Hacker News new | past | comments | ask | show | jobs | submit login

Here's a figure from the article posted by Eric Topol on twitter: https://twitter.com/EricTopol/status/729688822053670912

Combined with the stat you mentioned, it's a strange one. Maybe I'm looking at it wrong, but there's barely any physicians receiving above $50k in payments, so the biggest impact pharma payments have made is around a 5 % difference in prescribing. Looking at the variation at around $0-$10k, 5 % seems like a tiny change, since you have docs prescribing anywhere from 0 % and 80 %. Something else is driving it, my guess would be good personal relationships with sales reps? In that case, focusing on limiting that would have a much better impact on fixing the issue.

The big variation could also be something completely innocent like trust in a specific brand due to experience in the clinic, etc. Perhaps someone more knowledgeable can chime in.




Nothing quite like extrapolating a line across your entire data set when 99% of the data points are in the bottom 5% of the range.




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

Search: