Umberson and Montez
Journal of Health and Social Behavior
51(S) S54–S66
© American Sociological Association 2010
DOI: 10.1177/0022146510383501
http://jhsb.sagepub.com
Social Relationships and
Health: A Flashpoint for
Health Policy
Debra Umberson1 and Jennifer Karas Montez1
Abstract
Social relationships—both quantity and quality—affect mental health, health behavior, physical health, and
mortality risk. Sociologists have played a central role in establishing the link between social relationships
and health outcomes, identifying explanations for this link, and discovering social variation (e.g., by gender
and race) at the population level. Studies show that social relationships have short- and long-term effects
on health, for better and for worse, and that these effects emerge in childhood and cascade throughout
life to foster cumulative advantage or disadvantage in health. This article describes key research themes in
the study of social relationships and health, and it highlights policy implications suggested by this research.
Keywords
relationships, social support, social integration, stress, cumulative disadvantage
Captors use social isolation to torture prisoners of
war—to drastic effect. Social isolation of otherwise healthy, well-functioning individuals eventually
results in psychological and physical disintegration, and even death. Over the past few decades,
social scientists have gone beyond evidence of
extreme social deprivation to demonstrate a clear
link between social relationships and health in the
general population. Adults who are more socially
connected are healthier and live longer than their
more isolated peers. This article describes major
findings in the study of social relationships and
health, and how that knowledge might be translated into policy that promotes population health.
Key research findings include: (1) social relationships have significant effects on health; (2) social
relationships affect health through behavioral, psychosocial, and physiological pathways; (3) relationships have costs and benefits for health;
(4) relationships shape health outcomes throughout the life course and have a cumulative impact
on health over time; and (5) the costs and benefits
of social relationships are not distributed equally in
the population.
WHAT DO WE MEAN BY “SOCIAL
RELATIONSHIPS”?
Social scientists have studied several distinct features of social connection offered by relationships
(Smith and Christakis 2008). Social isolation refers
to the relative absence of social relationships.
Social integration refers to overall level of involvement with informal social relationships, such as
having a spouse, and with formal social relationships, such as those with religious institutions and
volunteer organizations. Quality of relationships
includes positive aspects of relationships, such as
emotional support provided by significant others,
and strained aspects of relationships, such as conflict and stress. Social networks refer to the web of
1
University of Texas at Austin
Corresponding Author:
Debra Umberson, University of Texas at Austin,
Department of Sociology, 1 University Station A1700,
Austin, TX 78712
E-mail: umberson@prc.utexas.edu
S55
Umberson and Montez
social relationships surrounding an individual, in
particular, structural features, such as the type and
strength of each social relationship. Each of these
aspects of social relationships affects health. We
discuss the broad effects of these features of relationships for health, and, for ease of discussion, we
use the terms “social relationships” and “social
ties” interchangeably throughout this article.
SOCIAL RELATIONSHIPS
BENEFIT HEALTH
Many types of scientific evidence show that
involvement in social relationships benefits health.
The most striking evidence comes from prospective studies of mortality across industrialized
nations. These studies consistently show that individuals with the lowest level of involvement in
social relationships are more likely to die than
those with greater involvement (House, Landis,
and Umberson 1988). For example, Berkman and
Syme (1979) showed that the risk of death among
men and women with the fewest social ties was
more than twice as high as the risk for adults with
the most social ties. Moreover, this finding held
even when socioeconomic status, health behaviors,
and other variables that might influence mortality,
were taken into account. Social ties also reduce
mortality risk among adults with documented medical conditions. For instance, Brummett and colleagues (2001) found that, among adults with coronary artery disease, the socially isolated had a risk
of subsequent cardiac death 2.4 times greater than
their more socially connected peers.
In addition to mortality, involvement in social
relationships has been associated with specific
health conditions as well as biological markers
indicating risk of preclinical conditions. Several
recent review articles provide consistent and compelling evidence linking a low quantity or quality
of social ties with a host of conditions, including
development and progression of cardiovascular
disease, recurrent myocardial infarction, atherosclerosis, autonomic dysregulation, high blood
pressure, cancer and delayed cancer recovery, and
slower wound healing (Ertel, Glymour, and Berkman 2009; Everson-Rose and Lewis 2005; Robles
and Kiecolt-Glaser 2003; Uchino 2006). Poor
quality and low quantity of social ties have also
been associated with inflammatory biomarkers and
impaired immune function, factors associated with
adverse health outcomes and mortality (KiecoltGlaser et al. 2002; Robles and Kiecolt-Glaser
2003). Marriage is perhaps the most studied social
tie. Recent work shows that marital history over
the life course shapes a range of health outcomes,
including cardiovascular disease, chronic conditions, mobility limitations, self-rated health, and
depressive symptoms (Hughes and Waite 2009;
Zhang and Hayward 2006).
HOW DO RELATIONSHIPS
BENEFIT HEALTH?
Once the clear link between social relationships
and health was established, scientists devoted
themselves to explaining how this occurs. Generally speaking, there are three broad ways that
social ties work to influence health: behavioral,
psychosocial, and physiological.
Behavioral Explanations
Health behaviors encompass a wide range of personal behaviors that influence health, morbidity,
and mortality. In fact, health behavior explains
about 40 percent of premature mortality as well as
substantial morbidity and disability in the United
States (McGinnis, Williams-Russo, and Knickman
2002). Some of these health behaviors—such as
exercise, consuming nutritionally balanced diets,
and adherence to medical regimens—tend to
promote health and prevent illness, while other
behaviors—such as smoking, excessive weight gain,
drug abuse, and heavy alcohol consumption—tend
to undermine health. Many studies provide evidence that social ties influence health behavior
(see a review in Umberson, Crosnoe, and Reczek
2010). For example, Berkman and Breslow’s
(1983) prospective study in Alameda County
showed that greater overall involvement with formal (e.g., religious organizations) and informal
(e.g., friends and relatives) social ties was associated with more positive health behaviors over a
ten-year period. Being married (Waite 1995), having children (Denney 2010), and ties to religious
organizations (Musick, House, and Williams 2004)
have all been linked to positive health behaviors
(although, notably, as we will discuss below, marriage and parenthood have also been associated
with behaviors that are not beneficial to health—
including physical inactivity and weight gain).
