NIH-PA Author Manuscript NIH-PA.docx

N
IH
-P
A
A
uthor M
anuscript
N
IH
-P
A
A
uthor M
anuscript
N
IH
-P
A
A
uthor M
anuscript

NIH Public Access
Author Manuscript
Psychosom Med. Author manuscript; available in PMC 2012
January 1.
Published in final edited form as:
Psychosom Med. 2011 January ; 73(1): 67–74.
doi:10.1097/PSY.0b013e3182002116.
Modeling Social Influences on Human Health
Kate Karelina, Ph.D.1 and A. Courtney DeVries, Ph.D.1,2
1 Department of Neuroscience, The Ohio State University,
Columbus, OH 43210 USA
2 Institute of Behavioral Medicine Research, The Ohio State
University, Columbus, OH 43210
USA
Abstract
Social interactions have long-term physiological, psychological
and behavioral consequences.
Social isolation is a well recognized but little understood risk
factor and prognostic marker of
disease, and can have profoundly detrimental effects on both
mental and physical well-being,
particularly during states of compromised health. In contrast,
the health benefits associated with
social support (both reduced risk and improved recovery) are
evident in a variety of illnesses and
injury states; however, the mechanisms by which social
interactions influence disease
pathogenesis remain largely unidentified. The substantial health
impact of the psychosocial
environment can occur independently of traditional disease risk
factors and is not accounted for
solely by peer-encouraged development of health behaviors.

Instead, social interactions are
capable of altering shared pathophysiological mechanisms of
multiple disease states in distinct
measurable ways. Converging evidence from animal models of
injury and disease recapitulates the
physiological benefits of affiliative social interactions and
establishes several endogenous
mechanisms (inflammatory signals, glucocorticoids and
oxytocin) by which social interactions
influence health outcomes. Taken together, both clinical and
animal research are undoubtedly
necessary in order to develop a complete mechanistic
understanding of social influences on health.
Keywords
Social interaction; health; experimental models; oxytocin
Social influences on health
Social interactions shape humans from early development
through senescence and have a
strong impact on many aspects of physiology and behavior.
Indeed, social interaction is
essential for proper cognitive, affective and behavioral
development (1). Among adults, the
social environment remains an important determinant of health
and well being; ample
evidence suggests that positive social support accelerates and
improves patient recovery
from cancer, cerebrovascular and cardiovascular disease (CVD),
atherosclerosis, and other
chronic diseases with an inflammatory component (2–5). This
has led to a substantial
interest in the capacity to which the social environment affects
physiological systems,
particularly during health challenges. The benefits of a positive
social environment are

particularly salient in chronic disease states, in which emotional
social support can be
perceived as being equally or more important than instrumental
and informational support
Correspondence should be addressed to: Kate Karelina, Ph.D.,
Department of Neuroscience, The Ohio State University, 750
Biomedical Research Tower, 460 W 12th Ave., Columbus, OH
43210, Fax: 614-292-3464, Phone: 614-688-4665, karelina.
[email protected]
Conflict of interest: All authors declare that they have no
conflicts of interest in publishing this manuscript
mailto:[email protected]
Karelina and DeVries Page 2
N
IH
-P
A
A
uthor M
anuscript
N
IH
-P
A
A

uthor M
anuscript
N
IH
-P
A
A
uthor M
anuscript
(6). In contrast, social isolation and loneliness can have
profoundly detrimental effects on
mental and physical health (7). While this observation is not
novel in the medical
community (8–10), it has only relatively recently begun to gain
momentum in both clinical
and animal research. The addition of evaluating patients’ social,
along with cognitive and
physical states, while not yet considered common practice, is
gaining acceptance in hospitals
and clinics worldwide (11–12); however, despite growing
evidence implicating the social
environment as a modifying factor in disease outcomes, little is
known regarding the
mechanisms through which psychosocial factors influence
disease pathogenesis. Converging
evidence from experimental research suggests that socially
isolated animals mount a
quantitatively and qualitatively different pathophysiological
response to disease and physical
trauma compared to socially housed animals. Moreover, the

benefits of social housing in
animal models are remarkably consistent with clinical findings,
and are evident in a diverse
set of disease and injury models. The aim of this review is to
further illustrate the need for
integrative clinical and experimental research that encompasses
a more complete
understanding of the qualitative and quantitative consequences
of social experiences on
disease physiology.
Chronic diseases such as cardiovascular and cerebrovascular
disease, diabetes, cancer, and
autoimmune disorders accounted for 70% of all deaths in the
United States in 2005 (13). A
substantial research effort has elucidated a number of risk
factors (e.g. smoking, alcohol
consumption, high blood pressure, cholesterol, etc) that are
common across most chronic
disease states, as well as identified causal relationships and
mechanisms by which these
factors influence disease onset and outcome. Interestingly, even
after statistically controlling
for these risk factors, there still exists substantial inter-
individual variability in susceptibility
to disease and recovery. This naturally occurring variability can
be accounted for in part by
an additional class of risk factors: psychological stress
(including social isolation or
perceived lack of social support), that is predictive of disease
outcome independently of
other traditional risk factors (14–15). Importantly, the impact of
the social environment is
not only evident during the course of disease, but also
influences the development of
potentially debilitating consequences such as chronic pain,

long-term physical disability, and
psychological distress such as anxiety/depression (16).
A prevailing hypothesis is that social support improves health
by promoting healthy
behaviors. Indeed, social support is associated with better
medical compliance, increased
physical exercise, improved nutrition and low-to-moderate
tobacco and alcohol consumption
(17). While it is not surprising that social and peer support
increases the likelihood of
engaging in health behaviors (whether because of the pressure
to conform to social norms or
a potential increase of tangible resources), statistically
controlling for these behavioral
changes indicates that the benefits of social support remain
substantial (15,18–19). The
major implication of these findings is the need to identify an
endogenous physiological
mechanism through which social behaviors influence
physiological parameters that have
important health implications.
To date, it has been difficult to definitively establish a causal
relationship between
psychosocial factors and disease among the clinical population,
however, a number of
successful longitudinal studies have identified social isolation
as a predictor of physiological
measures that are known risk-factors for disease, including
onset of CVD, even 20 years
following initial assessment (19–20). Indeed, childhood social
isolation is associated with a
greater number of CVD and stroke risk factors, including high
body mass index, high blood
pressure and cholesterol, and low oxygen consumption in

adulthood (19). These data
demonstrate that the relationship between social isolation and
disease is not strictly
correlational; rather, early experiences of social isolation can
predict the development of risk
factors well into adulthood.
January 1.
N
IH
-P
A
A
uthor M
anuscript
N
IH
-P
A
A
uthor M
anuscript
N

IH
-P
A
A
uthor M
anuscript
Two important questions emerge in light of these studies. (1)
What are the shared
pathophysiological features among the disease states that are
influenced by social
experiences, and (2) in what way are they modifiable by the
psychosocial environment?
Inflammatory processes represent one common mechanism of
disease that underlies the
multiple pathophysiologies described in this review. Indeed,
chronic inflammation
increasingly ranks as an important risk factor for stroke and
CVD, as well as many cancers,
Alzheimer’s disease and major depression disorder (21–25). For
example, patients
presenting with systemic inflammation, such as systemic lupus
erythematosus and
rheumatoid arthritis, have a 4–10 fold increased risk of
developing CVD (reviewed in 26).
This relationship is particularly evident in coronary and
cerebral ischemia, as inflammation
is relevant both as an indicator of an underlying cause (i.e.
atherosclerosis 27) and as an
important mediator (22), rather than just a biomarker of
ischemia. As such, chronic
inflammation has become a main focus for monitoring and

prevention of stroke and CVD. In
particular, systemic levels of acute phase proteins such as C-
reactive protein (CRP) and
proinflammatory cytokines (primarily interleukin 6; IL-6)
predict the risk and prognosis of
stroke and CVD (21,28). Importantly, circulating CRP and IL-6
levels are reduced in
patients with sufficient social support, thus presenting one
mechanism by which social
experiences could influence stroke and CVD (29–30); however,
the intermediate signal that
interprets and translates the social environment into an altered
inflammatory response
remains unknown.
One signal that may serve to transduce social experiences into
an altered physiology is
oxytocin (OT). OT is a neuropeptide that is released during
social interactions. Exogenous
OT administration has been shown to increase pro-social
behaviors in humans, including the
ability to interpret emotions of others (31), interpersonal
communication and social
approach behavior (32). Anxiolytic properties of OT are also
evident in clinical studies.
Indeed, a particularly robust finding in the clinical literature is
the relationship between
endogenous OT (which is elevated during lactation) and stress
hyporesponsiveness.
Lactating and nursing women have attenuated stress responses
(33) and lower blood
pressure (34) compared to non-lactating controls. Moreover,
this stress buffering effect of
social interaction can be mimicked with exogenous OT
administration (35). Taken together,
the role of OT in meditating social behaviors as well as stress

physiology makes it an
attractive potential endogenous signaling mechanism by which
social behaviors influence
health.
Overall, the precise mechanisms through which psychosocial
factors influence the
pathophysiological response to disease remain unknown because
(1) the relevant clinical
studies are inconsistent in the way they define social support
and often don’t distinguish
between the various types of social support (i.e. functional,
emotional, informational, etc)
and importantly (2) the manipulations necessary to establish a
causal link between social
support and health outcomes cannot be conducted ethically in
humans. On the other hand,
environmental factors such as social housing are easily
modifiable in laboratory animals and
produce reliable and quantifiable physiological effects, and as
such can be used to establish
causation as well as to allow extensive characterization of the
physiological mechanisms
underlying social influences on disease.
Animal models of social influences on disease outcome
As the impact of social housing conditions (i.e. single vs.
pair/group housing) on rodent
welfare gains attention, it is becoming increasingly evident that
for some species, social
isolation is a profound psychological stressor. Chronic
individual housing of rodents induces
symptoms of “isolation syndrome” including depressive-like
behavior (36–37), stress and
anxiety-like behaviors (38–39), as well as aggression (40). In
addition, the physiological

