empirical research borderline

• Open access
• Published: 15 February 2023

Developmental predictors of young adult borderline personality disorder: a prospective, longitudinal study of females with and without childhood ADHD

• Sinclaire M. O’Grady 1 &
• Stephen P. Hinshaw 1 , 2  

BMC Psychiatry volume  23 , Article number:  106 ( 2023 ) Cite this article

7883 Accesses

2 Citations

11 Altmetric

Metrics details

Research on the precursors of borderline personality disorder (BPD) reveals numerous child and adolescent risk factors, with impulsivity and trauma among the most salient. Yet few prospective longitudinal studies have examined pathways to BPD, particularly with inclusion of multiple risk domains.

We examined theory-informed predictors of young-adult BPD (a) diagnosis and (b) dimensional features from childhood and late adolescence via a diverse (47% non-white) sample of females with ( n  = 140) and without ( n  = 88) carefully diagnosed childhood attention-deficit hyperactivity disorder (ADHD).

After adjustment for key covariates, low levels of objectively measured executive functioning in childhood predicted young adult BPD diagnostic status, as did a cumulative history of childhood adverse experiences/trauma. Additionally, both childhood hyperactivity/impulsivity and childhood adverse experiences/trauma predicted young adult BPD dimensional features. Regarding late-adolescent predictors, no significant predictors emerged regarding BPD diagnosis, but internalizing and externalizing symptoms were each significant predictors of BPD dimensional features. Exploratory moderator analyses revealed that predictions to BPD dimensional features from low executive functioning were heightened in the presence of low socioeconomic status.

Conclusions

Given our sample size, caution is needed when drawing implications. Possible future directions include focus on preventive interventions in populations with enhanced risk for BPD, particularly those focused on improving executive functioning skills and reducing risk for trauma (and its manifestations). Replication is required, as are sensitive measures of early emotional invalidation and extensions to male samples.

Peer Review reports

Borderline personality disorder (BPD) is a persistent and highly impairing condition characterized by intense and pervasive dysregulation of emotion, behavior, and cognition, and a pattern of highly unstable interpersonal relationships [ 1 ]. Individuals with BPD are at extremely high risk for suicide: Up to 10% of individuals with BPD die by suicide each year, 50 times higher than the rate in the general population [ 2 , 3 ]. BPD is also associated with significant personal (i.e., severe psychosocial impairment) and economic/public health (i.e., high rates of underemployment and increased disability) consequences. In fact, although individuals with BPD comprise 1–2% of the general population [ 4 ], they have extremely high rates of health service use, representing up to 20% of individuals receiving inpatient psychiatric treatment and 10% receiving outpatient psychiatric care [ 5 ]. Evidence-based psychological treatments (e.g., Dialectical Behavior Therapy [DBT]) exist for BPD, with strong evidence for efficacy [ 6 ]. Yet they are resource-intensive, with limited availability of expert providers.

Given the high morbidity, mortality, and public health consequences of BPD, an urgent need exists to identify individuals at risk for its development. Relevant research is accumulating. A leading model is Linehan’s Biosocial Theory, which proposes that BPD emerges from transactions between (a) biological vulnerabilities linked with both impulsivity and emotional sensitivity and (b) specific environmental influences such as social invalidation, adversity, or trauma [ 1 , 7 ]. Across development, such combinations give rise to increasingly extreme emotional, behavioral, cognitive, and interpersonal dysregulation for vulnerable individuals. Although empirical research has generated numerous potential risk factors for BPD [ 8 , 9 ], relatively few studies have examined longitudinal pathways to BPD. Identification of such would help strengthen theoretical approaches to the development of BPD. In particular, few studies have examined childhood risk factors for BPD in prospective designs [ 10 ]. Indeed, a 2016 systematic review of risk factors for BPD revealed that risk factors were assessed mainly during early adolescence ( M age  = 13 years), highlighting the need for further investigation of childhood variables and processes [ 9 ]. Such studies could further inform leading theories [ 1 , 11 ] regarding heightened periods of risk (and areas for intervention). In the present investigation, we examine both child and late-adolescent risk factors for later BPD.

Developmental risk factors for BPD

Adhd symptoms.

Impulsivity is a key feature in both the development and presentation of BPD [ 1 ]. It is also a core symptom of attention-deficit hyperactivity disorder (ADHD). In fact, several studies have reported high comorbidity between BPD and ADHD [ 12 , 13 ]. In a large national study of 34,000 adults in the United States, among adults with ADHD the lifetime comorbidity with BPD was 33.7% [ 14 ]. Comorbid ADHD and BPD is a particularly impairing combination [ 12 ]. As for childhood ADHD in relation to adult BPD, most prior research has involved retrospective reports of symptoms. Here, 50–60% of adults with BPD endorsed high levels of ADHD symptoms in childhood [ 15 ]. Philipsen et al. [ 16 ] reported estimates of ADHD among adult women with BPD to be especially high during childhood (41.5%) as compared to adulthood (16.1%). Severity of childhood ADHD symptoms is also associated with higher frequency of personality disorder diagnoses, including BPD, by adulthood [ 17 ].

Hypothesized mechanisms linking ADHD and BPD focus on the interaction, across development, between early impulsivity (a highly heritable trait) and emotion dysregulation, the combination of which, in turn, is shaped by adverse socialization processes (e.g., maltreatment, family reinforcement of emotional lability) [ 11 ]. Proposed neurobiological mechanisms focus on dysfunction in the prefrontal cortex, which is also implicated in emotion regulatory capacities [ 11 , 18 ]. Essentially, early ADHD is hypothesized to confer risk for later BPD because of its associated behavioral dysregulation, promoting environmental reinforcement of maladaptive behaviors and often leading to a pervasive and difficult-to-treat cycle of dysregulation.

As noted above, however, few prospective studies have examined the link between childhood ADHD and later BPD. Notable exceptions (see [ 19 , 20 , 21 ]) reveal that childhood ADHD predicts personality disorders, including BPD, later in life. In a follow-up investigation of “hyperactive” children in young adulthood, Fischer et al. [ 22 ] found that 14% of hyperactive participants met criteria for BPD compared to 3% of their comparison group. Miller et al. [ 13 ] found that among a group with childhood ADHD, 13.5% were diagnosed with BPD in adolescence compared to 1.2% in their comparison group. Furthermore, in the Pittsburgh Girls Study, Stepp et al. [ 18 ] found that high levels of ADHD symptoms during childhood predicted BPD in adolescence (see also [ 23 ] for parallel findings in males). Using latent-class analysis, Thatcher et al. [ 24 ] found that the presence of ADHD symptoms in adolescents, along with substance use disorders, predicted more severe BPD symptoms at young-adult follow-up.

The majority of relevant research has focused on the categorical diagnosis of ADHD rather than the core ADHD dimensions of (a) hyperactivity/impulsivity and (b) inattention. Exceptions include Carlson et al. [ 25 ], who reported data from a prospective, longitudinal study. Teacher-rated severity of both attentional disturbance and behavioral instability (including impulsivity) at age 12 was predictive of adult BPD. This finding was recently replicated by Beeney et al. [ 8 ] in a prospective study of females. Here, parent- and child-reported severity of hyperactivity, impulsivity, and inattention at ages 14–15 predicted BPD at ages 16–18. In a prospective study in twins, maternal- and teacher-rated symptoms of impulsivity at age 5 were related to borderline symptoms at age 12 [ 26 ]. Finally, a recent national, prospective longitudinal study of twins in Sweden found that the association between childhood ADHD symptoms, assessed at age 9 or 12, and adult BPD was primarily driven by impulsivity as opposed to inattention or hyperactivity [ 27 ].

Some adult research has examined presentations (or “types”) of ADHD as related to BPD. Among adults with ADHD, one study reported a higher prevalence rate of comorbid BPD in the Combined presentation of ADHD (ADHD-C; 24%), for which hyperactivity/impulsivity and inattention are key components, compared to comorbid BPD and the Inattentive presentation (ADHD-I; 10%) [ 28 ]. Using latent class analysis with adult females, [ 29 ] found that one pathway to adult BPD emanated from a childhood profile with at least low/moderate levels of hyperactive/impulsive symptoms but low levels of inattentive symptoms.

Finally, considered either categorically or dimensionally, ADHD is clearly linked with increased risk for self-harm (a common feature of BPD), including suicidal behavior and nonsuicidal self-injury (NSSI) (see [ 30 ] for a recent review). Notably, females with ADHD, especially those with high levels of impulsivity as characterized by the ADHD-C presentation, are markedly at risk for attempted suicide and moderate-to-severe NSSI [ 31 ]. Mechanisms linking ADHD with later self-harm include internalizing and externalizing symptoms, as well as peer victimization and peer rejection [ 30 ]. Risk for suicidality is greatly increased when females with ADHD have histories of childhood maltreatment [ 32 ].

Executive functioning (EF)

EF includes goal-oriented cognitive processes such as planning, inhibition, organization, set shifting, working memory, and problem solving. EF deficits have frequently been linked to both ADHD and BPD. For a review of ADHD and EF see Brown [ 33 ]; for a review of BPD and EF, see Garcia-Villamisar et al. [ 34 ]. In general, individuals with BPD show greater EF deficits than typically developing controls [ 35 ]. Individuals with BPD have demonstrated EF deficits in the domains of planning [ 36 , 37 ], working memory [ 38 ], response inhibition and problem solving [ 39 ], and motor inhibition [ 40 ]. Additionally, a meta-analysis revealed that BPD samples with higher rates of comorbid psychopathology performed worse on EF tasks compared to samples with lower rates of comorbidity [ 41 ]. However, few prospective longitudinal studies have examined childhood EF as predictive of later BPD, with the notable exception of Belsky et al. [ 26 ], who found that a composite measure of EF deficits at age 5 was related to BPD symptoms at age 12. Longitudinal investigation of global measures of EF and relations to BPD are critically needed.

Early internalizing and externalizing symptoms

BPD is commonly comorbid with a variety of other psychological disorders, both internalizing and externalizing in nature [ 42 ]. In a systematic review of risk factors for BPD from longitudinal research, Stepp et al. [ 43 ] found that 16 of 19 studies examining internalizing and externalizing psychopathology yielded predictions to later BPD [ 9 ]. That is, dimensions of internalizing (depression) and externalizing (substance use disorder) behaviors in adolescence were associated with subsequent adult BPD symptoms. Indeed, existing literature theorizes that adolescence is a sensitive period for the development of personality disorders—and that personality disorders are preceded by internalizing and externalizing symptoms, not the other way around [ 44 ]. Additionally, some evidence indicates that internalizing and externalizing symptoms in childhood are also related to later BPD symptoms (Belsky et al., [ 26 ]; Geselowitz et al., [ 10 ] but see Burke [ 23 ] & Stepp, [ 18 ], for negative results). Hypothesized mechanisms include emotion dysregulation, which characterizes both internalizing and externalizing psychopathology [ 11 ]. In short, greater understanding of the contribution and developmental timing of internalizing and externalizing symptoms has the potential to inform early interventions to prevent the development of clinically significant BPD symptoms.

Adverse childhood experiences/trauma

Consistent with Linehan’s Biosocial Model, a large body of research has linked a history of environmental invalidation and adversity—and at its extreme, trauma—to the development of BPD [ 1 ]. A history of physical abuse, sexual abuse, and neglect in childhood has long been linked with BPD [ 9 ]. For key prospective longitudinal investigations, see Johnson et al. [ 45 ] and Widom et al. [ 46 ]. Having a parent with psychopathology (including depression and substance use problems) has consistently been found to be a family-related risk factor for BPD [ 9 , 43 ]. Empirical research on the role of parenting and parent–child transactions has been limited [ 47 ], even though key theories posit that transactions between a child’s biological sensitivity and adverse environments (including family factors and parenting) both lead to and maintain BPD symptomology [ 1 ] as well self-harmful behaviors in early adulthood [ 48 ]. High levels of parental depression and parental stress have been linked to BPD [ 49 ], as well as escalation of negative affect and behaviors in mother-daughter conflict situations [ 47 ]. In a prospective, longitudinal study of twins followed from age 5 to 12, children who were physically maltreated or exposed to high maternal negative expressed emotion developed high levels of BPD characteristics [ 26 ], replicating other prospective studies [ 25 , 50 , 51 ]. Still, at least one study revealed that maternal parenting stress in adolescence was not related to adolescent BPD symptom severity [ 52 ]. More research is needed related to transactions between parents and their offspring in terms of the development of BPD [ 53 ].

Present study and hypotheses

In sum, numerous risk factors for BPD have been posited, including ADHD symptom dimensions, low executive functioning, early internalizing and externalizing psychopathology, and childhood adversity and trauma. Yet with the clear exception of Beeney et al. [ 8 ], little research has examined such factors simultaneously, limiting current understanding of the independent or combined contributions of such variables. Also, many studies examine delimited developmental periods (e.g., childhood to adolescence, adolescence to young adulthood). Finally, there is a dearth of research examining the core ADHD dimensions of hyperactivity/impulsivity and inattention as related to risk for later BPD (for a review, see Beauchaine et al. [ 11 ]).

We leverage a sample of females with childhood-diagnosed ADHD and a matched comparison sample followed prospectively from childhood through young adulthood. Consistent with Linehan’s Biosocial Model, recent developmental models of females with ADHD [ 11 ], and extant literature, we hypothesize that both childhood and adolescent (a) hyperactivity/impulsivity and (b) adversity/trauma will emerge as significant risk factors for BPD after adjusting for demographic covariates as well as additional evidence-based risk factors. We also predict that, by late adolescence, internalizing and externalizing symptoms will be significant predictors of young adult BPD. An exploratory aim is to examine if childhood socioeconomic status (SES) moderates associations between predictors of interest and later BPD. We aim to add to the literature on developmental risk factors for BPD to inform existing models of BPD development and prevention approaches.

Procedure and participants

The current data were drawn from an ongoing prospective, longitudinal study of females with and without carefully diagnosed childhood ADHD (see Hinshaw, [ 54 ] for more complete details). This study was approved by the Committee for the Protection of Human Subjects (CPHS) at the University of California, Berkeley. Participants were initially recruited across a metropolitan area from schools, mental health centers, pediatric practices, and through advertisements to participate in research-based, 5-week summer day camps between 1997–1999. Some participants were recruited through the general population whereas others were recruited through the healthcare system. These programs were designed to be enrichment programs featuring classroom and outdoor environments for ecologically valid assessment, rather than intensive therapeutic interventions. All participants and their families underwent a rigorous, multi-step psychodiagnostic assessment process (see below), after which 140 girls with ADHD and 88 age- and ethnicity-matched comparison girls were selected to participate in the childhood program (Wave 1; M age  = 9.6 years, range = 6–12 years).