Social ties influence health behavior, in part,
because they influence, or “control,” our health
habits (Umberson et al. 2010). For example, a
spouse may monitor, inhibit, regulate, or facilitate
S56
health behaviors in ways that promote a partner’s
health (Waite 1995). Religious ties also appear to
influence health behavior, in part, through social
control (Ellison and Levin 1998). Social ties can
instill a sense of responsibility and concern for others that then lead individuals to engage in behaviors that protect the health of others, as well as
their own health. Social ties provide information
and create norms that further influence health habits. Thus, in a variety of ways, social ties may
influence health habits that in turn affect physical
health and mortality.
Psychosocial Explanations
Research across disciplines and populations suggests possible psychosocial mechanisms to explain
how social ties promote health. Mechanisms
include (but are not limited to): social support,
personal control, symbolic meanings and norms,
and mental health. While most studies focus on
only one or two of these mechanisms, it is clear
that connections between mechanisms are complex, and that these interconnections may explain
the linkage between social ties and health better
than any single mechanism (Thoits 1995; Umberson et al. 2010).
Social support refers to the emotionally sustaining qualities of relationships (e.g., a sense that
one is loved, cared for, and listened to). Hundreds
of studies establish that social support benefits
mental and physical health (Cohen 2004; Uchino
2004). Social support may have indirect effects on
health through enhanced mental health, by reducing the impact of stress, or by fostering a sense of
meaning and purpose in life (Cohen 2004; Thoits
1995). Supportive social ties may trigger physiological sequelae (e.g., reduced blood pressure,
heart rate, and stress hormones) that are beneficial
to health and minimize unpleasant arousal that
instigates risky behavior (Uchino 2006). Personal
control refers to individuals’ beliefs that they can
control their life outcomes through their own
actions. Social ties may enhance personal control
(perhaps through social support), and, in turn, personal control is advantageous for health habits,
mental health, and physical health (Mirowsky and
Ross 2003; Thoits 2006).
Many studies suggest that the symbolic meaning
of particular social ties and health habits explains
why they are linked. For example, meanings
attached to marriage and relationships with children may foster a greater sense of responsibility to
stay healthy, thus promoting healthier lifestyles
Journal of Health and Social Behavior 51(S)
(Nock 1998; Waite 1995). Studies on adolescents
often point to the meaning attached to peer groups
(e.g., what it takes to be popular) when explaining
the influence of peers on alcohol, tobacco, and drug
use (Crosnoe, Muller, and Frank 2004). The meaning of specific health behaviors within social contexts may also vary. For example, Schnittker and
McLeod (2005) argue that racial-ethnic identity
may correspond with the meaning of certain health
behaviors, such as consuming particular foods or
avoiding alcohol, in ways that promote and sustain
those behaviors. Moreover, the notion of “meaning” may help explain health behavior contagion
across social networks: for example, the spread of
obesity across social networks appears to be influenced by perceptions of social norms about the
acceptability of obesity and related health behaviors (e.g., food consumption, inactivity) among
network members who are socially close, rather
than members who are simply geographically close
(Christakis and Fowler 2007; Smith and Christakis
2008). In a more fundamental way, greater social
connection may foster a sense of “coherence” or
meaning and purpose in life, which, in turn,
enhances mental health, physiological processes,
and physical health (Antonovsky 1987).
Mental health is a pivotal mechanism that works
in concert with each of the other mechanisms to
shape physical health (Chapman, Perry, and Strine
2005). For instance, the emotional support provided
by social ties enhances psychological well-being,
which, in turn, may reduce the risk of unhealthy
behaviors and poor physical health (Kiecolt-Glaser
et al. 2002; Thoits 1995; Uchino 2004). Moreover,
mental health is an important health outcome in and
of itself. The World Health Organization identifies
mental health as an essential dimension of overall
health status (World Health Organization 2007).
However, the prevalence of mental disorders and
their consequences for individuals and societies are
often underappreciated by policy makers and private insurers. Data from the National Comorbidity
Survey Replication indicate that 26.2 percent of
noninstitutionalized U.S. adults suffer from a mental disorder in a given year (Kessler et al. 2005). As
the leading cause of disability in both low- and highincome countries, mental disorders account for over
37 percent of the total years of healthy life lost due
to disability (Mathers et al. 2006).
Physiological Explanations
Psychologists, sociologists, and epidemiologists
have contributed a great deal to our understanding
Umberson and Montez
of how social processes influence physiological
processes that help to explain the link between
social ties and health. For example, supportive
interactions with others benefit immune, endocrine, and cardiovascular functions and reduce
allostatic load, which reflects wear and tear on the
body due, in part, to chronically overworked physiological systems engaged in stress responses
(McEwen 1998; Seeman et al. 2002; Uchino
2004). These processes unfold over the entire life
course, with effects on health. Emotionally supportive childhood environments promote healthy
development of regulatory systems, including
immune, metabolic, and autonomic nervous systems, as well as the hypothalamic-pituitary-adrenal
(HPA) axis, with long-term consequences for adult
health (Taylor, Repetti, and Seeman 1997). Social
support in adulthood reduces physiological
responses such as cardiovascular reactivity to both
anticipated and existing stressors (Glynn, Christenfeld, and Gerin 1999). Indeed, continuously
married adults experience a lower risk of cardiovascular disease compared with those who have
experienced a marital loss, in part due to the psychosocial supports conferred by marriage (Zhang
and Hayward 2006).
THE DARK SIDE OF SOCIAL
RELATIONSHIPS
While social relationships are the central source of
emotional support for most people, social relationships can be extremely stressful (Walen and Lachman
2000). For example, marriage is the most salient
source of both support and stress for many individuals (Walen and Lachman 2000), and poor
marital quality has been associated with compromised immune and endocrine function and depression (Kiecolt-Glaser and Newton 2001). Sociological research shows that marital strain erodes
physical health, and that the negative effect of
marital strain on health becomes greater with
advancing age (Umberson et al. 2006).
Relationship stress undermines health through
behavioral, psychosocial, and physiological pathways. For example, stress in relationships contributes to poor health habits in childhood, adolescence,
and adulthood (Kassel et al. 2003). Stress contributes to psychological distress and physiological
arousal (e.g., increased heart rate and blood pressure) that can damage health through cumulative
wear and tear on physiological systems, and by
leading people of all ages to engage in unhealthy
S57
behaviors (e.g., food consumption, heavy drinking,
smoking) in an effort to cope with stress and reduce
unpleasant arousal (Kassel, Stroud, and Paronis
2003). The propensity to engage in particular risky
health behaviors in response to stress appears to
vary over the life course. For example, stress is
associated with more alcohol consumption in
young adulthood and greater weight gain in midlife (Umberson et al. 2010). Relationship stress also
undermines a sense of personal control and mental
health, both of which are, in turn, associated with
poorer physical health (Mirowsky and Ross 2003).