January 1.
N
IH
-P
A
A
uthor M
anuscript
N
IH
-P
A
A
uthor M
anuscript
N
IH
-P
A
A

uthor M
anuscript
consequences of social isolation include autonomic
dysregulation (41), altered metabolism,
heart rate, and core body temperature (42–44), and suppression
of adult neurogenesis (45).
Many of the negative consequences of social isolation are
ameliorated simply by
manipulating social housing conditions (46–48).
Existing animal models of disease, in particular rodent models
of ischemia, atherosclerosis,
neuropathic pain, and wound healing – disease states well
known to be influenced by social
factors in the clinical population – are all sensitive to social
manipulations. As such, a
growing body of experimental literature has provided evidence
that the behavioral and
physiological consequences of social isolation include increased
susceptibility to disease,
delayed wound healing, and a disruption of functional recovery
following trauma or injury.
Importantly, there is a high degree of agreement among the
clinical and emerging rodent
data on the positive effects of social interaction on the
pathogenesis of these disease states,
and the implications of these data are now being extended to
study the mechanisms by
which social factors influence disease pathophysiology (table
1).
Cerebral Ischemia
Social housing has been shown to significantly affect measures

of cerebral ischemia
outcome in a series of studies employing well-characterized
mouse models of focal (stroke)
and global (cardiac arrest) cerebral ischemia (49–51). To study
social influences on ischemia
outcome, mice are housed singly (socially isolated), or are
socially housed in same-sex
groups or paired with an ovariectomized female. Using
experimental stroke models:
unilateral middle cerebral artery occlusion (MCAO) and cardiac
arrest/cardiopulmonary
resuscitation (CA/CPR), our lab (49–52) has reported that the
social housing condition is a
strong determinant of the pathophysiological response and long-
term survival following
experimental stroke. Mice housed in pairs are 60% more likely
to survive 7 days following
the ischemic damage (51). Moreover, social housing reduces
cell death, brain water content
(edema) and infarct size relative to social isolation in both the
stroke and cardiac arrest
models of ischemia (49–53). In addition, experimental stroke in
rodents produces functional
deficits similar to those that occur following a clinical stroke,
namely a reduction of
coordinated limb use contralateral to the affected hemisphere
followed by a reduction of
overall mobility depending on stroke severity. Social housing
reduces post-ischemic
functional deficits and accelerates recovery of locomotor
activity relative to social isolation
(50,52).
Neuropathic Pain
Social interaction is further implicated as a modulating factor of
the development of

neuropathic pain, which occurs as a result of a dysfunction or
injury to the central nervous
system. Neuropathic pain is modeled in mice by inducing a
peripheral nerve injury in the
hind limb. Subsequent development of neuropathic pain is then
measured as paw withdrawal
in response to an innocuous stimulus (small filaments applied to
the lateral side of the paw).
Following nerve injury, socially housed mice exhibit a
significant increase in paw
withdrawal thresholds relative to socially isolated mice,
indicating a reduction of
neuropathic pain response (54). Importantly, depressive-like
behavior, a common
consequence of chronic pain, is also reduced in the socially
housed animals (55). Taken
together with the behavioral data from the ischemia studies,
these data indicate that the
influence of social interaction extends beyond the course of
central nervous system injury
and continues to affect the development of arguably the most
debilitating consequences of
CNS damage including functional deficits, chronic pain and
depression.
January 1.
N
IH
-P

A
A
uthor M
anuscript
N
IH
-P
A
A
uthor M
anuscript
N
IH
-P
A
A
uthor M
anuscript
Wound Healing
Additional extensive insight into the neurobiological correlates
of social behavior and
pathophysiology is possible through comparison of closely
related rodent species that
exhibit different social systems. For example, the socially
monogamous Peromyscus

californicus mice exhibit a strong tendency toward forming
long-lasting bonds with a mate,
while the closely related Peromyscus leucopus exhibit behaviors
consistent with a
polygynous social system and tend to be solitary in nature (56–
57). These animals provide
an excellent natural model for studying the impact of social
structure and the accompanying
physiological and behavioral correlates on health outcomes. As
such, Peromyscus
californicus and leucopus were recently used to establish the
finding that the health-
promoting benefits of social interaction are further evident in
non-CNS injuries with an
inflammatory component (58). In order to assess social
influences on the rate of wound
healing, small (3.5 mm diameter) wounds are created using a
punch biopsy tool on the
dorsum. The wounds are then photographed and measured daily
for determination of healing
rate. Interestingly, social housing only facilitated wound
healing among the monogamous
Peromyscus californicus, whereas the healing rate among the
polygynous Peromyscus
leucopus remained similar between socially housed and isolated
mice (58). Likewise, social
interaction among Siberian hamsters, which form social bonds
with familiar conspecifics,
improves wound healing (59). Taken together, these data
suggest that the ability to form
social bonds significantly affects the extent to which social
interactions influence health
outcomes.
Mechanisms of social influences on health
Inflammation

The immune system actively regulates and sustains a state of
homeostasis under normal
physiological conditions. However, following trauma, the
organism enters a pro-
inflammatory and pro-thrombotic state (60). This state of
systemic inflammation involves
the activation of macrophages and T cells, the release of soluble
factors such as cytokines
and chemokines, leukocyte extravasation and the upregulation
of adhesion molecules,
among a multitude of other immune effectors (60). Whereas
acutely this is a coordinated
adaptive response intended to restore homeostasis and remove
pathogens and dead/dying
cells, there are serious long-term consequences associated with
tipping the balance toward a
chronic proinflammatory state.
Thus, social modulation of inflammatory processes represents a
plausible link between
social interactions and health. Indeed, a particularly robust
finding in the clinical literature is
an inverse relationship between social support and circulating
CRP and IL-6 in otherwise
healthy individuals (29–30). It follows that increased stroke and
CVD morbidity and
mortality rates in socially isolated individuals may be a
reflection of elevated chronic low-
grade inflammation (61).
Animal models provide evidence that the social housing
influence on disease is
accompanied by altered systemic and neuroinflammatory
responses. In parallel to clinical
findings, circulating IL-6 is reduced in socially housed mice

following experimental stroke
(51) or cardiac arrest (53). Moreover, the circulating
concentration of CRP, which is induced
by IL-6, is similarly reduced following MCAO in socially
housed animals and the reduction
of both markers is accompanied by smaller infarcts and
improved functional recovery
(50,52). Interestingly, while circulating IL-6 is reduced in
socially housed mice, the same
group of animals exhibits increased brain IL-6 protein
concentrations (52). These data are
consistent with findings that increased central IL-6 is
neuroprotective in stroke models (62–
63). Indeed, central pretreatment of socially housed animals
with a neutralizing antibody to
IL-6 eliminates the neuroprotection conferred by social housing
and increases infarct size to
January 1.
N
IH
-P
A
A
uthor M
anuscript
N

IH
-P
A
A
uthor M
anuscript
N
IH
-P
A
A
uthor M
anuscript
the same size observed in socially isolated animals (51). As
circulating IL-6 concentrations
remain among the most widely used early predictors of patients’
stroke severity and
outcome, further insight into the role of central IL-6 signaling
is essential for the
development of therapeutic treatments involving modulation of
inflammatory processes in
stroke and cardiac arrest patients.
Proinflammatory cytokines such as interleukin 1 beta (IL-1β),
tumor necrosis factor alpha
(TNF-α), and IL-6 are critical regulators of cerebral ischemia
outcome. These cytokines are

produced by microglia, which act the brain’s principal local
inflammatory response to brain
trauma and are thus activated within minutes of the onset of
cerebral ischemia (reviewed in
64). Although microglial activation (microgliosis) is a key
process to the removal of dead
cells and regeneration following brain injury, elevated and
prolonged microgliosis tips the
balance toward unchecked neuroinflammation which contributes
to further
neurodegeneration and inhibits recovery. While cerebral
ischemia increases brain microglial
expression, socially housed animals exhibit modest levels of
microgliosis relative to social
isolation (49,51,53). Importantly, the reduction of microgliosis
is accompanied by reduced
neuronal damage (49,53). These data provide compelling
evidence for a role for
neuroinflammation as a mechanism by which social interaction
influences health.
Stress/Glucocorticoids
A substantial research effort has focused on the role of stress
responses as a mediating
variable in the relationship between social interaction and
disease. Although basal
glucocorticoid concentrations are typically similar between
socially housed and isolated
rodents (36,65–66), stress reactivity is attenuated in socially
housed animals (39, 59, 67; but
see Sanchez et al., 1998). Acute stress responses may be
adaptive in many circumstances,
but chronic stress is well-documented as being detrimental to
health. In fact, chronic stress
(such as chronic social isolation) leads to altered immune
function (68–69), hypertension,

myopathy (70–71), and a range of psychological disorders (72–
73) to a similar extent in
humans and animals. Further, social stress, such as social
hierarchy disruption and intruder
aggression results in glucocorticoid insensitivity and tips the
balance toward a pro-
inflammatory state in mice, (74) leaving the organism more
susceptible to disease.
One approach to studying the deleterious effects of stress is to
identify the psychosocial
factors that increase an individual’s susceptibility to both
psychological and physical
stressors. The buffering effect of affiliative social interaction
against stress represents one
potential mechanism of social influences on health. For
example, acute restraint elevates
circulating cortisol concentrations and prolongs wound healing
in socially isolated Siberian
hamsters (P. sungorus), however, pair-housing eliminates the
stress-induced activation of
the HPA axis and ameliorates the effect of stress on wound
healing (58–59). In fact, restraint
stress has no impact on wound healing latency among pair
housed hamsters. Moreover, the
effect of restraint stress on wound healing in socially isolated
hamsters is likely mediated by
endogenous cortisol secretion, because adrenalectomized
hamsters heal more rapidly (59).
In addition, social disruption (an experimental model in which
established social hierarchies
are disrupted via introduction of an aggressive intruder male)
substantially elevates
circulating glucocorticoid concentrations and increases
susceptibility to viral infection (75)
and a variety of inflammatory diseases including pulmonary

inflammation (76), asthma (77)
and influenza (78). Taken together, the consequences of over-
stimulation of the HPA axis
and hypersecretion of glucocorticoids may be ameliorated by
stable and affiliative social
interactions.
January 1.
N
IH
-P
A
A
uthor M
anuscript
N
IH
-P
A
A
uthor M
anuscript
N