Following recruitment, all participants were screened for ADHD regardless of if they had already had a pre-established diagnosis. To establish a baseline diagnosis of ADHD, we used the parent-administered Diagnostic Interview Schedule for Children, 4 th ed. (DISC-IV) [ 55 ] and SNAP rating scale [ 52 ], Hinshaw, [ 54 ] for the diagnostic algorithm). Comparison girls could not meet diagnostic criteria for ADHD on either measure. Some comparison girls met criteria for internalizing disorders (3.4%) or disruptive behavior disorders (6.8%) at baseline, yet our goal was not to match ADHD participants on comorbid conditions but instead to obtain a representative comparison group. Exclusion criteria included intellectual disability, pervasive developmental disorders, psychosis, overt neurological disorder, lack of English spoken at home, and medical problems preventing summer camp participation. The final sample included 228 girls with ADHD-Combined presentation ( n  = 93) and ADHD-Inattentive presentation ( n  = 47), plus an age- and ethnicity-matched comparison sample ( n  = 88). Participants were ethnically diverse (53% White, 27% African American, 11% Latina, 9% Asian American), reflecting the composition of the San Francisco Bay Area in the 1990’s. Family income was slightly higher than the median local household income in the mid-1990s, yet income and educational attainment of families were highly variable, ranging from professional families to those receiving public assistance. On average, parents reported being married and living together (65.8%) at the baseline assessment.

Participants were then assessed 5 (Wave 2; M age  = 14.2 years, range = 11–18; 92% retention [data not included from this wave in the present study]), 10 (Wave 3; M age  = 19.6 years, range = 17–24 years; 95% retention), and 16 (Wave 4; M age  = 25.6 years, range = 21–29 years; 93% retention) years later. Data collection included multi-domain, multi-informant assessments, performed in our clinic for most individuals; when necessary, we performed telephone interviews or home visits. We obtained informed consent from all participants (for initial waves: all legal guardians for minors (if age was below 18 years) and parents; for later waves: all participants and parents). Participants received monetary compensation. For additional information see Hinshaw et al. [ 31 , 56 ], Owens et al. [ 57 ].

Predictor variables

Predictor variables were measured during the baseline assessment at Wave 1 (childhood), with repeated assessment of several key measures at Wave 3 (late adolescence), to incorporate risk factors in both developmental periods.

ADHD Symptom Severity: Swanson, Nolan, and Pelham rating scale, 4 th Ed. (SNAP-IV; Swanson, [ 58 ] ). We measured severity of both hyperactivity/impulsivity (SNAP-HI) and inattentive (SNAP-IA) symptoms using an average of parent- and teacher-report (childhood) or parent- and self-report (late adolescence) on a dimensionalized checklist of these two respective symptom domains (9 items for each) to obtain multi-informant composite scores (SNAP-HI: α = 0.950; SNAP-IA: α = 0.968). For example, items included “…this child is forgetful in daily activities” and “…this child blurts out answers to questions before the questions have been completed.” The severity of each symptom was scored 0 (not at all) to 3 (very much). Thus, scores of both hyperactivity/impulsivity and inattention symptoms ranged from 0–27, with higher scores indicating more severe symptomology. The SNAP-IV is a widely used scale of ADHD symptom severity in both research and clinical settings (e.g., MTA Cooperative Group, [ 59 ]). It has good internal consistency and test–retest reliability [ 60 ].

Executive Functioning: Rey Osterrieth Complex Figure (ROCF) [ 61 ]. We measured executive functioning using the ROCF, a laboratory-based cognitive task requesting that an individual copy and later recall a complex image composed of 64 segments. The ROCF measures multiple domains of executive functioning such as planning, inhibitory control, attention to detail, working memory, and organization. It is often considered a more “global” measure of executive functioning [ 62 ]. We analyzed the Copy condition of this task, during which participants are timed as they view the stimulus figure and draw the figure on a blank piece of paper. For scoring, we used the Error Proportion Score (EPS; the ratio of number of errors divided by the total number of segments drawn), a well-validated method of scoring the ROCF, indexing efficiency [ 63 ]. In previous research with this sample, only the Copy condition (versus Delayed Recall condition) differentiated girls with ADHD from our comparison sample at baseline. The ROCF EPS showed the largest effect size ( d  = 0.90) out of all other EF measures in our battery (Hinshaw et al., [ 64 ]; Sami et al., [ 63 ]). As well, childhood EPS predicts later academic and occupational functioning in comparison to other EF measures Miller et al., [ 62 ]).

Internalizing and Externalizing Symptoms: Child Behavior Checklist, Adult Self Report, and Adult Behavior Checklist (CBCL; ASR; ABCL) [ 65 , 66 ] . In childhood, we measured severity of internalizing (α = 0.89) and externalizing (α = 0.93) symptoms via parent-report on the Internalizing and Externalizing scales of the CBCL. In late adolescence, we averaged participant self-report on the Adult Self-Report (ASR) and parent-report on the Adult Behavior Checklist (ABCL) to obtain multi-informant composite scores. The ASR and ABCL are parallel versions of the CBCL for older individuals. We used T -scores ( M  = 50, SD  = 10) as dimensional symptom measures, with scores above 60 indicating elevated/at-risk and scores above 70 indicating clinically significant symptoms. For example, items included: “…. your child feels worthless or inferior” (internalizing) “…your child gets in many fights” (externalizing). The CBCL, ASR, and ABCL have good–excellent validity, test–retest reliability, and internal consistency [ 66 , 67 ].

Parent Psychopathology: Beck Depression Inventory (BDI-I; BDI-II) [ 68 , 69 ] . We measured depressive symptoms of the primary caregiver (typically the mother) using self-report on the BDI-I at Wave 1 and the BDI-II at Wave 3. Mothers rated each of the 21 items on a 4-point severity scale. For example, items included a choice between “1.) I do not feel sad. 2.) I feel sad. 3.) I am sad all the time and I can’t snap out of it. 4.) I am so sad or unhappy I can’t stand it.” Total possible scores could range from 0–63, with higher scores indicating greater severity of depression. The BDI is a widely used and extensively validated self-report measure of depression in adults [ 70 ].

Parenting Stress Due To Dysfunctional Interactions: Parenting Stress Index-Short Form (PSI-SF) [ 71 ] . We measured stress-inducing dysfunctional parent–child interactions using the PSI-SF, a widely used self-report measure assessing stress experienced by parents related to their role as a parent. In particular, we used the 12-item Parental-Child Dysfunctional Interaction (PCDI) subscale which measures a parent’s perception of dysfunction in the parent–child relationship that contributes to the parent’s feelings of parental stress. Participants’ mothers rated each item on a scale from 1 (strongly agree) to 5 (strongly disagree). For example, items included: “My child does not like me or want to be close.” Higher scores indicated higher levels of maternal parenting stress. The PSI-SF has demonstrated good test-test reliability, internal consistency, and validity, with the reliability of the subscales ranging from 0.68 to 0.85 and the internal consistency ranging from 0.80 to 0.87 [ 72 , 73 ]. In our sample, the internal consistency (Cronbach’s alpha) of the Parental-Child Dysfunctional Interaction subscale at Wave 1 and Wave 3 were 0.88 and 0.93 respectively.

Cumulative childhood adversity: Adverse Childhood Experiences questionnaire (ACE) [ 74 ] . We measured cumulative experiences of childhood adversity via retrospective report by on the ACE questionnaire at Wave 4, which assesses experiences of childhood abuse, neglect, and household dysfunction during the first 18 years of life. ACE scores ranged from 0–10, with higher scores indicating experiences of multiple types of childhood adversity. For example, items included: “Did you often or very often feel that no one in your family loved you or thought you were important or special?” The ACE questionnaire is a commonly used measure to assess for the cumulative effect of multiple forms of childhood adversity [ 75 ], and has good reliability and validity [ 76 ]—including at least moderate test–retest reliability of retrospective reports [ 77 ].

Criterion variables

These were measured at Wave 4 (Young Adulthood).

Borderline Personality Disorder Diagnosis. A licensed clinical psychologist or a graduate student in clinical psychology, under close supervision, conducted a clinical interview with participants using the Structured Clinical Interview for DSM-IV-TR (SCID) [ 78 ] and the Borderline Personality Disorder (BPD) module of the SCID-II (SCID-II) [ 79 ]. The SCID-II is a semi-structured interview widely used in both research and clinical practice, with research indicating good to excellent inter-rater reliability [ 80 ]. A participant met criteria for a diagnosis of BPD if the clinician rated the participant at or above threshold on five of the nine symptom traits. A single dichotomous variable (0 or 1) reflected a BPD diagnosis.

Borderline Personality Disorder Dimensional Features. Because both diagnostic interview and self-report measures may yield optimal assessment of BPD [ 81 ], we also included a dimensional measure of BPD in order to assess and validate the categorical measure of BPD. For a large subset of the sample, a 15-item self-report scale was included, based on the BPD module of the Structured Clinical Interview for DSM-5 Axis II disorders (SCID-II) [ 82 ]. However, every participant did not complete this self-report measure, as it was added after data collection began. Additionally, some participants completed only interviews and did not return their packet of questionnaires including this measure. Each item of the measure is rated dichotomously (0 = No, 1 = Yes), so that the total possible score ranged from 0–15, with higher scores indicating more features of BPD. For example, items included: “Have you often become frantic when you thought that someone you really cared about was going to leave you?” This scale is consistent with DSM-5 BPD criteria, and has been used in several other studies, with satisfactory internal reliability (α = 0.81) [ 83 , 84 ].

To ascertain whether domains of impairment were related specifically to BPD status, we added covariates empirically associated with BPD and associated predictors: (1) SES—a composite measure of parent report of family income and maternal education in childhood; (2) parent report of child’s race/ethnicity in childhood; and (3) participant age in young adulthood.

Data analytic plan

Statistical analyses were performed with RStudio, version 1.2.1335. First, we computed descriptive statistics and zero-order correlations across potential predictors, background variables of interest, and young adult BPD (measured both categorically and dimensionally). Second, we conducted a series of (a) binary logistic regressions to test whether each theory-informed predictor independently predicted a young-adult diagnosis of BPD and (b) parallel linear regressions regarding dimensional features of BPD. We calculated effect sizes of Cohen’s d for the dichotomous criterion and R 2 for the dimensional measure. Given the many initial predictors, we deployed the stringent criterion that a predictor be retained for subsequent analyses only if it displayed a medium (or larger) effect size in relation to the respective categorical or dimensional measure of BPD. For Cohen’s d, we considered effect sizes ≧ 0.2 as small, ≧ 0.5 as medium, and ≧ 0.8 as large; for R 2 , we considered ≧ 0.02 as small, ≧ 0.13 as medium, and ≧ 0.26 as large (Cohen, 1988). Third, we tested whether predictors meeting this criterion continued to do so when adjusting for sociodemographic covariates (baseline SES, participant race/ethnicity, and participant age), using (a) binary logistic regressions or (b) linear regressions, respectively.

We added predictors maintaining significance into separate models by developmental period (Model 1 = childhood; Model 2 = late adolescence), Given the small subset with a BPD diagnosis, we used Firth’s penalized likelihood method in binary logistic regressions to minimize bias introduced by several independent variables [ 85 ]. For exploratory moderator analyses, we conceptualized a moderator as a baseline factor that might reveal differential predictor-criterion associations at different levels of the putative moderator [ 86 ]. Our moderator of interest included baseline (Wave 1) socioeconomic status. Understanding that such analyses are non-hypothesis-driven, we placed interaction terms of the putative predictor x SES moderator at the third step of each significant predictor regression model.

Descriptive analyses and correlations

A total of 19 participants met criteria for a diagnosis of BPD. Fourteen (74%) had received a childhood diagnosis of ADHD (χ 2 (3, N  = 199) = 1.1, p  = 0.3, OR : 1.31, CI: 0.79, 2.17), with a majority of them having received a childhood diagnosis of ADHD-C (58%).

Tables 1 , 2 , 3 and 4 present intercorrelations among key variables. Because maternal- and teacher-report of childhood hyperactivity/impulsivity (W1 SNAP-HI), inattention (W1 SNAP-IA), overt aggression (W1 CSBS), and relational aggression (W1 CSBS), plus maternal- and self-report of late adolescent hyperactivity/impulsivity (W3 SNAP-HI), inattention (W3 SNAP-IA), externalizing symptoms (W3 ASR/ABCL), and internalizing symptoms (W3 ASR/ABCL) were highly correlated, we averaged ratings across mother and teacher (childhood) and mother and self (early adulthood) to create a composite score for each domain.

For young-adult categorical BPD diagnoses, significant childhood point biserial correlates included hyperactivity/impulsivity (W1 SNAP-HI; r pb  = 0.17, p  < 0.05), low executive functioning (W1 ROCF; r pb  = 0.22, p  < 0.01), and a history of overall adversity (W4 ACEs; r pb  = 0.32, p  < 0.01) (see Table 1 ). Significant late-adolescent point biserial correlates included hyperactivity/impulsivity (W3 SNAP-HI; r pb  = 0.40, p  < 0.01), inattention (W3 SNAP-IA; r pb  = 0.38, p  < 0.01), externalizing symptoms (W3 ASR/ABCL; r pb  = 0.39, p  < 0.01), internalizing symptoms (W3 ASR/ABCL ; r pb  = 0.34, p  < 0.01), and maternal psychopathology (W3 BDI-II ( r pb  = 0.16, p  < 0.05) (see Table 2 ).

Regarding young-adult dimensionally scored features of BPD, childhood hyperactivity/impulsivity (W1 SNAP-HI; r  = 0.43, p  < 0.01), childhood inattention (W1 SNAP-IA; r  = 0.29, p  < 0.01), low executive functioning (W1 ROCF; r  = 0.25, p  < 0.01), externalizing symptoms (W1 CBCL; r  = 0.36, p  < 0.01), internalizing symptoms (W1 CBCL; r  = 0.24, p  < 0.01), overt aggression (W1 CSBS; r  = 0.38, p  < 0.01), relational aggression (W1 CSBS; r  = 0.32, p  < 0.01), negative peer nominations (W1 Peer Report; r  = 0.35, p  < 0.01), and a cumulative history of childhood adversity (W4 ACEs; r  = 0.47, p  < 0.01) were significant correlates. Late adolescent hyperactivity/impulsivity (W3 SNAP-HI; r  = 0.52, p  < 0.01), inattention (W3 SNAP-IA; r  = 0.43, p  < 0.01), low executive functioning (W3 ROCF;; r  = 0.20, p  < 0.05), externalizing symptoms (W3 ASR/ABCL; r  = 0.63, p  < 0.01), internalizing symptoms (W3 ASR/ABCL ; r  = 0.57, p  < 0.01), maternal psychopathology (W3 BDI-II; r  = 0.25, p  < 0.01), and maternal parenting stress due to dysfunctional interactions (W3 PSI-PCDI; r  = 0.23, p  < 0.05) were all significantly correlated with young adult BPD features (Table 4 ).

