A particular Minnesota Multiphasic Personality Inventory (MMPIJ high-point configuration was found to describe borderline personality disorder within a general alcoholic inpatient population. We also partially replicated, in an Icelandic sample of alcoholics, the findings of Nace, Saxon, and Shore (1983) on the clinical and demographic characteristics of borderline patients. Subjects were 51 male and female inpatient alcoholics consecutively admitted to an Icelandic psychiatric hospital. Gunderson's Diagnostic Interview for Borderlines and the Michigan Alcoholism Screening Test were used to diagnose borderline personality disorder and alcoholism, respectively. As predicted, a particular MMPI high-point configuration characterized alcoholics with borderline personality disorder but did not characterize nonborderline patients. Borderline patients were also more likely to be younger, to abuse other psychotropic drugs besides alcohol, and to have a family history of mental illness. The rinding of a borderline personality-disordered subgroup among alcoholics contests the view of alcoholism as a unitary disease and has implications for treatment.Global conceptualizations of alcoholism that focus on ethanol consumption and dependency have been criticized by a number of authors (Morey, Skinner, & Blashfield, 1984;Pattison, Sobell, & Sobell, 1977;Skinner, Glaser, & Annis, 1982). It has been suggested that there are different types of alcoholics and that efficient classification should be a fundamental step in the scientific study of the alcoholic (Kline & Snyder, 1985;Svanum & Dallas, 1981).One suggested diagnostic type of alcoholic is the schizoid personality with severe alcohol dependence (Morey et al., 1984).Borderline personality-disordered alcoholics form a similar diagnostic subgroup, and they have been recommended for a special form of therapy (Hellman, 1981). In a careful study, 12.8% of alcoholics in an inpatient alcoholic program were found to have borderline personality disorder (Nace, Saxon, & Shore, 1983). The borderline patients were distinguished, among other symptoms, by more frequent polydrug use, by combined use of alcohol and drugs, by familial mental illness, by prior psychotherapy, and by a continuing need for outpatient treatment. Although they were younger than the nonborderline patients, they This article is based on the first author's master's thesis at Acadia University under the supervision of the second author. We wish to thank Elizabeth Blackmer and Patrick O'Neill for their helpful comments on this study. For their generous help with the project, we wish to thank

Methodological variations in the scoring and interpretation of the MMPI and their effects on discrimination between borderline and non‐borderline personality disordered alcoholics were investigated. Subjects were 49 male and female inpatient alcoholics in an Icelandic psychiatric hospital. Gunderson's Diagnostic Interview for Borderlines and the Michigan Alcoholism Screening Test were used to diagnose borderline personality disorder and alcoholism, respectively. Scoring and interpretation of the MMPI were varied in terms of the use and non‐use of high F‐scale profiles, and their impact on the frequency of various code types among borderline and non‐borderline personality disordered alcoholics was considered. It was found that such methodological variations do not affect the frequency of some profile types, and, consequently, the discrimination between the diagnostic groups. Studies and coding systems should consider methodological variations in the scoring and interpretation of MMPI profiles and their consequent effects on diagnosis.

No abstract