• comorbidity;
  • diagnostic classification;
  • somatoform disorder


  1. Top of page
  2. Abstract

Abstract  A total of 283 patients with somatoform disorder (SFD) seen in a psychiatry clinic were surveyed and their diagnostic subtypes, demographic features, and comorbidities, analyzed. The results indicate that: (i) SFD comprises 5.8% of first-visit outpatients; (ii) undifferentiated SFD (USFD) and SFD not otherwise specified (SFD-NOS) account for the majority of patients; (iii) there are 1.7-fold more women than men; (iv) age of onset is lower in patients with somatization disorder or body dysmorphic disorder and higher in patients with hypochondriasis or pain disorder; (v) the mean number of years of education was 11.2 years; and (vi) comorbid illness were seen in 24.8% of patients, and included mood disorder, anxiety disorder, and personality disorder, as well as borderline intellectual functioning and mental retardation. The data indicate that the majority of patients with SFD are given a diagnosis of residual category, such as USFD or SFD-NOS, and that the age of onset varies depending on the diagnostic subtype. SFD was more frequently seen in women, associated with comorbidities.


  1. Top of page
  2. Abstract

Somatoform disorders (SFD) have been studied or surveyed mainly in primary care settings1,2 or in the general population.3,4 These studies indicate that SFD is a common disease with a prevalence of 4–5%,3,4 and in the field of primary care it is an important disease, constituting 15–20% of all patients.1,2 However, only a few studies of SFD have been conducted in psychiatric practice,5–7 and a wide range of basic characteristics such as demographic features or incidence by diagnostic subtype have been reported.

According to DSM-IV-TR criteria, the common feature of SFD is ‘the presence of physical symptoms that suggest a general medical condition and not fully explained by a general medical condition’,8 and can be classified into diagnostic subtypes of somatization disorder (SZD), undifferentiated somatoform disorder (USFD), conversion disorder (CD), pain disorder (PD), hypochondriasis (HC), body dysmorphic disorder (BDD), and somatoform disorder not otherwise specified (SFD-NOS).

However, as suggested by the statement that ‘the grouping of these disorders in a single section is based on clinical utility rather than on assumption regarding shared etiology or mechanism’, it is reasonable to state that SFD is a heterogeneous set of diseases with various disease syndromes.

Because SFD patients are clinically heterogeneous, there are no reports of any treatment that is highly effective against all SFD, and effective pharmacotherapy has been reported for only certain SFD subtypes of HC,9 BDD,10 and PD.11

Thus, we surveyed 283 patients with SFD seen as outpatients in the psychiatry department at Niigata University Medical and Dental Hospital, and analyzed their demographic features (age, sex, and years of education), the proportion of each diagnostic subtype, and comorbidities.


  1. Top of page
  2. Abstract

The subjects were 283 patients (105 men, 178 women) with the diagnosis of SFD according to DSM-IV-TR criteria among 4838 first-visit patients seen in the Department of Psychiatry, Niigata University Medical and Dental Hospital during the 6-year period April 1998–March 2004. The DSM-IV (-TR) diagnoses were made at the time of the initial visit using an approximately 1-h semistructured interview by psychiatrists with >10 years of clinical experience. These patients were also surveyed for age, sex and years of education. This study had been approved by the Ethical Committee of the Niigata University Graduate School of Medical and Dental Sciences and all subjects gave written informed consent after the nature of the survey and the statistical uses of the survey data had been explained to them.

For statistical analysis, the t-test was used, and P < 0.05 was considered significant.


  1. Top of page
  2. Abstract

Demographic characteristics

From a total sample of 4838 patients who visited the Department of Psychiatry, Niigata University Medical and Dental Hospital during the 6-year period, 283 (5.8%) received a DSM-IV (-TR) diagnosis of SFD (Table 1).

Table 1.  Demographic characteristics (n = 283)
Diagnosisn%Women : menAge of onset (Mean ± SD)Age at evaluation (Mean ± SD)Years of education (Mean ± SD)
  1. NOS, not otherwise specified.

Somatization disorder93.28.017.1 ± 4.922.8 ± 8.411.3 ± 3.1
Undifferentiated somatoform disorder15053.01.938.3 ± 19.043.4 ± 19.411.5 ± 2.5
Conversion disorder289.91.826.6 ± 15.928.2 ± 16.810.4 ± 3.9
Pain disorder248.51.449.1 ± 18.451.5 ± 18.012.3 ± 2.6
Hypochondriasis51.81.564.8 ± 9.165.6 ± 8.910.8 ± 1.6
Body dysmorphic disorder41.41.021.7 ± 6.621.8 ± 6.611.8 ± 2.6
Somatoform disorder NOS6322.31.338.1 ± 23.438.3 ± 23.510.3 ± 3.6
Total2831001.737.6 ± 20.540.9 ± 20.811.2 ± 3.0

Out of these 283 patients with SFD, 150 patients (53.0%) had USFD and 63 patients (22.3%) had SFD-NOS. Other diagnoses were BDD (1.4%), HC (18%), and SZD (3.2%).

Age of onset and age at first visit

The mean age of onset and age at first visit in the patients studied were 37.6 ± 20.5 years and 40.9 ± 20.8 years, respectively (Table 1). By diagnostic subtype, the mean ages of onset for SZD and BDD were 17.1 ± 4.9 years and 21.7 ± 6.6 years, respectively. In contrast, the age of onset was much higher in HC at 64.8 ± 9.1 years, indicating a difference according to diagnostic subtype. Similarly, differences were seen in age at evaluation. The mean age of onset for USFD, which was the most common SFD, was 38.3 ± 19.0 years.

The duration from age of onset to age at first visit was approximately 3 years for SFD overall. The duration was short at <1 year for HC, BDD, and SFD-NOS, and long at approximately 5 years for SZD and USFD.

Male : female ratio

Among the subjects, 178 were women, which was 1.7-fold higher than male patients (Table 1). With regard to diagnostic subtype, there was a preponderance of women in all subtypes, except BDD, for which there were equal numbers of men and women. The ratio of women to men ranged from 8.0 (SZD) to 1.3 (SFD-NOS).

Mean number of years of education

The mean number of years of education was 11.2 ± 3.0 years, especially low in SFD-NOS patients (10.3 ± 3.6 years) and CD patients (10.4 ± 3.9 years; Table 1). Even in the PD patients, who had the highest number of years of education among the SFD patients, the duration was 12.3 ± 2.6 years.

Comorbidities in SFD patients

Comorbidities were seen in 70 patients (24.8%; Table 2). In axis I diagnosis, mood disorder was most common, and existed in 20 patients (7.1%, 12 patients with depressive disorder NOS, seven patients with major depressive disorder, and one patient with bipolar I disorder), followed by anxiety disorder in nine patients (3.2%, three patients with anxiety disorder NOS, two with panic disorder with agoraphobia, two with general anxiety disorder, one with specific phobia, one with obsessive-compulsive disorder), five patients with dissociative disorder NOS (1.8%), five patients with cognitive disorder (1.8%, three patients with cognitive disorder NOS, two patients with dementia NOS), four patients with adjustment disorder (1.4%), and one patient each with psychotic disorder NOS, alcoholism, and Asperger disorder (0.4%).

Table 2.  Comorbidity (n = 283)
  1. NOS, not otherwise specified.

Any axis I disorder  4616.3
 Mood disorders  207.1
  Major depressive disorder72.5  
  Depressive disorder NOS124.2  
  Bipolar I disorder10.4  
 Anxiety disorders  93.2
  Panic disorder20.7  
  Specific phobia10.4  
  Obsessive-compulsive disorder10.4  
  Generalized anxiety disorder20.7  
  Anxiety disorder NOS31.1  
 Dissociative disorder NOS  51.8
 Cognitive disorders  51.8
  Dementia NOS20.7  
  Cognitive disorder NOS31.1  
 Adjustment disorders  41.4
 Psychotic disorder NOS  10.4
 Alcohol dependence  10.4
 Asperger disorder  10.4
Any axis II disorder  248.5
 Personality disorders  124.2
  Paranoid personality disorder10.4  
  Schizotypal personality disorder10.4  
  Borderline personality disorder72.5  
  Avoidant personality disorder10.4  
  Dependent personality disorder10.4  
  Personality disorder NOS10.4  
 Mental retardation  51.8
 Borderline intellectual functioning  72.5

In axis II diagnosis, 12 patients had borderline intellectual functioning or mental retardation (4.2%, seven patients with borderline intellectual functioning, and five patients with mental retardation), and 12 patients had personality disorder (4.2%, seven patients with borderline, one patient each with paranoid, schizotypal, avoidant, dependent and NOS).


