In this study, we compiled 12 previous databases to examine, among a large sample (N=972) of psychiatric and nonpsychiatric patients, the relationship, if any, between borderline personality symptomatology and sexual impulsivity. Two types of sexual impulsivity were assessed through two self-report items that queried participants about (1) having sex with individuals whom respondents hardly knew (i.e., casual sexual relationships) and (2) promiscuity. Across the entire sample, those with borderline personality symptomatology evidenced at least twice the rate of endorsement of these items compared to those without such symptomatology. This proportion was also closely approximated in the various subsamples (i.e., psychiatric, internal medicine, and other medical subsamples). These findings indicate that a substantial minority of patients with borderline personality symptomatology exhibit sexual impulsivity in the form of casual sexual relationships as well as promiscuity.
According to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR),1 one criterion for borderline personality disorder (BPD) is impulsivity in at least two functional areas, including sexual behavior. Sexual impulsivity refers to engaging in sexual behavior without much forethought. Examples include having sexual relationships with strangers, with numerous partners (i.e., promiscuity), and not using adequate protection (i.e., high-risk sexual behaviors).
In addition to the DSM-IV-TR,1 a number of other assessments for BPD have a criterion related to sexual impulsivity, including the Borderline Syndrome Index,2 the Diagnostic Interview for Borderline Patients,3 the Borderline Personality Disorder Scale,4 the Diagnostic Interview for Personality Disorders,5 the Zanarini Rating Scale for Borderline Personality Disorder,6 the Personality Disorder Examination,7 the Personality Diagnostic Questionnaire,4,8 and the Self-Harm Inventory (SHI).9 However, is there genuine empirical support for this particular criterion?
In the literature, there appears to be very modest support. In a case report, Pelsser10 described an individual with borderline personality and sexual promiscuity. In addition, O’Boyle11 reported promiscuity among a case series of four female patients with BPD. With regard to clinical impressions, Stone12 stated that “over 25 percent” of his outpatients with BPD had histories of promiscuity. In an empirical study, Allan13 classified 71 women as “low risk” versus “high risk” according to their sexual behavior; those in the latter subsample were significantly more likely to be diagnosed with BPD. In a second empirical study, Hull, Clarkin, and Yeomans14 examined 71 hospitalized women with BPD; 46 percent had entered into sexual relationships with partners that they did not know well. As for meta-analytic approaches, Neeleman15 examined six empirical studies and concluded that patients with BPD evidence heightened sexual impulsivity. In addition to the preceding data, according to two investigative groups, comorbid substance abuse appears to heighten the risk for sexual impulsivity among those with BPD.16,17
However, not all studies have confirmed the finding of sexual impulsivity among patients with BPD. For example, Zanarini et al18 found that 41 percent of such patients avoided sexual relationships. Conversely, among 85 patients diagnosed with compulsive sexual behavior, Lloyd et al19 encountered only one patient who met the criteria for BPD.
To summarize, in the area of BPD and sexual impulsivity, bonafide empirical studies are few in number, sample sizes are generally small, and some data actually refute this relationship. In the following study, we combined several previous study samples to magnify the sample size, and examined the relationship between BPD and two specific forms of sexual impulsivity—casual sexual relationships and promiscuity.
Participants. Participants consisted of 763 women and 209 men (N=972) from three general types of settings: psychiatric settings (one inpatient site) (n=365),20–23 internal medicine (one outpatient site) (n=337),24–28 and other types of medical settings (i.e., bariatric medicine, n=121; family medicine, n=35; and buprenorphine treatment center, n=114).29–31
The psychiatric samples consisted of 105 men and 260 women, with a mean age of 35.91 years (SD=11.02). Of these 365 respondents, 75.6 percent were Caucasian, 18.6 percent African-American, 3.0 percent Hispanic, 0.8 percent Asian, and 1.4 percent Native-American; 0.5 percent indicated “Other” for Race/Ethnicity.
The internal medicine samples consisted of 12 men and 325 women, with a mean age of 42.35 years (SD=11.98). Of these 337 respondents, 85.4 percent were Caucasian, 9.0 percent African-American, 0.9 percent Hispanic, 1.2 percent Asian, and 0.3 percent Native-American; 2.1 percent indicated “Other” for Race/Ethnicity.
The other medical samples consisted of 92 men and 178 women, with a mean age of 40.21 years (SD=10.57). Of these 270 respondents, 88.9 percent were Caucasian, 7.4 percent African-American, 2.2 percent Hispanic, 0.4 percent Asian, and 0.3 percent Native-American; 0.7 percent indicated “Other” for Race/Ethnicity.
Procedure. Participants were drawn from 12 individual studies that utilized both the Personality Diagnostic Questionnaire-R32 (PDQ-R) or Personality Diagnostic Questionnaire-4 (PDQ-4)8 and the SHI.9 All participants were recruited from the caseloads of the clinical investigators. For each cohort, data were collected over an 18 to 24 month period.
The items for the current study relating to sexual impulsivity were as follows: 1) the PDQ-R32 or PDQ-48 item referring to casual sexual relationships (both contain the identical query, “I have done things on impulse that can get me into trouble…[such as] having sex with people I hardly know”) and (2) SHI item 11 (“Have you ever intentionally, or on purpose,…been promiscuous [i.e., had many sexual partners]?”). We divided the sample according to cut-off scores on the PDQ (i.e., a score of 5 or higher is highly suggestive of BPD), and then compared those with versus without borderline personality symptomatology with regard to the two queries relating to sexual impulsivity.
