Five main findings have emerged from this study. The first is that remissions lasting two to eight years are extremely common for borderline patients (and axis II comparison subjects). In fact, cumulative rates at 16-year follow-up for borderline patients ranged from a high of 99% for a two-year remission to 78% for an eight-year remission. These results are consistent with and extend our findings pertaining to the first 10 years of prospective follow-up in this study where we found that 93% of borderline patients had achieved a remission lasting two years (5
). However, the results of the current study found that remissions, regardless of length, occurred significantly more rapidly for axis II comparison subjects than for borderline patients. This finding may reflect the greater severity of borderline psychopathology than the axis II psychopathology found in the other personality disorder group. Our findings pertaining to a two-year remission are also consistent with the follow-back findings of Paris (2
) and the follow-along findings of the Collaborative Longitudinal Personality Disorders Study (6
The second main finding is that recurrences of borderline personality disorder were relatively rare, ranging from 36% after a two-year remission to 10% after an eight-year remission. Axis II comparison subjects reported even lower rates of recurrence, ranging from 7% after a two-year remission to 4% after an eight-year remission. These results are consistent with and extend our findings pertaining to the first 10 years of prospective follow-up in this study where we found that 30% of borderline patients had a recurrence of borderline personality disorder after a two-year long remission (5
). However, the results of the current study found that recurrences occurred significantly more slowly for axis II comparison subjects than for borderline patients. This finding is not surprising given the relative severity of borderline psychopathology. However, the opposite pattern was found in the Collaborative Longitudinal Personality Disorders Study where those with another personality disorder (who had as their primary axis II diagnosis either avoidant or obsessive-compulsive personality disorder) had higher rates of recurrence than those with borderline personality disorder (25% vs. 11%). This difference may be due to the substantially lower rates of retention in the Collaborative Longitudinal Personality Disorders Study after 10 years of prospective follow-up than in the McLean Study of Adult Development after 16 years of prospective follow-up (surviving borderline patients: 66% vs. 88%; surviving patients with another personality disorder: 69% vs. 83%). It may also be due to the fact that 13% of the other personality disorder group in the Collaborative Longitudinal Personality Disorders Study had a secondary diagnosis of borderline personality disorder at the time of study entry (17
The third main finding is that recovery from borderline personality disorder occurred at a lower rate and more slowly than recovery from another personality disorder. In terms of recoveries lasting two years, 60% of borderline patients and 85% of axis II comparison subjects attained this important outcome. In terms of recoveries lasting eight years, these rates dropped to 40% and 75% respectively. It is not surprising that axis II comparison subjects achieved this outcome more rapidly and at a higher overall rate regardless of the length of recovery than those in the borderline group. This is so because previous research by our group has found that borderline patients have far more difficulty functioning consistently and competently in the vocational arena than axis II comparison subjects (18
). We have also found that borderline patients are significantly more likely to support themselves through federal disability benefits than axis II comparison subjects (19
). And yet these between-group differences are striking and represent a new finding.
The fourth main finding is that the loss of recovery was substantially but not significantly more common for borderline patients than axis II comparison subjects, in part due to the relatively small number of events in each group. Rates of loss of recovery after a two-year recovery were 44% for borderline patients and 28% for axis II comparison subjects. Rates of loss of recovery after an eight-year recovery were 20% and 9% respectively. These results too, which are new, probably reflect the unstable vocational performance of borderline patients relative to those with other forms of axis II psychopathology (18
The fifth main finding is that recoveries occurred at lower rates than remissions for those in both study groups. Only 60% of borderline patients achieved a recovery lasting two years as opposed to 99% of borderline patients achieving a two-year remission. This discrepancy was smaller for those with another personality disorder as 99% had a two-year remission during the course of 16 years of prospective follow-up and 85% attained a two-year recovery involving symptomatic remission of their primary axis II diagnosis and good concurrent social and vocational functioning. In terms of remissions and recoveries lasting eight years, comparable figures for those in the borderline group were 78% remitted vs. 40% recovered. In the other personality disorder group, these figures were 97% remitted and 75% recovered. Clearly those in the other personality disorder group were more likely to both remit and recover, perhaps due to the lower level of severity of their axis II psychopathology or perhaps due to their greater capacity to perform vocationally or the confluence of the two.
