distribute not do copy, press - ecri

ECRI INSTITUTE PSO DEEP DIVE: MEDICATION SAFETY Share Learn ProtectTM

In-depth look at medication events

Systems-focused learning

Leadership strategies

Online resources

PRESS COPY, D

O NOT D

ISTRIB

UTE

ECRI INSTITUTE PSORonni P. Solomon, J.D., Executive Vice President and General Counsel; Karen P. Zimmer, M.D., M.P.H., FAAP, Editor, PSO Navigator; Clinical Director; Amy Goldberg-Alberts, M.B.A., FASHRM, CPHRM, Program Director; Barbara G. Rebold, B.S.N., M.S., R.N., CPHQ, Director, PSO Operations

Paul A. Anderson, Director, Risk Management Publications; Theresa V. Arnold, D.P.M., Manager, Clinical Analysis; Sharon Bradley, R.N., C.I.C., Infection Preventionist; Andrea Fenton, Web Editor; Kelly C. Graham, B.S.N., R.N., Patient Safety Analyst; Charlotte Huber, R.N., M.S.N., Senior Patient Safety Analyst; William M. Marella, M.B.A., Director, Patient Safety Reporting Programs; Patricia Neumann, R.N., M.S., Senior Patient Safety Analyst; Catherine Pusey, R.N., M.B.A., Senior Patient Safety Analyst; Madelyn S. Quattrone, Esq., Legal Editor, Risk Management Publications; Senior Risk Management Analyst; Sheila Rossi, M.H.A., Senior Patient Safety Analyst; Melva Sanzon, Reporting and Business Reporting Analyst; Kathleen M. Shostek, R.N., A.R.M., FASHRM, CPHRM, Senior Risk Management Analyst; Patricia Teggatz, M.A., Administrative Coordinator; Cynthia Wallace, CPHRM, Senior Risk Management Analyst; Andrea J. Zavod, Managing Editor

ADVISORY COUNCILKatrina Belt, M.P.A., Healthcare Authority for Baptist Health; Doug Bonacum, M.B.A., CSP, CPHQ, CPHRM, Kaiser Permanente; Darrel A. Campbell, Jr., M.D., University of Michigan Health System; Nancy E. Foster, American Hospital Association; Stephen T. Lawless, M.D., M.B.A., The Nemours Foundation/Alfred I. DuPont Hospital for Children; David C. Levin, M.D., Jefferson Health System; Michael A. Olympio, M.D., Wake Forest University School of Medicine; Richard G. Roberts, M.D., J.D., University of Wisconsin School of Medicine and Public Health; Robert M. Wachter, M.D., University of California, San Francisco

MISSION STATEMENTThe ECRI Institute PSO is a federally listed patient safety organization that is a component of ECRI Institute.

ECRI Institute, a nonprofit organization, dedicates itself to bringing the discipline of applied scientific research in healthcare to uncover the best approaches to improving patient care. As a pioneer in this science for 40 years, ECRI Institute marries experience and independence with the objectivity of evidence-based research.

ECRI Institute PSO Deep Dive: Medication Safety is published by ECRI Institute, 5200 Butler Pike, Plymouth Meeting, PA 19462, USA; (610) 825-6000 (telephone); (610) 834-1275 (fax); [email protected] (e-mail).

For more information about ECRI Institute PSO, send an e-mail to [email protected].

Share Learn ProtectTMShare Learn ProtectTM

Share Learn ProtectTM

Share Learn ProtectTMShare Learn ProtectTM

Share Learn ProtectTM

ACKNOWLEDGEMENTS

Lessons learned from a medication event at one organization can prevent the same or a similar error from occurring at another facility if the information learned in the aftermath is disseminated to other organizations. Many organizations encourage caregivers to externally report medication errors so that the information can be shared anonymously with others in order to prevent other similar errors.

ECRI Institute PSO thanks its collaborating PSOs and member organizations for sharing their medication events for this deep-dive analysis so that the lessons learned from the aggregated medication events can be shared with ECRI Institute PSO members.

ECRI Institute PSO encourages its members to review the findings from this report and to enlist a multidisciplinary team of pharmacists, physicians, nurses, and others to discuss the applicability of the findings to their own organizations. Medication safety should be an ongoing focus of every healthcare organization with the goal of reducing the frequency and severity of errors related to medication use. ECRI Institute PSO members can use this deep-dive analysis to begin to achieve that goal.

In addition to the review provided by ECRI Institute PSO’s Advisory Council, ECRI Institute PSO acknowledges the following individuals for their insights about this report:

X Michelle M. Bell, R.N., B.S.N., FISMP, Patient Safety Liaison, Pennsylvania Patient Safety Authority, Harrisburg, Pennsylvania

X Mary Blanco, R.N., M.S.N., CPHQ, Chief Quality Officer, Brandywine Hospital, Coatesville, Pennsylvania

X Stephanie Uses, Pharm. D., M.J., Staff Pharmacist, Crozer-Keystone Health System, Taylor Hospital, Ridley Park, Pennsylvania

PRESS COPY, D

O NOT D

ISTRIB

UTE

ECRI INSTITUTE PSO DEEP DIVE: MEDICATION SAFETY

©2011 ECRI Institute. May be disseminated, for internal educational purposes solely at the subscribing site. For broader use of these copyrighted materials, please contact ECRI Institute to obtain proper permission.DECEMBER 2011 iii


Acknowledgements ii

Table of Contents iii

Executive Summary v

Background 1

Methods 2

Results 3-4

Distribution of Medication Errors by Medication Process Node 5-8

Contributing Factors to Medication Administration Events 9-11

Risk Reduction Strategies to Improve Medication Safety by Strength of Impact 12-13

Improving Medication Administration Safety 14

Where to Start? 15

Minimizing Distractions and Interruptions 16

Workload 17

Technology Solutions 18-19

Safe IV Practices 20-21

High-Alert Medications 22-23

General Safety Strategies for Medication Administration 24-25

Education 26

Online Resources 27

References 27-28

Appendix 29

Table of Contents

PRESS COPY, D

O NOT D

ISTRIB

UTE

PRESS COPY, D

O NOT D

ISTRIB

UTE


©2011 ECRI Institute. May be disseminated, for internal educational purposes solely at the subscribing site. For broader use of these copyrighted materials, please contact ECRI Institute to obtain proper permission.DECEMBER 2011 v

Key Recommendations X Provide leadership support for medication safety initiatives.

X Adopt a systems approach to medication safety.

X Use multiple techniques (e.g. event reporting, chart review) to track and identify medication errors.

X Involve frontline staff in medication safety initiatives.

X Ensure that pharmacists are readily available to caregivers at all times to answer questions.

X Emphasize safety strategies that eliminate hazards so that errors cannot reach patients.

X Standardize formats for medication orders.

X Label all medications, including any syringe or solution bag containing medications.

X Purchase infusion pumps with dose error reduction features.

X Limit the number of concentrations available for each infusion solution.

X Limit the variety of high-alert medi-cations available in the organization’s formulary.

X Require independent doublechecks of high-alert drugs to confirm infor-mation such as correct drug, dose, and patient.

X Involve patients and their families in the medication process.

Medication mishaps are the most common errors in healthcare. Indeed, medica-

tion errors represent the most frequently reported events submitted to ECRI

Institute PSO—comprising about 30% of all events.

The sheer number of drugs administered in healthcare facilities increases

the likelihood that medication errors will occur if risk reduction systems are

not in place.

To assist healthcare facilities in learning from medication errors, ECRI

Institute PSO asked its members and members of collaborating PSOs to

submit at least 10 medication events over a five-week period to enable ECRI

Institute PSO to identify patterns and trends from the aggregated data and

share the findings, as well as its recommendations.

WHAT ECRI INSTITUTE PSO FOUND Participating PSO members submitted 695 medication events during the five-week period starting April 15, 2011, and ending May 20, 2011. Events occurred most frequently during the medication administration stage. The analysis provides recom-mendations to improve the safety of medication administration, particularly with intravenous infusions, which represented the largest share of administration events when analyzed by administration route. The analysis also underscores the need to identify system-based causes of medication errors rather than the current tendency to focus on human performance.

Executive Summary

PRESS COPY, D

O NOT D

ISTRIB

UTE

PRESS COPY, D

O NOT D

ISTRIB

UTE


©2011 ECRI Institute. May be disseminated, for internal educational purposes solely at the subscribing site. For broader use of these copyrighted materials, please contact ECRI Institute to obtain proper permission.DECEMBER 2011 1

Background

XX Medication mishaps are the most common errors in healthcare.

XX Every hospitalized patient is at risk for at least one medication error per day.

XX Many medication events are preventable.

XX By analyzing medication errors, institutions can identify risks in their medication-use process and apply lessons to prevent their recurrence.

Medication mishaps are the most common errors in healthcare. Indeed, medica-tion errors represent the most frequently reported events submitted to ECRI Institute PSO, comprising about 30% of all events. In its 2006 report Preventing Medication Errors, the Institute of Medicine suggests that every hospitalized patient is at risk for at least one medication error per day and that many of these events are preventable (IOM).

The sheer number of drugs adminis-tered in healthcare facilities increases the likelihood that medication errors will occur if risk reduction systems are not in place. Additionally, the medication-use process is complex, encompassing several phases, including (but not limited to) prescribing, dispensing, administering, and monitoring.

Error Prevention Is Top PriorityPreventing harm from medications, or adverse drug events (ADEs), remains a top priority for healthcare organizations and policymakers. Consider the following ongo-ing initiatives:

X The federal government’s Partnership for Patients, a public-private initiative to improve the quality and safety of health-care has set a goal to reduce preventable ADEs in hospitals by 50% by 2013.

About half of all ADEs are considered preventable.

X The Joint Commission, which lists medication errors among the top 10 categories of sentinel events reported to the accrediting organization (Joint Commission “Summary”), also targets medication safety in many of its initia-tives. Almost one-quarter of its Sentinel Event Alerts issued since 1998 have covered medication safety. Additionally, medication safety is addressed in several National Patient Safety Goals, including goals to use two patient identifiers and to maintain and communicate accurate patient medication information. Medica-tion safety is also incorporated into Joint Commission accreditation standards.

X The National Quality Forum’s (NQF) list of serious reportable patient safety events, which is used by several state-based event reporting programs, includes any patient death or serious injury asso-ciated with a medication error, such as those involving the wrong drug, wrong dose, wrong patient, wrong time, wrong rate, wrong preparation, or wrong route of administration (NQF). NQF’s list rep-resents events that are largely preventable.

By analyzing medication errors— both actual events and close calls, or near

misses—institutions can identify the risks in their medication-use process that allow errors to occur and apply lessons to pre-vent their recurrence. To assist healthcare facilities in learning from medication errors, ECRI Institute PSO asked its members and members of collaborating PSOs to submit medication events over a specified time period so that ECRI Institute PSO could identify patterns and trends from the aggre-gated data and share the findings, as well as recommendations. Under the Patient Safety and Quality Improvement Act, PSO mem-bers can report the data on a privileged and confidential basis.

This report summarizes ECRI Institute PSO’s analysis of the nearly 700 medication-related events submitted.* In addition to summarizing general findings about events in the medication-use process, the report focuses on one stage of the medication-use process—medication administration, which represented the largest share of medication events reported to ECRI Institute PSO dur-ing the five-week period.

* ECRI Institute PSO routinely collects a large number of event and near-miss reports, including, but not limited to, reports of medication errors. For this report, ECRI Institute PSO limited its review to the medication-related event reports spe-cifically submitted for purposes of this analysis.

AT A GLANCE

PRESS COPY, D

O NOT D

ISTRIB

UTE


©2011 ECRI Institute. May be disseminated, for internal educational purposes solely at the subscribing site. For broader use of these copyrighted materials, please contact ECRI Institute to obtain proper permission. DECEMBER 20112

In March 2011, ECRI Institute PSO issued a call to action to its member healthcare organizations to collaborate in an initiative, called a “deep-dive” analysis, to learn from medication errors and to share strategies to reduce errors and improve care within member facilities. ECRI Institute PSO asked its members and members of collaborat-ing PSOs to submit at least 10 medication events during a five-week reporting period. Because facilities could select the medication events they reported during the five-week period, the data represent a snapshot of all medication errors that occurred during the five-week period. There is much that can be learned from the reported events to improve medication safety.

Member facilities submitted the events using ECRI Institute PSO’s patient safety event reporting system. Because ECRI

Institute PSO and the collaborating PSOs are federally certified as PSOs, providers that report medication events and near misses have the assurance that the data is protected from legal discovery and will be kept confidential. However, ECRI Institute PSO can share de-identified findings and lessons learned from its analysis of the aggregated data.

Common Reporting FormatECRI Institute PSO’s event reporting system uses the Agency for Healthcare Research and Quality’s (AHRQ) common definitions and reporting formats, which allow PSOs to collect information from providers and standardize how patient safety events are represented. AHRQ’s common reporting format for patient safety events involving medications includes up to

25 questions about the event. For example, reporters are asked to identify the node, or stage, in the medication process when the event originated.

ECRI Institute PSO has further modi-fied the questions for medication events by asking reporters to identify process factors that contributed to the medication error or near miss, as well as systemwide issues, or contributing factors, that were involved. For example, process factors for medica-tion events that occur during administration might include medication administration delays or failure to discontinue a medication. Contributing factors to the event can range from caregiver communication failures to environmental distractions, such as noise or interruptions.

Methods

AT A GLANCE

XX ECRI Institute PSO member organizations collaborate to learn about medication errors.

XX Deep-dive analysis evaluates medication events during a five-week reporting period.

PRESS COPY, D

O NOT D

ISTRIB

UTE



MS11

535

0 50 100 150 200 250 300

Neonates(0 to 28 days)

Infants(>28 days to <1year)

Children(1 to 12 years)

Adolescents(13 to 17 years)

Adults(18 to 64 years)

Older adults(64 to 84 years)

Aged adults(85+ years)

Not given, unknown

15 (2%)

16 (2%)

40 (6%)

17 (2%)

281 (49%)

224 (32%)

76 (11%)

26 (4%)

Age

Number (N=695)Percentages do not add up to 100% due to rounding.

Participating PSO members submitted 695 medication events during the five-week period starting April 15, 2011, and ending May 20, 2011. Eighty healthcare facili-ties—including acute care, children’s, and long-term care—participated in the initia-tive. The majority of events were submitted by acute care hospitals.

The majority of the reports, 631 or 91%, were classified as incidents, which are defined as patient safety events that reach the patient, whether or not the patient was harmed. All but 2 of the remaining reports were near misses (62 or 9% of the total), which are defined as patient safety events

that did not reach the patient. The 2 remain-ing reports involved an unsafe condition, or a circumstance that increases the probability of a patient safety event.

Distribution of Medication Errors by Patient AgeAs illustrated in Figure 1, the majority of the medication events affected adults between 18 and 64 years of age (40%) and older adults between the 65 and 84 years of age (32%). Just over 12% of the patients were pediatric patients 17 years of age and under.

Results

XX Participating PSO members submit 695 medication events during a five-week reporting period.

XX Analysis of events by patient age, care areas, and severity.

AT A GLANCE

Figure 1. Ages of Patients Involved in Medication Events, April 15 to May 20, 2011

PRESS COPY, D

O NOT D

ISTRIB

UTE



Care Areas Involved in Medication ErrorsOf the 538 reports that indicated the care area where the medication event occurred, the majority happened on the medical/ surgical floor (52%), followed by critical care settings (11%) and the emergency depart-ment (ED) (9%). See the Table for a list of the top five areas where medication events occurred.

Severity of Medication ErrorsFacilities reporting medication events to ECRI Institute PSO indicated that 8% of the medication errors reached the patient and contributed to or resulted in harm, which ranged from temporary harm to death. Reports resulting in patient harm are those in Categories E through I using the National Coordinating Council for Medica-tion Error Reporting and Prevention’s (NCC MERP) Index for Categorizing Medication Errors. See Figure 2 for a breakdown of the reports by the NCC MERP harm score. The harm score was provided for 690 of the 695 reported medication events.

MS11

536

A (circumstance that has the capacity to cause harm)

B (error occurs but does not reach the patient)

C (error reaches the patient but does not cause harm)

D (error reaches the patient and requires monitoring to confirm no harm and/or intervention required)

E (error contributes to or results in temporary harm to the patient and requires intervention)

F (error contributes to or results in temporary harm and requires initial or prolonged hospitalization)

G (error contributes to or results in permanent patient harm)

H (error requires intervention to sustain life)

I (error may have contributed to or resulted in patient’s death)

Harm Score

Number (N=695)Percentages do not add up to 100% due to rounding.

0 100 200 300 400 500

18 (3%)

70 (10%)

406 (59%)

144 (21%)

36 (5%)

9 (1%)

1 (<1%)

4 (1%)

2 (<1%)

Figure 2. Medication Events by Harm Score, April 15 to May 20, 2011

Table. Top Five Care Areas for Medication Events, April 15 to May 20, 2011

Care Area Number Percent of Total (N = 538)

Medical/surgical floor 277 52%

Critical care (includes critical care, intensive care, and medical intensive care units) 61 11

ED 46 9

Telemetry unit 26 5

Pediatric floor 18 3

PRESS COPY, D

O NOT D

ISTRIB

UTE



MS11

537

Prescribing

Event occurred during specific node(N=473)

Event involvedmultiple nodes(N=194)

0

50

100

150

200

250

300

350

Dispensing Administering Monitoring

40

80 70

185

320

178

3719

Figure 3. Node in Medication-Use Process for Medication Event’s Origination*

*Node not identified for 28 of the 695 events.

Distribution of Medication Errors by Medication Process Node

XX Most events in the deep-dive analysis occur during medication administration.

XX Intravenous-related errors are the most frequent of administration-only events analyzed by the route of medication delivery.

ECRI Institute PSO analyzed the distribution of medication errors by the medication-use process. Prescribing, dispensing, administer-ing, and monitoring are identified as the nodes in the medication-use process in which events originate. Although some analyses of the medication-use process also list tran-scribing as one of the nodes, ECRI Institute PSO’s evaluation found that transcribing errors occurred during all the various nodes analyzed. For example, an incomplete order

was classified as a prescribing error instead of a transcription error, and a misinterpreted order was grouped as a dispensing or admin-istration error.

Of the 695 events analyzed, 473 were more specific to one node of the medication-use process. Although errors can occur in any phase of the medication process, ECRI Institute PSO facilities indicated that most events specific to one node occurred during administration of the medication (67.7%),

followed by dispensing (16.1%), prescribing (8.5%), and monitoring (7.8%).

Many events can include multiple nodes of the medication-use process. For example, an error in the ordering stage may continue into the dispensing and administration nodes. Another 194 of the 695 medica-tion events involved multiple nodes in the medication-use process. When events involving multiple nodes were analyzed, the majority of events occurred during both the dispensing (95.4%) and administering nodes (91.8%). See Figure 3 for a breakdown of medication events by node. There were 28 events for which no node of the medica-tion-use process was identified.

Analysis of Events by NodeWhen the percentage of all events occurring during each of the four nodes—regardless of whether the event occurred during a spe-cific node or multiple nodes—are calculated, events occurred most frequently during the administration node (71.7%), followed by dispensing (37.6%), prescribing (17.3%), and monitoring (8.1%).

Medications can be administered in a variety of ways—for example, by infusing an intravenous (IV) solution, by swallow-ing a tablet, or by rubbing an ointment into the skin. ECRI Institute PSO analyzed the

AT A GLANCE

PRESS COPY, D

O NOT D

ISTRIB

UTE



320 reports of administration-only events to identify the most common routes for these events. For almost 30% of the events reported, however, the route of administra-tion was not given. Of the 320 reports for administration-only errors, IV-related errors were the most frequently occurring events. The breakdown of administration events by route of administration is as follows:

X IV (36.9% of administration-only events)

X Oral (18.1%)

X Subcutaneous injection (7.8%)

X Inhalation (2.8%)

X Intramuscular injection (1.6%)

X Feeding tube (0.1%)

X Other (i.e., six events were associated with six other routes of administration, such as nasal spray, skin spray, eye drops) (1.8%)

ECRI Institute PSO analyzed the admin-istration events in more detail by looking at the types of errors involved with the top-three problematic routes of administra-tion—IV infusions, oral medications, and subcutaneous injections.

INTRAVENOUS ADMINISTRATION ERRORSThe frequency of IV administration errors may be partly due to the frequency of IV use in a hospital and the complex, error-prone aspects of infusion pump programming. In some hospital units, about half of all medications that a nurse administers are delivered intravenously. One study of medication administration in the intensive care unit found that a nurse administered 13 medications per patient and that 7 of these medications were either IV medications or IV fluids (Moss et al.).

As the reports submitted to ECRI Institute PSO demonstrate, IV therapy is

prone to mistakes. The devices are pro-grammed at the bedside and have a range of programming options, making it possible, for example, to leave out a decimal point or add a zero when setting the infusion rate, resulting in a 10- or 100-fold overdose. Some pumps have small screens, increasing the risk of selecting the wrong medication or dose. Additionally, there is increased risk of error because of variability in the names of drugs used for infusions, dosing concentrations, dose limits, and infusion rates (AHRQ).