IH
-P
A
A
uthor M
anuscript
Oxytocin
The health-promoting effects of social interaction are marked
by substantial changes in
neuroimmune and neuroendocrine markers. To date, a specific
underlying mechanism for
these effects remains unclear, however converging evidence
from a rapidly growing
collection of studies implicates a mediating role for OT. OT is
produced in high
concentrations in the supraoptic (SON) and paraventricular
(PVN) nuclei of the
hypothalamus, which in turn project to the posterior pituitary,
where OT is released to its
central and peripheral targets (79). The biological activity of
OT is mediated by the oxytocin
receptor (OTR) and to a lesser extent, three vasopressin
receptors. The OTR, a G protein-
coupled receptor, is abundantly present in several regions, both
in the brain and periphery;
however, several structures (particularly nuclei of the
hypothalamus and amygdala) are
particularly important for the onset and maintenance of the
effects of OT on social behavior.
Numerous studies have implicated both causal and regulatory
roles for OT in the context of

animal social behavior. Indeed, central release of OT regulates a
wide range of social
behaviors, including pair-bonding, mother-infant bonding,
social recognition, and
aggression (80–82). Central infusion of OT facilitates the onset
of social behaviors (83–85),
while an OT receptor antagonist (OTA) nearly completely
eliminates them (86). OT also
suppresses the hypothalamic-pituitary-adrenal (HPA) axis in
several species (reviewed in
68), which further facilitates aspects of social behavior such as
social recognition and
formation of pair-bonds (reviewed in 87).
A substantial gap in knowledge about the precise physiological
mechanisms of the health-
promoting effects of social interactions has begun to close as a
result of the increasing use of
animal models for a variety of disease/injury states. In addition
to the ability to precisely
control for social conditions, the use of animal models allows
for direct manipulation of the
endogenous physiological signals (i.e. OT) believed to be
responsible for social influences
on disease. Pharmacological manipulations of central OT
signaling have recently identified
that the positive impact of social housing on health outcomes
can be eliminated by blocking
endogenous OT receptors. Treatment with an oxytocin receptor
antagonist (OTA) prevents
the ability of social interaction to reduce cerebral infarct size
and inflammation as well as to
increase anti-oxidant defenses following an ischemic event.
Moreover, central
administration of OT to socially isolated rodents mimics the
beneficial effects of social

housing on cerebral ischemia pathophysiology, thus establishing
a mechanistic link for OT
as a mediating signal by which social interactions confer health-
promoting effects (88). Pair-
housing also facilitates wound healing in hamsters, an effect
that requires central OT
signaling (59). Indeed, treatment of pair-housed hamsters with
an OTA eliminated the
protection of social housing, increased wound size and delayed
healing. However, the
benefits of pair-housing were recapitulated in socially isolated
hamsters by central treatment
with an OT agonist, resulting in similar wound size and healing
time as in pair-housed
hamsters (59). In addition, OT is involved in pain modulation,
and has been shown to have
potent anti-nociceptive effects (89). Indeed, reduced allodynia
(pain in response to a
stimulus that does not typically evoke pain) in pair-housed mice
is also mediated by central
OT signaling (Norman et al., 2010). A critical finding in these
studies is that social
facilitation of pain and healing is blocked by central OTA,
indicating a receptor-dependent
mechanism for this environment-physiology interaction.
The mechanisms through which OT influences disease
progression and outcome appear to
involve anti-inflammatory and antioxidant properties of OT. OT
administration alleviates
tissue damage in a variety of animal models of injury including
renal (90) and hepatic (91)
ischemia/reperfusion injury, as well as sepsis-induced multiple
organ damage (92), skin
injury (93) and colitis (94). The protective actions of OT in
these models are primarily anti-

inflammatory, resulting in decreased levels of TNFα and IL-6
(90–91) as well as decreased
neutrophil infiltration to the site of injury (90–92,94). In
addition, OT has recently been
January 1.
N
IH
-P
A
A
uthor M
anuscript
N
IH
-P
A
A
uthor M
anuscript
N
IH

-P
A
A
uthor M
anuscript
shown to have antioxidant properties (95). Subcutaneous OT
treatment …
MEMORANDUMto:
Professor Bakerfrom:
Yirou Xiaodate:
May 24, 2020subject:
STRATEGY FOR REDUCING LONG WORKING HOURS IN
BANKS
Description
Of the many working problems and challenges that exist in the
workplace for bankers, long working hours is the most
consistent across banks. The upcoming recommendation report
addressed in this memo hopes to provide working
recommendations to reduce the long working hours for bankers
to the number recommended by the International Labor
Organization (ILO).
Objective
The Massachusetts government Division of Banks (DOB) should
look into the suggestions highlighted in the recommendation
report to propose formidable solutions to the long working
hours that bankers in Massachusetts are subjected to, and that
contravene guidelines by the ILO.
Information

It is virtually impossible to have a workplace that is devoid of
challenges. These problems range from harassment, high
turnover rates, understaffing, poor compensation, and conflicts
amongst others. It is, therefore, instrumental for the authorities
responsible and all the stakeholders involved to take
intervention measures to ensure these problems are addressed
promptly and appropriately in a way that satisfies the aggrieved
parties.
Fundamentally, the banking industry is one that has
continuously faced the problem of understaffing and the
consequent long working hours for tellers and other bank
employees. A 2019 report by Makkar and Basu indicated that
accountants spend an average of 70 percent of their time at
work translating to about ten hours a day. These figures are
worse for bank tellers who also fall under accountants. The U.S.
labor department and the ILO recommend for workers to spend
only six to eight hours a day at work. The Bureau of Labor
Statistics (2019) revealed that the ratio of tellers to the
customers they serve is about a thousand people to one teller.
On average, most banks in Massachusetts serve about 5000
customers a day, which results in severe implications for the
tellers.
Firstly, overworked bankers are likely to suffer health
challenges emanating from long working hours most of which
are spent sitting. Some health implications associated with
sitting for prolonged periods include obesity and the inherent
heart problems, back problems, increased blood pressure,
abnormal cholesterol levels, and low metabolism among other.
Socially, these long working hours are associated with poor
socializing skills, anxiety, loneliness, and sometimes
depression. In addition to these health implications, the long
working hours have been associated with accounting errors due
to fatigue. Moreover, the turnover rate for banks is significantly
high compared to other service industries.
For these reasons, the upcoming recommendation report
advocates for several intervention measures to reduce the long

working hours from ten to the ILO recommended six to eight
hours. The first suggestion for banks is to employ more
accountants to reduce overworking. More employees provide for
a better employee-to-customer ratio. Consequently, the working
hours decrease significantly and so does fatigue and the turn-
over rate. The other suggestion is for the banks to adapt a
system where accountants work in shifts allowing them to have
enough rest. These recommendations make practical sense and
will help the affected banks to comply with ILO regulations.
Audience
The target audience for the report is the Massachusetts
government Division of Banks (DOB) which regulates banks in
Massachusetts. In its supervisory role, the DOB has the mandate
to ensure every financial institution in its jurisdiction provides
favorable working conditions to its employees. Implementing
these recommendations might encounter some resistance from
banks who might oppose the idea of adding more employees as
a financial liability. However, the DOB has the mandate to
implement international, local, and federal standards most of
which have not been fulfilled by banks in Massachusetts.
Conclusion
This memo presides a recommendation report that hopes to
reduce the persistent problem of long working hours for
accountants in Massachusetts. The report will be addressed to
the Massachusetts government Division of Banks (DOB).
Among the suggestions offered include hiring more bankers to
reduce the teller to customer ratio and introducing shifts as an
alternative.
5
2
Current Directions in Psychological

Science
21(1) 60 –65
© The Author(s) 2012
Reprints and permission:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0963721411429456
http://cdps.sagepub.com
Illness perceptions are frameworks or working models that
patients construct to make sense of their symptoms and medi-
cal conditions. A patient’s cognitive representation of his or
her illness then guides behavior directed at managing the con-
dition (Leventhal, Nerenz, & Steele, 1984). An illness percep-
tion comprises a number of interrelated beliefs about an illness
and what it means for the patient’s life. The major components
include how the illness was caused, how long it will last, what
the consequences of the illness are for the patient’s life and
family, the symptoms that are part of the illness, and how the
condition is controlled or cured. Research has demonstrated
that patients’ perceptions of their illness along these dimen-
sions vary widely, even between patients with similar illnesses
or injuries.
Patients build models of their illness based on previous per-
sonal or family experiences with their disease or information
they may have received from medical staff or the general
media. These illness models may be specific to the individual
and may differ considerably from those of the clinicians giv-
ing treatment. Despite this, there is now a large body of litera-
ture showing that patients’ illness perceptions are related to a
range of important health outcomes, including functioning,
health care utilization, adherence, and mortality. The growing
interest in patients’ perceptions of their illness is reflected in
the rapidly increasing number of publications in this area over
the past 15 years. While much of the published research has
focused on the role of illness perceptions in explaining coping

and outcomes in patients with a wide range of health condi-
tions (see Hagger & Orbell, 2003), here we will describe some
more recent developments. In particular, we provide an over-
view of research on new approaches for investigating and
modifying illness perceptions in order to improve health out-
comes and treatment adherence.
Recent studies of how patients’ perceptions develop follow-
ing medical testing and diagnosis have provided some insights
into how illness perceptions change in response to diagnostic
information. A study of patients undergoing coronary computer
tomographic (CT) angiography for diagnosis of heart disease
shows how patients’ perceptions change rapidly to incorporate
diagnostic information that they either have or do not have heart
disease (Devcich, Ellis, Broadbent, Gamble, & Petrie, 2011).
The results of this study, illustrated in Figure 1, show that
patients generally prepare themselves for an unfavorable diag-
nosis, indicated by higher levels of concern and perceived con-
sequences in both disease and nondisease groups prior to
testing.
When the outcome is favorable, with no heart disease apparent
from the test, illness concern and perceived consequences drop
immediately in the nondisease group. Notice also how members
of the nondisease group tend to attribute their good result to
their ability to control the disease, indicated by higher ratings
of
personal control over the disease following the test—in a sense,
they claim credit for the favorable outcome. The patients
Corresponding Author:
Keith J. Petrie, Psychological Medicine, Faculty of Medical and
Health
Sciences, University of Auckland, Private Bag 92019,
Auckland, New
Zealand

E-mail: [email protected]
Patients’ Perceptions of Their Illness:
The Dynamo of Volition in Health Care
Keith J. Petrie1 and John Weinman2
1Department of Psychological Medicine, University of
Auckland, New Zealand and
2Psychology Department, Institute of Psychiatry, King’s
College, United Kingdom
Abstract
A patient constructs a cognitive representation of his or her
illness that guides behavior aimed at managing that illness.
Patients’
models of their illness share a common structure made up of
beliefs about the cause of an illness, the symptoms that are part
of the condition, the consequences of the illness for the
patient’s life, how the illness is controlled or cured, and how
long the
illness will last. Illness perceptions can be measured using
questionnaires and also assessed in patients’ drawings, which
readily
show how an illness is visualized. Illness perceptions change
rapidly in response to diagnostic results and have been
associated
with emotional distress, recovery, and disability, as well as with
treatment-related behavior such as adherence. Interventions
based around changing inaccurate or unhelpful perceptions of
illness are an important emerging area of health psychology.
Keywords
illness perceptions, outcome, assessment, intervention,
adherence

