Somatoform disorders: Comorbidity with other DSM-III-R psychiatric diagnoses in Greece

Somatoform Disorders: Comorbidity With Other DSM-III-R Psychiatric Diagnoses in Greece George Garyfallos, Aravela Adamopoulou, Anastasia Karastergiou, Maki Voikli, Nikos Ikonomidis, Stamatis Donias, John Giouzepas, and Evangelos Dimitriou From a t o t a l s a m p l e of 1,448 psychiatric outpatients, 175 (12.1%) received a diagnosis of a somatoform disorder according to DSM-III-R criteria. One hundred t w e n t y - t w o (70%) of these patients had another current axis I diagnosis, and this rate increased to 79% (139 of 175) when lifetime psychiatric diagnoses were recorded. The most frequent comorbid diagnoses were depressive disorders, i.e., dysthymia and major depression, and then anxiety disorders, mainly panic disorder. One hundred ten (63%) of the somatoform patients met the criteria for a personality disorder, significantly higher than the rate (52%} for the rest of the total sample (n = 1,273), who were used as a control group. The most frequent comorbid personality disorders were histrionic, dependent, and personalities of cluster B in general. Hypochondriasis was the only somatoform disorder that was additionally signifi-

p

HYSICIANS have recognized somatizing patients for centuries. These cases have been given a variety of overlapping labels, notably "hysteria," "hypochondriasis," and "melancholia."~ In 1980, DSM-III 2 introduced the term "somatoform disorders" for a group of disorders for which the essential features are physical symptoms suggesting physical disorder without any demonstrable organic findings or known physiological mechanisms and with positive evidence or strong presumption that the symptoms are linked to psychological factors or conflicts. The DSM-III somatoform disorders comprise five specific entities: somatization disorder that is the antecedent of Briquet's syndrome, conversion disorder, psychogenic pain disorder, hypochondriasis, and atypical somatoform disorder. DSM-III-R 3 adds two new entities, body dysmorphic disorder and undifferentiated somatoform disorder, while DSM-IV 4 preserves all seven entities in the section of somatoform disorders. A field of particular research interest is the comorbidity of somatoform disorder with other psychiatric disorders. There are studies reporting that somatoform disorders in general or some in particular manifest a high comorbidity with other axis I diagnoses, mainly depressive and anxiety disorders. 5-14Similarly interesting is the correlation of somatoform disorders with axis II diagnoses, i.e., personality disorders, a field that has not been heavily researched, with the exception of the

cantly related to obsessive-compulsive personality disorder. Somatoform patients with a concomitant personality disorder manifested more severe overall psychopathology as measured by the Minnesota Multiphasic Personality Inventory (MMPI) and a worse level of functioning than those without. The results of the present study show t h a t (1) patients with s o m a t o form disorders have a high rate of comorbidity with other clinical syndromes and personality disorders, and (2) the presence o f a personality disorder is related to more severe overall psychopathology and a worse level of functioning. All of the above indicate that special attention must be paid to the interaction between somatoform disorders, other clinical syndromes, and personality structure at the level of both clinical and research practice.

Copyright© 1999by W.B. Saunders Company

relation of somatization disorder with histrionic and antisocial personality disorder. 5,15-18There are a few studies that investigated this topic more broadly, and found that somatoform disorders in general 19,2° or some of them, such as somatization disorder21,22 or hypochondriasis, 12 are highly correlated with personality disorders. This correlation is particularly interesting, as there are studies supporting the view that the presence of a personality disorder considerably influences the symptom severity, outcome, number of relapses, and treatment response of a clinical syndrome. These studies refer to patients with depressive and/or anxiety disorders.23-32 The aim of the present study, which is the first of its kind in Greece, is to investigate (1) the comorbidity of somatoform disorders, both as a total and as specific entities, with other clinical syndromes (axis I) and personality disorders (axis II), and (2) whether patients with a concomitant diagnosis of a personality disorder have more severe psychopathol-

From the Community Mental Health Center of the Northwestern District of ThessalonikL Thessaloniki; and the B University Department of Psychiatry, Aristotelian University of Thessaloniki, Thessaloniki, Greece. Address reprint requests to George Garyfallos, M.D., Community Mental Health Center, 20 Papadopoulou St.-Sikies, 566 25 Thessaloniki-Greece. Copyright © 1999 by W.B. Saunders Company 0010-440X/99/4004-0006510.00/0

Comprehensive Psychiatry,Vol. 40, No. 4 (July/August), 1999: pp 299-307

299

300

GARYFALLOS ET AL

ogy compared with patients without a diagnosis of a personality disorder.

