Bedside Barcode Technology: Impact on Medication Administration Tasks in an Intensive Care Unit

Abstract

Purpose

To evaluate the effect of bedside barcode medication administration (BCMA) on specific tasks performed by nurses during medication administration in an intensive care unit (ICU).

Methods

A prospective observational study was conducted on paper-based medication administration (PBMA) and BCMA in an ICU. A prevalidated time and motion observation instrument was used to measure time spent by nurses on tasks associated with PBMA and BCMA. Frequency of each task performed and mean duration of time dedicated to each task were obtained using descriptive statistics. Analysis of covariance was conducted to assess the difference in time spent with various nursing tasks between 2 periods. Factors considered as covariates were age, gender, body weight, length of stay of the patients in the ICU, number of drugs administered, and number of nurses involved.

Results

A total of 101 PBMA and 151 BCMA were analyzed. The mean duration of total medication administration time was found to be significantly higher during BCMA(377.8 ± 232.8 s) than during PBMA(313.3 ± 224.3 s). During BCMA, nurses spent more time on the following medication administration tasks: drug preparation(159.4 vs 27.3 s; P < .05), assisting physician(5.2 vs. 0.4 s; P < .05), universal precautions(3.6 vs 0.0 s; P < .05), and conversing with patients(21.6 vs 15.0 s; P < .05). During BCMA, nurses spent less time on the following medication administration tasks: giving medications(24.7 vs 35.0 s; P < .05), documenting medications(11.3 vs 17.3 s; P < .05), and cross-checking drugs with orders(7.4 vs 7.6 s; P =.04).

Conclusion

During PBMA, nurses spent more time on documenting medications and less time conversing with patients. This changed significantly after implementation of BCMA. BCMA resulted in nurses spending less time on documentation and more time conversing with patients during medication administration process.

Keywords

bedside barcode medication administration medication administration tasks time

Medication administration is a complex process, and multiple steps are required to ensure accuracy of administration and thereby patient safety. Approximately 1.5 million people per year are affected by preventable medication errors in hospitals.¹ More than 30% of preventable adverse drug events take place during the administration stage of the medication use process.^2,3 Long-established paper-based medication administration (PBMA) does not reduce chances for human error in the process of delivering medications.^4,5 An approach to preventing medication errors has been the inclusion of technology in medication administration process.⁶ Electronic medical records, electronic health records, electronic personal health records, wireless handheld devices, and barcode-assisted medication administration (BCMA) are being incorporated into primary care practice and hospital settings to improve patient safety and streamline clinicians' workflow.^7,8 BCMA reduces medication administration errors by allowing documentation at every stage of the process. A nurse receives a warning message in the BCMA system if there is any kind of disparity with the 5 rights of medication administration, that is, right patient, right drug, right dose, right route, and right time.⁹ There was evidence of an 86% reduction in medication administration errors in all 172 Veterans Affairs Medical Centers after the intervention of a barcode point of care system.¹⁰ A 2009 study in medical-surgical units reported improved accuracy in documenting and labeling medications at the patient's bedside.¹¹ The barcode point of care system is becoming a standard of care in health care information technology and helps improve patient safety by providing access to accurate information on medication errors.¹²

A wealth of literature addresses the application of barcode technology in reducing medication errors during medication administration processes, but very few studies have addressed the benefits of the technology regarding efficiencies of nursing tasks related to medication administration. A thorough understanding of the scope of nurses' workflow in the inpatient environment is critical to the successful implementation of BCMA technology.^4,13 A study reported that there was a 30% decrease in documentation time after technological intervention in the intensive care unit (ICU).¹⁴ A previous study evaluating the impact of broad categories of medication administration activities indicated that BCMA has reduced time spent on administration activities and has increased time spent on direct patient care activities¹⁵; however, the influence of BCMA on specific tasks within these activities has not been reported.

This observational time-and-motion study was conducted to compare the amount of time spent by nurses to perform each of the specific medication administration tasks during PBMA and BCMA in an ICU.

