafeteria, and
aer school)
——
Individual/
organizational Channel
Gessell et al.,  []Socialnetworks
Social network evolution over
duration of intervention
Social network survey developed to
assess changes in social
relationships
Social network, social
support
Individual/
interpersonal Tar g e t

Journal of Obesity
Table  provides a descriptive overview of these  stud-
ies.
e vast majority of studies (𝑁=22)featuredsocial
support as the social relational construct, whether alone
(𝑁=20) or in combination with social capital []or
social cohesion []. e ways that social support was incor-
porated into interventions ranged widely between interven-
tions related to the provisioning of professional advice and
telephone consultation, to motivational workbooks, to the
inclusion of a family or friend in the program itself, and
to instructional sessions or interactive group sessions. e
measurement of social support varied considerably across
interventions from no measures, formal survey instruments,
to informal qualitative assessments. Ten of the twenty studies
did not measure social support despite the fact that the
construct was included in the abstract or description of the
intervention. Two studies featured social cohesion, one of
which specically examined family cohesion []andthe
othersocialcohesioninaweight-lossgroup[]. Social
cohesion was measured in both studies with the use of
(dierent) questionnaire scales. Six studies featured social
networksasthemainsocialrelationalconstruct,although
the way in which social networks were incorporated varied
considerably. Social networks were observed to be used as
a study recruitment strategy [], a structure for transmit-
ting health programs and social inuence [–], and a
changeable entity which might evolve in the intervention
[]. ree of the six studies measured the social network
component, which included a study-specic survey [], a
qualitative report of social inuence [], and a quantitative
report of social inuence []. ere were no studies that
addressed social trust, collective ecacy, or social capital
exclusively. Interventions focused primarily on the individual
levelbutoccasionallyspannedintotheinterpersonalrealm
duetotheuseofsmallersupplementarycomponents.For
example, a school-based intervention program tailored for
adolescent girls sent home four parent newsletters/progress
reports which reported their children’s time spent in physical
activity, sedentary behaviours, and self-reported fruit and
vegetable consumption. In addition, the newsletters included
information meant to increase awareness and encourage par-
ents to support their children’s physical activity and dietary
behaviors []. Such an intervention would be considered
primarily an intervention at the individual level with minimal
crossover into the interpersonal level. One study intervened
at the organizational level [], and no studies were found
to intervene at community or political levels. ere were
a number of studies which were seemingly conducted at
asocialecologicallevelbeyondtheindividualbutupon
closer examination were in fact targeting individuals within
broader settings rather than targeting change at a higher
social ecological level itself. For example, the “Choose to
Move for +(Positive) Living” intervention drew participants
from a community-based “Stay the Course” obesity support
group and sought to determine the inuence of psychosocial
aspects of the (physical activity and heart healthy living)
program on increasing physical tness, perceived social
support and quality of life, and stage of health behaviour
change for physical activity. ese program’s objectives were
T : Descriptive overview of intervention studies found per-
taining to social relational constructs and obesity.
Social relational construct 𝑁=30
Social support 
Social cohesion 
Social network 
Social trust —
Collective ecacy —
Social capital —
Multiple social relational constructs
Social support-social cohesion 
Social support-social capital 
Social ecological level targeted 𝑁=30
Single level target
Individual 
Interpersonal environment —
Organizational environment —
Community —
Political environment —
Multiple level target
Individual-interpersonal 
Individual-organizational 
Interpersonal-organizational —
Individual-interpersonal-organizational 
eory or model 𝑁=30
Health belief model  (+)
Stages of change (transtheoretical model)  (+)
Social learning theory (social cognitive theory)  (+)
eory of planned behaviour 
Social support theory  (+)
Social comparison/inuence/modeling theory  (+)
Ecological approaches (CBPR, SEM) 
Multiple theories, models, or approaches 
No reference to theoretical rationale 
Conceptual role of social relational construct 𝑁=30
Intervention channel 
Ancillary resource 
Intervention target 
+ in theor y section indicates t he addition of partia l references of multiple the-
ories. SEM: social ecological model; CBPR: community-based participatory
research.
individual-oriented, and although the intervention appeared
to operate as a holistic, community-based program, it did
not intervene at the community level []. Similarly, the
Ke’Ano Ola: Moloka’i’s community-based healthy lifestyle
program was conducted in thecommunityandwasbasedon
principles of community-based participatory research. Yet,
the intervention targeted individual nutrition education [].
