The evidence for getting better
BPD has one of the worst reputations in the diagnostic manual and one of the better outcome profiles in it. That gap is not a matter of optimism — it is a matter of which decade's information you happened to read first.
What the long-term studies found
Two large prospective follow-up studies tracked people with BPD for a decade and more. This is the broad picture, stated without overselling it.
| Question | What the data showed |
|---|---|
| Do people stop meeting the criteria? | The large majority do, over long-term follow-up. Remission rates rise steadily across ten to sixteen years, and this holds across studies rather than resting on one. |
| Does it stay away? | Mostly. Recurrence after sustained remission is comparatively uncommon — lower than relapse rates for major depression, which is not what the reputation would predict. |
| Which symptoms go first? | The behavioural ones — self-harm, impulsivity, crises. These decline earliest and most reliably. |
| Which are slowest? | Chronic emptiness, abandonment sensitivity and relationship difficulty. These are the ones that persist, which is why people can be in technical remission and still find life hard. |
| Does treatment matter? | It accelerates it substantially. It is also worth knowing that improvement occurred for many participants during periods without treatment, which contradicts the idea that nothing changes without intervention. |
| Does symptom remission mean a good life? | Not automatically, and this is the honest caveat. Social and occupational functioning improved more slowly than symptoms did. Not meeting criteria is not the same as having the life you want, which is the gap chapter 18 addresses. |
Why the reputation is so much worse than the data
Three things produced the reputation, and none of them is the evidence.
The first is the era. Personality disorders were defined as enduring and inflexible, and for decades that definition was read as a prediction. The longitudinal studies that contradicted it are relatively recent, and a great deal of older material remains in circulation, in textbooks and in clinicians trained before it.
The second is clinical sampling. Practitioners see people during crises and frequently do not see them afterwards. Someone who recovers stops attending, which means the clinical impression is assembled almost entirely from people who are currently unwell — a selection effect strong enough to produce a pessimism nothing in the outcome data supports.
The third is that the label attracted genuine hostility. 'Difficult', 'manipulative', 'untreatable' — language that appeared in professional settings and leaked into everything else. Some clinicians declined to treat it. That history is real, it is improving, and it left a residue that people still meet in waiting rooms and search results.
The practical consequence is worth stating plainly, because it changes behaviour: if you read in the first week after your diagnosis that this is a life sentence, that was wrong, and it is worth going back and unlearning it deliberately. Expecting no improvement is one of the more reliable ways to stop doing the things that produce it.
Most of what is written about BPD prognosis was true in 1985 and has not been true for some time.
Honest answers
Including the ones where the honest answer is less comforting.
Is BPD curable?
'Cure' is not really the vocabulary. Most people reach the point of no longer meeting the diagnostic criteria, and most remain more emotionally sensitive than average. Whether that counts as cured is a question about words rather than about outcomes.
Does everyone get better?
No. The studies show high remission rates, not universal ones, and functioning improved more slowly than symptoms. A page that promised otherwise would be lying, and you would be right not to trust the rest of it.
Do I need treatment for this to happen?
It helps considerably and it is not the only route — improvement occurred for many people during untreated periods. That is a reason for hope if services are inaccessible, not a reason to decline them.
How old is this evidence?
The main prospective follow-up work runs from the 1990s onward with sixteen-year and ten-year reports. It is the reason clinical guidance changed, and it is why anything you read describing BPD as untreatable is out of date rather than merely pessimistic.
What if I've had it for twenty years?
Duration is a weaker predictor than the studies' pessimists expected. People in their forties and fifties reach remission. Starting late is worse than starting early and it is substantially better than not starting.
Does the book make claims about this?
It states that recovery is realistic and points to the long-term evidence rather than to reassurance, and it is fairly careful not to promise it to any individual reader.