Our follow-up of 12-year-olds with borderline symptoms to age 18 revealed three main findings. First, at a time in life when young people take a leap toward greater social, economic, and personal maturity, study members with a history of borderline symptoms were held back by psychosocial difficulties. Difficulties were evident in numerous areas (personality; psychopathology; vocational, health, and social functioning; and experiences of victimization); observed by multiple informants; and assessed through multiple methods, including official records. Differences in outcomes were striking: young people with the highest borderline symptom scores at age 12 were nearly three times more likely to engage in suicidal and self-harming behavior, to find themselves without training or job opportunities, to have a criminal record, and to have experienced victimization compared with their cohort peers with lower symptom scores. These findings show that adolescent borderline symptoms observed as early as at age 12 forecast meaningful individual differences in young people’s lives.
Second, although many 12-year-olds experienced behavioral and emotional problems alongside their borderline symptoms, borderline symptoms added incremental value to predicting most poor outcomes over and above these other problems, indicating that the later-life impairments associated with adolescent borderline pathology are insufficiently described by these problems. Notably, behavioral and emotional problems of 12-year-olds accounted for associations with nearly all of their psychiatric diagnoses at age 18, but did not account for associations with many other adverse outcomes. This finding illustrates that psychiatric diagnoses do not capture the full scope of life challenges associated with adolescent borderline symptoms and shows that it is important to look beyond psychiatric status when testing the predictive validity of adolescent borderline symptoms.
Third, borderline symptoms of 12-year-olds were under considerable genetic influence, and genetically identical twins of children with elevated borderline symptoms were at increased risk for poor outcomes even if they did not have equally elevated borderline symptoms themselves. This finding raises three issues. First, it raises the question of why twins with the same genetic susceptibility do not share similar borderline symptoms. Our previous study pointed to twins’ unique environmental experiences as a possible explanation: we reported that twins in the same families experienced different levels of harsh parental treatment and that adolescents’ genetic vulnerability interacted with harsh parental treatment in the etiology of borderline symptoms.12x12Belsky, D.W., Caspi, A., Arseneault, L. et al. Etiological features of borderline personality related characteristics in a birth cohort of 12-year-old children. Dev Psychopathol. 2012;
Crossref | PubMed | Scopus (81) | Google ScholarSee all References This finding is consistent with diathesis-stress models of borderline personality, which propose that it is the interaction between children’s genetically influenced, early emerging temperamental difficulties and an invalidating, abusive, and ineffective caregiving environment that increases risk for borderline problems (and other poor outcomes) in a transactional process across development.12x12Belsky, D.W., Caspi, A., Arseneault, L. et al. Etiological features of borderline personality related characteristics in a birth cohort of 12-year-old children. Dev Psychopathol. 2012;
Crossref | PubMed | Scopus (81) | Google ScholarSee all References,32x32Crowell, S.E., Beauchaine, T.P., and Linehan, M.M. A biosocial developmental model of borderline personality: elaborating and extending linehan’s theory. Psychol Bull. 2009;
Crossref | PubMed | Scopus (497) | Google ScholarSee all References Second, our findings indicate that adolescent borderline symptoms reflect broader genetic risk for poor outcomes, rather than being the cause of these outcomes. This finding does not undermine the prognostic significance of adolescent borderline symptoms. However, it suggests that adolescents remain at risk for adverse psychosocial outcomes even after symptom reduction.10x10Winograd, G., Cohen, P., and Chen, H. Adolescent borderline symptoms in the community: prognosis for functioning over 20 years. J Child Psychol Psychiatry. 2008;
