Henry: I had forgotten all about your earlier explanation about why the placebo may have been 'active.' To be honest, I am quite skeptical about this explanation, as I was when you first mentioned it. I suppose we could give the company the benefit of the doubt, but we seem to be doing that an awful lot lately.
I am in the health science profession and am familiar with hundreds of clinical studies. I do not recall one study where the placebo component was thought to be active. Lidakol is a very simple molecule. It should be a simple matter to construct a placebo by just eliminating the addition of Lidakol. Again, we could just give the company the benefit of the doubt.
Thus, when I said that "the best explanation as to why Lidakol and placebo show the same time to healing is that Lidakol is ineffective", I am assuming that the placebo is ineffective in the first place. If the placebo is active, then your statement is correct.
Finally, I believe you are suggesting it is okay to make a direct comparison between the 4-5 days of healing with Lidakol with the 8-10 days it 'normally' takes an oral lesion to heal. This direct comparison can only be made if all of these patients were enrolled within the same clinical trial with the same inclusion criteria . Otherwise, you just cannot make any scientific comparison.
For instance, say I am a famous dermatologist practicing at Stanford University. Being a tertiary center, I am likely to get a lot of patients that are difficult to treat by the community physicians. The easy patients will not be referred, but the ones with herpetic outbreaks every 3 weeks will be sent to me. There may be a lot of immunocompromised patients (e.g with HIV) that have difficult to treat Herpes as well in my practice. In my practice, an untreated lesion could take 8-10 days, or even 15 days on average to heal. I publish my findings, and lo and behold, these numbers become dogma.
On the other hand, a community dermatologist in Des Moines, Iowa could have a bunch of patients with mild outbreaks every 2 years. These patients take 4-5 days to heal untreated.
Thus you have to construct a trial in which all arms (whether active drug, placebo, or no treatment) are matched for age, severity of outbreaks, number of outbreaks, etc. Only then can one make valid comparisons.
If Lidak really wanted to prove their assertion that the placebo may be active, they should contruct a trial with 3 treatment arms: Lidakol, placebo, and no treatment. |
