Cognitive Therapy and Research, Vol. 28, No.

5, October 2004 (
C 2004), pp. 629–644

The Psychopathology of Self-Assessed Health:

A Cognitive Approach to Health Anxiety
and Hypochondriasis
Paula G. Williams1

This paper reviews the relevant cognitive research on health anxiety and hypochondri-
asis, as well as research examining temperament/personality and social developmental
factors relevant to health and illness cognition and associated negative emotional states.
A cognitive developmental model is proposed which integrates individual difference
variables, selective attention/cognitive factors, and social learning processes that may
serve as vulnerability and maintenance factors in the etiology of health anxiety and
hypochondriasis. Future directions for the study of cognition and hypochondriasis are
suggested.
KEY WORDS: health anxiety, hypochondriasis, health cognition, self-assessed health, personality.

Self-assessments of health are central to health management. Judgments about

the status of health and illness influence self-care decision-making, health care uti-
lization, and communication with health care providers. Assessing health status
and taking care of health are necessary and unavoidable life activities. Unfortu-
nately, for some individuals, self-assessments of health become a source of pre-
occupation, emotional distress, and severe, sometimes chronic, disability. In our
current diagnostic system, this is characterized as hypochondriasis. In nonclinical
populations, less debilitating presentations of these characteristics have been termed
hypochondriacal tendencies or health anxiety, and individuals with these tenden-
cies are often referred to as the worried well. The purpose of this paper is to re-
view the relevant literature on the cognitive–emotional aspects of the etiology and
maintenance of health anxiety and hypochondriasis. First, health anxiety and
hypochondriasis are defined and distinguished from related conditions. Next,
current cognitive models of hypochondriasis are outlined and relevant literature
is reviewed. Finally, temperament/personality characteristics and social learning pro-
cesses that may serve as vulnerability and maintenance factors for health

1 Department of Psychology, 380 S. 1530 E. Room 502, University of Utah, Salt Lake City, Utah 84112;
e-mail: paula.williams@psych.utah.edu.

629
0147-5916/04/1000-0629/0 
C 2004 Springer Science+Business Media, Inc.
630 Williams

anxiety and hypochondriasis are discussed and an integrative cognitive model is

proposed.

HYPOCHONDRIASIS DEFINED

The central feature of hypochondriasis is preoccupation with the belief that one
has a serious disease, based on misinterpretation of bodily symptoms. This belief
occurs in the absence of known organic pathology and persists despite appropri-
ate medical evaluation and reassurance. Current diagnostic criteria (Diagnostic and
Statistical Manual of Mental Disorders—Text Revision; American Psychiatric Asso-
ciation, 2000) for hypochondriasis specify that these beliefs have persisted for at least
6 months and caused significant disability (i.e., emotional distress and/or functional
disability). Although the prevalence rates for clinical hypochondriasis are typically
not high (e.g., Barsky, Wyshek, Klerman, & Latham, 1990; Escobar et al., 1998),
transient hypochondriasis and health anxiety are relatively common (Looper &
Kirmayer, 2001). The costs of this constellation of conditions are enormous, including
overuse of health services, unnecessary medical tests, missed work, and subjective
distress.
Although it is beyond the scope of this paper to fully characterize the research,
there has been a long-standing controversy about both where hypochondriasis “be-
longs” (with respect to diagnostic category) and the adequacy of differentiating
hypochondriasis from other disorders. Currently, hypochondriasis is included with
the somatoform disorders, a category that also includes somatization disorder, so-
matoform pain disorder, body dysmorphic disorder, and conversion disorder. The
shared focus of these diagnoses is a dysfunction in the perception of the body. The
most frequently suggested alternative is that hypochondriasis be included in the anx-
iety disorders, based upon shared features, as well as comorbidity with other anxiety
disorders (Noyes, 2001).
Common differential diagnostic dilemmas include distinguishing hypochondri-
asis from somatization disorder, panic disorder, and illness phobia. Somatization
disorder is characterized by multiple unexplained physical symptoms of particular
number and type, whereas there is not specification regarding the number or nature
of symptoms in hypochondriasis. Other distinctions include the chronicity and age of
onset (somatization disorder is chronic and onset is before age 30; hypochondriasis
criteria specify 6 months duration and it can occur at any age) and sex ratio (som-
atization disorder is more common in women; the sex ratio for hypochondriasis is
equal; Noyes, 2001).
Despite the distinction in diagnostic criteria, it is not uncommon for clinical re-
search to consider hypochondriasis and somatization disorder as a single category.
This strategy may derive from common features between the two categories includ-
ing high symptom reporting and high rates of health care utilization. Relatedly, it
may also derive from the use of nonspecific measures such as the MMPI Scale 1 to
measure both clinical phenomena. Although this scale is labeled “hypochondriasis,”
high scores on this scale may be obtained by patients across the range of somato-
form disorders and by objectively ill medical patients. It is also clear that the term
somatization is used to describe high symptom reporters and should be distinguished
Health Anxiety and Hypochondriasis 631

from the clinical syndrome of somatization disorder. Despite some common features,
most recent research suggests that hypochondriasis and somatization disorders are
distinguishable categories (Cloninger, Sigvardsson, von Knorring, & Behman, 1984;
Kirmayer & Robbins, 1991) with different proposed etiological and maintaining fac-
tors. For example, family studies have shown high rates of antisocial personality dis-
order in male relatives of female somatization patients (Lilienfeld, Van Valkenburg,
Larntz, & Akiskal, 1986) and links between antisocial behavior and somatization in
families of clinic-referred children (Frick, Kuper, Silverthorn, & Cotter, 1995). These
findings suggest that somatization disorder may be better categorized as a personality
disorder. Indeed, somatization disorder is typically of long duration and interpersonal
in nature. Although rigorous etiologial research is lacking for hypochondriasis, much
of the current cognitive research suggests that vulnerability for the development of
hypochondriasis, as well as maintaining factors, may be similar to those for several
of the anxiety disorders.
Hypochondriasis and panic disorders share in common the catastrophic inter-
pretation of bodily symptoms. As Warwick and Salkovskis (1990) note, in panic, the
perceived catastrophe is immediate (e.g., “I’m having a heart attack”) and typically
related to autonomic arousal symptoms (e.g., racing heart, hyperventilation, sweaty
palms). In hypochondriasis, the catastrophic thinking is not immediate (e.g., “I have
cancer and will die in the future”) and the misperceived symptoms are typically not a
consequence of autonomic arousal (e.g., distended abdomen, a lump). It is notewor-
thy, however, that panic attacks can and do occur in individuals with hypochondriasis
(Fava, Grandi, Saviotti, & Conti, 1990). Nevertheless, hypochondriasis and panic dis-
order are considered distinguishable diagnoses, but may coexist (Barsky, Barnett, &
Cleary, 1994).
Both hypochondriasis and illness phobia have illness-related fear as a promi-
nent feature. However, hypochondriasis may be distinguished from illness phobia
by determining whether the fear is related to exposure to disease or “catching” a
disease (phobia) or the fear that one already has a disease (hypochondriasis).

