3

I was using this spray meant to hasten recovery from the common cold. A study [1], sponsored by the pharmaceutical company that produces the spray, stated it reduced virus by 92% (this value only appear on the promotional material), and so shortens the duration of the common cold.

Meanwhile a doctor told me that these kinds of things don't work.

I found some "similar" studies [2,3, 4, 5 ,6] but I lack the competence to assert their value.

Does someone have some perspective on this? Are these believable studies? Was it well designed, and has it been duplicated?

If so, why wouldn't the pharmaceutical produced a certified and expensive drug instead of a no-prescription drug? I would think anything that would hasten common cold would get an enormous buzz (and profit...).

[1] https://clinicaltrials.gov/ct2/show/NCT01944631 [2] http://www.mrmjournal.com/content/9/1/57 [3] http://www.medscape.com/viewarticle/819411 [4] http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2923116/ [5] http://www.ncbi.nlm.nih.gov/pubmed/25411637 [6] http://respiratory-research.com/content/14/1/124

2
  • 1
    It is preferable that you edit your original question if you have new data rather than that you ask the question again. – anongoodnurse May 4 '15 at 4:19
  • I thought about that. But then I can't mark the answer given there as correct, which is unfair to those who answer it. I also made a link here in the comments. Should I still edit one of them? – Fernando César May 4 '15 at 9:45
2

One might wonder why they gave an ambiguous and questionably relevant figure like "reduces virus by 92%" rather than saying "people recovered more quickly" or something like that. The answer is that in reference 1, which used real-world measures of cold severity and duration, the drug was a total failure. You can see this by looking on the results tab at the "95% confidence interval." This is intended to represent a range of plausible values for the true effect of the drug, extrapolating from the sample of people that was studied (it gives a range of values because you can never know the exact value -- due to random chance, the people who were included in a study won't perfectly represent the entire population of people who might use the drug). By scientific convention, if the 95% confidence interval includes zero, the evidence isn't strong enough to say that the drug has an effect.

The study looked at several outcomes, including patients' ratings of their symptoms, their impression of whether the drug worked, and the duration of their cold. For the first four outcomes, the confidence interval includes zero. For #5, duration of the cold, they don't report that kind of confidence interval, but they report a p-value which tells you the same thing (if p>.05, then by convention the evidence is not strong enough). For #6 they used a test where the critical value for the confidence interval is 1 and not 0.

In other words, the clinical trial of this particular drug gives us little reason to believe that it works. That doesn't mean that it doesn't work, but they certainly haven't made a convincing case.

3
  • I like your input, but it seems you only address the first study. Is that right? – Fernando César May 4 '15 at 9:50
  • The answer only addresses the first study because (if I understood the poster correctly) it's a clinical trial of the specific product in question. I assumed that the others are similar products, which would be less relevant. – octern May 4 '15 at 10:19
  • The first was is the one that supports the 92% claim (I think). I'm not sure if the other studies agree or not... – Fernando César May 4 '15 at 13:57

Not the answer you're looking for? Browse other questions tagged or ask your own question.