The trusted source for
healthcare information and
Drug Blues: Antidepressants, Efficacy, and Effectiveness
Abstract & Commentary
By Russell H. Greenfield, MD
Synopsis: Upon evaluating data from the STAR*D project, researchers concluded that phase III clinical trials often do not recruit representative populations of depressed outpatients, making the recommendations drawn from the studies' conclusions of limited applicability to general clinical practice.
Source: Wisniewski SR, et al. Can Phase III trial results of antidepressant medications be generalized to clinical practice? A STAR*D report. Am J Psychiatry 2009;166:599-607.
Data from the STAR*D project, a large multisite, prospective, sequentially randomized clinical trial of outpatients aged 18-75 years with nonpsychotic major depressive disorder (MDD), were assessed by researchers interested in the generalizability of findings of studies on treatment of MDD. STAR*D was initially designed to help define which therapeutic interventions might be most effective for outpatients with nonpsychotic MDD with a history of suboptimal clinical outcome to initial therapy. Standard demographic and self-reported psychiatric history were collected at baseline, together with data from a number of validated tools to assess severity of depression, including the HAM-D, 16-item Quick Inventory of Depressive SymptomatologyClinician Related, the Quick Inventory of Depressive SymptomatologySelf-Report, and the Psychiatric Diagnostic Screening Questionnaire. Measures obtained were used to estimate the presence of atypical, anxious, and melancholic symptoms. A representative SSRI was chosen (citalopram) and administered at an initial dose of 20 mg/d, which was increased to 40 mg/d by week 4 and then again to 60 mg/d by week 6. Study protocol instructed patients to make clinic visits at weeks 2, 4, 6, 9, and 12 (with an optional visit at week 14), and for practitioners to make management decisions at weeks 4, 6, 9, and 12 after enrollment. Primary outcome measure was based on the self-rated Quick Inventory of Depressive Symptomatology, with remission defined as a score of 5 or less (equivalent to a score of 7 or less on the 17-item HAM-D) at week 9. The secondary outcome was response, defined as a reduction of 50% or more from baseline score on the self-rated Quick Inventory of Depressive Symptomatology at the last assessment.
A total of 22.2% of the enrolled subjects met typical entry criteria for phase III clinical trials (efficacy sample) and 77.8% did not (non-efficacy sample). The efficacy sample had a shorter average duration of illness as well as lower rates of family history of substance abuse, prior suicide attempts, and anxious and atypical symptom features, yet more of the subjects in this group were seen in psychiatric speciality care clinics. Efficacy participants also tolerated citalopram better, though there was no significant difference between the two groups with respect to dosage used. In addition, they had higher response (51.6% vs 39.1%) and remission (34.4% vs 24.7%) rates. Participants in the efficacy sample were also likely to be younger, more educated, white, non-Hispanic, employed, married, and privately insured, and to have a higher income. The researchers concluded that phase III clinical trials may not recruit representative depressed patients, and the findings of these studies may not reflect characteristics or treatment responses of the general population. The outcomes of phase III trials for the treatment of MDD may thus be more optimistic than results commonly seen in clinical practice.
There has long been a debate within medical circles regarding efficacy (high on internal validity at the expense of generalizability and favored by researchers) and effectiveness (high on external validity at the expense of careful controls, and preferred by most practitioners).1 The findings of well-controlled trials may not easily translate into clinical outcomes in "real world" patients who frequently present with comorbidities and other confounding factors. And yet, approval of a new antidepressant drug requires at least two phase III clinical trials to demonstrate safety and (apparently) presumed effectiveness. Such trials employ strict inclusion and exclusion criteria that typically exclude a substantial portion of the broader population of depressed patients. In the setting of depression, where placebo response is recognized to be high, how is a clinician to interpret the results of phase III trials when facing a patient in need of help? This question has no easy answer; in fact, the authors posit that the research methodology employed in phase III trials, at least in addressing MDD, may need to be revisited.
Is it not surprising, then, that at a time when questions arise about the clinical relevance of phase III trials, when questions abound regarding appropriate research methodology in integrative health care (quantitative vs qualitative, for example), and when debate rages on the topic of what constitutes an evidence basis supporting a given therapy, that some of our patients question our recommendations? As practitioners we are sometimes left to ponder how we know what we know, as well as what do we really know? What's new is never really new: "Teach thy tongue to say I know not and thou wilt progress" (Maimonides).
1. Stricker G: The relationship between efficacy and effectiveness. Prevent Treatment 2000;3: article 10.