Bryan Jewell

Compulsive Buying and Borderline Personality Symptomatology

In this study, the authors investigated the relationship between compulsive buying and borderline personality disorder (BPD) symptomatology--two disorders possibly linked through impulsivity. Using a survey methodology in a cross-sectional consecutive sample of nonemergent female outpatients from an obstetrics/gynecology clinic, the authors assessed compulsive buying with the compulsive buying scale (CBS) and BPD symptomatology through the BPD scale of the personality diagnostic questionnaire-4 (PDQ-4) and the self-harm Inventory (SHI). In this sample, 8% of Caucasian women and 9% of African-American women scored positively for compulsive buying. The correlations between scores on the CBS and the PDQ-4 and SHI were 0.43 and 0.41, respectively-both statistically significant at the p < .001 level. Compared to Caucasian women, African-American women demonstrated statistical associations between the BPD measures and the CBS scale scores that were significantly larger. Findings suggest relationships between compulsive buying and BPD, particularly among African-American women.

Compulsive Buying: Associations with Self-Reported Alcohol and Drug Problems

Randy A. Sansone, MD,1,2,3 Joy Chang, BS,4 Bryan Jewell, MD,5 Martin Sellbom, PhD6 1Department of Psychiatry, Wright State University School of Medicine, Dayton, Ohio 2Department of Internal Medicine, Wright State University School of Medicine, Dayton, Ohio 3Department of Psychiatry Education, Kettering Medical Center, Kettering, Ohio 4Wright State University School of Medicine, Dayton, Ohio 5Department of Obstetrics/Gynecology, Wright State University School of Medicine, Dayton, Ohio 6Department of Psychology, The University of Alabama, Tuscaloosa, Alabama

An examination of racial differences in self-harm behavior

The prevalence of diverse and multiple self-harm behaviors according to race/ethnicity has been infrequently explored. In the present study, using a cross-sectional design and a self-report survey methodology, we examined the prevalence rates of and statistically significant differences in multiple self-harm behaviors among 200 White and 146 African-American women from an obstetrics/gynecology clinic. Compared with African-American women, we found that White women reported generally higher prevalence rates of self-harm behaviors (20 of 21) and significantly higher percentages with regard to seven specific self-harm behaviors. In addition, we found support that self-harm behavior reflects the same latent construct across both groups.

Compulsive buying: Relationship with body mass index

Compulsive buying has historically been associated with various self‐regulatory disturbances, including eating pathology (e.g., binge eating). Therefore, a relationship between scores on a measure of compulsive buying, the Compulsive Buying Scale (CBS), and body mass index (BMI) in adulthood was hypothesized.

Childhood trauma and compulsive buying

Abstract Objective. Childhood trauma has been empirically associated with various types of self-regulatory difficulties in adulthood. However, according to the extant literature, no study has examined relationships between various types of childhood trauma and compulsive buying behavior in adulthood. Methods. Using a self-report survey methodology in a cross-sectional consecutive sample of 370 obstetrics/gynecology patients, we examined five types of childhood trauma before the age of 12 years (i.e. witnessing violence, physical neglect, emotional abuse, physical abuse, sexual abuse) in relationship to compulsive buying as assessed by the Compulsive Buying Scale (CBS). Results. All forms of trauma demonstrated statistically significant correlations with the CBS. Using a linear regression analysis, both witnessing violence and emotional abuse significantly contributed to CBS scores. Further analyses indicated that race did not moderate the relationship between childhood trauma and compulsive buying. Conclusions. Findings indicate that various forms of childhood trauma are correlated with compulsive buying behavior, particularly witnessing violence and emotional abuse.

Preventing wounds from healing and borderline personality symptomatology.