Predictors of young adult BPD diagnosis

In the binary logistic regressions with Firth’s penalized likelihood method, conducted to assess independent predictors of the dichotomous outcome of meeting (vs. not meeting) diagnostic criteria for BPD in young adulthood, we initially tested whether each predictor of interest was significantly associated with BPD, followed by inclusion of (a) covariates and (b) other significant predictor variables according to developmental period (Table 5 ).

Among childhood predictors, hyperactivity/impulsivity ( p  < 0.05; d  = 0.58) and low executive functioning ( p  < 0.01; d  = 0.76) each predicted BPD diagnostic status in young adulthood, but only low executive functioning maintained significance after adjusting for covariates ( p  < 0.05). As well, the childhood ACE score was a significant predictor, even with adjustment for covariates ( p  < 0.001; d  = 1.14). Regarding for late-adolescent predictors, hyperactivity/impulsivity ( p  < 0.001; d  = 1.45), inattention ( p  < 0.001; d  = 1.36), externalizing ( p  < 0.001; d  = 1.41), and internalizing ( p  < 0.001; d  = 1.23) symptoms each predicted young adult BPD, adjusting for covariates. Maternal psychopathology did not survive inclusion of covariates ( p  = 0.093).

Finally, we entered all predictors with a medium or larger effect size (Cohen’s d ≧ 0.5) that had maintained significance after inclusion of covariates into models divided by developmental period. In childhood, low executive functioning ( p  = 0.012) and the ACE score maintained significance ( p  = 0.003). In the late-adolescent predictor model, only inattentive symptoms maintained marginal significance ( p  = 0.059), but hyperactivity/impulsivity ( p  > 0.05), internalizing symptoms ( p  > 0.05), and externalizing symptoms ( p  > 0.05) did not.

Predictors: Young adult dimensional BPD features

Via linear regressions, childhood hyperactivity/impulsivity ( p  < 0.001; R 2  = 0.182), inattention ( p  < 0.001; R 2  = 0.079), low executive functioning ( p  = 0.004; R 2  = 0.054), externalizing symptoms ( p  < 0.001; R 2  = 0.121), internalizing symptoms ( p  = 0.004; R 2  = 0.051), overt aggression ( p  < 0.001; R 2  = 0.137), relational aggression ( p  < 0.001; R 2  = 0.094), negative peer nominations ( p  < 0.001; R 2  = 0.114), and maternal psychopathology ( p  = 0.045; R 2  = 0.020) independently predicted young-adult features of BPD. Of these, only childhood hyperactivity/impulsivity ( p  < 0.001), inattention ( p  < 0.001), externalizing symptoms ( p  < 0.001), internalizing symptoms ( p  < 0.05), overt aggression ( p  < 0.001), relational aggression ( p  < 0.001), and negative peer nominations ( p  < 0.001) maintained significance after adjusting for covariates. Maternal parenting stress due to dysfunctional interactions became significant after adjusting for covariates ( p  < 0.05). The ACE score significantly predicted young adult BPD features ( p  < 0.001; R 2  = 0.213), even after adjusting for covariates ( p  < 0.001).

For late-adolescent predictors, hyperactivity/impulsivity ( p  < 0.001; R 2  = 0.265), inattention ( p  < 0.001; R 2  = 0.177), low executive functioning ( p  < 0.05; R 2  = 0.033), externalizing symptoms ( p  < 0.001; R 2  = 0.398), internalizing symptoms ( p  < 0.001; R 2  = 0.317), maternal psychopathology ( p  < 0.01; R 2  = 0.053), and maternal parenting stress due to dysfunctional interactions ( p  < 0.05; R 2  = 0.043) each independently predicted features of BPD in young adulthood. Of these, hyperactivity/impulsivity ( p  < 0.001), inattention ( p  < 0.001), externalizing symptoms ( p  < 0.001), internalizing symptoms ( p  < 0.001), maternal psychopathology ( p  < 0.05), and maternal parenting stress due to dysfunctional interactions ( p  < 0.01) maintained significance after adjusting for covariates. Low executive functioning did not.

Finally, we entered predictors with a medium (or above) effect size ( R 2 ≧ 0.13)—that had maintained significance after inclusion of covariates—into separate models by developmental period. In childhood, only childhood hyperactivity/impulsivity ( p  < 0.01) and the ACE score maintained significance ( p  < 0.001), but overt aggression did not ( p  > 0.05). As for late-adolescent predictors, only externalizing ( p  < 0.001) and internalizing symptoms ( p  < 0.01) maintained significance—but not hyperactivity/impulsivity ( p  > 0.05) or inattention ( p  > 0.05).

Exploratory moderator analyses

Regarding categorical young adult BPD diagnosis, no predictor x moderator interactions emerged as statistically significant. For young adult dimensional BPD features, only an interaction between (a) low childhood executive functioning (predictor) and (b) low childhood socioeconomic status (moderator) (W1 SES; Δ R 2  = 0.022, p  < 0.05) emerged as statistically significant. Here, it was the combination of low executive functioning and low baseline SES that predicted higher levels of BPD dimensional features.

Leveraging a well-characterized longitudinal female sample with and without carefully diagnosed childhood ADHD, we examined theory-informed predictors of young adult BPD—considered both categorically and dimensionally—from childhood and late-adolescent timespans. Although we emphasize caution regarding interpretation of findings due to our small sample size, this investigation extends research from our laboratory on developmental predictors of self-harm behaviors [ 87 ] to include borderline personality disorder as a criterion measure. We note that individuals with BPD—a condition characterized by intense and pervasive dysregulation of emotion, behavior, cognition, and relationships—may or may not engage in self-harm.

First, regarding our categorical measure of BPD, using binary logistic regressions with correction for small sample size, we found that—as hypothesized—a cumulative history of childhood adversity, as measured by the ACE score, predicted BPD diagnosis. Low EF in childhood was also a significant predictor, even after adjusting for ACE scores and demographic covariates. Regarding late-adolescent predictors, hyperactivity/impulsivity, inattention, internalizing, and externalizing symptoms each independently predicted young adult BPD diagnosis after adjusting for covariates, but maternal depression did not. In stringent analyses accounting for all independently significant late adolescent predictors, only symptoms of inattention were independently (albeit marginally) related to young adult BPD diagnosis.

Second, with respect to our dimensional measure of BPD features, we found—consistent with hypotheses—that both childhood hyperactivity/impulsivity and a cumulative history of childhood adversity from the ACE score predicted young adult BPD features, with adjustment for covariates. Furthermore, childhood inattention, externalizing symptoms, internalizing symptoms, overt aggression, relational aggression, negative peer nominations, and maternal parenting stress due to dysfunctional interactions also independently predicted young adult BPD features after adjusting for covariates. Yet in the final model, including all childhood predictors with a medium (or larger) effect size that had survived covariates, only childhood hyperactivity/impulsivity and the ACE score maintained significance. As for late-adolescent predictors of the dimensional outcome, hyperactivity/impulsivity, inattention, externalizing symptoms, internalizing symptoms, maternal psychopathology, and maternal parenting stress due to dysfunctional interactions maintained significance after adjusting for covariates, but low executive functioning did not. In the final analysis, adding all surviving predictors in the same model, only late-adolescent externalizing and internalizing symptoms maintained significance. Finally, as for exploratory moderator analyses, we found an interaction between low childhood executive functioning and low socioeconomic status at baseline was significant, suggesting that socioeconomic disadvantage may compound the predictive effects of low executive functioning with respect to later BPD dimensional scores.

Overall, the child and adolescent predictors of later BPD are largely consistent with those from previous investigations [ 8 , 10 , 18 , 25 , 26 , 43 ], emerging here from a carefully controlled prospective investigation. Regarding ADHD symptoms, almost 75% of women who met criteria for BPD in young adulthood had diagnoses of childhood ADHD, most often characterized by high levels of impulsivity (ADHD-C). This finding is consistent with both cross-sectional and longitudinal research, as well as theoretical models of the developmental course of individuals with high levels of early impulsivity, related to BPD as an end-point [ 11 ].

For ADHD dimensions, our findings add to the limited number of studies examining hyperactive, impulsive, and inattentive symptoms and their severity across development, especially beginning in childhood [ 8 , 25 , 26 , 27 ]. That hyperactivity/impulsivity in childhood and adolescence did not significantly predict later categorical BPD diagnosis was unexpected and may relate to our small sample size. Yet regarding our dimensional measure, when adjusting for covariates and other predictors, hyperactivity/impulsivity in childhood did significantly predict later BPD features. This finding is consistent with the only two other known studies to our knowledge that have examined prospective associations between childhood impulsivity and later BPD [ 26 , 27 ]. Two studies have found prospective prediction from both adolescent hyperactivity/impulsivity and inattention to later BPD [ 8 , 25 ]—along this line, note our marginally significant prediction of categorical BPD from late-adolescent inattentive symptoms: In both Carlson et al. [ 25 ] and a recent machine learning study of 128 variables related to risk for BPD, inattention in adolescence emerged as an important predictor [ 8 ]. Each dimension of ADHD appears to play an important role in the development of BPD symptoms.

The finding linking low EF in childhood to young adult BPD diagnosis is also consistent with previous (yet limited) research. In the only known prospective longitudinal study to date examining childhood EF as predictive of later BPD, a composite measure of EF at age 5 predicted BPD symptoms at age 12 [ 26 ]. When we examined low childhood EF and BPD dimensional features, our findings were not significant. Still, moderator analyses revealed that girls with both low EF and low socioeconomic status were at especially risk for high levels of BPD features. Perhaps difficulties in low EF are related to high or clinically significant BPD in the context of socioeconomic disadvantage.

Regarding internalizing and externalizing symptoms, we found that high levels of each were related to later BPD features after adjusting for other important predictors—but not when we measured BPD categorically. Furthermore, multiple forms of aggression in childhood including both overt and relational aggression, plus negative peer nominations, predicted young adult BPD features, but these findings did not maintain significance in the presence of other important predictors. Thus, symptoms of aggression and peer preference are important in childhood as risk factors for later BPD, yet other factors—childhood hyperactivity/impulsivity and trauma-remain statistically superior. Indeed, there was substantial overlap in our measures of aggression, negative peer nomination, hyperactivity/impulsivity, and broadband externalizing symptoms. Overall, our key findings replicate those from Stepp et al. [ 43 ], who showed that adolescent internalizing and externalizing symptoms predict adult BPD (for additional research, see Belsky et al., [ 26 ]; and Geselowitz et al., [ 10 ]). Adolescence appears to be a particularly sensitive period during which vulnerability for the development of severe and pervasive dysregulation across the lifespan may be realized [ 44 ].

Although maternal psychopathology—plus both child and late-adolescent maternal parenting stress due to dysfunctional interactions—were each independently associated with young adult BPD features after adjusting for covariates, their effect sizes were small. Any implications require replication. Findings from other research indicate that parental invalidation and negative parenting practices may well be stronger predictors [ 53 ].

Consistent with a large body of research linking a history of childhood adversity/trauma with later BPD [ 88 ], we found that a cumulative history of childhood adversity, measured by the ACE scale, was a crucial predictor of both young-adult BPD diagnosis and dimensional features. This measure of cumulative history of childhood adversity is retrospective—and may therefore be better characterized as a subjective experience of childhood trauma rather than objective experiences of childhood trauma. These findings are consistent recent data finding that risk of psychopathology is high among individuals with subjective reports of childhood maltreatment regardless of whether these experiences were validated by objective measures [ 89 ]. As well, the ACE measure we used constitutes the gold standard in the field.

Our results support theories that transactions between dispositional and environmental factors over time can lead to a cycle of dysregulation of emotion, behavior, and cognition as well as difficult interpersonal relationships [ 1 , 11 ]. Indeed, findings support Linehan’s Biosocial Theory plus recent developmental models of females with ADHD [ 11 ]. That is, behaviorally expressed impulsivity may be a risk factor for a range of outcomes, including BPD. As development progresses, children with trait impulsivity may experience childhood trauma—linked with, for example, intergenerational trauma, maladaptive parenting practices (especially in relation to the child’s impulsive behavior), and/or parents’ own behavioral impulsivity—which may then transactionally escalate the development of BPD. The original impulsivity may, via heterotypic continuity, come to be expressed as a combination of internalizing and externalizing dimensions, leading to BPD [ 11 , 90 ]—a suggestion requiring further empirical investigation. Future research should include prospective temporal assessment of these domains, as well as other environmental mediators (e.g., peer relationship influences, substance use) and valid measures of behavioral parental invalidation, to assess multi-factor etiological influences.

Clinical implications

Although our sample size is too small to draw definitive clinical recommendations, we provide several ideas for possible clinical and public health implications, emphasizing caution in interpretation of results related to study limitations (see below for more detail in this regard). First, findings highlight the longstanding effects of early experiences of adversity and trauma. Prevention of these childhood experiences, especially through public health initiatives, cannot be overemphasized. Second, our results reveal the importance including global EF deficits in childhood as indicators of risk for BPD, in addition to the focus on childhood impulsivity. These findings have implications for guiding early clinical assessment and intervention (e.g., through early EF skills training) to prevent later BPD. In short, we highlight the need for interventions before the adolescent period, which appears to be an especially sensitive time of risk [ 11 ].

Children with histories of adversity/trauma and/or deficits in EF could receive interventions targeting emotion dysregulation, a mechanism linked to the development of BPD, such as Dialectical Behavior Therapy for Children (DBT-C) [ 91 ], Parent–Child Interaction Therapy (PCIT) [ 92 ], or creative combinations of these therapies [ 93 ]. Widespread assessment of early risk factors to identify individuals at risk remains a challenge. We also recommend that evidence-based treatments for severe emotion dysregulation (i.e., DBT) include remediation of EF deficits.

Limitations and future directions

Our study has several important limitations. First, our sample size is small for the categorical BPD variable, with only 19 females meeting diagnostic criteria for BPD in young adulthood, clearly limiting statistical power. Note that we used Firth’s penalized likelihood method to statistically account for our small sample [ 94 ]. Furthermore, only a subset of our sample completed the self-report dimensional BPD measure. We emphasize the need for replication and cautious interpretation of findings. Second, we did not have symptom-level data available for our categorical measure of BPD, preventing us from evaluating clinician-assessed dimensions of BPD symptoms. Future research would benefit from examining dimensional severity of BPD symptoms, as well as specific traits, some of which have recently been linked to increased risk for a suicide attempt [ 95 ]. Third, several measures—including those of BPD features, cumulative trauma history, and ADHD (in part) were self-reported—and may thus be subject to bias. Fourth, we were not able to peform mediator analyses and therefore cannot add to the literature on potential “driving” mechanisms between childhood and adulthood (e.g., emotion dysregulation). Fifth, we did not separate predictor symptom domains of hyperactivity vs. impulsivity, as psychometrics are superior when using the full 9-item Hyperactivity/Impulsivity scale. As well, this measure is more consistent with the DSM’s layout of symptoms. Although we support the separation of theses symptoms in future research—see the excellent national analysis by Tiger et al. [ 27 ]—we elected to include the full 9-item scale. Sixth, our measure of externalizing symptoms in late adolescence (ASR, ABCL) included measures of aggression, but given our multiple testing, we did not examine aggression per se during this developmental window. Future research would benefit from examining aggressive symptoms across development, given empirical research and theory linking high levels aggression and peer problems with later BPD [ 11 , 18 ]. Seventh, there is controversy over whether the Rey-Osterrieth Complex Figure Test captures meaningful variance in executive functioning [ 96 ]. It could be that this measure is a better index of visual-motor integration and overall neuropsychological functioning than of executive functioning [ 97 , 98 ]. Future research should investigate different domains of neuropsychological functioning and BPD development. Eighth, there is definitional overlap between ADHD and BPD, given that both are characterized by impulsivity, which could account for some of the present results. Finally, a key limitation is the timing of our BPD measure—we measured BPD only during young adulthood, but some participants may already have met criteria for BPD in adolescence.