In terms of medication types associated with IV medication errors in the reports submitted to ECRI Institute PSO, antibiotics were identified with the largest number of IV administration events (28) when infor-mation about the medication was provided. Vancomycin, which is classified as a glyco-peptide antibiotic (Harrison’s), was associated with the most events involving antibiot-ics—10, or 35.7% of the 28—followed by Zosyn (piperacillin and tazobactam), named in 14.3% of the IV administration-only errors involving antibiotics.

Many IV infusions involve high-alert drugs, such as insulin, anticoagulants, and chemotherapeutic agents. The Institute for Safe Medication Practices (ISMP), which publishes a list of high-alert medications, defines them as drugs that bear a heightened risk of causing significant harm when they are used in error (ISMP). IV medication errors with high-alert drugs can be especially harmful because they involve direct admin-istration of drugs into the vasculature; this often has immediate effects, leaving little time to correct or reverse the drug’s effects. Of the 118 IV-related administration errors, 55 (46.6%) involved a high-alert medica-tion, and 9 of the events, according to the reports, resulted in patient harm, including two deaths. Significantly, these two fatal events were the only two medication errors reported to ECRI Institute PSO that may

have contributed to or resulted in patient deaths; both occurred during administration of IV infusions, and both events involved high-alert medications.

The most frequently identified high-alert medications for IV administration errors were heparin, an anticoagulant (identified in seven reports); insulin (listed in five reports); and diltiazem, an antiarrhythmic (identified in three reports).

ECRI Institute PSO examined the 118 IV events and found that the most commonly reported types of IV-related medication errors were for the following reasons:

X Drug not given (22.9% of IV-related errors)

X Wrong pump rate (20.3%)

X Wrong drug (16.9%)

X Wrong dose (13.6%)

Drug not given. Several events described delays in administering a piggyback IV medication because of failure to open the tubing clamp so that the medication can be infused through the main IV line, as in the following report:

The IV antibiotic was hung but did not infuse because the tubing was clamped to the secondary set.

Sometimes, a caregiver forgot to connect the IV line to the patient, as illustrated by the following case:

A nurse interrupted a patient-controlled anal-gesia (PCA) infusion to draw blood from the IV line but failed to reconnect the PCA pump. The patient was without pain medication until the error was discovered.

Wrong rate. Several wrong-rate events involved programming the pump to deliver an infusion at the wrong rate, as in the fol-lowing report:

Levophed (norepinephrine bitartrate) was administered at an incorrect rate. The nurse ran the

PRESS COPY, D

O NOT D

ISTRIB

UTE



Levophed at the maintenance IV rate instead of the correct, and much slower, rate.

Programming errors have the potential for harm if the wrong amount of medica-tion—either too much or too little—is deliv-ered. Even minor variations in dosages can result in serious adverse effects (Hicks and Becker). In the above case, the programming error resulted in the patient receiving too much of a high-alert medication—Levophed is an adrenergic agonist—although according to the report, the patient was not harmed.

Programming errors that result in drug dosing errors can be difficult to detect unless the organization uses newer infusion pumps that incorporate medication safety software like drug libraries with dosing limits to warn users of programming errors before IV medications are administered. In one report, a rate-programming error resulted in the patient receiving 100 times the amount of heparin that was ordered:

The nurse programmed the pump with heparin to run at 200 cc/hour versus 200 units/hour, or 2cc/hour.

Heparin is another high-alert medication and can cause patient harm or death from overdoses of the blood-thinning agent. In the event reported to ECRI Institute PSO, the patient was monitored for any adverse effects.

Errors in calculating the patient’s weight can also lead to pump-rate errors, as in the following report, which involved a heparin infusion:

The nurse miscalculated the patient’s weight. As a result, the patient was given the wrong bolus, and the heparin dose was running at the wrong rate. No heparin drip protocol sheet was used to document the infusion.

Wrong drug. The infusion events analyzed by ECRI Institute PSO underscore the risk of administering the wrong drug in an infu-sion. One such event may have contributed

to or resulted in a patient death when the patient was given a wrong-drug infusion of lidocaine, a high-alert antiarrhythmic:

A patient was given an IV infusion of lidocaine for arrhythmia. When the IV infusion was discontinued, the patient was to receive saline intravenously. Later, a nurse found the patient had pulled the IV line and was unresponsive; it was discovered that the lidocaine bag was empty, and the saline bag was full.

Another wrong-drug IV infusion error was nearly fatal because the patient mistakenly received a paralyzing agent— succinylcholine, a high-alert medication—instead of ketamine, a general anesthetic. Because paralyzing agents can affect the muscles for breathing, a patient who receives succinylcholine requires a ventilator. A patient receiving ketamine can still breathe spontaneously and need not be ventilated. In the following case, the patient was not on a ventilator when succinylcholine was mistak-enly given:

The [anesthesia provider] mistakenly grabbed a syringe containing succinylcholine instead of ket-amine. Although the medication label on the syringe was correct, the syringe was supposed to be stored separately and have an orange warning label. The [anesthesia provider] attributed the error to being in a hurry and lack of a warning label.

Wrong-drug IV errors occurred for a variety of reasons. For example, a verbal order for Solu-Medrol (methylprednisolone sodium succinate) was misheard as Depo-Medrol (methylprednisolone acetate), and the order was not read back for confirma-tion; an antibiotic infusion was mistakenly pulled from an automated medication dis-pensing cabinet because it was in the same bin as potassium chloride, the infusion that was ordered; or the wrong IV fluid was hung without double-checking the contents of the infusion bag.

ORAL ADMINISTRATION ERRORSEvents occurring with medications delivered by mouth sometimes involved similar errors as those found among the IV events. Of the 58 events involving oral medications, 29.3% were identified as a wrong-dose error, 20.7% were because a drug was not given, and 15.5% of events resulted from admin-istering the wrong drug. Though there were wrong-patient errors for events involving both IV and oral administration of drugs, the wrong-patient errors represented a larger share of events for medications taken orally—10.3% for oral administration events versus about 6% for IV-related events.

As with IV events, a large share of oral administration events involved high-alert medications (39.6%). The largest number of events with high-alert medications, 12 or 52.2% of the events, occurred with opioids, and another 5 events (21.7%) arose with anticoagulants.

Examples of events reported for oral medications are given below by type of event.

Wrong dose. Among the immediate causes of the 17 wrong-dose errors for medica-tions administered orally were caregiver distraction and inaccurate information about a patient’s weight. In the following report, the caregiver mistakenly doubled a patient’s dose of lorazepam:

The nurse inadvertently misread a medication order for 0.5 mg of lorazepam as two lorazepam 0.5 mg and administered 1 mg.

Drug not given. In the following report, a transcription error resulted in a patient miss-ing a drug dose:

A physician’s order for a medication was omit-ted from the medication administration record (MAR). The error was found during the night shift chart check. The prior day’s medication dose had been omitted.

PRESS COPY, D

O NOT D

ISTRIB

UTE



Other errors involved drug omissions or the failure to give an ordered dose, resulting in problems requiring additional care. For example, in one report, the nurse thought that an order for a patient to not take any food or fluids by mouth—identified in the patient’s record with the acronym NPO—also applied to the patient’s anti-arrhythmia medicine provided in pill form:

A nurse did not administer two doses of amio-darone because the patient was NPO. There was an order to take all medications. Patient experienced arrhythmias and multiple shocks from the implant-able cardioverter defibrillator. A rapid response team was called, and the patient was transferred to a higher level of care and required additional medications.

Wrong drug. The nine wrong-drug events of oral medications occurred for a variety of reasons: the automated dispensing cabi-net had the wrong medication; one medica-tion was incorrectly substituted for another; a patient, who indicated an allergy to

hydromorphone, was still given the opioid, a high-alert medication; and a patient was given several different medications without a physi-cian’s order, resulting in temporary harm.

Wrong patient. When reasons were given for any of the six wrong-patient drug errors for oral medications, they included the following: the drug was mistakenly administered to the intended patient’s roommate, the patient was not identified by name before the drug was administered, and the nurse had multiple patients on varying doses of the same medi-cation and gave the wrong dose to one of the patients.

DRUG INJECTION ADMINISTRATION ERRORSOf the 25 events occurring with a subcuta-neous injection of a drug, wrong-dose errors represented 36%, followed by wrong-drug errors (20%) and drug-not-given events and delays in administering an injection (both

categories represented 12% of events involv-ing subcutaneous injections). Significantly, 88% of the events entailed a high-alert medication—either insulin or anticoagulants, such as heparin and low-molecular-weight heparin.

For example, one report described a near-miss, wrong-dose event with insulin:

A sliding-scale insulin regimen was supposed to be discontinued but remained in the MAR the following day.

In another report, a patient received too much of a low-molecular-weight hepa-rin because the patient’s recorded weight was misrepresented as kilograms rather than pounds:

Lovenox (enoxaparin) was given based on 1mg/kg and administered every 12 hours. The patient’s weight was recorded as 114 kg rather than the actual weight of 52kg (114 pounds), resulting in a medication overdose.

PRESS COPY, D

O NOT D

ISTRIB

UTE



While it may seem that an accident is the result of an obvious error or problem, further analysis is necessary to identify a number of underlying circumstances or systems issues that caused the error. Unfor-tunately, with many administration events, the tendency is to focus primarily on the individual performance of the nurse who gave the medication. Consider the following report, which cited the nurse’s lack of expe-rience as contributing to the event when illegible handwriting and transcription errors were more likely the cause:

The handwritten transcription of an order in the MAR incorrectly stated the patient was to receive 2.5 mg of methotrexate weekly. The nursing student administered this amount. The actual dose should have been 20 mg, or 8 tablets of methotrex-ate, each containing 2.5 mg.

To prevent medication administration errors, nurses are often reminded to adhere to the “five rights” of safe medication administration: right patient, right drug, right time, right dose, and right route of administration (Cohen). For some caregiv-ers, the list of rights to follow has been expanded to include considerations such as right reason, right documentation, right response, and right to refuse.

Without systems in place at the orga-nization to help caregivers administer

medications safely, errors will occur. How can a nurse provide the right dose of a new medication if information about the medi-cation or a pharmacist is not readily avail-able? Isn’t a mistake more likely to occur if two patients with the same last name are admitted to the same hospital room? One nurse has taken the idea of the five rights to identify nurses’ rights for safe medication administration, ranging from the right to have a complete and clear order to the right to stop, think, and be vigilant when adminis-tering medications (Cook).

