Journal of Personality Disorders, 24(6), 812–822, 2010

© 2010 The Guilford Press

COMORBID ATTENTION-DEFICIT/
HYPERACTIVITY DISORDER IN BORDERLINE
PATIENTS DEFINES AN IMPULSIVE SUBTYPE
OF BORDERLINE PERSONALITY DISORDER
Marc Ferrer, MD, Óscar Andión, MSc, Josep Matalı́, MSc,
Sergi Valero, PhD, José Antonio Navarro, MD,
Josep Antoni Ramos-Quiroga, MD, PhD, Rafael Torrubia, PhD,
and Miguel Casas, MD, PhD

In order to examine the impulsive profile of a BPD sample with comor-

bid ADHD, adult patients who met criteria for BPD were assessed for
ADHD with the CAADID and the WURS. A high rate of ADHD in the
BPD sample was found, with sixty-nine (38.1%) BPD patients diagnosed
as having comorbid adult ADHD. BPD-ADHD group had higher rates of
general substance use disorder (59.4% vs. 38.4%), antisocial personal-
ity disorder (7.2% vs. 0.9%) and obsessive-compulsive personality dis-
order (21.7% vs. 6.3%). The BPD group without comorbid adult ADHD
showed a higher rate of mood disorders (62.5% vs. 37.7%), panic disor-
ders (54.5% vs. 23.1%) and benzodiazepine abuse (18.8% vs. 5.8%).
Only in BPD patients without ADHD was comorbid avoidant personality
disorder found. BPD patients could be distinguished in two clear sub-
groups related to the adult ADHD comorbidity. BPD-ADHD patients
showed a more homogeneous and impulsive profile while BPD without
ADHD comorbidity had more anxiety and depressive disorders.

Several studies have noted numerous similarities between borderline per-

sonality disorder (BPD) and attention deficit-hyperactivity disorder
(ADHD) such as symptomatology in adulthood, and have explored possible
relationships between both disorders (Andrulonis, Glueck, Stroebel, & Vo-
gel, 1982; Andrulonis, Glueck, Stroebel, Vogel, & Shapiro, 1981; Andru-
lonis & Vogel, 1984; Fossati, Novella, Donati, Donini, & Maffei, 2002;
Miller et al., 2008; O’Leary, 2000; Philipsen, 2006; Philipsen et al., 2008).
In fact, the seven symptom clusters proposed by Wender for the diagnosis
of adult ADHD, consisting of inattentiveness, hyperactivity, mood lability,

From Hospital Universitari Vall d’Hebron, Barcelona (M. F., O. A., S. V., J. A. N., J. A. R.-Q.,
M. C.); and Universitat Autònoma de Barcelona (M. F., Ó. A., J. A. N., J. A. R.-Q., R. T., M. C.).
Address correspondence to Marc Ferrer, MD, Psychiatry Department, Hospital Universitari
Vall d’Hebron, Passeig Vall d’Hebron 119–129, 08035 Barcelona, Spain; E-mail: maferrer@
vhebron.net

812
COMORBID ATTENTION-DEFICIT/HYPERACTIVITY 813

irritability and hot temper, impaired stress tolerance, disorganization, and

impulsivity, are also largely observed in patients with BPD (Fossati et al.,
2002). Moreover, in recent studies, ADHD diagnosis in childhood was
found in over 40% of adults in BPD samples (Fossati et al., 2002; Philipsen
et al., 2008). Furthermore, in the Philipsen and colleagues study (2008),
16.1% of the adult BPD sample was diagnosed as having a comorbid
ADHD in adulthood.
The polythetic criteria set for BPD results in 151 possible combinations
of criteria for BPD diagnosis turns it into a heterogeneous syndrome. Such
clinical heterogeneity has led to a search for latent variables within the
diagnosis based on empirical methods (Skodol et al., 2002) and BPD ethi-
ology theories (Oldham, 2006). Researchers commonly recognized impul-
sivity as an important component of BPD construct (Links, Heslegrave, &
van Reekum, 1999; Moeller, Barrat, Dougherty, Schmitz, & Swann, 2001)
and has been proposed as a definitory feature of a BPD clinical subtype
(Oldham, 2006). ADHD is very common in subjects with BPD and both
disorders share key features such as emotional instability and impulsivity
(Fossati et al., 2002). Adult patients with BPD and comorbid ADHD have
been proposed as a more homogeneous but still representative subgroup
of BPD (Rüsch et al., 2007) which could be considered as part of the BPD
impulsive subtype.
The main aim of this study was to examine if BPD with comorbid ADHD
is a more homogeneous impulsive BPD subtype. Our hypothesis is that
patients with BPD and comorbid ADHD will have a higher rate of impulsive
related comorbid disorders, more suicidal attempts, self harm behaviors,
and emergency room visits and a higher score in psychometric tools to
assess impulsivity, compared to those BPD patients without a comorbid
ADHD.