Methods

A prospective observational pre-post time-and-motion study was conducted to compare medication administration time between 2 medication administration processes – PBMA and BCMA. Specific medication administration tasks conducted by nurses in an ICU in a large tertiary hospital were observed and evaluated. Approval for this study was received from the Committee for the Protection of Human Subjects at the University and the Institutional Review Board at the hospital. Data collection assistants and investigators were trained on the Health Insurance Portability and Accountability Act (HIPAA) and protection of human subjects before data collection was initiated.

Sampling

The PBMA sample was collected by reviewing and evaluating medication administration of patients whose medications were documented by nurses in the paper-based system or medication administration record (MAR). The BCMA sample included medication administrations that were documented using the BCMA technology. The BCMA technology includes tethered scanners attached to wireless workstations on wheels. The BCMA software is bi-directionally interfaced with the pharmacy information system.

Data Collection Instrument

A prevalidated 2-part data collection instrument was used to record all tasks performed by individual nurses.^14–17 The first part of the instrument consisted of 3 sections. The first section was used to obtain information specific to the medication administration, such as date of medication administration, start time and end time of medication administration, total time, phase of medication administration (ie, whether PBMA or BCMA), and number of nurses involved in the medication administration process. The second section included demographic characteristics of the patients, and the third section obtained information on drug characteristics. The second part of the data collection instrument included a tool to document workflow tasks performed by nurses during the medication administration process. The instrument was used to document specific tasks performed by nurses as well as the time spent on each of the tasks. Workflow tasks were further categorized as general broad activities: direct patient care activities, indirect patient care activities, administration activities, and miscellaneous/other activities. A task was defined as the action necessary to achieve a specific goal. Definitions of each task are provided in Table 1 .

Table 1.

Definitions of medication administration tasks performed by nurses¹⁴

Activities	Tasks	Operational definitions
Direct patient care activity	Assessing vital signs	Administration of a medication may be determined by some predefined values of these vital signs, such as manually assessing temperature, blood pressure, pulse rate, and respiratory rate.
	Drug preparation	Includes tasks after the nurse puts on gloves to prepare and mix medications. It involves retrieving medications, preparation or making solution of medication, removing medications from vials or package, and preparing injections.
	Hanging intravenous (IV) fluids/medications	Executing the task of hanging IV bags or IV medications.
	Inserting IV catheters	Task of inserting IV catheter in piggy bags or at the insertion site of the patient's body.
	Programming/checking IV pumps	Task of adjusting the rate of fluid flow through the catheter tubes by programming the IV pumps.
Indirect patient care activity	Assisting physicians	Assisting physician with procedures or obtaining patient data related to medication administration.
	Observing IV flow	Specific observation of IV flow through IV sites to ensure fluid infusion.
	Observing monitors	Looking at the IV monitors.
	Performing tasks related to universal precautions	Putting on gloves and removing gloves.
Administration activity	Administering medication	Task of giving the medications to patients.
	Cross-checking drug with orders	Actual time of comparing delivered drug with medication orders. This task is accomplished at the time the nurse pulls out the medications from the electronic medication storage cabinet. It also includes the task of verifying the orders in the electronic medication storage cabinet interface with those indicated in the medication administration record.
	Documenting medications	Task performed to record, either electronically or by a paper-based system, the administration of medications.
	Other documentation	Task performed to record relevant vital signs before/after drug administration or details about reasons of any delay in administering medications to the patient.
Miscellaneous activity	Conversing with other personnel in the unit	Conversing with nursing staffs, primary care assistants, pharmacists, and/or other health care professionals, except the physician.
	Hand washing	Washing hands as and when necessary and after completion of each medication administration.
	Conversing with patients	Informing the patients about medications that the nurse is supposed to give and explaining the mode of action.
Other activity	Reported as and when observed	When a task did not meet any of the definitions, we added it and classified it under other activity and noted down that task.