Table 1. Demographic and Clinical Characteristics of the Sample (N = 175) Characteristic

METHOD The study sample consisted of consecutive outpatients who attended the Community Mental Health Centers of the Northwestern District and the B University Department of Psychiatry of Aristotelian University of Thessaloniki in Greece over a period of 5 years (1990 to 1994) and received a diagnosis of somatoform disorder according to DSM-III-R 3 criteria. A DSM-III-R diagnosis (as previously a DMS-III diagnosis and now a DSM-IV diagnosis) is standard practice in both Centers, and all scientific personnel who are involved in diagnostic interviews are trained and experienced in its use. Furthermore, the great majority have been working in this scientific field for over one decade. The axis I diagnoses of the present study were made at the disposition conference by consensus of the whole therapeutic team. This procedure guarantees a high axis I diagnostic reliability, as also shown in other studies performed in these settings.28,33 For the axis II diagnosis of personality disorders, the Structured Clinical Interview for DSM-III-R axis II (SCIDII) was used in conjunction with the SCID Personality Questionnaire. 34 The instruments have been translated into Greek, and their diagnostic sensitivity has been validated for the Greek population, as The current level of functioning was assessed with the Global Assessment of Functioning (GAF) scale of the DSM-III-R. All remaining patients who attended both Centers during this period (1990 to 1994), followed this diagnostic procedure, and received any other axis I and/or axis II diagnosis were used as a control group for axis II diagnoses. Finally, all patients under study who had at least a ninth-grade education completed the Minnesota Multiphasic Personality Inventory (MMPI), which was adapted for use in Greece. 36 The MMPI, a test for assessing patient personality characteristics, is also a useful device in assessing psychopathology37 and thus has been used for this purpose in many comparative studies between groups of patients. MMPI results that were invalid (i.e., false-positive or false-negative) were excluded. It is important to emphasize that all patients participated in the study voluntarily after informed oral consent was obtained.

RESULTS

From a total sample of 1,448 patients who attended both Centers, 175 (12.1%) received a DSM-III-R diagnosis of a somatoform disorder. Their demographic and clinical characteristics are presented in Table 1. Patients with a somatoform disorder did not differ significantly regarding any demographic characteristic in comparison to patients (n = 1,273) with other diagnoses (P > .05). The most frequent somatoform disorder was conversion disorder, followed by somatization and undifferentiated somatoform disorder. In conversion disorder, there were significantly more women, (81%, 46 of 57) as compared with hypochondriasis (52%, 12 of 23, chi-square [×2] = 6.70, P < .05), undifferentiated somatoform disorder (60%, 21 of

Sex Male Female Age (yr) Education (yr) Marital status Single Married Divorced/widowed Somatoform entity Somatization Undifferentiated Conversion Hypochondriasis Somatoforrn pain Body dysmorphic Somatoform NOS Age of onset (yr) Duration (yr)

No.

Mean -+ SD

54 121

%

31 69 36.1 _+ 11.5 10.2 _+ 4.5

61 100 14

35 57 8

36 35 57 23 4 6 14

21 20 33 13 2 3 8 25.5 _+ 9.4 10.6 -+ 6.2

35, ×z = 4.70, P < .05), and the other remaining diagnoses (64%, 813 of 1,273, ×2 = 10.22, P < .01). Similarly, in somatization disorder, there was a significant preponderance of females (78%, 28 of 36) as compared with hypochondriasis (×2 = 4.22, P < .05), and a tendency for statistical significance in comparison to patients with other diagnoses (×z = 2.95, P < . 1) except somatoform. Of the somatoform patients, 70% qualified for another diagnosis of current disorder and 79% for another lifetime disorder (Table 2). The most frequent comorbid disorders were depressive disorders (49% current and 61% lifetime), mainly major depression or dysthymia, followed by anxiety disorders (30% current and 35% lifetime), mainly panic disorder with or without agoraphobia. There were differences between the specific somatoform disorders regarding the frequency of comorbid diagnoses. Thus, the frequency of "any" axis I diagnosis was as follows: in somatization disorder, 78% (28 of 36) current and 89% (32 of 36) lifetime; in hypochondriasis, 78% (12 of 23) current and 87% (20 of 23) lifetime; in undifferentiated somatoform disorder, 77% (27 of 35) current and 86% (30 of 35) lifetime; and in conversion disorder, 56% (32 of 57) current and 65% (37 of 57) lifetime. Conversion disorder coexists less frequently with any axis I diagnoses compared with other somatoform disorders. This comparison is statistically significant for the somatization disorder (×2 = 4.51, P < .05 current and ×2 = 5.41, P < .05 lifetime),

SOMATOFORM DISORDER, COMORBIDITY

301

Table 2. Comorbid Axis I Diagnoses in Patients With Somatoform Disorders (n = 175) Current

Lifetime

Diagnosis

No.

%

No.

