AHRP |
InfoMail |
|
| MediaCoverage
News Stories on Human ResearchProtection and March 1, 2002 Anti-Depressant Drug Trials are Rigged FYI “When you take any medicine you assume it’s been foundto be effective for your condition.” But a Brown University analysis of 31 antidepressanttrials published from 1994 to 1998 in five leading psychiatric journals, foundthat between 60% and 85% of patients who are most likely to be prescribedantidepressants were excluded. Such a selection process ("cherrypicking") inevitably skews the results, thereby undermining the publishedfindings and claims about the efficacy of antidepressants. "If antidepressants are, in fact, not effective forsome of these large subgroups of depressed individuals, their prescriptionincurs an unjustifiable exposure of risks and side effects, and alternativetreatments need to be considered." Treating Depression Antidepressant drug trials turn away most of the depressedpopulation Studies establishing the effectiveness of antidepressantsare based on highly selective samples of depressed patients. New research byBrown University psychiatrists found as many as 85 percent of depressed patientstreated in an outpatient setting would be excluded from the typical study todetermine whether an antidepressant works. Trials to determine the effectiveness of antidepressantshave historically evaluated only a small subset of depressed individuals with avery specific clinical profile. People diagnosed with other psychiatric problemsand people with mild depression are among those excluded, says the study, whichappears in the March 2002 American Journal of Psychiatry. “When you take any medicine you assume it’s been foundto be effective for your condition,” said Mark Zimmerman, associate professorof psychiatry and human behavior, director of outpatient psychiatry at RhodeIsland Hospital, and the study’s lead researcher. “No one knows for surewhether antidepressants are effective for most of the patients we treat.” As few as 15 percent of 346 depressed patients evaluatedin the Rhode Island Hospital Department of Psychiatry outpatient clinic wouldhave met the eligibility requirements of a standard drug trial, depending on thecriteria. To determine whether the clinic patients would qualify forthe drug studies, the researchers reviewed inclusion and exclusion criteria usedin 31 antidepressant trials published from 1994 to 1998 in five leadingpsychiatric journals. Many of the studies excluded patients who had psychoticfeatures, a history of manic episodes, suicide risk, unstable medical illnesses,or a history of drug or alcohol abuse. Several also excluded subjects witheating disorders, obsessive-compulsive disorder or panic disorder. Nearly all of the studies excluded patients who fell belowa cutoff score on a measure of symptom severity, even though they were diagnosedwith major depression. “We are not aware of any other medical condition inwhich individuals with the disorder are routinely excluded because they are notsick enough,” said Zimmerman. “Drug companies are concerned that individuals with milddepression will respond just as well to a placebo as they will to antidepressantmedication,” said Zimmerman. “However, this represents a sizable number ofindividuals who are prescribed these medicines, especially by primary carephysicians.” The researchers conducted diagnostic evaluations ofpatients at the Rhode Island Hospital Department of Psychiatry outpatientpractice, a group ranging in age from 16 to 65. Excluding patients with any ofthe features commonly used in efficacy trials eliminated two-thirds of thepatients. If patients with an anxiety disorder were also excluded, then morethan 85 percent of the patients would not have qualified for a drug study ofantidepressants – yet more than 90 percent of the patients in the study forwhom prescribing information was available were being treated withantidepressants at the time of the evaluation. Some extrapolation of antidepressant studies by clinicianswill always be necessary, Zimmerman said. It would be impossible to establishthe effectiveness of antidepressant medications in every conceivable populationof depressed patients. But the current practice of limiting studies to only “pure”moderate-to-severely ill depressives may skew the findings of drug trials, headded. If antidepressants are, in fact, not effective for some of these largesubgroups of depressed individuals, their prescription incurs an unjustifiableexposure of risks and side effects, and alternative treatments need to beconsidered. Opening antidepressant trials to patients with a widerrange of symptoms would allow researchers to learn whether any specific subsetsrespond or do not respond to a drug. The question now is whether governmentmandates are necessary to make trials more inclusive, Zimmerman said. There islittle motivation for drug companies – whose primary aim is to show that theirmedication is safe and that it works for some patients – to do this. “Drug companies have been correct in assuming that ifthey show their medicine works for a highly select group of depressed patients,physicians will use it for all patients,” said Zimmerman. This study was supported by grants from the NationalInstitutes of Mental Health. Zimmerman worked with Jill I. Mattia, clinicalassistant professor, and Michael A. Posternak, research fellow, in theDepartment of Psychiatry and Human Behavior at Brown University. FAIR USE NOTICE: This may contain copyrighted (©) material the use of which has not always been specifically authorized by thecopyright owner. Such material is made available to advance understanding ofecological, political, human rights, economic, democracy, scientific, moral,ethical, and social justice issues, etc. It is believed that this constitutes a’fair use’ of any such copyrighted material as provided for in section 107 ofthe US Copyright Law. In accordance with Title 17 U.S.C. Section 107, thismaterial is distributed without profit to those who have expressed a priorgeneral interest in receiving similar information for research and educationalpurposes. For more information go to: http://www.law.cornell.edu/uscode/17/107.shtml |