System-Based Causes of Medication ErrorsAdditionally, focusing solely on the five rights ignores the human factors and systems-based causes of errors. A human factors approach analyzes an individual’s performance within the context of the surrounding environment. A nurse who accidently misprograms an infusion pump by omitting a decimal point may still think that the setting is correct, even after check-ing the display. Human factors researchers call this confirmation bias, underscoring the importance of independent double checks. Asking another nurse to independently check the infusion dose and confirm the setting will reduce the likelihood of error.

A systems approach to safety views acci-dents as the result of multiple faults within a system that occur together in an unan-ticipated interaction (Reason). Did poor lighting, inadequate staffing, or poorly designed medical devices, along with a long list of other systems issues, create a situation that enabled the error to occur?

Of the 320 administration-only events submitted to ECRI Institute PSO over a five-week period, only 13 events identi-fied contributing factors for the event. According to the information presented in the reports, staffing issues, such as inex-perienced staff, were identified for 38% of events. In descending order, the other categories of contributing factors identified were: team coordination challenges, such as communication failures, and workforce matters, such as failing to follow required protocol (identified in 24% of events for both factors); operating environment deficiencies, such as distractions (7%); management issues, such as inadequate ori-entation of staff (3%); and patient-related factors, such as noncompliance (3%). Of all the contributing factors, communication failures were the most frequently listed.

The results underscore the current ten-dency to focus on human failures rather than the systems-based causes of medication

Contributing Factors to Medication Administration Events

XX The findings underscore the mistaken tendency to focus on human failures rather than the system-based causes of medication errors.

XX A medication safety plan incorporates a hierarchy of error-reduction techniques.

AT A GLANCE

PRESS COPY, D

O NOT D

ISTRIB

UTE



errors. Here is another medication-adminis-tration event that lists the nurse’s failure to follow the medication order for the patient as the contributing factor when a noisy environment may have caused the nurse to mishear the doctor’s order:

The nurse took a verbal order and administered Depo-Medrol (methylprednisolone acetate) intra-venously. A review of the order after the physician added it to the patient’s chart indicated that the order was for Solu-Medrol (methylprednisolone sodium succinate) intravenously. Depo-Medrol is an intramuscular injection.

Identifying root causes is one of the main goals of event investigation and depends heavily on identifying the systems

issues that set people up to make mistakes. The focus should not be on those closest to the error, such as the nurse who administers the wrong drug. This type of approach lim-its an event investigation to the proximate causes of errors, or the more superficial and readily apparent causes of an error. Proximate causes are the most immedi-ate reasons for a mishap—a nurse pulled the wrong medication from a dispensing cabinet or gave a medication to the wrong patient after bypassing the organization’s bar-code medication administration system. By contrast, a root-cause analysis identifies the systems issues that caused the problem. Instead of asking, “Why did the person make the mistake?” a root-cause analysis

continues to ask, “Why did the mistake occur?” in order to identify ways to prevent the error (Scanlon and Karsh). Were two look-alike drugs stored next to each other in a dispensing cabinet, making it likely for the nurse to select the wrong drug? If the issue is human performance, ask what systems or processes caused such a performance problem. Are staff performing workarounds because the system is too hard to follow?

Hierarchy of Error-Reduction TechniquesExperts in system safety have developed a hierarchy of error-reduction techniques based on the impact that they can have in preventing errors (ECRI Institute

Low Impact

ModerateImpact

HighImpact

—Simplify the process

—Standardize to reduce process variability

—Minimize choices

—Increase detectability

—Optimize redundancy—Document

—Educate or train

—Implement policies

—Automate

—Incorporate forcing functions

—Incorporate fail-safe

mechanisms

MS11

546

Figure 4. Strength of Error-Reduction Strategies

PRESS COPY, D

O NOT D

ISTRIB

UTE



“Medication”). The high-impact strategies, which “design out” the hazard, are consid-ered the most powerful and most desirable strategies because they can eliminate haz-ards. For example, an organization can reduce the risk of medication overdoses by dispensing medications in single-dose units only. High-impact strategies also incor-porate fail-safe mechanisms and forcing functions that provide a barrier or safe-guard to prevent a hazard from adversely affecting the process. An example of a forcing function is an infusion pump that will not allow delivery of an infusion if a programmed dose exceeds preset limits.

Moderate-impact strategies do not eliminate a hazard but use techniques, such

as standardization, process simplification, warnings, and alarms, to reduce the likeli-hood that errors will occur. While effective, these error-reduction techniques are rated as moderate because they are highly dependent on the behavior of people using the system. For example, audible alarms on infusion pumps increase caregivers’ ability to detect possible errors. But if the audible alarm on an infusion pump is set to a low volume so that a sleeping patient is not disturbed, nurses at a nearby nursing station may never hear the pump alarm.

Low-impact strategies use special policies and procedures and education to reduce errors. For example, a policy requir-ing annual training on IV infusion safety

is intended to ensure the competency of nurses who administer IV medications. By itself, the policy will not prevent mishaps, but, in combination with other measures, such as using infusion pumps with forcing functions, it will contribute to overall safety.

In sum, consider the various high-, moderate-, and low-impact approaches listed in Figure 4 when developing safety strategies for the medication-use process. A safety plan can include all three approaches; however, choose those safety strategies that will have a higher impact in preventing medication errors as often as possible. The goal is to redesign the medication-use process to make it harder for errors to reach the patient.

PRESS COPY, D

O NOT D

ISTRIB

UTE


©2011 ECRI Institute. May be disseminated, for internal educational purposes solely at the subscribing site. For broader use of these copyrighted materials, please contact ECRI Institute to obtain proper permission. DECEMBER 201112 DECEMBER 201112

LEADERSHIP STRATEGIESLow Impact Moderate Impact High ImpactShow leadership support for medication safety initiatives and

provide the resources to support change.

Provide resources and support for educational programs about medication safety for all those involved in the medication-use process.

Involve frontline staff in evaluating medication safety initiatives to ensure that the changes make sense and will be adopted.

Adopt a systems approach to medication safety with an emphasis on addressing the systems-related issues that enable people to make mistakes.

Assess the overall medication-use process to identify its strengths and weaknesses.

Apply proactive risk assessment to evaluate high-risk medication processes.

Use multiple approaches to track and identify medication errors (e.g., incident report review, direct observation, chart review, and trigger tools for targeted medical record review).

Use root-cause analysis to understand the underlying causes of medication errors and to identify strategies to prevent recurrence of the errors.

Evaluate measures to improve workflow in patient care areas and to minimize distractions of caregivers administering medications.

Whenever new technologies are adopted, evaluate whether they can introduce new types of errors and address those risks.

Improve the medication-use process by making it harder for errors to reach patients.

Consider technologies designed to enhance safety of medication administration: bar-code medication administration systems, automated dispensing cabinets, and electronic medication administration records.

SAFE MEDICATION ADMINISTRATION PRACTICESLow Impact Moderate Impact High ImpactEnsure that comprehensive information about the patient

(e.g., weight, diagnosis, allergies, etc.) is easily available to caregivers involved in the medication management process.

Make comprehensive reference information concerning medications and related health data readily accessible to providers.

Create a hierarchy of time-critical medications and other scheduled and non-time-critical medications to minimize errors that can occur when caregivers feel pressed to precisely time medication administration. Dosing schedules should be based on this hierarchy, which would allow those patients who need prompt dosage to get it.

Involve patients and their families in the medication process.

Require at least two patient identifiers when administering medications (e.g., name, date of birth, or address).

Address the safe use of look-alike or sound-alike medications by developing a list of such medications provided in the organization; taking action to prevent errors involving the confusion of drug names and look-alike drugs; and annually reviewing and revising the list of such drugs.

Standardize formats for medication orders to limit the use of verbal orders.

In written orders, prohibit error-causing abbreviations and require a zero before a decimal point, as well as eliminate trailing decimal points and zeroes.

Establish dosing protocols for drugs such as insulin that often have multiple protocols for adjusting insulin dosages to keep patients’ blood sugar levels stable.

Provide caregivers with dose-determination charts to eliminate the risk of errors in calculating dosages.

Label all medications, including any syringe of a medication or solution.

Whenever feasible, institute a 24-hour pharmacy service so that pharmacists are readily available to caregivers to answer questions that arise (if this is not possible, ensure that providers have access to an on-call pharmacist when a pharmacist is not on-site).

Limit nurse access to the pharmacy during pharmacy after hours by providing automated dispensing cabinets that offer a limited variety and amount of prepackaged pharmaceuticals.

Dispense medications in single-unit doses.

©2011 ECRI Institute. May be disseminated, for internal educational purposes solely at the subscribing site. For broader use of these copyrighted materials, please contact ECRI Institute to obtain proper permission.

Risk Reduction Strategies to Improve Medication Safety by Strength of Impact

PRESS COPY, D

O NOT D

ISTRIB

UTE



SAFE IV PRACTICESLow Impact Moderate Impact High ImpactProvide education on IV medication administration to all

healthcare professionals, including agency and per diem staff, involved with IV infusions and evaluate their understanding of the information.

Annually evaluate healthcare professionals’ competency in IV administration techniques.

Standardize infusion pumps available in the organization.

Limit the number of concentrations available for each infusion solution.

Require pharmacy preparation of IV solutions.

Establish standardized dosing methods for medications.

Before an infusion of a high-alert drug, require independent doublechecks of the drug, dose concentration, line connection, patient identification and infusion pump setting.

When multiple lines are used to administer several solutions, label each infusion line and require independent doublechecks to trace each line to the insertion site.

Purchase infusion pumps with dose error reduction features to reduce the risk of administration errors with IV infusions.

SAFE PRACTICES FOR HIGH-ALERT MEDICATIONSLow Impact Moderate Impact High ImpactIdentify all high-alert medications provided in the organization.

Develop educational programs to heighten clinical practitioners’ awareness of high-alert medication risks.

Limit the variety of high-alert medications available in the organization’s formulary.

Eliminate high-alert drugs from clinical areas.

Limit the number of concentrations and volumes of high-alert medications.

Require independent doublechecks of doses and preparations of high-alert medications.

Label high-alert drugs with a prominent sticker, such as one in the shape of a stop sign, reminding staff to conduct an independent doublecheck of the drug dose.

Label all neuromuscular blocking agents with a warning that clearly identifies the drug, whether in a vial or a syringe, as a paralyzing agent.

Have a pharmacist review all high-alert medication orders.

Use premixed infusion products and remove vials of high-alert drugs from stock.

ECRI Institute PSO has prepared a toolkit for healthcare facilities that enables them to consider each safety strategy for medication administration discussed in this report, identify any action required by the facility, and assign a deadline for implementation. The toolkit is available as a supplement to this report. A downloadable version of the toolkit is available on your PSO member site.