%

Major depression Dysthymia Depression NOS Cyclothymia Depressive disorders (total) Panic disorder with or without agoraphobia Generalized anxiety Social phobia Simple phobia Obsessive-compulsive disorder Anxiety NOS Anxiety disorders (total) Sexual disorders Substance abuse Depersonalization Any axis I disorder No axis I disorders

39 40 9 4 86

22 23 5 2 49

58 41 10 4 107

33 23 6 2 61

30 6 5 5 5 6 52 3 5 1 122 53

17 3 3 3 3 3 30 2 3 1 70 30

37 8 7 7 5 6 62 3 7 1 139 36

21 5 4 4 3 3 35 2 4 1 79 21

while there appeared to be a tendency for statistical significance regarding hypochondriasis and undifferentiated somatoform disorder (P < . 1). Furthermore, there was a significant difference between somatization and conversion disorder concerning comorbidity with a depressive disorder as both current (somatization 24 of 36 and conversion 21 of 57, × 2 = 7.86, P < .01) and lifetime diagnoses (somatization 30 of 36 and conversion 28 of 57, ×2 = 11.00, P < .001). Finally, the hypochondriacal subgroup was the only somatoform subgroup where comorbid depressive and anxiety disorders were equally present (12 of 23, 52%), while in all other subgroups, the most prevalent comorbid diagnoses were depressive disorders. The somatoform patients significantly more frequently exhibited any personality disorder and histrionic, dependent, and cluster B personalities compared with the patients with other diagnoses (Table 3). Somatization disorder was significantly more frequently combined (v control group) with any personality disorder (25 of 36 v 664 of 1,273, X2 = 4.19, P < .05) and histrionic (13 of 36 v 92 of 1,273, ×2 = 39.61, P < .001), dependent (five of 36 v 47 of 1,273, X2 = 7.09, P < .01), and cluster B personalities (16 of 36 v 263 of 1,273, X2 = 11.81, P < .01). Furthermore, there was also a tendency for statistical significance regarding antisocial personality disorder (two of 36 v 12 of 1,273, X2 = 3.36, P < . 1). Both patients with somatization and antiso-

cial personality disorder were male. Thus, the prevalence of these personality disorders in male patients with somatization disorder was two of eight (25%). Of 12 antisocial patients in the control group, 11 were male (11 of 460, 2.4%). The statistical comparison showed that male subjects with somatization disorder more frequently had an antisocial personality disorder than patients with other diagnoses (two of eight v 11 of 460, ×2 = 7.69, P < .01). However, since the number of subjects is too small, these results must be considered indicative only. Conversion disorder had a significantly higher comorbidity (v control group) with any personality disorder (38 of 57 v 1,273, X2 = 4.60, P < .05) and histrionic (18 of 57 v 92 of 1,273, ×2 = 42.64, P < .001), dependent (six of 57 v 47 of 1,273, X2 = 6.66, P < .01), and cluster B personalities (27 of 47 v 263 of 1,273, X2 = 22.83, P < .001). Hypochondriasis differed significantly versus the control group on any personality disorder (17 of 23 v 664 of 1,273, X2 = 4.29, P < .05) and histrionic (five of 23 v 92 of 1,273, × 2 = 4.93, P < .05), obsessive-compulsive (five of 23 v 89 of 1,273, X2 = 5.40, P < .05), and cluster C personalities (nine of 23 v 184 of 1,273, X2 = 10.66, P < .01). Finally, undifferentiated somatoform disorder was less frequently combined with personality disorders (51%) than other somatoform disorders, differTable 3. Distribution of Personality Disorders in Patients With Somatoform Disorders and Those With Other Diagnoses (N = 1,448) Somatoform Disorder Other Diagnosis (n = 175) (n = 1,273) Personality Disorder

No.

%

No.

%

Any Cluster A Cluster B Cluster C Paranoid Schizoid Schizotypal Antisocial Borderline Histrionic Narcissistic Avoidant Dependent Obsessive-compulsive Passive-aggressive NOS Self-defeating

110 8 63 34 6 0 3 2 22 42 6 6 16 15 5 27 5

63 5 36 19 3 0 2 1 13 24 3 3 9 9 3 15 3

664 60 263 184 19 7 41 12 204 92 27 62 47 89 25 201 26

52* 5 21t 14 1 <1 3 1 16 7t 2 5 4t 7 2 16 2

* P < .01, t P < .001: X2 with Yates' correction when necessary, d f = 1.

302

GARYFALLOS ET A L

ing from the control group only on histrionic personality disorder (six of 35 v 92 of 1,273, X2 = 4.83, P < .05). Out 175 somatoform patients, 77 completed the MMPI, and their tests were valid. They did not differ significantly (P > .05) by sex, age, marital status, and the presence of a personality disorder versus those who did not complete the test. Table 4 presents the statistical comparison between somatoform patients with (n = 48) and without (n = 29) a personality disorder for MMPI scores on the clinical scales, the sum of clinical scales, and research scales such as A (anxiety), Mas (manifest anxiety), D1 (depression), Dep (subjective depression), Mor (poor moral), Dy (dependency), Soc (social maladjustment), and Es (ego strength). Patients with a personality disorder had significantly higher scores on all but three (Mf, Ma, and Si) scales in comparison to those without a personality disorder. Finally, somatoform patients with a personality disorder had a significantly lower GAF score, i.e., a worse level of functioning, than those without (55.1 +__11.1 v 59.9 ___ 10.7, t = 2.81, d f = 173, P < .01).