A number of interventions did not include a theoretical
rationale or explanation related to the social aspect of the

Journal of Obesity 
intervention or program (𝑁=9). Most interventions refer-
enced the stages of change model (or transtheoretical model)
and social cognitive theory (social learning theory) (𝑁=4
and 𝑁=7, resp.). e typology was developed to identify the
wayinwhichthesocialrelationalconstructswereincluded
in the interventions. e role that researchers considered the
social constructs to play in obesity prevention was reected
in the placement of the construct along the intervention
pathway. Two of the thirty studies reviewed featured social
relational constructs (social networks []andsocialsupport
[]) as intervention targets. Of the studies which featured
social support as the social relational construct, twelve of
these operationalized social support as an ancillary resource
with the remaining seven studies operationalizing social
support as a channel. e studies that featured social cohesion
asthesocialrelationalconstructoperationalizeditasanancil-
lary resource. Social networks were mainly operationalized in
these studies as a channel to deliver the intervention itself [,
,]. Two studies included social networks as an ancillary
resource [,], and one exceptional study conceptualized
social networks as an intervention target [].
4. Discussion
epurposeofthisstudywastoreviewthetypesofobe-
sity interventions targeting social relational constructs and
characterize the degree to which these interventions have
addressed key social relational constructs in intervention
design and implementation. Social support was the predom-
inant social relational construct targeted [,,,–
], treated as a mediator or channel [], or used as the
control treatment in a trial []. Social support was not
always clearly dened, with a diverse range of social support
(peer, family, group, and professional) being delivered either
inperson through peer groups or professional therapy or
remotely through such tools as handbooks, newsletters, or
electronic support messages. e measurement of social sup-
port also varied across interventions (e.g., perceived versus
actualsupport).Socialsupportwasoenassumedtobe
inherent in any intervention that involved a support group.
For example, monthly meetings of overweight/obese individ-
uals who might share their challenges with healthy eating or
physical activity were considered to be inherently supportive
and equally available to all participants. As a result, many
interventions failed to measure whether participants actually
received social support. Only four of the  studies which
focused on social support included a theoretical rationale or
evidence for addressing social support in the intervention
[,,,]. e dierent functions of social support
(informational, emotional, t angible, and belonging) were out-
linedinonlytwoofthestudiesfeaturingsocialsupport[,
]. e nondierentiation of social support highlights the
atheoretical treatment of social support as an agent of change
in reducing obesity. Five studies mentioned social support
in combination with social cohesion as shared attributes of
peer support groups but did not distinguish between these
two dierent social relational constructs by denition or mea-
surement [,,,,]. Overall, social networks were
largely limited to methodological applications, as a means
of study recruitment or disseminating information related to
behavioural change. Little attention was given to the network
measures or the eects that social networks might have on
health. One exception was Gessell et al.’s [] study in which
they examined the evolution of social networks over the
duration of an obesity prevention intervention. In terms of
other social relational constructs, there were no studies which
discussed social trust, collective ecacy, or social capital. e
lack of interventions targeting these higher ecological social
network or relational variables suggests that there is still
much work to do in translating social capital work into actual
interventions, specically obesity. In addition, there may be
alackoffamiliaritywith,orcondenceintheuseof,“more
complex” social interventions in public health practice. Social
support was inconsistently dened, measured, and applied
in the current collection of the literature; this might imply
that health researchers are dierentially receptive to including
social support in an intervention, as compared to other social
relational constructs. Social support may seem intuitive and
most easily intervened on amidst the diering denitions
and approaches to measuring social capital; the sophisticated
methods of social network analysis; and the vagueness of
social cohesion and collective ecacy (and challenges of
measurement).
e social ecological model provides a framework from
which to discern and compare the complexity of the dif-
ferent interventions examined in the current review. While
intraindividual factors, including beliefs, knowledge, and
skills, are important aspects in the behaviour change process,
interventions which are limited to targeting change at an indi-
vidual level fail to address the importance of broader social,
physical, economic and political contexts. e breakdown
of study types by social ecological level was shown to be
pyramid-shaped with the vast majority of studies focused on
the individual [,–,,,–,,]andafew
interventions that included components which spanned into
the interpersonal [,,,,,,] or organizational
realms [,,]. Within organizational realms, interven-
tions tended to target making nutritional or physical activity
resources available. For example, in a school setting, play-
grounds and school yards were made accessible for children
to play aer end of curricular program, and school canteens
were obliged to have fresh fruit and freshly made juices [].
