How can psychiatrists and psychotherapists optimize their collaboration and individual skills to facilitate treatment success?
Although the focus of some debate, a number of excellent studies comparing psychotherapy or medications alone versus combination therapy have reported two important findings. First, the combination of psychotherapy and medication is better than either treatment alone (primarily in severely ill or chronic patients) and second, biological symptoms (e.g., sleep disturbance, agitation) generally respond better to medications, whereas psychological and interpersonal deficits are more effectively treated by psychotherapy (Barlow et al., 2000; Dewan and Pies, 2001; Keller et al., 2000). This paper addresses several key issues that are important in making combined treatment work effectively. First, medications have powerful psychological meaning for both patients and therapists. Second, when indicated, combined treatment can be provided by a psychiatrist (integrated treatment) or a psychiatrist/psychopharmacologist plus a non-physician psychotherapist (split treatment).
While I present data suggesting that integrated treatment by a psychiatrist is the treatment of choice on both theoretical and economic grounds, split treatment continues to be widely practiced. Psychotherapy by itself is a complex procedure that requires great skill and has the potential for both benefit and harm. Adding another modality (medication) and another partner (a prescribing physician) requires even greater sensitivity and skill on the part of the clinician.
When medications are used in addition to psychotherapy, it is important to emphasize that combined therapy does not mean that only half the usual attention needs to be paid to each modality. Medications (e.g., antidepressants, anxiolytics and sleeping medications) are sometimes prescribed long after they would have been discontinued if the patient had been treated with medication alone. Also reported are cases where primary care physicians (and some psychiatrists who primarily practice psychotherapy) add subtherapeutic amounts of medications to ongoing therapy due to their discomfort with psychopharmacological agents (Dewan, 1992).
Every patient and therapist brings their own unique and personal attitudes toward medication. Therefore, therapists need to carefully assess their own reasons for considering medications and also look for the psychological reactions and meaning, both obvious and covert, that medications have for each patient.
Some patients derive a psychological benefit from being given medications since they consider it a caring, nurturing act that feeds their dependency needs or validates their suffering. Other patients see it as an imposition of external control or as a statement by the therapist that they are not strong enough to solve their problems by themselves, which may contribute to noncompliance. Offering medications to patients in denial of their illness means that they have to confront their worst nightmare and acknowledge that they are very sick. Some patients with bipolar disorder will take antipsychotic medications for a short time but refuse the long-term use of mood stabilizers because it forces them to recognize that they are suffering from a chronic illness (Dewan, 1992).
Not offering medications is also interpreted in different ways. Some patients see it positively, feeling that the therapist "must be interested in me as a person and not just in my symptoms" or "I am competent enough to do it by myself." Angry and dependent patients may regard it negatively, as a withholding of support or prolongation of their agony. This is particularly potent since our culture vigorously promotes the false idea that a pill can fix everything. Other patients feel they are not being taken seriously, or are not considered sick enough, or even that the therapist thinks they are faking their symptoms. Some patients so overvalue their medication that they will carry around the unfilled prescription as a soothing, and often very effective, good luck charm or transitional object.
Psychotic patients may have unusual, idiosyncratic associations to the names of medications. One of my patients vehemently opposed taking Stelazine (trifluoperazine) ("I hate it! It reminds me of my sister Stella"), but graciously agreed to taking an equivalent drug, Mellaril (thioridazine) ("That's fine. It will make me mellow, right doc?") (Dewan, 1992).
Therapists also have strong biases and reactions toward the use of medications. Sometimes therapists do not refer for medication evaluation because they erroneously believe that relief from medications will undermine psychotherapy or that medications are not needed because there is a psychological explanation for the symptoms. Some therapists oppose the use of benzodiazepines, believing them to be addicting. Thus, patients are deprived even when such medications could be enormously helpful and prescribed safely. The therapist's unrecognized feelings, be it fear, hate or sexual attraction toward a particular patient, may unfairly dictate the addition of medication as a way of distancing, controlling or even punishing the patient (Dewan, 1992).
