eicu-code
eicu-code copied to clipboard
MIT-LCP
vasopressor dosage
Searching through the infusiondrug table is a bit puzzling. Some infusions have rates entered in the name while others don't. For e.g. drugname = 'Epinephrine ()' just has a number under drugrate, but there's no unit attached. What can we infer here?
Good question... if a continuous infusion is documented directly in the eICU software, the users must explicitly define the drug concentration and therefore you will have both the drug rate and pump rate and be able to calculate dose based on the concentration provided. However, some of the infusion interface messages that originate from the EHR at the bedside will only send the pump rate which is defined in mL/hr. So if it's a drug like propofol which is almost always 10% concentration, it will be easy to determine drug amount given based on ml/hr.. But for other drugs, you may only be able to establish a range based on standard concentrations used. In some cases, the concentration may be available in the drug name from the medications table so you can use that for your calculation.
On Wed, Mar 6, 2019 at 10:28 AM Rishikesan Kamaleswaran < [email protected]> wrote:
Searching through the infusiondrug table is a bit puzzling. Some infusions have rates entered in the name while others don't. For e.g. drugname = 'Epinephrine ()' just has a number under drugrate, but there's no unit attached. What can we infer here?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/MIT-LCP/eicu-code/issues/68, or mute the thread https://github.com/notifications/unsubscribe-auth/AP1JVK_7MXNgYLF6tQAGEgQUiE9MUZq9ks5vT96xgaJpZM4bhJwj .
Great question. I just had a similar problem with trying to use vasopressin data and realised the same issue. The dosage is critical as we purposefully make double strengths to lessen the volume load. Also not every hospital make same drug infusions (e.g 4 mg noradrenaline in 50 ml saline or 4 mg in 100 ml saline bag) in Australia.
I would love to know what other researchers did with their study! (Or if there is a proper way to match the variables to make all sense that I didn’t realise, please let me know too. Any help is gold here!)
Dropping back on this older post, anyone have any solutions? I'm trying to clean up vasopressor data and are seeing many clear errors in transcription that aren't real doses and I'm not sure how to clean these up easily......