Another study program modied the cafeteria food service
program (the contents of vending machines), and physical
education programs [] and another intervention included
implementing short PA breaks during lessons [].
e prominence of individual-level obesity interventions
was matched by the greater reliance on theoretical perspec-
tives built on individual psychosocial and behavioral models
andconstructs.Interventionstendedtobedrivenbytheories
largely centered on behavioral psychology, including social
cognitive theory, the transtheoretical model, and the theory
of planned behavior. e lack of social theory in intervention
planning limits the development of higher ecological level
interventions on obesity. For example, an obesity intervention
which is based solely on social cognitive theory would likely
lack the breadth to investigate or address the range of

Journal of Obesity
environmental factors that might impact person’s odds of
being obese.
Furthermore, the frequent reference to self-ecacy in
the selected interventions requires additional attention. Self-
ecacy—which comprises an individual’s motivation, locus
of control, and behavioural choices, intentions, and actions
with respect to their goals, tasks, and challenges—was oen
included as a predictor, mediator, or moderator of overweight
andobesityriskfactorsandstatus.etheoreticalemphasis
on personal responsibility and control belies the use of
concepts related to social, political, and organizational change
[]. is is not to detract from the value of individually ori-
ented theories []. However, mounting evidence suggests that
innovative strategies for addressing and preventing obesity at
a population level should entail theories and approaches that
operate from an ecological perspective [].
ere were a range of outcomes found in the set of
interventions. Obesity-related outcomes included () anthro-
pometric indicators, such as body mass index or body fat
percentage, () physiological measures of cholesterol, blood
pressure, and blood sugar, and () behavioural risk factors
such as physical activity, dietary patterns and knowledge,
screen time, sedentary time, and smoking. A number of
studies included psychological and psychosocial outcomes,
such as depressive symptoms, self-ecacy, and motivation,
while some studies also included social indicators, such as
social support.
e conceptualization of a social relational construct as
an intervention target would suggest that the researchers view
the particular construct as integral to the obesity pathway.
Yet, within our sample of interventions, social relational
constructs were predominantly incorporated as a channel
through which to deliver the intervention, or a nonessential
intervention resource. Accordingly, these social relational
constructs may be seen as being useful but not amenable
characteristic in and of themselves. Although the fram-
ing of the rationale of some studies suggests a concep-
tual emphasis being put on the respective social relational
constructs,itisapparentthatthisemphasisdoesnotcarry
through in practice. When examining the studies collectively,
these ndings suggest either (i) a possible stagnation of
intervention research that builds on dierent social relational
constructs as they contribute to obesity or (ii) the idea that
the conceptualization, implementation, and evaluation of
interventions which incorporate social relational constructs
and theories beyond the individual are dauntingly complex
and inaccessible among researchers.
Despite the comprehensiveness of our search strategy, our
search criteria may have favored the discovery of smaller
scale interventions that would be communicated in more
traditional academic outlets. Accordingly, one limitation of
our study may have been the potential exclusion of broad-
er policy planning interventions that might target more
upstream social political determinants of obesity. Upstream
social interventions might consist of one or more social
relational constructs or address multiple levels of the social
ecological framework. Nevertheless, the lack of interventions
on social relational constructs suggests a limited landscape of
social relational interventions being implemented or incor-
porated in broader policy interventions.