http://crossmark.crossref.org/dialog/?doi=10.1177%2F09637214
11429456&domain=pdf&date_stamp=2012-01-31
Patients’ Perceptions of Their Illness 61
diagnosed with heart disease have a higher belief that their
future treatment will control the disease.
Assessing Illness Perceptions
Early work on assessing patients’ illness perceptions relied on
open-ended questions from structured interviews. More reli-
able measurement was possible with the development of the
Illness Perceptions Questionnaire (IPQ; Weinman, Petrie,
Moss-Morris, & Horne, 1996). The IPQ provided assessment
of five components of illness representations: identity, time-
line, consequences, cause, and control/cure. A later revised
version of the scale (IPQ-R) added more items; split the
control dimension into personal and treatment control; and
added a cyclical timeline, an overall comprehension of illness
factor, and an emotional-representation-of-illness scale (Moss-
Morris et al., 2002). More recently, a shorter 9-item scale has
been developed. The Brief Illness Perception Questionnaire
(BIPQ) can be administered quickly and has acceptable psy-
chometric properties (Broadbent, Petrie, Main, & Weinman,
2006). The BIPQ is suited to assessment in clinical situations
in which time constraints preclude patients completing the full
80-item IPQ-R.
A recent development in the assessment of illness percep-
tions has been the use of drawings to uncover how patients
think about their illness. The advantage of a drawing is that it
provides the opportunity to pick up a patient’s idiosyncratic
beliefs or misconceptions about the illness that may play a role
in determining future outcomes (Broadbent, Petrie, Ellis,
Ying, & Gamble, 2004). For instance, a patient who has had a

recent heart attack may show in a drawing that his or her heart
has been so severely damaged that any activity and strain on
the heart may bring on another heart attack. Some characteris-
tics of patients’ drawings, such as greater areas of heart dam-
age drawn and larger drawings of the heart over time, have
been shown to be associated with important outcomes such as
a slower return to work following a heart attack (Broadbent,
Ellis, Gamble, & Petrie, 2006) and functional status in patients
with heart failure (Reynolds, Broadbent, Ellis, Gamble, &
Petrie, 2007).
A drawing also offers the opportunity to see how the patient
views the effect of treatment on his or her condition. Figure 2
shows the drawings of two patients before and after undergo-
ing heart surgery. The first patient, whose drawings are shown
on the top panels, underwent heart bypass surgery, and the
drawings illustrate how the patient’s perceptions of his heart as
clogged and diseased change over a relatively short period of
time. The drawings of the second patient, who had heart valve
surgery, also show improved perceptions of the health of his
heart, as well as a focus on the rhythm of the heart, which is an
important indicator of recovery for this patient.
Illness Perceptions and Outcomes
There is a rapidly growing literature on how patients’ illness
perceptions relate to later outcomes from illness. The outcomes
researchers have examined fall into four major groups: emo-
tional distress, recovery and disability, survival, and treatment-
related behavior such as adherence. The central question in
this research is this: How does the way a patient views his or
her illness or injury affect how the patient deals with the chal-
lenges of the illness and eventual outcome? This research is
important because it helps inform psychological interventions
and can also provide a basis for identifying patients at

Fig.1. Illness perceptions (concern about illness, perception of
illness consequences, perception
of personal control, and belief in ability of treatment to control
the disease) before and after CT
angiography test results for patients without (left) and with
(right) heart disease (differences are based
on before–after paired sample t tests; adapted from Devcich et
al., 2011).
62 Petrie, Weinman
an early stage who are at risk of not coping well with their
illness.
A number of studies have looked at how illness perceptions
predict depression and distress in various illnesses. A meta-
analysis of illness perceptions and outcomes found consistent
relationships between psychological distress as an outcome
and higher perceived consequences, low control/cure beliefs,
and longer timeline perceptions (Hagger & Orbell, 2003). In a
more recent study, Dickens et al. (2008) measured illness per-
ceptions of 269 first-time heart attack patients and examined
their levels of depression over the following year. The
researchers found that a patient who developed depression fol-
lowing his heart attack was more likely to believe at baseline
that his heart condition would last a long time and was unlikely
to be cured.
A number of other studies have now shown that when
patients hold negative perceptions about their illness, these
beliefs are associated with greater likelihood of future disabil-
ity and slower recovery (e.g., Galli, Ettlin, Palla, Ehlert, &
Gaab, 2010; Kaptein et al., 2010). While it seems possible that
negative perceptions are indicative of poorer prognosis, stud-

ies have generally found that illness perceptions not only fail
to relate closely to objective measures of disease severity but
also are often better predictors of outcome. For example, ill-
ness perceptions have even been found to predict mortality in
a recent study of patients with renal failure. After controlling
for both clinical factors and depression, Chilcot, Wellsted, and
Farrington (2011) found that perceptions of the ability of treat-
ment to control the disease were a significant predictor of all-
cause mortality. One possible explanation of this is that illness
perceptions can differentially activate illness-coping proce-
dures such as treatment adherence, which can strongly influ-
ence health outcome.
Adherence problems often arise because there is a poor fit
between the patient’s model of the illness and the nature of the
treatment recommendation, which means that the treatment
does not really make sense to the patient. For example, heart
attack patients who believe that their illness had been caused
by their own lifestyle have been found to be more adherent to
lifestyle-change advice, whereas those who identified stress or
genes as key causes were much less inclined to make these
changes (Weinman, Petrie, Sharpe, & Walker, 2000). Timeline
beliefs can also influence treatment adherence in a very direct
way; Halm, Mora, and Leventhal (2006) have shown that
patients who perceive their asthma as only present when they
experience symptoms are much less likely to take daily pre-
venter medication than other patients who perceive their
asthma as a more permanent condition.
Changing Perceptions to Improve
Health Outcomes
Given the consistent findings on the relation between illness
perceptions and outcome and the emerging evidence from
longitudinal studies, it is not surprising that there is now con-
siderable interest in developing interventions to modify dys-

functional illness beliefs in order to improve patient outcomes.
Since direct links have been found between illness perceptions
and adherence, illness-perception-based interventions have
also been developed in this area.
Fig 2. Two heart surgery patients’ drawings of their hearts
before and immediately after surgery and 3 months later.
These interventions aim to modify a patient’s perception
of his or her condition to increase the goodness-of-fit between
the patient’s model of illness and his or her treatment.
An example of this approach is provided by Karamanidou,
Weinman, and Horne (2008), who improved adherence to
phosphate-binding medication in patients with end-stage renal
disease by providing a concrete illustration of how the medica-
tion works, thereby strengthening the patients’ understanding
of the need to take the medicine regularly. The intervention
included a simple demonstration using a plastic container to
represent the patient’s stomach; phosphate solution was poured
in to represent high-phosphate food. Phosphate-binding medi-
cation was then introduced, and patients were asked to describe
what they saw as they observed the medication binding and
solidifying—thus providing a simple concrete image of the
mode of action of the medication. Another recent example
of this approach to improve treatment adherence is provided
by Petrie, Perry, Broadbent, and Weinman (2011), who devel-
oped a text-messaging intervention to increase adherence to
preventer-medication treatment in young adults with asthma.
After an initial assessment of his or her illness and treatment
beliefs, each patient was sent targeted text messages aimed at
modifying unhelpful beliefs over the next few weeks. This
resulted in significant changes in beliefs and in treatment

adherence, which persisted for 6 months after the text mes-
sages had stopped.
Illness-perception interventions have also been successfully
developed for heart attack patients, which have not only suc-
cessfully changed patients’ beliefs about their heart attacks but
also improved recovery and return to work (e.g. Broadbent,
Ellis, Thomas, Gamble, & Petrie, 2009). The potential for
illness-perception-based interventions with patients who have
suffered heart attacks is considerable, partly because there is
an extensive literature linking illness perceptions and outcome
in this patient group and partly because it is possible to inter-
vene at a very early stage prior to hospital discharge. However,
successful illness-perception-based interventions have also
been developed for other patient groups, including those with
type 2 diabetes (Keogh et al., 2011) and psoriasis (Fortune,
Richards, Griffiths, & Main, 2004). All of these studies have
demonstrated that fairly brief, straightforward psychoeduca-
tional interventions can be used for identifying and modifying
negative illness beliefs, which in turn can result in improve-
ments in a range of health-related outcomes. Given that adjust-
ment to a chronic illness or recovery from an acute illness
often depends upon effective self-management behaviors, it is
clear that targeting patients’ maladaptive illness perceptions
should be integrated into routine care.
Future Research
Examining patients’ perceptions of their illness opens up a
new approach to examine and intervene in a number of prob-
lematic areas of health care. Here we identify what we believe
are a few of the many potential applications of examining
illness perceptions in the health area and also some of the
issues that require attention in order to fulfill the promise of
this approach.

One of the most obvious applications of examining illness
perceptions is the identification of patients who are at risk
of coping poorly with the demands of their illness. There is
now a solid evidence base available from meta-analyses (e.g.
Hagger & Orbell, 2003) and individual studies with major ill-
nesses enabling us to identify a profile of illness perceptions
that puts patients at greater risk of poor outcomes. These pro-
files differ somewhat by illness but tend to center around neg-
ative views of the consequences and control of the illness that
may lead to greater emotional distress and a lower engagement
with treatment. Identification of at-risk patients offers the
potential to intervene with the goal of improving illness adjust-
ment, treatment, and outcome.
Poor adherence to treatment regimens and recommenda-
tions is a problem that besets most areas of health care. As
previously discussed, the fit between the patient’s view of his
or her illness and its prescribed treatment is a key area for
interventions designed to improve adherence. Often the diffi-
culty here is to develop effective methods that provide patients
with a more complete understanding of how and why their
treatments match with their illness. This can require either a
change in a patient’s understanding of the illness or of the rec-
ommended treatment or both. The development of successful
interventions in this area can have powerful effects on a
patient’s understanding of his or her treatment and subsequent
adherence to the treatment (Karamanidou et al., 2008).
Another important area of application of illness-perception
research is in the diagnostic area and improving patients’
responses to feedback from medical testing when investiga-
tions prove negative. Despite having normal results, many
patients remain concerned about their conditions and continue
to inappropriately seek medical investigations and treatment
for their symptoms. Recent research has demonstrated that
changing patients’ ideas about the meaning of diagnostic-test

results for their health prior to undergoing testing can reduce
concern about symptoms, improve reassurance, and reduce
future symptom reports (Petrie et al., 2007). As illness percep-
tions are very sensitive to information given in clinical inter-
views following diagnostic testing (Devcich et al., 2011), this
area has considerable potential for interventions to affect sub-
sequent health care costs.
The modifiability and sensitivity to change of illness per-
ceptions raises important questions about their origin and
development over time and, apart from the myocardial infarc-
tion studies mentioned previously, few studies have tracked
the early progression of illness perceptions in newly diagnosed
patients, particularly those with chronic conditions. Such stud-
ies would yield valuable information not only about the sorts
of factors influencing the formation of patients’ models but
also about whether there are critical periods during which
interventions may be particularly effective. On the more theo-
retical side, there is also scope for both experimental and
64 Petrie, Weinman
quasi-experimental research to increase our understanding of
the ways component illness perceptions link with each other to
form an organized cluster of beliefs about the illness and the
ways these beliefs relate to emotional processes and to illness-
related coping procedures. Recently, Leventhal, Leventhal,
and Breland (2011) have attempted to locate the processes
involved in illness cognition within a broader cognitive sci-
ence and neuroscience framework and outline the potential
that this approach can offer for further theory development
and practical applications. Their analysis points to the huge
potential for future interdisciplinary research aimed at uncov-
ering the nature of the many complex processes involved in

