The rise in the number of reported cases of dissociative disorders, particularly dissociative identity disorder, in the United States is interpreted by some as a result of the better recognition by clinicians of these diagnoses and by others as a result of overdiagnosis (DSM-IV, pp. 479 and 486). Several studies have been conducted to assess the frequency of dissociative disorders in psychiatric inpatients. Rates vary between 5.0% and 58.3% for dissociative disorders in general and between 0.5% and 12.0% for dissociative identity disorder (1–7 and 1995 personal communication from Knudson).
The aim of our study was to determine the frequency of dissociative disorders among adult inpatients in a Dutch general psychiatric hospital, using the Structured Clinical Interview for DSM-IV Dissociative Disorders, Revised (SCID-D) (8). Since there is no available indication of the reliability of the SCID-D in an inpatient population, our second aim was to determine the interrater reliability of the SCID-D. To test the interrater reliability, both blind interviewing and blind rating had to be involved.
The Dissociative Experiences Scale (9) is a brief, 28-item, self-report questionnaire designed to screen for dissociative pathology. The items on this scale relate to experiences of disturbance in identity, memory, awareness, and cognition as well as feelings of depersonalization and derealization or associated feelings such as absorption. The Dissociative Experiences Scale is the most widely used self-report measure for dissociation.
The SCID-D (8) is a semistructured diagnostic interview. It evaluates the presence and severity of symptoms in five areas relevant to the dissociative disorders.
From September 1994 to October 1995, all consecutively admitted adult patients (N=270) at the Psychiatric Hospital Willibrord (situated in the West of the Netherlands) were invited by their clinician to participate. Verbal and written informed consent was obtained from each patient after the procedures were fully explained. Characteristics such as age, sex, length of stay, number of admissions, and clinical discharge diagnosis were gathered from all admitted patients.
The Dissociative Experiences Scale was administered independently by the research assistant, usually between 1 and 2 weeks after admission. Patients scoring 25 and higher as well as a random selection of patients (one out of every four) scoring lower than 25 were approached for a further SCID-D interview. The SCID-D was also administered to patients with a clinical diagnosis of dissociative disorder and a Dissociative Experiences Scale score lower than 25. The research assistant presented all subjects randomly to the interviewers.
All interviews were audio/videotaped to allow independent rating. Both the interviewer and the rater were blind to all research data as well as to the ratio of subjects with high Dissociative Experiences Scale scores to those with low scores.
A total of 43 SCID-D interviews were independently scored by both the interviewer and the rater. Interrater reliability was assessed for the presence or absence of a dissociative disorder and the severity rating of the five dissociative symptoms. The frequency of dissociative disorders was calculated on the basis of the number of patients who received this diagnosis from both the interviewer and the rater.
Participants who completed the Dissociative Experiences Scale were compared with those who did not participate or did not complete the Dissociative Experiences Scale. For these analyses, two-tailed t tests were used for interval data and chi-square tests were used for the nominal data. Interrater reliability was calculated by using kappa for nominal data and weighted kappa for ordinal data.
Out of the 270 consecutively admitted patients, 28 patients were excluded because they were discharged within 48 hours (13 patients), didn’t speak Dutch (11 patients), or died before entering the study (four patients). A total of 242 patients were eligible for the study; 120 patients (49.6%) did not participate: 91 patients refused; in 19 cases the psychiatrist refused permission to contact the patient; five patients left the hospital unexpectedly; three patients were unable to complete the Dissociative Experiences Scale; one patient was excluded because he participated in the pilot study; and one patient committed suicide before being interviewed. A total of 122 (50.4%) of the 242 eligible patients, 64 men and 58 women, successfully completed the Dissociative Experiences Scale.
The characteristics of participants were compared with those of the nonparticipants. There were no significant differences in sex, age, marital status, length of stay, or number of admissions. Participants did not differ from nonparticipants in the number who had a discharge diagnosis of psychotic disorder (excluding schizophrenia), depression, posttraumatic stress disorder, substance use disorder, and axis II personality disorders. Significantly more of the nonparticipants (N=48, 40.0%) than participants (N=27, 22.1%) were diagnosed as having schizophrenia (χ2=8.48, df=1, p<0.01).
The mean Dissociative Experiences Scale score of the 122 participating patients was 19.95 (median=13.93, SD=18.13, range=0.00–86.07). Thirty-six patients (29.5%) scored 25 or higher on the Dissociative Experiences Scale.
Fifty-six patients were interviewed with the SCID-D: 34 patients with a Dissociative Experiences Scale score of 25 or higher (16 women and 18 men), a random selection of 21 patients with a Dissociative Experiences Scale score below 25 (seven women and 14 men), and one woman with a clinical diagnosis of a dissociative disorder and a Dissociative Experiences Scale score below 25. Two patients with a high Dissociative Experiences Scale score and one other patient with a clinical diagnosis of dissociative disorder and a Dissociative Experiences Scale score below 25 refused to participate in the interview.
Interrater reliability was established on the basis of 43 SCID-D interviews. A very high agreement (weighted kappa) was reached between interviewer and rater on the five severity ratings: amnesia (kappa=0.96), depersonalization (kappa=0.92), derealization (kappa=0.96), identity confusion (kappa=0.98), and identity alteration (kappa=0.85) (all significant at p≤;0.001).
Total agreement was reached between interviewer and rater on the absence or presence of dissociative disorders (kappa=1.0, z=6.56). Total agreement was reached as well on the type of dissociative disorder.
Ten patients (8.2%), all women, received a dissociative disorder diagnosis (95% confidence interval=3.9%–15.1%); two of these women (1.6%) met the criteria for dissociative identity disorder, and eight (6.6%) were diagnosed as having a dissociative disorder not otherwise specified. Two male patients (1.6%) had factitious dissociative identity disorder.
On the basis of this study, we estimate that the frequency of dissociative disorders among psychiatric inpatients willing to participate in such a study is 8.2%. Among 122 patients, two (1.6%) had a dissociative identity disorder. These rates seem to be lower than those found previously in most North American studies (1–4), but they are similar to previous findings in European studies (5, 7, and 1995 personal communication from Knudson).
These variations need further analysis. They might be due to differences in the methodologies used (e.g., the choice of clinical diagnostic instrument) (6) or to cultural differences in the interpretation of symptoms.
Some limitations need to be mentioned. First, more of the nonparticipants than participants had a clinical diagnosis of schizophrenia at discharge. As far as we know, this selectivity is a general problem in clinical psychiatric research. Second, the study group is rather small. Therefore, our data need confirmation in larger samples.
Received March 10, 1997; revisions received June 2 and Dec. 1, 1997, and Sept. 7, 1999; accepted Sept. 25, 1999. From the Psychiatric Hospital Willibrord, Heiloo, the Netherlands; and the Department of Psychiatry, Vrije Universiteit. Address reprint requests to Dr. Draijer, Department of Psychiatry, Vrije Universiteit, Valeriusplein 9, 1075 BG Amsterdam, Netherlands; firstname.lastname@example.org (e-mail). Supported by grant NR:4343 from the Dutch National Foundation of Mental Health.