Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

On the 17 item HAM-D here are the typical score ranges [1]

Not depressed: 0–7 Mild (subthreshold): 8–13 Moderate (mild): 14–18 Severe (moderate): 19–22 Very severe (severe): >23

So, a 4 point difference vs placebo is what I would consider a "clinically significant" improvment, e.g. about the difference between mild vs. moderate, or moderate to severe.

https://en.wikipedia.org/wiki/Hamilton_Rating_Scale_for_Depr...



based on a look at the actual contents of the HAM-D, four points isn't exactly an impressive improvement for a $11,000 treatment

a four point improvement could be a person stops fidgeting during the interview

daily reminder that psychology is the worst science in terms of the replication crisis


If you think that’s bad, go look up some SSRI meta analyses. The average HAM-D improvement vs placebo is only about 2 points.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: