Psychiatric comorbidity among patients with hypochondriasis

Psychiatry and Primary Care Recent epidemiologic studies have found that most patients with mental illness are seen exclusively in primary care medicine. These patients often present with medically unexplained somatic symptoms and utilize at least twice as many health care visits as controls. There has been an exponential growth in studies in this interface between primary care and psychiatry in the last 10 years. This special section, edited by Wayne J. Katon, M.D., will publish informative research articles that address primary care-psychiatric issues.

Psychiatric Comorbidity Among Patients with Hypochondriasis Russell Noyes, Jr., M.D., Roger G. Kathol, M.D., Mary M. Fisher, M.A., Brenda M. Phillips, M.D., Michael T. Suelzer, Ph.D., and Catherine L. Woodman, M.D.

Abstract: The purpose of this study was to determine the nature and extent of comorbidity among patients with DSM-111-R hypochondriasis and to examine the relationships between this disorder and coexisting psychiatric illness. For this purpose, patients seen in a general medicine clinic were screened using measures of hypochondriacal attitudes and somatic symptoms. Those scoring above an established cutoff were given a structured diagnostic interview. In this manner, 50 patients who met DSM-111-R criteria for hypochondriasis and 50 age- and sex-matched controls were identified. The presence of other psychiatric disorders (current and past) was determined by means of the same diagnostic interview. More hypochondriacal subjects (62.0%) had lifetime comorbidity than did controls (30.0%). Major depression, the most frequent comorbid disturbance, was usually current and most often had an onset after that of hypochondriasis. Panic disorder with agoraphobia, the most frequent anxiety disorder, was also current but often began before or at the same time as hypochondriasis. Few subjects met criteria for somatization disorder but a third qualified for a subsyndromal form of this disorder. The data show that, in medical outpatients with hypochondriasis, mood and anxiety disorders frequently coexist. This comorbidity is subject to varying interpretations including overlap of symptom criteria, treatment-seeking bias, and the possibility that hypochondriasis predisposes to or causes the comorbid disorder, as seems likely in the case of depression. In some instances hypochondriasis may be an associated feature of another illness. Departments of Psychiatry and Internal of Iowa, College of Medicine, Iowa City, Address reprint requests to: Dr. Noyes, University of Iowa Hospitals and Clinics,

78 ISSN 0163-8343/94/$7.00

Medicine, University Iowa 200 Hawkins Drive, Iowa City, IA 52242.

Introduction Hypochondriasis is defined as a preoccupation with the fear of having, or the belief that one has, a serious disease, based on the person’s interpretation of physical signs or symptoms [l]. The DSM-III-R criteria further specify that there is no physical explanation for symptoms but that fear of disease persists despite reassurance. Although the classification lists hypochondriasis as a separate disorder, its status has not yet been firmly established. The disturbance is frequently associated with other psychiatric illnesses and may represent an associated feature of anxiety or depressive disorders. Also, hypochondriasis shares clinical manifestations with other somatoform disturbances and may be difficult to distinguish from somatization disorder. According to Kenyon [2], who studied psychiatric patients, hypochondriasis is almost always secondary to other psychiatric illnesses. On the other hand, Pilowsky [3] observed important differences in the demographic and clinical characteristics of patients with primary and secondary hypochondriasis and concluded that hypochondriasis has independent status. The proportion of patients identified as having hypochondriasis, in the absence of other psychiatric illness, has ranged from almost none to as high as 58% [2-51. However, most studies have involved General Hospital Psychiatry 16, 78-87, I!394 0 1994 Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010

PsychiatricComorbidityand Hypochondriasis psychiatric patients among whom secondary hypochondriasis is prevalent. Surveys of clinical populations have shown that the symptoms of anxiety, depression, and hypochondriasis commonly co-occur. With respect to anxiety, Kellner et al. [6] observed that patients with DSM-III hypochondriasis had higher ratings of anxiety than matched psychiatric outpatients without hypochondriasis. Also, Noyes et al. [7] reported that anxiety and phobic symptoms were prominent in a small series of patients with illness phobia, a subtype of hypochondriasis. In their examination of DSM-III-R comorbidity among hypochondriacal medical outpatients, Barsky et al. [8] found that 85.7% had at least one coexisting anxiety disorder compared with 35.6% of controls. The majority met criteria for generalized anxiety disorder (71.4%) but a substantial proportion reported phobias, including agoraphobia (42.9%), and a smaller percentage had panic disorder (16.7%). Also, hypochondriacal features have been observed in patients with panic and agoraphobia [9-111. In patients with panic disorder with agoraphobia, Noyes et al. [12] found hypochondriasis scores equal to those of hypochondriacal psychiatric patients studied by Pilowsky [13]. Similarly, Fava et al. [14] and Starcevic et al. [15] reported significant hypochondriacal concerns in panic and agoraphobic patients. Strong associations between hypochondriasis and depression have also been observed [6,16]. For example, in medical outpatients, Barsky et al. [17] found depressive symptoms highly correlated (r = 0.57) with hypochondriacal attitudes. Kenyon [2] found that 82% of psychiatric inpatients with secondary hypochondriasis had primary depression. Also, two-thirds of hypochondriacal outpatients scored in the depressed range on the Zung Depression Scale [18]. Most recently, Barsky et al. [8] identified coexisting mood disorders in over half of hypochondriacal outpatients they studied using the Structured Clinical Interview for DSM-III-R. Looked at another way, studies of depressed patients have identified hypochondriacal features in a significant proportion [19-211. Fava et al. [22] found that depressed medical patients had higher levels of hypochondriasis than nondepressed patients. In their population, the correlation between depression and hypochondriasis was 0.47. The distinction between hypochondriasis and somatization disorder has received little study. Hypochondriasis is defined in DSM-III-R as a set of health worries or attitudes whereas somatization