COGNITIVE MODELS OF SELF-ASSESSED HEALTH

The development of a model of health anxiety and hypochondriasis must rest

upon the understanding of processes involved in health cognition, in general. Clearly,
self-assessments of health and illness are a reflection of shifts in actual health status.
However, a large body of research suggests that self-assessments of health are not
veridical; that is, they are imperfectly correlated with measures of objective health.
Therefore, attention has turned to understanding the psychosocial factors that influ-
ence self-assessed health. The prevailing models of symptom perception, reporting,
and health behavior are cognitive in nature (e.g., Cioffi, 1991; Leventhal, Meyer, &
Nerenz, 1980; Pennebaker, 1982) suggesting that there are stable mental represen-
tations that guide self-assessments of health. Indeed, research using experimental
information-processing paradigms supports the view that individual differences in
self-assessed health are reflected in schematic processing (i.e., preferential attention
allocation) of health-relevant information (Williams, Wasserman, & Lotto, 2003).
632 Williams

Specifically, individuals with higher symptom reports show greater interference in

color-naming health-relevant words in a modified Stroop task and facilitation in mak-
ing “me-not me” judgments for health-relevant descriptors. Moreover, high symptom
reporters among healthy young adults are more accurate than low symptom reporters
in recognizing previously presented health-relevant descriptors. Additionally, there
is a large body of evidence indicating that global health ratings (i.e., a rating of
overall health from “poor” to “excellent”) predict mortality above and beyond ob-
jective indicators of health (Idler & Benyamini, 1997). Thus, within the normal range,
self-assessments of health may have better predictive utility with respect to health
self-care behavior, health care utilization, and health outcomes than do objective
indicators of health.
However, self-assessed health might be considered to operate along a contin-
uum with varying degrees of accuracy (in relation to underlying pathophysiology),
efficiency, and associated emotional distress. At one end of the hypothetical contin-
uum, illness status is underestimated and, presumably, appropriate self-care activities
are not initiated, whereas at the other end, illness status is overestimated and may
be fraught with anxiety, and, presumably, inappropriate use of health services may
ensue (i.e., overuse or avoidance). Health anxiety and hypochondriasis would ap-
pear to fall at the high end of this theoretical continuum, although understanding the
entire continuum is clearly necessary to put these conditions into context. Several
important questions arise from consideration of cognitive models of self-assessed
health in relation to hypochondriasis: Do individuals with hypochondriacal con-
cerns show selective attention to health-relevant information? Is there evidence that
health-related information processing is biased in individuals with hypochondriacal
concerns? From a cognitive perspective, what might explain the development and
persistence of health-related anxiety? What are the implications of both the cognitive
and the emotional aspects of hypochondriacal tendencies for understanding health
self-care and health care utilization? To answer these and related questions, recent
cognitive research on hypochondriasis is considered.

A COGNITIVE APPROACH TO HYPOCHONDRIASIS

Although historical accounts of hypochondriasis reflected a range of perspec-

tives, cognitive models have been the predominant recent approach to the under-
standing and treatment of health anxiety and hypochondriasis. These models have
focused on the hypothesis that individuals with hypochondriacal tendencies tend
to misinterpret benign physical sensations as signs of disease (Barsky & Klerman,
1983). It is also hypothesized that hypochondriacal individuals overestimate the prob-
ability of serious illnesses and perceive bodily symptoms as more dangerous than
they actually are (Warwick & Salkovskis, 1990). To account for the maintenance
of hypochondriasis, Warwick and Salkovskis (1990) hypothesize that a combination
of a cognitive bias toward attending to information that confirms illness concerns
and “avoidance” behaviors (e.g., medical consultations, bodily checking) serve to
perpetuate health anxiety. The latter has been compared to the compulsions of in-
dividuals with obsessive–compulsive disorder in that the behaviors may initially, but
only temporarily, alleviate anxiety.
Health Anxiety and Hypochondriasis 633

Selective Attention to Illness Information: Self-Reports

The hypothesis that individuals with hypochondriacal tendencies overattend to
and misinterpret benign physical sensations has generally found support in the empir-
ical literature examining self-reported health cognition. Hypochondriasis is related to
self-reported sensitivity to normal physiological states and minor bodily sensations,
as measured by the Somatosensory Amplification Scale (Barsky, Wyshek, & Latham,
Klerman, 1990). Moreover, high scores on this scale predict the persistence of tran-
sient hypochondriasis (Barsky, Wyshek, & Klerman, 1990). Individuals with health
anxiety may hold the general belief that most symptoms are a sign of physical illness.
For example, in a task in which participants were asked to sort common and ambigu-
ous symptoms into the categories “healthy” or “not healthy,” hypochodriacal patients
considered more symptoms to be indicative of disease than did nonhypochondriacal
patients (Barsky, Coeytaux, Sarnie, & Cleary, 1993). Scores on this task were related
to self-reported somatosensory amplification, but not with medical care utilization,
suggesting that different cognitive processes may be related to decision-making re-
garding health care use. There has been some indication that both somatization
and hypochondriasis patients overattend to and misinterpret body sensations (by
self-report), but that hypochondriasis patients have a lower threshold for seeking
medical consultation in relation to such sensations. Rief, Hiller, and Margraf (1998)
found that a scale they labeled “Intolerance of Bodily Complaints” distinguished
hypochondriacal patients from those with somatization. This scale contains items
such as “I consult a doctor as soon as possible when I have bodily complaints” and
“If I don’t observe my body often, I could become seriously ill without noticing it.”

Selective Attention to Illness Information: Information-Processing Approaches

One limitation of using self-report measures to examine attention to body sen-
sations is that attentional processes may operate outside of conscious awareness.
Additionally, self-report measures are susceptible to self-presentational style in-
fluences (e.g., social desirability, self-handicapping, self-deception). Experimental
information-processing paradigms offer non-self-report methods to examine atten-
tional processes. Unfortunately, there have been few studies of this sort focused on
hypochondriasis. In one approach to the use of non-self-report methodology, Brown,
Kosslyn, Delamater, Fama, and Barsky (1999) designed a task in which participants
were asked to identify perceptually degraded words. Hypochondriacal patients did
not show a bias toward identifying health versus nonhealth words in this task. How-
ever, individuals high in hypochondriasis did exhibit better recall for health versus
nonhealth words in a free-recall task. Similarly, Pauli and Alpers (2002) found that
somatoform pain patients with comorbid hypochondriasis exhibit a memory bias in
favor of recalling pain words embedded in a list of other negative, positive, or neutral
words; somatoform pain patients without the comorbidity did not exhibit the bias.
However, these types of tasks (identifying words and free recall) involve a conscious
processing component—individuals may choose not to identify a word if they are
unsure about its identity (indeed, Brown et al., 1999, noted that many participants
were reluctant to make guesses) and individuals may choose not to list a word in
634 Williams

a free-recall task if uncertain about its prior presentation. Thus, self-presentational

style differences may influence the outcome of such tasks.
Preferred methods for examining preferential attention allocation to particular
content domains utilize tasks that do not rely on self-report (e.g., reaction time).
In particular, one approach is to examine interference (i.e., disruption in perfor-
mance because of selective attention); another approach examines facilitation (i.e.,
speeded performance because of selective attention). In a recent example of such
an approach, Lecci and Cohen (2002) examined attention to health-relevant infor-
mation in relation to components of self-reported hypochondriacal tendencies. In
a variation on the modified Stroop task, individuals scoring high in somatosen-
sory amplification showed interference in color naming health-related words, but
only under illness concern activation (Lecci & Cohen, 2002). Additionally, Owens,
Asmundson, Hadjistavropoulos, and Owens (2004) found that individuals high in
health anxiety (assessed using the Illness Attitudes Scale [IAS], a commonly used
measure of hypochondriacal tendencies; Kellner, Abbott, Winslow, & Pathak, 1987)
showed selective attention to illness-related threat information in comparison to
non-illness-related threat information in a modified Stroop task. These findings pro-
vide evidence for preferential attention allocation to health-relevant information in
individuals with hypochondriacal tendencies and validate the often-suggested no-
tion that hypochondriacal cognitions are “triggered” by health-related events. Lecci
and Cohen (2002) suggest that individuals with hypochondriacal tendencies may
frequently put themselves in situations that activate illness concern, such as fre-
quent medical consultations or seeking out medical information. Adopting health-
focused goals may also serve the function of activating health concern (Karoly &
Lecci, 1993; Lecci, Karoly, Ruehlman, & Lanyon, 1996). Thus, the behaviors that
some have termed avoidance (i.e., intended to alleviate anxiety) may in fact be
self-perpetuating.
Selective attention to health-relevant information has also been demonstrated
in a paradigm examining facilitation. Hitchcock and Mathews (1992) found that
individuals with hypochondriacal tendencies were faster to correctly recognize illness
words than neutral words in a task examining decision speeds for recognition of words
presented prior in either illness threat, social threat, or neutral sentences. Speeded
performance on this task suggests that individuals with hypochondriacal tendencies
have selective attention toward illness-related threat.

Risk Perception and Misinterpretation of Benign Symptoms

In addition to a hypothesized selective attention toward health-relevant infor-
mation, individuals with health anxiety and hypochondriasis are thought to engage
in catastrophic interpretations of benign symptoms. In examining this, Hitchcock
and Mathews (1992) found that endorsement of catastrophic thoughts and, to a
lesser extent, emotional thoughts in relation to ambiguous situations involving bod-
ily sensations related to total score on IAS. Extreme groups on IAS showed bias
for both social and illness threat in a sentence-recognition task designed to examine
interpretations, suggesting that there may be interpersonal factors related to clinical
hypochondriasis.
Health Anxiety and Hypochondriasis 635

Several studies have examined both catastrophic interpretation of symptoms

and risk perception for various negative health outcomes. For example, Haenen, de
Jong, Schmidt, Stevens, and Visser (2000) found that hypochondriacal patients (based
upon structured diagnostic interviews) make higher estimates of negative outcomes
in judging ambiguous health versus nonhealth scripts than do nonhypochondriacal
controls. Similarly, individuals high in health anxiety have been found to rate them-
selves to be at greater risk of medical complications than those lower in health anxiety
(Hadjistavropoulos, Craig, & Hadjistavropoulos, 1998).
Marcus (1999) found that hypochondriasis was related to attributing ambiguous
symptoms to catastrophic illness and overestimating the frequency of such illnesses.
Using the same Symptoms and Outcomes Scale, Marcus and Church (2003) found
that individuals high in hypochondriacal tendencies provided higher probability rat-
ings that a variety of symptoms were indicative of a serious illness, even when a broad
range of mood and anxiety symptoms were controlled. Thus, whereas negative af-
fectivity is significantly correlated with hypochondriasis, dysfunctional beliefs about
illness appear to be unique to hypochondriasis.
There is compelling evidence that most individuals demonstrate an optimistic
bias with respect to judging their own health risk compared to others their same
age (Weinstein, 1984, 1987). Barsky et al. (2001) examined the risk perceptions for
both health- and non-health-related future negative events by hypochondriacal pa-
tients and nonhypochondriacal medical patients. Although hypochondriacal patients
exhibited an optimistic bias with respect to estimating illness risk, they had higher
total risk scores than did nonhypochondriacal individuals. This effect was seen only
for disease risk—hypochondriacal patients did not score higher on estimates of ac-
cidents or criminal victimization. The authors note that it could be argued that the
“cognitive distortion” observed in their data was with the nonhypochondriacal pa-
tients because their estimates were unrealistically optimistic, whereas those of the
hypochondriacal patients, though still optimistic, were closer to reality. These find-
ings suggest that further research is needed to fully characterize the accuracy versus
bias of risk perception processes in hypochondriasis.