PRESS COPY, D

O NOT D

ISTRIB

UTE



Hospitals are responsible for administer-ing thousands of medication doses to their patients every day. One 450-bed hospital’s estimate of two million medication doses administered annually (Mutter) represents about 5,500 medication doses every day. Given the likelihood identified from this analysis that the majority of medication errors occur during drug administration, healthcare facilities can improve safety for thousands of patients by turning their atten-tion to safe administration practices.

Detecting Administration Errors Is ChallengingIn addition to their high frequency, admin-istration errors are among the most serious medication errors. Preventing harmful errors at the administration stage is particularly challenging because there is

no one between the nurse and the patient who might detect the potential error and intercept it. In a widely referenced systems analysis of ADEs in the medication-use process, researchers found that most ADEs occurred in the ordering stage (39%) and administration stage (38%) (Leape et al.). The researchers included intercepted ADEs in their study and found that only 2% of ADEs were intercepted at the administra-tion node; the interception rate was higher for all the other nodes.

There is no one solution to reduce the frequency and severity of medication administration mistakes; medication safety approaches should harness a combina-tion from the hierarchy of error-reduction techniques.

To assist healthcare organizations in selecting error-reduction strategies for

medication administration, ECRI Institute PSO has grouped the strategies discussed in this report by the impact—whether high, medium, or low—in preventing errors. Refer to “Risk Reduction Strategies to Improve Medication Safety by Strength of Impact” on pages 12 and 13 for a summary of the strategies by their error-reduction impact.

Improving Medication Administration Safety

XX Medication administration events are among the most serious medication errors.

XX Healthcare facilities can improve patient safety by turning their attention to safe medication administration practices.

XX Medication safety must harness a combination of error-reduction techniques.

AT A GLANCE

ECRI Institute PSO has prepared a toolkit for healthcare facilities that enables them to consider each safety strategy for medication administration discussed in this report, iden-tify any action required by the facility, and assign a deadline for implementation. The toolkit is available as a supplement to this report. A downloadable version of the tool-kit is available on your PSO member site.

PRESS COPY, D

O NOT D

ISTRIB

UTE



A good starting point to improve medication administration is to gain an understanding of process breakdowns in this stage of the medication-use process. In other words, redesign starts with under-standing how errors occur. ECRI Institute and ISMP have developed assessment tools of the overall medication process, which can help organizations identify strengths and weaknesses with their medication adminis-tration processes (see “Online Resources” for information on accessing these and other tools for medication safety).

Many organizations also use proactive risk assessment methods—failure mode and effects analysis (FMEA), for exam-ple—to evaluate particular aspects of their medication processes and to identify and implement improvement strategies. Sample FMEAs for medication administration are available from the Institute for Healthcare Improvement’s website (see “Online Resources”).

Additionally, having senior executives conduct walkarounds to regularly visit hospital units and engage in meaningful discussions with clinicians and frontline staff about the issues and challenges they face in delivering safe patient care opens the

door for many significant opportunities for medical error reduction and patient safety improvement.

Additional information can be gleaned with the following methods to review medication practices: incident report review, direct observation of the medication-use process, chart review, and trigger tools for targeted medical record review (Meyer-Massetti et al.). Events identified by incident review and event reporting should be analyzed in a structured, step-by-step investigation, or root-cause analysis, to understand why the event happened and its underlying causes and to identify what can be done to prevent recurrence.

Any safety measures identified through root-cause analysis should be applied to the entire organization, not just the unit involved in the error. Staff should be provided with feedback about the analy-sis—and the error-prevention strategies put in place—so that they understand that their event reporting leads to safer patient care.

Essential Ingredients for Medication Safety InitiativesWhether looking at medication adminis-tration in a proactive risk assessment or

retroactively in a root-cause analysis of a serious event, there are two essential ingredients.

First, the organization’s leadership must support the initiative and be involved in improvement efforts, explain to staff why an improvement initiative is important, and whenever there is a change in process or introduction of new technology, support the necessary training to implement these changes (Mutter). Second, any review of the medication process must involve the frontline staff who complete the particular tasks under review. They know what is and is not working, and, if they are expected to implement process changes, their input is necessary to ensure that the changes make sense.

In short, the only way to sustain a rede-signed medication administration process is to ensure that staff understand and rec-ognize the importance of the changes and participate in developing the improvements (Kliger et al.).

Where to Start?

XX Evaluate the medication-use process to identify strengths and weaknesses.

XX Employ proactive and reactive risk assessment tools (e.g. failure mode and effects analysis, root-cause analysis) to identify and implement medication safety strategies.

XX Use event reporting, direct observation, chart review, and trigger tools to analyze and improve medication practices.

AT A GLANCE

PRESS COPY, D

O NOT D

ISTRIB

UTE



As many of the medication administration events reported to ECRI Institute PSO sug-gested, distractions are strongly linked to errors. It is not uncommon for a nurse, who is about to administer a medication, to be interrupted with a question about another patient or to be paged to assist with a differ-ent patient; a distracted nurse can then forget to unclamp an IV line, misread or mishear a drug order, or inadvertently pull the wrong medication from an automated dispensing cabinet. Researchers have found that each interruption of a nurse preparing and admin-istering medications is associated with a 13% increase in clinical errors (Westbrook et al.). Once interrupted, the nurse might forget to check a patient’s identification against the

MAR or might set the wrong IV rate, as the researchers found in their observations of 98 nurses preparing and administering more than 4,200 medications to 720 patients.

Measures that have been tried at some organizations to reduce interruptions include the following (George et al.; Kliger et al.; Relihan et al.; Westbrook et al.):

X Creating a “protected hour” when nurses only focus on reconciling medica-tion orders, administering medications, and charting administration of the medications.

X Providing nurses with a brightly colored vest to wear when administering medica-tions that signals to other staff that the

individual is in the process of giving patients their medications and should not be interrupted. Some organizations even label the vest to state, “Do not interrupt, mediation round in progress.”

X Setting aside a segregated “safe zone” for preparing and administering medications where intrusions are prohibited.

Of course, these measures are only effective if all staff—not just those administering medications—are aware of these strategies and agree not to interrupt a nurse who, for example, is in a designated area for preparing and administering medications.

Minimizing Distractions and Interruptions

XX Consider measures, such as setting aside a separate area for nurses preparing medications, to minimize interruptions and distractions.

XX Ensure that all staff are aware of these measures and their purpose.

AT A GLANCE

PRESS COPY, D

O NOT D

ISTRIB

UTE



Inevitably, any discussion about medication administration errors will raise questions about workload. Are nurses expected to accomplish too many tasks during their shifts, forcing them to work quickly and take short-cuts to complete a task? For example, one hospital found that its medication administra-tion events spiked on Wednesdays when the number of surgery cases, as well as surgical admissions to the inpatient unit, were highest. The errors were particularly high during eve-ning shift changes on that day.

In response, the hospital hired a shift coordinator to ensure that the workload was evenly distributed among nurses. The coordinator also assisted with some docu-mentation tasks so that nurses could spend more time with patients. The initiative reduced the spike in medication adminis-tration errors when admissions to the unit were highest. (Mutter)

Assess Impact of Workload Resdistribution Of course, any measure to change staff workload should be assessed for its impact on other caregivers. If nursing shifts are shortened, does the workload increase for the other clinicians who remain on the unit after the nursing shift change? (Scanlon and Karsh) And the measure should be evaluated to determine whether it achieves the desired goals. For example, research-ers thought having dedicated medication nurses available to exclusively administer medications would reduce medication administration errors. But a randomized, controlled trial found that the dedicated medication nurses did not have any differ-ence in error rates than general nurses who provided comprehensive care, including drug administration (Greengold et al.).

The researchers concluded that the two hospitals involved in the study required additional systems-based solutions to improve medication administration.

Additionally, some organizations may want to explore the set-up for their patient care areas to evaluate whether nurses can complete their tasks efficiently. Are medications and IV supplies stored close by so that the nurse need only walk a few steps to obtain the necessary supplies to prepare to administer medications? Is the set-up for medications standardized in an automated dispensing cabinet so nurses know where a medication is stored in the cabinet? Providing a work environment that reduces nurses’ inefficiency will help to promote patient safety. (George et al.)

Workload

XX Consider measures to improve workload distribution among nurses during medication administration.

XX Evaluate the impact of any workload redistribution on other caregivers.

XX Examine patient care areas to evaluate whether nurses can complete tasks for medication preparation and administration efficiently.

AT A GLANCE

PRESS COPY, D

O NOT D

ISTRIB

UTE



Technology Solutions

XX Use technology, such as bar-code medication administration systems, that incorporate functions to make it harder for errors to reach patients.

XX Allow sufficient time to plan for the adoption of technology to minimize medication administration errors.

XX Identify the potential for technology to introduce new errors and address those concerns.

XX Test the technology in a pilot setting before implementing a full-scale adoption.

Just as electronic prescribing systems are promoted to reduce errors in the prescribing node, bar-code medication administra-tion systems (also referred to as bar-code point-of-care technology), electronic MARs, automated dispensing cabinets, and “smarter” infusion pumps (discussed in Safe IV Practices) have been identified as technologies that can improve medication administration.

Bar-code medication administration sys-tems utilize bar-code technology to match a unique code printed on a medication with an ordered medication at the point of care. A licensed caregiver, such as a nurse or respira-tory therapist, scans a bar-coded label on the patient’s identification wristband and on the medication packaging; the system then con-firms that the patient has been prescribed that particular medication. If the scans do not match, the system issues a warning that requires the caregiver to investigate and intervene to correct the problem.

Some hospitals have integrated their bar-code medication administration systems with electronic MARs so that the pharmacist can electronically enter a physician’s order in the MAR. The nurse can also automatically

document administration of a patient’s medications with the bar-code scanner. An electronic MAR can also issue alerts to the nurse about next doses that are due or over-due doses, as well as cautions about certain medications or drug interactions.

In a recent study, researchers at a Massachusetts hospital found that using bar-code technology with an electronic MAR reduced medication errors associated with transcription and administration. The study, which classified errors involving early or late administration as timing errors and all others as nontiming errors, found that implementa-tion of the system resulted in a 41% relative reduction in nontiming medication errors and a 51% reduction in ADEs associated with nontiming errors. Timing errors fell by 27%, although the rate of ADEs associated with these errors did not change significantly. (Poon et al.)

Other studies have shown that bar-code medication administration systems can reduce medication errors by as much as 86% (“Medication errors occurring”).

Automated dispensing cabinets allow caregivers quick access to medications and

can be safer than using stock supplies and drug cabinets in the care unit. Pharmacy personnel periodically stock each cabinet with the medications needed for that care area. When a drug must be administered to a patient, the nurse accesses the appropriate storage compartment by entering the request on a computer-entry screen or keyboard incorporated into the cabinet.

