The Journal of Nervous & Mental Disease 

Volume 188(1), January 2000, pp 45-47

Medically Self-Harming Behavior and Its Relationship to Borderline Personality
Symptoms and Somatic Preoccupation among Internal Medicine Patients

[Brief Reports]

Sansone, Randy A. M.D.; Wiederman, Michael W. Ph.D.; Sansone, Lori A. M.D.

----------------------------------------------

In the psychiatric literature, there has been a long-standing association
between self-harm behavior and borderline personality disorder. Indeed, the
diagnostic criteria for borderline personality disorder in DSM-IV (American
Psychiatric Association, 1994) include two criteria for self-harm behavior:
"impulsivity in at least two areas that are potentially self-damaging" and
"recurrent suicidal behavior, gestures, or threats, or self-mutilating
behavior." In response to this association, investigators have developed a
measure of self-harm, the Self-Harm Inventory (SHI; Sansone et al., 1998),
which, as expected, has predictive value in the diagnosis of borderline
personality symptomatology. Using the Diagnostic Interview for Borderlines (DIB;
Kolb & Gunderson, 1980) as the standard, the SHI demonstrated a diagnostic
accuracy of 84% (Sansone et al., 1998). It is evident that there is some
association between self-harm behavior and borderline personality symptomatology.
However, the clinical impressions and data that support this association have
evolved primarily from experience and investigations in psychiatric settings. It
is unknown whether this association is present in the primary care setting.

Patients in medical settings may manifest self-harm behavior through the
intentional sabotage of their medical care (Sansone et al., 1997). Sansone et
al. (1997) found that nearly 7% of patients in an outpatient family medicine
setting acknowledged the intentional sabotage of their medical care through at
least one survey endorsement (1% of respondents acknowledged 4 or more such
behaviors). However, the association between medically self-harming behavior and
borderline personality disorder symptomatology was not assessed in that study.

The focus of the current study was to explore, in an ambulatory internal
medicine clinic, the association between medically self-harming behavior and
borderline personality symptomatology. Because Hudziak et al. (1996) found that
25% of patients with borderline personality disorder met diagnostic criteria for
somatization disorder, we also examined the relationship between medically
self-harming behavior and somatic preoccupation. Our hypothesis was that, as in
psychiatric settings, self-harm behaviors would demonstrate a positive
relationship to borderline personality symptoms as well as somatic preoccupation.

Methods

Participants. Participants were 33 men and 85 women (N = 118) who presented for
medical care to the ambulatory clinic of an internal medicine training program.
This program is located in a mid-sized, mid-western city and provides care to
individuals residing in two middle-to-lower socioeconomic suburban areas. All
participants were age 18 or older, cognitively intact, and sufficiently
medically stable to complete questionnaires. At check-in for a scheduled
appointment, each candidate was asked by the receptionist to participate in the
project (i.e., sample of convenience). If willing, participants were asked to
complete materials onsite, which took approximately 15 minutes. Completion of
materials functioned as informed consent.

Of the 137 candidates who were invited to participate, 118 completed measures
for a response rate of 86.1%. The large majority (91.5%) indicated being white,
non-Hispanic, with the remaining participants (8.5%) indicating black/African-American
as their racial/ethnic group. Most participants (84.7%) had attained at least a
high school diploma, with 17.8% reporting a bachelor's degree or greater for the
highest level of educational attainment.

Measures. The research booklet contained a cover page that explained the purpose
of the study (i.e., "to better understand physical symptoms and their relationship
to psychological concerns"). In addition to requests for demographic information,
the booklet contained measures for self-harm behaviors, borderline personality
symptomatology, and somatic preoccupation.

Self-Harm Inventory: The Self-Harm Inventory (SHI; Sansone et al., 1998) is a
22-item, yes/no questionnaire that explores respondents' history of self-destructive
behaviors. Items are preceded by the phrase, "Have you ever intentionally, or on
purpose" and include behaviors such as, over-dosed, cut yourself on purpose,
burned yourself on purpose, and hit yourself. Among the items in the inventory,
there are three that specifically refer to self-harm behaviors with medical
care. These are prevented wounds from healing, made medical situations worse on
purpose, and abused prescription medication.