Still, key strengths include a carefully diagnosed, ethnically and socioeconomically diverse sample of females; emphasis on multi-domain and multi-informant measures; high sample retention; and a prospective (and ongoing) longitudinal design. Moreover, we included stringent use of covariates and statistical penalization. Finally, we examined multiple domains of risk for BPD simultaneously and included several measures of BPD symptomology.

The current findings add to existing research on developmental pathways to BPD, especially among females with ADHD. Future directions should include replication, further examination of dimensions of both ADHD and EF, and distinct types of traumatic life events across development [ 99 ] as related to later BPD. Sensitive measures of early emotional invalidation are also necessary [ 53 ].

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Abbreviations

Adverse childhood experiences scale

Attention-deficit/hyperactivity disorder

Attention-deficit/hyperactivity disorder, combined presentation

Attention-deficit/hyperactivity disorder, inattentive presentation

Beck depression inventory

• Borderline personality disorder

Child behavior checklist

Children’s social behavior scale

Executive functioning

Hyperactivity/impulsivity

Inattention

Nonsuicidal self-injury

Parenting stress index

Parent–child dysfunctional interactions

Rey osterrieth complex figure task

Structured clinical interview for DSM disorders

Swanson, Nolan, and Pelham questionnaire

Crowell SE, Beauchaine TP, Linehan MM. A biosocial developmental model of borderline personality: Elaborating and extending Linehan’s theory. Psychol Bull. 2009;135(3):495–510.

Article   PubMed   PubMed Central   Google Scholar  

Pompili M, Girardi P, Ruberto A, Tatarelli R. Suicide in borderline personality disorder: a meta-analysis. Nord J Psychiatry. 2005;59(5):319–24.

Article   PubMed   Google Scholar  

Olfson M, Blanco C, Marcus SC. Treatment of adult depression in the United States. JAMA Intern Med. 2016;176(10):1482–91.

Torgersen S, Kringlen E, Cramer V. The prevalence of personality disorders in a community sample. Arch Gen Psychiatry. 2001;58(6):590–6.

Article   CAS   PubMed   Google Scholar  

Comtois KA, Carmel A. Borderline personality disorder and high utilization of inpatient psychiatric hospitalization: Concordance between research and clinical fiagnosis. J Behav Health Serv Res. 2016;43(2):272–80.

Linehan MM, Korslund KE, Harned MS, Gallop RJ, Lungu A, Neacsiu AD, et al. Dialectical behavior therapy for high suicide risk in individuals with borderline personality disorder: A randomized clinical trial and component analysis. JAMA Psychiat. 2015;72(5):475–82.

Article   Google Scholar  

Linehan MM. Cognitive-behavioral treatment of borderline personality disorder. Guilford Press; 1993.

Beeney JE, Forbes EE, Hipwell AE, Nance M, Mattia A, Lawless JM, et al. Determining the key childhood and adolescent risk factors for future BPD symptoms using regularized regression: comparison to depression and conduct disorder. J Child Psychol Psychiatry. 2021;62(2):223–31.

Stepp SD, Lazarus SA, Byrd AL. A systematic review of risk factors prospectively associated with borderline personality disorder: Taking stock and moving forward. Personal Disord. 2016;7(4):316–23.

Geselowitz B, Whalen DJ, Tillman R, Barch DM, Luby JL, Vogel A. Preschool age predictors of adolescent borderline personality symptoms. J Am Acad Child Adolesc Psychiatry. 2021;60(5):612–22.

Beauchaine TP, Hinshaw SP, Bridge JA. Nonsuicidal self-injury and suicidal behaviors in girls: The case for targeted prevention in preadolescence. Clin Psychol Sci. 2019;7(4):643–67.

Matthies SD, Philipsen A. Common ground in Attention Deficit Hyperactivity Disorder (ADHD) and Borderline Personality Disorder (BPD)-review of recent findings. Borderline Personal Disord Emot Dysregul. 2014;1(3):1–13.

Google Scholar  

Miller TW, Nigg JT, Faraone SV. Axis I and II comorbidity in adults with ADHD. J Abnorm Psychol. 2007;116(3):519–28.

Bernardi S, Faraone SV, Cortese S, Kerridge BT, Pallanti S, Wang S, et al. The lifetime impact of attention-deficit hyperactivity disorder: Results from the National Epidemiologic Survey on Alcohol and Related Conditions. Psychol Med. 2012;42(4):875–87.

Andrulonis PA. Disruptive behavior disorders in boys and the borderline personality disorder in men. Ann Clin Psychiatry. 1991;3(1):23–6.

Philipsen A, Limberger MF, Lieb K, Feige B, Kleindienst N, Ebner-Priemer U, et al. Attention-deficit hyperactivity disorder as a potentially aggravating factor in borderline personality disorder. Br J Psychiatry. 2008;192:118–23.

Matthies S, Van Elst LT, Feige B, Fischer D, Scheel C, Krogmann E, et al. Severity of childhood attention-deficit hyperactivity disorder-a risk factor for personality disorders in adult life? J Pers Disord. 2010;25(1):101–14.

Stepp SD, Burke JD, Hipwell AE, Loeber R. Trajectories of Attention Deficit Hyperactivity Disorder and oppositional defiant disorder symptoms as precursors of borderline personality disorder symptoms in adolescent girls. J Abnorm Child Psychol. 2012;40(1):7–20.

Miller CJ, Flory JD, Miller R, Harty SC, Newcorn JH, Halperin JM. Childhood ADHD and the emergence of personality disorders in adolescence: A pospective follow-up study. J Clin Psychiatry. 2008;69(9):1477–84.

Rasmussen P, Gillberg C. Natural outcome of ADHD with developmental coordination disorder at age 22 years: a controlled, longitudinal, community-based study. J Am Acad Child Adolesc Psychiatry. 2000;39(11):1424–31.

Rey JM, Morris-Yates A, Singh M, Andrews G, Stewart GW. Continuities between psychiatric disorders in adolescents and personality disorders in young adults. Am J Psychiatry. 1995;152(6):895–900.

Fischer M, Barkley RA, Smallish L, Fletcher K. Young adult follow-up of hyperactive children: self-reported psychiatric disorders, comorbidity, and the role of childhood conduct problems and teen CD. J Abnorm Child Psychol. 2002;30(5):463–75.

Burke JD, Stepp SD. Adolescent disruptive behavior and borderline personality disorder symptoms in young adult men. J Abnorm Child Psychol. 2012;40(1):35–44.

Thatcher DL, Cornelius JR, Clark DB. Adolescent alcohol use disorders predict adult borderline personality. Addict Behav. 2005;30(9):1709–24.

Carlson EA, Egeland B, Sroufe LA. A prospective investigation of the development of borderline personality symptoms. Dev Psychopathol. 2009;21(4):1311–34.

Belsky DW, Caspi A, Arseneault L, Bleidorn W, Fonagy P, Goodman M, et al. Etiological features of borderline personality related characteristics in a birth cohort of 12-year-old children. Dev Psychopathol. 2012;24(1):251–65.

Tiger A, Ohlis A, Bjureberg J, Lundström S, Lichtenstein P, Larsson H, et al. Childhood symptoms of attention-deficit/hyperactivity disorder and borderline personality disorder. Acta Psychiatr Scand. 2022;146(4):370–80.

Article   CAS   PubMed   PubMed Central   Google Scholar  

Cumyn L, French L, Hechtman L. Comorbidity in adults with attention-deficit hyperactivity disorder. Can J Psychiatry. 2009;54(10):673–83.

van Dijk F, Lappenschaar M, Kan C, Verkes RJ, Buitelaar J. Lifespan attention deficit/hyperactivity disorder and borderline personality disorder symptoms in female patients: A latent class approach. Psychiatry Res. 2011;190(2–3):327–34.

Hinshaw SP, Nguyen PT, O’Grady SM, Rosenthal EA. Annual Research Review: Attention-deficit/hyperactivity disorder in girls and women: underrepresentation, longitudinal processes, and key directions. J Child Psychol Psychiatry. 2022;63(4):484–96.

Hinshaw SP, Owens EB, Zalecki C, Huggins SP, Montenegro-Nevado AJ, Schrodek E, Swanson EN. Prospective follow-up of girls with attention-deficit/hyperactivity disorder into early adulthood: Continuing impairment includes elevated risk for suicide attempts and self-injury. J Consult Clin Psychol. 2012;80(6):1041–51. https://doi.org/10.1037/a0029451 .

Meza JI, Owens EB, Hinshaw SP. Response inhibition, peer preference and victimization, and self-harm: Longitudinal associations in young adult women with and without ADHD. J Abnorm Child Psychol. 2016;44(2):323–34.

Brown TE. A new understanding of ADHD in children and adults: Executive function impairments. New York, NY: Routledge; 2013.

Book   Google Scholar  

Garcia-Villamisar D, Dattilo J, Garcia-Martinez M. Executive functioning in people with personality disorders. Curr Opin Psychiatry. 2017;30(1):36–44.

Haaland V, Esperaas L, Landrø NI. Selective deficit in executive functioning among patients with borderline personality disorder. Psychol Med. 2009;39(10):1733–43.

Gvirts HZ, Harari H, Braw Y, Shefet D, Shamay-Tsoory SG, Levkovitz Y. Executive functioning among patients with borderline personality disorder (BPD) and their relatives. J Affect Disord. 2012;143(1–3):261–4.

Ruocco AC. The neuropsychology of borderline personality disorder: a meta-analysis and review. Psychiatry Res. 2005;137(3):191–202.

Hagenhoff M, Franzen N, Koppe G, Baer N, Scheibel N, Sammer G, et al. Executive functions in borderline personality disorder. Psychiatry Res. 2013;210(1):224–31.

Williams GE, Daros AR, Graves B, McMain SF, Links PS, Ruocco AC. Executive functions and social cognition in highly lethal self-injuring patients with borderline personality disorder. Personal Disord. 2015;6(2):107–16.

Lampe K, Konrad K, Kroener S, Fast K, Kunert HJ, Herpertz SC. Neuropsychological and behavioural disinhibition in adult ADHD compared to borderline personality disorder. Psychol Med. 2007;37(12):1717–29.

Unoka Z, Richman M J. Richman. Neuropsychological deficits in BPD patients and the moderator effects of co-occurring mental disorders: A meta-analysis. Clin Psychol Rev. 2016;44:1–12.

Eaton NR, Krueger RF, Keyes KM, Skodol AE, Markon KE, Grant BF, et al. Borderline personality disorder co-morbidity: Relationship to the internalizing-externalizing structure of common mental disorders. Psychol Med. 2011;41(5):1041–50.

Stepp SD, Olino TM, Klein DN, Seeley JR, Lewinsohn PM. Unique influences of adolescent antecedents on adult borderline personality disorder features. Personal Disord. 2013;4(3):223–9.

Sharp C, Vanwoerden S, Wall K. Adolescence as a sensitive period for the development of personality disorder. Psychiatr Clin North Am. 2018;41(4):669–83.

Johnson JG, Cohen P, Brown J, Smailes EM, Bernstein DP. Childhood maltreatment increases risk for personality disorders during early adulthood. Arch Gen Psychiatry. 1999;56(7):600–8.

Widom CS, Czaja SJ, Paris J. A prospective investigation of borderline personality disorder in abused and neglected children followed up into adulthood. J Pers Disord. 2009;23(5):433–46.

Whalen DJ, Scott LN, Jakubowski KP, McMakin DL, Hipwell AE, Silk JS, et al. Affective behavior during mother-daughter conflict and borderline personality disorder severity across adolescence. Personal Disord. 2014;5(1):88–96.

Gordon CT, Hinshaw SP. Parenting stress as a mediator between childhood ADHD and early adult female outcomes. J Clin Child Adolesc Psychol. 2017;46(4):588–99.

Infurna MR, Fuchs A, Fischer-Waldschmidt G, Reichl C, Holz B, Resch F, et al. Parents’ childhood experiences of bonding and parental psychopathology predict borderline personality disorder during adolescence in offspring. Psychiatry Res. 2016;246:373–8.

Crawford TN, Cohen PR, Chen H, Anglin DM, Ehrensaft M. Early maternal separation and the trajectory of borderline personality disorder symptoms. Dev Psychopathol. 2009;21(3):1013–30.

Johnson JG, Cohen P, Chen H, Kasen S, Brook JS. Parenting behaviors associated with risk for offspring personality disorder during adulthood. Arch Gen Psychiatry. 2006;63(5):579–87.

Schuppert HM, Albers CJ, Minderaa RB, Emmelkamp PMG, Nauta MH. Severity of borderline personality symptoms in adolescence: relationship with maternal parenting stress, maternal psychopathology, and rearing styles. J Pers Disord. 2015;29(3):289–302.

Musser N, Zalewski M, Stepp S, Lewis J. A systematic review of negative parenting practices predicting borderline personality disorder: Are we measuring biosocial theory’s ‘invalidating environment’? Clin Psychol Rev. 2018;65(January):1–16. Available from: https://doi.org/10.1016/j.cpr.2018.06.003

Hinshaw SP. Preadolescent girls with attention-deficit/hyperactivity disorder: I. Background characteristics, comorbidity, cognitive and social functioning, and parenting practices. J Consult Clin Psychol. 2002;70(5):1086–98. https://doi.org/10.1037//0022-006X.70.5.1086 .

Shaffer D, Fisher P, Lucas CP, Dulcan MK, Schwab-Stone ME. NIMH Diagnostic Interview Schedule for Children Version IV (NIMH DISC-IV): description, differences from previous versions, and reliability of some common diagnoses. J Am Acad Child Adolesc Psychiatry. 2000;39(1):28–38.

Hinshaw SP, Owens EB, Sami N, Fargeon S. Prospective follow-up of girls with attention-deficit/hyperactivity disorder into adolescence: Evidence for continuing cross-domain impairment. J Consult Clin Psychol. 2006;74(3):489–99. https://doi.org/10.1037/0022-006X.74.3.489 .

Owens EB, Zalecki C, Gillette P, Hinshaw SP. Girls with childhood ADHD as adults: Cross-domain outcomes by diagnostic persistence. J Consult Clin Psychol. 2017;85(7):723–36. https://doi.org/10.1037/ccp0000217 .