Cognitive Response to Medical Reassurance

A final area of cognitive research relevant to hypochondriasis is the examination
of the cognitive processing of medical reassurance by individuals with hypochondri-
acal tendencies. Although the study was not conducted in a medical setting, Haenan
et al. (2000) found that hypochondriacal patients did not differ from controls with
respect to the effects of providing reassuring information on estimations of neg-
ative outcomes. The authors concluded that their results are not consistent with
the notion that individuals high in hypochondriasis are immune to reassuring in-
formation. A limitation to the study, however, is that the reassurance was linked
to hypothetical vignettes. In a more personal health context, there is evidence that
patients high in health anxiety recall medical reassurance as “less certain” in ruling
out a serious health problem compared to patients low in health anxiety when ob-
taining medical consultation for unexplained symptoms (Lucock, White, Peake, &
Morley, 1998). These findings provide preliminary evidence that individuals with
636 Williams

hypochondriacal tendencies may process information provided to them by health

care providers differently. This is clearly an important area for future cognitive
research.
In summary, a convincing body of research has demonstrated the central role
of cognitive processes in health anxiety and hypochondriasis. Nonetheless, a full ac-
count of hypochondriacal tendencies must incorporate individual difference factors
(e.g., temperament/personality) and social processes (e.g., social learning, attach-
ment) that are related to health and illness cognition and associated negative emo-
tional states. To develop an integrative, cognitive model of hypochondriasis, person-
ality and social learning processes that have been linked to health anxiety are now
considered.

Temperament, Personality, and Hypochondriasis

Of the traditional personality variables (e.g., Five Factor Model; McCrae &
Costa, 1987), neuroticism (N; and highly related variables such as Negative Affec-
tivity, Trait Anxiety) has been the most consistently associated with health anxiety
and hypochondriasis (McClure & Lilienfeld, 2001). Neuroticism is associated with
poorer self-assessed health in general (e.g., Watson & Pennebaker, 1989) and is sig-
nificantly associated with hypochondriacal concerns (Cox, Borger, Asmundson, &
Taylor, 2000; Ferguson, 2000). Moreover, individuals diagnosed with hypochondriasis
report higher levels of neuroticism (Noyes et al., 1994).
Although Cox et al. (2000) report significant negative correlations between
hypochondriasis and extraversion and concientiousness, these relations drop to non-
significant levels when the effects of neuroticism are controlled. Noyes et al. (2003)
found similar effects. Thus, of the five factors, neuroticism shows the most robust
relation to hypochondriacal tendencies. However, as described below, other per-
sonality dimensions (particularly conscientiousness) may be important moderators
in the relation between neuroticism and the development of health anxiety and
hypochondriasis.
Why is neuroticism related to health anxiety? Relevant to this question is the lit-
erature on the development and maintenance of anxiety, in general. In recent years,
neuroticism has been characterized in terms of the Behavioral Inhibition System
(BIS), characterized by sensitivity to signals of punishment and nonreward, within
Gray’s (1982, 1987) model of behavioral motivation. Anxiety is thought to be the
manifestation of a highly active BIS. In contrast, the Behavioral Activation Sys-
tem (BAS) is associated with sensitivity to reward and removal of punishment and
has been linked to the personality dimension extraversion (or trait positive affect
[PA]). The affective dimensions of neuroticism/negative affect (NA) and extraver-
sion/PA (and the hypothesized behavioral motivational systems underlying them)
have formed the basis for many current models of anxiety and depression. For exam-
ple, Clark and Watson’s (1991) tripartite model of anxiety and depression considers
anxiety to be a manifestation of high NA, whereas depression is a manifestation of
high NA and low PA.
From a cognitive perspective, sensitivity to punishment cues among individ-
uals high in neuroticism may be assessed using information-processing paradigms
Health Anxiety and Hypochondriasis 637

that examine attention to threat. Indeed, there is now considerable evidence that
vulnerability to anxiety is associated with selective processing of threatening stimuli
(Mathews & MacLeod, 1994). Attentional biases inherent in clinical anxiety and trait
anxiety have been demonstrated using the emotional Stroop task (e.g., MacLeod &
Rutherford, 1992; Mathews & MacLeod, 1985; Mogg, Mathews, & Weinman, 1989)
and the dot probe task (MacLeod & Mathews, 1988; MacLeod, Mathews, & Tata,
1986; Mogg, Mathews, & Eysenck, 1992).
Despite the evidence that the anxiety-relevant personality dimensions as well
as clinical anxiety states are related to selective attention to threat, several recent re-
views have noted discrepancies in the literature (Lonigan, Vasey, Phillips, & Hazen,
2004; Mathews & MacLeod, 1994; Matthews & Mackintosh, 1998). In particular,
some studies have found that state anxiety may lead to shifting attention away from
threatening stimuli (e.g., Mathews & Sabastian, 1993). It has been suggested that stim-
uli presented for sufficiently long intervals (i.e., allowing conscious apprehension of
the nature of the stimuli) may allow for effortful redirection of attention whereas
stimuli presented for very short intervals (i.e., participants cannot consciously iden-
tify the stimuli) do not allow for such an attentional shift (Mogg, Kentish, & Bradley,
1993). Lonigan et al. (2004) provide an integrative account of these findings by con-
sidering the temperament dimension termed effortful control (EC). Whereas neu-
roticism/NA and extraversion/PA are considered “reactive” motivational traits, EC is
related to the ability to engage executive control processes and “override” reactivity
(e.g., Posner & Rothbart, 2000). The adult personality dimension of the Five Fac-
tor Model most closely aligned to this temperament factor is conscientiousness (or
in Tellegan’s [1985] model, constraint). Thus, conscientiousness/EC may moderate
the effects of neuroticism/NA on selective attention to threat cues when conscious
processing of stimuli is possible.
With respect to understanding the cognitive processes underlying health anx-
iety and hypochondriasis, the extant literature on selective processing of threat in
anxiety has obvious applicability. Threat may take a variety of forms and, arguably,
assessing personal health status can be a threatening endeavor. Thus, within the cog-
nition and anxiety framework described above, bodily sensations and illness-related
information (e.g., from the media, from health care providers) can be conceptual-
ized as threat stimuli. Individuals high in neuroticism are vulnerable to selectively
attend to such threat cues. Conscientiousness (EC) may moderate this tendency.
This hypothesized interaction between N and C may be particularly useful in under-
standing why some high N individuals develop chronic hypochondriacal tendencies
and others do not. That is, high N/high C individuals may be able to successfully
shift attention away from threatening health stimuli over time, whereas (hypothet-
ically) high N/low C individuals are not able to regulate in this fashion. Although
this framework allows for the consideration of why some high N individuals develop
clinical levels of anxiety, it does not account for the specificity of threat focus. An
integrative cognitive model of health anxiety and hypochodriasis must be able to
account for why some high N individuals develop anxiety disorders, others develop
hypochondriasis, and some develop neither or both. To inform this issue, social and
developmental factors that have been related to hypochondriacal tendencies are now
considered.
638 Williams

Social Developmental Factors and Hypochondriasis

Although longitudinal research on the development of hypochondriasis has been
scarce, considerable attention has been devoted to the interpersonal histories of indi-
viduals with hypochondriacal tendencies. For instance, significant correlations have
been found between measures of childhood separation anxiety and hypochondriasis
(Noyes et al., 2002). The authors suggest that separation anxiety may be considered
to be an indirect measure of insecure attachment and is hypothesized to be a vul-
nerability factor for the development of psychopathology in adulthood. The role
of temperament and personality in these findings is unclear; however, Noyes et al.
(2003) report that fearful attachment style is positively correlated with neuroticism
and when both are included in regression analyses, neuroticism remains the strongest
predictor of hypochondriasis. These findings suggest that attachment style is a more
general vulnerability factor (i.e., not specific to hypochondriasis), although it may
have direct bearing on the interpersonal expression (e.g., adverse interactions with
medical professionals) of hypochondriasis. Noyes et al. (2003) couch this in terms
of an interpersonal model of hypochondriasis that posits that, because of negative
attachment histories, these individuals are unskilled at getting emotional needs met
in appropriate ways and, hence, use hypochondriacal complaints to garner emotional
support from others.
Whereas attachment style may be a risk factor for a variety of forms of psy-
chopathology, it has been suggested that childhood experience with significant illness
is a specific antecedent of hypochondriasis (Barsky, Wool, Barnett, & Cleary, 1994).
For example, Noyes et al. (2002) found that when patients with hypochondriasis
were presented with a list of traumatic childhood events, they were more likely to
report having a serious illness or injury during childhood, a parent with a hazardous
occupation, a family member with an alcohol or drug problem, chronic illness in a
family member, and serious illness in a close friend than were nonhypochondria-
cal individuals. Thus, the events that differentiated hypochondriacal patients from
nonhypochondriacal patients all involved health-related threat. The authors also
found modest correlations between hypochondriasis and measures of parental over-
concern in relation to illness. Indeed, individuals with hypochondriacal concerns
report more instrumental and vicarious learning experiences around bodily symp-
toms (Watts & Stewart, 2000). Further, in studies of maternal factors that influence
parental response to children’s illness, mothers’ own self-assessed health is a stronger
predictor of more extreme caretaking behavior and encouragement of sick role be-
havior in their children than is maternal neuroticism (Scalzo, Williams, & Holmbeck,
in press).
In summary, the literature suggests that there are illness-specific social and en-
vironmental factors that are important in the development of health anxiety. Unfor-
tunately, the supporting studies are retrospective. It is unclear whether or not the
reported histories represent selective memory on the part of adults with hypochon-
driasis or if these individuals actually had more health-related threat experiences
as children. The answer to this question awaits prospective empirical study and/or
examination in individuals whose medical history is documented (e.g., adult survivors
of childhood cancer). Of additional interest is whether or not personal or vicari-
ous illness experiences in adulthood are sufficient to trigger hypochondriasis in the
Health Anxiety and Hypochondriasis 639

Fig. 1. Proposed cognitive developmental model of the etiology and maintenance of health anxiety
and hypochondriasis.

absence of the types of childhood experiences described above. In other words, it

is clear that personal experience of a health scare or knowing someone with a seri-
ous illness as an adult may be a proximal precipitating factor for health anxiety or
hypochondriasis; whether or not a learning history related to illness threat moderates
this relationship awaits empirical investigation.

A Cognitive Developmental Model of Health Anxiety and Hypochondriasis

The research literature investigating the cognitive processes underlying health
anxiety and hypochondriasis provides compelling evidence for the applicability of a
cognitive model of the development and maintenance of these phenomena. A pro-
posed model is depicted in Fig. 1. Given the shared vulnerability factors between
hypochondriasis and other anxiety disorders, a comprehensive model must account
for the development of health anxiety and hypochondriasis per se. Clearly, neuroti-
cism is a vulnerability factor for health anxiety, but it is also related to other forms of
psychopathology (especially anxiety disorders, depression, and insomnia). The cog-
nition and anxiety literature suggests that neuroticism can be characterized as the
behavioral manifestation of a temperament-related propensity to selectively attend
to punishment cues or threat. Threat can take a variety of forms and for individuals
with hypochondriasis the hypothesis is that bodily symptoms and health status have
become threatening targets. Consideration of the social learning and environmen-
tal histories of individuals with hypochondriasis, as well as the illness socialization
research in general, suggests that prior experience with illness during childhood
may be an important moderating variable. That is, illness- and injury-related events
(either vicarious or directly experienced) and interpersonal learning experiences
involving illness (either modeled or directly experienced) represent the specific social
developmental factors that may make vulnerable children (i.e., high N) begin a tra-
jectory toward health anxiety (vs. other forms of psychopathology). Moreover, prior
social learning experiences that lead to general attachment difficulties, as well as
illness-specific learning experiences (e.g., parental overconcern about illness), may
640 Williams

contribute to the interpersonal style thought to maintain hypochondriasis once it has