It may seem obvious that strained and conflicted social interactions undermine health, but
social ties may have other types of unintended
negative effects on health. For example, relationships with risk-taking peers contribute to increased
alcohol consumption, and having an obese spouse
or friend increases personal obesity risk (Christakis and Fowler 2007; Crosnoe et al. 2004). This
“social contagion” of negative health behaviors
operates via multiple mechanisms (Smith and
Christakis 2008). One key mechanism is social
norms. Perceived social norms about drinking
behavior influence alcohol consumption among
young adults (Thombs, Wolcott, and Farkash
1997), and friendship norms about dieting influence unhealthy weight control (Eisenberg et al.
2005). Unsupportive social ties may also present
barriers to improving health behaviors and outcomes. For example, Nagasawa and colleagues
(1990) found that negative social environments
and their perceived barriers predicted poor compliance to medical regimens among diabetes patients.
Caring for one’s social ties may also involve
personal health costs. For example, providing care
to a sick or impaired spouse imposes strains that
undermine the health of the provider, even to the
point of elevating mortality risk for the provider
(Christakis and Allison 2006). Caring for a sick or
impaired spouse is associated with increased physical and psychiatric morbidity, impaired immune
function, poorer health behavior, and worse health
for the provider (Schulz and Sherwood 2008).
Moreover, the recipient of care may be negatively
affected by interpersonal interactions with stressed
caregivers (Bediako and Friend 2004). Middleaged adults, particularly women, often experience
exceptionally high caregiving demands as they
contend with the challenge of simultaneously rearing children, caring for spouses, and looking after
aging parents (Spain and Bianchi 1996). The combination of smaller families (to share in the caregiving of aging parents) and an aging population
S58
mean that the multigenerational demands of social
ties may become more pronounced in the future.
CUMULATIVE ADVANTAGE AND
DISADVANTAGE
All Americans are not at equal risk for risky health
behaviors, morbidity, and premature mortality.
Throughout life, we are exposed to social conditions that promote or undermine health, and over
time these exposures accumulate to create growing
advantage or disadvantage for health in socially
patterned ways. Thus, social variation in relationships/health processes provides information that
may be used to address social disparities in health.
The most salient social ties for health vary over
the life course, with parents having the greatest
influence on children’s health, peers becoming particularly important in adolescence, intimate partners
becoming most important in adulthood, and adult
children taking an elevated role in later life (Umberson et al. 2010). The principal explanatory mechanisms may also vary over the life course. For
example, stressful family interactions may have
their greatest impact on children’s health, while peer
pressure and the social meaning of health habits
(e.g., pressure to experiment with tobacco, alcohol,
and drugs) may have their greatest impact in adolescent relationships, and social control of health habits
may be most important in adult relationships.
Some effects of social ties are more immediate,
while others slowly build over time. For example,
at any given point in time, ongoing social ties affect
mental health and health behavior—for better or for
worse. These effects may or may not dissipate over
time, but recent work on the effects of distressed,
disrupted, and emotionally unsupportive childhood
environments on adult health shows that these
effects reverberate throughout the life course (Crosnoe and Elder 2004; Palloni 2006; Shaw et al.
2004). Certainly, chronic isolation or strain in
social ties take an increasing toll over time on a
host of health indicators including allostatic load
(Seeman et al. 2002), blood pressure (Cacioppo
et al. 2002), physical health (Umberson et al. 2006),
and mortality risk (Berkman and Syme 1979).
COSTS AND BENEFITS OF SOCIAL
RELATIONSHIPS: INEQUALITIES
Both quantitative (size and diversity) and qualitative (benefits and costs) aspects of social ties are
Journal of Health and Social Behavior 51(S)
demographically patterned and socially constructed. Regarding size, women tend to have
larger confidant networks than men, as do whites
compared with blacks, better-educated adults
compared with less-educated, and, to a lesser
extent, younger adults (McPherson, Smith-Lovin,
and Brashears 2006). Moreover, the diversity
of social ties varies in patterned ways with,
for example, better-educated adults engaged in
more diverse personal networks (McPherson et al.
2006). Sociodemographic variation in quantitative
aspects of social ties may partly explain parallel
variation in health disparities because both size
(Brummett et al. 2001) and diversity (Cohen et al.
1997) of social ties enhance health. People with a
greater number of ties have a larger pool of confidants from which to connect and to receive social
support and health-relevant information.
In general, we know little about how the benefits and costs of social ties vary across sociodemographic groups, but some evidence suggests that
there is variation. Most attention has been devoted
to gender differences, particularly in the context of
marriage. Historically, marriage has conferred
more health gains for men than for women. Men
not only experience greater health benefits through
the positive lifestyle and health behaviors that
often accompany marriage (Waite 1995), they also
experience fewer costs from spousal caregiving,
childrearing, caring for aging parents, and balancing work/family demands (Spain and Bianchi
1996). The availability, costs, and benefits of
social ties may also vary by race. For instance,
blacks are less likely to be married than whites. Yet
evidence regarding costs and benefits is mixed.
African Americans may experience more marital
strain (Broman 1993; Umberson et al. 2005; cf.
Kiecolt, Hughes, and Keith 2008) and receive
fewer economic gains from marriage compared to
whites (Willson 2003), yet some studies find African Americans have historically received more
health benefits from marriage than whites (Kiecolt
et al. 2008; Liu and Umberson 2008). Disparities
in the quantity and quality of social ties exist
across socioeconomic statuses as well. More educated adults have a larger number of close confidants and may experience less stress in their
relationships. For instance, women with a high
school degree or less are roughly twice as likely to
divorce within 10 years of their first marriage compared with women having at least a bachelor’s
degree (Martin 2006). Notably, differential access,
benefits, and costs to social ties across sociodemographic groups are not immutable; recent work
Umberson and Montez
S59
shows that these differentials have changed significantly over time (Liu and Umberson 2008;
McPherson et al. 2006).
health means reduced health care costs as well as
better quality of life for Americans, regardless of
their age.