To the Editor: Borderline personality disorder is a personality dysfunction characterized by chronic self-harm behavior. In psychiatric populations, self-harm behavior in borderline personality disorder traditionally manifests as self-mutilation (eg, cutting, scratching, burning oneself), sadomasochistic relationships with others, and suicide attempts. However, in medical settings, self-harm behavior may take on more somatic overtones and medical themes, including preventing wounds from healing. A relationship between intentionally preventing wounds from healing and borderline personality disorder symptomatology has been reported in both psychiatric1 and internal medicine samples,1,2 but not in obstetrics/gynecology samples. Method. Participants consisted of 370 consecutive obstetrics/gynecology outpatients who ranged in age from 18 to 61 years (mean = 26.33, SD = 7.51). Exclusion criteria were medical or cognitive impairment of sufficient severity to preclude the successful completion of a survey. Participants were primarily white (54.3%) or African American (39.4%). Most had never been married (72.7%), with 16.5% being married, 6.9% divorced, 3.3% separated, and 0.6% widowed. Approximately 88% had at least completed high school, with 13.3% having completed a college degree. Most reported government (79.2%) or private insurance (5.5%), with the remaining 15.3% reporting self-pay or no insurance. As patients arrived at the clinic, one researcher (J.C.) solicited each for participation, informally assessed exclusion criteria, and invited candidates to complete a 4-page survey. The cover page of the survey contained the elements of informed consent, and completion of the survey was presumed to be implied consent (specified on the cover page). The survey contained (1) the Self-Harm Inventory,3 which has 1 particular query, “Have you ever intentionally, or on purpose, prevented wounds from healing?” and (2) the borderline personality disorder scale of the Personality Diagnostic Questionnaire-4 (PDQ-4),4 a 9-item, true/false, self-report version of the criteria for borderline personality disorder as described in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition. Data were collected in October 2010. The project was approved by an institutional review board. Results. In this sample, 13 participants (3.5%) reported preventing wounds from healing and 75 participants (20.3%) met criteria for borderline personality symptomatology according to the PDQ-4, using the traditional cutoff score of 5. The correlation between preventing wounds from healing and borderline personality symptomatology was r = 0.30. Likewise, as determined via χ2 analysis, the prevalence of prevented wounds from healing among participants with borderline personality symptomatology was statistically significantly higher than among participants without borderline personality symptomatology, χ21 = 26.76, P < .001 (N = 370). The potential limitations of this study include the self-report methodology, possible overinclusiveness of the borderline personality measure (ie, risk of false positives), and an atypical sample type (eg, low-income single women). However, in this sample, preventing wounds from healing was reported by nearly 1 in 25 participants and was associated with borderline personality symptomatology—a finding of high relevance to the general health care of women. It may be that preventing wounds from healing is functioning as a self-harm equivalent in the context of borderline personality disorder.

Rage: associations with compulsive buying.

In the extant literature, associations between compulsive buying and comorbid disorders of impulse control have been infrequently studied. McElroy et al. [1] examined a sample of 20 psychiatric patients with compulsive buying and reported that 40% had adjunctive impulse control disorders. Mueller et al. [2] examined 171 patients with compulsive buying who presented for treatment trials (i.e., a treatmentseeking sample) and found that 11% had intermittent explosive disorder. Because of the scant existing literature and the possible common underlying dynamic of disinhibition (i.e., disinhibition has been empirically related to impulsivity [3–5], and impulsivity has been empirically related to various dysfunctional behaviors including aggression [6], compulsive buying [7], substance abuse [4], binge drinking [8] and bulimia nervosa [5]), we examined in this study associations between compulsive buying and rages, either rage reactions or road rages— two phenomena related to impulsivity [6,7]. Participants were 350 consecutive females, ages 18 years or older, who were seeking nonemergent care through a university-affiliated obstetrics/gynecology outpatient clinic. We elected this sample because of accessibility, female gender and its more naturalistic nature (i.e., not a treatmentseeking sample for pathological buying). Exclusion criteria were intellectual, medical, psychiatric and/or cognitive impairment of sufficient severity that would preclude the successful completion of a survey (i.e., participants who appeared at the outset to be too impaired were not invited to participate). As patients arrived at the clinic, one researcher (JC) solicited each, informally assessed exclusion criteria and invited candidates to complete a four-page survey. The cover page of the survey contained the elements of informed consent, and completion of the survey was presumed to be implied consent (specified on the cover page). The research survey contained (1) two items to assess lifetime history of rages (i.e., “Have you ever had any rage reactions—i.e., outof-control amplified anger?” and “Have you ever had any road rages—i.e., violent uncontrolled anger during driving?”), and (2) the Compulsive Buying Scale (CBS), a seven-