Data Collection Process

Consent from the nurse manager was obtained before initiating data collection. The nurse manager informed all working nurses in the unit of the scope of the study. Data collection for PBMA commenced on April 18, 2008, and ended on May 16, 2008. BCMA technology (intervention) was implemented during July 2008. Data collection for BCMA was started on May 6, 2009, and ended on June 26, 2009. A 10-month period was considered adequate for nurses to be accustomed to the BCMA process. Medication administration time for each task was noted using 2 electronic stop watches. One watch measured overall medication administration time, while the other was used to measure the specific time for each task. Time spent by each nurse on specific tasks defined in Table 1 for the medication administration process was recorded. Total medication administration time for a specific patient began when the nurse started the medication administration process-related activity either inside or outside the patient room and ended when the nurse exited the patient room and completed the medication administration-related tasks.

Data validation for medication administration time was accomplished by simultaneously obtaining data using 3 data collection assistants before starting actual data collection. Data collectors were trained graduate students who had at least 6 months of experience in observing and recording information in the ICU. Five medication administrations were collected using this process, and interrater reliability calculated was 0.96. All watches used by data collectors were calibrated to minimize any variation in the measuring instrument.

Nursing tasks during the morning shift were considered in this study, as the majority of the medication administrations happened during this shift. The data were collected on consecutive days, except for Saturday and Sunday. The data collection assistant began the data collection process by observing a particular nurse during the medication administration process; using a systematic random process, every second nurse who entered the room was selected for observation. Nurses who were assigned regularly to the ICU and were familiar with the system were included in this study. The data collectors were anticipated to work with 4 to 5 nurses per shift per day. An informed consent document was used to explain the purpose of the study and the data collection process to the nurses. Nurses were informed that their performance measures would be kept confidential. Names or any other identifiers of specific nurses were not recorded. The data collection assistants observed the nurses from outside the patients' rooms and recorded all pertinent information with a previously developed and standardized data collection instrument.¹⁴ The data collectors were at the unit for more than 6 months to familiarize themselves with the process and to make the nurses comfortable with their presence. The data collectors could observe activities within and outside each room. The workstation on wheels was available for every administration. Workstations were sometimes shared between rooms. Nurses were requested to ignore the presence of the data collection assistants and to perform their duties in a normal manner. Data collection assistants were instructed not to interfere with the activities of the nurses. Further, to acclimatize them to the ICU setting and to reduce any Hawthorne effect, data collection assistants interacted with nurses in the ICU for over 6 months before initiating data collection.

Demographic characteristics (age, gender) of the patients, length of stay of the patients in the ICU, number of comorbidities present during admission, and detailed information about drugs administered were obtained from the hospital data management system.

Statistical Analysis

A sample size of 138 total medication administrations for an effect size of 0.15 with an α of 0.05 and a power of 0.95 was considered adequate for this study.¹⁸

After completion of the data collection process, data were entered in a Microsoft Excel sheet and analyzed using SAS 9.1(SAS Institute Inc, Cary, North Carolina). Mean(SD) duration of time spent on each task was calculated and descriptive statistics were reported. Chi-square tests were conducted to assess the effect of covariates. Analysis of covariance (ANCOVA) was performed to determine difference in time related to all medication administration tasks across 2 periods. Figure 1 demonstrates the model tested in this study. All statistical analyses were conducted with a number of factors acting as covariates, such as patients' characteristics (age, sex, length of stay at the ICU, and number of comorbidities) and medication administration characteristics (total number of drugs administered, number of oral drugs administered, number of topical drugs administered, and number of nurses involved).

Figure 1.

Operational model tested in a study evaluating the impact of bedside barcode technology on medication administration process. BCMA = bedside barcode medication administration; ICU = intensive care unit; PBMA = paper-based medication administration.