Table 4. Mean MMPI Scores on Clinical and Sum of Clinical and Research Scales for Somatoform Patients With and Without a Personality Disorder Personality Disorder Scale

Yes (n = 48)

No (n = 29)

Hs

74.1 -+ 10.8

66.5 -+ 10.2t

D

65.5 -+ 9.6

59.1 -+ 8.8t

Hy

68.1 -+ 9.3

60.9 +- 9.8t

Pd

60.6 -+ 7.3

56.5 -+ 7.8t

Mf

52.3 -+ 9.6

52.2 -+ 8.9

Pa

61.5 -4- 8,7

56.6 +- 8.1"

Pt

66.7 -4- 8.3

60.9 -+ 6.51-

Sc

61.6 _+ 8.5

57.2 _+ 6.9*

Ma

53.0 - 7.5

51.6 -+ 11.0

Si

55.3 -+ 8.9

51.8 +_ 9.7

620.0 _+ 50.2

573.1 -- 43.2¢

A

59.6 _+ 8.2

54.4 -- 8.8*

Mas

61.3 -4- 8.3

56.4 - 8.6*

D1

62.8 +_ 8.8

56.4 _+ 8.41-

Dep

59.5 -+ 8.3

54.5 -+ 7.41"

Mor

59.0 -+ 9.3

54.6 -+ 7.3*

Dy

59.4 +- 8.7

55.5 -+ 6.6*

Soc

55.1 -4- 10.2

51.0 -+ 7.1"

Es

36.5 -+ 8.7

42.2 -+ 7.5t

Sum

NOTE. For Es, a higher score indicates a better psychological state. * P < .05, t P <

.01, a t p < .001: 2-tailed Student ttest, d r = 75.

DISCUSSION

The percentage of 12.1% of the total sample of the present study presenting with a somatoform disorder is higher than that reported in previous studies (0.6% to 5%) from outpatient clinics. 19,38-4° The difference may be at least partly due to cultural differences, and is congruent with other research findings for Greek and other culturally similar populations. 41-46 The sex distribution in specific somatoform disorders is in line with the findings of other studies in clinical or community samples. 6,2° Furthermore, DSM-IV 4 reports that the sex ratio (women:men) in conversion disorders varies from 2:1 to 10:1 (5:1 in the present study), while hypochondriasis and undifferentiated somatoform disorder are equally common in males and females (1:1 and 1.5:1 in the present study, respectively). For somatization disorder, the data of the present study show that although it appears more frequently in women (3.8:1), it is not rare in men, a finding congruent with DSM-IV 4 reports for higher frequencies in Greek and PuertoRican versus US men. The findings of the present study suggest that somatoform disorders both as a whole and as specific entities show a high comorbidity with other axis I diagnoses. Previous studies on this issue reported similar results, 11-13,47 with the exception of one based on small sample, which found considerably lower rates, z° Almost all previous studies agree that the most common comorbid disorders with somatization disorder are depressive disorders, a finding consistent with the present study. The lifetime prevalence rate of major depression in the present study (53%) is below the lowest range (54% to 94%) in the international literature. 5-~°,16 However, these studies did not report even a single case of dysthymia or depressive disorder NOS. On the contrary, the present study found that the rate of comorbid dysthymia was 31% and the rate of any comorbid depressive disorder 83%, very similar to the reported rates of major depression of the prior studies. 5,7,16 Regarding hypochondriasis, the lifetime prevalence rate of any comorbid depressive disorder (52%) in the present study is between the rates found by two other studies (55% and 44%), 12,47and the comorbidity of somatoform disorders as a whole with any depressive disorder is lower compared with the study by Rief et al. 13 (61% v 87%), possibly due to the high frequency of conversion disorder in the

SOMATOFORM DISORDER, COMORBIDITY

present study. Conversion disorder manifested lower comorbidity with any depressive disorder, as well as with any axis I diagnoses, versus all of the other somatoform disorders. A possible explanation for this finding could be the "la belle indifference," i.e., a relative lack of concern about the nature or implications of the symptoms frequently shown by these patients. In addition to the high comorbidity of depressive and somatoform disorder, a considerable comorbidity with anxiety disorders was also found, as greater than one third (35%) of the patients had an anxiety disorder as a lifetime diagnosis, similar to the rate (40%) reported by Rief et al. 13 The more common comorbid anxiety disorder was panic disorder with or without agoraphobia manifested at a higher frequency than found by Rief et a113 (21% v 13%). Their study found that agoraphobia without panic was the more frequent comorbid disorder, a diagnosis that has been assigned in none of the patients in the present study. However, the finding of the present study is in line with the classic view that in clinical settings a diagnosis of agoraphobia without a current (or history of) panic disorder is very rare. 3,4 In contrast to the rest of the specific somatoform entities that were correlated more frequently with depressive versus anxiety disorder, hypochondriasis manifested an equal comorbidity with the above-mentioned diagnoses. Of the two other relevant studies, one found a higher comorbidity of hypochondriasis with anxiety disorders,12 and the other with depressive disorders. 47 The frequent comorbidity of somatoform disorders with both depressive and anxiety disorders raises a question regarding what exactly is the nature of the comorbidity. In a study like this, one cannot clearly answer the question. However, one can suggest some hypotheses. One possibility may be that somatoform disorders are masked expressions of depression or anxiety and, in the case of comorbidity, are subsymptoms of a depressive and/or an anxiety syndrome. However, the use in the present study and other similar studies of DSM-III-R/DSM-IV criteria does not support this view. A second possibility may be that depressive and anxiety disorders are complications of the somatoform disorders. As somatoform patients suffer chronically from unexplained somatic symptoms and many of them live as physically disabled persons, they develop depression and anxiety. Pa-