illness cognition and health outcome.
The ability to identify how patients view their illness offers
the potential to examine how medical staff and patients differ
in their views of the same illness. Given the large number of
potential areas of misunderstanding and miscommunication
that exist in health care, it is perhaps surprising that more work
has not been undertaken in this area. Surveys that have asked
patients to identify the location of body organs like the heart,
kidney, and lungs have found that a large proportion of patients
cannot correctly identify the location of these major organs
(Weinman, Yusuf, Berks, Rayner, & Petrie, 2009). Given the
rudimentary medical knowledge that many patients use to
construct models of their illness, there is considerable poten-
tial to document the areas of major differences between
patients and doctors in their views of an illness or injury and
the opportunities for each to “talk past each other” when dis-
cussing future care. Any health care consultation offers the
opportunity to simply ask the patient what he or she thinks is
wrong, what he or she believes caused the condition, and what
he or she thinks about the proposed treatment.
While health care interventions based around changing
patients’ illness perceptions have considerable promise, there
is much work to do before that promise is realized. More study
is needed on the types and timing of psychological interven-
tions that are effective in changing patients’ perceptions of
their illness and on who is best to deliver such interventions.
Given the pressures to lower health care costs, it is important
that the interventions that are developed not only improve
health care outcomes but also are scalable so they reach across
larger populations of patients.
Recommended Reading
Hagger, M. S., & Orbell, S. (2003). (See References). A meta-

analysis of published studies examining illness perceptions,
cop-
ing variables, and outcomes.
Leventhal, H., Meyer, D., & Nerenz, D. R. (1980). The common
sense representation of illness danger. In S. Rachman. (Ed.),
Con-
tributions to medical psychology (Vol. 2, pp. 17–30.). New
York,
NY: Pergamon Press. An important early theoretical paper
outlin-
ing the common-sense self-regulatory model.
Petrie, K. J., & Weinman, J. (Eds.). (1997). Perceptions of
health
and illness: Current research and applications. London,
England:
Harwood Academic. An edited book covering the theory and
measurement of illness perceptions as well as their application
to
prevention and illness management.
Skelton, J. A., Croyle, R. T. (Eds.). (1991). Mental
representation
in health and illness. New York, NY: Springer. An important
early book with interesting applications of the illness-
perception
approach and some fascinating historical and cultural examples.
Declaration of Conflicting Interests
The authors declared that they had no conflicts of interest with
respect to their authorship or the publication of this article.
References

Broadbent, E., Ellis, C. J., Gamble, G., & Petrie, K. J. (2006).
Changes
in patient drawings of the heart identify slow recovery
following
myocardial infarction. Psychosomatic Medicine, 68, 910–913.
Broadbent, E., Ellis, C. J., Thomas, J., Gamble, G., & Petrie, K.
J.
(2009). Further development of an illness perception based
inter-
vention for myocardial infarction patients: A randomised con-
trolled trial. Journal of Psychosomatic Research, 67, 17–23.
Broadbent, E., Petrie, K. J., Ellis, C. J., Ying, J., & Gamble, G.
(2004).
A picture of health—Myocardial infarction patients’ drawings
of
their hearts and subsequent disability: A longitudinal study.
Jour-
nal of Psychosomatic Research, 57, 583–587.
Broadbent, E., Petrie, K. J., Main, J., & Weinman, J. (2006).
The
Brief Illness Perception Questionnaire (BIPQ). Journal of Psy-
chosomatic Research, 60, 631–637.
Chilcot, J., Wellsted, D., & Farrington, K. (2011). Illness
perceptions
predict survival in haemodialysis patients. American Journal of
Nephrology, 33, 358–363.
Devcich, D. A., Ellis, C. J., Broadbent, E., Gamble, G., &
Petrie,
K. J. (2011). The psychological impact of test results following
diagnostic coronary CT angiography. Health Psychology. DOI:

10.1037/a0026502.
Dickens, C., McGowan, L., Percival, C., Tomenson, B., Cotter,
L.,
Heagerty, A., & Creed, F. (2008). Negative illness perceptions
are associated with new-onset depression following myocardial
infarction. General Hospital Psychiatry, 30, 414–420.
Fortune, D. G., Richards, H. L., Griffiths, C., & Main, C.
(2004).
Targeting cognitive behaviour therapy to patients’ implicit
model
of psoriasis: Results from a patient preference controlled trial.
British Journal of Clinical Psychology, 43, 65–82.
Galli, U., Ettlin, D. A., Palla, S., Ehlert, U., & Gaab, J. (2010).
Do
illness perceptions predict pain-related disability and mood
in chronic orofacial pain patients? A 6-month follow-up study.
European Journal of Pain, 14, 550–558.
Hagger, M. S., & Orbell, S. (2003). A meta-analytic review of
the
common-sense model of illness representations. Psychology &
Health, 18, 141–184.
Halm, E. A., Mora, P., & Leventhal, H. (2006). No symptoms,
no
asthma. The acute episodic disease belief is associated with
poor
self-management among inner-city adults with persistent
asthma.
Chest, 129, 573–580.
Kaptein, A. A., Bijsterbosch, J., Scharloo, M., Hampson, S. E.,
Kroon,

H. M., & Kloppenburg, M. (2010). Using the common sense
model of illness perceptions to examine osteoarthritis change: A
6-year longitudinal study. Health Psychology, 29, 56–64.
Karamanidou, C., Weinman, J., & Horne, R. (2008). Improving
hae-
modialysis patients’ understanding of phosphate-binding
medica-
tion: A pilot study of a psychoeducational intervention designed
to change patients’ perceptions of the problem and treatment.
British Journal of Health Psychology, 13, 205–214.
Keogh, K. M., Smith, S. M., White, P., McGilloway, S., Kelly,
A.,
Gibney, J., & O’Dowd, T. (2011). Psychological family
interven-
tion for poorly controlled type 2 diabetes. American Journal of
Managed Care, 17, 105–113.
Leventhal, H., Leventhal, E. A., & Breland, J. (2011). Cognitive
sci-
ence speaks to the “common-sense” of chronic illness manage-
ment. Annals of Behavioral Medicine, 41, 152–163.
Leventhal, H., Nerenz, D. R., & Steele, D. J. (1984). Illness
repre-
sentations and coping with health threats. In A. Baum & J.
Singer
(Eds.), A handbook of psychology and health (Vol. 4, pp. 219–
252). Hillsdale, NJ: Erlbaum.

Moss-Morris, R., Weinman, J., Petrie, K. J., Horne, R.,
Cameron,
L. D., & Buick, D. (2002). The Revised Illness Perception
Ques-
tionnaire (IPQ-R). Psychology & Health, 17, 1–16.
Petrie, K. J., Müller, J. T., Schirmbeck, F., Donkin, L.,
Broadbent, E.,
Ellis, C. J., . . . Reif, W. (2007). Effect of providing
information
about normal test results on patients’ reassurance: Randomised
controlled trial. British Medical Journal, 334, 352–355.
Petrie, K. J., Perry, K., Broadbent, E., & Weinman, J. (2011). A
text
message program designed to modify patients’ illness and treat-
ment beliefs improves self-reported adherence to asthma pre-
venter medication. British Journal of Health Psychology.
Advance
online publication. doi:10.1111/j.2044-8287.2011.02033.x
Reynolds, L., Broadbent, E., Ellis, C. J., Gamble, G., & Petrie,
K. J.
(2007). Patient drawings illustrate psychological and functional
status in heart failure. Journal of Psychosomatic Research, 63,
525–532.
Weinman, J., Petrie, K. J., Moss-Morris, R., & Horne, R.
(1996). The
Illness Perception Questionnaire: A new method for assessing
ill-
ness perceptions. Psychology and Health, 11, 431–446.
Weinman, J., Petrie, K. J., Sharpe, N., & Walker, S. (2000).
Causal

attributions in patients and spouses following a heart attack and
subsequent lifestyle changes. British Journal of Health Psychol-
ogy, 5, 263–273.
Weinman, J., Yusuf, G., Berks, R., Rayner, S., & Petrie, K. J.
(2009).
How accurate is patients’ anatomical knowledge: A cross-
sectional
questionnaire study of six patient groups and a general public
sample. BMC Family Practice, 10, 43. doi:10.1186/1471-2296-
10-43.
Involving Family in Psychosocial
Interventions for Chronic Illness
Lynn M. Martire and Richard Schulz
Department of Psychiatry, University of Pittsburgh School of
Medicine, and University Center for Social and
Urban Research, University of Pittsburgh
ABSTRACT—Interactions with close family members have
consequences for the emotional and physical well-being of
individuals who are dealing with a chronic physical illness.
Therefore, inclusion of a close family member in psycho-
social interventions for chronic illnesses is a logical treat-
ment approach that has the potential to boost the effects

of intervention on the patient and also benefit the family
member. However, randomized, controlled studies indicate
that such family-oriented interventions generally have
small effects. The efficacy of these treatment approaches
might be enhanced by targeting specific interactions that
emerging research identifies as promoting or derailing
healthy behaviors and by better incorporating strategies
from family caregiver interventions. In addition, family-
oriented interventions should be more fully evaluated, by
assessing the benefits for both patients and family mem-
bers. Future research in this area can tell us much about
how and when to involve family in treatment of specific
chronic illnesses and, in turn, may inform conceptual
models of the impact of family interactions on health.
KEYWORDS: chronic illness, psychosocial interventions,
family, social support
Family is not an important thing, it’s everything.
—Michael J. Fox