Results

Sample Characteristics

Sample characteristics are described in Table 2 . The total sample was 252: 101 during PBMA and 151 during BCMA. Interruptions occurred during both PBMA and BCMA. An interruption was defined as the suspension of a current task due to an unexpected event.¹⁹ Interrupted administrations(15 during PBMA, and 5 during BCMA) were deleted from the final sample. Distribution of patients' age, number of drugs administered, number of nurses involved, and length of stay of the patients in the ICU were different in the pre- and postintervention periods (Table 2). Most patients were 65 years and older, and significantly more drugs were administered during BCMA than PBMA. In 62.4% of the medication administrations during the PBMA period, 2 nurses together performed medication administration; however, only one nurse was responsible for medication administration during BCMA.

Table 2.

Characteristics of medication administrations in PBMA and BCMA approaches

		PBMA (n=101)	BCMA (n=151)
Variables	Categories	Frequency(%)	Frequency(%)	P value^*
Age of patients, years	<35	9(8.9)	3(2.0)	<.05
	35–54	24(23.8)	21(13.9)
	55–64	23(22.8)	34(22.5)
	65 +	45(44.5)	93(61.6)
Gender of patients	Male	71(70.3)	105(69.5)	.89
	Female	30(29.7)	46(30.5)
No. of drugs administered	<6	48(47.5)	95(62.9)	<.05
	6–10	47(46.5)	29(19.2)
	>10	6(5.9)	27(17.9)
No. of nurses involved in medication administration process	1	38(37.6)	151(100.0)	<.05
	2	63(62.4)	0(0.0)
Length of stay of patients in ICU, days	≤10	85(84)	100(67)	<.05
	11–20	8(8)	26(17)
	21–30	4(4)	14(9)
	31–40	4(4)	11(7)

Note: BCMA = bedside barcode medication administration; ICU = intensive care unit; PBMA = paper-based medication administration.

The P values (significance level .05) were calculated with the use of the chi-square test.

Time Spent on Medication Administration Tasks

The mean(±SD) duration for total medication administration time was found to be 313.3 s(± 224.3) during PBMA and 377.8 s(± 232.8) during BCMA. Time spent on various medication administration tasks during the 2 periods is provided in Table 3 . Assessing vital signs, programming/checking intravenous (IV) pumps, preparing the drug, inserting IV catheters, and hanging IV fluids/medications were further categorized as direct patient care activities, as these tasks were performed by nurses at the patient's bedside. Observing monitors, observing IV flow, assisting physicians, and performing tasks related to universal precautions were categorized as indirect patient care activities, as these tasks were not performed at the bedside but enabled direct patient care activities. Administering medication, documenting medications, other documentation, and cross-checking drug with orders were categorized as administration activities, as these tasks were related to medication administration. Hand washing, conversing with patients, and conversing with other personnel on the unit were categorized as miscellaneous activities, as these were nonclinical tasks and were peripherally related to medication administration. There were few tasks that were identified as they happened, for example, positioning patients before administering medications, viewing warnings in the bedside barcode technology, and reacting to warnings during the BCMA observation period. As these tasks were not in the instrument, they were classified as other activity. Only a few tasks were categorized as other activity, and they were not included in the analysis. The 3 tasks on which nurses spent the most time during PBMA were giving medications to the patients(35.0 s), drug preparation(27.3 s), and documenting medications(17.3 s). The 3 tasks on which nurses spent most time during BCMA were drug preparation(159.4 s), giving medications(24.7 s), and conversing with patients(21.6 s). The ANCOVA result indicated that during BCMA nurses spent more time on the following medication administration tasks: drug preparation(159.4 vs 27.2 s; P < .05), assisting physician(5.2 vs 0.4 s; P < .05), universal precautions(3.6 vs 0.0 s; P < .05), and conversing with patients(21.6 vs 15.0 s; P < .05). During BCMA, nurses spent less time on the following medication administration tasks: giving medications(24.7 vs 35.0 s; P < .05), documenting medications(11.3 vs 17.3 s; P < .05), and cross-checking drugs with orders(7.4 vs 7.6 s; P = .04).

Table 3.