5. Conclusion
To address the problem of obesity, there is a need for
public health programs to intervene at social ecological
levels beyond the individual. Intervening on interpersonal,
organizational, or community levels may be more eective
and sustainable in the long term in reducing individual risk
of obesity. e apparent lack of social network as opposed
to individual support interventions addressing obesity high-
lights a key gap existing between research and practice. While
social epidemiological research has examined the inuence
of social networks, social capital, and social environments
on obesity, this research has yet to be translated into the
design of social relational or network interventions that
address obesity. While social support may be an important
component of such interventions, there is a need to consider
more carefully the importance of social relationships and the
social environment on the onset and establishment of obesity.
e ndings of the current study suggest a vast potential for
methods and evidence from social health research to further
advances in addressing the obesity epidemic.
Acknowledgment
is paper was supported in part by the National Institutes of
Health Research (Grant no. ).
References
[] D. L. Katz, “Competing dietary claims for weight loss: nding
the forest through truculent trees,” Annual Review of Public
Health, vol. , pp. –, .
[] S.K.Kumanyika,E.Obarzanek,N.Stettleretal.,“Population-
based prevention of obesity: the need for comprehensive pro-
motion of healthful eating, physical activity, and energy bal-
ance: a scientic statement from American Heart Association
Council on Epidemiology and Prevention, Interdisciplinary
Committee for prevention (formerly the expert panel on popu-
lation and prevention science),” Circulation,vol.,no.,pp.
–, .
[] A. Jain, “Treating obesity in individuals and populations,” Brit-
ish Medical Journal,vol.,no.,pp.–,.
[] G. Egger and B. Swinburn, “An “ecological” approach to the
obesity pandemic,” British Medical Journal,vol.,no.,
pp. –, .
[] T. A. Glass and M. J. McAtee, “Behavioral science at the cross-
roads in public health: extending horizons, envisioning the
future,” Social Science and Medicine,vol.,no.,pp.–,
.
[]K.M.Booth,M.M.Pinkston,andW.S.C.Poston,“Obesity
and the built environment,” JournaloftheAmericanDietetic
Association,vol.,no.,pp.–,.
[] J.O.HillandJ.C.Peters,“Environmentalcontributionstothe
obesity epidemic,” Science,vol.,no.,pp.–,
.
[] M. B. Schwartz and K. D. Brownell, “e need for courageous
action to prevent obesity,” in Obesity Prevention and Public
Health,D.CrawfordandR.W.Jeery,Eds.,OxfordUniversity
Press, New York, NY, USA, .

Journal of Obesity 
[] K. Glanz and D. B. Bishop, “e role of behavioral science
theory in development and implementation of public health
interventions,” Annual Review of Public Health,vol.,pp.–
, .
[] L. Cohen, D. P. Perales, and C. Steadman, “e O word: why the
focus on obesity is harmful to community health,” California
JournalofHealthPromotion,vol.,no.,pp.–,.
[] B. Swinburn and G. Egger, “Preventive strategies against weight
gain and obesity,” Obesity Reviews,vol.,no.,pp.–,
.
[] R. G. Wilkinson, Unhealthy Societies: e Aictions of Inequal-
ity,Routledge,London,UK,.
[] D.A.Cohen,B.K.Finch,A.Bower,andN.Sastry,“Collective
ecacy and obesity: the potential inuence of social factors on
health,” Social Science and Medicine,vol.,no.,pp.–,
.
[] R. J. Sampson, S. W. Raudenbush, and F. Earls, “Neighborhoods
and violent crime: a multilevel study of collective ecacy,”
Science,vol.,no.,pp.–,.
[] I.Kawachi,B.P.Kennedy,andR.Glass,“Socialcapitalandself-
rated health: a contextual analysis,” American Journal of Public
Health,vol.,no.,pp.–,.
[] S. Moore, M. Daniel, C. Paquet, L. Dub´
e, and L. Gauvin, “Asso-
ciation of individual network social capital with abdominal
adiposity, overweight and obesity,” Journal of Public Health,vol.
,no.,pp.–,.
[] N.A.ChristakisandJ.H.Fowler,“espreadofobesityina
large social network over  years,” e New England Journal of
Medicine,vol.,no.,pp.–,.
[] B. N. Uchino, “Social support and health: a review of physiolog-
ical processes potentially underlying links to disease outcomes,”
Journal of Behavioral Medicine,vol.,no.,pp.–,.