303

tients who believe that the psychiatric symptoms are a consequence of their physical symptoms frequently adopt this view. A third view may be that the disorders share common underlying processes. Torgersen48 found a high frequency of anxiety disorders, especially generalized anxiety disorder, in the co-twins of somatoform-disordered twins and supported the existence of a link between the two disorders, although this link may not be genetic in origin. In a family study, 49 it was found that the relatives of children with major depression frequently manifested somatization disorder and anxiety disorders. Katon et al. 1° speculated that a neurophysiologic dysfunction and resulting autonomic nervous system dysregulation may cause panic disorder, depression, and somatic symptoms. In addition, it has also been reported that depressive mood fosters illness-related memories and a negative view of one's health. 5° Dysregulation of serotonin metabolism may also constitute a common underlying process of somatoform and depressive disorders. 13 Furthermore, data from family and twin studies 49,51-53 suggest a common predisposition between depressive and anxiety disorders. Therefore, it is possible that a shared underlying diathesis between depressive, anxiety, and somatoform disorders in conjunction with developmental experiences and life events lead to the manifestation of these clinical syndromes simultaneously or in different phases of the patient's life. An important question can be raised regarding the reason that the patients in this Greek sample manifested comorbidity so similar to samples in other countries with different cultures. Perhaps the similarity relates to the diagnostic system, i.e., DSM-III-R, which is unable to detect cultural differences. Against this hypothesis is the fact that the use of DSM-III-R in the present study produced considerably higher rates of diagnosis of somatoform disorders versus other studies performed in outpatient clinics (12.1% v 0.6% to 5%). Another more plausible hypothesis could be that Greeks sustain their fundamental cultural values and norms, which encourage them to express psychological distress through somatic language. Therefore, they manifest higher rates of somatoform disorders. This pattern is characteristic of agricultural societies, where there is a reliance on manual employment. 54 Perfect physical health is essential for survival, and the concept of illness is conceived as

304

somatic illness. 54 However, in recent years, the structure of Greek society has gradually been transformed from agricultural to more industrialized or "westernized." In addition, cultural differences may have decreased due to information transfer throughout the world. All of these factors point to the fact that while Greeks continue to have higher rates of somatoform disorders, they manifest comorbid disorders, i.e., mainly depression and/or anxiety, to the same extent as patients in other cultures. There are very few studies investigating the comorbidity between somatoform disorders and personality disorders. The rate of 63% for any personality disorder of the present study is higher than the rate (48%) reported by Snyder and Strain z° in a different setting and without a structured interview for the assessment of personality disorders. For the relationship of personality disorders to specific somatoform entities such as somatization disorder and hypochondriasis, the rates of the present study, i.e., 69% and 74%, respectively, are similar to those found by other investigators. 12,21,22 In addition, all previous studies are in agreement with the present study that somatoform disorders both as a whole and as specific entities, i.e., somatization disorder and hypochondriasis, are related more frequently to personality disorders than the corresponding control groups. 12,21,z2,55There are no studies investigating the co-occurrence of conversion disorder or undifferentiated somatoform disorder with personality disorders. The results of the present study show that conversion disorder coexisted frequently with personality disorder (67%), while undifferentiated somatoform disorder did not (51%). Histrionic personality disorder was the most highly correlated personality disorder both with the somatoform disorders as a whole and with all the specific somatoform subgroups. It was mainly present in somatization and conversion disorder, but also in undifferentiated somatoform disorder, where it was the only personality disorder discriminating that somatoform entity from the control group. DSM-III-R and DSM-IV texts state that somatization and conversion disorders are commonly associated with histrionic personality disorder, while other studies 15,~7 support the somatization-histrionic correlation. Dependent personality disorder is also another personality disorder that discriminated all of the somatoform disorders (except undifferentiated) from the control

