Pre-DSM-lll. In the Midtown Manhattan study, Srole et al. (1962) reported a prevalence rate of 6% for sociopathy. In the Stirling County study using DSM-I diagnoses, Leighton (1963) reported that 11% of males and 5% of females received a sociopathic diagnosis. Merikangas and Weissman (1991) reviewed the prevalence rates for personality disorder reported in pre-DSM-III studies that permitted the exclusion of alcoholism and drug abuse (which were classified as personality disorders in some older systems) from other personality disorders (Bremer, 1951; Essen-Moller et al., 1956; Langner and Michael, 1963; Leighton, 1959). They concluded that, in spite of nonuniformity in diagnostic definition, the reported rates were quite similar. Approximately six to nine percent of the samples were characterized as having a major personality disturbance. In these early data reviewed by Merikangas and Weissman, there is an indication that the overall sex ratio for personality disorders is about equal, with differences for specific disorders. Prevalence is fairly even across age groups with a slight increase in later life, higher rates in urban than rural populations, and higher rates in lower socioeconomic groups compared to higher groups.
Post-DSM-lll. Since the publication of DSM-III in 1980 there have been several studies that provided data on the prevalence of having any personality disorder. Nestadt and his coworkers, in a series of reports (1990, 1991, 1992) detailed the results of a follow-up assessment for personality disorders of the Epidemiologic Catchment Area (ECA) study. At the Baltimore site of the ECA, a total of 3,481 individuals were interviewed with the Diagnostic Interview Schedule (Robins et al., 1981), the General Health Questionnaire (Goldberg, 1974), and the Mini-Mental Status Exam (Folstein et al., 1975) as part of the ECA. Subsequent to the ECA data collection, the Clinical Reappraisal was carried out and included all subjects who had been ''screened positive'' for psychopathology and a random sample of 17% of the 3,481 original respondents. Of the 1,086 subjects selected for inclusion in the Clinical Reappraisal, 810 agreed to participate for a response rate of 75%. Board-certified or board-eligible psychiatrists, blind to first stage information, interviewed the subjects. The psychiatrists used the Standardized Psychiatric Examination (SPE), a semistructured interview that averaged one and one-half to three hours to complete. The authors described the SPE as assessing personal history, medical and psychiatric problems and present mental status; it includes an inventory and direct question approach. Nestadt et al. (1993) reported a prevalence of 5.9% for a definite diagnosis of personality disorder and a prevalence of 9.3% for their combined ''definite'' plus ''provisional'' diagnostic categories.
Zimmerman and Coryell (1989a) reported the rates of DSM-III personality disorders assessed through the use of the Structured Interview for DSM-III Personality Disorders (SIDP) (Stangl et al., 1995) among a nonpatient sample of 797 individuals. There are several features of the study that qualify somewhat the interpretation of their results. The sample that they studied was a mixture of relatives of normal controls (n = 185), relatives of schizophrenic probands (n = 131), relatives of probands with psychotic depression (n = 247), relatives of probands with nonpsychotic depression (n = 235), and relatives of probands with other psychiatric disorders (n = 10). Eleven individuals refused the SIDP interview, yielding the final sample size of 797. Their sample is problematic because certain personality disorders may have a familial relationship to axis I disorders. To the extent that a personality disorder has a familial relationship to an axis I disorder, the rate of the personality disorder among first-degree relatives of probands with the axis I disorder will be elevated and results may not generalize to the general population. Therefore, the prevalence of some personality disorders may be inflated in the sample that Zimmerman and Coryell studied. Approximately three-fourths of the sample was interviewed by telephone (72.9%) and one-fourth by face-to-face interviews (27.1%). The authors found no difference in the frequency of axis II diagnoses between telephone and face-to-face interviews. The prevalence of any DSM-III personality disorder diagnosis using the SIDP, including mixed personality disorder, was 17.9%. These investigators also adminis tered the Personality Diagnostic Questionnaire (Hyler, 1983) to their subjects (Zimmerman and Coryell, 1990). Prevalence rates were fairly similar, but the PDQ produced higher rates of schizotypal, compulsive, dependent, and borderline personality disorders. The SIDP yielded higher rates of antisocial and passive-aggressive personality disorders. More individuals were diagnosed with a personality disorder by the SIDP, but the PDQ diagnosed multiple personality disorders more often. The results of the PDQ are not tabulated separately because the same sample was utilized for both instruments.
Casey and Tyrer (1986) carried out a study of 200 randomly selected community residents in the United Kingdom. They administered the Present State Examination (9th edition) (Wing et al., 1974) for axis I disorders and the Personality Assessment Schedule (PAS; Tyrer et al., 1979) for the assessment of personality disorders. The PAS obtains information from an informant as well as the subject. Ordinal ratings are made on 24 personality traits. A computer algorithm is applied to determine personality disorder diagnoses according to the International Classification of Diseases (World Health Organization, 1978). Personality disorders were diagnosed in 26 of their subjects (13%). The personality disorder with the highest prevalence was explosive personality, which probably corresponds most closely to the DSM-III diagnoses of antisocial personality disorder or intermittent explosive disorder on axis I. There were no differences in prevalence between their urban and rural samples and males and females did not differ in the overall rate of personality disorder; women had a higher prevalence of asthenic personality disorder. They did not find a relationship between neurotic depression or the combined category of neuroses and personality disorder. Subjects with a personality disorder were found to have significantly poorer social functioning.
Reich and colleagues (1989) conducted a random mailed survey of the adult population of Iowa City, Iowa. Surveys were mailed to 401 subjects; 240 surveys were returned for a response rate of 62.1%. Data were collected using the Personality Diagnostic Questionnaire (Hyler et al., 1983). Diagnoses were based on meeting the requisite number of criteria for a given personality disorder and meeting the impairment distress scale criterion for the presence of a disorder. The rate for receiving any axis II disorder was 11%.
Maier and his coworkers (1992) studied a sample of 452 subjects in the mixed urban-rural Rhein-Main area of Germany. Subjects were recruited to serve as controls in a family study of affective disorders and schizophrenia. The sample of control probands (n = 109) was selected randomly without regard to psychiatric status and was stratified by age, sex, residential area, and educational status. Maier et al.'s sample included the probands, their mates, and all first-degree relatives over age 20 who agreed to take part in face-to-face interviews, for a total of 452. Subjects were administered the Schedule for Affective Disorders and Schizophrenia: Lifetime Version (Mannuzza et al., 1986) for axis I diagnoses and the Structured Clinical Interview for DSM-III Axis II (Spitzer et al., 1987) for axis II diagnoses. The rate of receiving any personality disorder diagnosis was 10.0%. They followed DSM-III-R diagnostic criteria with the exception of exclusionary criteria; for example, schizotypal personality disorder could be diagnosed in the presence of schizophrenia.
We recently administered the SIDP-IV to a sample of 693 male twins who were members of the Vietnam Era Twin (VET) Registry. The SIDP-IV yields DSM-IV
diagnoses. The sample was randomly selected from the VET Registry (details of construction of the Registry have been reported elsewhere, Eisen et al., 1989; Henderson et al., 1990). The men were between the ages of 45 and 55 years. The sample is limited because it includes only males, participants were screened for mental health at military induction during early adulthood, and the observations of members of a twin pair are not statistically independent. The prevalence of having any DSM-IV personality disorders among these men was 7.6%. These data are presented here for the first time and are identified in the tables as ''Lyons and Jerskey.''
Was this article helpful?