SOCIAL TIES: AN INVESTMENT IN
POPULATION HEALTH
PUBLIC POLICY: SOCIAL TIES AND
HEALTH OF THE POPULATION
Research shows that social ties influence multiple
and interrelated health outcomes, including health
behaviors, mental health, physical health, and mortality risk. Thus, a policy focus on social ties may
prove to be a cost-effective strategy for enhancing
health and well-being at the population level
(McGinnis et al. 2002; Mechanic and Tanner
2007). Social ties may be unique in their ability to
affect a wide range of health outcomes and to
influence health (thus cumulative health outcomes)
throughout the entire life course. Moreover, interventions and policies that strengthen and support
individuals’ social ties have the potential to
enhance the health of others connected to those
individuals. For example, reducing strain and
improving health habits of a married person may
benefit the health of both partners, as well any
children they care for.
Recent work also shows that some health outcomes can “spread” widely through social networks. For example, obesity increases substantially
for those who have an obese spouse or friends
(Christakis and Fowler 2007), and happiness
appears to spread through social networks as well
(Fowler and Christakis 2008). These findings suggest that the impact of social ties on one person’s
health goes beyond that person to influence the
health of broader social networks. Thus, policies
and interventions should capitalize on this natural
tendency for health-related attitudes and behaviors
to spread through social networks by incorporating
these amplification effects into the mechanics of
interventions and their cost-benefit estimates
(Smith and Christakis 2008).
Finally, enhanced relationship/health linkages
can be viewed as preventive medicine. While
social ties may serve to improve health outcomes
for those who develop serious health conditions,
social ties may help prevent these conditions from
developing in the first place. Policies that promote
and protect social ties should have both short-term
and long-term payoffs. If social ties foster psychological well-being and better health habits throughout the life course, then social ties can add to
cumulative advantage in health over time—a
worthwhile goal for an aging population. Better
Social ties and their connection to health have
important implications for health policy. Indeed,
some existing social policies and programs implicitly and indirectly incorporate social ties as mechanisms for enhancing population health and wellbeing. For example, many programs concerned
with health of the elderly (e.g., home health services and meal deliveries) direct attention to the
impact of social isolation/connection on health.
Healthy People 2010, a nationwide health promotion plan developed by the Department of Health
and Human Services, recognizes that social ties
play an important role in influencing health habits
(U.S. Department of Health and Human Services
2000). The Healthy Marriage Initiative recognizes
that marriages characterized by supportive interactions benefit the health of children as well as
spouses (U.S. Department of Health and Human
Services n.d.). The Family Medical Leave Act
(FMLA) allows eligible employees to take up to 12
weeks of unpaid, protected leave over a 12-month
period to attend to certain medical and familyrelated needs, such as the birth of a child or caring
for an immediate family member (U.S. Department of Labor 2009).
Yet in some cases these policies and programs
do not benefit the populations that need them the
most, or they unintentionally undermine the health
of the target population and others in their social
network. For example, FMLA may benefit those
who receive care, but it also may be financially
prohibitive for caregivers who do not have an
employed spouse or enough savings to support
them through the time off work, yet those with the
fewest financial resources and social ties may need
assistance the most. Further, in rare cases, experimental programs have reported worse health outcomes among subgroups of participants. A
randomized experiment of the effects of support
groups for women with breast cancer found that,
compared to women in the control group, the
physical functioning of women who participated in
the peer discussion group improved if they reported
low levels of emotional support from their partners
at baseline, but it deteriorated if they reported initially high levels (Helgeson et al. 2000). Another
S60
psychosocial intervention tested individualized
emotional and instrumental support services in an
effort to improve one-year survival outcomes of
adults recovering from myocardial infarction. This
study found that, compared to a control group, men
in the intervention group exhibited similar mortality rates while women exhibited higher mortality
rates during the one-year follow-up (Frasure-Smith
et al. 1997). Thus, we must develop a policy foundation that integrates scientific evidence on the
linkages between social ties and health, and that
foundation must do two things: (1) ensure that
policies and programs benefit the populations that
need them; and (2) maximize health-related benefits for recipients while minimizing costs for providers and recipients.
POLICY FOUNDATION
Poor mental and physical health and unhealthy
behaviors exact a huge toll on individuals, families,
and society. Solid scientific evidence establishing
the causal impact of social ties on health provides
the impetus for policy makers to ensure that U.S.
health policy works to protect and promote social
ties that benefit health. Scientific evidence supports
the following premises, and it is from this empirical
footing that we can build a policy foundation for
promoting both social ties and health:
1. Social ties affect mental health, physical
health, health behaviors, and mortality risk.
2. Social ties are a potential resource that can
be harnessed to promote population health.
3. Social ties are a resource that should be
protected as well as promoted.
4. Social ties can benefit health beyond target
individuals by influencing the health of
others throughout social networks.
5. Social ties have both immediate (mental
health, health behaviors) and long-term,
cumulative effects on health (e.g., physical
health, mortality), and thus represent
opportunities for short- and long-term
investment in population health.
Although social ties have the potential to benefit health, policy efforts must recognize that social
ties also have the potential to undermine health,
and that the link between social ties and health
may vary across social groups. For example, gender, race, and age are associated with different
levels and types of responsibilities, strains, and
resources in social ties that then influence personal
Journal of Health and Social Behavior 51(S)
health habits as well as the health of significant
others. In order to be effective, policies and interventions must account for the ways in which social
constraints and resources influence health across
social groups (House et al. 2008). Moreover, care
must be taken to develop strategies that increase the
power of social ties to enhance individual health
without imposing additional strains on care providers. Thus, we suggest two additional policy components for the basic foundation suggested above:
6. Caveat: social ties—overburdened, strained,
conflicted, abusive—can undermine health.
7. The costs and benefits of social ties are not
distributed equally in the population (e.g.,
age, socioeconomic status, gender, race
variation).
POLICY GOALS
How can policy makers use the scientific findings
on social ties and health to advance population
health and reduce social disparities in health? They
can begin by addressing six fundamental goals.
Promote Benefits of Social Ties
Support and promote positive features of social
ties (e.g., supportive interactions, healthy lifestyle
norms). For example, Health and Human Service’s
Healthy Marriage Initiative is designed to promote
positive marital interactions that may foster mental
and physical health of couples and their children.