Regardless of an individual’s celebrity, few circumstances in
adulthood are more stressful than a chronic illness, and family
plays an important role in psychological adjustment and symp-
tom management. Emotionally and instrumentally supportive
actions on the part of family members, as well as family
conflict
and criticism, have been linked with patients’ emotional well-
being, health behaviors, immune function, blood pressure, and
illness events (Kiecolt-Glaser & Newton, 2001; Schmaling &
Sher, 2000). These associations have been observed across ill-
nesses as diverse as cardiovascular disease, chronic pain dis-
orders, arthritis, cancer, renal disease, and Type 2 diabetes, as
well as in healthy individuals who are at risk for illness. In turn,
illness in a loved one can erode family members’ psychological
and physical well-being over time and compromise their ability
to be supportive, especially when the illness is life threatening
or the patient requires assistance with daily activities.
Psychosocial or behavioral interventions for chronic illness,
such as patient education, support groups, and cognitive-

behavioral therapies, have been shown to have effects on health
and emotional well-being that surpass improvements attained
with usual medical care alone (i.e., medication or surgery).
Because of the links between family relationships and chronic-
illness management, some researchers have incorporated a close
family member such as the spouse in these interventions. The
rationale for involving a family member in treatment can be
found in the biopsychosocial model of health and illness and
specific marital and family-systems frameworks. These con-
ceptual models and frameworks have been supported by
empirical evidence showing that close social relationships, es-
pecially the marital relationship, affect biological systems,
health behaviors, and psychological well-being. Therefore,
improving the quality of interactions with a close family
member
or involving that individual in disease management should
promote adjustment to chronic illness. Specific treatment ap-
proaches range from enlisting the family member’s help in

changing the patient’s health behaviors (e.g., training the
spouse
of a patient with chronic pain to help the patient practice pro-
gressive muscle relaxation) to also addressing issues of how
spouses can provide effective emotional and instrumental sup-
port (e.g., counseling for couples dealing with the wife’s breast
cancer).
Incorporating a close family member in psychosocial treat-
ment may have a positive impact on patient health behaviors,
Address correspondence to Lynn M. Martire, 121 University
Place,
Room 508, University of Pittsburgh, Pittsburgh, PA 15260; e-
mail:
[email protected]
CURRENT DIRECTIONS IN PSYCHOLOGICAL SCIENCE
90 Volume 16—Number 2Copyright r 2007 Association for
Psychological Science
emotional well-being, and symptomatology as the result of
increased empathy and supportiveness of the family member. In
addition, the family member’s caregiving burden may be eased
by the validation of his or her caregiving experiences and im-

proved interactions with the patient. In the following sections,
we summarize our findings from two recent systematic reviews
of the literature that compared family-oriented psychosocial
intervention with usual medical care alone and with patient-
oriented psychosocial intervention.
ARE FAMILY INTERVENTIONS MORE BENEFICIAL
THAN USUAL MEDICAL CARE?
In our first review of the literature, we included studies that
compared family-oriented psychosocial interventions (which
included usual medical care) to usual medical care that did not
include a psychosocial component. We reviewed the findings of
70 randomized, controlled studies that compared these two
approaches and that recruited a family member, such as the
spouse or an adult son or daughter, for every patient (Martire,
Lustig, Schulz, Miller, & Helgeson, 2004). These interventions
focused on different illness populations; used psychological,
social, or behavioral approaches; and were targeted at either the
patient’s closest family member (the primary caregiver in some

studies) or both patient and family member. Because we in-
cluded different illnesses in this review, our focus was on out-
comes that were generalizable across illnesses, and we
examined
effects on the patient (i.e., depressive and anxiety symptoms,
relationship satisfaction, physical disability, and mortality) as
well as on the family member (i.e., depressive and anxiety
symptoms, relationship satisfaction, and caregiving burden).
Almost half of the studies focused on populations with
dementia;
the remainder focused on those with cardiovascular disease,
cancer, general medical frailty, chronic pain, stroke, rheumatoid
arthritis, and traumatic brain injury.
For patients, interventions that included only patients’ spou-
eses had small, positive effects on reducing depressive symp-
toms across various illnesses, but interventions that included
mixed groups of family members (e.g., spouses and adult chil-
dren) did not decrease depressive symptoms. In addition, fam-
ily-oriented interventions that focused on individuals dealing

with hypertension or cardiovascular disease resulted in a small
decreased risk for patient mortality, especially if the interven-
tions included mixed groups of family members and used be-
havioral treatment approaches. For family members, we found
that family-oriented interventions slightly reduced the psycho-
logical burden of caregiving, as well as depressive and anxiety
symptoms. This latter finding has important implications for
family members’ physical health because caregiver burden and
distress have been linked to an increased risk for morbidity and
mortality (Schulz & Beach, 1999).
ARE FAMILY INTERVENTIONS MORE BENEFICIAL
THAN PATIENT INTERVENTIONS?
From a behavioral-medicine perspective, the more interesting
question than whether family interventions are more beneficial
than usual medical care is whether targeting both the patient
and
the family member with psychosocial or behavioral strategies is
better than targeting only the patient with these strategies (with

both treatment approaches including usual medical care; see
Fig. 1). In other words, is there an advantage to patient and
family member when the family member is included in a psy-
chosocial intervention? Unfortunately, few studies have been
Illness Parameters
- Physical symptoms
- Treatment side effects
Emotional Well-Being
- Depressive symptoms
- Anxiety
- Perceived stress
Family Factors
- Effective emotional support
and assistance to patient
- Burden on family member
Health Behaviors
- Diet, exercise
- Smoking, alcohol use
- Medication adherence
Biological Parameters
- Cardiovascular, immune,
and neuroendocrine
functioning
Functional Status
- Overall illness severity

- Physical disability
Patient-Oriented
Psychosocial Intervention
Family-Oriented
Psychosocial Intervention
Fig. 1. Heuristic model showing domains of functioning (i.e.,
emotional well-being, health behaviors, and family factors) that
mediate parameters of
chronic illness in their effects on a patient’s functioning.
Intervention that incorporates or involves a close family
member may have added benefit as
compared to intervention aimed only at the patient, due to the
former’s effects on the third domain of functioning (i.e., family
factors). The type of
family-oriented intervention depicted in this model includes
standard content (e.g., education regarding illness etiology and
cognitive-behavioral
skills training for illness management) and incorporates a close
family member by treating that individual as a collaborator in
the patient’s inter-
vention and/or by addressing his or her personal concerns,
burden, and supportiveness of the patient. Examples of specific
constructs are provided
for each domain of functioning.
Volume 16—Number 2 91
designed to answer this question, and those show mixed
findings

for patients and focus little attention on whether family
members
benefited from being included.
The findings of 12 randomized, controlled studies comparing
family-oriented intervention to patient-oriented intervention
were recently reviewed (Martire, 2005). Approximately half of
these studies showed significant improvements over time for
those receiving family intervention or reported that there was
also a statistically significant advantage of family intervention
over patient intervention. For example, individuals with chronic
low back pain who attended exercise sessions in combination
with couple-oriented behavioral therapy showed greater less-
ening of pain and pain behavior (e.g., grimacing, limping,
groaning) and greater diminishment of the impact of pain on
their lives than did individuals who received only an exercise
intervention. Our own research with osteoarthritis patients and
their spouses showed that patients felt they managed their
arthritis more effectively if they received a couples-oriented

education and support intervention than if they received edu-
cation and support with other patients only (Martire et al.,
2003).
Consistent with these positive findings, a more recent study
showed that problem-solving therapy for cancer patients and
their significant others reduced the patients’ psychological
distress more than did patient-focused problem-solving therapy
(Nezu, Nezu, Felgoise, McClure, & Houts, 2003). Significant
improvements over time as the result of family-oriented inter-
vention have also been reported regarding blood-pressure con-
trol in hypertension; stress and cardiovascular complications in
postcardiac surgery; and pain, psychosocial adjustment, and
number of medical visits in chronic pain.
In contrast to the positive findings for family-oriented inter-
ventions, an educational intervention for individuals with
rheumatoid arthritis and their significant others resulted in
decreased self-efficacy and increased fatigue, whereas a similar
patient-oriented intervention was found to enhance self-efficacy
and reduce fatigue (see review by Martire, 2005). Such unex-

pected negative effects of family-oriented intervention may oc-
cur in studies that do not address communication issues between
patients and partners, or partners’ personal concerns. We return
to this issue in the next section.
In the remaining studies, the more efficacious approach for
patients depended on factors such as patient gender and specific
type of intervention (i.e., educational versus behavioral
approach; see review by Martire, 2005). For example, a couple-
oriented behavioral program for obese individuals with Type 2
diabetes resulted in more weight loss for female patients than
did
a patient-oriented program, whereas male patients lost more
weight in the patient-oriented program. (Spouses who partici-
pated in the couples program lost more weight than spouses of
individuals in the patient-oriented program, regardless of their
gender.) In addition, studies focused on rheumatoid arthritis or
osteoarthritis showed that family-oriented interventions that
used cognitive-behavioral rather than educational approaches

resulted in greater reductions in joint swelling or pain-related
outcomes than did cognitive-behavioral interventions for pa-
tients only.
These studies illustrate how research can reveal for whom and
under what conditions family intervention may be especially
beneficial, by identifying patient, family member, or
intervention
characteristics that moderate the effects of family intervention.
A recent study showed that a couple-oriented intervention for
breast cancer patients that was designed to enhance support
exchanges (e.g., effective communication, problem solving as a
team, respecting differences in coping styles) was most helpful
to
patients with unsupportive partners (Manne et al., 2005). This
study, as well as others (e.g., Helgeson, Cohen, Schulz, &
Yasko,
2000), raises an important question about patient-oriented in-
terventions that have targeted individuals with unsupportive
relationships and have had small effects on patient outcomes:
Would the effects of these interventions have been stronger had

the interventions included a family member?
HOW CAN WE ENHANCE THE EFFICACY OF
FAMILY INTERVENTIONS?
As described above, the findings of randomized trials indicate
that family-oriented interventions do not consistently outper-
form patient-oriented psychosocial interventions. There are
several explanations why the promise of family interventions
has
not been fulfilled. Methodological flaws, such as failing to
ensure
the full participation of family members, may explain why this
approach has had null or weak effects in some studies. In
addition, family-oriented interventions have not consistently
targeted family interactions that affect health and issues sur-
rounding the burden of illness on a family. Interventions appear
to be more beneficial for patients (Martire et al., 2004) and for
family members (Martire, 2005) when they address such issues.
Research on family caregiver interventions illustrates the value
of particular strategies such as stress management, skills

training, and validation of the family member’s experiences as a
provider of support (Schulz et al., 2002).
But how can we explain the generally small effects of inter-
ventions targeting family support or the relationship between
patient and family member? One explanation may be found in
the empirical literature that is the foundation for many family-
oriented interventions for chronic illness. Early correlational
research revealed that patients sometimes perceive well-
intentioned family actions or communications as unhelpful, and
that overprotective or solicitous behaviors may be perceived as
helpful by patients but also cause them to be more physically
inactive and dependent (Lyons, Sullivan, Ritvo, & Coyne,
1995).
Thus, past family-oriented interventions may not have been
successful in (a) reducing the frequency of family actions or
communications that derail healthy behaviors and distress
patients or (b) bolstering interactions that promote healthy be-
haviors and emotional well-being.