disorder is defined in terms of medical history [l]. Both are characterized, according to these criteria, by multiple somatic symptoms. DSM-III [23] excluded hypochondriasis, once a diagnosis of somatization disorder had been made, but this hierarchical rule, along with most others, was eliminated from DSM-III-R [l]. Few studies have examined the co-occurrence of these disorders, but Tyrer et al. [24] found that 39% of hypochondriacal psychiatric patients met criteria for Briquet’s syndrome and Oxman and Barrett [25] found hypochondriacal symptoms in 38% of family practice patients with somatization disorder. Similarly, Kirmayer and Robbins [26] found that 32% of hypochondriacal family medicine patients met criteria for subsyndromal somatization disorder. Most recently, Barsky et al. [8] found the prevalence of DSM-III somatization disorder to be 21% in a sample of medical outpatients who met criteria for hypochondriasis. The purpose of this study was to examine the comorbidity in a representative sample of patients from a general medicine clinic who met DSM-III-R criteria for hypochondriasis. Our intent in this preliminary study was to explore the relationships that exist between hypochondriasis and coexisting disorders.

Methods Patients seen in the Medicine Clinic of the University of Iowa Hospitals and Clinics between October 1990 and June 1991 were screened for hypochondriasis by means of a self-administered questionnaire which consisted of a modified version of the Whiteley Index [13] and the Somatic Symptom Inventory [27,28]. The cutoff established by Barsky was shown to have a sensitivity of 90% and a specificity of 67% in identifying medical clinic patients with hypochondriasis (Barsky, personal communication, 199O).l Patients who scored above this cutoff were contacted by an experienced psychiatric interviewer (MMF) and, while attending the clinic, were given a structured interview and questionnaires in order to establish the diagnosis of hypochondriasis, identify coexisting psychiatric disorders, and assess the symptoms and illness characteristics of hypochondriasis. Control subjects were randomly selected patients attending the same diagnostic clinic who scored below the established ‘Whiteley Index x 5.6 + Somatic Symptom

Inventory

= 290

or more

79

R. Noyes,

Jr. et al.

cutoff. These patients, who were matched for age (within 5 years) and sex with hypochondriasis patients, were personally interviewed in the same manner. A total of 1406 new patients were seen in the Medicine Clinic of which 1182 (84.1%) completed screening questionnaires. Of these, 163 (13.8%) scored above the cutoff for hypochondriasis and were approached about further participation. Of the 95 who agreed to be interviewed, 54 met criteria for hypochondriasis, 34 did not meet these criteria, and 7 gave incomplete information. Thirtysix of the remaining patients who scored above the cutoff refused to be interviewed and 32 were unavailable or failed to take part for other reasons. Four hypochondriasis subjects met criteria for somatization disorder and were excluded from most analyses to create a more homogeneous sample. As they became available, subjects who scored below the established cutoff and who matched hypochondriacal subjects in terms of age and sex were approached about participation in the study. Of the 73 who were approached in this manner, 50 were interviewed, 17 refused to participate, and 6 failed to take part for other reasons. Three of the latter met criteria for hypochondriasis. Patients were asked to complete the screening questionnaire as soon as they arrived for their appointment. Once the questionnaire had been scored, eligible patients were approached for interviews lasting about 1% hours while they awaited tests and examinations. Where necessary, interviews were completed by telephone. To begin with, the Structured Clinical Interview for DSMIII-R (SCID) was administered [29,30]. This widely used instrument has shown reliability comparable to that of other major diagnostic instruments [31]. The section on hypochondria&s was developed by Barsky et al. 1321 and shown by these authors to have good reliability and validity. All subjects were required to meet DSM-III-R criteria for hypochondriasis and any controls who met these criteria were excluded. Subjects were included who had hypochondriasis of less than 6 months duration. The presence or absence of coexisting disorders was also determined using the SCID (except for adjustment disorders and eating disorders). When anxiety, mood, or substance use disorders were identified, an effort was made to determine whether they had been present during the past 6 months and what the temporal relationship between any coexisting disorders and hypochondriasis had been. Subjects were screened for