This initiative uses a multifaceted approach,
including public awareness campaigns on responsible parenting and the value of healthy marriages,
as well as educational and counseling services
delivered through local organizations such as
schools and faith-based organizations. This goal
should also speak to policies that deny marriage to
same-sex couples. The absence of legal marriage
may reduce the benefits of committed partnerships
for the health of individuals in gay and lesbian
relationships (Herek 2006; King and Bartlett 2006;
Wienke and Hill 2009).
Do No Harm
Avoid policies, programs, and interventions that
increase relationship burdens and strains or that
undermine positive features of relationships. For
example, many programs for the sick and elderly
increase caregiving responsibilities for family
members—responsibilities that may impose stress
S61
Umberson and Montez
on caregivers and on family relationships. This
problem is exacerbated by hospital and insurance
policies that force family members to provide
medical care at home. These costs are borne more
heavily by women, minorities, and those with
fewer socioeconomic resources. Policy efforts
should recognize that specific programs may benefit some groups but harm others.
relation to health. These existing strategies can be
mapped onto a general strategy of promoting positive relationship/health linkages. This will make
gaps and overlaps between strategies more apparent, and it will allow greater coordination of services for helping professionals and for citizens
seeking services.
Provide Help Where Help Is Most Needed
Reduce Social Isolation
This addresses one of the most fundamental findings from research on social ties and health: The
most socially isolated Americans are those at
greatest risk of poor health and early mortality
(Brummett et al. 2001). Policies can reduce the
risk of social isolation in the first place by enhancing our educational system to impart socialemotional skills, interests in civic engagement, and
meaningful employment (Greenberg et al. 2003);
by ensuring that all communities are economically
developed and contain public places to safely congregate and exercise (Mechanic and Tanner 2007);
and by fostering stable marriages and families for
all Americans. Notably, some groups are more
likely than others to experience social isolation.
For example, widowhood increases the risk of
social isolation. Women are more likely than men
to be widowed, and widowhood affects a higher
proportion of African Americans than other races,
and at earlier ages; among those aged 65 to 74,
24.3 percent of African Americans are widowed
compared to 14.8 percent of whites (U.S. Census
2009). Coordinated programs could help identify
socially isolated adults, perhaps through their physicians, and they could mobilize local resources to
offer social and instrumental support to these individuals.
Reduce Harm
Prevent and alleviate negative features of social
ties. For example, work to reduce strains for those
who provide care to children, sick or impaired
significant others, and the elderly, remaining cognizant of unintended effects on caregivers. In addition, prevent or alleviate harm caused by negative
social ties, such as abusive parent-child relationships and strained marriages.
Coordinate Policies and Programs
Many existing policies and programs, at least
implicitly, address some aspect of social ties in
Some populations are at greater risk for illness and
disease than others, and these groups should receive
higher priority in policy efforts. In particular, some
populations are more likely to be socially isolated
(e.g., the poor and the elderly), and some are more
likely to be burdened by caring for others in their
social networks (e.g., women, especially African
American women). Existing policies should also be
re-evaluated to ensure that they help the populations that need them most. For instance, the FMLA
may be entirely unhelpful for low-income adults
who have no alternative source of income and are
more likely to be without alternative sources of
instrumental and emotional support.
FUTURE RESEARCH
Social scientists can advance this policy agenda by
addressing several specific issues. First, it is
important to identify individuals most at risk, as
well as explanations for heightened risk. Individuals who are socially isolated may be at the greatest
health risk. Several studies suggest that the relationship between social ties and health is nonlinear
so that individuals with no social ties or very few
social ties exhibit the most pronounced risk of poor
health (Brummett et al. 2001; Cohen et al. 1997;
Seeman et al. 2002). Despite the considerable
evidence linking social isolation to poor health
outcomes, the causal mechanisms are poorly
understood. We need to investigate the possibility
that differences between socially isolated and
socially integrated adults—in health behaviors,
emotional and instrumental support networks,
physiological responses to anxiety, or other mechanisms—explain the linkage. Sociologists should
direct attention to the social distribution of isolation and the possibility that the consequences of
social isolation vary across social groups.
Second, the broader social context—as structured by age, class, race, and gender—influences
the formation and quality of social ties as well as
the processes through which social ties affect
S62
health. However, the ways in which these structural variables shape social ties are not well understood, and few studies consider how these structural
variables might modify relationship/health linkages. Likewise, social ties may shape the way that
structural variables influence health. For instance,
marital status may alter the inverse association
between educational attainment and mortality
risks, at least for men (Montez et al. 2009). This
type of research is needed in order to identify atrisk populations as well as explanatory mechanisms linking social ties to health outcomes across
social groups.
Third, past work on social ties and health habits
tends to emphasize the benefits of social ties for
health, yet research on stress clearly shows that
strained social ties undermine health. Given the
ability of social ties to have both positive and
negative effects on health, existing research has
likely underestimated the true impact of social ties
on health. Future research should consider how the
positive and negative facets of social ties work
together to influence health outcomes, as well as
consider how this balance may vary over the life
course and across social groups.
A growing body of theoretical and empirical
work illustrates how social conditions foster cumulative advantage and disadvantage for health over
the life course. This may be a case of the rich getting richer while the poor get poorer, in that
strained and unsupportive relationships in childhood launch into motion a cascade of factors—
such as increased risk for depression, low personal
control, and poor health habits—that lead to poorer
health and more strained and less supportive relationships across the life course. Scholars should
consider this cascading process, and they should
identify at-risk populations as well as the most
important modifiable risk and protective factors in
their social relationships. Scholars should also help
to clarify when social ties impact health habits, as
well as identify which social ties are most important to health at different life stages.
In addition, future research will benefit from
methodological considerations, including a greater
focus on prospective survey designs and corresponding longitudinal analyses, dyadic information about social relationships, and qualitative
data. Prospective designs are essential in order to
consider how relationship/health linkages and
explanatory mechanisms unfold over time. This
approach fits with the life course notion that determinants of current health originate early in life and
accumulate across the life span (Ben-Shlomo and
Kuh 2002). Taking full advantage of prospective
Journal of Health and Social Behavior 51(S)
surveys through longitudinal data analysis and
wider application of multilevel modeling could
shed more light on the social processes involved in
building, sustaining, and benefiting from social
ties across the life course.
Most studies on social ties and health use
individual-level data, as surveys typically collect
information from one member per household.
However, social ties, by definition, involve more
than one person. Studies that include dyads show
that individuals in the same relationship often
experience and report on their relationship in quite
different ways (Proulx and Helms 2008). Independent reports, as well as discrepancies between
reports, may be linked to health outcomes. We
should take advantage of existing longitudinal data
sets that include more than one focal individual.
New data collection efforts should go beyond the
individual to include data from a range of linked
social ties. As recent work shows, including reports
from several network members may reveal important relationship/health linkages that go beyond
one individual (Smith and Christakis 2008).
Finally, most research on social ties and health
has relied on assessment of quantitative data
sources. Quantitative data are essential for identifying patterns between variables in the general population and, particularly, for revealing how social
location (e.g., as defined by life course stage, race,
and gender) is associated with regularity in social
experiences (e.g., relationships and health). However, population-level data are limited in their ability to reveal rich social contexts that allow us to
analyze the meanings, dynamics, and processes that
link social ties to health over time. Thus, blending
qualitative and quantitative methods provides the
opportunity to build on the strengths of both methodologies and to address how structure and meaning coalesce to shape health outcomes at the
population level (Pearlin 1992). Information
obtained from qualitative data may also suggest
new explanations (e.g., new psychosocial mechanisms or connections between mechanisms) for
relationship/health linkages, and for group differences in those linkages, and those explanations can
be further explored using population-level data.
CONCLUSION
Solid scientific evidence shows that social relationships affect a range of health outcomes, including
mental health, physical health, health habits, and
mortality risk. Sociologists have played a major
role in establishing these linkages, in identifying
Umberson and Montez
explanations for the impact of social relationships
on health, and in discovering social variation (e.g.,
by age and gender) in these linkages at the population level. The unique perspective and research
methods of sociology provide a scientific platform
to suggest how policy makers might improve population health by promoting and protecting Americans’ social relationships. Recent and projected
demographic trends should instill a sense of urgency
in developing policy solutions. Specifically, the
confluence of smaller families, high divorce rates,
employment-related geographical mobility, and
population aging means that adults of all ages, and
in particular the elderly, will be at increasing risk of
social isolation and shrinking family ties in the
future (Cacioppo and Hawkley 2003).
ACKNOWLEDGMENTS
An earlier version of this article was presented at the 2009
annual meeting of the American Sociological Association.
This research was supported by National Institute on Aging
grant RO1AG026613 (PI: Debra Umberson), National Institute of Child Health and Human Development grant 5 R24
HD042849 (PI: Mark D. Hayward) and 2 T32 HD007081
(PI: Robert A. Hummer) awarded to the Population Research
Center at the University of Texas at Austin.
REFERENCES
Antonovsky, Aaron. 1987. Unraveling the Mystery of
Health. San Francisco, CA: Jossey-Bass.
Bediako, Shawn M. and Ronald Friend. 2004. “IllnessSpecific and General Perceptions of Social Relationships in Adjustment to Rheumatoid Arthritis:
The Role of Interpersonal Expectations.” Annals of
Behavioral Medicine 28:203–10.
Ben-Shlomo, Yoav and Diana Kuh. 2002. “A Life Course
Approach to Chronic Disease Epidemiology: Conceptual Models, Empirical Challenges, and Interdisciplinary Perspectives.” International Journal of
Epidemiology 31:285–93.
Berkman, Lisa F. and Lester Breslow. 1983. Health and
Ways of Living: The Alameda County Study. New
York: Oxford University Press.
Berkman, Lisa F. and Leonard Syme. 1979. “Social Networks, Host Resistance, and Mortality: A Nine-Year
Follow-up Study of Alameda County Residents.”
American Journal of Epidemiology 117:1003–1009.
Broman, Clifford L. 1993. “Race Differences in Marital
Well-Being.” Journal of Marriage and Family 55:724–32.
Brummett, Beverly H., John C. Barefoot, Ilene C. Siegler,
Nancy E. Clapp-Channing, Barbara L. Lytle, Hayden B.
Bosworth, Redford B. Williams, and Daniel B. Mark.
S63
2001. “Characteristics of Socially Isolated Patients
with Coronary Artery Disease Who Are at Elevated Risk for Mortality.” Psychosomatic Medicine
63:267–72.
Cacioppo, John T. and Louise C. Hawkley. 2003. “Social
Isolation and Health, with an Emphasis on Underlying Mechanisms.” Perspectives in Biology and Medicine 46:S39–S52.
Cacioppo, John T., Louise C. Hawkley, L. Elizabeth
Crawford, John M. Ernst, Mary H. Burleson, Ray B.
Kowalewski, William B. Malarkey, Eve Van Cauter, and
Gary G. Berntson. 2002. “Loneliness and Health: Potential Mechanisms.” Psychosomatic Medicine 64:407–17.
Chapman, Daniel P., Geraldine S. Perry, and Tara W.
Strine. 2005. “The Vital Link between Chronic Disease and Depressive Disorders.” Preventive Chronic
Disease 2:1–10.
Christakis, Nicholas A. and Paul D. Allison. 2006. “Mortality after the Hospitalization of a Spouse.” The New
England Journal of Medicine 354:719–30.
Christakis, Nicholas A. and James H. Fowler. 2007. “The
Spread of Obesity in a Large Social Network over
32 Years.” The New England Journal of Medicine
357:370–79.
Cohen, Sheldon. 2004. “Social Relationships and
Health.” American Psychologist 59:676–84.
Cohen, Sheldon, William J. Doyle, David P. Skoner,
Bruce S. Rabin, and Jack M. Gwaltney, Jr. 1997.
“Social Ties and Susceptibility to the Common
Cold.” Journal of the American Medical Association
277:1940–44.
Crosnoe, Robert and Glenn H. Elder, Jr. 2004. “From
Childhood to the Later Years: Pathways of Human
Development.” Research on Aging 26:623–54.
Crosnoe, Robert, Chandra Muller, and Kenneth Frank.
2004. “Peer Context and the Consequences of Adolescent Drinking.” Social Problems 51:288–304.
Denney, Justin T. 2010. “Family and Household Formations and Suicide in the United States.” Journal of
Marriage and Family 72:202–13.
Eisenberg, Marla E., Dianne Neumark-Sztainer, Mary
Story, and Cheryl Perry. 2005. “The Role of Social
Norms and Friends’ Influences on Unhealthy WeightControl Behaviors among Adolescent Girls.” Social
Science and Medicine 60:1165–73.
Ellison, Christopher G. and Jeffrey S. Levin. 1998. “The
Religion-Health Connection: Evidence, Theory, and
Future Directions.” Health Education and Behavior
25:700–20.
