Does BPD get better?

You looked it up, and what you found was the word "personality," which sounds permanent, next to a list of things you recognised, which was worse. Nobody handed you the follow-up studies. They are the most useful thing anyone with this diagnosis can read, and they have been published for over a decade.

The short answer: yes, and more reliably than almost anything else in psychiatry. Two large prospective studies followed people with borderline personality disorder for a decade or more. In the sixteen-year follow-up of 290 hospitalised patients, 99 percent had achieved a remission lasting at least two years and 78 percent one lasting eight. In the ten-year follow-up of 175 treatment-seeking outpatients, 85 percent remitted on a twelve-month definition and only 11 percent relapsed. What does not resolve on the same schedule is functioning: recovery — remission plus consistent work or school and at least one close relationship — reached 60 percent at two years and 40 percent at eight, and in the outpatient study global functioning moved from 53 to 57 over the decade. So the symptoms lift. The life does not rebuild itself while they are lifting, and that is what treatment has to be aimed at.¹ ²

The sixteen-year numbers, in full

The McLean Study of Adult Development enrolled 290 inpatients who met criteria on two separate instruments, aged 18 to 35, and re-interviewed them every two years. Remission meant no longer meeting criteria for BPD or another personality disorder for at least the stated period. Recovery meant that plus a Global Assessment of Functioning score of 61 or higher — which in practice required at least one emotionally sustaining close relationship and the ability to work or study consistently.¹

Read the last four rows together with the first four, because they carry the practical lesson: the longer a remission has lasted, the less likely it is to break. A two-year remission recurred 36 percent of the time; an eight-year one, 10 percent. Time in remission is itself protective.

The trajectory is also worth seeing as a curve rather than an endpoint. The proportion in a two-year remission ran 35, 55, 76, 88, 91, 95, 97, 99 percent at years 2, 4, 6, 8, 10, 12, 14 and 16. Recovery ran 14, 27, 36, 43, 47, 50, 56, 60.¹ Most of the symptom improvement happens early; most of the functional improvement happens late, and slowly.

The ten-year outpatient numbers

The Collaborative Longitudinal Personality Disorders Study followed 175 treatment-seeking patients across nineteen clinical sites, which makes it a better mirror of ordinary outpatient life than an inpatient cohort.²

• Cumulative remission over ten years: 91 percent using a two-month definition, 85 percent using a twelve-month one.
• Relapse: 11 percent on the twelve-month definition, 21 percent on the two-month one.
• Criteria met fell from a mean of 6.7 to 4.3 in the first year, then by about 0.29 criteria a year to 1.7 at ten years. Only 9 percent still met five or more.
• Global functioning rose from 53 to 57. Nobody in the BPD group had good functioning at baseline; 21 percent did at ten years, against 48 percent of the comparison group with other personality disorders and 61 percent of the group with major depression.
• Full-time employment rose from 19 percent to 36 percent. Married or cohabiting rose from 23 percent to 41 percent.

The authors' summary sentence: "The 10-year course of BPD is characterized by high rates of remission, low rates of relapse, and severe and persistent impairment in social functioning."²

Why the gap between remission and recovery matters more than either number

Symptom remission is what treatment measures. Recovery is what you were asking about.

The same split shows up in the treatment trials. In the two-year follow-up of a randomised trial comparing dialectical behaviour therapy with general psychiatric management, two-thirds of participants had achieved diagnostic remission — and at 36 months, 53 percent were neither employed nor in school and 39 percent were receiving psychiatric disability support.³ In the eight-year follow-up of mentalization-based treatment, the treated group was still better than usual care on suicidality, diagnosis, service use and medication, and the authors still wrote that "their general social function remains impaired."⁴

Two practical consequences.

One: pick a treatment that has a functional target in it. Ask, in the first session, what you and the clinician will be measuring in six months besides symptoms — hours worked or studied, a specific relationship, something that would show up on a calendar. The evidence-based treatments all have room for this; some programmes never get to it.