Personality Diagnostic Questionnaire. Of the 972 respondents, 441 (45.4%) exceeded the cut-off score indicative of clinically significant borderline personality symptomatology. Of these respondents, 205 (46.5%) endorsed the PDQ item, “I have done things on impulse…[such as]…having sex with people I hardly know.” In contrast, of the 531 respondents who did not exceed the PDQ cut-off score for BPD symptomatology, only 88 (16.6%) endorsed the preceding PDQ item pertaining to casual sex. The difference between the proportions of respondents endorsing the casual sex PDQ item in each group was highly statistically significant: X2 (df=1)=58.67, p<0.0001.
Self-Harm Inventory. A total of 958 participants completed the SHI. Of the 434 (45.3%) such respondents who exceeded the PDQ cut-off score for borderline personality symptomatology, 184 (42.4%) endorsed the SHI item, “Have you ever intentionally, or on purpose, been promiscuous (i.e., had many sexual partners)?” In contrast, of the 524 (54.7%) respondents who did not exceed the PDQ cut-off score for BPD symptomatology, only 108 (20.6%) endorsed the preceding SHI promiscuity item. The difference between the proportions of respondents endorsing the SHI promiscuity item in each group was highly statistically significant: X2 (df=1)=52.15, p<0.0001.
Additional analyses. To determine whether these findings held across the three different subsamples, we conducted the same analyses separately for each. In the psychiatric samples (n=365), 270 (74.0%) respondents exceeded the PDQ cut-off score for borderline personality symptomatology. Compared to psychiatric respondents who did not exceed the PDQ cut-off score (n=95), the borderline personality symptomatology group was much more likely to endorse the PDQ item relating to casual sexual relationships [47.4% vs. 22.1%, X2 (df=1)=19.32, p<0.001] as well as the SHI promiscuity item [42.6% vs. 24.5%, X2 (df=1, n=359)=17.12, p<0.001].
In the internal medicine samples (n=337), 79 (23.4%) respondents exceeded the PDQ cut-off score for borderline personality symptomatology. Compared to respondents who did not exceed the PDQ cut-off score (n=258), the borderline personality symptomatology group was much more likely to endorse the PDQ item relating to casual sexual relationships [38.0% vs. 12.8%, X2 (df=1)=11.12, p<0.001] as well as the SHI promiscuity item [35.9% vs. 18.5%, X2 (df=1, n=332)=9.40, p<0.001].
In the other medical samples (n=270), 92 (34.1%) respondents exceeded the PDQ cut-off score for borderline personality symptomatology. Compared to respondents who did not exceed the PDQ cut-off score (n=178), the borderline personality symptomatology group was much more likely to endorse the PDQ item relating to casual sexual relationships [51.1% vs. 19.1%, X2 (df=1)=17.18, p<0.001] as well as the SHI promiscuity item [46.2% vs. 21.8%, X2 (df=1, n=265)=15.63, p<0.001].
These data indicate that, compared to those without borderline personality symptomatology, those with these symptoms evidence a higher prevalence rate of two types of sexual impulsivity—casual sexual relationships and multiple sexual partners. In all samples of patients with borderline personality symptomatology, a significant minority reported sexually impulsive behaviors. In addition, while the prevalence of substantial BPD symptomatology was markedly higher in the psychiatric subsample than in the other subsamples, as expected, within all samples the prevalence of these behaviors was approximately double among those with borderline personality symptomatology compared to those without these symptoms. These data support the concept that a substantial minority of patients with borderline personality symptomatology evidence casual sexual behavior as well as promiscuity, thereby supporting the inclusion of this criterion in various assessments for BPD.
Note that in these samples, there was a very high prevalence of borderline personality symptomatology. This may be explained in several ways. First, both PDQ measures are known to be over-inclusive. In other words, these measures tend to over-diagnose individuals with BPD. Because of our awareness of this phenomenon, we elected to use the term borderline personality symptomatology throughout this paper, rather than borderline personality disorder. Second, the psychiatric samples were solicited from inpatient settings, where the rates of BPD would be expected to be high. Third, all of the internal medicine participants were obtained from a resident-provider clinic. Existing payor data from this clinic indicate that 65 percent of the patients have government insurance and another 10 percent are fee-for-service (i.e., sliding fee scale). These payor types appear to harbor high levels of psychological dysfunction, as evidenced by the high prevalence of borderline personality symptomatology.
Despite the use of numerous datasets from a number of different clinical settings, findings were consistent across subsamples. This observation appears to strengthen the ability to generalize these data to various clinical populations.
This study has a number of potential limitations including the self-report nature of the data, the over-inclusive diagnostic nature of the PDQ, participants’ interpretation and/or understanding of the queries (e.g., the term promiscuity), and the compilation of various data sets. In addition, our queries about sexual impulsivity were limited to casual sexual relationships and promiscuity—the items that were present in the two reviewed measures. Other items might be explored in future studies such as high-risk behaviors. However, to our knowledge, this is the largest sample to date in this field of study and findings are consistent with all previous studies—i.e., that many patients with borderline personality symptomatology (i.e., a significant minority) evidence sexual impulsivity—yet, many do not.
Randy A. Sansone, Dr. Sansone is a Professor in the Departments of Psychiatry and Internal Medicine at Wright State University School of Medicine in Dayton, Ohio, and Director of Psychiatry Education at Kettering Medical Center in Kettering, Ohio.
Michael W. Wiederman, Dr. Wiederman is a professor in psychology in the Department of Human Relations at Columbia College in Columbia, South Carolina.