Content area
Full Text
Journal of Personality Disorders, 4(3), 257-272, 1990
'& 1990 The Guilford Press
ESTIMATING THE PREVALENCE OF BORDERLINE PERSONALITY DISORDER IN THE COMMUNITY
Marvin Swartz, MD, Dan Blazer, MD, PhD, Linda George, PhD, and Idee Winfield, PhD
The authors use a new diagnostic algorithm derived from the
Diagnostic Interview Schedule (the DIS/Borderline Index) to
identify a borderline personality disorder among 19- to
55-year-olds at the Duke site of the Epidemiologic Catchment
Area project. A criterion score of 1 1 or more symptoms from the 24-item DIS/Borderline Index identifies 1.8% of the
sample. The borderline diagnosis is significantly higher among females, the widowed, and the unmarried; and there is a trend
toward the diagnosis in younger, non-White, urban, and poorer respondents. Extensive psychiatric comorbidity and
high use of mental health services are found in the borderline group.
Borderline personality disorder is a clinically recognized and relatively well-defined personality disorder. Although there has been considerable prog
ress in studying borderline personality disorder in clinical populations, few
estimates of the prevalence and correlates of borderline personality disorder in the community are available.
Estimates of the community prevalence of borderline personality disorder
have been hampered by criterion variance and selection bias. Prevalence estimates have varied widely as a function of the stringency of criteria for the diagnosis and the type of setting in which estimates have been made.
Using Kernberg's (1984) broad clinical construct of borderline personality organization, Gunderson (1984) estimated that 15% of the general popula
tion suffer from a borderline spectrum disorder. Employing case criteria
akin to DSM-III, community prevalence estimates have ranged from 0.2% (Merikangas 6k Weissman, 1986: Weissman & Meyers, 1980) to 2-4% (Gun derson 6k Zanarini, 1987). In clinical settings prevalence tends to increase
This research was supported in part by grant from the A. W. Mellon Foundation to the Duke
University Medical Center Clinical Epidemiology Unit, and by grant from the NIMH
(MH35386) to the Department of Psychiatry, Duke University Medical Center. An earlier
version of this paper presented at the 141st Annual Meeting of the American Psychiatric
Association. Montreal, Quebec. May 10, 1988.
Address all correspondence to Dr. M. Swartz, Psychiatric Epidemiology and Health Services
Research Program, Box 3173. Duke University Medical Center. Durham, North Carolina
27710.
257
258 SWARTZ ET AL.
with...