|
|
|
|
|
|
by tyingq
2584 days ago
|
|
|
Having worked in healthcare IT, I'm less astonished than I should be about approximate matches and score thresholds being used to confirm "same entity" for pharma data. What could go wrong? (Not knocking the author...the practice is unfortunately needed because providers won't provide good data) |
|
|
It wasn't quite like that. It was to help the pharmaceuticals work together to "pool" talented researchers who had the required training to proceed in the clinical trials. If a bunch of companies don't share this data, they drive up their costs tremendously in trying to find and recruit qualified doctors for their trials, pay to train them, and then have them conduct studies, only to find out that the doctor in question doesn't recruit any candidates themselves, or fails to report results.
If, however, they pool their resources, they can find qualified doctors who have already taken the training and are known to participate well in trials. They can then correlate that with populations who might benefit from the drugs in question and save a fortune (and possibly many lives), by reducing the cost-to-market.
The matching algorithm was actually very strong, but there was still a human step, from the pharmaceutical companies, to verify that these were the researchers they were seeking.