Over the past decade, medicine in general and managed care organizations (MCOs) in particular have moved toward the primary care model. The aim is to provide primary, holistic care in which one doctor takes care of all the patient's needs. However, the only mental health care practitioner capable of providing comprehensive biopsychosocial care--the psychiatrist--is being replaced by a supposedly less costly but fragmented split treatment model.
Using a theoretical model, I first challenged the widely held assumption that split treatment was less expensive than integrated treatment (Dewan, 1997). In a more elaborate study, I collected the 1998 payment schedule from seven large MCOs with a combined market share of 54% and 67.8 million covered lives (Dewan, 1999). Medicare, covering 36.9 million people, was also added. As can be seen in the Table, (Due to copyright restraints, we are not able to publish this table on the Internet. Please see our print publication.) there are both substantial cost and time savings with integrated treatment. Furthermore, "When time away from work or child care plus the expense of traveling are factored in, the cost benefit analysis favors integrated care from a psychiatrist even more." Even when a social worker was substituted for the psychologist, the cost for integrated treatment was only $16 higher (Dewan, 1999).
In an elegant study, Goldman et al. (1998) used an MCO's database to retrospectively evaluate 1,517 depressed patients followed for 18 months. It is revealing that, despite this MCO's stated aim of trying to "directly refer to psychiatrists for both psychotherapy and pharmacotherapy," only 191 patients (13%) were in integrated treatment. Patients in split treatment needed more psychotherapy sessions (21.2 versus 10.4), more medication visits (6.3 versus four), and more total outpatient care (26.2 versus 14.7 visits). Instead of split treatment leading to savings, it cost $518 more per patient ($1,854 versus $1,336). For all 1,326 patients in split treatment, the MCO paid an extra $686,868! The authors concluded:
Splitting psychotherapy and pharmacotherapy is a practice and point of view that has in effect been legislated without evidence...This study contradicts the pervasively held belief that split treatment is more cost-effective.
In fact, these data indicate that integrated treatment provided by a psychiatrist is the preferred theoretical (biopsychosocial) and economic model. However, there are no studies comparing the therapeutic outcomes of integrated versus split treatment. There is an urgent need for studies in this area.
In 1980, the American Psychiatric Association attempted to clarify split treatment relationships and described three models. In the consultative model, the psychiatrist provides a limited consultation but no care. In the private practice collaborative model, responsibility is shared equally. However, in fact the psychiatrist carries the greater liability risk because they are deemed the head of the team by the legal system and have deeper pockets. In the supervisory model, most often seen in clinics, the psychiatrist is responsible for the initial diagnosis, formulation of the treatment plan, and for supervising, directing and monitoring all aspects of the therapist's work.
Collaboration between disciplines has many advantages for the patient and the collaborators. The patient receives greater amounts of time and expertise, which may lead to better adherence to medications. Collaboration provides an invaluable opportunity for professional and emotional support of each other on an ongoing basis, especially in times when the patient is in crisis or when treatment has a disastrous outcome such as a suicide.
Unfortunately, therapists and psychiatrists often do not take the additional time to interact regularly and instead work on parallel tracks. Further, interdisciplinary competition, whether conscious or not, allows for the undermining of each other's work or working at cross purposes and lends itself to splitting by the patient. To avoid failure, the therapist and the psychiatrist need to build a mutually trusting and respectful relationship, one that clearly recognizes the special and differing skills that each partner brings to the collaboration.
Meyer and Simon (1999) offered a model letter that a psychiatrist would send to a referring therapist that addresses the important elements each partner needs to know about the other: their qualifications, certification, liability coverage, experience, clinical orientation, who will provide what part of the treatment, how to contact each other after hours, vacation coverage (do not cover each other; get someone from your own discipline), how emergencies will be handled and confidentiality. The patient needs to know that all information will be shared between the collaborating partners and should sign appropriate releases at the outset.