Mean time to perform various medication administration tasks: PBMA vs BCMA

		PBMA (n=101)	BCMA (n=151)
Activities	Tasks	Mean time, s(±SD)	Mean time, s(±SD)	P value^*
Direct patient care activity	Drug preparation	27.27(±45.69)	159.45(±122.52)	<.05
	Program/check IV pumps	8.41(±15.74)	7.54(±30.99)	.02
	Hang IV medications	5.09(±15.70)	3.11(±9.57)	.61
	Assessing vital signs	4.74(±22.65)	7.17(±18.51)	<.05
	Insert IVs	2.62(±11.72)	5.04(±39.46)	.47
Indirect patient care activity	Observe IV flows	8.77(±21.62)	0.41(±2.48)	<.05
	Observe IV monitors	3.79(±9.85)	0(±0)	<.05
	Assist physician	0.47(±2.73)	5.20(±20.39)	<.05
	Universal precautions	0.00(±0)	3.60(±5.76)	<.05
Administration activity	Give medications	35.06(±56.49)	24.77(±41.36)	<.05
	Documenting meds	17.32(±27.37)	11.37(±7.27)	<.05
	Cross check drug with orders	7.69(±21.63)	7.43(±16.10)	.04
	Other documentation	3.53(±12.36)	1.94(±13.52)	.14
Miscellaneous activity	Conversing with patients	15.04(±29.25)	21.66(±25.54)	<.05
	Conversing with others in unit	7.79(±19.06)	7.07(±13.60)	.16
	Hand washing	6.31(±7.23)	4.04(±4.64)	.16

Note: BCMA = bedside barcode medication administration; ICU = intensive care unit; PBMA = paper-based medication administration.

P values (significance level .05) were based on analysis of covariance (ANCOVA).

Discussion

The study indicated that the time nurses spent in conversing with patients (eg, providing drug information to the patients about potential side effect and mechanism of action) during medication administration increased after implementation of BCMA in the ICU, while time spent on documenting medications decreased. In a BCMA system, direct access to patients' clinical and pathology results, patients' comorbidities and allergy information, drug information, drug-drug interaction information, and hospital policies related to medication administration may contribute to nurses' ability to conduct administration activities faster. Nurses spent significantly more time in assessing vital signs and assisting physicians during BCMA than during PBMA. After implementation of BCMA, the time nurses spent on “other documentation” was reduced, as they were not required to document in a paper chart. Although not evaluated in this study, the quality of care may have also improved after implementation of BCMA. However, further research is required to confirm this empirical evidence.

To improve efficiency and quality of care, many hospitals now rely on technology to enhance accuracy and efficiency in documentation.²⁰ The study reported that nurses spent less time in documenting medicines but spent significantly more time with patients during BCMA than during PBMA, which was a good indication of an improvement in quality of care. Amount of time spent on the drug preparation task was significantly higher during BCMA, possibly because number of drugs administered was much higher than during PBMA. As a majority of the medication administrations involved elderly patients (age ≥65) during BCMA and were not oral solid dosage forms, as reported by the observer, nurses were required to crush medications and it took time. This fact might explain why nurses spent more time in drug preparation during BCMA.

Certain limitations were identified during the study. First, nurses may be subject to the Hawthorne effect as they were observed during the medication administration process, though this effect was similar for both the PBMA and BCMA periods. The effect was minimized by making sure data collection assistants spent adequate time in the ICU before initiating data collection and nurses were comfortable with their presence. Second, this study was conducted in an ICU; therefore, we cannot generalize these findings to other patient care units as patient characteristics may be specific to an ICU. In our study, patients' age in the BCMA period was higher. Thus effects observed for time savings could be conservative as these patients typically have more comorbidities and used more medications in our study. Third, the nature of a time-and-motion study does not allow the assessment of the quality of the interaction between patients and nurses. Further research is needed to evaluate whether the increase in direct patient care activity increases the quality of interaction between patients and nurses and thus improves patient care and safety.

Conclusion

During PBMA, nurses spent more time on administering and documenting medications and spent less time conversing with patients. After implementation of BCMA, the time spent by nurses on administering medications and documenting medications was significantly reduced, and they spent more time conversing with patients.