GARYFALLOS ET AL

group. DSM-III-R and DSM-IV support the high correlation of dependent personality disorder with conversion disorder, and Stern et al. 22 support the correlation with somatization disorder. Of particular interest is the association of somatization and antisocial personality disorder. This assumption stems largely from Guze's 56 view that hysteria and sociopathy have a common familial origin, whether genetic, social, or both. Some years later, Lilienfeld et al. 15,57 suggested that histrionic individuals develop antisocial personality disorder if they are male and somatization disorder if female, and the three conditions are reportedly expressions of a shared hereditary diathesis. 56 More recently, Frick et al. 58 concluded that behavioral disinhibition may be a common predisposition that underlies both antisocial behavior and somatization. On the contrary, findings from community samples, 5,a~primarycare patients, 2a and psychiatric outpatient services 22 did not find any significant association between the two disorders. The results of the present study show that male somatizers significantly more frequently manifested antisocial personality disorder than men in the control group. However, as the numbers are two small, this finding is only indicative and further research is necessary. Another important finding of the present study is that somatoform patients with a concomitant personality disorder manifested more severe overall psychopathology and a worse level of functioning than those without. In the only study that used a similar approach and examined only patients with somatization disorder, Stern et al. 22 did not find differences between these two subgroups on any scale of the 90-item Symptoms Checklist. However, there are reports for other clinical syndromes with results similar to the present study. It was found that in patients with major depression, 59 agoraphobia, 6° social phobia, 32,61 obsessive-compulsive disorder 62 or anxiety and/or depressive disorders in general, 28 the presence of an axis II diagnosis indicates more severe psychopathology. Furthermore, it has been suggested that the co-occurrence of a personality disorder and a clinical syndrome predicts a negative treatment outcome, more relapses, a poorer prognosis, 23-32 and worse psychosocial/occupational functioning. 26,6%64The last finding has also been confirmed for the somatoform patients by the results of the GAF and the Soc scale of the MMPI in the present study. It is not clear whether there is a preexisting personality disorder that leads to a more

SOMATOFORM DISORDER, COMORBIDITY

305

severe psychopathology of the clinical syndrome, i.e., somatoform disorder, or whether a s y n d r o m e with a more severe clinical picture influences the patient's personality and "creates" personality disorders. All of the above indicate that somatoform disorders frequently co-occur with personality disorders and that the presence of a personality disorder is related to a more severe psychopathology. This underscores the importance of personality assessment in evaluation of the s y n d r o m e in clinical practice. However, prospective longitudinal studies are necessary to understand better the nature of the relationship b e t w e e n these s y m p t o m disorders and personality disorders. In conclusion, the present study using a large sample of Greek outpatients indicates that somatoform disorders are frequently associated with other clinical syndromes, m a i n l y anxiety and depressive disorders, as well as personality disorders. Regarding these c o m o r b i d disorders, there are similarities but also differences b e t w e e n the specific somato-

form entities. For instance, conversion disorder is less frequently c o m b i n e d with other axis I diagnoses, undifferentiated somatoform disorder exhibits lower comorbidity with personality disorder, and histrionic personality disorder is the c o m m o n c o m o r b i d feature for all somatoform disorders. Finally, it seems that cultural factors contribute to the formation of some particularities regarding the manifestation of somatoform disorders in the Greek population. At the same time, some sociocultural changes in recent years contribute to the developm e n t of similarities to people in other countries regarding the comorbidity of somatoform disorders with other psychiatric syndromes. Further research is necessary through various approaches (i.e., clinical, neurobiologic, family, genetic, psychosocial, cross-cultural, treatmentoutcome, etc.) to clarify better the interactions of somatoform disorders both with other clinical synd r o m e s - m a i n l y anxiety and depressive d i s o r d e r s - and with the personality structure itself.

REFERENCES 1. Lipowski J. Somatization: the concept and its clinical application. Am J Psychiatry 1988;145:1358-1368. 2. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. Ed. 3. Washington, DC: American Psychiatric Press, 1980. 3. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. Ed. 3. Rev. Washington, DC: American Psychiatric Press, 1987. 4. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. Ed. 4. Washington, DC: American Psychiatric Press, 1994. 5. Liskow B, Othmer E, Penick E, DeSouza C, Gabrielli W. Is Briquet's syndrome a heterogenous disorder? Am J Psychiatry 1986;143:626-629. 6. Schwartz M, Blazer D, George L, Landerman R. Somatization in a community population. Am J Psychiatry 1986;143:14031408. 7. Morrison J, Herbstein J. Secondary affective disorder in women with somatization disorder. Compr Psychiatry 1988;29: 433-440. 8. Orenstein H. Briquet's syndrome in association with depression and panic: a reconceptualization of Briquet's syndrome. Am J Psychiatry 1989;146:334-338. 9. Brown F, Golding J, Smith R. Psychiatric comorbidity in primary care somatization disorder. Psychosom Med 1990;55: 445-451. 10. Katon W, Lin E, Von Korff M, Rasso J, Lipscomb P, Bush T. Somatization: a spectrum of severity. Am J Psychiatry 1991;148:34-40. 11. Simon G, Von Korff M. Somatization and psychiatric disorder in the NIMH Epidemiologic Catchment Area Study. Am J Psychiatry 1991;148:1494-1500. 12. Barsky A, Wyshak G, Klerman G. Psychiatric comorbid-

