ORLANDO — The use of a computer protocol to achieve tight glycemic control lowered insulin administration errors, compared with a paper-based protocol, according to a study of simulated patients in an intensive care unit.
The computer format also improved satisfaction in ICU nurses, Dr. Anthony Y. Lee of Columbus Children's Hospital in Ohio said at the annual congress of the Society of Critical Care Medicine.
Dr. Lee and colleagues at the University of Maryland Medical Center, Baltimore, recruited 51 medical ICU nurses to complete seven simulated patient scenarios using both the standard paper-based insulin protocol and a computer version of the protocol.
The scenarios included a clinical case description, a current insulin dose, and new and previous blood glucose levels. The nurses were given instructions on using the paper and computer versions of the protocol and had to indicate the new insulin dose and time of the next blood glucose check.
The simulated situations produced 357 paper responses and 357 computer responses showing a significant reduction in errors using the computer format. Use of the paper protocol resulted in 82 insulin-dosing errors, compared with 4 errors using the computer system. It seemed that the same study participant committed all four errors using the computer protocol.
Errors in the timing of the next blood glucose check fell from 47 with the paper-based format to 8 with the computer format. The time to completion fell from 9 minutes with the paper-based format to 6 with the computer program.