March 1, 2012
When doctors ditch pens, medical errors drop
The title alone just makes sense, the type of thing that you don’t really need a study to find. I find the description of what this software actually does even more interesting.
The computer system studied here reviewed the patient’s current medications, alerted the prescriber to any potential conflicts, and then sent the order off, electronically, to the pharmacy
Wow, even less in need of a study. Of course, that’s more because of the title of the blog post. The importance of the study is to find how much more accurate. The post says errors drop 60% and the NPR blog post says that the software UI still needs work, so that number should get better.
I was still curious what the 60% meant. It’s a good number, but if there were 1,000 errors before vs 10 errors, that’s a big difference. I tried to read through the study. During the study there were 1,923 admissions, and 11,168 errors. That’s 5.8 per person! That sounds crazy, no? Thankfully the serious rate was less than 3 serious errors per thousand admissions.
I’d say that 60% is a strong number. The study to do is which is more cost effective, this software or giving everyone in the hospitals lessons to improve their handwriting.