Prescription Medication Abuse Among Women in an Obstetrics/Gynecology Clinic

To the Editor: In a previous study,1 we confirmed among internal medicine outpatients a relationship between the self-reported abuse of prescription medications and 2 self-report measures of borderline personality symptomatology. We found no differences in gender,2 which is in contrast to the extant literature on borderline personality suggesting that males invariably evidence higher rates of alcohol/substance misuse.3–7 In the present study, we examined the relationship between the self-reported abuse of prescription medications, borderline personality symptomatology, and alcohol abuse in an obstetrics/gynecology sample. Method. Participants were female outpatients, aged 18 years or older, seeking nonemergent care in a university-affiliated obstetrics/gynecology clinic. Exclusion criteria were psychiatric, medical, cognitive, or intellectual difficulties of a severity that would preclude the candidate’s ability to complete a survey. As patients arrived at the clinic, one researcher (J.C.) solicited each, informally assessed exclusion criteria, and invited candidates to complete a 4-page survey. The cover page of the survey contained the elements of informed consent, and completion of the survey was presumed to be implied consent (specified on the cover page). We used item 13 of the Self-Harm Inventory8 to assess prescription medication abuse (“Have you ever intentionally, or on purpose, abused prescription medication?”). To assess borderline personality symptomatology, we used the borderline personality scale of the Personality Diagnostic Questionnaire-4 (PDQ-4),9 a self-report measure that is based directly on the criteria for borderline personality in the DSM-IV. We used item 6 of the Self-Harm Inventory to determine history of alcohol abuse (“Have you ever intentionally, or on purpose, abused alcohol?”).8 Data were collected in October 2010. This project was approved by an institutional review board. Results. The 371 participants ranged in age from 18 to 61 years (mean = 26.35, SD = 7.47). Participants were primarily white (55.0%) or African American (38.8%). Most had never been married (72.8%), with the remaining individuals being married (16.8%), divorced (6.6%), separated (3.3%), or widowed (0.5%). Approximately 88% had at least completed high school, with 13.5% having completed a college degree. Most participants reported government insurance (78.4%) or private insurance (6.1%), with the remaining 15.5% reporting self-pay or no insurance. The prevalence of self-reported prescription substance abuse, borderline personality symptomatology, and alcohol abuse in this sample was 11.5%, 20.5%, and 22.4%, respectively. The correlation between the abuse of prescription medication and presence of borderline personality symptomatology per the PDQ-4 borderline personality scale was 0.36 (P < .001). Likewise, the correlation between the abuse of prescription medication and the abuse of alcohol was 0.40 (P < .001). The limitations of this study include the self-report nature of all data, including personality disorder diagnosis (eg, risk of false positives), and the somewhat atypical nature of the sample (eg, single women of low socioeconomic status). However, findings indicate that prescription substance abuse is reported by a significant minority of patients and demonstrates relationships with both borderline personality symptomatology and the abuse of alcohol. These interrelationships are important in terms of fully assessing patients who present with any 1 of these 3 syndromes.

Hair pulling and borderline personality symptomatology among obstetrics/gynecology outpatients

To the Editor: In a previous study of internal medicine outpatients, we found a statistically significant relationship between hair pulling and borderline personality symptomatology,1 suggesting associations with impulsive behavior rather than compulsive behavior.2 In this study, we reexamined this association, as well as potential racial differences, in a consecutive sample of obstetrics/gynecology outpatients. Method. Participants were consecutive female outpatients, aged 18 years or older, who were seeking nonemergent care through a university-affiliated obstetrics/gynecology outpatient clinic. Exclusion criteria were intellectual, medical, psychiatric, and/or cognitive impairment of sufficient severity to preclude the successful completion of a survey. A total of 373 patients participated in the study. The mean age of participants was 26.44 years (SD = 7.47), with ages ranging from 18 to 61 years. Most participants were white (54.2%) or African American (39.1%). Approximately 88% had at least a high school diploma, and 13% had completed college. The majority were never married (71.9%), 16.9% were married, 6.8% were divorced, and the remainder were separated or widowed. About 80% reported government insurance, 5.5% were privately insured, and 14.3% had no insurance or elected to self-pay. As patients arrived at the clinic, one researcher (J.C.) solicited each, informally assessed exclusion criteria, and invited candidates to complete a 4-page survey. The cover page of the survey contained the elements of informed consent, and survey completion was presumed to be implied consent (specified on the cover page). In the survey, we inquired about (1) demographics; (2) hair pulling (ie, with yes/no response options, “As an adult, have you engaged in hair pulling [the repeated urge to pull out scalp hair, eyelashes, eyebrows, or other body hair, resulting in bald patches]?); and (3) borderline personality symptomatology using 2 self-report assessments—the borderline personality disorder scale of the Personality Diagnostic Questionnaire-4 (PDQ-4)3 and the Self-Harm Inventory (SHI).4 The study was approved by 2 institutional review boards. Results. Twenty-seven participants (7.2%) reported hair pulling. Among the participants, 59.3% with hair pulling were PDQ-4–positive (cutoff score of 5) versus 17.3% without hair pulling (χ21 = 27.14, P < .001; N = 373). Likewise, 48.0% of participants with hair pulling were SHI-positive (cutoff score of 5) versus 18.5% without hair pulling (χ21 = 12.39, P < .001; N = 373). In examining racial differences, PDQ-4–positive white participants did not exhibit a statistically significant higher frequency of hair pulling (42.9% vs 25.3%; χ21 = 2.07, P = .15; N = 200), whereas PDQ-4–positive African American women did (80.0% vs 8.3%; χ21 = 41.53, P < .001; N = 143). The same pattern emerged with SHI-positive participants: nonsignificance in white women (41.7% vs 25.4%; χ21 = 1.53, P = .22; N = 200), but statistical significance in African American women (60.0% vs 9.9%; χ21 = 20.17, P < .001; N = 142). Findings expand on our previous study in internal medicine outpatients1 and affirm a relationship between hair pulling and borderline personality symptomatology, but only in African American women. The potential limitations of this study include the self-report nature of the data, potential overinclusiveness of the borderline personality symptomatology measures, and sample characteristics (ie, low-income women). However, the sample was consecutive, naturalistic, and large. Findings suggest that hair pulling may be associated with borderline personality symptomatology and that there may be racial differences.