  1. Top of page
  2. Abstract

To our knowledge, surveys or studies of SFD in a large sample are restricted to the report by Yamada and Nakajima7 and the report by Garyfallos et al.5 Both surveys were performed in psychiatric practice approximately 10 years prior to the present survey, according to DSM-III-R diagnostic criteria.12 The former involved 310 Japanese patients with SFD out of 4144 outpatients overall, and the latter involved 175 Greek patients out of 1448 outpatients.

The present study shows that SFD constitutes 5.8% of all psychiatric outpatients. The incidence of SFD is lower than that in the primary care setting.1,2 This finding is consistent with the two previous studies. This difference is caused possibly by the reflection of the tendency for SFD patients to seek a primary care physician rather than a psychiatrist.1,13 The incidence of SFD in the present study is essentially in agreement with the figure of 7.5% reported by Yamada and Nakajima7 but only one half of the 12.1% reported by Garyfallos et al.5 It has been reported that the development of somatization symptoms is affected by cultural differences,14,15 so the difference in the incidence of SFD may be due to cultural differences between Greece and Japan. We need further research to determine the reason for these differences.

Figure 1 compares SFD diagnostic subtypes between our data and the data of Yamada and Nakajima7 and Garyfallos et al.5 In the present study, USFD and SFD-NOS constituted approximately 80% of all SFD, and other diagnostic subtypes, such as BDD, HC, and SZD, were uncommon. The trend that many of those with SFD had diagnosis of SFD-NOS or USFD, and that other diagnostic subtypes were uncommon, was also seen in studies conducted both for psychiatric7 and for primary care patients.1 Thus, these trends are considered to be a feature of SFD patients. Garyfallos et al. reported that the incidence of SZD and CD was high,5 a trend not seen in the present study. However, this trend may not be a true trend in SFD patients but rather, as explained by Kroenke et al.,16 may be attributed to the overly inclusive diagnostic criteria for SFD-NOS and USFD, in contrast to the strict diagnostic criteria for SZD. Thus, the SFD diagnostic subtypes are not practical.17–21 For this reason, Kroenke et al. proposed diagnostic criteria of a multisomatoform disorder: ‘there exist 3 or more medically unexplained, currently bothersome physical symptoms for more than 2 years’.16 Interestingly, there were differences between the present investigation and the previous studies in the proportion of diagnostic subtypes other than SFD-NOS and USFD.5,7 These differences include: (i) SZD and CD were more common in the study by Garyfallos et al.,5 compared to the other two studies; (ii) USFD is more frequent in the present study compared to the other studies; and (iii) HC is less frequent in the present study compared to the other two studies. While previous studies investigated both in psychiatric practice as in the present study, DSM-III-R rather than DSM-IV (-TR) was used as the diagnostic criteria. With respect to potential differences in the diagnostic criteria, change in the SFD diagnostic criteria during the revision from DSM-III-R to DSM-IV involved only SZD. There is a high degree of diagnostic agreement in SZD between DSM-III-R and DSM-IV,22 so it is unlikely that differences in the diagnostic criteria need to be considered here.


Figure 1. Diagnostic subtypes of somatoform disorder.

Download figure to PowerPoint

In contrast, as indicated in Fig. 1 and Table 3, the patients studied by Garyfallos et al., compared to the other studies, had a lower mean age and included a higher proportion of women. Thus, the aforementioned difference (i) may be due to subjects' characteristics regarding age and gender and, as indicated here, the cultural differences in the manifestation of somatization symptoms. The difference (ii) may be due to decreases in the threshold for visiting a psychiatrist that occurred during the 10 years that separates the survey conducted by Yamada and Nakajima and the current survey. That is, the present survey may have included patients with SFD especially with the diagnostic subtype of USFD who would previously have seen a general physician. Finally, regarding point (iii), the current survey does not address the difference. Based on the incidence of HC, a larger survey is desirable.