ity in DSM-III-R hypochondriasis. Arch Gen Psychiatry 1992; 49:101-108. 13. Rief W, Schaefer S, Hiller W, Fichten M. Lifetime diagnoses in patients with somatoform disorders: which care first? Eur Arch Psychiatry Clin Neurosci 1992;241:236-240. 14. Rief W, Hiller W, Geissner E, Fichter M. A two year follow up study of patients with somatoform disorders. Psychosomatics 1995;36:376-386. 15. Lilienfeld S, Van Valkenburg C, Larentz K, Akiskal H. The relationship of histrionic personality disorder to antisocial personality and somatization disorders. Am J Psychiatry 1986; 143:718-722. 16. Liskow B, Penick E, Powell B, Haefele W, Campbell J. lnpatients with Briquet's syndrome: presence of additional psychiatric syndromes and MMPI results. Compr Psychiatry 1986;27:461-470. 17. Morrison J. Histrionic personality disorder in women with somatization disorder. Psychosomatics 1989;30:433-437. 18. Smith GR. Somatization Disorder in the Medical Setting. Washington, DC: American Psychiatric Press, 1991. 19. Fabrega H, Mezzich J, Jacob R, Ulrich R. Somatoform disorder in a psychiatric setting. Systematic comparisons with depression and anxiety disorders. J Nerv Ment Dis 1988;176:431439. 20. Snyder S, Strain J. Somatoform disorders in the general hospital inpatient setting. Gen Hosp Psychiatry 1989;11:288293. 21. Rost K, Akins R, Brown F, Smith R. The comorbidity of DSM-III-R personality disorders in somatization disorder. Gen Hosp Psychiatry 1992;14:322-326. 22. Stern J, Murphy M, Bass C. Personality disorder in patients with somatization disorder. A controlled study. Br J Psychiatry 1993;163:785-789. 23. Zimmerman M, Coryell W, Pfohl B, Corenthal C, Stangl

306

D. ECT response in depressed patients with and without a DSM-II! personality disorder. Am J Psychiatry 1986;143:10301032. 24. Pfhol B, Coryell W, Zimmerman M, Stangl D. Prognostic validity self-report and interview measures of personality disorder in depressed inpatients. J Clin Psychiatry 1987;48:468-472. 25. Black D, Bell S, Hulbert J, Nasrallah A. The importance of axis II inpatients with major depression. A controlled study. J Affect Disord 1988;14:115-122. 26. Shea T, Pilkonis P, Beckham E. Personality disorders and treatment outcome in the NIMH Treatment of Depression Collaborative Research Program. Am J Psychiatry 1990; 143:711718. 27. Sanderson W, Wetzler S, Beck A, Betz E Prevalence of personality disorders in patients with major depression and dysthymia. Psychiatry Res 1992;42:93-99. 28. Garyfallos G, Adamopoulou A, Voikli M, Saitis M, Kirtsos P, Moutzoukis C. DSM-III-R personality disorders among patients with depressive and/or anxiety disorders. J Pers Disord 1994;8:320-332. 29. Klass E, DiNardo P, Barlow D. DSM-III personality diagnoses in anxiety disorder patients. Compr Psychiatry 1989; 30:251-258. 30. Nurnberg G, Raskin M, Levin P. Borderline personality disorder as a negative prognostic factor in anxiety disorders. J Pers Disord 1989;3:205-216. 31. Noyes R, Reich J, Christiansen J, Suelzer M, Pfhol B, Coryell W. Outcome of panic disorder. Relationship to diagnostic subtypes and comorbidity. Arch Gen Psychiatry 1990;47:809818. 32. Garyfallos G, Adamopoulou A, Donias S, Bitsios P, Voikli M, Paraschos A. Social phobia: comorbidity with other DSM-III-R psychiatric disorders. Encephalos 1995;32:377-386. 33. Garyfallos G, Adamopoulou A, Karastergiou A, Voikli M, Milis V, Donias S, et al. Psychiatric comorbidity in Greek patients with generalized anxiety disorder. Psychopathology (In press). 34. Spitzer R, Williams J, Gibbon M. Structural Clinical Interview for DSM-III-R (SCID). New York, NY: New York State Psychiatric Institute, Biometric Research, 1987. 35. Manos N, Hatgisavas S, Monas K, Donias S, Simos G. Validating the SCID Personality Questionnaire. Presented at the Eighth World Congress of Psychiatry; October 12-19, 1989; Athens, Greece. 36. Manos N. Adaption of the MMPI in Greece. In: Butcher J, Spielberger C (eds): Advances in Personality Assessment. Hillsdale, NJ: Erlbaum, 1985:159-208. 37. Beutler L, Crago M. Self-report measures of psychotherapy outcome. In: Lambert M, Christensen E, Dejulio S (eds): The Assessment of Psychotherapy Outcome. New York, NY: Wiley, 1983:453-497. 38. Koenigsberg H, Kaplan H, Gilmore M, Cooper A. The relationship between syndrome and personality disorder in DSM-III: experience with 2,462 patients. Am J Psychiatry 1985;142:207-212. 39. Alnaes R, Torgersen S. DSM-III symptom disorders (axis I) and personality disorders (axis II) in an outpatient population. Acta Psychiatr Scand 1988;78:348-355. 40. Mezzich J, Fabrega H, Coffman G, Haley R. DSM-III disorders in a large sample of psychiatric patients: frequency