METHOD
PARTICIPANTS

Participants were selected over a whole sample of two hundred thirty-three

adult patients between 18 and 50 years old previously diagnosed as BPD
by general practitioners and subsequently referred to the Borderline Per-
sonality Disorder Treatment Program of the Hospital Universitari Vall
d’Hebron, Barcelona, Spain, over a two-year period from 2005 to 2007.
This study received ethical approval from the Ethics Committee. Written
consent was obtained from all the participants before entering the study.

MEASURES

Personality disorders were assessed using the Structured Clinical Inter-

view for DSM-IV Axis II Disorder (SCID-II; First, Gibbon, Spitzer, Williams,
& Benjamin, 1999). Axis I disorders were assessed using the Structured
814 FERRER ET AL.

Clinical Interview for DSM-IV Axis I Clinical Version (SCID-I; First, Spitzer,
Gibbon, & Williams, 1997). The Symptom Checklist-90-Reviewed (SCL-90-
R; Derogatis, 1999) was used to assess general severity of psychopathol-
ogy, paranoid ideation, psychoticism, and interpersonal sensibility. Impul-
sivity was assessed by the Barrat Impulsivity Scale-11 (BIS-11; Oquendo
et al., 2001). Suicidal behavior, self harm behavior, and visits to emer-
gency room were assessed using the corresponding items of the revised
Diagnostic Interview for Borderline (DIB-R; Zanarini, Gunderson, Fran-
kenburg, & Chauncey, 1989).
The Conners’ Adult ADHD Diagnostic Interview for DSM-IV (CAADID;
Epstein, Johnson, & Conners, 2006) has been shown to be a reliable and
accurate interview by assessing ADHD in adults (Epstein & Kollins, 2006).
The CAADID interview assesses the ADHD DSM-IV criteria presence in
childhood and also the continuity of those during adulthood. Consistent
with the methodology of previous studies that used the Wender Utah Rat-
ing Scale (WURS; Rodrı́guez-Jiménez et al., 2001) in BPD samples (Fossati
et al., 2002; Philipsen et al., 2008), ADHD was assessed using a conserva-
tive WURS cuff-off ( ≥ 46).

PROCEDURE

Patients were evaluated in three different interviews. During the first inter-
view, all patients underwent an initial screening to disregard history or
current symptoms of a serious organic condition that might be associated
with the development of psychiatric symptoms, current diagnosis of
schizophrenia, bipolar I disorder, or substance dependence disorder, and
met at least 5 of the 9 DSM-IV-TR (Frances, First, & Pincus, 2005) BPD
criteria. Participants with substance abuse without DSM-IV-TR depen-
dence criteria were included in the study. Those participants who did not
meet the exclusion criteria were then assessed for DSM-IV criteria of
ADHD using the CAADID.
During the second and third interviews, participants were evaluated to
assess Axis I and Axis II disorders using the Structured Clinical Interview
for DSM-IV for Axis II Disorders (SCID-II; First et al., 1999) and then the
Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I; First
et al., 1997). These interviews were carried out on two separate days by
psychologists trained to use these instruments and without knowledge of
the first interview results. Twenty of the total SCID interviews adminis-
tered were audio-recorded in order to analyze interrater reliability. Excel-
lent median interrater reliability was found for Axis I (kappa > 0.73) and
Axis II disorders (kappa = 0.71–0.91). Finally, self-report evaluations us-
ing WURS, SCL-90-R, and BIS-11 were collected during the second and
third interviews. Data from those patients whose ADHD diagnosis in child-
hood was not confirmed for both WURS and CAADID was not included in
the results analysis.
COMORBID ATTENTION-DEFICIT/HYPERACTIVITY 815

DATA ANALYSIS

Between-group comparisons involving categorical data were analyzed with

the Chi-Square test, and between-group comparisons involving continu-
ous data were analyzed using the Student’s t-test. The U-Man Whitney test
was used when variables did not show a normal distribution. Although
Bonferroni-type correction is not recommended in exploratory studies
(Bender & Lange, 2001), it was applied to the p-values for the main effects
of axis I disorders, axis II disorders, and self-report measures. This means
that statistical significance reached p < 0.002 (p-value of 0.05/24) for axis
I disorders, p < 0.004 (p-value of 0.05/11) for axis II disorders, and p <
0.005 (p-value of 0.05/9) for self-report measures. Results with Bonfer-
roni-type correction and significance at p < 0.05 were reported to keep the
exploratory approach flexibility.

