Earlier Recovery Signals Continued Progress in BPD

In spite of the grave and seemingly intransigent nature of borderline personality disorder, a number of persons with it do get better over time. This finding has emerged over the past 20 years or so in the work of Thomas McGlashan, M.D., a professor of psychiatry at Yale University; Joel Paris, M.D., chair of psychiatry at McGill University; and other psychiatric researchers.

This finding, of course, raises a question: Why do a number of persons with the disorder do better as time goes by? “We don’t know,” Paris admitted in an interview with Psychiatric News. “But I’ll give you several possibilities. One is that many disorders associated with impulsiveness improve over time. You see that with antisocial personality, you see it with drug abuse, and you see it with bulimia. If you look at community studies, people are less impulsive at 40 than at 20.”

He added that another explanation is that “people do better because they learn to avoid things that give them trouble and to focus on things where they do better.”

In any event, if a number of persons with borderline personality disorder improve over the long haul, is there any way to forecast which ones will do better? Quite possibly, it appears, although there are some conflicting answers in the research literature.

For instance, limited emotional instability at the start of illness predicts a favorable outcome, McGlashan found a few years ago. In contrast, Paris, along with Hallie Zweig-Frank, Ph.D., a research associate at Jewish General Hospital in Montreal, failed to find that that is the case. They have found, however, that making good progress against the disorder in one’s earlier adult years presages even greater progress against the disorder as time goes on. They reported these findings in the March/April Comprehensive Psychiatry.

The research that led to these findings by Paris and Zweig-Frank started in the 1950s. Paris and several other researchers reviewed, between 1958 and 1978, the hospital charts of some 3,000 persons who had been admitted to a city hospital because of suicide attempts, substance abuse, or brief psychotic episodes.

Using a variation of the Diagnostic Interview for Borderlines that was adapted for chart review, Paris and his coworkers identified 322 persons who had borderline personality disorder. All of these people were under age 35.

Then in 1986 and 1987, that is, on average 15 years later, Paris and his coworkers followed up 100 of the original 322 subjects to determine their borderline personality status. The 100 subjects were comparable to the original sample on sex, education, and initial Diagnostic Interview for Borderlines scores, so that the researchers considered them representative of the original subjects.

The investigators then assessed their borderline personality disorder status with the Diagnostic Interview for Borderlines, Revised, and the Health-Sickness Rating Scale, which was equivalent to the later-developed Global Assessment of Functioning scale.

The results constituted both good and bad news. The bad news was that 9 percent of the subjects had committed suicide after they were initially diagnosed with borderline personality disorder. The good news, however, was that 75 percent of the subjects no longer appeared to be borderline. Paris and his coworkers reported these results in the November/December 1987 Comprehensive Psychiatry.

Then in 1999 Paris and Zweig-Frank assessed the borderline personality disorder status of 64 of the 100 subjects. They were comparable to the 100 subjects on sex, education, marital status, 1986-1987 Diagnostic Interview for Borderlines scores, and 1986-1987 Health-Sickness Rating Scale scores. At this point, the subjects were, on average, 51 years old, and it had been, on average, 27 years since they had been initially diagnosed with borderline personality disorder.

Paris and Zweig-Frank used five yardsticks to measure subjects’ borderline personality disorder status—the Diagnostic Interview for Borderlines, Revised; Schedule for DSM-III-R Diagnosis; Global Assessment of Functioning scale; Symptom Checklist-90 (a self-report of psychiatric symptoms); and Social Adjustment Scale (a self-report of social adaptation).

Only 8 percent of the subjects still met borderline personality disorder criteria at this point in their lives, Paris and Zweig-Frank reported in the November/December 2001 Comprehensive Psychiatry.

And now Paris and Zweig-Frank have investigated whether the borderline personality status of these 64 subjects in 1999 might have been predicted by certain aspects of their lives when they were initially diagnosed with the disorder 27 years earlier or at least by certain aspects of their lives 15 years earlier, when they were reassessed.

They could find no significant link between the subjects’ 1999 status and their initial Diagnostic Interview for Borderlines scores 27 years earlier. They also uncovered no links between their 1999 status and their socioeconomic status in 1986 or between their 1999 status and any childhood adversity they recalled experiencing, such as physical abuse, sexual abuse, unaffectionate parents, or controlling parents.

However, Paris and Zweig-Frank did find a significant association between the subjects’ 1999 borderline personality disorder status and their disorder status 15 years earlier. Their level of functioning earlier in adulthood seemed to predict level of functioning later in adulthood, whereas initial severity of illness, demographic factors, and childhood experiences did not.

Or as Paris and Zweig-Frank summed it up in their study report: “Those on the road to recovery at the 15-year follow-up were also those who continued to progress later in adulthood.” ▪