On 10 Mar 2010 at 22:15, Jim Clark wrote, referring to the claim 
that anti-depressants may do poorly in placebo trials because  
the placebo effect is getting stronger.

> I'm not sure it does make sense.  If the placebo effect is getting stronger, 
> then shouldn't the med effect
> also get stronger because of its placebo effect?  That is, med effect is 
> hypothesized to be combination of
> placebo + med, hence the rationale for the placebo control group.  So you 
> have something like: 
> 
> placebo(1990) + med versus placebo(1990) in 1990
> 
> and
> 
> placebo(2010) + med versus placebo(2010) in 2010
> 
> I do not see how this can lead to a narrowing of the med effect?

Interesting point.

As Jim notes, the overall therapeutic effect of a medication 
consists of two parts: a true drug effect plus its placebo effect. 
So if the placebo is getting stronger, then the overall medication 
therapeutic effect ( = drug + placebo) should also get stronger.

Perhaps the paradox can be resolved in one of two ways.

First, if the true drug effect remains constant over time, but the 
placebo contribution has increased, then placebo now makes up 
a greater proportion of the overall therapeutic effect. So the 
difference between medication and placebo conditions would 
get smaller.

e.g., in arbitrary units:

Before:
overall therapeutic effect [OTE] (10) = drug (5) + placebo (5)
placebo effect (5) vs OTE (10) = 50% of OTE

After:
OTE (15) = drug (5) + placebo (10)
placebo effect (10) vs OTE (15) = 67% of OTE

So the difference between placebo and OTE has decreased 
over time.  However, this would also imply, as noted above,  that 
OTE has also increased, and the drug is now more effective 
than initially (when compared with no drug, not with placebo).  I 
wonder if there are any data to support this.

Second, what if the OTE remains constant but the proportion 
contributed by placebo increases? This would imply that the true 
drug effect declines over time.

In this case:

Before
OTE (10) = drug (5) + placebo (5)
placebo effect (5) vs OTE (10) = 50% of OTE

After
OTE (10) = drug (3) + placebo (7)
placebo effect (7) vs OTE (10) = 70% of OTE
So again the difference between  placebo and OTE would 
decrease over time.

In both scenarios of increased placebo action over time, the 
difference in effectiveness between the drug and placebo 
conditions would diminish, and thus the likelihood of finding a 
significant difference in a drug trial.  

 Right?

Stephen

--------------------------------------------
Stephen L. Black, Ph.D.          
Professor of Psychology, Emeritus   
Bishop's University               
e-mail:  sblack at ubishops.ca
2600 College St.
Sherbrooke QC  J1M 1Z7
Canada
-----------------------------------------------------------------------

---
You are currently subscribed to tips as: [email protected].
To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=1210
or send a blank email to 
leave-1210-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu

Reply via email to