- Home>
- Specialty Care>
- Psychiatry>
- Top Story
Does STAR*D Make the All-Stars?
Researchers try to evaluate depression treatments by emulating real-world clinical decision-making.
With federally funded "big science" studies, evidence-based psychiatry is coming of age in a manner analogous to other medical specialties with studies of diabetes mellitus, hypertension, coronary heart disease, and cancer. One of the largest and best-known psychiatric studies is the multisite Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study, which concerns the treatment of real-world patients with major depression, including many with comorbidities. Key results of this four-wave study have appeared in major journals (see Journal Watch Psychiatry Apr 5 2006; Jul 17 2006; Oct 16 2006).
The acute treatment protocol, which initially enrolled more than 4000 patients, mimicked what many clinicians do sequentially in practice. Step 1 involved citalopram treatment. If that didn’t work (i.e., lack of remission or drug intolerance), patients entered step 2. Either they switched to sertraline, bupropion sustained-release, or venlafaxine extended-release, or they received citalopram augmentation with bupropion SR or buspirone. If these regimens didn’t work, patients entered step 3. They switched to either mirtazapine or nortriptyline, or while maintaining their step 2 agent, they received augmentation with either T3 or lithium carbonate. Finally, if these didn’t work, patients switched to tranylcypromine or to a combination of venlafaxine XR and mirtazapine (step 4). After step 1, some patients chose cognitive therapy alone or as augmentation to citalopram; if these patients continued, they were then offered the same medication choices as in step 2 and, later, in step 3. Obviously, the number of patients in each cell fell substantially at each step.
The most recent report (Am J Psychiatry 2006 Nov; 163:1905-17) summarizes the overall results. Remission rates for the four acute-treatment steps, respectively, were 36.8%, 30.6%, 13.7%, and 13.0%. The overall cumulative remission rate was estimated to be 67%. The authors note that when more treatment steps were required, lower remission rates were seen, underscoring the need for future studies to identify the best multi-sequenced treatments.
One strategic aspect of STAR*D’s design is that its protocols were primarily focused on medications rather than on patients. Are alternative strategies possible? For example, several thousand patients with complex major depressions similar to patients recruited in STAR*D could be enrolled into a "practice research network." Given even freer rein to intervene as each patient’s characteristics dictate, clinicians could explicate their decisions clearly, justifying each sequential step in relation to that patient’s characteristics. Outcomes would be measured yearly for as long as possible. In such a study, we might find clinicians even more inventive with medication choices, e.g., atypical antipsychotics, stimulants, other agents, or drug combinations. In addition to receiving medications, some patients might be treated with ECT, therapeutic light, vagal stimulation, individual therapy, couples therapy, physical exercise — or combinations of these. Could remission rates in such a study be higher than 67%?
STAR*D, its many authors, and its many more contributors have given us a marvelous study, and we have learned a lot. Overall results regarding remission indicate modest improvement, reflecting both the seriousness of this mood disorder and the limitations of available treatments. To meet the treatment challenges of major depression, we have many options, which may include designing studies with even greater complexity and pointing our telescopes at different types of STARs.
— Joel Yager, MD
Published in Journal Watch Psychiatry December 29, 2006