RESULTS
Of the total sample of 233 patients referred to the BPD Program for diagno-
sis and treatment, 181 (77.7%) met diagnostic criteria for BPD. A total of
fifty-two (22.3%) were excluded for the following reasons: 6 (2.6%) met cri-
teria for schizophrenia, 4 (1.7%) for bipolar I disorder, 19 (8.1%) for sub-
stance dependence disorder and 23 (9.9%) for a lack of agreement about
ADHD diagnoses. When the demographic variables of included and ex-
cluded participants were compared, significant differences were only
found for level of education (χ2 = 10.23, df = 1, p = 0.001), and gender {χ2 =
6.25, df = 1, p = 0.012; [33 (63.5%) vs. 138 (76.2%)]}, with lower educa-
tional level and more women in the group of those who were included.
Of the 181 participants who were diagnosed as BPD, 69 (38.1%) were
diagnosed as suffering a comorbid ADHD (BPD-ADHD group), while 112
(61.9%) did not meet diagnostic criteria enough for adult ADHD (BPD
group). The most frequent comorbid ADHD subtype was the combined
one, with 41 patients (22.7%), see Table 1.
The BPD group and the BPD-ADHD group were found to be very similar

TABLE 1. Rates of Comorbid ADHD and ADHD

Subtypes in BPD Sample
Borderline Patients n %
No ADHD Comorbidity 112 61.9
BPD-ADHD Comorbidity 69 38.1
BDP-ADHD Subtypes
ADHD Inattentive Type 22 12.2
ADHD Hyperactive/Impulsive Type 6 3.3
ADHD Combined Type 41 22.7
Notes. Abbreviations: ADHD = attention deficit–hyper-
activity disorder, BPD = borderline personality disor-
der, BPD-ADHD = borderline personality disorder with
comorbid attention deficit–hyperactivity disorder.
816 FERRER ET AL.

in their demographic characteristics. Groups did not differ significantly in

age (t = 0.64, df = 179, p = 0.53), gender (χ2 = 2.75, df = 1, p = 0.09), mari-
tal status (χ2 = 2.61, df = 2, p = 0.27), or occupation (χ2 = 2.24, df = 3, p =
0.52). The only statistical difference between groups was educational level
(χ2 = 10.23, df = 1, p = 0.001), see Table 2.
Groups did not differ in the total number of comorbid Axis I disorders
(Z = 1.65, p = 0.10). However, between-group differences were found for
mood disorders (χ2 = 10.56, df = 1, p = 0.001), substance abuse disorders
(χ2 = 7.56, df = 1, p = 0.006), and anxiety disorders (χ2 = 7.56, df = 1, p =
0.002). In the BPD group, a higher prevalence of major depression (49.1%
vs. 27.3%; χ2 = 6.56, df = 1, p = 0.01), benzodiazepine abuse (18.8% vs.
5.8%; χ2 = 4.09, df = 1, p = 0.04), and panic disorder (54.5% vs. 23.1%;
χ2 = 14.78, df = 1, p < 0.001) was observed. In the BPD-ADHD group,
higher rates of cannabis abuse (31.9% vs. 15.1%; χ2 = 8.52, df = 1, p =
0.004) and nonspecific anxiety disorder (10.1% vs. 0.9%; χ2 = 9.37, df = 1,
p = 0.02) were found. However, at the Bonferroni-corrected probability
level, only mood disorders, anxiety disorders, and panic disorders remain
significant. In addition, the higher rate of cocaine abuse in the BPD-ADHD
group approached significance (24.6% vs. 16.0%; χ2 = 2.72, df = 1, p =
0.09), see Table 3.
Differences in the total number of comorbid personality disorders was
also found not to be significant (Z = 1.47, p = 0.15). But when specific per-

TABLE 2. Sociodemographic Differences Between BPD and BPD-ADHD Groups

BPD-ADHD
n (181) BPD (112) (69)
Mean (SD) Mean (SD) Mean (SD) t (df ) p
Age 25.89 (6.58) 26.13 (6.53) 25.49 (6.64) 0.64 (179) 0.53