80

somatization disorder using an abbreviated list of symptoms recommended by Swartz et al. [33]. A structured interview, designed to obtain information regarding demographic and illness characteristics, was developed for this study. This interview contained storable items covering the history of hypochondriasis. These included age of onset, duration of hypochondriasis, severity of hypochondriasis (O-absent to 4-extreme), and impairment in functioning related to hypochondriasis (Onone to 4-extreme). Social class was determined according to the Hollingshead Two-Factor Index of Social Position [34]. A series of self-rated questionnaires designed to obtain information about symptoms, functional status, and personality functioning were also administered. Patients completed the anxiety, depression, and somatization subscales of the Symptom Checklist-90 (SCL-90) 1281. Symptoms listed in the SCL-90 were rated for the past week on ordinal scales (l-not at all to 5-extremely). Overall functioning was assessed by means of the Global Assessment Scale [35]. Personality functioning was assessed by means of the Impairment/Distress Scale of the Personality Diagnostic Questionnaire (PDQ) [36,37]. This subscale, made up of five items, correlates highly (r = 0.75) with the total score for the 163-item PDQ. We modified this subscale by substituting 5-point, ordinal scales (l-not at all to 5-a great deal) for the dichotomous response categories used in the PDQ. The scale provides a measure of impairment in personality function without indicating the specific nature of the disturbance. Subjects also completed the Eysenck Personality Inventory (EPI) which yields scores on two dimensions of personality, Neuroticism and Extroversion [38]. Sensitivity to unpleasant but nonpathological bodily sensations and environmental stimuli was measured by means of an ll-item Somatosensory Amplification Scale developed by Barsky et al. 1391 Responses to items on the scale are scored on a 5-point ordinal scale (l-not at all to 5-extremely). The scale was shown to have high internal consistency and good test-retest reliability [39]. Medical morbidity was assessed by means of global ratings made by the medicine clinic physicians and by record audits. Global ratings for Severity of Physical Disease, the extent to which Disease Explains Symptoms, and Unrealistic Fear of Illness (hypochondriasis) were made on 9-point ordinal scales ranging from l-absent or not at all to 9-extreme or completely. Definitions for points on

Psychiatric Comorbidity and Hypochondriasis these scales were provided. Both resident and staff physicians, who were unaware of the patients’ study diagnoses, made ratings at the time of their examinations. Statistical analyses included f-tests for continuous and ordinal data and Chi-square or Fisher’s exact tests for categorical data. Pearson’s correlation coefficients were used to examine relationships between continuous variables. One-way analyses of variance were used to examine the relationships among three diagnostic groups with Tukey’s tests for subsequent pairwise comparisons.

respect to mean educational level (3.5 ? 0.9 vs 3.2 * 1.1, ns), occupational class (4.3 2 1.5 vs 4.2 * 1.8, ns), and social class (44.2 t 13.3 vs 42.1 ? 15.3, ns). Fifty-six percent of both groups were married. Hypochondriacal and control subjects were also similar with respect to severity of aggregate physical disease as rated by examining physicians (mean 3.8 2 1.4 vs 4.0 + 1.6, ns). The mean age of onset of hypochondriasis was 36.9 ? 11.3 years and the median duration of illness was 19 months (range 2-144 months). Six subjects had had hypochondriasis for less than 6 months (range 2-5 months). More hypochondriacal subjects met criteria for coexisting DSM-III-R disorders than did controls. As is shown in Table 1, 62.0% of hypochondriacal subjects qualified for one or more other lifetime disorders compared with 30.0% of control subjects. With respect to specific disorders, more hypochondriacal subjects had lifetime depressive and anxiety disorders, but not substance use disorders. Twenty-eight percent of hypochondriasis subjects met criteria for current major depression which accounted for the difference between groups with respect to depressive disorders. With one exception, hypochondriacal and control subjects with anxiety disorders met criteria currently (during the past month) but few substance use disorders were current (two hypochondriasis and two control subjects). An effort was made to determine the tem-

Results Thirteen point eight percent of the patients screened scored above the cutoff for hypochondriasis. Of these, 61.4% met criteria for current hypochondriasis, as determined by the SCID. This represented 8.5% of the patients originally screened. When hypochondriasis subjects whose illness had lasted less than 6 months and who also met criteria for somatization disorder were excluded, 6.9% met criteria for hypochondriasis. Forty women and 10 men with a mean age of 39.6 ? 12.1 years comprised the hypochondriacal sample. Nonhypochondriacal controls consisted of 40 women and 10 men with a mean age of 39.9 5 12.3 years. Hypochondriacal and control subjects were similar with Table

