Patterns of Self-Harm Behavior Among Womenwith Borderline Personality Symptomatology:Psychiatric versus Primary Care SamplesRandy A. Sansone, M.D., Michael W. Wiederman, Ph.D.,Lori A. Sansone, M.D., and Doug Monteith, M.D.Abstract: The current study investigated differences in self-harm behavior among individuals with borderline personalitysymptomatology from two different clinical settings. Partici-pants were women, between the ages of 18 and 45, from anoutpatient mental health setting or a primary care setting. Eachparticipant completed the Self-Harm Inventory (SHI) and theborderline personality scale of the Personality DiagnosticQuestionnaire-Revised (PDQ-R). Using a predetermined cut-off for substantial borderline personality symptomatology onthe SHI, group comparisons with2analyses indicated that“overdosed” and “hit self” were significantly more common inthe mental health subsample whereas “abused laxatives” wassignificantly more common in the primary care subsample.Using a predetermined cut-off for borderline personality on thePDQ-R, chi-square analyses indicated that “overdosed” and“hit self” remained significantly more common in the mentalhealth subsample. Despite these differences, there was remark-able similarity of symptoms between groups. The implicationsof these findings are discussed. © 2000 Elsevier Science Inc.IntroductionFew epidemiological studies exploring self-harmbehaviors are available. Past research has ad-dressed specific behaviors such as suicide at-tempts/completions [1–4] and self-inflicted burns[5] as well as focused on particular populationssuch as those from inpatient mental health settings[6] or mentally disabled populations [7,8]. How-ever, self-harm behaviors pertain to individualsfrom a variety of clinical settings, including mentalhealth and primary care. Indeed, some researchindicates that many individuals with emotional dis-turbance are seen predominantly in primary caresettings [9].Whether there are behavioral differences amongself-harming individuals seen in mental health ver-sus primary care settings is unknown. If there are,such differences may influence or determine theindividual’s chosen setting for treatment. Becausethere appears to be an association between self-harm behaviors and borderline personality, thistype of diagnostic identification might enable faircomparison among subjects from different settings.The following study was undertaken to explorewhether there are differences in reported self-harmbehaviors between two populations of individualswith borderline personality symptoms—one froman outpatient mental health setting and one from aprimary care setting.MethodSamplesInitial samples consisted of women ages 18–45,recruited from an outpatient psychiatric clinic (n⫽43) and a primary care medical clinic (n⫽72). Thisentire sample was part of a previous study explor-ing body weight, body image, and depression (San-sone RA, Wiederman MW, Sansone LA, MonteithD, manuscript submitted for publication).Departments of Psychiatry and Internal Medicine at WrightState University School of Medicine in Dayton, Ohio (R.A.S.),Psychiatry Education,Kettering MedicalCenter (R.A.S.),Depart-ment of Human Relations at Columbia College in Columbia,South Carolina (M.W.W.), Kettering Medical Center PhysiciansIncorporated in Dayton, Ohio (L.A.S.), University of TexasSouthwest Medical Center in Dallas, Texas (D.M.).Address reprint requests to: Randy A. Sansone, M.D., Sy-camore Primary Care Center, 2115 Leiter Road, Miamisburg,OH, 45342.General Hospital Psychiatry 22, 174–178, 2000174© 2000 Elsevier Science Inc. All rights reserved.ISSN 0163-8343/00/$–see front matter655 Avenue of the Americas, New York, NY 10010PII S0163-8343(00)00074-8For the psychiatric sample, each consecutive pa-tient who was scheduled for evaluation at auniversity-based outpatient psychiatric clinic wasinvited to participate. This clinicsetting is located ina general office building rather than a medical col-lege campus. All patient candidates in this clinic arepre-screened for stability as outpatients and mostare non-psychotic. The majority are seeking psychi-atric evaluation for affective disorders or medica-tion evaluation through a referral from a non-physician mental health professional. From thisoriginal sample, all women who were invited toparticipate in the previous study agreed to do so,but one did not complete the questionnaire afterenlistment.The primary care sample of women was re-cruited from an HMO setting by a female familyphysician. Candidates were approached if they metthe following general inclusion criteria: 1) 18 to 45years of age; 2) cognitively intact; and 3) not inacute medical distress. This sample, recruited dur-ing a previous study (Sansone RA et al., manuscriptsubmitted for publication), consisted of two sub-samples: 1) controls (i.e., individuals who pre-sented for non-emergent medical services) and 2)patients whowere identifiedat the time of primary-care service as having depression or having beenprescribed antidepressant medication. Participantswere not recruited consecutively (i.e., the samplewas one of convenience). Of the 88 women whowere approached for the previous study, 72 agreedto participate.MethodAll participants completed the Self-Harm Inventory(SHI) [10], a 22-item, self-report, yes/no question-naire that explores subjects’ self-harm behaviors.Each item is preceded by the statement, “Have youever intentionally, or on purpose...” Items in-clude, “overdosed,” “burned yourself,” “attemptedsuicide,” “cut yourself,” and “engaged in sexuallyabusive relationships.” A score on the SHI repre-sents the number of endorsed items (i.e., “yes” re-sponses), each of which represents a pathologicalresponse (i.e., there are no non-pathological itemsin the inventory). Scores on the SHI have beenshown to highly correlate with borderline person-ality symptomatology [10] as measured by both theborderline personality scale of the Personality Di-agnostic Questionnaire-Revised (PDQ-R) [11] aswell as the Diagnostic Interview for Borderlines(DIB) [12]. In previous research, using a cut-offscore of 5 resulted in 83.7% of cases being accu-rately classified as borderline or not using the DIBas the criterion [10].Ultimately 27 women in the psychiatric sampleand 24 women in the primary care sample scored inthe range on the SHI suggestive of borderline per-sonality symptomatology (Table 1). In the
View Full Document
Unlocking...