In addition to exploring self-destructive behavior, the SHI appears to be
predictive of borderline personality symptomatology. Each endorsement on the SHI
is in the pathological direction and a total score is the summation of "yes"
responses. The endorsement of 5 or more items is highly suggestive of borderline
personality according to the diagnostic criteria of the DIB (Kolb and Gunderson,
1980). The SHI correctly identified 84.7% of subjects according to the DIB
(Sansone et al., 1998).

Personality Diagnostic Questionnaire-Revised: The borderline personality scale
of the Personality Diagnostic Questionnaire-Revised (PDQ-R; Hyler and Rieder,
1987) is an 18-item, self-report measure that is based upon the DSM-III-R
(American Psychiatric Association, 1987) criteria for borderline personality
disorder. The PDQ-R is reported as a useful screening tool for borderline
personality in both clinical (Dubro et al., 1988; Hyler et al., 1990) and
nonclinical (Johnson and Bornstein, 1992) samples including the use of the
freestanding borderline subscale (Patrick et al., 1995).

Bradford Somatic Inventory: Somatic preoccupation was measured using the
Bradford Somatic Inventory (BSI; Mumford et al., 1991), a 46-item, yes/no
questionnaire that was developed with a cross-cultural perspective. Individual
items consist of somatic symptoms most frequently reported by anxious and
depressed patients. Examples of items include: Have you had severe headaches;
Have you had pain or tension in your neck and shoulders; Has your mouth or
throat felt dry; and Have you felt a lack or energy much of the time? Two items
relating to male respondents ("Have you had difficulty getting a full erection"
and "Have you felt that you have been passing semen in your urine") were deleted
due to concerns about their relevance and applicability in the study setting.
This resulted in a total of 44 items for the scale. Scores were computed based
on the total number of symptoms endorsed.

Results

Five respondents reported ever having intentionally prevented wounds from
healing, five reported making medical situations worse on purpose, and seven
reported having abused prescription medication. These 17 endorsements of
medically self-harming behavior on the SHI were distributed among a total of 15
different respondents.

In examining the prevalence of borderline personality symptomatology among those
with versus without medically self-harming behaviors, a chi-square analysis was
undertaken. Before analysis, the medically self-harming items were eliminated
from the SHI scoring to avoid diagnostic overlap, with the remaining scale
constituting 19 items. According to the scoring of the SHI, among the subsample
reporting medically self-harming behavior (N = 15), 80.0% exceeded the cutoff
for substantial borderline personality symptomatology compared with 14.6% for
the remainder of the sample. This was a statistically significant difference
([chi]2 [1, N = 118] = 31.77, p 

With regard to scores on the PDQ-R, 86.7% of those participants reporting
medically self-harming behaviors exceeded the cutoff for substantial borderline
personality symptomatology compared with 42.7% of the remainder of the sample,
which was statistically significant using chi-square analysis ([chi]2 [1, N =
118] = 10.13, p 

Finally, in examining subsamples according to the endorsements on the BSI using
a one-way analysis of variance, those with medically self-harming behaviors had
significantly higher scores on somatic preoccupation (mean = 22.27, SD = 10.22)
compared with the remaining respondents (mean = 17.21, SD = 9.10, F[1,116] =
3.92, p d = .54).

Discussion

The findings of this study suggest that, as in psychiatric settings, self-harming
behavior is associated with borderline personality symptomatology. In comparison
with a psychiatric setting, whether self-harming behavior in the medical setting
has consistent or more frequent medical self-sabotage overtones remains unknown.
However, the general impression is that self-harming behavior, regardless of
setting or type, is frequently associated with borderline personality symptomatology.

These data indicate that medically self-harming behavior appears to be
associated with somatic preoccupation, as well. This finding needs to be
interpreted with some caution. For example, it may be that the somatic measure
used in this study, the BSI, is actually measuring somatic manifestations of a
mood disorder. If so, given that mood disorders are reported to be commonplace
among those with borderline personality (Prasad et al., 1990; Soloff et al.,
1991; Zanarini et al., 1989), the association may merely be an adjunctive
finding of individuals with this type of personality disturbance.

As a counter argument, the affective syndrome in patients with borderline
personality disorder is reported as unique (de Bonis et al., 1998; Rogers et
al., 1995). These syndromes often have interpersonal, self-conceptual, and
existential elements (e.g., self-condemnation, emptiness, abandonment fears,
negative self-view). In contrast, the BSI elicits specific somatic symptoms, not
self-related issues, which may indicate some genuine somatic preoccupation among
this subsample.