Table 3.  Comparison of sex ratio and age at evaluation
DiagnosisWomen : menAge at evaluation (Mean ± SD)
Present studyYamada and Nakajima7Garyfallos et al.5Present studyYamada and Nakajima7Garyfallos et al.5
  • Value not reported in original article but calculated based on the figure indicated.

  • NOS, not otherwise specified; NR, not reported.

Somatization disorder8.02.0NR22.8 ± 8.432.1 ± 14.6NR
Undifferentiated somatoform disorder1.92.0NR43.4 ± 19.442.5 ± 15.8NR
Conversion disorder1.84.8NR28.2 ± 16.833.3 ± 14.6NR
Pain disorder1.41.7NR51.5 ± 18.038.3 ± 17.1NR
Hypochondriasis1.51.2NR65.6 ± 8.945.8 ± 15.4NR
Body dysmorphic disorder1.01.0NR21.8 ± 6.629.2 ± 0.0NR
Somatoform disorder NOS1.30.8NR38.3 ± 23.540.0 ± 18.3NR
Total1. ± 20.839.936.1 ± 11.3

In the present survey the mean disease onset and age at evaluation varied greatly depending on the diagnostic subtype, as SZD and BDD were seen in younger patients and HC in older patients. The former is in agreement with the notation in DSM-IV-TR,8 while the reasons for the latter are unclear. The reason for the longer period from disease onset to evaluation in the SZD and USFD is unclear because the reason for visiting a psychiatry department was not surveyed. It is possible that after onset of the disease, the patient may have visited a primary care physician prior to a psychiatrist.

Table 3 shows a comparison of the data from the present and previous studies for age at evaluation and male : female ratio in SFD patients. Among the patients surveyed here, women constituted 1.7-fold the number of men, and among the diagnostic subtypes, SZD, USFD, and CD were particularly notable for the high women : men ratio, in agreement with the previous studies, and also with the notation in DSM-IV-TR.

In the patients studied, axis I comorbidities were seen in 16.3% of the patients, mainly mood disorders and anxiety disorders, which comprised 7.1% and 3.2%, respectively.

The axis I comorbidities in SFD have been reported by Rief et al. in 30 patients with SFD6 and by the previously mentioned report from Garyfallos et al.5 Comorbidities of mood disorders and anxiety disorders were seen in 63% and 36%, respectively, in the former report and in 49% and 30%, respectively, in the latter report. The present percentages are lower than these figures, but the present data are for presence or absence of comorbidities at the initial evaluation, and thus may be an underestimate.

In the present patients axis II comorbidities were seen in 8.5%; personality disorders were seen in 4.2%. Reports on axis II comorbidities in SFD are limited to the report by Snyder and Strain in a psychiatry liaison service at a general hospital23 and the previously mentioned report by Garyfallos et al. Personality disorders were seen in 48% in the former and 63% in the latter report. The latter study reported that most personality disorders were histrionic or borderline, and that 67% of CD patients had a coexistent personality disorder. Compared to these earlier studies the incidence of personality disorders in the present study is low. Reasons for this observation may be that as with the axis I comorbidities, the incidence may have been an underestimate because the data were taken at the initial evaluation and that compared to the study of Garyfallos et al. the number of CD was low.

The present study surveys the diagnostic classification, demographic features, and comorbidities in SFD patients seen in psychiatric outpatients. The diagnoses of USFD or SFD-NOS, the residual category of SFD, were given to the majority of patients, and only a few patients were diagnosed with other subtypes, suggesting that there is room for improvement in the diagnostic subtypes. Among the SFD subtypes, there exist differences in age at disease onset and other parameters, but there also exist common features among the SFD such as the greater number of women and the shorter duration of education. One reason for the disagreement between the present data and prior studies may be the possibility that there are cultural differences (e.g. peoples in some culture may be more likely to report multisomatoform complaints) that may affect the incidence of various SFD subtypes.

The present study is a cross-sectional study based on initial evaluation and is likely to underestimate comorbidities. It addition, it remains possible that some patients with an initial diagnosis of SFD had the diagnosis changed later. In the future, it will be necessary to conduct a longitudinal and prospective study to investigate the diagnostic stability and prognosis of SFD. It is also desirable to conduct a study involving a larger number of patients over a longer period to clarify the relationship between somatization symptoms and personality in patients with psychiatric disease.