Mental health care utilization and compulsive buying

Compulsive buying appears to be highly comorbid with other psychiatric disorders, particularly mood and anxiety disorders (Black, Repertinger, Gaffney, & Gabel, 1998; McElroy et al., 1994; Mueller et al., 2010; Schlosser, Black, Repertinger, & Freet, 1994). Therefore, it would seem likely that compulsive buyers would evidence higher levels of mental health care utilization than others – a previously unexamined relationship in the literature and the focus of the present investigation. Participants in this study were consecutive female outpatients, ages 18 years or older, who were seeking nonemergent care through a university-affiliated obstetrics/ gynaecology outpatient clinic. Exclusion criteria were intellectual, medical, psychiatric and/or cognitive impairment of sufficient severity that would preclude the successful completion of a survey. As patients arrived at the clinic, one researcher (JC) solicited each, informally assessed exclusion criteria and invited candidates to complete a four-page survey. The cover page of the survey contained the elements of informed consent, and completion of the survey was presumed to be implied consent (specified on the cover page). The research survey contained: (1) four yes/no items to assess mental health care utilization that are typical inquiries in the clinical setting (i.e. ‘Have you ever been... seen by a psychiatrist, hospitalized in a psychiatric hospital, in counseling, on medication for your nerves?’), and (2) the Compulsive Buying Scale (CBS), a seven-item self-report measure with Likert-style response options (Faber & O’Guinn, 1992). The CBS has good criterion, face and construct validities (Faber & O’Guinn, 1992). Total scores were calculated based on the average of the seven items, with possible total scores ranging from 1 to 5. In the current sample, the internal consistency reliability estimate was .75 (coefficient alpha). This project was approved by an institutional review board. The average age of the 373 participants was 26.44 years (SD = 7.47), ranging from 18 to 61 years. Most participants were Caucasian (54.2%) or African American (39.1%). Approximately 88% of participants had at least a high school diploma and 13% had completed college. The majority were never married (71.9%), with 16.9% being married, 6.8% divorced, and the remaining separated or widowed. About 80% reported government insurance, 5.5% were privately insured and 14.3% had no insurance or elected to self-pay. We calculated point bi-serial correlations for the individual items reflecting mental health care utilization to account for their binary distributions. To account for familywise error due to multiple analyses, we set a conservative alpha at .013 (.05/4 correlations). The results indicated that CBS scores were positively associated with a history of being in counselling (r = .21, p < .001), being treated by a psychiatrist (r = .17, p = .001) and psychiatric hospitalization (r = .15, p = .004), but not with being prescribed psychotropic medication (r = .11, p = .038). A global index of mental health care utilization, reflected in a count of the number of positive responses to the four questions, was also examined. CBS scores were significantly correlated with this index (r = .20, p < .001). Mental health care utilization and compulsive buying