The prevalence of somatic preoccupation has not been well studied nor well
categorized in the primary care setting, but it appears to be a common
occurrence (Righter and Sansone, 1999). The PRIME-MD 1000 study (Linzer et al.,
1996) reported the prevalence of DSM-IV somatoform disorder as 14%, which
appears relatively high. In a study using physician ratings (Peveler et al.,
1997), 19% of patients exhibited unexplained physical symptoms. In a study of
newly referred patients (Van Hemert et al., 1993), 45% had poorly explained
symptoms and 38% had unexplained symptoms. In comparing somatic preoccupation
with DSM-III-R (American Psychiatric Association, 1987) somatization disorder,
Kirmayer and Robbins (1991) reported a prevalence of 1% for the former and 26.3%
for the latter.

How might all of this be explained? It may be that there are two subpopulations
of patients with substantial borderline personality symptomatology, one with
predominant psychological symptoms (i.e., the psychiatric subpopulation) and the
other with predominant physical symptoms (i.e., the primary care subpopulation).
Both share the common dynamic feature of self-destructive behavior, although the
primary care subpopulation may manifest more medically oriented behaviors.
Unlike their psychiatric counterparts in which affective symptoms are expressed
more through interpersonal, self-conceptual, and existential manifestations, the
primary care subpopulation may manifest affective symptoms through more somatic
preoccupation. Each group may correspondingly seek out providers who meet their
respective needs, with the psychiatric group seeking to resolve psychological
issues and the primary group seeking to resolve somatic concerns.

From a primary care perspective, we suggest further examination of the
relationship between somatic preoccupation and borderline personality symptomatology.
It may be that somatic preoccupation is a marker for borderline personality
symptomatology among at least some patients. However, the specific relationship
and associated risk factors need to be identified.

We have been very careful in this paper to use the term, "borderline personality
symptomatology" instead of "borderline personality disorder." First and
foremost, personality disorders are very difficult to accurately diagnose.
Second, many of the available measures appear to be over-inclusive. For example,
note the prevalence rates in our data. Among those who did not endorse medically
self-defeating behaviors, the prevalence of substantial borderline personality
symptomatology was 14.6% based on the SHI and 42.7% based on the PDQ-R. These
are likely to be inflated diagnostic figures, but what we believe is important
is the relatively greater prevalence of personality pathology among those who
endorsed medically self-defeating behaviors. We also believe that by eliminating
the medically self-defeating items when scoring the SHI for borderline
personality symptomatology, it was a more conservative measure than initially
intended.

There are a number of limitations with the current study. First, there is the
self-report nature of the data. Second, there is the difficulty of measuring a
phenomenon such as somatic preoccupation (i.e., few self-report scales exist).
Third, we did not incorporate possible covariates in the analyses, including the
presence or absence of depression (which we did not measure). Fourth, we focused
on only a few medically self-harming behaviors rather than a wide range of such
behaviors. Finally, this was a sample of convenience (although the majority of
candidates agreed to participate).

In conclusion, these data suggest that medically self-harming behaviors are
associated with borderline personality symptomatology according to two different
measures of the construct that focus on very different aspects of the syndrome.
In addition, medically self-harming behaviors are also associated with somatic
preoccupation (at least as measured by the BSI). Based on these findings, when
consulting on primary care patients, psychiatrists need to consider these
empirical relationships when evaluating patients. Whether there are genuinely
different subgroups of borderline patients according to treatment provider and
setting remains to be explored.

Randy A. Sansone, M.D.

Michael W. Wiederman, Ph.D.

Lori A. Sansone, M.D.



References

American Psychiatric Association (1987) The Diagnostic and Statistical Manual of
Mental Disorders (3rd ed, rev). Washington, DC: Author. 

American Psychiatric Association (1994) The Diagnostic and Statistical Manual of
Mental Disorders (4th ed). Washington, DC: Author. 

De Bonis M, De Boeck P, Lida-Pulik H, Hourtane M, Feline A (1998) Self-concept
and mood: A comparative study between depressed patients with and without
borderline personality disorder. J Affect Disord 48:191-197. Full Text
Bibliographic Links Document Delivery 

Dubro AF, Wetzler S, Kahn MW (1988) A comparison of three self-report questionnaires
for the diagnosis of DSM-III personality disorders. J Pers Disord 2:256-266. 