  1. Top of page
  2. Abstract
  • 1
    Fink P, Sorensen L, Engberg M, Holm M, Munk-Jorgensen P. Somatization in primary care: Prevalence, health care utilization, and general practitioner recognition. Psychosomatics 1999; 40: 330338.
  • 2
    De Waal MWM, Arnold IA, Eekhof JAH, Van Hemert AM. Somatoform disorders in general practice. Prevalence, functional impairment and comorbidities with anxiety and depressive disorders. Br. J. Psychiatry 2004; 184: 470476.
  • 3
    Escobar JI, Burnam MA, Karno M, Forsythe A, Golding JM. Somatization in the community. Arch. Gen. Psychiatry 1987; 44: 713718.
  • 4
    Swartz M, Blazer DG, George L, Landerman R. Somatization disorder in a community population. Am. J. Psychiatry 1986; 143: 14031408.
  • 5
    Garyfallos G, Adamopoulou A, Karastergiou A et al. Somatoform disorders: Comorbidities with other DSM-III-R psychiatric diagnoses in Greece. Compr. Psychiatry 1999; 40: 299307.
  • 6
    Rief W, Schaefer S, Hiller W, Fichter MM. Lifetime diagnoses in patients with somatoform disorders: Which came first? Eur. Arch. Psychiatry Clin. Neurosci. 1992; 241: 236240.
  • 7
    Yamada N, Nakajima S. DSM-III-R disorders in a large sample of psychiatric patients; frequency, sex difference, and age. Seishin Igaku 1996; 38: 767775 (in Japanese).
  • 8
    American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th edn, text revision. American Psychiatric Association, Washington, DC, 2000.
  • 9
    Fallon BA, Liebowitz MR, Salman E et al. Fluoxetine for hypochondriacal patients without major depression. J. Clin. Psychopharmacol. 1993; 13: 438441.
  • 10
    Phillips KA, Albertini RS, Rasmussen SA. A randomized placebo-controlled trial of fluoxetine in body dysmorphic disorder. Arch. Gen. Psychiatry 2002; 59: 381388.
  • 11
    Sindrup SH, Bach FW, Madsen C, Gram LF, Jensen TS. Venlafaxine versus imipramine in painful polyneuropathy: A randomized, controlled trial. Neurology 2003; 60: 12841289.
  • 12
    American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 3rd edn, text revision. American Psychiatric Association, Washington, DC, 1987.
  • 13
    Grabe HJ, Meyer C, Hapke U et al. Specific somatoform disorder in the general population. Psychosomatics 2003; 44: 304311.
  • 14
    Escobar JI, Gomez J, Tuason VB. Depressive phenomenology in north and south American patients. Am. J. Psychiatry 1983; 140: 4751.
  • 15
    Mezzich JE, Raab ES. Depressive symptomatology across the Americas. Arch. Gen. Psychiatry 1980; 37: 818823.
  • 16
    Kroenke K, Spitzer RL, De Gruy FV III et al. Multisomatoform disorder. An alternative to undifferentiated somatoform disorder for the somatizing patient in primary care. Arch. Gen. Psychiatry 1997; 54: 352358.
  • 17
    Escobar JI, Gara MA. DSM-IV somatoform disorders: Do we need a new classification? Gen. Hosp. Psychiatry 1999; 21: 154156.
  • 18
    De Gucht V, Fischler B. Somatization: A critical review of conceptual and methodological issues. Psychosomatics 2002; 43: 19.
  • 19
    Martin RD. The somatoform conundrum: A question of nosological values. Gen. Hosp. Psychiatry 1999; 21: 177186.
  • 20
    Rief W, Hiller W. Somatization: Future perspectives on a common phenomenon. J. Psychosom. Res. 1998; 44: 529536.
  • 21
    Rief W, Hiller W. Toward empirically based criteria for classification of somatoform disorders. J. Psychosom. Res. 1999; 46: 507518.
  • 22
    Yutzy SH, Cloninger CR, Guze SB et al. DSM-IV field trial: Testing a new proposal for somatization disorder. Am. J. Psychiatry 1995; 152: 97101.
  • 23
    Snyder S, Strain J. Somatoform disorders in the general hospital inpatient setting. Gen. Hosp. Psychiatry 1989; 11: 288293.