begun.
A comprehensive model of hypochondriasis must also account for why some in-
dividuals exhibit hypochondriacal tendencies, but never develop clinical hypochon-
driasis. Many individuals experience periodic concern over the status of their health,
but these episodes are short-lived and respond to medical reassurance. By defini-
tion, the diagnosis of hypochondriasis contains the feature of being unresponsive to
medical reassurance. From a cognitive perspective, why is it that some individuals,
even if prone to health anxiety, are able to shift their attention away from this per-
ceived threat? The cognition and anxiety literature suggests that consideration of the
temperamental factor EC and the related personality factor conscientiousness might
be useful. This factor is related to ability to self-regulate and allows an individual to
overcome reactive motivational tendencies (i.e., BIS, BAS) when necessary. Thus, in-
dividuals high in neuroticism and low in EC would be especially prone to persistent
anxiety (Lonigan et al., 2004). Considering temperamental/personality and social
developmental factors, the individual most at risk for the development of hypochon-
driasis is someone high in neuroticism, low in EC, and who has a history in which
there were significant illness-related events and/or for whom there was significant
parental overconcern and attention to bodily symptoms.
This hypothesized cognitive model can also be expanded to consider comorbidi-
ties between hypochondriasis and other affective disorders. Over time, an individual
with the propensities described above may also meet criteria for other anxiety dis-
orders, depending on where the focus of threat is at any given time. As Salkovskis
and Warwick (2001) note, “critical precipitating incidents” such as a personal health
scare or intensive media coverage of a disease can trigger health anxiety. Thus, even
if an individual has a history of panic disorder or generalized anxiety, the focus of
attention to threat may shift to health (if the vulnerability conditions described above
are present).
With respect to comorbidity with depression, the literature suggests that low
positive affect in conjunction with high negative affect characterize depressed mood.
Therefore, extraversion–introversion (or trait positive affectivity) may moderate the
relation between hypochondriasis and depression. Introverted individuals may be
particularly prone to withdraw under stressful circumstances leading to both the
removal of positive (i.e., rewarding) experiences and diminishing opportunities to
receive social support (which may buffer negative experiences). Because individuals
with hypochondriasis have the presumed negative cognitive bias with regard to their
health and may become functionally disabled as a result, they may be particularly
prone to co-occurring depression. Indeed, poor self-assessed health is an antecedent
risk factor for depressive symptoms (Williams, Colder, Richards, & Scalzo, 2002)
and the hypothesized mediating factor is functional disability (Lewinsohn, Seeley,
Hibbard, Rohde, & Sack, 1996).

Future Directions
Many of the hypothesized relationships depicted in Fig. 1 await empirical test-
ing. In particular, longitudinal developmental research examining the etiological
Health Anxiety and Hypochondriasis 641

pathways to health anxiety and hypochondriasis is sorely needed. Of particular in-

terest are the factors that are specific to the development of hypochondriasis versus
other forms of psychopathology. The factors that influence the transition from sub-
clinical health anxiety to clinical hypochondriasis need to be determined. From a
developmental perspective, although health anxiety may be evident in childhood,
hypochondriacal tendencies are likely first observed in early adulthood when indi-
viduals begin making health-related self-care decisions independent from parents.
Thus, research examining the transition between adolescence and early adulthood
may be especially relevant to understanding the behavioral aspects of hypochondri-
acal tendencies (i.e., overuse of health services, functional disability in relation to
poor self-assessed health).
Although prior research has generally supported the hypothesized cognitive un-
derpinnings of hypochondriasis, this area of inquiry could benefit from expanding the
experimental paradigms. Because of the limitations in using self-report measures of
cognition outlined above, the use of experimental information-processing paradigms
that do not rely on self-report seems essential. Given the conceptual overlap with
the study of cognition and anxiety, much can be learned from the relatively greater
attention to such methodologies in that line of research.
The interface of cognition, emotion, and physiology in hypochondriasis is still
insufficiently understood. A variety of information-processing paradigms may offer
interesting approaches to more directly examine how health-related information
is (or is not) associated with emotional information. For example, the effects of
priming health-related information processing with emotion information can inform
how these factors are represented cognitively.
The research on cognition, emotion, and hypochondriasis suggests that a cog-
nitive approach to the treatment should be particularly effective. Indeed, Looper
and Kirmayer (2002) report that CBT has been shown to effectively treat hypochon-
driasis. Moreover, consistent with the health-specific nature of the behavior and
cognition inherent in hypochondriasis, symptom-focused strategies appear to offer
greater benefit than more general stress management CBT approaches. Clearly, how-
ever, there is much to be learned about how to effectively treat clinical hypochon-
driasis once it has begun. Additionally, very little attention has been devoted to
secondary prevention (i.e., early identification) of hypochondriasis. The proposed
cognitive model suggests that there may be multiple points of both intervention
and, possibly, prevention of hypochondriasis. It may well be that some of the de-
velopments in the treatment of anxiety disorders could be applied to hypochondri-
asis. For example, there is preliminary evidence that attentional bias to threat can
be altered using an information-processing approach (Hazen, Vasey, & Schmidt,
2001; MacLeod, Rutherford, Campbell, Ebsworthy, & Holker, 2002). Such research
suggests intriguing possible applications to the treatment of health anxiety and
hypochondriasis.
Beyond the specifics of understanding the psychopathology of self-assessed
health, is the need to better characterize what is optimal in terms of attention to health
and concomitant health self-regulatory behavior. Overattention to symptoms and
overuse of health services has received the most attention and, unfortunately, many
premature assumptions have been made about the cognitive processes underlying
642 Williams

such phenomena. For example, it is typically assumed that individuals who report
more symptoms are “distorting” or are “biased.” In truth, we know very little about
the accuracy of health- and illness-related information processing. It may well be that
a relative lack of attention to health-relevant information is equally if not more dam-
aging (and costly) in the long run. Thus, it seems imperative to continue to construct
a cognitive model of self-assessed health that can be used as a template to inform our
understanding of when and why making judgments about health becomes a source
of distress and dysfunction.