Taken together, the results pertaining to high rates of remission of 2–8 years in length represent very good news for borderline patients, their families, and the mental health professionals treating them. The same can be said of the relatively low rates of recurrence following these lengthy remissions. Given these findings pertaining to symptomatic outcome, it seems fair to suggest that borderline personality disorder has a better symptomatic outcome than other major mental illnesses with which it is frequently compared, such as major depression (20
) or bipolar I disorder (22
However, our results pertaining to recovery are more sobering, particularly when compared with the rates of recovery attained by axis II comparison subjects. While 50% of borderline patients attained a two-year recovery after 10 years of prospective follow-up (5
), only 60% attained this outcome after an additional six years of prospective follow-up. And this 60% compares unfavorably to the 85% of axis II comparison subjects who attained this outcome by the time of the 16-year follow-up period. Perhaps even more sobering is the fact that only 40% of borderline patients but 75% of axis II comparison subjects attained a recovery that lasted eight years or more (or half the length of prospective follow-up).
Cumulative rates of longer remissions and recoveries and the recurrences and losses of recovery that follow them might be expected to be lower than those for less lengthy outcomes due to the shorter risk periods involved. However, when differences in risk periods are taken into account, our results suggest that it is relatively more difficult to achieve longer remissions and recoveries than shorter ones. And that the lower rates of recurrence and loss of recovery are not solely artifacts of shorter risk periods but represent stable changes in symptomatology and psychosocial functioning.
As noted above, we have previously found that vocational impairment is the main reason that borderline patients fail to attain and/or maintain recovery involving both symptomatic remission and good social and vocational functioning (18
). The reasons for the vocational dysfunction that drives the difficulty borderline patients have attaining and maintaining recovery are unclear and may differ from patient to patient. Some of this difficulty may be due to residual symptoms of borderline personality disorder. This difficulty may also be due, at least in part, to concurrent axis I disorders. However, our experience suggests that some borderline patients may be more goal oriented and more competent than others. Or looked at another way, some patients may be more resilient than others.
We have previously suggested a rehabilitation model to deal with this impairment in functioning (5
). Links (24
) and Gunderson et al. (6
) have also suggested this approach to dealing with the vocational dysfunction common among seriously ill borderline patients.
However, not all therapists, family members, or friends are supportive of efforts to help the borderline patient attain a solid adult adaptation in the vocational realm. Some may be concerned that the stress of trying to work may lead to an upsurge in suicidal urges or episodes of self-mutilation. Others may believe that a borderline patient should not attempt to work until most of his or her more serious symptoms are fully resolved. Yet others may be concerned about the patient's losing access to the almost unlimited psychiatric treatment provided by Medicare, which is typically provided to those receiving Social Security Disability benefits.
In the end, some borderline patients will attain a good vocational adjustment and overall recovery with minimal support from those to whom they are close. Yet others will be able to attain this adaptation with much support and personal struggle. And yet others simply cannot or will not work consistently and/or competently and thus, will not attain our multifaceted definition of recovery. They too will need support as they bear the shame and disappointment attendant on failing to achieve the life they had once dreamt of and planned.
Poor health-related outcomes join vocational impairment as areas where a more guarded prognosis has been found (25
). In particular, we have found that obesity and obesity-related illnesses are both common among borderline patients (25
) and associated with poor vocational functioning (27
). We have also found that chronic PTSD, lack of exercise, a family history of obesity, and aggressive polypharmacy are risk factors for obesity in borderline patients (26
). Further investigation of the interconnection of these poor prognostic factors is warranted.
This study has a number of limitations. The first is that all subjects were initially inpatients. It may well be that borderline patients who have never been hospitalized are less severely ill symptomatically and less impaired psychosocially and thus, more likely to remit more rapidly and attain a good global outcome over time. The second is that the majority of those in both study groups were in non-intensive outpatient treatment over time (28
) and thus, the results may not generalize to untreated subjects.
Taken together, the results of this study suggest that sustained symptomatic remission is substantially more common than sustained recovery from borderline personality disorder. They also suggest that sustained remissions and recoveries are substantially more difficult for borderline patients to attain and maintain than those with other forms of personality disorder.