Hyler SE, Lyons M, Rieder RO, Young L, Williams JBW, Spitzer RL (1990) The
factor structure of self-report DSM-III Axis II symptoms and their relationship
to clinicians' ratings. Am J Psychiatry 147:751-757. Bibliographic Links
Document Delivery 

Hyler SE, Rieder RO (1987) Personality Diagnostic Questionnaire-Revised. New
York: New York State Psychiatric Institute. 

Hudziak JJ, Boffeli TJ, Kriesman JJ, Battaglia MM, Stragner C, Guze SB (1996)
Clinical study of the relation of borderline personality disorder to Briquet's
Syndrome (Hysteria), somatization disorder, antisocial personality disorder, and
substance abuse disorders. Am J Psychiatry 153:1598-1606. Ovid Full Text
Bibliographic Links Document Delivery 

Johnson JG, Bornstein RF (1992) Utility of the Personality Diagnostic Questionnaire-Revised
in a nonclinical sample. J Pers Disord 6:450-457. Document Delivery 

Kirmayer LJ, Robbins JM (1991) Three forms of somatization in primary care:
Prevalence, co-occurrence, and sociodemographic characteristics. J Nerv Ment Dis
179:647-655. Bibliographic Links Document Delivery 

Kolb JE, Gunderson JG (1980) Diagnosing borderline patients with a semistructured
interview. Arch Gen Psychiatry 37:37-41. Bibliographic Links Document Delivery 

Linzer M, Spitzer R, Kroenke K, Williams JB, Hahn S, Brody D, de Gruy F (1996)
Gender, quality of life, and mental disorders in primary care: Results from the
PRIME-MD 1000 study. Am J Med 101:526-533. Full Text Bibliographic Links
Document Delivery 

Mumford DB, Bavington JT, Bhatnagar KS, Hussain Y, Mirza S, Naraghi MM (1991)
The Bradford Somatic Inventory: A multiethnic inventory of somatic symptoms
reported by anxious and depressed patients in Britain and the Indo-Pakistan
subcontinent. Br J Psychiatry 158:379-386. Bibliographic Links Document Delivery

Patrick J, Links P, Van Reekum R, Mitton MJE (1995) Using the PDQ-R BPD scale as
a brief screening measure in the differential diagnosis of personality disorder.
J Pers Disord 9:266-274. Document Delivery 

Peveler R, Kilkenny L, Kinmonth AL (1997) Medically unexplained physical
symptoms in primary care: A comparison of self-report screening questionnaires
and clinical opinion. J Psychosom Res 42:245-252. Full Text Bibliographic Links
Document Delivery 

Prasad RB, Val ER, Lahmeyer HW, Gaviria M, Rodgers P, Weiler M, Altman E (1990)
Associated diagnoses (comorbidity) in patients with borderline personality
disorder. Psychiatr J Univ Ott 15:22-27. Bibliographic Links Document Delivery 

Righter EL, Sansone RA (1999) Managing somatic preoccupation. Am Fam Physician
59:3113-3120. 

Rogers JH, Widiger TA, Krupp A (1995) Aspects of depression associated with
borderline personality disorder. Am J Psychiatry 152:268-270. Ovid Full Text
Bibliographic Links Document Delivery 

Sansone RA, Wiederman MW, Sansone LA (1998) The Self-Harm Inventory (SHI):
Development of a scale for identifying self-destructive behaviors and borderline
personality disorder. J Clin Psychol 54:973-983. Full Text Bibliographic Links
Document Delivery 

Sansone RA, Wiederman MW, Sansone LA, Mehnert-Kay S (1997) Sabotaging one's own
medical care. Arch Fam Med 6:583-586. Bibliographic Links Document Delivery 

Soloff PH, Cornelius J, George A (1991) The depressed borderline: One disorder
or two? Psychopharmacol Bull 27:23-30. Bibliographic Links Document Delivery 

Van Hemert AM, Hengeveld MW, Bolk JH, Rooijmans HG, Vandenbroucke JP (1993)
Psychiatric disorders in relation to medical illness among patients of a general
medical out-patient clinic. Psychol Med 23:167-173. Bibliographic Links Document
Delivery 

Zanarini MC, Gunderson JG, Frankenburg FR (1989) Axis I phenomenology of
borderline personality disorder. Compr Psychiatry 30:149-156.