n (%) n (%) n (%) !2 (df ) p

Sex
Female 138 (76.2) 90 (80.4) 48 (69.6) 2.75 (1) 0.09
Education1
Did not complete primary
education* 5 (2.8) 0 (0.0) 5 (7.2) 10.23 (1) 0.001
Completed primary
education 138 (76.2) 97 (86.6) 41 (59.4)
High School Diploma 35 (19.3) 14 (12.5) 21 (30.4)
University Degree 3 (1.7) 1 (0.9) 2 (2.9)
Marital status
Never married 141 (77.9) 83 (74.1) 58 (84.1) 2.61 (2) 0.27
Married or Cohabiting 24 (13.3) 18 (16.1) 6 (8.7)
Divorced, Separated, or
Widowed 16 (8.8) 11 (9.8) 5 (7.2)
Occupation
Student 30 (16.5) 17 (15.2) 13 (18.8) 2.24 (3) 0.52
Never employed 49 (27.1) 28 (25.0) 21 (30.4)
Unemployed or on disability 45 (24.9) 28 (25.0) 17 (24.6)
Employed 57 (31.5) 39 (34.8) 18 (26.1)
Notes. Abbreviations: BPD = borderline personality disorder, BPD-ADHD = borderline per-
sonality disorder with comorbid attention deficit–hyperactivity disorder.
*In Spain, “primary education” lasts until age 16
1
Recoded as dichotomed variables
COMORBID ATTENTION-DEFICIT/HYPERACTIVITY 817

TABLE 3. Comparison of BPD and BPD-ADHD groups by Axis I Disorders

BPD-ADHD
BPD (112) (69)
Axis I Disorders Median (SD) Median (SD) Z p
Number of Axis I Disorders 3.00 (2.05) 4.00 (2.26) 1.65 0.10

!2
n (%) n (%) (df = 1) p
Mood Disorders 70 (62.5) 26 (37.7) 10.56 0.001*
Major Depression 55 (49.1) 19 (27.3) 6.56 0.01
Dysthymia 13 (11.6) 7 (10.1) 0.02 0.87
Bipolar Disorder II 2 (1.8) 0 (0.0) 1.17 0.28
Substance Abuse Disorders 43 (38.4) 41 (59.4) 7.56 0.006
Alcohol Abuse 19 (16.9) 9 (13.0) 0.28 0.60
Cannabis Abuse 17 (15.1) 22 (31.9) 8.52 0.004
Hallucinogens Abuse 4 (3.6) 3 (4.3) 0.12 0.73
Amphetamine Abuse 8 (7.1) 1 (1.4) 2.65 0.10
Cocaine Abuse 18 (16.0) 17 (24.6) 2.72 0.09
Benzodiazepine Abuse 21 (18.8) 4 (5.8) 4.09 0.04
Anxiety Disorders 88 (78.6) 39 (56.5) 9.92 0.002*
Panic Disorder 61 (54.5) 16 (23.1) 14.78 <0.001*
Agoraphobia 8 (7.1) 10 (14.5) 2.18 0.14
Anxiety Disorder ns 1 (0.9) 7 (10.1) 9.37 0.02
Social Phobia 20 (17.9) 7 (10.1) 1.56 0.21
Simple Phobia 21 (18.8) 10 (14.5) 0.67 0.41
Generalized Anxiety 9 (8.0) 2 (2.9) 1.71 0.19
Obsessive-compulsive Disorder 4 (3.6) 2 (2.9) 0.03 0.87
PTSD 18 (16.1) 13 (18.8) 0.47 0.49
Somatoform Disorders 12 (10.7) 3 (4.3) 2.28 0.13
Eating Disorders 17 (15.2) 14 (20.3) 0.79 0.38
Notes. BPD = borderline personality disorder; BPD-ADHD = borderline personality dis-
order with comorbid attention deficit–hyperactivity disorder; PTSD = post–traumatic
stress disorder.
*Significant after Bonferroni correction.

sonality disorders were analyzed, significant differences arose. The BPD-

ADHD group had higher rates than the BPD group in comorbid antisocial
personality disorder (7.2% vs. 0.9%; χ2 = 5.38, df = 1, p = 0.02) and obses-
sive-compulsive personality disorder (21.7% vs. 6.3%; χ2 = 9.59, df = 1, p =
0.002). Comorbid avoidant personality disorder (AvPD) was observed only
in the BPD group (n = 37; 33%; χ2 = 28.65, df = 1, p < 0.001). Only antiso-
cial personality disorder did not remain significant at Bonferroni-corrected
probability level. Post-hoc analysis of AvPD in the BPD group indicated
that only women presented comorbid AvPD (χ2 = 8.75; df = 1; p = 0.003) as
can be seen in Table 4.
Suicidal behaviors were found to be more frequent in the BPD-ADHD
group than in the BPD group (72.9% vs. 54.5%; χ2 = 6.85, df = 1, p =
0.009), although it is not significant at the Bonferroni-corrected probabil-
ity level. The groups did not differ either in self-harm behaviors (59.8% vs.
62.3%; χ2 = 0.11, df = 1, p = 0.74) or in number of emergency room visits
(40.2% vs. 37.7%; χ2 = 0.11, df = 1, p = 0.74). Finally, Table 5 shows how
the BPD group exhibited significantly higher SCL-90 R scores on the gen-
eral symptomatic index, paranoid ideation, psychoticism, and interper-
sonal sensibility scales (Z = 4.30, p < 0.001; Z = 3.67, p < 0.001; Z = 2.33,
818 FERRER ET AL.