92 Volume 16—Number 2
Family-Oriented Interventions
More recent conceptual and empirical work might be useful in
developing family interventions that have greater impact and
more consistent effects. Here, we provide examples of relevant
research in two of several useful areas: autonomy support and
social control. We chose to highlight these two areas of research
because they focus on health behaviors, and management of
today’s most common chronic illnesses often requires patients
to
make substantial changes in diet and exercise and to adhere to a
medication regimen despite unpleasant side effects. Autonomy
support refers to behaviors that are characterized by warmth,
empathy, and understanding for an individual’s situation;
patient-centered communication; and the provision of choices
for making health behavior changes. A program of research on
individuals with Type 2 diabetes has shown that patients with
health-care providers who are more supportive of their

autonomy
feel that they are better able to regulate their blood glucose and
show improved glucose control over 1 year (Williams,
Freedman,
& Deci, 1998). Complementing these findings, our work has
demonstrated that older adults with disabling arthritis who feel
that they have no choice over the amount, timing, and manner
of physical assistance from their spouses experience increased
depressive symptoms over time (Martire, Stephens, Druley, &
Wojno, 2002).
Emerging work in the area of health-related social control also
might be useful to incorporate in family interventions. Social
control is thought to be distinct from social support and refers
to
an individual’s attempts to regulate or influence the behaviors
of
another person through actions, affective responses, and cor-
rective feedback. Early theory on social control suggested that
it
may deter poor health practices but at the same time cause dis-

tress by evoking irritation, resentment, or guilt. However, more
recent research has identified tactics that may promote healthy
behaviors and also be appreciated by patients—tactics such
as persuasion (e.g., efforts to convince or motivate), modeling
(enacting the behavior), reinforcement, and using logic (e.g.,
pointing out positive consequences; Lewis & Butterfield, 2005).
As these lines of research move forward and are conceptually
and empirically synthesized with previous research, findings
can
inform the development of psychosocial interventions for spe-
cific illness populations. Specifically, future interventions may
include content aimed at teaching family members how to sup-
port patients’ need to make their own choices and carry out
daily
activities independently. This type of family-oriented interven-
tion may show stronger and more consistent advantages over
psychosocial treatments focused only on the patient.
CONCLUSIONS
There is not yet strong evidence for the efficacy of family-

oriented interventions. However, including family members in
health care is an approach that has much face validity.
Moreover,
the small effects that have generally been observed across
studies, particularly with regard to the emotional well-being of
family members, make a compelling argument for further re-
search using well-designed studies that can more fully reveal
what type of patient may benefit from this approach.
Future research may better evaluate the added benefit of
family-oriented interventions by including specific features that
have often been disregarded in past studies, such as random
assignment of participants and adequate statistical power to
detect between-group differences. In this research, it would be
optimal to assess outcomes for both the patient and the family
member. Finally, we have much to learn about the causal
mechanisms linking positive and negative aspects of close
social
relationships to physical health (e.g., psychological, behavioral,
and biological pathways), and such knowledge is critical for

developing family-oriented interventions that truly make a dif-
ference for patients and their families.
Recommended Reading
Cohen, S., Underwood, L.G., & Gottlieb, B.H. (2000). Social
support
measurement and intervention: A guide for health and social
scientists. New York: Oxford University Press.
Lyons, R.R., Sullivan, M.J.L., Ritvo, P.G., & Coyne, J.C.
(1995).
(See References)
Martire, L.M. (2005). (See References)
Martire, L.M., Lustig, A.P., Schulz, R., Miller, G.E., &
Helgeson, V.S.
(2004). (See References)
Schmaling, K.B., & Sher, T.G. (2000). (See References)
Acknowledgments—Preparation of this manuscript was sup-
ported in part by National Institutes of Health Grants K01
MH065547, R24 HL076852-076858 (Pittsburgh Mind-Body
Center), and R01 NR008272.
REFERENCES
Helgeson, V.S., Cohen, S., Schulz, R., & Yasko, J. (2000).

Group support
interventions for women with breast cancer: Who benefits from
what? Health Psychology, 19, 107–114.
Kiecolt-Glaser, J.K., & Newton, T.L. (2001). Marriage and
health: His
and hers. Psychological Bulletin, 127, 472–503.
Lewis, M.A., & Butterfield, R. (2005). Antecedents and
reactions to
health-related social control. Personality and Social Psychology
Bulletin, 31, 416–427.
Lyons, R.R., Sullivan, M.J.L., Ritvo, P.G., & Coyne, J.C.
(1995). Rela-
tionships in chronic illness and disability. Thousand Oaks, CA:
Sage.
Manne, S.L., Ostroff, J.S., Winkel, G., Fox, K., Grana, G.,
Miller, E.,
et al. (2005). Couple-focused group intervention for women
with
early stage breast cancer. Journal of Consulting and Clinical
Psychology, 73, 634–646.
Martire, L.M. (2005). The ‘‘relative’’ efficacy of involving
family in
psychosocial interventions for chronic illness: Are there added

benefits to patients and family members? Families, Systems &
Health, 23, 312–328.
Volume 16—Number 2 93
Martire, L.M., Lustig, A.P., Schulz, R., Miller, G.E., &
Helgeson, V.S.
(2004). Is it beneficial to involve a family member? A meta-
analytic review of psychosocial interventions for chronic
illness.
Health Psychology, 23, 599–611.
Martire, L.M., Schulz, R., Keefe, F.J., Starz, T.W., Osial, T.A.,
Dew,
M.A., et al. (2003). Feasibility of a dyadic intervention for
man-
agement of osteoarthritis: A pilot study with older patients and
their spousal caregivers. Aging and Mental Health, 7, 53–60.
Martire, L.M., Stephens, M.A.P., Druley, J.A., & Wojno, W.C.
(2002).
Negative reactions to received spousal care: Predictors and
consequences of miscarried support. Health Psychology, 21,
167–176.

Nezu, A.M., Nezu, C.M., Felgoise, S.H., McClure, K.S., &
Houts, P.S.
(2003). Project Genesis: Assessing the efficacy of problem-
solving
therapy for distressed adult cancer patients. Journal of
Consulting
and Clinical Psychology, 71, 1036–1048.
Schmaling, K.B., & Sher, T.G. (2000). The psychology of
couples and
illness: Theory, research, & practice. Washington, DC:
American
Psychological Association.
Schulz, R., & Beach, S.R. (1999). Caregiving as a risk factor
for mor-
tality: The Caregiver Health Effects Study. Journal of the
American
Medical Association, 282, 2215–2219.
Schulz, R., O’Brien, A., Czaja, S., Ory, M., Norris, R., Martire,
L.M.,
et al. (2002). Dementia caregiver intervention research: In
search
of clinical significance. Gerontologist, 42, 589–602.
Williams, G.C., Freedman, Z.R., & Deci, E.L. (1998).
Supporting

autonomy to motivate patients with diabetes for glucose control.
Diabetes Care, 21, 1644–1651.
94 Volume 16—Number 2
Family-Oriented Interventions
<<
/ASCII85EncodePages false
/AllowTransparency false
/AutoPositionEPSFiles true
/AutoRotatePages /None
/Binding /Left
/CalGrayProfile (Dot Gain 20%)
/CalRGBProfile (sRGB IEC61966-2.1)
/CalCMYKProfile (U.S. Web Coated 050SWOP051 v2)
/sRGBProfile (sRGB IEC61966-2.1)
/CannotEmbedFontPolicy /Error
/CompatibilityLevel 1.4
/CompressObjects /Tags
/CompressPages true
/ConvertImagesToIndexed true
/PassThroughJPEGImages true
/CreateJobTicket false
/DefaultRenderingIntent /Default
/DetectBlends true
/DetectCurves 0.0000
/ColorConversionStrategy /CMYK
/DoThumbnails false
/EmbedAllFonts true
/EmbedOpenType false
/ParseICCProfilesInComments true
/EmbedJobOptions true
/DSCReportingLevel 0

/EmitDSCWarnings false
/EndPage -1
/ImageMemory 1048576
/LockDistillerParams false
/MaxSubsetPct 100
/Optimize true
/OPM 1
/ParseDSCComments true
/ParseDSCCommentsForDocInfo true
/PreserveCopyPage true
/PreserveDICMYKValues true
/PreserveEPSInfo true
/PreserveFlatness true
/PreserveHalftoneInfo false
/PreserveOPIComments true
/PreserveOverprintSettings true
/StartPage 1
/SubsetFonts true
/TransferFunctionInfo /Apply
/UCRandBGInfo /Preserve
/UsePrologue false
/ColorSettingsFile ()
/AlwaysEmbed [ true
]
/NeverEmbed [ true
]
/AntiAliasColorImages false
/CropColorImages true
/ColorImageMinResolution 300
/ColorImageMinResolutionPolicy /OK
/DownsampleColorImages true
/ColorImageDownsampleType /Average
/ColorImageResolution 150
/ColorImageDepth -1
/ColorImageMinDownsampleDepth 1
/ColorImageDownsampleThreshold 1.50000

/EncodeColorImages true
/ColorImageFilter /DCTEncode
/AutoFilterColorImages true
/ColorImageAutoFilterStrategy /JPEG
/ColorACSImageDict <<
/QFactor 0.76
/HSamples [2 1 1 2] /VSamples [2 1 1 2]
>>
/ColorImageDict <<
/QFactor 0.15
>>
/JPEG2000ColorACSImageDict <<
/TileWidth 256
/TileHeight 256
/Quality 30
>>
/JPEG2000ColorImageDict <<
/TileWidth 256
/TileHeight 256
/Quality 30
>>
/AntiAliasGrayImages false
/CropGrayImages true
/GrayImageMinResolution 300
/GrayImageMinResolutionPolicy /OK
/DownsampleGrayImages true
/GrayImageDownsampleType /Average
/GrayImageResolution 150
/GrayImageDepth -1
/GrayImageMinDownsampleDepth 2
/GrayImageDownsampleThreshold 1.50000
/EncodeGrayImages true
/GrayImageFilter /DCTEncode
/AutoFilterGrayImages true
/GrayImageAutoFilterStrategy /JPEG

/GrayACSImageDict <<
/QFactor 0.76
>>
/GrayImageDict <<
/QFactor 0.15
>>
/JPEG2000GrayACSImageDict <<
/TileWidth 256
/TileHeight 256
/Quality 30
>>
/JPEG2000GrayImageDict <<
/TileWidth 256
/TileHeight 256
/Quality 30
>>
/AntiAliasMonoImages false
/CropMonoImages true
/MonoImageMinResolution 1200
/MonoImageMinResolutionPolicy /OK
/DownsampleMonoImages true
/MonoImageDownsampleType /Average
/MonoImageResolution 300
/MonoImageDepth -1
/MonoImageDownsampleThreshold 1.50000
/EncodeMonoImages true
/MonoImageFilter /CCITTFaxEncode
/MonoImageDict <<
/K -1
>>
/AllowPSXObjects false
/CheckCompliance [
/None
]