Swanson JM. Assessment and treatment of ADD students. Irvine: K.C. Press; 1992.

MTA Cooperative Group. A 14-month randomized clinical trial of treatment strategies for attention-deficit/hyperactivity disorder. Arch Gen Psychiatry. 1999;56(12):1073–86.

Bussing R, Fernandez M, Harwood M, Hou Wei, Garvan CW, Eyberg SM, et al. Parent and teacher SNAP-IV ratings of attention deficit hyperactivity disorder symptoms: psychometric properties and normative ratings from a school district sample. Assessment. 2008;15(3):317–28.

Osterrieth PA. The test of copying a complex figure: A contribution to the study of perception and memory. Arch Psychol (Geneve). 1944;30:286–356.

Miller M, Nevado-Montenegro AJ, Hinshaw SP. Childhood executive function continues to predict outcomes in young adult females with and without childhood-diagnosed ADHD. J Abnorm Child Psychol. 2012;40(5):657–68. https://doi.org/10.1007/S10802-011-9599-Y .

Sami N, Carte ET, Hinshaw SP, Zupan BA. Performance of girls with ADHD and comparison girls on the Rey-Osterrieth Complex Figure: evidence for executive processing deficits. Child Neuropsychol. 2003;9(4):237–54. https://doi.org/10.1076/CHIN.9.4.237.23514 .

Hinshaw SP, Carte ET, Sami N, Treuting JJ, Zupan BA. Preadolescent girls with attention-deficit/hyperactivity disorder: II. Neuropsychological performance in relation to subtypes and individual classification. J Consul Clin Psychol. 2002;70(5):1099–111. https://doi.org/10.1037/0022-006X.70.5.1099 .

Achenbach TM. Manual for the child behavior checklist and revised child behavior profile. Burlington, VT: University Associates in Psychiatry; 1991.

Achenbach TM, Rescorla LA. Manual for the ASEBA adult forms & profiles. Burlington, VT: University of Vermont, Research Center for Children, Youth, & Families; 2003.

Nakamura BJ, Ebesutani C, Bernstein A, Chorpita BF. A psychometric analysis of the Child Behavior Checklist DSM-oriented scales. J Psychopathol Behav Assess. 2009;31(3):178–89.

Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory for measuring depression. Arch Gen Psychiatry. 1961;4(6):561–71.

Beck AT, Steer RA, Brown GK. Manual for the beck depression inventory. San Antonio, TX. San Antonio, TX: Psychological Corporation; 1996.

Sprinkle SD, Lurie D, Insko SL, Atkinson G, Jones GL, Logan AR, et al. Criterion validity, severity cut scores, and test-retest reliability of the Beck Depression Inventory-II in a university counseling center sample. J Couns Psychol. 2002;49(3):381–5.

Abidin RR. Parenting Stress Index (PSI) manual. 3rd ed. Charlottesville, VA: Pediatric Psychology Press; 1995.

Haskett ME, Ahern LS, Ward CS, Allaire JC. Factor structure and validity of the parenting stress index-short form. J Clin Child Adolesc Psychol. 2006;35(2):302–12.

Abidin RR, Brunner JF. Development of a parenting alliance inventory. J Clin Child Psychol. 1995;24(1):31–40.

Felitti VJ, Anda RF, Nordenberg D, Williamson DF, Spitz AM, Edwards V, et al. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults The Adverse Childhood Experiences (ACE) Study. Am J Prev Med. 1998;14(4):245–58.

Petruccelli K, Davis J, Berman T. Adverse childhood experiences and associated health outcomes: A systematic review and meta-analysis. Child Abus Negl. 2019;97:104127.

Hardt J, Rutter M. Validity of adult retrospective reports of adverse childhood experiences: review of the evidence. J Child Psychol Psychiatry. 2004;45(2):260–73.

Dube SR, Williamson DF, Thompson T, Felitti VJ, Anda RF. Assessing the reliability of retrospective reports of adverse childhood experiences among adult HMO members attending a primary care clinic. Child Abuse Negl. 2004;28(7):729–37.

First MB, Spitzer RL, Gibbon M, Williams JBW. Structured Clinical Interview for DSM-IV-TR Axis I Disorders, research version, non-patient edition (SCID-I/NP). New York: Biometrics Research, New York State Psychiatric Institute; 2002.

First MB, Spitzer RL, Gibbon M, Williams JBW. Structured Clinical Interview for DSM–IV Personality Disorders (SCID–II). Washington, D.C.: American Psychiatric Press; 1997.

Huprich SK, Paggeot AV, Samuel DB. Comparing the Personality Disorder Interview for DSM-IV (PDI-IV) and SCID-II borderline personality disorder scales: an item-response theory analysis. J Pers Assess. 2015;97(1):13–21.

Hopwood CJ, Morey LC, Edelen MO, Shea MT, Grilo CM, Sanislow CA, et al. A comparison of interview and self-report methods for the assessment of borderline personality disorder criteria. Psychol Assess. 2008;20(1):81–5.

First MB, Williams JB, Benjamin LS, Spitzer RL. Structured clinical interview for DSM-5® personality disorders (SCID-5-PD): With the structured clinical interview for DSM-5® Screening Personality Questionnaire (SCID-5-SPQ). Arlington, VA: American Psychiatric Association; 2016.

Hyland P, Karatzias T, Shevlin M, Cloitre M. Examining the discriminant validity of complex posttraumatic stress disorder and borderline personality disorder symptoms: Results from a United Kingdom population sample. J Trauma Stress. 2019;32(6):855–63.

Frost R, Murphy J, Hyland P, Shevlin M, Ben-Ezra M, Hansen M, et al. Revealing what is distinct by recognising what is common: distinguishing between complex PTSD and Borderline Personality Disorder symptoms using bifactor modelling. Eur J Psychotraumatol. 2020;11:1836864.

Firth D. Bias reduction of maximum likelihood estimates. Biometrika. 1993;80(1):27–38.

Kraemer HC. Discovering, comparing, and combining moderators of treatment on outcome after randomized clinical trials: A parametric approach. Stat Med. 2013;32:1964–73.

Meza JI, Owens EB, Hinshaw SP. Childhood predictors and moderators of lifetime risk of self-harm in girls with and without attention-deficit/hyperactivity disorder. Dev Psychopathol. 2021:1–17. https://doi.org/10.1017/S0954579420000553 .

Stepp SD, Scott LN, Jones NP, Whalen DJ, Hipwell AE. Negative emotional reactivity as a marker of vulnerability in the development of borderline personality disorder symptoms. Dev Psychopathol. 2016;28(1):213.

Danese A, Widom CS. Objective and subjective experiences of child maltreatment and their relationships with psychopathology. Nat Hum Behav. 2020;4(8):811–8.

Ahmad SI, Hinshaw SP. Attention-Deficit/Hyperactivity Disorder, Trait impulsivity, and externalizing behavior in a longitudinal sample. J Abnorm Child Psychol. 2017;45(6):1077–89.

Perepletchikova F, Nathanson D, Axelrod SR, Merrill C, Walker A, Grossman M, et al. Randomized clinical trial of Dialectical Behavior Therapy for preadolescent children with disruptive mood dysregulation disorder: Feasibility and outcomes. J Am Acad Child Adolesc Psychiatry. 2017;56(10):832–40.

Luby JL, Barch DM, Whalen D, Tillman R, Freedland KE. A randomized controlled trial of parent-child psychotherapy targeting emotion development for early childhood depression. Am J Psychiatry. 2018;175(11):1102–10.

Zalewski M, Maliken AC, Lengua LJ, Gamache Martin C, Roos LE, Everett Y. Integrating dialectical behavior therapy with child and parent training interventions: A narrative and theoretical review. Clin Psychol Sci Pract. 2020; Advance online publication. https://doi.org/10.1111/cpsp.12363 .

Adhikary AC, Rahman MS. Firth’s penalized method in Cox proportional hazard framework for developing predictive models for sparse or heavily censored survival data. J Stat Comput Simul. 2021;91(3):445–63.

Yen S, Peters JR, Nishar S, Grilo CM, Sanislow CA, Shea MT, et al. Association of borderline personality disorder criteria with suicide attempts: Findings from the Collaborative Longitudinal Study of Personality Disorders over 10 years of follow-up. JAMA Psychiat. 2020;02115(2):187–94.

Snyder HR, Miyake A, Hankin BL. Advancing understanding of executive function impairments and psychopathology: bridging the gap between clinical and cognitive approaches. Front Psychol. 2015;6:328.

Davies SR, Field AR, Andersen T, Pestell C. The ecological validity of the Rey-Osterrieth Complex Figure: Predicting everyday problems in children with neuropsychological disorders. J Clin Exp Neuropsychol. 2011;33(7):820–31.

Weber RC, Riccio CA, Cohen MJ. Does Rey Complex Figure copy performance measure executive function in children? Appl Neuropsychol Child. 2013;2(6–12). https://doi.org/10.1080/09084282.2011.643964 .

McLaughlin KA, Sheridan MA. Beyond cumulative risk: A dimensional approach to childhood adversity. Curr Dir Psychol Sci. 2016;25(4):239–45.

Download references

Acknowledgements

We express gratitude to the young women who have participated in our ongoing investigation, and their families, as well as the members of the Hinshaw lab both past and present who have made this research possible.

Preparation of this research was supported by Grant R01 MH45064 from the National Institute of Mental Health.

Author information

Authors and affiliations.

Department of Psychology, University of California, Berkeley, 2121 Berkeley Way West, Berkeley, CA, 94720-1650, USA

Sinclaire M. O’Grady & Stephen P. Hinshaw

Department of Psychiatry and Behavioral Sciences, University of California, 675 18th Street, San Francisco, San Francisco, CA, 94107, USA

Stephen P. Hinshaw

You can also search for this author in PubMed   Google Scholar

Contributions

S.M.O. played a lead role in conceptualization and data curation, formal analysis, writing of original draft, and writing of review and editing. S. P. H. played lead role in the original study design, methodology, data collection, as well as supervision of S. M. O., and an equal contribution in the conceptualization, drafting, and critical revision of this work and writing. All authors read and approved the final manuscript.

Authors’ information

S. M. O. is a psychology doctoral student at the University of California, Berkeley studying developmental pathways to self-harm, with emphasis on trait impulsivity, childhood maltreatment, emotion dysregulation, and suicide. S. P. H. is a Distinguished Professor of Psychology at the University of California, Berkeley, and Professor of Psychiatry and Behavioral Sciences at the University of California, San Francisco. He investigates developmental psychopathology and mental illness stigma.

Corresponding author

Correspondence to Sinclaire M. O’Grady .

Ethics declarations

Ethics approval and consent to participate.

This study and its experimental protocols were approved by the ethics committee at the University of California Berkeley’s Committee for Protection of Human Subjects. All methods were performed in accordance with the relevant guidelines and regulations of the Declaration of Helsinki. Written informed consent was obtained from all participants. Written informed consent was obtained from all participants’ parents/legal guardians if their age was less than 16 years. Written informed consent was obtained from parents/legal guardians and participants for vulnerable participants.

Consent for publication

Not applicable.

Competing interests

All authors have no conflicts of interest to declare.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

O’Grady, S.M., Hinshaw, S.P. Developmental predictors of young adult borderline personality disorder: a prospective, longitudinal study of females with and without childhood ADHD. BMC Psychiatry 23 , 106 (2023). https://doi.org/10.1186/s12888-023-04515-3

Download citation

Received : 25 September 2022

Accepted : 02 January 2023

Published : 15 February 2023

DOI : https://doi.org/10.1186/s12888-023-04515-3

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Attention-deficit hyperactivity disorder
• Risk factors
• Longitudinal studies
• Adverse childhood experiences

BMC Psychiatry

ISSN: 1471-244X

• Submission enquiries: [email protected]
• General enquiries: [email protected]

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Peer-reviewed

Research Article

The value of psychological treatment for borderline personality disorder: Systematic review and cost offset analysis of economic evaluations

Affiliation Illawarra Health and Medical Research Institute and School of Psychology, University of Wollongong, Wollongong, New South Wales, Australia

* E-mail: [email protected]

• Denise Meuldijk, 
• Alexandra McCarthy, 
• Marianne E. Bourke, 
• Brin F. S. Grenyer

• Published: March 1, 2017
• https://doi.org/10.1371/journal.pone.0171592
• Reader Comments

Borderline Personality Disorder (BPD) is a common mental health condition with high patterns of service utilisation of inpatient and community treatment. Over the past five years there has been significant growth in research with economic data, making this systematic review a timely update.

Empirical studies written in English or German, published up to December 2015, and cited in major electronic databases were examined using the PRISMA systematic review method. Papers were included that had one of the following: data related to cost of BPD to society, the individual, the carer or families; cost benefits of interventions. Reported cost data were inflated to the year 2015 and converted into US- dollars (USD $) using purchasing power parities.

We identified 30 economic evaluations providing cost data related to interventions for BPD across 134,136 patients. The methodological quality was good, almost all studies fulfilled ≥ 50% of the quality criteria. The mean cost saving for treating BPD with evidence-based psychotherapy across studies was USD $2,987.82 per patient per year. A further mean weighted reduction of USD $1,551 per patient per year (range $83 - $29,392) was found compared to treatment as usual. Evidence-based psychological treatment was both less expensive as well as more effective, despite considerable differences in health cost arrangements between individual studies and countries. Where it was able to be calculated, a significant difference in cost-savings between different types of evidence-based psychotherapies was found.

Individuals with BPD consistently demonstrate high patterns of service utilization and therefore high costs. The findings of this review present a strong argument in favour of prioritizing BPD treatments in reimbursement decisions, both for the affected individual and the family. The provision of evidence based treatment, irrespective of the type of psychological treatment, may lead to widespread reductions in healthcare costs.

Citation: Meuldijk D, McCarthy A, Bourke ME, Grenyer BFS (2017) The value of psychological treatment for borderline personality disorder: Systematic review and cost offset analysis of economic evaluations. PLoS ONE 12(3): e0171592. https://doi.org/10.1371/journal.pone.0171592

Editor: Christian Schmahl, Central Institute of Mental Health, GERMANY

Received: September 23, 2016; Accepted: January 23, 2017; Published: March 1, 2017

Copyright: © 2017 Meuldijk et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the paper and its Supporting Information files.

Funding: This study was funded in part by the University of Wollongong. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. No additional external sources of funding were received.

Competing interests: The authors have declared that no competing interests exist.

Introduction

In a budget constrained health care system, the cost and benefits of psychological interventions are increasingly of interest as decisions regarding resource allocation have forced individuals to re-examine the cost effectiveness of psychological treatments [ 1 , 2 ]. In particular clinicians are increasingly faced with numerous challenges surrounding decisions regarding treatment options which need to be both ethically informed and based on available research findings. Concurrent with the debate on health care reform, and the relative dearth of evidence-based treatment for Personality Disorders (PD), clinicians have become increasingly interested in evaluating the cost-effectiveness of psychological treatments [ 3 ]. Attention to the cost-benefit analysis is crucial in weighing the benefits versus disadvantages of specific interventions and treatment approaches [ 4 ], clearly defining the method, duration and outcomes of interventions. Managed care providers demand that treatments have demonstrated efficacy, minimize costs, and reduce future utilization of expensive resources.

