The Impact of Bar Coding on Nurses' Time

Sunday, November 1, 2009

Andrew Hanson, RN, BSN, CCRN-CSC
Surgical Intensive Care Unit, Froedtert Hospital, Milwaukee, WI

Learning Objective 1: Understand the differences in medication administration time pre to post bar coding.

Learning Objective 2: Identify factors influencing medication administration time.

Background and Significance:  Nurses spend up to 40% of their work time administering medications. It has been proposed that technological advancements, such as use of bar-code medication administration (BCMA), could contribute to decreasing the risk of medication errors. Research focusing on implementation of BCMA technology has occurred concomitantly with implementation of an electronic medication administration process. The results of these studies indicate technology has resulted in positive patient outcomes but longer administration times. No study has examined the implementation of BCMA with an electronic medication system already in place.
Purpose:  The purpose of this study is to evaluate the impact of BCMA on medication administration time. This is one component of a larger study which included patient and nurse satisfaction and med errors.
Method: A pre-post descriptive observational design was conducted during a two week time period with observations obtained prior to initiation of bar coding and 3 months later.
Sample Description: Medication administration episodes were observed and timed. An episode was defined as a nurse giving medication(s) at any singular point in time.
Setting: This study took place on two medical/surgical units averaging 10,067 and 15,492 medications per month.
Procedure: Written consent was obtained for every nurse and patient participating in an observation. An observation protocol was developed and observers were trained and periodically monitored for adherence to protocol. A standardized observation form was used. Time was measured via a stop watch.  Each medication administration episode was then entered into Access Database and transferred to SPSS 14 for analysis.
Results/Outcomes: Actual Medication Administration Time was calculated for each individual observation allowing comparison between groups by statically controlling for the number of medications in each administration episode. Medication administration time increased (1minute 15 seconds) post bar coding using a two tailed T test for independent samples. (T test -9.27; DF= 382; P value = 0.00)  Medication time was confounded by med type, route, preparation, and status of med order (new or ongoing).  Logistic regression was used to create a propensity score for each observation to control for these confounding variables.  Results of an ANCOVA demonstrated statically significant increase in medication administration time persisted pre to post barcode when confounding variables were held constant (F 6.992; DF 374; Significance 0.009).
Conclusions/Implications: As anticipated the addition of bar coding did increase medication administration time.  However, the introduction of bar coding resulted in "real time" documentation of medications; a clinical practice that dramatically increases patient safety and enhances communication among health care providers. Bar coding represents a change in practice; some nurses expressed discomfort preparing medication in patients presence while others identified it as an opportunity to teach patients about their medications.