1. Number (percent) of hypochondriasis coexisting disorders

Disorder Major depression, current Major depression, past Dysthymia Alcohol abuse/dependence Drug abuse/dependence Panic/agoraphobia Social phobia Simple phobia Generalized anxiety disorder Obsessive-compulsive disorder Other disorders Any depressive disorder Any anxiety disorder Any substance use disorder Any current disorder Any disorder

Hypochondriasis N = 50 14 6 4 7 4 8 2 1 0 0 0 22 11 10 27 31

(28.0) (12.0) (8.0) (14.0) (8.0) (16.0) (4.0) (2.0) (0.0) (0.0) (0.0) (44.0) (22.0) (20.0) (54.0) (62.0)

and control subjects

with

Normal controls N = 50 3 6 1 6 5 3 0 1 0 0 1 9 3 9 7 15

(6.0) (12.0) (2.0) (12.0) (10.0) (6.0) (0.0) (2.0) (0.0) (0.0) (2.0) (18.0) (6.0) (18.0) (14.0) (30.0)

X2*

P

8.58 0.00 1.90 0.88 0.12 2.55 2.04 0.00 0.00 0.00 1.01 7.90 5.32 0.07 17.83 10.31

0.005 ns ns ns ns ns ns ns ns ns 0.;5 0.02 0.x5 0.005

81

R. Noyes, Jr. et al.

poral relationship of coexisting disorders and hypochondriasis. In three subjects, depressive disorders (major depression or dysthymia) preceded the onset of hypochondriasis by 6 months or more, in four, the onset of a depressive disorder and hypochondriasis were coincident, and in 13 the onset of a depressive disorder followed the onset of hypochondriasis by 6 months or more. In four subjects the onset of an anxiety disorder (panic disorder, social phobia, or simple phobia) preceded that of hypochondriasis by 6 months or more, in four, the onset of anxiety and hypochondriasis were coincident, and in three, the onset of anxiety followed the onset of hypochondriasis by 6 months or more. When hypochondriacal subjects with comorbid depression were compared with subjects without comorbid depression (Table 2), few differences in demographic or illness characteristics were found. Also, there was little difference in the severity of hypochondriasis as measured by the Whiteley Index or the Somatization subscale of the SCL-90.

However, subjects with coexisting depression had more severe depressive symptoms, were more impaired in overall functioning as measured by the Global Assessment Scale, and fewer of them were employed. Hypochondriacal subjects with comorbid anxiety were younger and had an earlier onset of hypochondriasis compared with subjects with no coexisting disorder, although these differences were not statistically significant. They also had more severe anxiety symptoms and were rated as more hypochondriacal (Unrealistic Fear of Illness) by clinic physicians than subjects without coexisting depressive or anxiety disorders. Like the subjects with comorbid depression, subjects with comorbid anxiety were more impaired in overall functioning, as rated on the Global Assessment Scale (GAS). Although only four hypochondriacal subjects met DSM-III-R criteria for somatization disorder, an additional 14 subjects met less stringent criteria for this disorder (5 or more of 11 somatization disorder symptoms) proposed by Swartz et al. [33].

Table 2. Mean (%) values for demographic and illness characteristics of hypochondriacal subjects with coexisting depression and anxiety disorders and no coexisting disorder No coexisting Coexisting Coexisting disorde8 N = 22

depression N = 17

N = 11

P

Age (mean)

40.6

Sex (% women)

81.8% 59.1% 50.0% 40.6 38.3 27.5 18.2 26.7 35.3 69.3 12.4 10.7 44.8 27.7 3.9 4.5 3.7

42.5 76.5% 64.7% 17.6% 48.7 39.1 42.2 23.2 39.0 40.4 57.5 14.4 10.5 46.1 29.9 4.1 4.7 4.1

33.2 81.8% 36.4% 63.6% 44.3 31.1 24.5 24.1 33.6 36.1 57.0 13.4 10.7 47.6 33.4 3.2 4.0 5.4

ns ns

Marital status (% married) Employment status (% employed) Social class Onset of hypochondriasis Duration of hypochondriasisf Anxiety symptoms (SCL-90) Depressive symptoms (SCL-90) Somatic symptoms (SCL-90) Global Assessment Scale Neuroticism (EPI) Extroversion (EPI) Whiteley Index Somatosensory amplification Severity of disease Disease explains symptoms Unrealistic fear of illness

anxiety

o.o:F ns ns 0.0; 0.005” o.o;tYl”,b ns ns 0.1: ns 0.0;~

“Tukey’s comparing coexisting disorder and coexisting depression, p < .05. qukey’s no coexisting disorder and coexisting anxiety, p < .05. Tukey’s coexisting depression and coexisting anxiety, p < .05. dIncludes subjects without coexisting mood or anxiety disorders. ‘Social class according to the Hollingshead Index of Social Position (lo-highest to 80-lowest). fh4ean duration in months.