Two: do not wait for symptoms to clear before starting the rest of your life. The curve above says the functional half takes longer and moves slower, which is an argument for starting it earlier, not for postponing it until you feel ready.

What predicts a better course

The predictors that come out of these cohorts are unglamorous and mostly actionable.

• Absence of a substance use disorder was the strongest predictor of recovery in the 24-year analysis of the McLean cohort.⁵ If a substance problem is in the picture, it is not a side issue to be dealt with later; it is the single biggest lever in the data. Substance use and what to treat first →
• Time already spent in remission, as above.¹
• Fewer prior hospitalisations was associated with lower risk of both suicide and premature non-suicide death in the same cohort.⁶

The mortality question, honestly

People find the figure "about 10 percent of people with BPD die by suicide" and stop reading. Where it comes from matters: follow-back studies of patients discharged from long-stay specialist units decades ago, including a 27-year follow-up of 64 patients that reported 10.3 percent.⁷ The psychiatrist who co-authored that figure has since written that "lower rates (3%–6%) have been reported in prospectively followed cohorts."⁷

In the McLean cohort followed forward for 24 years, 5.9 percent died by suicide. A meta-analysis of prospective clinical cohorts put the range at 2 to 5 percent.⁶ ⁸

Two things that are less quoted and matter more. In those same 24 years, 14.0 percent died of causes that were not suicide — a larger number, and one that points at physical health, smoking, substance use and the medical care people with this diagnosis often do not get.⁶ And of those who died by suicide, 87.5 percent were not in recovery before death; of those who died of other causes, 88 percent were not.⁶ Recovery is not only about quality of life.

None of this makes the risk small. It makes it a risk that treatment, and time, measurably change.

Q&A

Q: How long does it take? A: In the outpatient study, the largest single drop in symptoms happened in the first year — from 6.7 criteria to 4.3 — and then continued slowly for a decade.² In the inpatient cohort, three-quarters had achieved a two-year remission by year six.¹ Meaningful change in months, substantial change in years.

Q: Does treatment actually change the curve, or does everyone improve anyway? A: Both cohorts were largely in treatment, so they cannot separate the two. What the randomised trials show is that structured psychotherapy beats usual care on symptom severity, on moderate-quality evidence, and that the named treatments do not differ much from one another.⁹ The honest statement is that improvement is the norm and that structured treatment improves on the usual pace.

Q: If it gets better on its own, why bother with therapy? A: Because of the ten years in the middle, the suicide risk concentrated in them, and the functional gap that does not close by itself. "It remits eventually" is not a treatment plan.

Q: I'm 45 and it hasn't remitted. Do these numbers apply to me? A: The cohorts enrolled people aged 18 to 35 and 18 to 45, so the studies say less about a course that begins or persists later. What they do say is that recurrence after a long remission is uncommon, and that the strongest recovery predictor — an untreated substance problem — is treatable at any age.⁵

Q: Do these studies apply to me if I've never been hospitalised? A: The McLean cohort were inpatients, which is the most severe end; the CLPS cohort were outpatients, which is closer to most people. Both found high remission. If anything, being less severely ill at the start is associated with a better course, not a worse one — but that is an inference, not a finding from these papers.

Q: What about people who never get diagnosed at all? A: Community surveys find many more people meeting criteria than clinics ever see, and their course has not been followed the way these cohorts were. The follow-up literature describes treated, severely affected populations.