/PDFX1aCheck false
/PDFX3Check false
/PDFXCompliantPDFOnly false
/PDFXNoTrimBoxError true
/PDFXTrimBoxToMediaBoxOffset [
0.00000
0.00000
0.00000
0.00000
]
/PDFXSetBleedBoxToMediaBox true
/PDFXBleedBoxToTrimBoxOffset [
0.00000
0.00000
0.00000
0.00000
]
/PDFXOutputIntentProfile (None)
/PDFXOutputConditionIdentifier ()
/PDFXOutputCondition ()
/PDFXRegistryName ()
/PDFXTrapped /False
/CreateJDFFile false
/Description <<
/CHS
<FEFF4f7f75288fd94e9b8bbe5b9a521b5efa76840020004100640
06f006200650020005000440046002065876863900275284e8e9a
d88d2891cf76845370524d53705237300260a853ef4ee54f7f7528
0020004100630072006f0062006100740020548c0020004100640
06f00620065002000520065006100640065007200200035002e00
3000204ee553ca66f49ad87248672c676562535f00521b5efa7684
00200050004400460020658768633002>
/CHT
<FEFF4f7f752890194e9b8a2d7f6e5efa7acb76840020004100640
06f006200650020005000440046002065874ef69069752865bc9a

d854c18cea76845370524d5370523786557406300260a853ef4ee5
4f7f75280020004100630072006f0062006100740020548c002000
410064006f0062006500200052006500610064006500720020003
5002e003000204ee553ca66f49ad87248672c4f86958b555f5df25e
fa7acb76840020005000440046002065874ef63002>
/DAN
<FEFF004200720075006700200069006e0064007300740069006
c006c0069006e006700650072006e0065002000740069006c0020
006100740020006f0070007200650074007400650020004100640
06f006200650020005000440046002d0064006f006b0075006d00
65006e007400650072002c0020006400650072002000620065006
400730074002000650067006e0065007200200073006900670020
00740069006c002000700072006500700072006500730073002d0
07500640073006b007200690076006e0069006e00670020006100
660020006800f8006a0020006b00760061006c006900740065007
4002e0020004400650020006f0070007200650074007400650064
00650020005000440046002d0064006f006b0075006d0065006e0
074006500720020006b0061006e002000e50062006e0065007300
2000690020004100630072006f00620061007400200065006c006
c006500720020004100630072006f006200610074002000520065
006100640065007200200035002e00300020006f00670020006e0
079006500720065002e>
/DEU …
Affective Reactivity to Daily Stress and 20-Year Mortality Risk
in Adults
With Chronic Illness: Findings From the National Study of
Daily Experiences
Jessica J. Chiang
Northwestern University
Nicholas A. Turiano

West Virginia University
Daniel K. Mroczek and Gregory E. Miller
Northwestern University
Objective: Daily stress processes have been previously linked to
health-related outcomes, but implica-
tions for longevity remain unclear. The present study examined
whether daily stress exposure and/or
affective responses to daily stressors predicted mortality risk
over a 20-year period. Based on the
hypothesis that chronic illness confers vulnerability to
deleterious effects of stress, we also examined
whether its presence accentuated the association between daily
stress processes and later mortality risk.
Method: Participants were 1,346 middle-aged adults from the
survey of Midlife Development in the
United States who also completed the National Study of Daily
Experiences. Participants reported on their
experiences of stress and affect for 8 consecutive evenings, and
mortality data were collected over the
next 20 years, using the National Death Index and other
methods. Results: There was a positive
association between total number of stressors experienced
across days and mortality risk. There was also
a positive association between increases in negative affect on
stressor days relative to nonstressor days
and risk for mortality. The presence of a chronic illness
moderated this association such that negative
affective reactivity predicted mortality risk among individuals
with at least one chronic illness but not
among otherwise healthy individuals. This association was
independent of sociodemographic character-
istics, typical levels of negative affect on nonstressor days, and
total number of endorsed stressors.
Conclusion: These results suggest that greater increases in

negative affect in response to stress in
everyday life may have long-term consequences for longevity,
particularly for individuals with chronic
illness.
Keywords: daily experience, stress reactivity, negative affect,
chronic disease, longevity
Acute stressful experiences, such as interpersonal conflict and
work deadlines, are ubiquitous in everyday life. Research
suggests
that these everyday experiences with stress are consequential
for
physical health. For instance, individuals who report more
stress in
their daily lives endorse more somatic and infectious illness
symp-
toms and have smaller antibody responses to ingested antigens
(DeLongis, Folkman, & Lazarus, 1988; Stone et al., 1994;
Stone,
Reed, & Neale, 1987). More daily stress is also associated with
alterations in biological processes thought to contribute to the
development and worsening of diseases, including elevated
blood
pressure, stress hormones, and inflammatory biomarkers
(Chiang,
Eisenberger, Seeman, & Taylor, 2012; Stawski, Cichy, Piazza,
&
Almeida, 2013; Uchino, Berg, Smith, Pearce, & Skinner, 2006).
Among individuals with existing chronic diseases, such as
asthma,
rheumatoid arthritis, irritable bowel syndrome, and diabetes,
higher daily stress is associated with more severe illness-related
symptoms (Halford, Cuddihy, & Mortimer, 1990; Levy, Cain,
Jarrett, & Heitkemper, 1997; Stone, Broderick, Porter, & Kaell,

1997).
Of importance is that not all individuals confronting stress
develop poor health, which has been attributed, in part, to vari-
ability in people’s affective responses to stress (Almeida,
Piazza,
Stawski, & Klein, 2011; Lovallo & Gerin, 2003). Perceptions of
threat elicit increases in negative affect and decreases in
positive
affect, which, in turn, can modify patterns of cardiac, vascular,
endocrine, metabolic, and immune functioning (Holmes, Krantz,
Rogers, Gottdiener, & Contrada, 2006; Kiecolt-Glaser,
McGuire,
Robles, & Glaser, 2002; Lovallo & Gerin, 2003; Pressman &
Cohen, 2005). Repeated and/or heightened activation of these
systems over time may exact a toll, ultimately altering the
function
of tissues and organs in ways that contribute to morbidity and
This article was published Online First November 20, 2017.
Jessica J. Chiang, Institute for Policy Research, Northwestern
Univer-
sity; Nicholas A. Turiano, Department of Psychology, West
Virginia
University; Daniel K. Mroczek, Department of Psychology and
Depart-
ment of Medical Social Sciences, Northwestern University;
Gregory E.
Miller, Institute for Policy Research and Department of
Psychology, North-
western University.
The present study was supported by the National Institute on
Aging
(P01-AG020166; R01-AG018436) and the National Heart, Lung,

and
Blood Institute (R01-HL122328; F32-HL134276).
Correspondence concerning this article should be addressed to
Jessica J.
Chiang, Institute for Policy Research, Northwestern University,
1801 Ma-
ple Avenue, Suite 2450, Evanston, IL 60201. E-mail:
[email protected]
northwestern.edu
THIS ARTICLE HAS BEEN CORRECTED. SEE LAST PAGE
T
hi
s
do
cu
m
en
t
is
co
py
ri
gh
te
d
by

th
e
A
m
er
ic
an
P
sy
ch
ol
og
ic
al
A
ss
oc
ia
ti
on
or
on
e
of

it
s
al
li
ed
pu
bl
is
he
rs
.
T
hi
s
ar
ti
cl
e
is
in
te
nd
ed
so

le
ly
fo
r
th
e
pe
rs
on
al
us
e
of
th
e
in
di
vi
du
al
us
er
an
d

is
no
t
to
be
di
ss
em
in
at
ed
br
oa
dl
y.
Health Psychology © 2017 American Psychological Association
2018, Vol. 37, No. 2, 170 –178 0278-6133/18/$12.00
http://dx.doi.org/10.1037/hea0000567
170
http://dx.doi.org/10.1037/hea0000567
mortality (McEwen & Seeman, 1999). Thus, individuals who

react
strongly (i.e., greater increases in negative affect or decreases
in
positive affect) to stress repeatedly day after day may be more
vulnerable to health problems associated with exposure to
stress.
Indeed, greater stress-related increases in negative affect in
daily
life have been linked to smaller antibody responses to ingested
antigens (Stone, Marco, Cruise, Cox, & Neale, 1996), higher
cortisol levels (Jacobs et al., 2007), lower heart rate variability
(Sin, Sloan, McKinley, & Almeida, 2016), and self-reports of
mood problems and chronic illness (Charles, Piazza, Mogle, Sli-
winski, & Almeida, 2013; Piazza, Charles, Sliwinski, Mogle, &
Almeida, 2013). Moreover, larger reductions in positive affect
in
response to daily stress have been associated with higher levels
of
inflammatory biomarkers (Sin, Graham-Engeland, Ong, &
Almeida, 2015), lower sleep efficiency and quality (Ong et al.,
2013), and more depressive symptoms (O’Neill, Cohen, Tolpin,
&
Gunthert, 2004). Of particular interest, more positive affect
reac-
tivity has been linked to increased mortality risk in the VA Nor-
mative Aging Study (Mroczek et al., 2015). In this study of 181
men, larger decreases in positive affect on stressor days
compared
with nonstressor days were associated with greater risk for mor-
tality 10 years later. This association was independent of men’s
frequency of stress exposure and typical experiences of negative
and positive affect in daily life. Interestingly, greater increases
in
negative affect in response to stress did not predict mortality
risk.

To our knowledge, the study conducted by Mroczek et al.
(2015) is the only study to date that has explored how daily
stress
and affective processes relate to longevity. Although
provocative,
findings from this initial study were based on a small, all-male
sample, raising concerns regarding generalizability.
Furthermore,
although this study used daily reports of physical symptoms and
bodily pain to adjust for preexisting health problems, it did not
explicitly consider the role of chronic diseases. The most
prevalent
chronic illnesses in America, like cardiovascular disease and
var-
ious cancers, can alter people’s affective and biological
responses
to stress (e.g., Costanzo, Stawski, Ryff, Coe, & Almeida, 2012;
Kop et al., 2008; van Der Pompe, Antoni, & Heijnen, 1996), and
forecast shorter lifespans even among those who are
successfully
treated (Hudson et al., 1998; Ronkainen et al., 2001). These
observations raise the possibility that chronic diseases
contributed
to the mortality risks associated with greater positive affective
reactivity (i.e., daily stress-related decreases in positive affect)
in
the Mroczek et al. study. Alternatively, chronic disease may
func-
tion as a moderator of this association, creating an underlying
vulnerability that accentuates the mortality risks associated with
stressor exposure and/or affective reactivity. Indeed, there is
evi-
dence to suggest that stress-evoked changes in the
cardiovascular
and inflammatory systems are magnified in patients with a

NIH-PA Author Manuscript NIH-PA.docx

NIH-PA Author Manuscript NIH-PA.docx

Recommended

Recommended

More Related Content

Similar to NIH-PA Author Manuscript NIH-PA.docx

Similar to NIH-PA Author Manuscript NIH-PA.docx (20)

More from hallettfaustina

More from hallettfaustina (20)

Recently uploaded

Recently uploaded (20)

NIH-PA Author Manuscript NIH-PA.docx