82

Psychiatric Comorbidity

Subjects with DSM-III-R somatization disorder had an earlier onset and longer duration of hypochondriasis than did subjects without somatization disorder (Table 3). They also had more severe hypochondriasis; more severe anxiety, depressive, and somatic symptoms; greater neuroticism; and poorer overall functioning, as measured by the GAS, than nonhypochondriacal subjects. However, because of small numbers we did not compare this group with the subsyndromal and no somatization disorder groups using statistical tests. Hypochondriacal subjects with subsyndromal somatization disorder fell between those with DSM-III-R somatization disorder and those with no somatization disorder but few differences between the subsyndromal and no somatization disorder groups were statistically significant. However, when subjects meeting criteria for full and subsyndromal somatization disorder were combined (N = 18) and compared with subjects without somatization disorder (N = 36) several differences reached statistical significance. Subjects with coexisting somatization disorder had more severe anxiety (25.4 -+ 10.0 vs 20.5 +- 7.0, p < 0.05) and

and Hypochondriasis

somatic symptoms (41.3 -t- 9.1 vs 35.9 2 8.9, p < 0.05). Their level of overall functioning, as assessed by the GAS, was lower (56.1 + 12.7 vs 63.8 k 9.1, p < 0.02). They showed trends toward greater neuroticism (15.2 t 5.2 vs 12.6 + 4.9, p = 0.07) and hypochondriacal attitudes (50.2 2 10.0 vs 45.1 2 5.6, p = 0.05), and had higher physician ratings of hypochondriasis (Unrealistic Fear of Illness)(5.4?1.7vs3.9? 1.8,p
Table 3. Mean (%) values for demographic and illness characteristics of hypochondriacal subjects with and without coexisting somatization disorder

Age (mean) Sex (% women) Marital status (% married) Employment status (% employed) Social class Onset of hypochondriasis Duration of hypochondriasis Anxiety symptoms (SCL-90) Depressive symptoms (SCL-90) Somatic symptoms (SCL-90) Global Assessment Scale Neuroticism (EPI) Extroversion (EPI) Whiteley Index Somatosensory amplification Severity of disease (GAS) Disease explains symptoms Unrealistic fear of illness Somatization disorder checklist

Somatization disorder N=4

Subsyndromal somatization disorder N = 14

No somatization disorder N = 36

41.5 50.0% 50.0% 25.0% 39.3 29.5 145.3 33.5 44.0 43.5 44.0 16.3 6.5 58.3 58.3 2.8 2.1 6.1 8.3

36.0 92.9% 50.0% 50.0% 50.0 33.9 27.6 23.1 34.3 40.6 59.6 14.9 11.3 47.9 29.8 3.4 3.4 5.1 6.0

41.0 75.0% 58.3% 38.9% 41.9 38.2 33.5 20.5 31.7 35.9 63.8 12.6 10.4 45.1 29.7 4.0 4.8 3.9 2.4

‘t-tests or, in the case of sex, marital status, and employment status, Chi-square tests compare subsyndromal somatization and no somatization disorder groups only.

PQ ns ns ns 0.:: ns ns ns ns ns ns ns ns ns ns 0.;. 0.05 0.0001 disorder

83

R. Noyes,

Jr. et al.

scoring high had greater Neuroticism scores (15.3 -+ 4.9 vs 11.4 2 4.1, p < 0.005). They also were rated as having more hypochondriasis (Unrealistic Fear of Illness) (4.8 rt 2.0 vs 3.7 + 1.6, p < 0.05).

Discussion There are a number of limitations to this study that need to be considered in interpreting the findings. A large number of patients who attended the medicine clinic in which this study was conducted did not participate. Some refused and others were not available or not contacted before leaving the clinic. This means that bias may have entered into the selection of subjects for interview. Because we do not have information about these persons we do not know whether they differed from subjects who were interviewed. Also, in many instances, the numbers were small, making firm conclusions impossible. Some data, such as ages of onset and temporal relationship of disorders, involved retrospective recall of doubtful reliability. In addition, this study may have been influenced by reporting biases. Some patients may have exaggerated and others may have minimized the extent of their health worries. The extent to which tendencies of this kind may have influenced our findings is unknown. Finally, our study was conducted in a diagnostic screening clinic of a tertiary care medical center. Consequently, the results may not be generalizable to other primary care settings. We found a high rate of psychiatric comorbidity among patients with hypochondriasis from a general medicine clinic. Sixty-two percent of our hypochondriacal subjects met criteria for at least one additional lifetime disorder compared with 30% of controls. Our findings replicate those of Barsky et al. [8] who used the Diagnostic Interview Schedule to screen a similar sample. In fact, rates for major depression, panic disorder, and alcohol abuse/ dependence were practically identical in the two studies, for both hypochondriacal and control subjects. The difference in overall comorbidity-62.0% with any psychiatric disorder in our study compared with 88.1% reported by Barsky et al. [8]-appears to be due to differences in the frequency with which dysthymic disorder and generalized anxiety disorder were diagnosed. Although many hypochondriacal and control subjects reported worries other than health (e.g., finances), our interviewer did not regard them as unrealistic or excessive. This is a difficult judgment to make and may have led to underdiagnosing of generalized