[]J.F.Sallis,N.Owen,andE.B.Fisher,“Ecologicalmodelsof
health behavior,” in Health Behavior and Health Education:
eory, Research, and Practice,K.Glanz,B.K.Rimer,andK.
Viswanath, Eds., Jossey-Bass, San Francisco, Calif, USA, th
edition, .
[] M. J. Wood, “Understanding and computing Cohen’s Kappa: a
tutorial,” , WebPsychEmpiricist, http://wpe.info/ .
[] M.Bjelland,I.H.Bergh,M.Grydelandetal.,“Changesinado-
lescents’ intake of sugar-sweetened beverages and sedentary
behaviour: results at  month mid-way assessment of the
HEIA study—a comprehensive, mu lti-component school-bas ed
randomized trial,” International Journal of Behavioral Nutrition
and Physical Activity,vol.,article,.
[] R. E. Lee, D. P. O’Conner, R. Smith-Ray et al., “Mediating eects
of group cohesion on physical activity and diet in women of
color: health is power,” AmericanJournalofHealthPromotion,
vol. , no. , pp. e–e, .
[] J. De Niet, R. Timman, C. Rokx, M. Jongejan, J. Passchier, and
E. van den Akker, “Somatic complaints and social competence
predict success in childhood overweight treatment,” Interna-
tionalJournalofPediatricObesity,vol.,no.,pp.e–e,
.
[] T.M.Leahey,J.G.omas,J.G.LaRose,andR.R.Wing,“A
randomized trial testing a contingency-based weight loss inter-
vention involving social reinforcement,” Obesity,vol.,no.,
pp. –, .
[] S. Kim, D. Koniak-Grin, J. H. Flaskerud, and P. A. Guarnero,
“e impact of lay health advisors on cardiovascular health
promotion: using a community-based participatory approach,”
e Journal of Cardiovascular Nursing,vol.,no.,pp.–,
.
[]T.M.Leahey,R.Kumar,B.M.Weinberg,andR.R.Wing,
“Teammates and social inuence aect weight loss outcomes in
a team-based weight loss competition,” Obesity,vol.,no.,
pp. –, .
[] A.A.Gorin,R.R.Wing,J.L.Favaetal.,“Weightlosstreatment
inuences untreated spouses and the home environment: evi-
dence of a ripple eect,” International Journal of Obesity,vol.,
pp. –, .
[] S. Ashida, A. V. Wilkinson, and L. M. Koehly, “Social inuence
and motivation to change health behaviors among Mexican-
origin adults: implications for diet and physical activity,” Ameri-
can Journal of Health Promotion,vol.,no.,pp.–,.
[] M. Shaw-Perry, C. Horner, R. P. Trevi˜
no,E.T.Sosa,I.Hernan-
dez, and A. Bhardwaj, “NEEMA: a school-based diabetes risk
prevention program designed for African-American children,”
Journal of the National Medical Association,vol.,no.,pp.
–, .
[] S. B. Gessell, K. D. Bess, and S. L. Barkin, “Understanding the
social networks that form within the context of an obesity pre-
vention intervention,” Journal of Obesity,vol.,ArticleID
, p. , .
[] D. R. Lubans, P. J. Morgan, A. D. Okely et al., “Preventing
obesity among adolescent girls,” Archives of Pediatric Adolescent
Medicine,vol.,no.,pp.–,.
[] P. D. Angelopoulos, H. J. Milionis, E. Grammatikaki, G.
Moschonis, and Y. Manios, “Changes in BMI and blood pres-
sure aer a school based intervention: the CHILDREN study,”
European Journal of Public Health,vol.,no.,pp.–,
.
[] J. A. Peterson and P. Ward-Smith, “Choose to move for positive
living: physical activity program for obese women,” Holistic
Nursing Practice, vol. , no. , pp. –, .
[] K. S. Gellert, R. E. Aubert, and J. S. Mikami, “Ke`
eAno Ola:
Molaka`
ei`
es community-based healthy lifestyle modication
program,” American Journal of Public Health,vol.,no.,pp.
–, .
[] R. F. Kushner, D. J. Blatner, D. E. Jewell, and K. Rudlo, “e
PPET study: people and pets exercising together,” Obesity,vol.