TABLE 4. Comparison of BPD and BPD-ADHD Groups by Rates of Addi-

tional Personality Disorders
BPD-ADHD
BPD (112) (69)
Median (SD) Median (SD) Z p
Personality Disorders 1.00 (1.21) 1.0 (1.44) 1.4 70.15

!2
n (%) n (%) (df = 1) p
Cluster A 40 (35.7) 29 (42.0) 0.72 0.40
Paranoid 37 (33.0) 26 (37.7) 0.41 0.52
Schizotypal 5 (4.5) 8 (11.6) 3.26 0.07
1
Cluster B 23 (20.5) 19 (27.5) 1.17 0.28
Histrionic 16 (14.3) 13 (18.8) 0.66 0.42
Antisocial 1 (0.9) 5 (7.2) 5.38 0.02
Cluster C 56 (50.0) 26 (37.7) 2.12 0.11
Avoidant 37 (33.0) 0 (0.0) 28.65 <0.001*
Dependent 23 (20.5) 13 (18.8) 0.08 0.78
Obsessive-compulsive 7 (6.3) 15 (21.7) 9.59 0.002*
Notes. 1Cluster B personality disorder diagnoses does not include BPD diagno-
ses; BPD = borderline personality disorder; BPD-ADHD = borderline personal-
ity disorder with comorbid attention deficit–hyperactivity disorder.
*Significant after Bonferroni correction.

p = 0.02, and Z = 5.28; p < 0.001, respectively), and lower on the BIS-11
total score (Z = 5.28, p < 0.001).

DISCUSSION
The main aim of this study was to examine if BPD with comorbid ADHD
represents a more homogeneous impulsive BPD subtype. According to our
hypothesis, compared to the BPD group, the BPD-ADHD group showed a

TABLE 5. Clinical Profile Comparison of BPD and BPD-ADHD Groups

BPD-ADHD
BPD group group
(n = 112) (n = 69) !2
n (%) n (%) (df = 1) p
Suicidal behaviors1 61 (54.5) 51 (72.9) 6.85 0.009
Self-harm behaviors1 67 (59.8) 43 (62.3) 0.11 0.74
Emergency room visits1 45 (40.2) 26 (37.7) 0.11 0.74

Mean (SD) Mean (SD) Z p

WURS 35.49 (20.60) 59.03 (15.37) 6.64 <0.001*
SCL-90-R
General symptomatic index scale 2.43 (0.77) 1.77 (0.50) 4.30 <0.001*
Paranoid ideation scale 2.05 (0.93) 1.36 (0.68) 3.67 <0.001*
Psychoticism scale 1.61 (0.79) 1.32 (0.63) 2.33 0.02
Interpersonal sensitivity scale 2.62 (0.94) 1.84 (0.81) 4.99 <0.001*
BIS-11 68.73 (14.48) 82.20 (11.54) 5.28 <0.001*
Notes. BPD = borderline personality disorder; BPD-ADHD = borderline personality disorder
with comorbid attention deficit–hyperactivity disorder; WURS = Wender Utah Rating Scale;
SCL-90-R = 90–item Symptom Checklist Revised; BIS-11 = Barrat Impulsiveness Scale
11th version.
1
Recoded as dichotomed variables
*Significant after Bonferroni correction.
COMORBID ATTENTION-DEFICIT/HYPERACTIVITY 819

higher rate of impulsive related comorbid disorders and suicidal behaviors

and a higher score in the impulsivity self-report instrument.
First of all, the BPD-ADHD comorbidity rate observed in our study is
consistent with the earliest studies that investigated relationships between
ADHD and BPD (Andrulonis et al., 1982; Andrulonis et al., 1981; Andru-
lonis & Vogel, 1984). Nevertheless, although none of these studies focused
specifically on the ADHD diagnosis, they did not exclude other learning
disabilities, a common confounding factor in the diagnosis of ADHD (Van
Reekum, 1993) and did not use clinical interviews like CAADID that can
evaluate ADHD during adulthood. The prevalence of comorbid ADHD ob-
served in our BPD sample (38.1%) is higher than that 16.1% observed in
the Philipsen and colleagues study (2008). This difference can be ex-
plained by two facts; men were not included and only the combined ADHD
type was studied, potentially resulting in an underestimation of comorbid-
ity rates.
According to our hypothesis, BPD and BPD-ADHD groups showed a dif-
ferential comorbidity axis I pattern. While the BPD-ADHD group was found
to be more frequently associated to substance abuse disorder, classically
linked to impulsivity (Bornovalova, Lejuez, Daughters, Rosenthal, &
Lynch, 2005; Brodsky, Malone, Ellis, Dulit, & Mann, 1997; LeGris & van
Reekum, 2006), the BPD group showed a higher prevalence of mood and
anxiety disorders. Focusing on the differences in the substance abused by
both groups, cannabis abuse rate was significantly higher in the BPD-
ADHD group and a higher rate of cocaine abuse approaching significance,
while benzodiazepine abuse rate was significantly more in the BPD group.
Moreover, a higher rate of suicidal behaviors and a higher score in impul-
sivity measured with the BIS-11 were also observed in BDP-ADHD group.
Compared to the BPD group, the BPD-ADHD group also showed a signif-
icantly higher prevalence of those personality disorders classically related
to impulsive behaviors. While antisocial and obsessive-compulsive person-
ality disorders were observed to be more prevalent in BPD-ADHD group,
AvPD was only described in the BPD group. It is possible that the appear-
ance of AvPD only in the BPD group could be due to the relative high fre-
quency of AvPD in BPD patients (Oldham et al., 1995; Zanarini et al.,
1998) in comparison to ADHD patients (Biederman, 2005). Another expla-
nation could be based on personality differences between the groups. In
fact, according to Cloninger’s model, ADHD is characterized as an explo-
sive/borderline personality type with high novelty seeking, harm avoid-
ance, and low reward dependence (Anckarsäter et al., 2006) while female
BPD patients are characterized by high levels of harm avoidance, but not
novelty seeking (Barnow et al., 2007).
Because of the similarity between some symptoms of BPD and adult
ADHD, such as impulsivity, angry outbursts, instability of affect, and feel-
ings of boredom (Fossati et al., 2002), the possibility that the clinical sever-
ity of BPD could be influenced by ADHD symptomatology must be consid-
ered. Our study found statistically-significant differences between the BPD
820 FERRER ET AL.

and the BPD-ADHD groups on the General Symptomatic Index scale of

the SCL-90-R, indicating more severe symptomatology in the BPD group.
Moreover, the BPD group scored significantly higher on the paranoid ide-
ation scale and the psychoticism scale of the SCL-90-R, symptoms related
to BPD and which are not part of ADHD diagnostic criteria (Philipsen,
2006). According to these, the greater clinical severity proved in BPD group
should be explained for characteristic BPD symptoms better than for
ADHD comorbidity.
The differences in the comorbidity between BPD and BPD-ADHD groups
observed in this study agree with the BPD subtyping system proposed by
Oldham (2006). This model proposed 5 BPD subtypes: Affective, impulsive,
aggressive, dependent, and empty subtype. Data on comorbid disorders
exhibited by the BPD-ADHD group in this study, such as antisocial per-
sonality disorder, obsessive-compulsive personality disorder and sub-
stance abuse disorders, as well as the higher rate of suicidal behaviors,
agree with Oldham’s (2006) criteria for impulsive subtype. Nevertheless,
the BPD group showed a wide range of additional comorbid disorders that
can be associated with the affective, aggressive, dependent, and empty
Oldham’s subtypes, indicating a less homogeneous symptom profile.
These findings suggest that BPD-ADHD group may be a particular BPD
subtype with a relatively homogeneous profile of impulsive symptoms and
additional impulsive-style comorbid disorders.
The main limitation of our study is common to those which try to evalu-
ate ADHD during adulthood, where ADHD childhood diagnosis must be
assessed retrospectively, with a potential lack of objectivity. Our study
agrees with previous evidence about a significant phenomenological asso-
ciation between ADHD and BPD and only follow-up investigations, as the
one carried out by Miller and colleagues (2008), may help to clear up the
relationship between childhood ADHD and adult BPD. Another limitation
of our study is related to the exploratory design. To confirm the results,
the corresponding hypothesis should be tested in further confirmatory
studies.
In conclusion, according to DSM-IV diagnostic features, BPD should be
described as a unidimensional disorder which covers a heterogeneous
syndrome (Paris, 2005). A way to a better understanding of the disorder
could be to study more homogeneous groups of BPD patients. In our
study, like in Rüsch and colleagues (2007) study, ADHD comorbidity has
been used as an external variable to establish a BPD clinical subtype char-
acterized by a high impulsivity. Understanding the disorder in terms of
more homogeneous subgroups offers a way to target areas of treatment,
clarify etiological pathways and address the problems associated with di-
agnostic co-occurrence in this disorder (Sanislow et al., 2002).
To date, little is known about the specific effects of comorbid ADHD in
BPD. In that sense, future research should investigate if ADHD in child-
hood may be considered a risk factor that predisposes an impulsive BPD
subtype and if this BPD subtype in adulthood can be prevented by treating
COMORBID ATTENTION-DEFICIT/HYPERACTIVITY 821

ADHD in childhood. Along the same lines, the efficacy of ADHD pharmaco-
logical treatment against the impulsivity should be tested with controlled
studies with this impulsive BPD subtype. Finally, it could be interesting
to confirm our results and to explore the existence of other BPD clinical
subtypes.

REFERENCES

Anckarsäter, H., Stahlberg, O., Larson, T., Dulit, R. A., & Mann, J. J. (1997).
Hakansson, C., Jutblad, S. B., Niklas- Characteristics of borderline personal-
son, L., et al. (2006). The impact of ity disorder associated with suicidal
ADHD and autism spectrum disorders behavior. American Journal of Psychia-
on temperament character, and per- try, 154, 1715–1719.
sonality development. The American Derogatis, L. R. (1999). Manual guide of the
Journal of Psychiatry, 163, 1239– symptom checklist r0 Revised (SCL-90-
1244. R). TEA ediciones, S.A., Madrid (Spain).
Andrulonis, P. A., Glueck, B. C., Stroebel, Epstein, J. N., & Kollins, S. H. (2006). Psy-
C. F., & Vogel, N. G. (1982). Borderline chometric properties of an adult
personality subcategories. The Journal ADHD diagnostic interview. Journal of
of Nervous and Mental Disease, 170, Attention Disorders, 9, 504–514.
670–679. Epstein, J. N., Johnson, D., & Conners, C. K.
Andrulonis, P. A., Glueck, B. C., Stroebel, (2006). Conners’ adult ADHD diagnos-
C. F., Vogel, N. G., & Shapiro, A. L. tic interview for DSM-IV. North Tona-
(1981). Organic brain dysfunctions wanda, NY: Multi-Health Systems, Inc.
and the borderline syndrome. The Psy- First, M. B., Gibbon, M., Spitzer, R. L.,
chiatric Clinics of North America, 4, Williams, J.B.W., & Benjamin, L. S.
47–66. (1999). The structured clinical inter-
Andrulonis, P. A., & Vogel, N. G. (1984). view for DSM-IV axis II disorder (SCID-
Comparison of borderline personality II). New York: Biometrics Research De-
subcategories to schizophrenic and af- partment, New York State Psychiatric
fective disorders. The British Journal of Institute.
Psychiatry, 144, 358–363. First, M. B., Spitzer, R. L., Gibbon, M., &
Barnow, S., Herpertz, S. C., Spitzer, C., Stop- Williams, J.B.W. (1997). Structured
sack, M., Preuss, U. W., Grabe, H. J., clinical interview for DSM-IV (SCID-I)
et al. (2007). Temperament and char- (User’s guide and interview) Clinical
acter in patients with borderline per- version. New York: Biometrics Re-
sonality disorder taking gender and search Department, New York State
comorbidity into account. Psychopath- Psychiatric Institute.
ology, 40, 369–378. Fossati, A., Novella, L., Donati, D., Donini,
Bender, R., & Lange, S. (2001). Adjusting for M., & Maffei, C. (2002). History of
multiple testing—when and how? childhood attention deficit/hyperac-
Journal of Clinical Epidemiology, 54, tivity disorder symptoms and border-
343–349. line personality disorder: A controlled
Biederman, J. (2005). Attention-deficit/hy- study. Comprehensive Psychiatry, 43,
peractivity disorder: A selective over- 365–377.
view. Biological Psychiatry, 57, 1215– Frances, A., First, M. B., & Pincus, H. A.
1220. (2005). DSM-IV-TR. Guı́a de uso. Ed.
Bornovalova, M. A., Lejuez, C. W., Daugh- Elsevier – Masson.
ters, S. B., Rosenthal, M. Z., & Lynch, LeGris, J., & van Reekum, R. (2006). Neuro-
T. R. (2005). Impulsivity as a common psychological correlates of borderline
process across borderline personality personality disorder and suicidal be-
and substance use disorders. Clinical haviour. Canadian Journal of Psychia-
Psychology Review, 25, 790–812. try, 51, 131–142.
Brodsky, B. S., Malone, K. M., Ellis, S. P., Links, P. S., Heslegrave, R., & van Reekum,
822 FERRER ET AL.

