Improving Transfusion Safety in the Operating Room With a Barcode Scanning System Designed Specifically for the Surgical Environment and Existing Electronic Medical Record Systems: An Interrupted Time Series Analysis

BACKGROUND

Manual processes for verifying patient identification before blood transfusion and documenting this pretransfusion safety check are prone to errors, and compliance with manual systems is especially poor in urgent operating room settings. An automated, electronic barcode scanner system would be expected to improve pretransfusion verification and documentation.

METHODS

Audits were conducted of blood transfusion documentation under a manual paper system from January to October 2014. An electronic barcode scanning system was developed to streamline transfusion safety checking and automate documentation. This system was implemented in 58 operating rooms between October and December 2014, with follow-up compliance audits through December 2015. The association of barcode scanner implementation with transfusion documentation compliance was assessed using an interrupted time series analysis. Anesthesia providers were surveyed regarding their opinions on the electronic system. In mid-2016, the scanning system was modified to transfer from the Metavision medical record system to Epic OpTime. Follow-up analysis assessed performance of this system within Epic during 2017.

RESULTS

In an interrupted time series analysis, the proportion of units with compliant documentation was estimated to be 19.6% (95% confidence interval [CI], 10.7-25.6) the week before scanner implementation, and 74.4% (95% CI, 59.4-87.4) the week after implementation. There was a significant postintervention level change (odds ratio 10.80, 95% CI, 6.31-18.70; P < .001) and increase in slope (odds ratio 1.14 per 1-week increase, 95% CI, 1.11-1.17; P < .001). After implementation, providers chose to use the new electronic system for 98% of transfusions. Across the 2 years analyzed (15,997 transfusions), the electronic system detected 45 potential transfusion errors in 27 unique patients, and averted transfusion of 36 mismatched blood products into 20 unique patients. A total of 69%, 86%, and 88% of providers reported the electronic system improved patient safety, blood transfusion workflow, and transfusion documentation, respectively. When providers used the barcode scanner, no transfusion errors or reactions were reported. The scanner system was successfully transferred from Metavision to Epic without retraining staff or changing workflows.

CONCLUSIONS

A barcode-based system designed for easy integration to different commonly used anesthesia information management systems was implemented in a large urban academic hospital. The system allows a single user with the assistance of a software system to perform and document pretransfusion safety verification. The system improved transfusion documentation compliance, averted potential transfusion errors, and became the preferred method of blood transfusion safety checking.

Early recognition of possible transfusion reactions using an electronic automatic notification system for changes in vital signs in patients undergoing blood transfusions

BACKGROUND

The study was designed to evaluate the ability of a novel electronic automatic notification system (EANS) to detect significant changes in transfusion-associated vital signs (VSs) during transfusion and to determine whether the EANS improved acute transfusion reaction (ATR) detection rates and suspected ATR reporting rates.

STUDY DESIGN AND METHODS

VSs were measured three times per unit or batch product transfused:-before, 15 minutes after commencement, and at the completion of the transfusion-and recorded on the EANS. Significant changes in VSs were defined as increased temperature (≥38°C or ≥1°C change in baseline temperature), 20 mm Hg or 20% increase or decrease in systolic blood pressure, or 20% increase in pulse rate. The 6-month periods preceding and after the introduction of the EANS were defined as "before" and "after." Data from these periods were used for comparison and evaluation.

RESULTS

During the after period, 945 notifications were reported from the EANS and 521 suspected ATR were detected. The suspected ATR reporting rates for the before and after were 0.29% (73/25 213) and 2.06% (521/25 304, P < .001) and the ATR detection rates before and after were 0.13% (33/25 213) and 0.49% (116/25 304, P < .001), respectively. Among 116 ATR cases, 49.1% could be detected only by significant changes in VSs.

CONCLUSION

The EANS was very effective in detecting ATRs that could have been overlooked by medical staff. Further data are needed to demonstrate the extent to which the introduction of an EANS may improve the safety of transfused patients.