GARYFALLOS ET AL

and specificity of diagnoses. Am J Psychiatry 1989;146:212219. 41. Adamopoulou A, Garyfallos G, Bouras N, Kouloumas G. Mental health and primary care in ethnic groups. Greek Cypriots in London: a preliminary investigation. Int J Soc Psychiatry 1990;36:243-251. 42. Garyfallos G, Adamopoulou A, Bouras N. Psychological problems in primary care: a comparison of Greeks to Greek Cypriots and English. Neurol Psychiatr 1991;12:70-79. 43. Pillowski I, Spence N. Ethnicity and illness behaviour. Psychol Med 1977;7:447-452. 44. Ierodiakonou C. Psychotherapeutic possibilities in a rural community mental health center in Greece. Am J Psychother 1983;37:544-551. 45. Graham S. Ethnic background and illness in a Pennsylvania county. Soc Probl 1956;4:76-82. 46. Escobar J, Burnam A, Karno M, Forsythe A, Golding J. Somatization in the community. Arch Gen Psychiatry 1987;44: 713-718. 47. Noyes R, Kathol R, Fisher M, Phillips B, Suelzer M, Woodman C. Psychiatric comorbidity among patients with hypochondriasis. Gen Hosp Psychiatry 1983;16:78-87. 48. Torgersen S. Genetics of somatoform disorders. Arch Gen Psychiatry 1986;43:502-505. 49. Weller R, Kapadia P, Weller F, Fristad M, Lazaroff L, Preskoru S. Psychopathology in families of children with major depressive disorders. J Affect Disord 1994;31:247-252. 50. Croyle R, Uretsky M. Effects of mood on self-appraisal of health status. Health Psychol 1987;6:239-253. 51. Weissman M, Gershon E, Kidd K, Prusoff B, Leckman J, Dibble E, et al. Psychiatric disorders in the relatives of probands with affective disorders: the Yale University National Institute of Mental Health Collaborative Study. Arch Gen Psychiatry 1984;41:13-21. 52. Kendler K, Neale M, Kessler R, Haeth A, Eaves L. Major depression and generalized anxiety disorder. Same genes (partly) different environment? Arch Gen Psychiatry 1992;52:374-383. 53. Kendler K, Waiters E, Neale M, Kessler R, Heath A, Eaves L. The structure of the genetic and environmental risk factors for six psychiatric disorders in women. Arch Gen Psychiatry 1995;52:374-383. 54. Minc C. Of new Australian patients, their medical lore and major anxieties. Med J Aust 1963;1:681-687. 55. Emerson J, Pankratz L, Joos S, Smith S. Personality disorders in problematic medical patients. Psychosomatics 1994; 35:469-473. 56. Guze S. The validity and significance of hysteria (Briquet's syndrome). Am J Psychiatry 1975;132:138-141. 57. Lilienfeld S. The association between antisocial personality and somatization disorders: a review and integration of theoretical models. Clin Psychol Rev 1992; 12:641-662. 58. Frick P, Kuper K, Silverthorn P, Cotter M. Antisocial behavior, somatization and sensation-seeking behavior in mothers of clinic-referred children. J Am Acad Child Adolesc Psychiatry 1995;34:805-812. 59. Shea T, Glass D, Pilkonis P, Watkins J, Docherty J. Frequency and implications of personality disorders in a sample of depressed outpatients. J Pers Disord 1987; 1:27-42. 60. Mavissakalian M, Hamann M. DSM-III personality disorder in agoraphobia. Compr Psychiatry 1986;27:451-479.

SOMATOFORM DISORDER, COMORBIDITY

61. Herbert J, Hope D, Bellack A. Validity of the distinction between generalized social phobia and avoidant personality disorder. J Abnorm Psychol 1992;101:332-339. 62. Mavissakalian M, Hamman M, Jones B. Correlates of DSM-III personality disorder in obsessive-compulsive disorder. Compr Psychiatry 1990;31:481-489. 63. Reich J. Effect of DSM-III personality disorders on

307

outcome of tricyclic antidepressant-treated nonpsychotic outpatients with major or minor depressive disorder. Psychiatry Res 1990;32:175-181. 64. Reich J, Perry C, Shera D, Dyck I, Vassile R, Goisman R, et al. Comparison of personality disorder in different anxiety disorder diagnoses: panic, agoraphobia, generalized anxiety and social phobia. Ann Clin Psychiatry 1994;6:125-134.