Footnotes

Acknowledgments

This study was partially funded by a research grant from the ASHP Research and Education Foundation.

References

Aspden

IOM'S Preventing Medication Errors Report. J Pharm Finance Econ Policy. 2008; 16(1): 3–4.

Pepper

G.A.

Errors in drug administration by nurses. Am J Health Syst Pharm. 1995; 52(4): 390.

Bates

D.W.

Cullen

D.J.

Laird

Incidence of adverse drug events and potential adverse drug events. JAMA. 1995; 274(1): 29.

AHRQ. Bar-Coded Medication Administration. 2008. http://healthit.ahrq.gov/portal/server.pt/gateway/PTARGS_0_1248_831403_0_0_18/08–0085_bcma.pdf. Accessed July 30, 2009.

Sublett

Technology's impact on reducing medication errors. Health Manage Technol. 2002; 23(11): 24–27.

Grigg

S.J.

Garrett

S.K.

Craig

J.B.

A process centered analysis of medication administration: identifying current methods and potential for improvement. Int J Industr Ergonomics. 2011; 41(4): 380–388.

Defining Key Health Information Technology Terms. Washington, DC: US Government Printing Office; 2008.

Keohane

C.A.

Bane

A.D.

Featherstone

Quantifying nursing workflow in medication administration. J Nurs Adm. 2008; 38(1): 19.

Westbrook

J.I.

Woods

Rob

M.I.

Dunsmuir

Day

R.O.

Association of interruptions with an increased risk and severity of medication administration errors. Arch Intern Med. 2010; 170(8): 683.

10.

Johnson

C.L.

Carlson

R.A.

Tucker

C.L.

Willette

Using BCMA software to improve patient safety in Veterans Administration Medical Centers. J Healthcare Inf Manage. 2002; 16(1): 46.

11.

Helmons

P.J.

Wargel

L.N.

Daniels

C.E.

Effect of bar-code-assisted medication administration on medication administration errors and accuracy in multiple patient care areas. Am J Health Syst Pharm. 2009; 66(13): 1202.

12.

Anderson

Wittwer

Using bar-code point-of-care technology for patient safety. J Healthc Qual. 2004; 26(6): 5–11.

13.

Larrabee

Brown

M.M.

Recognizing the institutional benefits of bar-code point-of-care technology. J Quality Patient Safety. 2003; 29(7): 345–353.

14.

Wong

D.H.

Gallegos

Weinger

M.B.

Clack

Slagle

Anderson

C.T.

Changes in intensive care unit nurse task activity after installation of a third-generation intensive care unit information system. Crit Care Med. 2003; 31(10): 2488.

15.

Dwibedi

Sansgiry

S.S.

Frost

C.P.

Effect of bar-code-assisted medication administration on nurses' activities in an intensive care unit: a time-motion study. Am J Health Syst Pharm. 2011; 68(11): 1026.

16.

Dasgupta

Sansgiry

S.S.

Jacob

S.M.

Frost

C.P.

Dwibedi

Tipton

Descriptive analysis of workflow variables associated with barcode-based approach to medication administration. J Nurs Care Quality. 2011; 26(4): 377.

17.

Fowler

S.B.

Sohler

Zarillo

D.F.

Bar-code technology for medication administration: medication errors and nurse satisfaction. Medsurg Nurs. 2009; 18(2): 103.

18.

Faul

Erdfelder

Lang

A.G.

Buchner

G^* Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods. 2007; 39(2): 175.

19.

Tang

Mazabob

Weavind

Thomas

Johnson

T.R.

A time-motion study of registered nurses' workflow in intensive care unit remote monitoring. AMIA Ann Sympos Proc. 2006; 759–763.

20.

Korst

Eusebio-Angeja

Chamorro

Aydin

Gregory

Nursing documentation time during implementation of an electronic medical record. J Nurs Adm. 2003; 33(1): 24–30.