Ertel, Karen A., Maria Glymour, and Lisa F. Berkman.
2009. “Social Networks and Health: A Life Course
Perspective Integrating Observational and Experimental Evidence.” Journal of Social and Personal
Relationships 26:73–92.
S64
Everson-Rose, Susan A. and Tené T. Lewis. 2005. “Psychosocial Factors and Cardiovascular Diseases.”
Annual Review of Public Health 26:469–500.
Fowler, James H. and Nicholas A. Christakis. 2008.
“Dynamic Spread of Happiness in a Large Social
Network: Longitudinal Analysis over 20 Years in the
Framingham Heart Study.” British Medical Journal
337:a2338.
Frasure-Smith, Nancy, François Lespérance, Raymond
H. Prince, Pierre Verrier, Rachel A. Garber, Martin
Juneau, Christina Wolfson, and Martial G. Bourassa.
1997. “Randomised Trial of Home-Based Psychosocial Nursing Intervention for Patients Recovering
from Myocardial Infarction.” Lancet 350:473–79.
Glynn, Laura M., Nicholas Christenfeld, and William
Gerin. 1999. “Gender, Social Support, and Cardiovascular Responses to Stress.” Psychosomatic Medicine 61:234–42.
Greenberg, Mark T., Roger P. Weissberg, Mary Utne
O’Brien, Joseph E. Zins, Linda Fredericks, Hank
Resnik, and Maurice J. Elias. 2003. “Enhancing
School-Based Prevention and Youth Development
Through Coordinated Social, Emotional, and Academic Learning.” American Psychologist 58:466–74.
Helgeson, Vicki S., Sheldon Cohen, Richard Schulz, and
Joyce Yasko. 2000. “Group Support Interventions
for Women with Breast Cancer: Who Benefits from
What?” Health Psychology 19:107–14.
Herek, G. M. 2006. “Legal Recognition of Same-Sex
Relationships in the United States: A Social Science
Perspective.” American Psychologist 61:607–21.
House, James S., Karl Landis, and Debra Umberson.
1988. “Social Relationships and Health.” Science
241:540–45.
House, James S., Robert F. Schoeni, George A. Kaplan,
and Harold Pollack. 2008. “The Health Effects of
Social and Economic Policy: The Promise and Challenge for Research and Policy.” Pp. 1–26 in Making
Americans Healthier: Social and Economic Policy as
Health Policy, edited by R. F. Schoeni, J. S. House,
G. A. Kaplan, and H. Pollack. New York: Russell
Sage Foundation.
Hughes, Mary Elizabeth and Linda J. Waite. 2009. “Marital Biography and Health at Mid-Life.” Journal of
Health and Social Behavior 50:344–58.
Kassel, Jon D., Laura R. Stroud, and Carol A. Paronis.
2003. “Smoking, Stress, and Negative Affect: Correlation, Causation, and Context across Stages of
Smoking.” Psychological Bulletin 129:270–304.
Kessler, Ronald C., Wai Tat Chiu, Olga Demler, and
Ellen E. Walters. 2005. “Prevalence, Severity, and
Comorbidity of 12-Month DSM-IV Disorders in the
National Comorbidity Survey Replication.” Archives
of General Psychiatry 62:617–27.
Journal of Health and Social Behavior 51(S)
Kiecolt, K. Jill, Michael Hughes, and Verna M. Keith.
2008. “Race, Social Relationships, and Mental
Health.” Personal Relationships 15:229–45.
Kiecolt-Glaser, Janice K., Lynanne McGuire, Theodore
F. Robles, and Ronald Glaser. 2002. “Emotions, Morbidity, and Mortality: New Perspectives from Psychoneuroimmunology.” Annual Review of Psychology
53:83–107.
Kiecolt-Glaser, Janice K. and Tamara L. Newton. 2001.
“Marriage and Health: His and Hers.” Psychological
Bulletin 127:472–503.
King, Michael and Annie Bartlett. 2006. “What Same Sex
Civil Partnerships May Mean for Health.” Journal of
Epidemiology and Community Health 60:188–191.
Liu, Hui and Debra J. Umberson. 2008. “‘The Times
They Are a Changin’: Marital Status and Health Differentials from 1972 to 2003.” Journal of Health and
Social Behavior 49:239–53.
Martin, Steven P. 2006. “Trends in Marital Dissolution
by Women’s Education in the United States.” Demographic Research 15:537–60.
Mathers, Colin D., Alan D. Lopez, and Christopher J. L.
Murray. 2006. “The Burden of Disease and Mortality
by Condition: Data, Methods, and Results for 2001.”
Pp. 45–240 in Global Burden of Disease and Risk
Factors, edited by A. D. Lopez, C. D. Mathers, M.
Ezzati, D. T. Jamison, and C. J. L. Murray. The World
Bank and Oxford University Press. Retrieved January
29, 2010 (http://files.dcp2.org/pdf/GBD/GBD03.pdf).
McEwen, Bruce S. 1998. “Stress, Adaptation, and Disease: Allostasis and Allostatic Load.” Annals of the
New York Academy of Sciences 840:33–44.
McGinnis, J. Michael, Pamela Williams-Russo, and
James R. Knickman. 2002. “The Case for More
Active Policy Attention to Health Promotion.” Health
Affairs 21:78–93.
McPherson, Miller, Lynn Smith-Lovin, and Matthew
E. Brashears. 2006. “Social Isolation in America:
Changes in Core Discussion Networks over Two
Decades.” American Sociological Review 71:353–75.
Mechanic, David and Jennifer Tanner. 2007. “Vulnerable
People, Groups, and Populations: Societal View.”
Health Affairs 26:1220–30.
Mirowsky, John and Catherine Ross. 2003. Social Causes
of Psychological Distress. 2nd ed. New York: Aldine
de Gruyter.
Montez, Jennifer Karas, Mark D. Hayward, Dustin C.
Brown, and Robert A. Hummer. 2009. “Why Is the
Educational Gradient of Mortality Steeper for Men?”
Journal of Gerontology: Social Sciences 64:625–34.
Musick, Marc A., James S. House, and David R.
Williams. 2004. “Attendance at Religious Services
and Mortality in a National Sample.” Journal of
Health and Social Behavior 45:198–213.