Previous research in the field has highlighted that individuals suffering Borderline Personality Disorder (BPD) pose a high economic burden on society due to their extensive use of treatment services [ 5 – 7 ]. This is understandable as BPD is present in 1–2% of the population, 10% of psychiatric outpatients and between 15% and 25% of inpatients [ 8 – 10 ]. Additionally patients with BPD present to services at a high risk of self-harm, with up to 10% of psychiatric outpatients committing suicide, a rate almost 50 times higher than the general population [ 11 ]. The widespread economic impacts of the condition is aggravated by the fact that individuals suffering BPD typically need several repeated treatments, including urgent interventions in emergency departments and seek help repeatedly and simultaneously from multiple sources [ 12 , 13 ].

Within mental health settings, Sansone and colleagues (2011) [ 6 ] found that patients with BPD symptomology appear to have a significantly higher turnover with primary care physicians and see a greater number of specialists than patients without these symptoms. Additionally, Bender and colleagues (2001) [ 5 ] found that compared to individuals with major depression, those with BPD were significantly more likely to use most types of psychiatric treatment. Finally, in an Australian community sample, Jackson and Burgess (2004) [ 14 ] found that individuals with BPD were more likely than other individuals to seek psychiatric or psychological consultations. Given the seemingly consistent pattern of high utilization of mental health services among individuals with BPD, these findings may reflect the underlying psychological processes of the disorder as well as a general pattern of demand for healthcare services [ 15 ].

Treatment guidelines for BPD support psychotherapy in the community as the treatment of choice [ 16 ]. Studies of different psychological therapies such as mentalization based treatment, transference focused therapy and dialectical behaviour therapy, have demonstrated positive outcomes in relation to symptomology and levels of service use [ 9 ]. However, due to the nature of BPD symptoms, the identification of psychotherapies and utilisation of interventions that are cost-effective with this population is of considerable importance. While many factors influence the likelihood that a treatment is effective, one of the primary factors has become the affordability of treatment. That is, the pure monetary value one saves or loses by investing in one treatment compared to another. Brazier and colleagues (2006) [ 17 ] completed a systematic review and a preliminary economic evaluation of available therapies for BPD patients [ 17 ]. In 2014, Brettschneider and colleagues [ 18 ] performed a systematic literature review of 15 existing economic evaluations of treatments for BPD up to 2012, but the economic evaluations reviewed used different comparators to define costs, which made comparability of the data difficult and the findings not sufficient to draw robust conclusions for all treatments. There has been no updated review over the past 5 years despite a surge in economic data available.

Therefore the present study aims to systematically review and synthesize the literature on this topic and provide an updated overview (i.e. up to December 2015). We are specifically focused at looking at the cost-benefit of recognised evaluated treatments, as the results can be used to gain insight into how much society is spending on BPD, and potentially how much can be saved if effective therapy is offered. The information can be helpful in setting priorities for health care efficiency research as high societal costs present a strong argument in favour of prioritizing BPD treatments in reimbursement decisions, in terms of health insurance benefits for affected individuals and funding grants to provide accessible and effective treatment services.

Materials and methods

Protocol and registration.

This review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) Statement for Reporting Systematic Reviews [ 19 ] and additional guidelines for conducting and reporting systematic reviews [ 20 ]. The protocol was registered by the International Prospective Register of Systematic Reviews (registration number: CRD42016037305).

Data sources

Literature was searched for relevant articles published up to December 2015 using the following online databases: Psychological and Behavioural Sciences, PsychInfo, Scopus, Medline, Web of Science and PubMed. Search terms used for each database included the following: Borderline Personality Disorder OR Personality Disorder AND cost OR cost analysis OR cost benefit OR cost effectiveness OR economic analysis OR (burden AND (society OR community OR individual OR carer OR carers)). This search identified papers directly related to the research questions.

Study selection

The eligibility of the studies was assessed in two steps. First titles and abstracts were screened. Subsequently, articles considered as relevant were obtained and the full text was screened. All original studies reporting cost or cost-effectiveness data of BPD were included.

Studies were eligible for further consideration in this review if they related to the costs or benefits of interventions for BPD (specifically cost or benefit to the individual, carer or family, or society either through direct (e.g. medical) or indirect (e.g. loss of productivity) costs or benefits. Studies were excluded if they were not in English or German, or did not publish original empirical data (e.g. newspaper articles, reviews, commentaries and editorials). Records identified by the searching process were screened by two authors (DM, AM) and checked by another (BG).

Data extraction and risk of bias

Two authors (DM and AM) extracted data from the included studies, which was then independently reviewed by another author (BG). This author was blind to prestige factors, including authors, institutions, journal titles, and publishers. The authors compared their findings and discrepancies were discussed and resolved to reach a consensus.

Data extracted included: country of publication , cost year , study sample characteristics (number of participants , age , gender , types of treatment compared) , cost (per population and per patient) , and if stated , type of economic evaluation .

Two categories of costs can be distinguished: direct and indirect costs; [ 21 , 22 ]. Direct costs include all treatment costs that arise directly from medical care. These include items such as the cost of psychiatric inpatient and outpatient care, psychological, rehabilitation, medication, or emergency room treatments. Indirect costs cover all costs that occur secondary to the disease. These include items such as production losses and sickness benefit payments due to days absent at work, as well as years lost due to mortality. In addition, the total costs (either indirect or direct) associated with the delivery of the intervention (intervention costs) were also presented.

For the purpose of this review, type of economic evaluation [ 21 ], was described and defined as follows:

• Full economic evaluations . A full economic evaluation does not only compare costs of at least two interventions but also their consequences (outcomes, effects) [ 21 ]. Effects can be measured in natural units (life years gained, parasuicide events avoided, artificial units (quality-adjusted life years [QALY] or disability-adjusted life years [DALY]) or monetary units measured by techniques like willingness-to-pay experiments. Depending on the effect measure employed, full economic evaluations are called cost-effectiveness analyses (natural units), cost-utility analyses (utility measures) or cost-benefit analyses (outcomes valued monetarily).
• Partial economic evaluations . Health economic studies considered to be partial economic evaluations do not make explicit comparisons between alternative interventions in terms of both costs (resource use) and consequences (effects). In partial economic evaluations only the costs of at least two alternatives are compared (i.e. cost analyses, cost-description studies and cost-outcome descriptions). Partial evaluations can be useful in that they can provide elements of information for a full evaluation and help answer questions not related to efficiency.

Whilst the methodologies of full economic evaluations and partical economic evaluations are distinct from each other, both types of economic evaluations were included for the purposes of our review. Debate exists as to whether the information derived from partial economic evaluations are of the same scientific value and usefulness as the information derived by full economic evaluations, nevertheless, including both types of economic evaluations will contribute useful evidence to an understanding of economic aspects of interventions.

Quality assessment

Quality assessment of the studies included was performed by means of the Consensus on Health Economic Criteria (CHEC) checklist. The CHEC-list has been developed using a Delphi method (three Delphi rounds; 23 international experts). The CHEC-list comprehends 19 criteria formulated as a question for answering either by "yes" or "no" [ 23 ].

The results of the quality assessment are displayed as percentage of studies fulfilling each criterion.

Cost-offset analysis

The focus of this review is on costing issues associated with the provision of interventions for the treatment of BPD, including costs, cost-effectiveness and cost offset (i.e., a reduction in health care costs attributable to the intervention provided). Cost offsets originate from the use of mental health services and related costs before, during and after (if applicable) a specific intervention. Interventions included were recognised psychological treatments that have separately been shown to have clinical effectiveness and therefore an evidence-base [ 9 ].Therefore data was extracted based on cost alone, without further consideration of clinical effectiveness as this had been separately established and is outside the scope of this review.

Cost offset PT.

The cost offset of psychotherapeutic interventions (PT) (Cost offset PT) was calculated by subtracting the total costs after the intervention provided from the total costs before the start of the intervention (i.e. pre-post differences in healthcare costs).

Cost offset PT vs. TAU.

The cost offset of a psychotherapeutic intervention PT vs. TAU (Cost offset PT vs. TAU) represents the difference in total costs after the intervention is provided compared to cost related to the provision of treatment as usual (TAU) (i.e. post- difference in healthcare cost related to PT vs. TAU).

In order to assess the costs and cost offsets, and to present a financial evaluation, we applied present-day financial costing standards to the data. Cost data were inflated and converted to 2015 US-$ purchasing power parities (PPP) to ensure comparability of the data [ 24 ].

If an included study did not report the cost year, the year in which the study was accepted for publication was used as a proxy. To further ensure comparability of the data, costs per patient were calculated if costs data related to groups or a population. Subsequently, if the time horizon chosen in the study was less or more than one year, costs were converted to one year costs.

Due to the variation in sample size, means weighted for sample size were calculated and reported when presenting the overall cost offset of PT and the overall cost offset of PT vs TAU of all the economic evaluations identified in this review, and where appropriate, compared using independent samples t test.

Search results

The search of electronic databases resulted in the identification of 4660 studies (4380 with duplicates removed). An additional 25 studies were identified through cross referencing and consultation with experts, yielding 4405 potentially relevant articles.

Of these, 4220 were excluded as their titles and/or abstracts clearly indicated that they did not meet the inclusion criteria.

Of the remaining 185 articles, 137 were excluded because primary focus was not BPD (n = 39), the articles were a secondary review without original data (n = 12), or the articles contained no cost or benefit data (n = 86).

Full text screening was performed for the remaining 48 articles. 19 articles were excluded of which 12 did not report on cost data, four were not mainly focused on BPD and three were review articles; not containing original data. Finally, 29 articles (30 evaluations) were considered in this review (see Fig 1 for PRISMA Flowchart).

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

https://doi.org/10.1371/journal.pone.0171592.g001

Characteristics of studies included

The characteristics of the included studies are summarized in Table 1 .

https://doi.org/10.1371/journal.pone.0171592.t001

Of the total 30 evaluations (29 articles) considered in this review, 19 evaluations were full evaluations and 11 evaluations were partial economic evaluations. Twelve evaluations were conducted in the United Kingdom (UK), five in the United States of America, four each in the Netherlands and in Australia, three in Germany and one each in Norway and Switzerland.

The majority of the evaluations employed the societal perspective. The time horizon of the evaluations ranged from 12 weeks to nine years. Most evaluations included less than 100 patients (n = 24), some even less than 50 (n = 16). We included five economic evaluations with a study population of n > 100 patients. The 30 evaluations bring the total number of patients in this review to n = 134,136.

In all evaluations the majority of patients were female; in most evaluations (n = 16) the proportion of female patients was larger than 80%. The overall weighted percent of female patients across the studies included was 89%.

The mean age of populations ranged from 22 years to 51 years. The weighted average age was 30.65 years across the 30 studies included.

Treatments compared

Regarding treatment comparisons, the studies generally fell into two following categories: 1) studies that examine (one or two competing) psychotherapeutic interventions or 2) studies examining psychotherapeutic interventions versus Treatment As Usual (TAU). A few studies included both categories [ 25 , 26 ]. In addition to the pre-post analysis comparing cost pre- and post intervention the study by Bateman & Fonagy (2003) [ 25 ] and the study by Palmer and colleagues (2006) [ 26 ] also examined the pre- and post- intervention costs in comparison to the provision of TAU. In two evaluations [ 3 , 27 ], the examined intervention was clarified and no comparison of treatments was made.

Studies comparing psychotherapeutic interventions.

A total of fifteen evaluations examined the use of mental health services and related costs associated with the provision of a psychological treatment (i.e. other than TAU) in BPD patients.

Of these, seven focused on the costs solely related to the provision of DBT [ 28 – 33 ] Of latter two studies DBT was compared to an alternative psychotherapeutic intervention [ 32 , 33 ]. Turner and colleagues (2000) [ 32 ] reported on a comparison of DBT with a client-centered therapy control condition (CCT) whereas Heard and colleagues (2000) [ 33 ] conducted a comparison between a stable psychotherapy in community (SCP) and DBT.

Other psychotherapeutic interventions examined in these fifteen evaluations are: Conversational Model (CM) [ 34 , 35 ]; Long term psychoanalytic psychotherapy (LT-P) [ 36 ]; Kvarstein and colleagues (2013) [ 37 ] provided the pre- and post- costs associated with Individual and group therapy giving in a step-down condition (SDC) and Outpatient Individual Psychotherapy (OPC). In two evaluations Schema Focused Therapy (SFT) and Transference Focused Psychotherapy (TFP) were the competing psychotherapeutic interventions [ 38 , 39 ]. Mentalization Based Therapy (MBT) was examined in the study by Bateman & Fonagy (2003) [ 25 ] whereas Palmer and colleagues (2006) [ 26 ] provided the cost associated with the provision of Cognitive Behavioural Therapy (CBT) in addition to TAU. See Table 1 .

Studies comparing psychotherapeutic interventions versus treatment as usual.

Fifteen evaluations studied a psychotherapeutic intervention against treatment as usual (TAU). Of these, eight evaluated dialectical behaviour therapy (DBT) with TAU [ 33 , 40 – 46 ]. Two studies evaluated Mentalization based therapy (MBT) with TAU [ 25 , 47 ]. The following psychotherapeutic interventions were investigated by one evaluation each, using TAU for comparison: Crisis intervention at the general hospital (CI) [ 48 ], Joint crisis plan plus treatment as usual (CP) [ 49 ], manual assisted cognitive behavioural therapy (MACT) [ 50 ], Cognitive behavioural therapy for personality disorders (CBT-PD) [ 51 ] and Cognitive Behavioural Therapy (CBT) in addition to TAU [ 26 ]. In all of these studies, TAU was considered as standard of care for all patients and could involve a range of therapeutic options.

Methodological quality of identified studies.

The results of the assessment of methodological quality are presented in Table 2 . Each criteria of the quality assessment was fulfilled by the majority of the studies (≥ 50% of the studies). Criteria 14 and criteria 19 were fulfilled by 12 (43%) of the 30 studies included. None of the included studies fulfilled all of the quality criteria ( Table 2 ). However, in the context of this review, the description of model details and cost data was suboptimal in most of the studies, with Pasciezny and Conner (2011) [ 43 ] being the exception (8 quality criteria were fulfilled).

https://doi.org/10.1371/journal.pone.0171592.t002

Cost categories

Most of the studies (n = 26) examined considered direct costs (see Table 1 ). These studies reported a diverse range of direct costs: costs for inpatient (costs of general and psychiatric hospital services) and outpatient treatment, rehabilitation, medication or emergency room treatments were assessed by most evaluations.