REFERENCES

American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed.,
text rev.). Washington, DC: Author.
Barsky, A. J., Coeytaux, Remy R., Sarnie, M. K., & Cleary, P. D. (1993). Hypochondriacal patients’ beliefs
about good health. American Journal of Psychiatry, 150, 1085–1089.
Barsky, A. J., Ahern, D. K., Bailey, E. D., Saintfort, R., Liu, E. B., & Peekna, H. M. (2001). Hypochondriacal
patients’ appraisal of health and physical risks. American Journal of Psychiatry, 158, 783–787.
Barsky, A. J., Barnett, M. C., & Cleary, P. D. (1994). Hypochondriasis and panic disorder: Boundary and
overlap. Archives of General Psychiatry, 51, 918–925.
Barsky, A. J., & Klerman, G. L. (1983). Overview: Hypochondriasis, bodily complaints, and somatic styles.
American Journal of Psychiatry, 140, 273–283.
Barsky, A. J., Wool, C., Barnett, M. C., & Cleary, P. D. (1994). Histories of childhood trauma in adult
hypochondriacal patients. American Journal of Psychiatry, 151, 397–401.
Barsky, A. J., Wyshek, G., & Klerman, G. L. (1990). Transient hypochondriasis. Archives of General
Psychiatry, 47, 746–752.
Barsky, A. J., Wyshek, G., Klerman, G. L., & Latham, K. S. (1990). The prevalence of hypochondriasis in
medical outpatients. Social Psychiatry and Psychiatric Epidemiology, 25, 89–94.
Barsky, A. J., Wyshek, G., Latham, K. S., & Klerman, G. L. (1990). The Somatosensory Amplification
Scale and its relationship to hypochondriasis. Journal of Psychiatry Research, 24, 323–334.
Brown, H. D., Kosslyn, S. M., Delamater, B., Fama, J., & Barsky, A. J. (1999). Perceptual and memory biases
for health-related information in hypochondriacal individuals. Journal of Psychosomatic Research,
47, 67–78.
Cioffi, D. (1991). Beyond attentional strategies: A cognitive–perceptual model of somatic interpretation.
Psychological Bulletin, 109, 25–41.
Clark, L. A., & Watson, D. (1991). Tripartite model of anxiety and depression: Psychometric evidence and
taxonomic implications. Journal of Abnormal Psychology, 100, 316–336.
Cloninger, C. R., Sigvardsson, S., von Knorring, A., & Behman, M. (1984). An adoption study of somato-
form disorders. Archives of General Psychiatry, 44, 573–588.
Cox, B. J., Borger, S. C., Asmundson, G. J. G., & Taylor, S. (2000). Dimensions of hypochondriasis and the
five-factor model of personality. Personality and Individual Differences, 29, 99–108.
Escobar, J. I., Gara, M., Waitzkin, H., Cohen Silver, R., Holman, A., & Comptom, W. (1998). DSM-IV
hypochondriasis in primary care. General Hospital Psychiatry, 20, 155–159.
Fava, G. A., Grandi, S., Saviotti, F. M., & Conti, S. (1990). Hypochondriasis with panic attacks. Psychoso-
matics, 31, 351–353.
Ferguson, E. (2000). Hypochondriacal concerns and the five factor model of personality. Journal of Per-
sonality, 68, 705–724.
Frick, P. J., Kuper, K., Silverthorn, P., & Cotter, M. (1995). Antisocial behavior, somatization, and sensation-
seeking behavior in mothers of clinic-referred children. Journal of the American Academy of Child
and Adolescent Psychiatry, 34, 805–812.
Gray, J. A. (1982). The neuropsychology of anxiety: An enquiry into the functions of the septo-hippocampal
system. New York: Oxford University Press.
Gray, J. A. (1987). The psychology of fear and stress. New York: Cambridge University Press.
Hadjistavropoulos, H. D., Craig, K. D., & Hadjistavropoulos, T. (1998). Cognitive and behavioral responses
to illness information: The role of health anxiety. Behaviour Research and Therapy, 36, 149–164.
Haenen, M. A., de Jong, P. J., Schmidt, A. J. M., Stevens, S., & Visser, L. (2000). Hypochondriacs’ estimation
of negative outcomes: Domain-specificity and responsiveness to reassuring and alarming information.
Behaviour Research and Therapy, 38, 819–833.
Health Anxiety and Hypochondriasis 643

Hazen, R. A., Vasey, M. W., & Schmidt, N. B. (2001, November). Attentional retraining for chronic worry.
Poster presented at the 35th annual meeting of the Association for the Advancement of Behavior
Therapy, Philadelphia.
Hitchcock, P. B., & Mathews, A. (1992). Interpretation of bodily symptoms in hypochondriasis. Behaviour
Research and Therapy, 30, 223–234.
Idler, E. L., & Benyamini, Y. (1997). Self-rated health and mortality: A review of twenty-seven community
studies. Journal of Health and Social Behavior, 38, 21–37.
Karoly, P., & Lecci, L. (1993). Hypochondriasis and somatization in college women: A personal projects
analysis. Health Psychology, 12, 103–109.
Kellner, R., Abbott, P., Winslow, W. W., & Pathak, D. (1987). Fears, beliefs and attitudes in DSM-III
hypochondriasis. Journal of Nervous and Mental Disease, 175, 20–25.
Kirmayer, L. J., & Robbins, J. M. (1991). Three forms of somatization in primary care: Prevalence, co-
occurrence, and sociodemographic characteristics. Journal of Nervous and Mental Disease, 179, 647–
655.
Lecci, L., & Cohen, D. J. (2002). Perceptual consequences of an illness-concern induction and its relation
to hypochondriacal tendencies. Health Psychology, 21, 147–156.
Lecci, L., Karoly, P., Ruehlman, L. S., & Lanyon, R. I. (1996). Goal-relevant dimensions of hypochondri-
acal tendencies and their relation to symptom manifestations and psychological distress. Journal of
Abnormal Psychology, 105, 42–52.
Leventhal, H., Meyer, D., & Nerenz, D. (1980). The common sense representation of illness danger. In S.
Rachman (Ed.), Medical psychology (pp. 7–30). New York: Pergamon.
Lewinsohn, P. M., Seeley, J. R., Hibbard, J., Rohde, P., & Sack, W. H. (1996). Cross-sectional and prospective
relationships between physical morbidity and depression in older adolescents. Journal of the Academy
of Child and Adolescent Psychiatry, 35, 1120–1129.
Lilienfeld, S. O., Van Valkenburg, C., Larntz, K., & Akiskal, H. S. (1986). The relationship of histri-
onic personality disorder to antisocial personality and somatization disorders. American Journal of
Psychiatry, 143, 718–722.
Lonigan, C. J., Vasey, M. W., Phillips, B. M., & Hazen, R. A. (2004). Temperament, anxiety, and the
processing of threat-relevant stimuli. Journal of Clinical Child and Adolescent Psychology, 33, 8–
20.
Looper, K. J., & Kirmayer, L. J. (2001). Hypochondriacal concerns in a community population. Psycho-
logical Medicine, 31, 577–584.
Looper, K. J., & Kirmayer, L. J. (2002). Behavioral medicine approaches to somatoform disorders. Journal
of Consulting and Clinical Psychology, 70, 810–827.
Lucock, M. P., White, C., Peake, M. D., & Morley, S. (1998). Biased perception and recall of reassurance
in medical patients. British Journal of Health Psychology, 3, 237–243.
MacLeod, C., & Mathews, A. (1988). Anxiety and the allocation of attention to threat. Quarterly Journal
of Experimental Psychology: Human Experimental Psychology, 38, 659–670.
MacLeod, C., Mathews, A., & Tata, P. (1986). Attentional bias in emotional disorders. Journal of Abnormal
Psychology, 95, 15–20.
MacLeod, C., & Rutherford, E. (1992). Anxiety and the selective processing of emotional information:
Mediating roles of awareness, trait and state variable, and personal relevance of stimulus materials.
Behaviour Research and Therapy, 30, 479–491.
MacLeod, C., Rutherford, E., Campbell, L., Ebsworthy, G., & Holker, L. (2002). Selective attention and
emotional vulnerability: Assessing the causal basis of their association through the experimental
manipulation of attentional bias. Journal of Abnormal Psychology, 11, 107–123.
Marcus, D. K. (1999). The cognitive–behavioral model of hypochondriasis: Misinformation and triggers.
Journal of Psychosomatic Research, 47, 79–91.
Marcus, D. K., & Church, S. E. (2003). Are dysfunctional beliefs about illness unique to hypochondriasis?
Journal of Psychosomatic Research, 54, 543–547.
Mathews, A., & MacLeod, C. (1985). Selective processing of threat cues in anxiety states. Behaviour
Research and Therapy, 23, 563–569.
Mathews, A., & MacLeod, C. (1994). Cognitive approaches to emotion and emotional disorders. Annual
Review of Psychology, 45, 25–50.
Mathews, A., & Sabastian, S. (1993). Suppression of emotional Stroop effects by fear arousal. Cognition
and Emotion, 7, 517–530.
Matthews, A., & Mackintosh, B. (1998). A cognitive model of selective processing in anxiety. Cognitive
Therapy and Research, 22, 539–560.
McClure, E. B., & Lilienfeld, S. O. (2001). Personality traits and health anxiety. In G. J. G. Asmundson,
S. Taylor, & B. J. Cox (Eds.), Health anxiety: Clinical and research perspectives on hypochondriasis
and related conditions (pp. 65–91). New York: Wiley.
644 Williams