84

anxiety disorder in this study. Suggestions have been made for clarifying the boundary between this disorder and normal anxiety in DSM-IV [40]. We found, as have previous investigators, that hypochondriacal patients report high levels of depressive and anxiety symptoms. In addition, we showed that a high proportion of patients with hypochondriasis meet criteria for depressive and anxiety disorders as identified by structured interview. We must now ask what this comorbidity means. First, the increase appears to be specific for depression and anxiety. No increase in substance use disorders was observed and most of these disorders were not current. Second, the data suggest that hypochondriasis is, at least in some patients, an independent syndrome. Nearly half (46%) of our sample had no other current disorder and 38% had no other lifetime disorder. Also, the rate of coexisting disorders may have been increased by overlap in symptoms of hypochondriasis and anxiety and depressive disorders. Both include multiple, unexplained somatic symptoms together with unrealistic worry that may center on health (i.e., apprehension, fear of dying, hopelessness). Assuming a degree of independence, we attempted to examine the relationship between hypochondriasis and both depressive and anxiety disorders. We were limited in doing this by not having patients with depression and anxiety for comparison. However, hypochondriacal subjects with depressive disorders (mostly current major depression) did not differ from subjects without depressive disorders in terms of demographic or illness characteristics. The onset of coexisting depression usually followed the onset of hypochondriasis and was associated with more severe symptoms and impairment in functioning. On the other hand, hypochondriacal subjects with anxiety disorders (mostly current panic disorder agoraphobia) were younger and had an earlier onset of hypochondriasis. Also, in these subjects, the onset of anxiety usually preceded or coincided with the onset of hypochondriasis. These findings suggest that the hypochondriasis associated with depressive disorders is often primary and that coexisting depression may result from or contribute to greater severity of hypochondriasis. Hypochondriasis associated with panic disorder may, on the other hand, be a secondary manifestation of the anxiety disorder. This conclusion is based upon a small number of observations and is only suggestive. However, it is consistent with recent studies showing strong hypochondriacal fea-

PsychiatricComorbidityand Hypochondriasis tures in patients with panic disorder, features that respond to treatment of the primary illness [12,13,15,41]. It is also consistent with an earlier onset of anxiety which in this study was a mean of 28 years. Patients with anxiety disorders may, then, be a subgroup that should be removed from the larger pool of hypochondriacal patients encountered in primary care. To examine this possibility more thoroughly, three groups should be compared: those with hypochondriasis, those with panic disorder, and those with both disorders. According to our hypothesis, patients with coexisting disorders would resemble patients with panic disorder in terms of demographic and illness characteristics but differ from those with hypochondriasis alone. We believe the boundary between hypochondriasis and panic disorder is relatively distinct; our comparison of patients with illness phobia, a subtype of hypochondriasis, and panic disorder showed a clear separation [7]. A surprisingly small proportion of our hypochondriacal subjects (7.4%) met DSM-III-R criteria for somatization disorder. However, the threshold for this disorder is high and, when less stringent criteria were applied, a third of our subjects met them [l]. Compared with the remaining twothirds, subjects with subsyndromal somatization disorder appeared more severely ill. They had more severe hypochondriacal symptoms and were recognized by examining physicians as being more hypochondriacal. In fact, we found little basis for distinguishing between hypochondriacal subjects with and those without somatization disorder except with regard to severity. Patients with somatization disorder are believed to differ from those with hypochondriasis in terms of age of onset, gender distribution, personality type, and psychiatric comorbidity [4244]. However, with our limited sample and small numbers we could only see trends in these directions. There has been little study of the distinction between somatization disorder and hypochondriasis, but Murphy [45] found that, while many patients with somatization disorder were worried about their health, not all were. Also, Noyes et al. [7] and others observed that, though most hypochondriacal patients have somatic symptoms, not all of them do. The DSM-IV Work Group [40] considered dropping the somatic symptom criterion from hypochondriasis thereby diminishing the definitional overlap between the two conditions. Clearly, more work is needed on the distinction between these disorders.

