I don't know who needs to hear this today, but the placebo effect probably isn't real and mostly describes common statistical failings of scientific research
-
-
Lots of people asking, so here's the basic explanation. For subjective outcomes (i.e. pain) there's some evidence that placebos have a benefit. However, this is problematic because it could be due to reporting bias
Show this thread -
This means that people could be reporting that they think they've improved when, objectively, they haven't Which is exactly what we see when we look at objective outcomes
Show this thread -
So, for example, you might find that people with hypertension report feeling better after taking a placebo, but they don't actually have lower blood pressure
Show this thread -
Or people may report a reduction in pain, but they don't take fewer pain relief medications despite the placebo subjectively improving their response
Show this thread -
This is all summarized very well in a 2001 systematic review in NEJMhttps://www.nejm.org/doi/full/10.1056/NEJM200105243442106 …
Show this thread -
Many of the claimed benefits to the placebo effect (blue pills work TWICE as well as red!) are based on surprisingly terrible research from decades ago If you want to know more, I strongly recommend listening to
@mikehall314 on Skeptics With A K, very comprehensive reviewsShow this thread
End of conversation
New conversation -
-
-
This idea was a bombshell for me when I first learned of it. I think the placebo effect holds up pretty well in some domains though. Hard to imagine 2.6% weight loss being attributable to regression to the mean. http://doi.org/10.1056/NEJMoa1411892 …pic.twitter.com/FeO2fswOpN
-
Regression to the mean perhaps not, but there are also study effects to consider. Most people don't have a study nurse following up their weight and HbA1c every week for a year!
- Show replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.