Can supportive therapy be modified to successfully treat patients with borderline personality disorder? By using a previously developed model, NIMH-funded researchers have found supportive therapy helpful in engaging patients in treatment, developing a therapeutic alliance and achieving treatment goals. Their outcome data may provide a new treatment approach for this difficult-to-treat population.
"Well, even borderlines can use handholding," said a colleague when he heard about the supportive-therapy treatment cell of a research project funded by the National Institute of Mental Health on borderline personality disorder (BPD). Barbara H. Stanley, Ph.D., was the principal investigator of this prospective, randomized study comparing Marsha Linehan, Ph.D.'s, dialectical behavioral therapy (DBT) to supportive therapy for patients with BPD and suicidal behavior.
To many psychiatric clinicians, supportive therapy has traditionally meant a type of treatment with limited goals that is suitable for patients who are too impaired to "really work" in therapy. The above offhand comment indicated that, despite the mounting evidence suggesting that supportive therapy is an effective treatment approach (Buckley, 1986; Wallerstein, 1989), such prejudices still linger among psychiatric professionals. For many psychiatrists, supportive therapy is synonymous with nonspecific treatment, with "just being nice to patients," as our colleague put it.
However, "being nice" to individuals with BPD will soon backfire, as inexperienced clinicians often discover. Core characteristics of patients with BPD such as splitting, idealization and devaluation--not to mention aggressive anger, acting out and psychotic regression--may soon leave clinicians regretting having a single "nice" bone in their bodies.
There are numerous models (or perhaps flavors) of supportive therapy, which range from psychodynamic to interpersonal to cognitive-behavioral to atheoretical (Kernberg, 1984; Luborsky, 1984; Novalis et al., 1993; Pinsker, 1997; Rockland, 1989). For our adaptation, we primarily used models defined by Pinsker (Hellerstein et al., 1994; Pinsker, 1997) and Novalis (Novalis et al., 1993). Pinsker's model defined supportive therapy as a treatment that emphasizes building self-esteem, reducing anxiety and enhancing coping mechanisms. Hellerstein's previous work at Beth Israel Medical Center had been involved in the development of this model. By this definition, supportive therapy is conversational in style and commonly uses techniques such as clarification, suggestion, praise, education and examination of the influence on present life of patterns originating in the past. Also, supportive therapy rarely uses techniques such as prolonged silent listening, neutrality, confrontation of resistance or transference interpretations. Wherever possible, therapy-induced anxiety is avoided (Hellerstein et al., 1998, 1994; Pinsker, 1997; Rosenthal et al., 1999).
Clearly this is a different approach from merely "being supportive"--this approach is disciplined, thoughtful and goal-oriented.
Is supportive therapy a good treatment approach for patients with BPD? At first glance, supportive therapy seems grossly inadequate as a treatment for individuals with BPD, particularly for those with current self-injurious and suicidal behaviors, which include cutting, overdoses, head-banging and face-gouging. Could a once-a-week, fairly unstructured therapy approach compete with the highly organized DBT approach (Linehan, 1993)? More than that, we worried if it would be safe to treat such patients with this modality.
It seemed that the only ethical way to adapt the Beth Israel supportive-therapy model was to develop it as an active psychotherapy approach, not a placebo condition and not just good clinical management. The Beth Israel model had been developed in an urban clinic setting, then applied to a psychotherapy research program with a relatively high-functioning employed Cluster C population (mostly patients with dependent and avoidant personality disorders [PDs]) (Hellerstein et al., 1998; Rosenthal et al., 1999). But we had also adapted this supportive-therapy model for individuals with schizophrenia who were substance abusers (Hellerstein et al., 1995), clearly a population with very different needs. Perhaps it could be used for individuals with BPD as well.
In adapting supportive therapy for BPD, we looked at core DSM-IV features of that personality disorder and then attempted to determine how supportive-therapy interventions might address them (Aviram et al., in press). For instance, the patient with BPD's "frantic efforts to avoid real or imagined abandonment" might be addressed by anticipatory guidance--that is, by planning ahead for the next such (often predictable) event. Patterns of unstable and intense interpersonal relationships might be addressed by clarification ("this is the pattern you find yourself in"), by education ("this is common with BPD") and by "striking while the iron is cold" (Pine, 1984)--finding ways to de-intensify personal relationships. Chronic feelings of emptiness, impulsive behavior and inappropriately intense anger could be addressed by various supportive therapy techniques. Emptiness could be addressed by clarification, psychoeducation and naming feelings. Impulsive behavior could be addressed by expanding available choices and discussing issues when the patient was no longer upset. Intense anger might be addressed by offering control.
