A recent study from Emory University in Atlanta, Georgia and published in the Journal of the American Medical Association used PET scans to predict whether clinically depressed patients would respond better to cognitive behavior therapy or medication. Results show that activity in a region of the brain associated with self-awareness and emotions is significantly correlated with response to treatment.
Study participants who showed lower activity in this brain region responded better to cognitive behavior therapy, a skills-based treatment approach that helps patients identify connections between thoughts and feelings. Patients with elevated activity in this region of the brain responded better to treatment with a commonly used depression drug, escitalopram (Lexapro).
I have mixed feelings about this study. One one hand, it represents an exciting step in the direction of developing more sophisticated methods of measurement in the treatment of mental illness. As a science, psychiatry is in its infancy. Ask a physicist to make predictions about the movements of celestial bodies, and she can tell you exactly how, when, and where almost any object in the cosmos will move. Ask a chemist what will happen if two substances are mixed, and he will produce a mathematical formula of the reaction at the atomic level. Ask a psychologist or psychiatrist how a person will respond to any given treatment, and the response will be vague and statistical. We simply don’t have a way to effectively measure psychological phenomena. We need studies like this to help us develop systems of measurement that will eventually allow psychiatry to become a more exact science.
On the other hand, the field of psychiatry is becoming increasingly narrow in its conception of mental illness. As psychiatry becomes aware of its scientific inadequacies, it has adopted a somewhat self-conscious, rigid positivism to compensate. Extreme importance has been placed on the importance of “empirically validating” approaches to treatment and diagnosis. To be sure, empirical data is vital to the field. The problem is that our ability to measure psychopathology and treatment outcomes is severely limited. If we reduce our ideas about treatment of mental illness to the narrowest definition suitable for measurement in a research setting, then I feel we are throwing the baby out with the bathwater. This is most apparent in the symptom-focused approaches being adopted by many health care organizations. To my mind, treatments based solely on this approach place narrow focus on observable and quantifiable symptoms without adequately considering the person or the relational context of the treatment itself. The same symptom can manifest in one person for reasons completely different than in another person. Moreover, people are more than collections of symptoms, and there is much to be said about the importance of the therapeutic relationship.
We live in exciting times. As our ability to understand the physical origins of thoughts and feelings increases, I expect a scientific, cultural, and philosophical revolution that will reach far beyond the field of psychiatry. Just as the development of the heliocentric model of the solar-system rocked Western culture to its core, I think the development of biological models of subjective experience will leave lasting effects on our understanding of what it means to be human. In the meantime, we shouldn’t lose sight of the centerpiece of effective psychotherapy: building healing relationships with people.