Hypochondriasis appears to be associated with a high level of personality psychopathology. Hypochondriacal subjects had higher personality disorder and neuroticism scores than controls and, among hypochondriacal subjects, greater Axis II pathology was associated with more severe hypochondriasis. Our findings are similar to those of Barsky et al. [8] who reported that two-thirds of their hypochondriacal patients had personality disorders. Also, our observation of increased neuroticism confirms earlier reports [46] and is consistent with observations of anxious and obsessional traits in hypochondriacal patients [2,3]. Some authors have regarded hypochondriasis as arising from long-standing personality disturbance [8,47]. Others have noted that hypochondriacal behavior may dominate personality and may even represent a personality disturbance [48]. Regardless of how it is conceptualized, there appears to be a strong relationship between personality and hypochondriasis just as there is with somatization disorder. The high rate of comorbidity among patients with hypochondriasis may have treatment implications. We showed that half of these patients have treatment-responsive syndromes, mostly depression and anxiety. Both types of disorders are responsive to antidepressant medication, which might prove beneficial for patients who, along with depression and anxiety, appear to have more severe hypochondriasis. The tricyclic antidepressant, imipramine, has been used to treat this disorder and appeared useful in an uncontrolled trial with illness phobia, a subtype of hypochondriasis [47,49]. Trials of drug therapy seem warranted. Cognitive behavioral therapy has also been recommended for hypochondriasis and may find application in patients with coexisting depression or anxiety [50,51]. Cognitive therapy has proven efficacy in depression and is being used in anxiety disorders with promising preliminary results [52]. We gratefully acknowledge the assistance and cooperation of lmogene Barloon and the entire stuff of the Medicine Clinic without which this study would not have been possible.

References 1.

American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 3rd ed, revised. Washington,DC, AmericanPsychiatricPress,

2.

Kenyon FE: Hypochondriasis:a clinical study. Br J

1987

Psychiatry

110:478488,

1964

8.5

R. Noyes, Jr. et al. 3. Pilowsky I: Primary and secondary hypochondriasis. Acta Psych&r Stand 46273-285, 1970 4. Ladee GA: Hypochondriacal Syndromes. New York, Elsevier, 1966 5. Kellner R: The prognosis of treated hypochondriasis: a clinical study. Acta Psychiatr Stand 67369-79, 1983 6. Kellner R, Slocumb JC, Wiggins RJ: The relationship of hypochondriacal fears and beliefs to anxiety and depression. Psychiatr Med 4:15-24, 1987 7. Noyes R, Wesner R, Fisher M: A comparison of patients with illness phobia and panic disorder. Psychosomatics 33:92-99, 1992 8. Barsky AJ, Wyshak G, Klerman GL: Psychiatric comorbidity in DSM-III-R hypochondriasis. Arch Gen Psychiatry 49:101-108, 1992b 9. Buglass D, Clark J, Henderson AS, Kreitman N, Presley AS: A study of agoraphobic housewives. Psycho1 Med 773-86, 1977 10. Harper M, Roth M: Temporal lobe epilepsy and the phobic anxiety-depersonalization syndrome. Part I: a comparative study. Compr Psychiatry 3:129-151, 1962 11. Sheehan DV, Ballenger J, Jacobson G: Treatment of endogenous anxiety with phobic, hysterical, and hypochondriacal symptoms. Arch Gen Psychiatry 37: 51-59, 1980 12. Noyes R, Reich J, Clancy J, O’Gorman TW: Reduction in hypochondriasis with treatment of panic disorder. Br J Psychiatry 149:631-635, 1986 13. Pilowsky I: Dimensions of hypochondriasis. Br J Psychiatry 113:89-93, 1967 14. Fava GA, Kellner R, Zieleznu M, Grandi S: Hypochondriacal fears and beliefs in agoraphobia. J Affect Dis 14:239-244, 1988 15. Starcevic V, Kellner R, Uhlenhuth EH, Pathak D: Panic disorder and hypochondriacal fears and beliefs. J Affect Dis 24:73-85, 1992 16. Pilowsky I, Chapman CR, Bonica KJ: Pain, depression, and illness behavior in a pain clinic population. Pain 4:183-192, 1977 17. Barsky AJ, Wyshak G, Klerman GL: Hypochondriasis: an evaluation of the DSM-III criteria in medical outpatients. Arch Gen Psychiatry 43:493-500, 1986 18. Jacobs TJ, Fogelson S, Charles E: Depression ratings in hypochondria. NY State J Med 68:3119-3122,1968 19. Lewis A: Melancholia: a clinical survey of depressive states. J Ment Sci 80:277-378, 1934 20. Kreitman N, Sainsbury P, Pearce K, Costain WF: Hypochondriasis and depression in outpatients at a general hospital. Br J Psychiatry 111:607-615, 1965 21. Stenbach A, Jalava J: Hypochondria and depression. Acta Psychiatr Stand 37(suppl 162):240-246, 1961 22. Fava GA, Pilowsky I, I’ierfederici A, Bernardi M, Pathak D: Depression and illness behavior in a general hospital: a prevalence study. Psychother Psychosom 38:141-153, 1982 23. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 3rd ed. Washington, DC, American Psychiatric Association, 1980 24. Tyrer I’, Lee I, Alexander J: Awareness of cardiac

86

25. 26.

27. 28. 29.

30.

31.

32.

33.

34. 35.

36.

37.