Developing a Model
The more that we worked on developing our model, the clearer it became that supportive therapy might actually be a good treatment approach for BPD. Borderline personality disorder has been conceptualized as a disorder of affective dysregulation and high impulsivity (Siever and Davis, 1991) and is also characterized by cognitive distortions and intense interpersonal reactivity. Supportive therapy techniques, including those mentioned above, might be helpful in improving affective regulation, decreasing impulsive behavior and gradually correcting cognitive distortions. And the "real" relationship with the supportive-therapy therapist might be very helpful in de-intensifying relationships and might provide a model for improving other relationships in the patient's life. Some aspects of supportive therapy might be particularly beneficial for patients with BPD; for instance, the goal of decreasing (rather than increasing) anxiety in therapy sessions. The supportive-therapy therapist rarely asks patients why they did something; instead, the therapist may make a comment that allows patients to choose whether or not to respond. The question why may be very threatening for people with a fragile sense of self.
Beyond this, there is supportive therapy's conversational style, rather than prolonged periods of silence that might provoke dissociation, paranoia or regression. Supportive therapy is conversational, but not merely conversation. The supportive-therapy therapist directs this dialogue toward mutually agreed-upon goals. Another potential advantage: Unlike psychodynamic therapy, supportive therapy minimizes the exploration of transferential feelings--transference is addressed only in situations where the therapy is threatened. This could also be calming and nonthreatening to the patient. The focus of supportive therapy on the present, rather than the past, might be particularly useful for patients with BPD, minimizing regression and identity diffusion. Finally, the fact that supportive therapy allows the patient to set the goals of treatment (in collaboration with the therapist) might be helpful for many individuals with BPD who often feel controlled and manipulated by others.
What about suicidal and self-injurious behavior? Unlike DBT, which has a strict protocol for dealing with such activities, many supportive-therapy models (Pinsker's, for instance) offer little overt guidance. Obviously, the supportive-therapy therapist wants to prevent patients from injuring themselves, but the main work in supportive therapy occurs when patients are calm. We therefore decided that the general clinical management approaches of the American Psychiatric Association's practice guidelines make sense in such situations (Jacobs et al., 2003; Oldham et al., 2001). These include contracting for safety. Patients have their therapists' beeper numbers and can contact them if acutely suicidal. Patients may call their therapist daily, or more often during times of crisis, and may increase the frequency of visits.
Clearly some valuable supportive therapy work could be done during such crises. For example, "Ms. A" paged her therapist one night, thinking her mood was depressed and suicidal. In speaking with her therapist, Ms. A began to recognize that she was angry at someone, and this helped her connect her feelings to actual events and to no longer feel suicidal.
However, the main goal in supportive therapy is prevention of such suicidal crises, in particular, by helping patients develop more adaptive alternatives. One key in supportive therapy is the word and. Patients may want to kill themselves or may feel compelled to cut or to take an overdose and yet they may have other choices: to call a friend, to go for a walk, to "do nothing" and so on. Suicidal behavior is always one choice and there may be other choices, which patients may not initially be aware of. In supportive therapy terms, the therapist tries to help the patient to improve adaptive skills--that is, to use more positive behaviors, both intrapsychic and interpersonal. The supportive-therapy therapist repeatedly, and in a low-key way, works with patients with BPD to introduce and into their world (see Aviram et al. [in press] for a fuller discussion of these issues).
It became clear early on in our work with these patients that there were other challenges. Harsh self-judgment clearly is common in patients with BPD: a sense of being worthless, an outsider, all bad and so on. Obviously, this may contribute greatly to self-injurious behavior. Therapists use core interventions of supportive therapy to address this issue, continually working to find things that are praiseworthy and adaptive, which may help to deflect such harsh feelings. With an angry, impulsive, desperate patient, finding something that the therapist can honestly praise may be a challenge, but it is essential. Patients with BPD often cannot see or acknowledge their real strengths, which the therapist can repeatedly point out: "It is important for you to be a good parent," "You showed a lot of initiative in dealing with that problem at work." Or even: "As frustrating as it is, it's clear that you're really trying to find better alternatives." Supportive therapy focuses on two areas: building self-esteem (i.e., feeling more positive about oneself) and enhancing psychological functions (i.e., using more adaptive defenses such as suppression or intellectualization, rather than using regression, splitting or projective identification).
One's sense in working with such patients over time is that one is building up from fragments of their personality, trying to reinforce and strengthen these using various supportive therapy techniques and trying to prevent the patient from being overwhelmed by aggressive or fearful reactions. Interventions like reframing, clarification, advice, education and anticipatory guidance are used repeatedly throughout the course of supportive-therapy treatment. Perhaps on a neural level, such interventions may dampen amygdalar hyperactivity and increase higher-level synaptic connections and, eventually, lead to top-down control over what have previously been impulsive limbic and paralimbic-driven behaviors (Siegel, 1999).
