Comparative Study of Intravenous Insulin Protocols: Paper-based versus Computer-based

Background: Scientific evidence exists to demonstrate that glycemic control produces a positive outcome for critically ill patients by decreasing mortality and morbidity. Results of published research have revealed a reduction of mortality in critically ill patients when serum blood glucose levels are maintained at a level of less than 150mg/dL. Recommendations from the Surviving Sepsis Campaign for treatment of patients in septic shock include use of intravenous insulin therapy to control hyperglycemia via the use of a validated protocol for insulin dose adjustment. Purpose: The purpose of this study was to determine whether glycemic control (less than 150mg/dL) was attained more effectively using an existing Rochester General Hospital Surgical Intensive Care Unit (RGH SICU) paper-based standardized insulin infusion protocol or a computer-based insulin infusion protocol. A secondary purpose was to determine how the use of each protocol affects the incidence of hypoglycemia. Research Questions: Is there a difference in the incidence of hypoglycemia (\u3c60mg/dL) between the paper-based protocol and the computer-based protocol? Is there a difference between the paper-based protocol and the computer-based protocol in attaining glycemic control (\u3c150mg/dL)? Is there a relationship between clinician adherence and correct insulin dosage? Hypotheses: There will be no relationship between time required to attain glycemic control and protocol used to achieve control. There will be no relationship between the incidence of hypoglycemia and protocol. There will be no relationship between clinician adherence and correct insulin dosage. Method: A quantitative, combined retrospective and prospective design was used to review medical records of 60 patients admitted to the SICU at RGH. For the retrospective part of the study, 30 charts were reviewed from 2006, which was prior to the development of the computerized insulin infusion protocol. The prospective part of the study began after the StatStrip Xpress Glucose Hospital Meters had been implemented by RGH. For the retrospective data collection, patients\u27 blood glucose measurements were monitored via the use of the Roche Advantage Accu-Chek Glucometers and Accu-Chek Comfort Curve test strips. The Pharmacy Department supplied the medical record numbers of these patients (n ~ 30). Data collected included age, gender, weight, prior history of diabetes, presenting illness/surgery, serum glucose on admission to SICU, use ofvasopressors, use of corticosteroids, use of total parenteral nutrition (TPN) or gastric tube feedings, use of mechanical ventilation and Acute Physiology and Chronic Health Evaluation (Apache II) scores related to illness severity. Analysis: Using SPSS 18.0, descriptive statistics were calculated, including measures of central tendency, shape of distributions, and measures of variability. Spearman\u27s correlations were used to determine the relationships between time required to attain glycemic control and protocol used to achieve control, incidence of hypoglycemia. Results: The rate of hypoglycemia for patients receiving the paper-based protocol was 10% and 3% for patients receiving the computer-based protocol. Both protocols achieved glycemic control within six hours of initiation of the intravenous insulin protocol. There was a moderately strong relationship (Spearman\u27s rho~ .667, p ~ 0.01) between clinician adherence and correct insulin dosage with the computer-based protocol, and a moderate relationship (Spearman\u27s rho~ .424, p ~ 0.05) between clinician adherence and correct insulin dosage with the paper-based protocol. Conclusions: The findings of this study did not support the hypothesis that a computerbased protocol is more effective at attaining glycemic control than a paper-based protocol. Although the difference in rate of hypoglycemia was not statistically significant between the paper-based protocol and the computer-based protocol, it was thought to be clinically significant for individual patients\u27 glycemic control. Dissemination: Study outcomes were presented to the faculty and students at Wegman \u27s School of Nursing at St. John Fisher College, the Nursing Research and Evidence-Based Practice Department, and the staff and management of the SICU at RGH. Potential use/impact at RGH: Both the Pharmacy and Nursing Departments at RGH held stakes in adopting an insulin infusion protocol as a standard of practice. Positive outcomes of this study includes improved glucose monitoring, improved patient outcomes, and decreased SICU length of stay (LOS). Potential National Contribution: Outcomes of this study may reinforce existing evidence related to glycemic control. Hospitals and insurers may benefit financially from improved patient outcomes and decreased LOS