Crossref | PubMed | Scopus (117) | Google ScholarSee all References Third, if borderline symptoms are not the cause of poor outcomes but are on the pathway from genetic risk to poor outcomes, more work is needed to understand how genetic risk influences both borderline symptoms and poor outcomes. A hypothesis consistent with diathesis-stress models of borderline personality is that genetic risk begins to manifest early in life, as a difficult temperamental profile characterized by high negative affect, poor impulse control, and high emotional sensitivity. A child’s difficult temperament subsequently increases risk for borderline pathology as well as for other poor outcomes, particularly when it is met by an invalidating caregiving environment.32x32Crowell, S.E., Beauchaine, T.P., and Linehan, M.M. A biosocial developmental model of borderline personality: elaborating and extending linehan’s theory. Psychol Bull. 2009;
Crossref | PubMed | Scopus (497) | Google ScholarSee all References
Our work expands on previous literature in three ways. First, although several studies have investigated the clinical and psychosocial outcomes of borderline personality disorder, a recent systematic review concluded that many of these studies have limitations, such as sampling bias, high rates of attrition, and a narrow range of psychosocial outcomes.11x11Winsper, C., Marwaha, S., Lereya, S.T., Thompson, A., Eyden, J., and Singh, S.P. Clinical and psychosocial outcomes of borderline personality disorder in childhood and adolescence: a systematic review. Psychol Med. 2015;
Crossref | PubMed | Scopus (25) | Google ScholarSee all References Our study overcomes some of these limitations because our cohort is nationally representative, follow-up of participants has occurred with extremely low attrition, and we report associations with a wide range of clinical and psychosocial outcomes. Second, there are very few studies testing associations between borderline pathology and exposure to victimization, particularly in adolescence. Our study extends the literature by showing that borderline symptoms in 12-year-olds predict exposure to different types of victimization, both inside and outside the home, during adolescence. Third, in addition to reporting that adolescents’ borderline pathology predicts poor outcomes, we find that these associations do not persist after accounting for familial influences shared between identical twins growing up in the same family. Although several studies have tested outcomes of adolescent borderline symptoms using a twin design,33x33Bornovalova, M.A., Hicks, B.M., Iacono, W.G., and McGue, M. Longitudinal twin study of borderline personality disorder traits and substance use in adolescence: developmental change, reciprocal effects, and genetic and environmental influences. Personal Disord Theory Res Treat. 2013;
Crossref | Scopus (34) | Google ScholarSee all References to our knowledge our study is the first to apply this approach to a wide range of clinical and psychosocial outcomes.
Our findings should be interpreted in light of some limitations. First, we did not make a formal diagnosis of borderline personality disorder. Without a replication in adolescents with a diagnosis of borderline personality disorder, we cannot be sure that our findings generalize to this population. However, our measure captures core diagnostic features of borderline personality disorder (affective instability, cognitive disturbance, impulsivity, and interpersonal dysfunction), and our previous study showed that the etiological factors, comorbidity, sex differences, and heritability of our measure of borderline symptoms are comparable to results from studies of borderline personality disorder in community samples.10x10Winograd, G., Cohen, P., and Chen, H. Adolescent borderline symptoms in the community: prognosis for functioning over 20 years. J Child Psychol Psychiatry. 2008;
Crossref | PubMed | Scopus (117) | Google ScholarSee all References,12x12Belsky, D.W., Caspi, A., Arseneault, L. et al. Etiological features of borderline personality related characteristics in a birth cohort of 12-year-old children. Dev Psychopathol. 2012;
Crossref | PubMed | Scopus (81) | Google ScholarSee all References,34x34Lenzenweger, M.F., Lane, M.C., Loranger, A.W., and Kessler, R.C. DSM-IV personality disorders in the National Comorbidity Survey Replication. Biol Psychiatry. 2007;
Abstract | Full Text | Full Text PDF | PubMed | Scopus (679) | Google ScholarSee all References,35x35Torgersen, S., Kringlen, E., and Cramer, V. The prevalence of personality disorders in a community sample. Arch Gen Psychiatry. 2001;
Crossref | PubMed | Scopus (824) | Google ScholarSee all References Second, a general weakness of discordant twin analyses is their higher likelihood of false-negative findings because the limited variation within twin pairs magnifies the impact of measurement error and reduces the precision of estimates.36x36Boardman, J.D. and Fletcher, J.M. To cause or not to cause? That is the question, but identical twins might not have all of the answers. Soc Sci Med. 2015;