38. 39. 40. 41.

function in anxious, phobic, and hypochondriacal patients. Psycho1 Med 10:171-174, 1980 Oxman TE, Barrett J: Depression and hypochondriasis in family practice patients with somatization disorder. Gen Hosp Psychiatry 2321329, 1985 Kirmayer LJ, Robbins JM: Three forms of somatization in primary care: prevalence, co-occurrence and sociodemographic characteristics. J Nerv Ment Dis 179647-655, 1991 Hathwaway SR, McKinley JC: The Minnesota Multiphasic Personality Inventory Manual, Revised. New York, The Psychological Corporation, 1951 Derogatis LR: The SCL-90-R: Administration, Scoring, and Procedures. Towson, MD, Clinical Psychometric Research, 1983 Spitzer RL, Williams JBW, Gibbon M, First MB: Structured Clinical Interview for DSM-III-R-NonPatient Edition, Washington, DC, American Psychiatric, 1990 Spitzer RL, Williams JBW, Gibbon M, First MB: The Structured Clinical Interview for DSM-III-R (SCID). I. History, rationale, and description. Arch Gen Psychiatry 49:624, 629, 1992 Williams JBW, Gibbon M, First MB, Spitzer RL, Davies M, Borus J, Howes MJ, Kane J, Pope HG, Rounsaville B, Wittchen H-U: The Structured Clinical Interview for DSM-III-R (SCID). II. Multisite test-retest reliability. Arch Gen Psychiatry 49630-636, 1992 Barsky AJ, Klerman GL, Wyshak G, Spitzer RL, Williams JBW: A structured diagnostic interview for hypochondriasis: a proposed criterion standard. J Nerv Ment Dis 180:2&27, 1992a Swartz M, Hughes D, George L, Blazer D, Landerman R, Bucholz K: Developing a screening index for community studies of somatization disorder. J Psychiatr Res 20:335-343, 1986 Hollingshead AB, Redlich FC: Social Class and Mental Illness. New York, Wiley and Sons, 1958 pp 398 407 Endicott J, Spitzer RJ, Fleiss JL, Cohen J: The Global Assessment Scale: a procedure for measuring overall severity of psychiatric disturbance. Arch Gen Psychiatry 33:766-771, 1976 Hyler S, Reider R, Spitzer R, Williams JB: Personality Diagnostic Questionnaire (PDQ). New York, New York State Psychiatric Institute, Biometrics Research, 1983 Hyler SE, Rieder RO, Williams JBW, Spitzer RL, Hendler J, Lyons M: The Personality Diagnostic Questionnaire: development and preliminary results. J Pers Disord 2:229-238, 1988 Eysenck HJ, Eysenck SBS: Manual for the Eysenck Personality Inventory. San Diego, Educational and Industry Testing Service, 1968 Barsky AJ, Wyshak G, Klerman CL: The Somatosensory Amplification Scale and its relationship to hypochondriasis. J Psychiatr Res 24:32%334, 1990b Task Force on DSM-IV: DSM-IV Options Book. Work in Progress. Washington, DC, American Psychiatric Association, 1991 Otto MW, Pollack MH, Sachs GS, Rosenbaum JF: Hypochondriacal concerns, anxiety sensitivity, and panic disorder. J Anx Dis 6:93-l&4, 1992

Psychiatric Comorbidity

42. Cloninger CR, Sigvardsson S, von Knorring A-L, Bohman M: An adoption study of somatoform disorders. II. Identification of two discrete disorder. Arch Gen Psychiatry 41:863-871, 1984 43. Barsky AJ, Wyshak G, Klerman GL: Transient hypochondriasis: a new somatoform diagnosis? Arch Gen Psychiatry 48:955, 1991 44. Yates WR: Transient hypochondriasis: a new somatoform diagnosis? Arch Gen Psychiatry 48:955, 1991 45. Murphy MR: Classification of the somatoform disorders. In Bass C (ed), Somatization: Physical Symptoms and Psychological Illness. London, Blackwell Scientific Publications, 1990, pp l&39 46. Bianchi GN: Origins of disease phobia. Aust NZ J Psychiatry 5:241-257, 1971

and Hypochondriasis

47. Kellner R: Somatization and Hypochondriasis. New York, Prager, 1986 48. Tyrer I’, Fowler-Divon R, Ferguson B, Kelemen A: A plea for the diagnosis of hypochondriacal personality disorder. J Psychosomatic Res 34:637642, 1990 49. Wesner RB, Noyes R: Imipramine: an effective treatment for illness phobia. J Affect Dis 22:4348, 1991 50. Barsky AJ, Geringer E, No01 CA: A cognitiveeducational treatment for hypochondriasis. Gen Hosp Psychiatry 10:322-327, 1988 51. Warwick HMC, Marks IM: Behavioral treatment of illness phobia and hypochondriasis: a pilot study of 17 cases. Br J Psychiatry 152:239-241, 1988 52. Hoffart A: Cognitive treatments of agoraphobia: a critical evaluation of theoretical basis and outcome evidence. J Anx Dis 7:75-91, 1993

87