Five of the studies in this review presented data on both direct and indirect costs [ 3 , 31 , 45 , 52 ]. Indirect costs items included mortality, early retirement, sickness absence, production losses, work disability and reduced quality of life. One study [ 38 ] focussed exclusively on indirect costs, no direct costs were examined.

Intervention related costs were reported by 16 studies (see Table 1 : Intervention Costs)

Three studies included in this review did not report any data on direct or indirect cost [ 27 , 44 , 46 ]. Goodman et al. (2011) [ 27 ] focused on the out-of-pocket expenses and insurance costs directly related to the treatment of BPD (i.e. therefore categorised as intervention costs in this review). In addition, Prendergast & McCausland (2007) [ 44 ] and Richter and colleagues (2013) [ 46 ] did not provide the actual (in)direct costs for inpatient and/or outpatient treatment in the paper, only cost-savings of providing DBT compared to TAU were provided (i.e. cost offset PT vs. TAU).

Cost outcome approach

The majority of studies (18 evaluations) presented data on post- intervention costs only [ 27 , 32 , 33 , 38 – 51 ]. Eleven studies presented data on healthcare costs before ( pre ) and after ( post ) providing the intervention [ 26 , 28 – 31 , 34 , 36 , 37 , 44 , 53 , 54 ]. One study reported only on pre-intervention costs [ 52 ]. See Table 1 .

Results of the economic evaluations

Primary analysis..

The calculated cost offsets associated with the provision of (one or two competing) psychotherapeutic interventions (PT) (i.e. Cost offset PT) and the additional cost-savings of implementing evidence-based psychotherapy compared to TAU (i.e. Cost offset PT vs. TAU) per patient per year, for each treatment approach separately, is shown in Table 1 . Three studies [ 3 , 27 , 52 ] only provided data on costs either before or after the intervention; therefore no cost-savings comparison (i.e. cost- off set calculation) could be made. In addition, we did not attempt to provide the overall cost savings of TAU. Only two studies included in this review reported on pre- and post- cost data associated with the provision of TAU [ 25 , 26 ]. However, these findings are not sufficient to offset to any extent the cost-savings of the provision of TAU in the treatment of patients with BPD. Therefore TAU-related cost-data was not reported.

All studies comparing cost pre- and post intervention (15 evaluations; total sample size n = 787) reported a cost saving found to be in the range from USD $4 to $56,024 per person/per year. The pre-post analysis resulted in a mean weighted cost-saving of USD $5,840.92 [SD = $10,816.56; SE = $1,479.57] per patient per year (see Total Table 1 Cost Offset PT). Only the study by Palmer and colleagues (2006) [ 26 ] reported an increase of $99 in costs per patient/per year, post- intervention.

Because single group pre-post tests have recognised limitations, we have run additional analysis without the non-controlled studies to overcome bias of our results. Seven non- controlled studies were identified as so [ 28 – 31 , 34 , 36 , 54 ] with a total sample size of n = 192. For these non-controlled studies, the pre-post analysis resulted in a mean weighted cost-saving of USD $14,682.56 [SD = $ 1,286.07; SE = $17,820.38] per patient per year.

Subsequenly, re-running our primary pre-post analysis of the costs of psychotherapeutic interventions (cost offset PT) excluding the non-controlled studies yielded similar outcomes. Without the inclusion of these seven non-controlled studies the overall weighted cost offset of PT was: USD $2,987.82 [SD = $4390.31; SE = $180.01] per patient per year across these studies.

The 15 evaluations (total sample size n = 1,415) comparing the healthcare costs of psychotherapeutic interventions for BPD to TAU related costs, mostly reported cost-savings. See Table 1 Cost Offset PT vs. TAU. Compared to treatment as usual, the additional weighted mean cost-savings of implementing evidence-based psychotherapy was USD $1,551.37 per person per year [SD = $6,574.17; SE = $174.77]. These cost-savings ranged from $83 to $29,392 per person/per year. Five evaluations [ 41 , 42 , 45 , 47 , 50 ] did not find a cost-saving and reported an increase in costs when comparing healthcare costs of psychotherapeutic interventions for BPD to TAU related costs (increase in costs ranged from $61 to $10,772). See Table 1 .

Secondary analysis.

We furthermore were interested in whether there were detectable differences in the cost-benefit of different types of psychological therapy. We were able to find six studies of DBT and seven non-DBT alternative approaches. There are not enough studies to directly compare two alternative approaches to date, we therefore pooled the seven non-DBT studies. DBT was compared to TAU in six evaluations [ 33 , 40 – 43 , 45 ]. The (weighted) mean cost-offset derived from the provision of DBT compared to TAU was USD $78.43 [SD = $3,1412.83; SE = $184.01] per patient per year across these studies (total sample size n = 344). Seven non-DBT evaluations focused on the post- healthcare costs related to the provision of other psychological approaches, namely: MBT [ 25 , 47 ]; CI [ 48 ]; CP [ 49 ]; CBT-PD [ 51 ]; CBT-TAU [ 26 ] and MACT [ 50 ]. Compared to TAU, a (weighted) mean increase in costs of USD -$1,150.03 [SD = $5,482.33; SE = $170.91] per patient per year was demonstrated across the studies (total sample size n = 1029).

The weighted mean difference in cost savings vs. TAU derived after the provision of DBT vs. non-DBT studies was: $1,228.43 (SE = $251.13; 95% confidence interval $735.63 to $1,721.30). This resulted in a significant weigthed mean difference ( t (953.33.) = 4.892, p = .000). Although the difference per patient is significant, it is important to recognise that treatments themselves begin from different cost bases–and DBT standard includes weekly 2.5 hours of group, one hour of individual and ancillary care and phone coaching–which is generally more intensive that the comparative treatments which are typically up to two hours individual per week. In addition, crisis care, because it diverts from hospitalisation, has a large cost-offset. Although there are many potential interpretations for this significant finding, it could suggest that there is a greater cost-benefit of non-DBT vs DBT approaches compared to TAU. However, caution should be used when interpreting these results.

Summary of main findings

This paper reviews data published in peer review journals on the economic evaluations of evidence-based treatments for BPD, to inform recommendations for current mental health care funding policy. The results indicate that providing evidence-based treatments for BPD is cost-effective and results in cost savings. Our 2015 US dollar cost-offset calculations indicated that each individual in BPD treatment showed a reduction of USD $2,988 per patient per year in total healthcare costs in the year following BPD treatment as compared to the year prior to BPD. Perhaps more importantly, this review demonstrated that compared to treatment as usual, that the provision of psychotherapy resulted in an additional cost-saving of USD $1,551 per patient per year.

Up to December 2015, thirty economic evaluations across 29 studies met the inclusion criteria of this review and provided valuable cost-data for calculating a cost-savings for the psychotherapies evaluated. Of the studies included in this review, fifteen evaluations examined the economic benefits of a psychotherapeutic intervention using pre- and/or post-measures, but without a control group. Another fifteen evaluations provided data on the use of mental health services and related costs of psychotherapeutic intervention compared to treatment as usual (TAU) or client centred therapy (CCT) as a control group. The costs data presented in three evaluations [ 3 , 27 , 52 ] could not be used to calculate cost-savings. However, because our aim was to include all original studies reporting cost or cost-effectiveness data of BPD and to reflect the financial burden associated with BPD, these studies were included in the review and used to maximise the available information about the economic value of the provision of psychological treatment of BPD. Examining the studies included, DBT was the most evaluated treatment for BPD (15 evaluations). Other psychotherapeutic interventions examined by the studies included in this review were CM (2 evaluations), MBT (2 evaluations), SFT (2 evaluations) and one each for CBT, CBT-PD CI, CP, LT-P, MACT, OPC, SCP, SDC and TFP, representing a variety of intervention approaches (see Table 2 for description of treatments). The mean cost-savings derived from the provision DBT compared to TAU, did differ significantly from the additional cost-savings as a result from other forms of psychotherapy (6 DBT vs. 7 non-DBT comparisons). These findings could suggest that there is a greater cost-benefit of non-DBT vs DBT approaches compared to TAU. However, this finding is exploratory in nature and should not be used to draw firm conclusion about the cost savings of DBT or other forms of psychotherapy as there are variations across studies in approach, setting, and context. Future studies should compare costs of alternative treatments within the same project under similar conditions.

Our results suggest that psychotherapy for borderline personality disorders, independent of the type of treatment, can lead to cost-savings. Comparison of the cost savings of DBT versus other forms of psychotherapies did not lead to a significant difference in cost reduction; strengthening the status of the use of any form of well evaluated psychological therapy as the main treatment of borderline personality disorders [ 9 , 55 ]. Our findings furthermore provide evidence to support the assertion that offering effective therapy in BPD generates cost-saving advantages in terms of both direct and indirect healthcare costs.

Strengths and limitations

Other reviews that have looked at economic evaluations [ 17 , 18 , 56 ], similarly found evidence for cost-effective treatments for patients with BPD. To the best of our knowledge, there is no recent study that estimated annual net cost-savings of providing treatment in patients with BPD in a systematic way as conducted here. Brettschneider and colleagues (2014) [ 18 ] older study reviewed 15 economic evaluations of psychological therapies for BPD, concluding that the economic evidence is not sufficient to draw conclusions and current evidence should be interpreted with caution due to methodological shortcomings. Our review updated this previous work and now includes 30 evaluations, providing firmer evidence for cost benefit.

Our estimates for cost benefit should be interpreted with caution as there was significant variation between studies. However, as most studies included a control or active treatment comparison, our estimates do reflect the entire evidence we were able to systematically review.

The key strength of our review is that it has used full resource-use data that have been reported by researchers who have either conducted an economic evaluation of clinical trials or reported healthcare costs associated with the provision of treatment for BPD. Moreover, the focus of our review was not limited to specific psychological therapies and/or economic evaluations performed alongside clinical effectiveness randomized controlled trials. Hence, in this review, adjustment of the cost-data to account for variation in the cost-methods used in the different economic evaluations allowed comparison between treatments and enhanced generalizability of the study results.

However, this review was restricted to a certain extent by some limitations. We found that among the economic evaluation studies there was a great heterogeneity in the definition of costs, as have previous reviews [ 18 ]. Although we made a concerted effort to identify cost-categories, in some cases classifying the reported healthcare costs as either direct or indirect costs was not clearly and based upon common-sense reasoning. Furthermore, the studies in this review reported data from different countries, which could possible affect the generalisability of study findings. There are differences in economic circumstances and in health systems across various countries resulting in corresponding differences in health outcomes and their costs. For example, the health service and societal perspective in the Netherlands is different compared to the UK and Australian system. In the Australian, there is universal coverage for health care services; with the federal government paying a large part of the cost of health services, whereas Health insurance in the Netherlands is mandatory and everyone has to take out their own basic healthcare insurance [ 57 ]. Although we have made a considerable effort by calculating costs using purchasing power parity PPP, this should be taken into consideration when interpreting the results.

Despite the demonstrated overall cost savings across the majority of the 30 evaluations included, the cost-offset calculations for six economic evaluations included in this review did not result in cost-savings [ 26 , 41 , 42 , 45 , 47 , 50 ]. In four of these [ 41 , 42 , 47 , 50 ] it may be that the extra costs associated with the alternative therapy compared to the costs of the provision of TAU may be due to the method of cost modelling in the studies [ 17 , 18 ]. In some cases, resource use was estimated by authors by regression models [ 17 ], and for these economic evaluations costs may have been overestimated. Furthermore, the healthcare costs incurred before and after the provision of CBT [ 26 ] and after the provision of DBT [ 45 ] compared to TAU, did not result in cost-savings. However, the cost differences in both studies were small and did not approach conventional levels of statistical significance. In addition, it must be noted that the intervention (CBT) examined by Palmer and colleagues (2006) [ 26 ] did result in an overall saving of USD $131 (per patient per year) in healthcare costs when compared with the provision of TAU.

This systematic review aimed to provide an overview of the cost savings associated with the provision of psychotherapeutic interventions for the treatment of BPD and was not intended as a statement of the clinical effectiveness of these treatments. Data was extracted based on cost alone, without further consideration of clinical effectiveness of these treatments and is outside the scope of this review. However, the evidence on economic outcomesalone is one factor informing clinical decision making in health care. Future studies may investigate the interaction between cost and clinical improvement, although this will rely upon an even larger pool of studies of sufficient methodological rigour to allow such an evaluation.

Although we were able to identify thirty economic evaluations in total, a paucity of material is apparent in this field, especially if we compare the modest number of economic findings with the larger number of clinical evaluations. We have found some papers aiming to evaluate the clinical and cost-effectiveness of evidence-based BPD treatment [ 58 – 60 ]. However, since these papers only outline the background and methods of randomised controlled trials and are still ongoing, these studies may inform future reviews.

Conclusions

Borderline Personality Disorder (BPD) is considered one of the most expensive mental disorders in terms of direct and indirect healthcare costs. Evidence-based treatment approaches for BPD are available and demonstrate potential cost-saving when implemented effectively. This review is the first to calculate the cost-savings associated with the provision of evidence-based treatment for patients with BPD using all published data available for review. Evidence for cost-effectiveness of psychological treatment was supported by our findings. Based on the findings of this systematic review, the provision of evidence-based psychological treatment to patients with BPD results in a reduction in costs associated with both the use of mental health services and related community costs, that significantly exceeds the cost of no treatment or treatment as usual.

Supporting information

S1 table. prisma checklist..

https://doi.org/10.1371/journal.pone.0171592.s001

Author Contributions

• Conceptualization: DM MEB BFSG.
• Data curation: DM.
• Formal analysis: DM.
• Funding acquisition: DM AM MEB BFSG.
• Investigation: DM AM MEB BFSG.
• Methodology: DM BFSG.
• Project administration: DM.
• Resources: DM AM MEB BFSG.
• Software: DM AM MEB BFSG.
• Supervision: MEB BFSG.
• Validation: DM AM MEB BFSG.
• Visualization: DM AM MEB BFSG.
• Writing – original draft: DM AM.
• Writing – review & editing: DM AM MEB BFSG.
• View Article
• PubMed/NCBI
• Google Scholar
• 13. Dimeff LA, & Koerner K. (2007) Dialectical Behaviour Therapy in Clinical Practice: Applications across Disorders and Settings.: New York, NY: Guildford.
• 24. Organisation for Economic Co-operation and Development (OECD) (2013) OECD Health Data: Economic references.
• 33. Heard HL (2000) Cost-effectiveness of dialectial behaviour therapy in the treatment of borderline personality disorder. (Doctor of Philosophy), University of Washington (9975993).

• Medical Books

Sorry, there was a problem.

Download the free Kindle app and start reading Kindle books instantly on your smartphone, tablet, or computer - no Kindle device required .

Read instantly on your browser with Kindle for Web.

Using your mobile phone camera - scan the code below and download the Kindle app.