Although the dispensing cabinets can vary significantly depending on the manu-facturer, they all store drugs in compart-ments that can be categorized according to function. For example, a unit might only store single-unit doses, dispensing the exact amount of the drug requested and prevent-ing the operator from accessing the remain-ing stock of drugs.

Automated dispensing cabinets are typi-cally networked to a pharmacy computer that maintains centralized control over drug dis-tribution. Organizations that use automated dispensing cabinets should monitor situations when a nurse must override the dispensing cabinet’s controls to obtain a medication without a pharmacist’s review and adopt measures to improve access to these medica-tions while ensuring patient safety.

AT A GLANCE

PRESS COPY, D

O NOT D

ISTRIB

UTE



Beware of Technology MisuseThese technologies are good examples of high-impact safety strategies that incorpo-rate automation and, sometimes, forcing functions. However, as with any technology, these systems are only effective if they are used properly.

For example, the Pennsylvania Patient Safety Reporting System (PSRS), an inde-pendent state agency that receives and analyzes reports of serious events and near misses in hospitals and other healthcare facilities, has issued advisories about misuse of both bar-code medication administration systems and automated dispensing cabinets because users circumvent the technology.

Dispensing cabinet users have used the inventory function to gain access to medica-tions without pharmacy screening and to remove medications for multiple patients while the cabinet is open (“Problems”).

In a report on bar-code medication administration systems, the Pennsylvania reporting program identified workarounds with this technology that can lead to errors. For example, the nurse might affix the

patient’s identification wristband to the bed-side rather than on the patient to expedite scanning or to avoid situations of having to wake a sleeping patient when an IV bag is hung (“Medication Errors Occurring”). Such workarounds can lead to serious harm.

A root-cause analysis of a patient death from an inadvertent IV infusion of fen-tanyl and bupivacaine instead of penicillin found that one of the proximate causes of the event was the ability of the nurse to circumvent the point-of-care bar-code technology, which would have detected the drug-selection error before the infusion was administered (Smetzer et al.).

The analysis further investigated the systems issues that led the nurse to use this particular workaround and that influenced the other proximate causes of the event.

Allow Time for Technology ImplementationExperts caution healthcare facilities not to underestimate the amount of planning and analysis necessary for selection and implementation of medication management

technologies and recommend the following guidelines (ECRI Institute “Medication”):

X Set clear objectives and goals for the system.

X Focus on the individual facility, and involve all caregivers who participate in medication management when assessing the current medication-use process.

X Match the technology to the workflow process when possible, but be prepared to modify the work process in order to capitalize on the technology’s advantages and safeguards.

X Test the technology in a pilot setting, and make any necessary changes to the full-scale implementation plan based on the findings from the pilot test.

X Offer training about using the technol-ogy to all staff members, and ensure that additional resources are available after the system is implemented and after any revisions are made.

X Use human-factors techniques in the implementation of new technologies to reduce workarounds from a poor system-user fit.

PRESS COPY, D

O NOT D

ISTRIB

UTE



Safe IV Practices

XX Purchase smarter infusion pumps with dose error reduction systems to reduce infusion errors caused by misprogramming.

XX Standardize infusion pumps available in the organization to enhance user familiarity with a pump’s operation.

XX Limit the number of concentrations available for each infusion solution.

XX Require pharmacy preparation of IV solutions. Limit nurse preparation of IV solutions to emergency situations, such as those in the ED and critical care unit.

Improvements in IV infusion pump technology with so-called “smarter pumps” have had an important role in reducing the risk of administration errors with IV infusions. The term “smarter” refers to a pump feature called a dose error reduction system (DERS); these systems are available on infusion pumps sold in the United States. A DERS is a set of software tools that check programmed doses against preset, hospital-developed limits specific to the drug and clinical application or location. These systems alert providers if programmed doses exceed the preset limits; some limits require confirmation before beginning delivery (called a “soft stop”), while others do not allow delivery at all (called a “hard stop”).

Use Smarter Infusion Pumps WiselyECRI Institute PSO believes that smarter infusion pumps can increase the safety of

infusion therapy by reducing infusion errors caused by misprogramming and by helping healthcare facilities standardize their medi-cation administration practices. However, pumps with DERS capability can prevent errors only if clinicians can find the right limits to support their dosing practices. Oth-erwise, they will find ways to circumvent the technology.

Facilities can adopt measures to ensure that smart infusion pumps are appropriately used so that users do not look for work-arounds. To seamlessly integrate the pump’s features into work practices, the smart infusion pump’s software for preset dose limits must be properly implemented and maintained. For example, facilities should establish a multidisciplinary team of nurses, pharmacists, and physicians to carefully develop drug libraries with predefined dose limits for the pumps. They should ensure regular review and maintenance of the

drug libraries so that information on new drugs and new uses for existing drugs can be updated and the need to modify existing dose limits can be monitored.

Drug libraries must be kept current with drug information from the facility’s pharmacy system and, if one is used, its computerized provider order-entry system so that drug names, IV drug concentrations, and other references to medications are used consistently in the organization. Some pumps have wireless capability so that pump libraries can be remotely updated; otherwise, the process of updating each pump library can be time consuming.

Facilities should regularly review data from the pump log to ensure user compli-ance and to support quality improvement activities. They should also provide train-ing and ongoing education for pump users regarding the purpose and function of

AT A GLANCE

PRESS COPY, D

O NOT D

ISTRIB

UTE



the drug library and the infusion pump’s operation.

Other measures to prevent IV infusion errors include the following:

X Standardize infusion pumps available in the organization so that users are com-petent with the pump’s operation; while different manufacturers’ pumps may look the same, they do not always operate similarly (Joint Commission “Infusion Pumps”). Purchase contracts with pump vendors should specify that the vendor will offer in-service education on the pump’s operation at the time of purchase and when upgrades are provided. Addi-tionally, the organization should have a system in place to monitor and respond to hazards, recalls, and alerts involving pumps used at its facilities.

X Limit the number of concentrations available for each infusion solution to

reduce the potential for wrong-concen-tration selection errors (Bates et al.).

X Require pharmacy preparation of IV solutions and limit nurse preparation of IV solutions to emergency situations, such as those in the ED and critical care unit. When nurses must mix IV solu-tions, mixing protocols should be readily available. (Smetzer and Cohen “Prevent-ing Drug Administration”)

X Establish standardized dosing methods for medications. The standard technique should be listed on preprinted or elec-tronic orders for applicable drugs, as well as listed on the MAR and drug container label. (Grissinger)

X Before an infusion—particularly an infusion of a high-alert medication—is administered, require an independent doublecheck of the drug, dose concentration, line connection, patient identification, and infusion pump setting.

An independent doublecheck is effective only when two clinicians separately check the information. (Smetzer and Cohen “Preventing Medication Errors”)

X When multiple lines are used to adminis-ter several solutions, label each infusion line and require an independent double-check to trace the line to the insertion site (Smetzer and Cohen “Preventing Medication Errors”).

X Provide education on IV medication administration, evaluate individuals’ understanding of the information, and ensure annual competency of admin-istration techniques. Education should include nurses, pharmacists, respiratory therapists, physicians, and anesthesia pro-viders. New staff and agency or per-diem staff should also be educated in the orga-nization’s IV medication administration procedures before providing patient care.

PRESS COPY, D

O NOT D

ISTRIB

UTE



High-Alert Medications

XX Refer to the list of high-alert medications provided in this report to identify all high-alert medications administered within the organization.

XX Eliminate high-alert drugs from floor stock in clinical areas.

XX Limit the variety of high-alert medications available in the organization’s formulary.

XX Require an independent doublecheck of doses and preparations of high-alert medications before they are administered.

High-alert medications were named in a large number of administration-only events, representing 37.8% (121) of these events. Among the most frequently identified high-alert medications were anticoagulants (involved in 21.5% of the 121 events), fol-lowed by opioids (20.7%), insulin (14%), and antiarrhythmic agents (9.1%).

As the events reported to ECRI Institute PSO found, many high-alert drugs are injected or infused parenterally (e.g., intra-venously, intramuscularly). These routes of administration provide a more direct means of delivering medication for faster and more effective results and are beneficial to patient care. However, high-alert drugs administered rapidly by direct routes can be dangerous when given erroneously, because there is little time to reverse their immedi-ate effects. Furthermore, when a high-alert drug is administered with an infusion device capable of rapid IV delivery, a mistake in

dosing or rate programming can have grave consequences.

High-Alert Drug Awareness Facility staff and physicians must be aware of the risks associated with high-alert drugs and take appropriate precautions to mini-mize these risks. Since November 1999, when the Joint Commission issued a Sentinel Event Alert on high-alert medications, the accrediting agency has addressed measures to prevent errors involving these drugs in its National Patient Safety Goals, Sentinel Event Alerts, and the medication management chapter of its accrediting standards (Joint Commission “High-Alert”).

Error prevention with high-alert medi-cations starts with an awareness of those high-alert drugs that have a heightened risk of causing significant harm when they are used in error. The Joint Commission’s medication management standards require accredited healthcare facilities to make a

list of high-alert medications used in their organizations and to develop manage-ment processes to prevent the occurrence of errors and sentinel events involving these medications (Joint Commission Comprehensive Accreditation).

There are some key strategies that can be used to prevent errors involving high-alert medications. Many of these strategies involve changes to existing processes, func-tions, and practices. Therefore, the approach to adopting and implementing them should be multidisciplinary; should include leaders, managers, and frontline staff; and should be part of an overall medication safety pro-gram. Facilities should engage physicians,

ISMP has published one of the most com-prehensive lists of high-alert medications. This list is reprinted in the Appendix and is also available online at ISMP’s website (see “Online Resources”).

AT A GLANCE

PRESS COPY, D

O NOT D

ISTRIB

UTE



pharmacists, nurses, patients, and various others with a stake in the medication process when planning and implementing these strat-egies. Key strategies include the following (ECRI Institute “High-Alert”; Cohen et al.):

X Eliminate high-alert drugs from floor stock in clinical areas. For instance, the Joint Commission’s medication man-agement standards require accredited facilities to remove concentrated electro-lytes from patient care areas.

X Limit the variety of high-alert medica-tions available in the organization’s formulary.

X Limit the number of concentrations and volumes of high-alert medications.

X Review manufacturers’ directions for administration of high-alert medications, and ensure the instructions are followed.

X When automated medication administra-tion systems are used, include warnings about high-alert drugs.

X Require independent doublechecks of doses and preparations of high-alert medications.

X Label high-alert drugs with a prominent sticker, such as one in the shape of a stop sign, reminding staff to conduct an inde-pendent doublecheck of the drug dose.

X Use premixed infusion products, and remove vials of high-alert drugs from stock.