R. (1999). Impulsivity: Core aspect of ish Journal of Psychiatry, 192, 118–

borderline personality disorder. Jour- 123.
nal of Personality Disorders, 13, 1–9. Rodrı́guez-Jiménez, R., Ponce, G., Monasor,
Miller, C. J., Flory, J. D., Miller, S. R., Harty, R., Jiménez-Giménez, M., Pérez-Rojo,
S. C., Newcorn, J. H., & Halperin, J. A., Rubio, G., et al. (2001). Valida-
J. M. (2008). Childhood attention-defi- tion in the Spanish population of the
cit/hyperactivity disorder and the Wender Utah rating scale for the retro-
emergence of personality disorders in spective evaluation in adults of atten-
adolescence: A prospective follow-up tion deficit/hyperactivity disorder in
study. Journal of Clinical Psychiatry, childhood. Revista de Neurologia, 33,
69, 1477–1483. 138–144.
Moeller, F. G., Barrat, E. S., Dougherty, Rüsch N., Weber M., Il’yasov K. A., Lieb K.,
D. M., Schmitz, J. M., & Swann, A. C. Ebert, D., Hennig J., et al. (2007) Infe-
(2001). Psychiatric aspects of impul- rior frontal white matter microstruc-
sivity. The American Journal of Psychi- ture and patterns of psychopathology
atry, 158, 1783–1793. in women with borderline personality
O’Leary, K. M. (2000). Borderline personality disorder and comorbid attention-defi-
disorder. Neuropsychological testing cit hyperactivity disorder. NeuroImage,
results. The Psychiatric Clinics of North 35, 738–747.
America, 23, 41–60. Sanislow, S. A., Grilo, C. M., Morey, L. C.,
Oldham, J. M. (2006). Borderline personality Bender, D. S., Skodol, A. E., Gunder-
disorder and suicidality. The American son, J. G., et al. (2002). Confirmatory
Journal of Psychiatry, 163, 20–26. factor analysis of DSM-IV criteria for
Oldham, J. M., Skodol, A. E., Kellman, H. D., borderline personality disorder: Find-
Hyler, S. E., Doidge, N., Rosnick, L., et ings from the collaborative longitudi-
al. (1995). Comorbidity of axis I and nal personality disorders study. The
axis II disorders. The American Journal American Journal of Psychiatry, 159,
of Psychiatry, 152, 571–578. 284–290.
Oquendo, M. A., Baca-Garcı́a, E., Graver, R., Skodol, A. E., Gunderson, J. G., Pfohl, B.,
Morales, M., Montalbán, V., & Mann, Widiger, T. A., Livesley, W. J., & Siever,
J. J. (2001). Spanish adaptation of the L. J. (2002). The borderline diagnosis
Barratt impulsiveness scale. European I: Psychopathology, comorbidity, and
Psychiatry, 15, 147–155. personality structure. Biological Psy-
Paris, J. (2005). The diagnosis of borderline chiatry, 51, 936–950.
personality disorder: Problematic but van Reekum, R. (1993). Acquired and devel-
better than the alternatives. Annals of opmental brain dysfunction borderline
Clinical Psychiatry, 17, 41–46. personality disorder. Canadian Jour-
Philipsen, A. (2006). Differential diagnosis nal of Psychiatry, 38, 4–10.
comorbidity of attention-deficit/hyper- Zanarini, M. C., Gunderson, J. G., Franken-
activity disorder (ADHD) and border- burg, F. R., & Chauncey, D. L. (1989).
line personality disorder (BPD) in The revised diagnostic interview for
adults. European Archives of Psychia- borderlines: Discriminating borderline
try and Clinical Neuroscience, 256, personality disorders from other Axis
42–46. II disorders. Journal of Personality Dis-
Philipsen, A., Limberger, M. F., Lieb, K., order, 3, 10–18.
Feige, B., Kleindienst, N., Ebner- Zanarini, M. C., Frankenburg, F. R., Dubo,
Priemer, U. et al. (2008). Attention- E. D., Sickel, A. E., Trikha, A., Levin,
deficit hyperactivity disorder as a A., et al. (1998). Axis II comorbidity of
potentially aggravating factor in bor- borderline personality disorder. Com-
derline personality disorder. The Brit- prehensive Psychiatry, 39, 296–302.