| 2004N-0330 - Joint Meeting of the Psychopharmacologic Drugs Advisory Committee and the Pediatric Advisory Committee|
|FDA Comment Number :||EC49|
|Submitter :||Ms. Eileen McGinn||Date & Time:||10/06/2004 04:10:04|
| PRESENTATION TO FDA AT THE PUBLIC HEARING ON SSRIs AND SUICIDALITY IN CHILDREN
September 13, 2004
The current process for the approval of drugs for the American market is scientifically flawed. Science requires reliability and validity. The trials for many psychotropic drugs are not reliable, nor are they valid.
Reliability in science requires that research be independently replicated in different labs by different researchers. In the case of drug trials, the same firm producing the drugs conducts all trials. This is an inevitable source of bias, and a breech of the scientific method.
There are several problems with respect to validity of drug trials. First and most important, people in a research study are supposed to represent the general population with the illness. Most trials systematically exclude many people, especially those with severe illness. The group studied does not represent the population with the illness, so we cannot generalize the results from these selected groups.
Second: Outcome measures are not clinically relevant. There is no blood test or brain scan to mark the presence of psychiatric illness, so researchers use rating scales similar to questionnaires to measure the severity of symptoms. On a scale of 1 to 20, clinicians may agree that a score under seven demonstrates wellness, while a score over seven shows that the person is ill. It would seem logical to use this cutoff score of seven to sort out the responders from the non-responders. However, this direct, simple method is rarely used. Instead, researchers use a mathematical formula based on a percent decrease in scores. A person may have a decrease in his score from 20 to 10, so he is grouped with the "responders" in the trial. Realistically, with a score of 10, he is still quite ill.
Third: The drop out rate for many trials approaches 50%. A drop out rate this high biases the results and calls into question any conclusions of the trial. The information from the dropouts, like adverse events or non-response, is rarely analyzed or reported. Scientific methods exist to deal with the dropout data, but they are rarely reported in the trial results.
Fourth: Trials are usually small and short. Small size means that less common adverse events are not captured. Short duration means that delayed harmful effects are not known by the end of the trial.
Fifth: Trials are not valid because the double blind is easily broken. All psychotropic drugs have obvious physical effects so that the participants and the researchers know who is getting the drug and who the placebo.
Sixth: There is no standard way to report harmful, adverse events. For example, similar events may be reported in different trials as activation,
| excitement, agitation or hypomania. This makes it difficult to determine the actual incidence of adverse events and to compare them across studies.
Compounding the basic scientific errors, these limitations of clinical trials are rarely discussed, even in the medical literature. Results of drug trials are promoted in an unscientific and careless way, and harmful effects of drugs are minimized, dismissed or denied.
In conclusion, the American people deserve a fair trial before a drug is approved for our market. We want the whole truth about drug trials and drugs.