Image Unavailable

• To view this video download Flash Player

The Borderline: Current Empirical Research (Progress in Psychiatry)

• ISBN-10 052133277X
• ISBN-13 978-0521332774
• Publisher Cambridge University Press
• Publication date October 9, 1986
• Language English
• Print length 118 pages
• See all details

Product details

• Publisher ‏ : ‎ Cambridge University Press (October 9, 1986)
• Language ‏ : ‎ English
• Hardcover ‏ : ‎ 118 pages
• ISBN-10 ‏ : ‎ 052133277X
• ISBN-13 ‏ : ‎ 978-0521332774
• Item Weight ‏ : ‎ 10.9 ounces

Customer reviews

• 5 star 4 star 3 star 2 star 1 star 5 star 0% 0% 0% 0% 0% 0%
• 5 star 4 star 3 star 2 star 1 star 4 star 0% 0% 0% 0% 0% 0%
• 5 star 4 star 3 star 2 star 1 star 3 star 0% 0% 0% 0% 0% 0%
• 5 star 4 star 3 star 2 star 1 star 2 star 0% 0% 0% 0% 0% 0%
• 5 star 4 star 3 star 2 star 1 star 1 star 0% 0% 0% 0% 0% 0%

Customer Reviews, including Product Star Ratings help customers to learn more about the product and decide whether it is the right product for them.

To calculate the overall star rating and percentage breakdown by star, we don’t use a simple average. Instead, our system considers things like how recent a review is and if the reviewer bought the item on Amazon. It also analyzed reviews to verify trustworthiness.

No customer reviews

• About Amazon
• Investor Relations
• Amazon Devices
• Amazon Science
• Sell products on Amazon
• Sell on Amazon Business
• Sell apps on Amazon
• Become an Affiliate
• Advertise Your Products
• Self-Publish with Us
• Host an Amazon Hub
• › See More Make Money with Us
• Amazon Business Card
• Shop with Points
• Reload Your Balance
• Amazon Currency Converter
• Amazon and COVID-19
• Your Account
• Your Orders
• Shipping Rates & Policies
• Returns & Replacements
• Manage Your Content and Devices

• Conditions of Use
• Privacy Notice
• Consumer Health Data Privacy Disclosure
• Your Ads Privacy Choices

New Open Access Journal from APS and Sage Expands Publishing Opportunity for Psychological Scientists

Applications now open for role of inaugural editor.

• Advances in Psychological Science Open
• APS Journals
• Open Practices

August 13, 2024 — The Association for Psychological Science (APS) and Sage announce the launch of Advances in Psychological Science Open , a fully open access journal that will publish high-quality empirical, technical, theoretical, and review articles, across the full range of areas and topics in psychological science. The journal will accept submissions in a variety of formats, including long-form articles and short reports, and APS is encouraging scientists to submit integrative and interdisciplinary research articles. 

“APS is always working to identify new ways to catalyze advances in psychological science,” said APS CEO Robert Gropp. “We are excited to announce that we are launching Advances in Psychological Science Open to provide a publication option for scientists who want a fully open access journal in which to share their research findings.”  

APS has launched a search for the inaugural editor of the journal with the goal of having an editor appointed to begin work in January 2025, with first acceptance of manuscripts in mid- 2025. Nominations, including self-nominations, for Founding Editor are welcomed.  Nominations of members of underrepresented groups in psychology are especially encouraged. For more information on how to submit a nomination, please refer to the open call here .  

“Sage has been committed to opening pathways for social and behavioral scientists since our founding nearly 60 years ago,” said Bob Howard, executive vice president, research at Sage. “We’re thrilled to build on our long-standing partnership with APS to launch a journal publishing high-quality, impactful research that will help shape the future of psychological science.” 

The new title is the seventh journal that APS will publish in partnership with Sage. Advances in Psychological Science Open adds to the rich ecosystem of APS publications that collectively meet the needs of the psychological science community. APS members will receive a significant discount on the open access fee for this new journal, adding to the suite of benefits that members already receive. 

For more information, please contact Scott Sleek at [email protected] .

About Sage  

Sage  is a global academic publisher of books, journals, and library resources with a growing range of technologies to enable discovery, access, and engagement. Believing that research and education are critical in shaping society, 24-year-old Sara Miller McCune founded Sage in 1965. Today, we are controlled by a group of trustees charged with maintaining our independence and mission indefinitely.  

See the related announcement: Advances in Psychological Science Open  Coming Soon

APS regularly opens certain online articles for discussion on our website. Effective February 2021, you must be a logged-in APS member to post comments. By posting a comment, you agree to our Community Guidelines and the display of your profile information, including your name and affiliation. Any opinions, findings, conclusions, or recommendations present in article comments are those of the writers and do not necessarily reflect the views of APS or the article’s author. For more information, please see our Community Guidelines .

Please login with your APS account to comment.

Incoming PSPI Editor Colleen Seifert Outlines Her Goals for the Journal

Colleen Seifert aims to expand the range of topics covered in the APS publication.

Research Briefs

Recent highlights from APS journals articles.

Recent highlights from APS journals articles on the link between self-esteem and eating disorders, how to be liked in first encounters, the effects of stress on rigid learning, and much more.

Privacy Overview

Health & Environmental Research Online (HERO)

• Learn about HERO
• Search HERO
• Projects in HERO
• Risk Assessment
• Transparency & Integrity

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Indian J Psychol Med
• v.39(2); Mar-Apr 2017

Empirical Reality of Dialectical Behavioral Therapy in Borderline Personality

M. s. reddy.

Asha Bipolar Clinic, Asha Hospital, Hyderabad, Telangana, India E-mail: moc.liamg@yddersmrd

M. Starlin Vijay

Introduction.

Borderline personality disorder (also known as emotionally unstable personality disorder) is a complex mental condition consisting of affective, cognitive-perceptual, anxiety, and stress-coping domains. Mental health professionals understand the problems in delivering required care and to a varying degree feel the pessimism with regard to its treatment. In this background, past two decades have witnessed various therapeutic attempts which over time have instilled much-needed optimism with the condition's treatment and prognosis.

One such therapeutic attempt is dialectical behavioral therapy (henceforth, DBT), which was introduced in the early 1990s by Linehan et al .[ 1 ] DBT has ever since garnered support of guidelines of many countries and has also been included in the American Managed Care system. While there are also other evidence-based therapies such as mentalization-based therapy, transference-focused therapy, schema-focused therapy, and dynamic deconstructive psychotherapy, none have found such popular appeal and encouragement by the professional guilds of psychiatry and psychology as did DBT. Hence, in this article, we will review the empirical evidence about DBT and examine whether its reputation equals its scientific basis.

WHAT IS DIALECTICAL BEHAVIORAL THERAPY?

DBT is an integrated psychotherapy comprising change techniques based on behavioral therapy on the one hand and acceptance techniques based on Zen Buddhism on the other hand. Both these techniques are used in a dialectic way to help patients handle and navigate through their complex affective and cognitive states.[ 2 ] The therapist has to move constantly between these two approaches, and to traverse this dialectic terrain, a personal practice of Zen Mindfulness and regular skills training is needed. Unlike cognitive behavioral therapy (CBT) which has a near-complete integration of its component cognitive and behavioral theories, DBT lacks such a good fit between the behavioral and dialectic theories because of their radically different philosophical presuppositions; for example – the behavioral stance would like to change a “problem behavior” to an “adaptive behavior” while the other stance might suggest a nonjudgmental acceptance of the same behavior and thereby leading to a stalemate. Acceptance techniques are taught through the practice of various components of mindfulness. Dialectical theory presupposes that there is no one truth and therefore works toward an integration of multiple perspectives within a particular complex human problem.[ 3 ]

DBT model has four stages in the treatment of borderline personality. Each Stage (I–IV) has certain specific goals,[ 2 ] such as:

• Reducing suicidal, therapy-interfering, and quality-of-life-interfering behaviors, and improving behavioral skills
• Treating issues related with past trauma. For example, exposure techniques for posttraumatic stress disorder
• Development of self-esteem, reclaiming ordinary happiness, and improving day-to-day behavioral skills
• Development of capacity for optimum experiencing and finding a higher purpose.

Stage I may take anywhere from 3 to 12 months (usually, it takes at least a year[ 4 ]) and then the therapy moves onto Stages II–IV. The patient has to attend one individual therapy session and a group skills training session in a week and also has to do the regular homework such as diary records. There are two important components during the therapy which are central–regular group supervision for individual and group therapists, and 24/7 patient access to the therapist by telephone to handle emergencies.

EMPIRICAL EVIDENCE

DBT is the only available therapy which led to creation of a large research data in patients with borderline personality.[ 5 ] There are multiple studies done by Linehan over the last three decades. Independent researchers like Verheul et al .,[ 6 ] Clarkin et al .,[ 7 ] and McMain et al .[ 8 ] have also investigated using DBT.

Evidence in parasuicide (this term used in the studies by Linehan includes suicide attempts and nonsuicidal self-injury)

Recent Cochrane review[ 9 ] reports that DBT when compared to treatment-as-usual (TAU) and community-treatment-by-experts (CTBE) shows no difference in the proportion of patients repeating self-harm or other outcomes such as suicidal ideation and depression, though frequency of repetition of self-harm reduces. Only one study[ 10 ] with weak methodology and small sample ( n = 24) which compared DBT with client-centered therapy (which is not studied or used in borderline personality management) showed difference in impulsivity, suicidality, parasuicidality, and depression. Another recent article which reviewed the impact of treatment intensity on suicidal acts and depression[ 11 ] concluded that more intensive therapies (i.e., more than 100 h per year) are not superior to less intensive therapies (i.e., <100 h in a year). They have found that CBT for personality disorder with 30 h in a year is not inferior to DBT with 84 h or more in a year.

Evidence in other domains

There is a lack of evidence favoring DBT on core personality features such as interpersonal instability, chronic emptiness, and boredom and identity disturbance or associated symptoms such as depression, suicidal ideation, survival and coping beliefs, overall life satisfaction, work performance, and anxious rumination.[ 4 , 6 , 12 ] DBT was no different in reducing depression than any comparator, be it TAU, CTBE, or general psychiatric management (GPM). All therapies showed a reduction in depression over time.[ 11 ]

Except for three studies, which are discussed below, all the other studies have weak methodology and small sample sizes.[ 12 , 13 ] In the majority of studies, DBT was compared to TAU, which has its own problems such as:

• It is difficult to know what treatment is given in TAU and it may change over a period of the study
• TAU sessions are not documented or recorded and thereby adherence and competence ratings cannot be done
• Patients in TAU comparatively get very less hours of treatment than in active structured treatments.

According to Chambless and Hollon,[ 14 , 15 ] psychological therapies have to fulfill five criteria to be called as “Empirically Supported Treatments.” These are superiority or equivalence to established/another treatment in studies with good methodological rigor ( n = at least 30 per group); at least nine single-case design experiments showing efficacy; availability of manuals; clear specification of patient characteristics; and effects should be demonstrated by at least two independent teams. Ost argues in a review of study methodologies[ 13 ] that DBT randomized controlled trials (RCTs) have significantly less stringent research methodology than CBT studies and that they do not fulfill criteria for empirically supported treatments. A 2009 RCT with 180 samples and good methodology[ 8 ] suggests equivalence between DBT and GPM, thereby making DBT “probable-empirically supported treatment”

We will discuss only three studies which are rigorous and show us the empirical reality of DBT. Strengths of these studies being adequate power, investigators with a balance of theoretical allegiances (two studies[ 7 , 8 ]), and investigated by an independent group (two studies[ 7 , 8 ]) [ Table 1 ].

Comparison of three well-designed and adequately powered studies investigating the comparative effectiveness of dialectical behavior therapy versus other valid treatments

These studies show no statistically significant between-group differences for pathology-related outcomes, though there are marginal or very small effects in terms of suicidality, anger, depression, etc.[ 5 ] There is a significant difference in the treatment hours between DBT and comparators. Follow-up of Linehan's initial studies shows high dropout rate and loss of efficacy over time.[ 16 , 21 ] Researchers recommend an adequately powered head-to-head comparison using a rigorous methodology with a structured psychotherapy with good evidence base in borderline personality management such as transference-focused, schema-focused, or mentalization-based therapy instead of comparison with waiting list or TAU groups.[ 13 ] Reviewers have also observed that common team approach,[ 22 ] easy access, and intensive relationship focus in therapy and supervision of therapists by peers[ 4 ] are common features between psychodynamic therapies and DBT which have shown positive results in borderline personality management.

LIMITATIONS OF DIALECTICAL BEHAVIORAL THERAPY RESEARCH AND CLINICAL IMPLICATIONS

• DBT has a demanding model of therapy. A patient has to attend two separate sessions which include 1 h of individual therapy and 2 h of group skills training every week along with regular homework assignments over at least 1 year of treatment. Therapist has to be available 24/7 for providing emergency behavioral coaching, however rules can be laid down in this to protect therapist from burnout. It can be very costly because of multiple sessions and involvement of highly qualified therapists if it is not delivered through the public health-care system
• All the DBT studies were of 1 year duration, however as pointed out earlier, Stage I itself many a times takes up to 1 year. Hence, we cannot suppose that studies have tested the whole therapy. Instead, they have tested the usefulness of just one, albeit an important stage of therapy. This might be the reason for the lack of evidence in domains of pathology other than parasuicide
• DBT needs therapists who are highly qualified (many studies by Linehan had doctoral-level professionals) and who have to be under regular supervision by attending 2-h consultation team meeting every week for learning skills and supervision. This presents problems with dissemination and resource usage, especially in nonacademic centers, community, and resource-poor settings like India
• Many reviewers including APA practice guidelines[ 13 , 23 ] suggest the need for studies by independent investigators. This is important because it ensures generalizability of findings and is an important criterion (criteria V) for empirically supported treatments
• DBT has consistent evidence exclusively for reduction in frequency of suicide reattempts and also has evidence in those with eating disorders and substance use disorders; based on this, some[ 4 , 6 ] have suggested that either we have to change from DBT, after the reduction of suicidality, to another therapy which is more targeted at core features of borderline personality; or we have to assume that DBT is a specific therapy for patients (mainly female) with life-threatening impulse control disorders rather than borderline personality disorder per se
• DBT when compared to other structured therapies does not fare well with regard to core features of borderline personality disorder except showing equivalence with regard to improvement in suicide attempts
• Although a Cochrane review concludes that psychotherapies, in general, are effective in the management of borderline personality,[ 5 ] it is not altogether clear as to the role of medication in the management and there are no rigorous or adequately powered studies comparing medication and psychotherapy. This is important to consider because selective serotonin reuptake inhibitors, antipsychotic agents, and mood stabilizers are commonly used in clinical settings for the borderline personality management.

DBT has to be appreciated as its research has instilled the much-needed optimism into the management of borderline personality disorder management in the early 1990s. DBT has specific utility in addressing suicide attempts in borderline personality without being generally effective in the overall personality management. However, the review of its research and discussion of its limitations show that the empirical reality is very different from its reputation and popular exaggeration. There is a need for future studies to design adequately powered RCTs comparing it to other structured therapies.