Ready to start a structured treatment rather than waiting it out? Filter therapists by approach, schedule and payment route → · The full BPD map → · How to check a DBT programme →

Sources

1. Zanarini MC, Frankenburg FR, Reich DB, Fitzmaurice G, "Attainment and stability of sustained symptomatic remission and recovery among patients with borderline personality disorder and axis II comparison subjects: a 16-year prospective follow-up study," American Journal of Psychiatry 169(5), 2012, 476–483 — remission ranges 78–99% and recovery 40–60%; recurrence and loss-of-recovery rates; the definitions of remission and of recovery at a Global Assessment of Functioning score of 61 or higher — pmc.ncbi.nlm.nih.gov.
2. Gunderson JG, Stout RL, McGlashan TH, et al., "Ten-year course of borderline personality disorder: psychopathology and function from the Collaborative Longitudinal Personality Disorders study," Archives of General Psychiatry 68(8), 2011, 827–837 — 91% and 85% cumulative remission; 11% relapse on the 12-month definition; criteria 6.7 to 1.7; GAF 53 to 57; 21% with good functioning at ten years; employment 19% to 36%; "high rates of remission, low rates of relapse, and severe and persistent impairment in social functioning" — pmc.ncbi.nlm.nih.gov.
3. McMain SF, Guimond T, Streiner DL, Cardish RJ, Links PS, "Dialectical behavior therapy compared with general psychiatric management for borderline personality disorder: clinical outcomes and functioning over a 2-year follow-up," American Journal of Psychiatry 169(6), 2012, 650–661 — "even though two-thirds of the participants achieved diagnostic remission and significant increases in quality of life, 53% were neither employed nor in school, and 39% were receiving psychiatric disability support after 36 months" — pubmed.ncbi.nlm.nih.gov.
4. Bateman A, Fonagy P, "8-year follow-up of patients treated for borderline personality disorder: mentalization-based treatment versus treatment as usual," American Journal of Psychiatry 165(5), 2008, 631–638 — 41 patients from the original trial; "their general social function remains impaired" — pubmed.ncbi.nlm.nih.gov.
5. Zanarini MC, Frankenburg FR, Glass IV, Fitzmaurice GM, "The 24-year course of symptomatic disorders in patients with borderline personality disorder and personality-disordered comparison subjects," Journal of Clinical Psychiatry 85(3), 2024, 24m15370 — absence of a substance use disorder the strongest predictor of recovery — pubmed.ncbi.nlm.nih.gov.
6. Temes CM, Frankenburg FR, Fitzmaurice GM, Zanarini MC, "Deaths by suicide and other causes among patients with borderline personality disorder and personality-disordered comparison subjects over 24 years of prospective follow-up," Journal of Clinical Psychiatry 80(1), 2019, 18m12436 — 5.9% died by suicide and 14.0% of non-suicide causes; "Most borderline patients who died either by suicide (87.5%) or non-suicide-related causes (88%) were not recovered before death"; prior hospitalisations as a predictor — pubmed.ncbi.nlm.nih.gov.
7. Paris J, Zweig-Frank H, "A 27-year follow-up of patients with borderline personality disorder," Comprehensive Psychiatry 42(6), 2001, 482–487 — "The total percentage of suicides from the original cohort has reached 10.3%" — pubmed.ncbi.nlm.nih.gov; Paris J, "Suicidality in borderline personality disorder," Medicina 55(6), 2019, 223 — "Follow-back research has found that suicide occurs in up to 10% of BPD cases… However, lower rates (3%–6%) have been reported in prospectively followed cohorts" — mdpi.com.
8. Álvarez-Tomás I, Ruiz J, Guilera G, Bados A, "Long-term clinical and functional course of borderline personality disorder: a meta-analysis of prospective studies," European Psychiatry 56, 2019, 75–83 — 11 studies, 837 participants; "Mean suicide rate ranged from 2% to 5%" — pubmed.ncbi.nlm.nih.gov.
9. Storebø OJ, Stoffers-Winterling JM, Völlm BA, et al., "Psychological therapies for people with borderline personality disorder," Cochrane Database of Systematic Reviews 5, 2020, CD012955 — 75 trials, 4,507 participants; BPD severity SMD −0.52, moderate-quality evidence; "Subgroup analyses found no evidence of a difference in effect estimates between the different types of therapies" — doi.org.

Paid for by participating therapists. Inclusion is computed from availability data — never purchased. No ads, no data sold.