Crossref | PubMed | Scopus (26) | Google ScholarSee all References Third, our study does not contain a measure of borderline symptoms at age 18, so we were unable to test the continuity of borderline symptoms. However, our findings show that 18-year-olds with a history of borderline symptoms show a personality profile characterized by emotional and interpersonal lability that is typical of adults with borderline personality disorder.37x37Saulsman, L.M. and Page, A.C. The five-factor model and personality disorder empirical literature: a meta-analytic review. Clin Psychol Rev. 2004;
Crossref | PubMed | Scopus (585) | Google ScholarSee all References Fourth, although a considerable portion of 12-year-olds with borderline symptoms went on to experience poor outcomes, there were also adolescents who bucked this trend and fared well despite their symptom history. Follow-on work is needed to investigate factors that predict variability in poor outcomes associated with borderline symptoms. Fifth, our study members are still young, so it is unclear how persistent their psychosocial difficulties will be. However, many of the outcomes we measured—low educational qualifications, cigarette smoking, personality dysfunction, having a criminal record, risky sexual behavior—are still meaningful because they represent barriers to leading a prosperous and healthy adult life. Moreover, previous work testing associations with some of the same outcomes we report, such as attainment and social support, has shown that adolescent borderline symptoms predict these outcomes up to age 33 years.10x10Winograd, G., Cohen, P., and Chen, H. Adolescent borderline symptoms in the community: prognosis for functioning over 20 years. J Child Psychol Psychiatry. 2008;
Crossref | PubMed | Scopus (117) | Google ScholarSee all References
Our findings have implications for health professionals working with adolescents who display borderline symptoms. First, our findings support the assessment of adolescents’ borderline symptoms in addition to other emotional and behavioral disorders if borderline pathology is suspected. Some clinicians are thought to prefer assessing only emotional and behavioral disorders in adolescents presenting with borderline symptoms, perhaps to avoid a stigmatizing diagnosis of personality disorder.6x6Miller, A.L., Muehlenkamp, J.J., and Jacobson, C.M. Fact or fiction: diagnosing borderline personality disorder in adolescents. Clin Psychol Rev. 2008;
Crossref | PubMed | Scopus (199) | Google ScholarSee all References,38x38Shiner, R.L. and Allen, T.A. Assessing personality disorders in adolescents: seven guiding principles. Clin Psychol Sci Pract. 2013;
Crossref | Scopus (20) | Google ScholarSee all References However, our findings and findings of others show that adolescents’ borderline symptoms provide independent prognostic information. Second, our findings argue in favor of early access to treatment for adolescents with borderline symptoms and against a wait-and-see approach.39x39Chanen, A., Sharp, C., and Hoffman, P. Global Alliance for Prevention and Early Intervention for Borderline Personality Disorder. Prevention and early intervention for borderline personality disorder: a novel public health priority. World Psychiatry. 2017;
Crossref | PubMed | Google ScholarSee all References Psychological treatments for adult patients with borderline personality disorder have been adapted for use with adolescents and show promise for improving symptoms.7x7Sharp, C. and Fonagy, P. Practitioner review: borderline personality disorder in adolescence—recent conceptualization, intervention, and implications for clinical practice. J Child Psychol Psychiatry. 2015;
Crossref | PubMed | Scopus (47) | Google ScholarSee all References In addition to treatment for personality pathology, adolescents with borderline symptoms need access to support services that help reduce the risk for future poor functioning, such as educational support services. Third, our findings show that adolescents’ borderline symptoms signal a longer-term need for care. Even if symptoms decrease after treatment, adolescents remain at risk for adverse outcomes because symptoms partly reflect genetic risk for future difficulties. Adolescents should be monitored and supported accordingly, particularly during the transition to adulthood when they face discharge from child and adolescent mental health services. Fourth, our findings imply that young people with a history of borderline symptoms may turn up on the doorsteps of many services, including mental health care services, unemployment offices, sexual health centers, courts, emergency departments, and social services. The breadth of poor outcomes among these young people requires an integrated treatment approach that involves coordination across multiple social and support services.