Himle (2001) enumerated what each partner can expect from the other. It is expected that the psychiatrist will ask the therapist for a clinical report with a reason for referral and then interview the patient. If medications are recommended, the reasons and specific options will be discussed with the patient. Since patients have very personal reactions to medications, the specific medication is best chosen as a collaborative venture to improve adherence. How to take the medication, expected improvement and potential side effects are clearly described and perhaps even written down so that both the patient and therapist are aware of them. The schedule for follow-up appointments will be clearly spelled out. After the first or second appointment, the psychiatrist and therapist will communicate directly and agree on a treatment plan. Finally, the psychiatrist will support the psychosocial treatment plan, but not discuss psychotherapeutic issues with the patient.
The therapist is expected to provide a written clinical summary with a reason for the referral, preferably followed by a brief phone call to the psychiatrist. The therapist will set the stage appropriately with the patient about the referral for medication consultation by describing the symptoms to be targeted and summarizing the research data that suggest medications are likely to be helpful. With psychotic patients, I emphasize medications as critical; with personality-disordered patients I present medications as a potentially useful adjunct to the more important work of psychotherapy; and with other patients, I present medications as an equal and often synergistic partner with ongoing psychotherapy. Patients need to be told explicitly that the consultation may or may not result in medication being prescribed. It is also important that the referral be for an open-ended consultation and not for a specific medication.
An important expectation of therapists is that they will promptly convey to the collaborating psychiatrist any clinical deterioration, suspected side effects or medical problems. This is critical, since the therapist will generally see the patient more frequently than the psychiatrist and is usually the best-informed member of the treatment team. It is important that therapists educate themselves as to the most common side effects of the frequently prescribed medications.
Finally, therapists are expected to fully support the medication regimen and are an important ally in improving adherence. If the therapist disagrees with or wants to change the medications, this discussion should take place directly with the physician and not through the patient. Similarly, specific questions from the patient about medications should be referred back to the prescribing physician.
The prevailing primary care model and the traditional biopsychosocial model allow psychiatrists to provide seamless, integrated combined treatment. In addition, for many conditions such as depression, integrated care is more cost-effective than split treatment. However, since split treatment continues to be widely practiced, it is important to pay particular attention to several aspects in order to make it work. The meaning of medications for both the patient and the therapist must be taken into account. Successful split treatment requires clarity of roles, respect for the many strengths that each profession brings to the enterprise and an appreciation of the powerful psychological dynamics of each. A thoughtful and sometimes energetic engagement is required by all partners in order to avoid potential pitfalls and benefit from the rich promise of medications and collaborative care.
American Psychiatric Association (1980), Guidelines for psychiatrists in consultative, supervisory or collaborative relationships with nonmedical therapists. Am J Psychiatry 137:1489-1491.
Barlow DH, Gorman JM, Shear MK, Woods SW (2000), Cognitive-behavioral therapy, imipramine, or their combination for panic disorder: a randomized controlled trial. [Published errata JAMA 284(19):2450; 284(20):2597.] JAMA 283(19):2529-2536 [see comment].
Dewan MJ (1992), Adding medications to ongoing psychotherapy: indications and pitfalls. Am J Psychother 46(1):102-110.
Dewan MJ (1997), Cost of care by a psychiatrist versus split treatment. NR295. Presented at the 150th Annual Meeting of the American Psychiatric Association. San Diego; May 20.
Dewan M (1999), Are psychiatrists cost-effective? An analysis of integrated versus split treatment. Am J Psychiatry 156(2):324-326 [see comments].
Dewan MJ, Pies RW (2001), The Difficult-To-Treat Psychiatric Patient. Washington, D.C.: American Psychiatric Publishing Inc.Goldman W, McCulloch J, Cuffel B et al. (1998), Outpatient utilization patterns of integrated and split psychotherapy and pharmacotherapy for depression. Psychiatr Serv 49(4):477-482.
Himle JA (2001), Medication consultation: the nonphysician clinician's perspective. Psychiatric Annals 31(10):623-628.
Keller MB, McCullough JP, Klein DN et al. (2000), A comparison of nefazodone, the cognitive behavioral-analysis system of psychotherapy, and their combination for the treatment of chronic depression. [Published erratum 345(3):232.] N Engl J Med 342(20):1462-1470 [see comments].
Meyer DJ, Simon RI (1999), Split treatment: clarity between psychiatrists and psychotherapists (Part 2). Psychiatric Annals 29(6):327-332.