Back To Lessons

Take Exam

Preventing Medication Errors

Pharmacy
Program/ACPE #: 380-000-05-026-H04
This program furnishes 2.0 hours of credit (0.2 CEU).

Publication Date: December 1, 2005 - Expiration Date: December 1, 2008


Author:

Kathleen A. Mahackian, PharmD
UC Davis Medical Center
Sacramento, California

The CE Solution, Inc. is approved by the Accreditation Council for Pharmacy Education as a provider of continuing pharmacy education.
This program reflects the opinion(s) of the author(s) and may not reflect those of the sponsor or publisher. While all reasonable attempts have been made to assure the accuracy of the information contained in this program based on current scientific knowledge at the time of publication, the reader is advised to evaluate their individual patient’s condition, compare information discussed or suggested with recommendations from other authorities, and refer to the official prescribing literature for the latest information on new or highly toxic drugs prior to administration or dispensing.

Statements made in this program have not been evaluated by the Food and Drug Administration. Discussion of published or investigational uses of drugs outside of approved labeling is offered for educational purposes only, and the sponsor and publisher of this program do not endorse such off-label use. Nutritional products discussed are not intended for the prevention, diagnosis, treatment, or cure of any disease.


  GOAL

To identify problems associated with the medication use process and provide strategies to eliminate or significantly reduce medication errors

LEARNING OBJECTIVES

Upon completion, the pharmacist should be able to:

    1. Identify two strategies that can prevent errors resulting from look-alike/sound-alike medication names;
    2. Describe the difference between root cause analysis and failure mode and effects analysis;
    3. List three resources for obtaining best practice information;
    4. Recognize how the formulary process can reduce medication errors;
    5. Identify five “high-alert” medications and select an error prevention strategy for each.

MEDICATION ERRORS TAKE CENTER STAGE

The Institute of Medicine (IOM) report estimated that between 44,000-98,000 patients die each year as a result of medical errors.1 Focusing on the medication error portion of the puzzle, it is estimated that approximately $2 billion is spent on treating hospitalized patients for preventable adverse drug events. One review of adverse drug reaction literature estimates that over 50% of drug-related admissions are preventable and therefore considered medication errors.2 After the release of the IOM report in 1999, the stage was set for reform.

Many individuals point to the release of the IOM report, "To Err is Human: Building a Safer Health System," as the catalyst for medication error reform. In reality, medication errors have been studied for over 30 years. Many high profile cases, such as the fatal error at Dana Farber Cancer Institute in 1995, served to focus media attention on the problem; and many groups, such as The National Coordinating Council for Medication Error Reporting and Prevention (NCC MERP), Institute for Safe Medication Practices (ISMP), and the National Patient Safety Foundation (NPSF) came into being years before the release of the IOM report. While much had been accomplished in pursuit of enhanced patient safety, there were really no incentives or requirements for hospitals, clinics, or pharmacies to implement changes to improve care. The IOM report changed all that, and there has been an increase in state and federal legislation aimed at reducing medical errors. Some of the currently proposed legislation address confidentiality, voluntary and mandatory reporting of errors, and legal protection.3,4 There has also been more of an emphasis on patient safety by the FDA, accrediting organizations such as the JCAHO, business communities, and professional societies.

ESTABLISHING A CULTURE OF SAFETY

Patient safety must be a priority throughout an organization, be it hospital, clinic, or community pharmacy. In order to prevent errors, an institution must have an error reporting process in place that is blame free. To encourage reporting, practitioners must feel that they can report errors in a non-punitive environment. A focus on "who" rather than "what" went wrong limits the institution's ability to identify and correct systems' problems and discourages reporting. Strategies to promote a blame-free environment include rewarding those who report errors, implementing a reporting mechanism that focuses on systems' problems, and prohibiting the use of error reports in performance evaluations.