Umberson and Montez
Nagasawa, Masako, Mickey C. Smith, James H. Barnes,
and Jack E. Finchman. 1990. “Meta-Analysis of
Correlates of Diabetes Patients’ Compliance with
Prescribed Medications.” The Diabetes Educator
16:192–200.
Nock, Steven L. 1998. Marriage in Men’s Lives. New
York: Oxford University Press.
Palloni, Alberto. 2006. “Reproducing Inequalities: Luck,
Wallets, and the Enduring Effects of Childhood
Health.” Demography 43:587–615.
Pearlin, Leonard I. 1992. “Structure and Meaning in
Medical Sociology.” Journal of Health and Social
Behavior 33:1–9.
Proulx, Christine M. and Heather M. Helms. 2008.
“Mothers’ and Fathers’ Perceptions of Change and
Continuity in Their Relationships with Young Adult
Sons and Daughters.” Journal of Family Issues
29:234–61.
Robles, Theodore F. and Janice K. Kiecolt-Glaser. 2003.
“The Physiology of Marriage: Pathways to Health.”
Physiology and Behavior 79:409–16.
Schnittker, Jason and Jane D. McLeod. 2005. “The Social
Psychology of Health Disparities.” Annual Review of
Sociology 31:75–103.
Schulz, Richard and Paula R. Sherwood. 2008. “Physical and Mental Health Effects of Family Caregiving.”
American Journal of Nursing 108:23–27.
Seeman, Teresa E., Burton H. Singer, Carol D. Ryff,
Gayle Dienberg Love, and Lené Levy-Storms. 2002.
“Social Relationships, Gender, and Allostatic Load
across Two Age Cohorts.” Psychosomatic Medicine
64:395–406.
Shaw, Benjamin A., Neal Krause, Linda M. Chatters,
Cathleen M. Connell, and Berit Ingersoll-Dayton.
2004. “Emotional Support from Parents Early in Life,
Aging, and Health.” Psychology and Aging 19:4–12.
Smith, Kirsten P. and Nicholas A. Christakis. 2008.
“Social Networks and Health.” Annual Review of
Sociology 34:405–29.
Spain, Daphne and Suzanne M. Bianchi. 1996. Balancing
Act: Motherhood, Marriage, and Employment among
American Women. New York: Russell Sage Foundation.
Taylor, Shelley E., Rena L. Repetti, and Teresa Seeman.
1997. “Health Psychology: What Is an Unhealthy
Environment and How Does It Get under the Skin?”
Annual Review of Psychology 48:411–47.
Thoits, Peggy. 1995. “Stress, Coping, and Social Support
Processes: Where Are We? What Next?” Journal of
Health and Social Behavior 35:53–79.
———. 2006. “Personal Agency and the Stress Process.”
Journal of Health and Social Behavior 47:309–23.
S65
Thombs, Dennis L., Bette Jean Wolcott, and Lauren G.
Farkash. 1997. “Social Context, Perceived Norms
and Drinking Behavior in Young People.” Journal of
Substance Abuse 9:257–67.
Uchino, Bert N. 2004. Social Support and Physical Health:
Understanding the Health Consequences of Relationships. New Haven, CT: Yale University Press.
———. 2006. “Social Support and Health: A Review of
Physiological Processes Potentially Underlying Links
to Disease Outcomes.” Journal of Behavioral Medicine 29:377–87.
Umberson, Debra, Robert Crosnoe, and Corinne Reczek.
2010. “Social Relationships and Health Behaviors
across the Life Course” Annual Review of Sociology
36:139–57.
Umberson, Debra, Kristi Williams, Daniel A. Powers,
and Meichu Chen. 2005. “As Good as It Gets? A
Life Course Perspective on Marital Quality.” Social
Forces 84:493–511.
Umberson, Debra, Kristi Williams, Daniel A. Powers,
Hui Liu, and Belinda Needham. 2006. “You Make Me
Sick: Marital Quality and Health over the Life Course.”
Journal of Health and Social Behavior 47:1–16.
U.S. Census Bureau. 2009. “America’s Families and
Living Arrangements: 2009.” Washington, DC: U.S.
Government Printing Office.
U.S. Department of Health and Human Services. 2000.
“Healthy People 2010: Understanding and Improving
Health.” Retrieved August 23 2009 (http://www.healthy
people.gov/Document/tableofcontents.htm#under).
———. n.d. “The Healthy Marriage Initiative.” Retrieved
August 23, 2009 (http://www.acf.hhs.gov/healthy
marriage/about/mission.html#goals).
U.S. Department of Labor. 2009. “Fact Sheet #28: The
Family and Medical Leave Act of 1993.” Retrieved
August 23, 2009 (http://www.dol.gov/esa/whd/regs/
compliance/whdfs28.pdf).
Waite, Linda J. 1995. “Does Marriage Matter?” Demography 32:483–508.
Walen, Heather R. and Margie E. Lachman. 2000. “Social
Support and Strain from Partner, Family, and Friends:
Costs and Benefits for Men and Women in Adulthood.”
Journal of Social and Personal Relationships 17:5–30.
Wienke, Chris and Gretchen J. Hill. 2009. “Does the ‘Marriage Benefit’ Extend to Partners in Gay and Lesbian
Relationships?: Evidence From a Random Sample of
Sexually Active Adults.” Journal of Family Issues
30:259–89.
Willson, Andrea E. 2003. “Race and Women’s Income Trajectories: Employment, Marriage, and Income Security
over the Life Course.” Social Problems 50:87–110.
S66
World Health Organization. 2007. “Mental Health:
Strengthening Mental Health Promotion. Fact Sheet
No 220.” Retrieved August 23, 2009 (http://www.
who.int/mediacentre/factsheets/fs220/en/index.html).
Zhang, Zhenmei and Mark D. Hayward. 2006. “Gender,
the Marital Life Course, and Cardiovascular Disease
in Late Midlife.” Journal of Marriage and Family
68:639–57.
Bios
Debra Umberson is professor of sociology and a faculty
associate in the Population Research Center at the University of Texas at Austin. Her research focuses on
Journal of Health and Social Behavior 51(S)
relationships and health across the life course. Her current research, supported by the National Institute on
Aging, considers how different types of relationships
influence health behaviors over the life course.
Jennifer Karas Montez is a doctoral candidate in the
Department of Sociology and Population Research
Center at the University of Texas at Austin. Her
research concerns socioeconomic and gender disparities in health and mortality in later life, and examines
how social, behavioral, and biological mechanisms
interact across the entire life course to produce those
disparities.