Phases of Treatment
In early stages of treatment, much time is spent dealing with suicidality--with self-injurious behavior being a key aspect--and helping patients to develop more adaptive alternatives. Other aspects include dealing with derealization/dissociation, idealization/devaluation, harsh self-evaluation, and anxiety and depression. Later in treatment, therapists focus on helping the patient develop positive aspects of their life-working on relationships, improving work functioning, and establishing and maintaining positive feelings about themselves. In this phase, patients may benefit from naming feelings ("when he speaks to you like that, it sounds like you feel enraged"), from anticipatory guidance ("you dealt with that very well last time, how would you like to address it next time?") and from offering control ("you can choose to walk away at that point, rather than to answer back"). Many patients find their intimate or work relationships improving over time, which gives them increased confidence.
Finally, given that our research study is based on a one-year treatment, there is the issue of termination. In supportive therapy, the therapist works to help the patient not regress around this phase. This includes a realistic discussion about the ending of the therapeutic relationship and the feelings that the patient may have, as well as planning for further treatment. Supportive therapy differs from many other treatment approaches in working to normalize and contain (rather than explore) the feelings around this phase of treatment. Patients are completing a "course" of treatment--they may take many such courses in life (similar to college courses) and the goal is to take something away from the treatment experience that may be useful in later life.
As our supportive-therapy treatment approach has developed, it appears to us that it remains within the umbrella of Pinsker's and Novalis' models. In general, the types of interventions used are ones that are standard with other populations, but we obviously have modified them to deal with individuals who may be volatile, fearful and impulsive. Perhaps most notably, there is the constant challenge of developing (and maintaining) the therapeutic alliance with patients with BPD. With patients who have Cluster C PDs the therapeutic alliance may not involve continual attention, whereas with patients with BPD the supportive-therapy therapist must constantly modulate distance from the patient, trying to not be too close, yet not too far. Otherwise, the patient may decompensate or flee from treatment. Therefore, our current model of supportive therapy is probably more relational than Pinsker's original definition. In its continual work on reframing and dealing with black-and-white thinking, it may have more of a cognitive slant as well.
It is also becoming clear that once-a-week supportive therapy is not suitable for every patient with BPD. While many patients make significant progress, for others, a once-weekly approach appears to be insufficient. Such individuals may need more frequent visits, day-treatment programs, family interventions, inpatient hospitalization or more aggressive medication treatment (Jacobs et al., 2003; Oldham et al., 2001). Nevertheless, at present it seems that supportive therapy may work for many (if not most) people with BPD in terms of engaging them in treatment, developing a good therapeutic alliance and working to attain treatment goals. This is all an impression, of course; we await our study's outcome data.
At this point, however, our impression is that supportive therapy may be useful both for research and for clinical settings. In research, supportive therapy may be a good treatment to compare to other approaches--it appears quite different from other approaches such as DBT or psychodynamic treatment. To truly demonstrate the superiority of a specific psychotherapeutic approach such as DBT, it makes sense to compare it to a disciplined, well-defined standard treatment approach, such as supportive therapy, rather than to poorly defined clinical management. Beyond that, it seems to us that in clinical settings, supportive therapy is a logical type of treatment to use with individuals with BPD. Not all patients with BPD would agree to take part in DBT, for one; also, supportive therapy may be easier and more intuitive for clinicians to learn. Psychiatry residency training now requires supportive-therapy supervision and training, so there will be a growing cohort of psychiatrists trained in this technique who could learn to adapt it for patients with BPD.
In conclusion, after three years, our clinical impression is that supportive therapy may be a promising treatment approach for patients with BPD. It appears to be adaptable for many treatment settings and probably is better than having an unfocused, eclectic approach with poorly defined goals and therapeutic interventions. Given findings from a longitudinal study of outcome in BPD (Stone, 1992), the flexible yet disciplined approach of the supportive-therapy therapist may meet the evolving needs of this population on a long-term basis.
Have we just reinvented handholding? Our guess is that we may have refined a treatment approach that is beneficial and effective. But clearly more data should be obtained.
This research was supported by NIMH Grant RO1 MH 57469, awarded to Barbara H. Stanley, Ph.D., in the division of neuroscience at the New York State Psychiatric Institute.
Dr. Hellerstein is clinical director of the New York State Psychiatric Institute and author of the upcoming book A Guide for the Journey, which focuses on new psychiatric treatments.
Dr. Aviram is a research scientist in the department of neuroscience at the New York State Psychiatric Institute and a psychotherapist in the supportive-therapy cell of the NIMH-funded grant discussed in this article.
Dr. Kotov is a research scientist in the department of neuroscience at the New York State Psychiatric Institute and a psychotherapist in the supportive-therapy cell of the NIMH-funded grant discussed in this article.