Effect of a computerized insulin dose calculator on the process of glycemic control.

Background: Glycemic control is important to patients' outcomes. However, the process of maintaining glycemic control is risk laden and labor intensive for nurses.

Objectives: To examine the effects of using a computerized insulin dose calculator to facilitate management of glycemic control for critically ill cardiac patients.

Methods: A prospective randomized controlled trial was conducted with a sample of 300 intensive care patients, 141 randomized to the calculator group and 159 in the control (paper protocol) group. A convenience sample of 44 intensive care nurses responded to a nurse satisfaction survey.

Results: A significantly higher percentage of glucose measurements were in the target range in the calculator group than in the control group (70.4% [SD, 15.2%] vs 61.6% [SD, 17.9%], Z = -4.423, P < .001), and glucose variance was significantly less in the calculator group (35.5 [SD, 18.3] mg/dL vs 42.3 [SD, 21.2] mg/dL, Z = -3.845, P < .001). Fewer hypoglycemic events occurred in the calculator group (7 vs 18), although this difference was not statistically significant. Nurse satisfaction was higher for the calculator group than for the control group (8.4 [SD, 1.4] vs 4.8 [SD, 2.4], Z = -5.055, P < .001). Nurses' deviation from the protocol was also less in the calculator group than in the control group.

Conclusions: Management of glycemic control and nurse satisfaction were improved with use of the dose calculator. Improving nurses' processes of care may improve nurses' use of time and patient care overall. Studies with larger sample sizes over time are needed to determine these relationships.