Hacker News new | ask | show | jobs
by stoorafa 1472 days ago
You’re right that there is non-random allocation to the exposure (taking glucosamine), but this is something that can be addressed to an extent with a bit of statistics.

I was disappointed to see the authors didn’t take the additional step of doing propensity score matching or weighting to account for propensity to take glucosamine.

Here’s a good reference about propensity score matching: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3144483/