, no. , pp. –, .
[] J. H. Rimmer, A. Rauworth, E. Wang, P. S. Heckerling, and
B. S. Gerber, “A randomized controlled trial to increase phys-
ical activity and reduce obesity in a predominantly African
American group of women with mobility disabilities and severe
obesity,” Preventive Medicine,vol.,no.,pp.–,.
[] M. R. Stolley, L. K. Sharp, A. Oh, and L. Schier, “A weight loss
intervention for African American breast cancer survivors,
,” Preventing Chronic Disease,vol.,no.,articleA,.
[] E. Hemmingsson, M. Hell´
enius, U. Ekelund, J. Bergstr¨
om, and
S. R¨
ossner, “Impact of social support intensity on walking in the
severely obese: a randomized clinical trial,” Obesity,vol.,no.
, pp. –, .
[] R. T. Anderson, A. King, A. L. Stewart, F. Camacho, and W.
J. Rejeski, “Physical activity counseling in primary care and
patient well-being: do patients benet?” Annals of Behavioral
Medicine, vol. , no. , pp. –, .
[] K. I. Gallagher, J. M. Jakicic, M. A. Napolitano, and B. H. Mar-
cus, “Psychosocial factors related to physical activity and weight
loss in overweight women,” Medicine and Science in Sports and
Exercise, vol. , no. , pp. –, .

 Journal of Obesity
[] T. L. Pettman, G. M. H. Misan, K. Owen et al., “Self-
management for obesity and cardio-metabolic tness: descrip-
tion and evaluation of the lifestyle modication program of a
randomised controlled trial,” International Journal of Behavioral
Nutrition and Physical Activity,vol.,article,.
[] M. Kiernan, S. D. Moore, D. E. Schoman et al., “Social support
for healthy behaviors: scale psychometrics and prediction of
weight loss among women in a behavioral program,” Obesity,
vol. , no. , pp. –, .
[] C. R. Kalodner and J. L. DeLucia, “e individual and combined
eects of cognitive therapy and nutrition education as additions
to a behavior modication program for weight loss,” Addictive
Behaviors, vol. , no. , pp. –, .
[] K. Casazza and M. Ciccazzo, “e method of delivery of nutri-
tion and physical activity information may play a rolein eliciting
behavior changes in adolescents,” Eating Behaviors,vol.,no.,
pp.–,.
[] P. Hajek, K. Humphrey, and H. McRobbie, “Using group
support to complement a task-based weight management pro-
gramme in multi-ethnic localities of high deprivation,” Patient
Education and Counseling,vol.,no.,pp.–,.
[] A.K.Yancey,W.J.McCarthy,G.G.Harrison,W.K.Wong,J.
M. Siegel, and J. Leslie, “Challenges in improving tness: results
of a community-based, randomized, controlled lifestyle change
intervention,” Journal of Women’s Health,vol.,no.,pp.–
, .
[] J. H. Cousins, D. S. Rubovits, J. K. Dunn, R. S. Reeves, A. G.
Ramirez, and J. P. Foreyt, “Family versus individually oriented
intervention for weight loss in Mexican American women,”
Public Health Reports,vol.,no.,pp.–,.
[]B.Gold,P.Buzzell,H.Leonard,S.Pintauro,andJ.Harvey-
Berino, “Minimal in-person support as an adjunct to internet
obesity treatment,” Annals of Behavioral Medicine,vol.,no.,
pp.–,.
[] V. Leblanc, V. Provencher, C. B´
egin, L. Corneau, A. Tremblay,
and S. Lemieux, “Impact of a Health-At-Every-Size intervention
on changes in dietary intakes and eating patterns in pre-
menopausal overweight women: results of a randomized trial,”
Clinical Nutrition,vol.,pp.–,.
[] D.A.Williamson,C.M.Champagne,D.W.Harshaetal.,“Eect
of an environmental school-based obesity prevention program
on changes in body fat and body weight: a randomized trial,”
Obesity,vol.,no.,pp.–,.
[] T. T.-K. Huang and T. A. Glass, “Transforming research strate-
gies for understanding and preventing obesity,” Journal of the
American Medical Association,vol.,no.,pp.–,
.