X Label all neuromuscular blocking agents with a warning that clearly identifies the drug, whether in a vial or a syringe, as a paralyzing agent.

X Have a pharmacist review all high-alert medication orders. If this is not possible, ensure that providers have access to an on-call pharmacist when a pharmacist is not on site.

Another important safety strategy for high-alert medications involves incorporat-ing specific safety measures available for

individual high-alert medications, particu-larly those medications associated with a large percentage of errors. ISMP says that about 11% of serious medication errors involve insulin misadministration and that almost 9% involve heparin (Focus).

Tips to improve heparin safety are included in the Joint Commission’s September 2008 Sentinel Event Alert on preventing errors related to commonly used anticoagulants (Joint Commission “Preventing Errors”). PSRS, the Pennsylvania event reporting program, has also published reports on preventing errors with heparin and insulin (“Medication Errors with the Dosing”; “Let’s Stop”). Additionally, organizations can search ISMP’s website for information on rec-ommended safety measures for specific high-alert medications.

PRESS COPY, D

O NOT D

ISTRIB

UTE



General Safety Strategies for Medication Administration

XX Ensure that comprehensive information about the patient is easily available to caregivers administering medications.

XX Make comprehensive and up-to-date reference information about medications readily accessible to caregivers.

XX Ensure that pharmacists are readily available to caregivers to answer questions that may arise.

XX Label all medications, including any syringe or solution bag containing medication.

XX Dispense medications in ready-to-use single doses.

XX Involve patients and their families in the medication process.

All healthcare organizations should have in place medication policies and procedures that address the medication-use process, including drug administration. The Joint Commission’s medication management standards for accredited healthcare facili-ties, as well as its National Patient Safety Goals, address many critical measures applicable to administration. In addition, ISMP, the American Society for Health-System Pharmacists, and others have published numerous recommendations on safe medication administration (see “Online Resources”). Recommendations from these organizations include the fol-lowing (Joint Commission “Comprehensive Accreditation”; Joint Commission “2011”;

Smetzer and Cohen “Preventing Drug Administration”):

X Ensure that comprehensive information about the patient (e.g., weight, diagnosis, allergies, laboratory and test results, cur-rent medications, pregnancy status) is easily available to caregivers involved in the medication management process.

X Require at least two patient identifiers when administering medications (e.g., name, date of birth, address). The patient’s room number and location can-not be used as an identifier.

X Address the safe use of look-alike or sound-alike medications by developing a list of such medications provided in the organization; taking action to prevent

errors involving the confusion of drug names and look-alike drugs (e.g., elimi-nating drugs that can be confused from the organization’s formulary, if possible; developing safe approaches to storing drugs that can be confused with other drugs); and annually reviewing and revis-ing the list of such drugs.

X Standardize formats for medication orders to limit the use of verbal orders. When verbal orders are needed (e.g., an emergent or urgent situation), require that the person taking the order write it down and read back the order.

X In written orders, prohibit error-causing abbreviations, require a zero before a decimal point, and eliminate trailing deci-mal points and zeroes.

AT A GLANCE

PRESS COPY, D

O NOT D

ISTRIB

UTE



X Make comprehensive reference informa-tion concerning medications and related health data readily accessible to providers.

X Establish dosing protocols for drugs such as insulin that often have multiple protocols for adjusting insulin dosages to keep patients’ blood sugar levels stable. Encourage physicians to adopt the orga-nization’s dosing protocols.

X Provide caregivers with dose-determina-tion charts to eliminate the risk of errors in calculating dosages.

X Dispense medications in ready-to-use single doses.

X Label all medications, including any syringe or solution bag containing medication. Ensure that specific areas and procedures that use medications

delivered in syringes (e.g., anesthesia, procedure rooms, diagnostic imaging) comply with the labeling requirement.

X Create a hierarchy of time-critical medications and other scheduled and non-time-critical medications to minimize errors that can occur when caregivers feel pressed to precisely time medication administration. Dosing schedules should be based on this hier-archy, which would allow those patients who need prompt dosage to receive it.

X Limit nurse access to the pharmacy after hours by providing automated dispensing cabinets that offer a limited variety and amount of prepackaged pharmaceuticals.

X Whenever feasible, institute a 24-hour pharmacy service so that pharmacists are

readily available to caregivers to answer questions that arise. If this is not possi-ble, ensure that providers have access to an on-call pharmacist when a pharmacist is not on site.

X Involve patients and their families in the medication process. Patients have a right to know about the medications they are receiving, how often they are admin-istered, and the possible side effects. Patients should be encouraged to speak up if they have questions about their medications, and providers should know to listen to patients’ concerns. Informed patients serve as another line of defense against medication errors if they are empowered to question anything about their medications that does not appear to be correct.

PRESS COPY, D

O NOT D

ISTRIB

UTE



Education

XX Educate caregivers, including temporary staff, about the organization’s medication policies and procedures.

XX Provide information and training, if needed, when new drugs or drug delivery systems are adopted.

XX Involve pharmacists in the development and delivery of medication safety education programs.

While education of those involved in the medication-use process will not, by itself, eliminate medication errors, educa-tional programs about the organization’s medication policies and procedures are an important component to heighten prac-titioners’ awareness of medication safety strategies and error prevention.

An education program should include review of the common types and causes of medication errors using case studies

as examples. Educators should stress the importance of timely and thorough report-ing of medication events and near misses. They should also underscore the importance of clarifying an order that is unclear.

Promote Pharmacy Staff ’s Know-HowEducational sessions should also be insti-tuted for any new drugs or drug delivery systems (e.g., catheters, infusion devices) to ensure that the drugs and devices are

used safely and effectively. Involvement of pharmacists in the development and delivery of medication safety education programs can increase the depth and visibility of such programs. Having pharmacists participate in these programs allows clinicians and staff to get to know the institution’s pharmacy staff and promotes use of their knowledge—a valuable resource in safe medication management.

AT A GLANCE

PRESS COPY, D

O NOT D

ISTRIB

UTE



ReferencesAgency for Healthcare Research and Quality

(AHRQ) Innovations Exchange. Intravenous infusion safety initiative prevents medication errors, leading to cost savings and high nurse satisfaction [online]. 2009 Nov 11 [cited 2011 Sep 19]. Available from Internet: http://www.innovations.ahrq.gov/content.aspx?id=2375.

Bates DW, Vanderveen T, Seger D, et al. Variability in intravenous medication practices: implica-tions for medication safety. Jt Comm J Qual Patient Saf 2005 Apr;31(4):201-10.

Cohen MR. Causes of medication errors. Chapter 4. In: Cohen MR, ed. Medication errors. Washing-ton (DC): American Pharmacists Association; 2007.

Cohen MR, Smetzer JL, Tuohy NR, et al. High-alert medications: safeguarding against errors. Chapter 14. In: Cohen MR, ed. Medication errors.

Washington (DC): American Pharmacists Association; 2007.

Cook MC. Nurses’ six rights for safe medication administration [online]. [cited 2011 Sep 19]. Available from Internet: http://www. massnurses.org/nursing-resources/ nursing-practice/articles/six-rights.

ECRI Institute:

High-alert medications. Healthc Risk Control 2004 Mar;4: Pharmacy and medications 1.4:1-7.

Medication safety. Healthc Risk Control 2011 May;4: Pharmacy and medications 1:1-32.

Focus on high-alert medications. PA PSRS Patient Saf Advis [online]. 2004 Sep [cited 2011 Sep 19]. Available from Internet: http:// patientsafetyauthority.org/ADVISORIES/AdvisoryLibrary/2004/Sep1(3)/Pages/06.aspx.

George EL, Henneman EA, Tasota FJ. Nursing implications for prevention of adverse drug events in the intensive care unit. Crit Care Med 2010 Jun;38(6 Suppl):S136-44.

Greengold NL, Shane R, Schneider P, et al. The impact of dedicated medication nurses on the medication administration error rate. Arch Intern Med 2003 Oct 27;163(19):2359-67.

Grissinger M. Lack of standard dosing units and measurements creates mishaps in intravenous drug administration. P T 2011 Jan;36(1):10, 21.

Harrison’s Practice. Drug monographs: vancomy-cin [online]. 2011 Aug 22 [cited 2011 Sep 30]. Available from Internet: http://www. harrisonspractice.com/practice/ub/view/DrugMonographs/156696/all/Vancomycin.

Hicks RW, Becker SW. An overview of intrave-nous-related medication administration errors

Online ResourcesXX American Society of Health-System

Pharmacists — Guidelines on preventing

medication errors in hospitals. Available from Internet: http://www.ashp.org/DocLibrary/BestPractices/MedMisGdlHosp.aspx.

XX ECRI Institute* — INsight™ web-based risk

assessment for medication safety practices. More information available at https://www.ecri.org/Products/Pages/INsight_Web_Based_Risk_Assessment_for_Medication_Safety_Practices.aspx.

— Medication safety. Healthc Risk Control 2011 May;4:Pharmacy and medications 1:1-32. Available from Internet to ECRI Institute PSO Plus members at: https://members2.ecri.org/Components/HRC/Pages/Pharm1.aspx#resource.

— Medication safety self-assessment questionnaire. Healthc Risk Control 2007 Sep;1:Self-assessment questionnaires 31:1-24. Available from Internet to ECRI Institute PSO Plus members at: https://members2.ecri.org/Components/HRC/Pages/SAQ31.aspx.

XX Institute for Healthcare Improvement — FMEA tools for medication

administration. Available from Internet: http://app.ihi.org/Workspace/tools/fmea/AllTools.aspx#10.

XX Institute for Safe Medication Practices — Acute care guidelines for timely

administration of schedule medications. Available from Internet: http://ismp.org/Tools/guidelines/acutecare/tasm.pdf.

— List of high-alert medications. Available from Internet:

http://www.ismp.org/Tools/highalertmedications.pdf.

— Medication safety self assessment for hospitals. Available from Internet: http://ismp.org/selfassessments/Hospital/2011/Default.asp.

— Medication safety tools and resources. Available from Internet: http://ismp.org/Tools/default.asp.

XX Joint Commission — National patient safety goals.

Available from Internet: http://www.jointcommission.org/standards_information/npsgs. aspx.

— Sentinel event alerts. Available from Internet: http://www.jointcommission.org/sentinel_event.aspx.

* For information on obtaining ECRI Institute reports, contact ECRI Institute PSO at [email protected].

PRESS COPY, D

O NOT D

ISTRIB

UTE



as reported to MEDMARX®, a national medi-cation error-reporting program. J Infus Nurs 2006 Jan-Feb;29(1):20-7.

Institute for Safe Medication Practices (ISMP). ISMP’s list of high-alert medications [online]. 2008 [cited 2011 Sep 19]. Available from Inter-net: http://www.ismp.org/Tools/ highalertmedications.pdf.

