Hacker News new | ask | show | jobs
by _heimdall 844 days ago
I'm a little rusty when it comes to statistical analysis, but is 168 total participants ranging from 1 to 56 years old really enough for any meaningful level of statistical significance?

Sure the efficacy rates seem really promising here, but that's an extremely small group and if there are any concerns over age-based factors the cohorts are even smaller.

Add the fact that this medication is meant for long term use and I really don't know how we can claim any level of certainty with regards to long term safety. The study only followed the 168 participants (including placebos) for 16-20 weeks.

1 comments

It went from like 6% to 68%, so yeah it’s statistically significant even at low numbers like 168.