was your dissertation work also done at a 'Cancer Treatment Place'?
if so, there'd be an increased probability that your samples actually constitute
a population that's biased, and that inherent bias must be explicated before the
results of your study can be resolved.
it might be the thing that biases the population toward needing treatment for
Cancer that is the thing that determines the seemingly-'contradictory' (correct
me if i'm wrong in that) results you obtained.
that's the 'bad news'.
the 'good news' is that, if, in fact, there is such an in-study population-bias
that derives in the fact that the population is taken from folks who're all
under treatment for Cancer, then the stuff of your experimental results becomes,
=at least=, a useful 'prying-bar' with respect to the disease of Cancer.
can you see it?
it's possible that, if your experimental population is biased away from the
general norm, then, in your experimental results, you have something that can
serve as an analogue of Pasteur's petrie dish.
'course, it's more-likely that that's not the case, but if there was such an
experimental-population bias, then the possibility is there, and it must be
considered, and followed-up.
so was the experimental population selected entirely from Cancer patients?
Steve Palmer wrote:
> My name is Stephen Palmer, and I am a postdoc at the Dept. of Pain
> Research at the M.D. Anderson Cancer Center, having recently defended my
> dissertation at SUNY Stony Brook.
> I am writing the group because my dissertation produced results I can't
> explain, and I am wondering if any of you have any suggestions about what
> may have occurred . My dissertation addressed the subject of placebo
> analgesia and endogenous opioids. Previous research has found that placebos
> can alleviate various kinds of pain, but that naloxone (an opioid
> antagonist) can reverse or weaken placebo analgesia.
> My study was intended to be a replication and extension of the previous
> research. I used a 2x2 design in which all subjects received an intravenous
> infusion, but half were told that they would receive a potent pain reliever,
> while the other half were told that they would only get saline. In
> actuality, some subjects received saline while others got naloxone. Cold
> pressor tests (a pain tolerance task) were given before and after the
> What I expected what that subjects who were told to expect a pain
> reliever but who got saline instead would show a reduction in pain (i.e.
> placebo analgesia). Subjects who got the pain reliever message plus
> naloxone, on the other hand, would show no change in pain, or at least a
> smaller reduction than the pain reliever/saline group. The groups who were
> told they were getting only saline (and really got either saline or
> naloxone) would show no change.
> This didnt happen. Instead, only the pain reliever/naloxone group
> showed pain relief! The other three groups showed either no change or
> increased pain. This finding is unlike anything in the prior literature,
> and I have no explanation for it.
> My question to you is, is there any literature out there which can
> explain these findings? Are there any other studies, perhaps unpublished
> ones, which have found the same results? I would appreciate any input you
> could give.