McCrae, R. R., & Costa, P. T. (1987). Validation of the five factor model across instruments and observers.
Journal of Personality and Social Psychology, 49, 710–727.
Mogg, K., Kentish, J., & Bradley, B. P. (1993). Effects of anxiety and awareness on color identification
latencies for emotional words. Behaviour Research and Therapy, 31, 559–567.
Mogg, K., Mathews, A., & Eysenck, M. (1992). Attentional bias to threat in clinical anxiety states. Cognition
and Emotion, 6, 149–159.
Mogg, K., Mathews, A., & Weinman, J. (1989). Selective processing of threat cues in anxiety states: A
replication. Behaviour Research and Therapy, 27, 317–323.
Noyes, R., Jr. (2001). Hypochondriasis: Boundaries and comorbidities. In G. J. G. Asmundson, S. Taylor,
& B. J. Cox (Eds.), Health anxiety: Clinical and research perspectives on hypochondriasis and related
conditions (pp. 132–160). New York: Wiley.
Noyes, R., Jr., Kathol, R. G., Fisher, M. M., Phillips, B. M., Suelzer, M. T., & Woodman, C. L. (1994).
One-year follow-up of medical outpatients with hypochondriasis. Psychosomatics, 35, 533–545.
Noyes, R., Jr., Stuart, S., Langbehn, D. R., Happel, R. L., Longley, S. L., Muller, B. A., et al. (2003). Test
of an interpersonal model of hypochondriasis. Psychosomatic Medicine, 65, 292–300.
Noyes, R., Jr., Stuart, S., Langbehn, D. R., Happel, R. L., Longley, S. L., & Yagla, S. J. (2002). Childhood
antecedents of hypochondriasis. Psychosomatics, 43, 282–289.
Owens, K. M. B., Asmundson, G. J. G., Hadjistavropoulos, T., & Owens, T. J. (2004). Attentional bias
toward illness threat in individuals with elevated health anxiety. Cognitive Therapy and Research, 28,
57–66.
Pauli, P., & Alpers, G. W. (2002). Memory bias in patients with hypochondriasis and somatoform pain
disorder. Journal of Psychosomatic Research, 52, 45–53.
Pennebaker, J. W. (1982). The psychology of physical symptoms. New York: Springer-Verlag.
Posner, M. R., & Rothbart, M. K. (2000). Developing mechanisms of self-regulation. Development and
Psychopathology, 12, 427–441.
Rief, W., Hiller, W., & Margraf, J. (1998). Cognitive aspects of hypochondriasis and the somatization
syndrome. Journal of Abnormal Psychology, 107, 587–595.
Salkovskis, P. M., & Warwick, H. M. C. (2001). Making sense of hypochondriasis: A cognitive theory of
health anxiety. In G. J. G. Asmundson, S. Taylor, & B. J. Cox (Eds.), Health Anxiety: Clinical and
research perspectives on hypochondriasis and related conditions (pp. 46–64). New York: Wiley.
Scalzo, C., Williams, P. G., & Holmbeck, G. N. (in press). Maternal self-assessed health and emotionality
predict maternal response to child illness. Children’s Health Care.
Tellegan, A. (1985). Structures of mood and personality and their relevance to assessing anxiety, with
an emphasis on self-report. In A. H. Tuma & J. D. Maser (Eds.), Anxiety and the anxiety disorders
(pp. 681–706). Hillsdale, NJ: Erlbaum.
Warwick, H. M., & Salkovskis, P. M. (1990). Hypochondriasis. Behaviour Research and Therapy, 28, 105–
117.
Watson, D., & Pennebaker, J. W. (1989). Health complaints, stress, and distress: Exploring the central role
of negative affectivity. Psychological Review, 96, 234–254.
Watts, M. C., & Stewart, S. H. (2000). Anxiety sensitivity mediates the relationships between childhood
learning experiences and elevated hypochondriacal concerns in young adults. Journal of Psychoso-
matic Research, 49, 107–118.
Weinstein, N. D. (1984). Why it won’t happen to me: Perceptions of risk factors and susceptibility. Health
Psychology, 3, 431–457.
Weinstein, N. D. (1987). Unrealistic optimism about susceptibility to health problems: Conclusions from
a community-wide sample. Journal of Behavioral Medicine, 10, 481–500.
Williams, P. G., Colder, C. C., Richards, M. H., & Scalzo, C. A. (2002). The role of self-assessed health in
the relationship between gender and depressive symptoms among adolescents. Journal of Pediatric
Psychology, 27, 509–517.
Williams, P. G., Wasserman, M. S., & Lotto, A. J. (2003). Individual differences in self-assessed health: An
information-processing investigation of health and illness cognition. Health Psychology, 22, 3–11.