Institute of Medicine (IOM). Preventing medication errors. Washington (DC): National Academies Press; 2006.

Joint Commission:

2011 National Patient Safety Goals [online]. 2011 Jul 7 [cited 2011 Sep 19]. Available from Internet: http://www.jointcommission.org/hap_2011_npsgs.

Comprehensive accreditation manual for hospitals: official handbook. Oakbrook Terrace (IL): Joint Commission; 2011 Mar.

High-alert medications and patient safety [online]. Sentinel event alert 1999 Nov 19 [cited 2011 Sep 19]. Available from Internet: http://www.jointcommission.org/sentinel_event_alert_issue_11_high-alert_medications_ and_patient_safety.

Infusion pumps: preventing future adverse events [online]. Sentinel event alert 2000 Nov 30 [cited 2011 Sep 19]. Available from Internet: http://www.jointcommission.org/sentinel_event_alert_issue_15_infusion_pumps_preventing_future_adverse_events.

Preventing errors relating to commonly used anticoagulants [online]. Sentinel event alert 2008 Sep 24 [cited 2011 Sep 19]. Available from Internet: http://www.jointcommission.org/sentinel_event_alert_issue_41_ preventing_errors_relating_to_commonly_used_anticoagulants.

Summary data of sentinel events reviewed by the Joint Commission [online]. 2010 Dec 31 [cited 2011 Sep 19]. Available from Internet: http://www.jointcommission.org/assets/1/18/SE_Data_Summary_4Q_2010_(v2).pdf.

Kliger J, Blegen MA, Gootee D, et al. Empower-ing frontline nurses: a structured intervention enables nurses to improve medication adminis-tration accuracy. Jt Comm J Qual Patient Saf 2009 Dec;35(12):604-12.

Leape LL, Bates, DW, Cullen DJ, et al. Systems analysis of adverse drug events. JAMA 1995 Jul 5;274(1):35-43.

Let’s stop the bleeding: preventing errors with hepa-rin therapy. PA PSRS Patient Saf Advis [online]. 2006 Dec [cited 2011 Sep 19]. Available from Internet: http://patientsafetyauthority. org/ADVISORIES/AdvisoryLibrary/2006/Dec3(4)/Pages/31.aspx.

Medication errors occurring with the use of bar-cod administration technology. Pa Patient Saf Advis [online] 2008 Dec [cited 2011 Sep 19]. Available from Internet: http:// patientsafetyauthority.org/ADVISORIES/AdvisoryLibrary/2008/Dec5(4)/Pages/ 122.aspx.

Medication errors with the dosing of insulin: problems across the continuum. Pa Patient Saf Advis [online] 2010 Mar [cited 2011 Sep 19]. Available from Internet: http:// patientsafetyauthority.org/ADVISORIES/ AdvisoryLibrary/2010/Mar7(1)/Pages/09.aspx.

Meyer-Massetti C, Cheng CM, Schwappach DL, et al. Systematic review of medication safety assessment methods. Am J Health Syst Pharm 2011 Feb 1;68(3):227-40.

Moss J, Berner E, Bothe O, et al. Intravenous medi-cation administration in intensive care: oppor-tunities for technological solutions. AMIA Annu Symp Proc 2008 Nov 6:495-9.

Mutter M. One hospital’s journey toward reduc-ing medication errors. Jt Comm J Qual Saf 2003 Jun;29(6):279-88.

National Quality Forum (NQF). NQF releases updated serious reportable events [press release online]. 2011 Jun 13 [cited 2011 Sep 19]. Available from Internet: http://www.

qualityforum.org/News_And_Resources/ Press_Releases/2011/NQF_Releases_Updated_Serious_Reportable_Events.aspx.

Poon EG, Keohane CA, Yoon CS, et al. Effect of bar-code technology on the safety of medica-tion administration. N Engl J Med 2010 May 6; 362(18):1698-707.

Problems associated with automated dispensing cabinets. PA PSRS Patient Saf Advis [online] 2005 Sep [cited 2011 Sep 19]. Available from Internet: http://patientsafetyauthority.org/ADVISORIES/AdvisoryLibrary/2005/sep2(3)/Pages/21.aspx.

Reason J. Managing the risk of organizational accidents. Brookfield (VT): Ashgate Press; 1997.

Scanlon MC, Karsh BT. Value of human factors to medication and patient safety in the intensive care unit. Crit Care Med 2010 Jun;38(6 Suppl):S90-6.

Smetzer J, Baker C, Byrne FD, et al. Shaping systems for better behavioral choices: lessons learned from a fatal medication error. Jt Comm J Qual Patient Saf 2010 Apr;36(4):152-63.

Smetzer JL, Cohen MR. Preventing drug administration errors. Chapter 11. In: Cohen MR, ed. Medication errors. Washington (DC): American Pharmacists Association; 2007.

Smetzer JL, Cohen MR. Preventing medication errors related to drug delivery devices. Chapter 12. In: Cohen MR, ed. Medication errors. Washington (DC): American Pharmacists Association; 2007.

Westbrook JI, Woods, A, Rob MI, et al. Association of interruptions with an increased risk and severity of medication administration errors. Arch Intern Med 2010 Apr 26;170(8):683-90.

PRESS COPY, D

O NOT D

ISTRIB

UTE

2011 ECRI Institute. May be disseminated, for internal educational purposes solely at the subscribing site. For broader use of these copyrighted materials, please contact ECRI Institute to obtain proper permission.DECEMBER 2011 29


2008 ECRI Institute. May be disseminated, for internal educational purposes solely at the subscribing site.For broader use of these copyrighted materials, please contact ECRI Institute to obtain proper permission.

Institute for Safe Medication Practices www.ismp.org

Institute for Safe Medication Practices

Classes/Categories of Medications

adrenergic agonists, IV (e.g., epinephrine, phenylephrine, norepinephrine)

adrenergic antagonists, IV (e.g., propranolol, metoprolol, labetalol)

anesthetic agents, general, inhaled and IV (e.g., propofol, ketamine)

antiarrhythmics, IV (e.g., lidocaine, amiodarone)

antithrombotic agents (anticoagulants), including warfarin, low-molecular-weightheparin, IV unfractionated heparin, Factor Xa inhibitors (fondaparinux), directthrombin inhibitors (e.g., argatroban, lepirudin, bivalirudin), thrombolytics (e.g.,alteplase, reteplase, tenecteplase), and glycoprotein IIb/ IIIa inhibitors (e.g., eptifi-batide)

cardioplegic solutions

chemotherapeutic agents, parenteral and oral

dextrose, hypertonic, 20% or greater

dialysis solutions, peritoneal and hemodialysis

epidural or intrathecal medications

hypoglycemics, oral

inotropic medications, IV (e.g., digoxin, milrinone)

liposomal forms of drugs (e.g., liposomal amphotericin B)

moderate sedation agents, IV (e.g., midazolam)

moderate sedation agents, oral, for children (e.g., chloral hydrate)

narcotics/opiates, IV, transdermal, and oral (including liquid concentrates, immediateand sustained-release formulations)

neuromuscular blocking agents (e.g., succinylcholine, rocuronium, vecuronium)

radiocontrast agents, IV

total parenteral nutrition solutions

ISMP’s List of High-Alert Medicationsigh-alert medications are drugs that bear a heightened risk ofcausing significant patient harm when they are used in error.

Although mistakes may or may not be more common with thesedrugs, the consequences of an error are clearly more devastating topatients. We hope you will use this list to determine which medica-tions require special safeguards to reduce the risk of errors. Thismay include strategies like improving access to information about

these drugs; limiting access to high-alert medications; usingauxiliary labels and automated alerts; standardizing the ordering,storage, preparation, and administration of these products; andemploying redundancies such as automated or independent double-checks when necessary. (Note: manual independent double-checksare not always the optimal error-reduction strategy and may not bepractical for all of the medications on the list).

Specific Medications

colchicine injection***

epoprostenol (Flolan), IV

insulin, subcutaneous and IV

magnesium sulfate injection

methotrexate, oral, non-oncologic use

opium tincture

oxytocin, IV

nitroprusside sodium for injection

potassium chloride for injection concentrate

potassium phosphates injection

promethazine, IV

sodium chloride for injection, hypertonic (greater than 0.9% concentration)

sterile water for injection, inhalation, and irrigation (excluding pour bottles) in containers of 100 mL or more

Background

Based on error reports submitted to the USP-ISMP Medication Errors ReportingProgram, reports of harmful errors in the literature, and input from practitioners andsafety experts, ISMP created and periodically updates a list of potential high-alertmedications. During February-April 2007, 770 practitioners responded to an ISMPsurvey designed to identify which medications were most frequently consideredhigh-alert drugs by individuals and organizations. Further, to assure relevance andcompleteness, the clinical staff at ISMP, members of our advisory board, and safetyexperts throughout the US were asked to review the potential list. This list of drugsand drug categories reflects the collective thinking of all who provided input.

H

© ISMP 2008. Permission is granted to reproduce material for internal newsletters or communications with proper attribution. Otherreproduction is prohibited without written permission. Unless noted, reports were received through the USP-ISMP Medication ErrorsReporting Program (MERP). Report actual and potential medication errors to the MERP via the web at www.ismp.org or by calling1-800-FAIL-SAF(E). ISMP guarantees confidentiality of information received and respects reporters’ wishes as to the level of detailincluded in publications.

© IS

MP

2008

***Although colchicine injection should no longer be used, it will remain on the list untilshipments of unapproved colchicine injection cease in August 2008. For details,please visit: www.fda.gov/bbs/topics/NEWS/2008/NEW01791.html.

Reprinted with permission from the Institute for Safe Medication Practices, Horsham, Pennsylvania.

Appendix

PRESS COPY, D

O NOT D

ISTRIB

UTE

X UNITED STATES 5200 Butler Pike, Plymouth Meeting, PA 19462-1298, USA Telephone +1 (610) 825-6000 XFax +1 (610) 834-1275

X EUROPE Weltech Centre, Ridgeway, Welwyn Garden City, Herts AL7 2AA, UK Telephone +44 (1707) 871 511 XFax +44 (1707) 393 138

X ASIA PACIFIC 11-3-10, Jalan 3/109F, Danau Business Centre, Taman Danau Desa, 58100 Kuala Lumpur, Malaysia Telephone +60 3 7988 1919 XFax +60 3 7988 1170

X MIDDLE EAST 708/16 Fairmont Hotel, Sheikh Zayed Road, PO Box 24459 Dubai, United Arab Emirates Telephone +971 4 3042460 XFax +971 4 3042461

PRESS COPY, D

O NOT D

ISTRIB

UTE

distribute not do copy, press - ecri

Documents