It is also important to recognize that many healthcare professionals are involved in the medication-use system. If dispensing were the only step in the medication-use process, the pharmacist alone could solve all of the medication error problems. Medication errors involve all aspects of the medication-use process: prescribing, documenting (transcribing), dispensing, administering, and monitoring. These processes are intimately related, and changes in one often affect another.

Due to the complexity of the medication-use process, improvements require collaborative efforts from all healthcare professionals. Development of a multidisciplinary committee to address medication errors in the hospital setting should be an integral part of a medication error management program. This committee should address all aspects of care in the health system, both inpatient and outpatient.

It is equally important for pharmacy practice sites other than hospitals to follow the model of encouraging error reporting and establishing a blame-free environment. While an interdisciplinary committee might not be an option in these settings, it is essential to involve all employees at the practice site in identifying and eliminating systems' weaknesses that will reduce medication errors and improve the practice. Including pharmacists and technicians in medication error analysis heightens their awareness to common medication errors and encourages problem solving. Implementing systems changes, then, is easier as there is "buy in" from those involved in the process.

To successfully establish a culture of safety, a hospital or community pharmacy should:

  • Create a blame free environment;
  • Reward and encourage error reporting;
  • Focus on systems' problems;
  • Use medication error data to drive changes and identify training opportunities;
  • When possible, include a multidisciplinary approach to error prevention.

Many institutions perform safety surveys to determine individual attitudes and perceptions regarding their error reporting process.5 These surveys, sometimes obtained from other industries, assist the institution in identifying problems with how their medication safety process is perceived by the staff. These surveys solicit input from staff members about ease of reporting, willingness to report (or fear of punitive action), and their understanding of the overall medication safety process. Analysis of survey results then addressing concerns expressed by the staff can lead to significant improvements. For smaller practice sites, such as clinics or community pharmacies, encouraging open discussions at staff meetings rather than formal safety surveys can be used to gauge staff attitudes and perceptions.

PRESCRIBING ISSUES

Whether hospital, independent, or chain store pharmacy, dispensing usually starts with a written or verbal order. There are multiple potential problems associated with the prescribing process that produces this order6,7:

  • Illegible handwriting;
  • Use of dangerous abbreviations;
  • Leading and trailing zeroes;
  • Look-alike/sound-alike drug names;
  • Incomplete prescriptions;
  • Unavailable patient information.

Hand-written Prescription Orders
Illegible handwriting can lead to misinterpretation of the order. Couple poor handwriting with either dangerous abbreviations or look-alike drug names, for example, and the likelihood of a medication error increases.

Legibility has been a problem with prescription writing for decades with various attempts at eliminating this problem. Many medical schools offer lectures on prescription writing in an attempt to avoid the development of "bad habits." These classes heighten awareness as to the importance of legibility, as well as teach other safe practices for prescription writing to future clinicians.

Another strategy for preventing errors due to legibility issues is to develop preprinted order forms. This approach can minimize legibility problems and address some of the other prescribing problems mentioned in this section (dangerous abbreviations, incomplete prescriptions). Computer Physician/Practitioner Order Entry (CPOE), which will be discussed later, is another strategy that has been shown to eliminate legibility errors.

The Use of Abbreviations
A patient safety goal of the JCAHO (Joint Committee on Accreditation of Healthcare Organizations) adopted in 2003 is to improve communication within healthcare organizations.8 A key objective is the standardization of abbreviations, acronyms, and symbols used throughout organizations. Part of the process includes identifying abbreviations that should never be used in practice, since abbreviations can have multiple meanings or be similar in appearance. Misinterpretation of abbreviations sometimes leads to significant patient harm. Both the ISMP as well as the NCC MERP have publicized lists of abbreviations that are frequently misunderstood and associated with errors.6 Examples of some of the more commonly used "dangerous abbreviations" are:

  • The letter "u" for units;
  • g for mcg;
  • QOD (every other day), QID (four times a day), and QD (daily);
  • D/C for discontinue or discharge.

For example:

  • The letter u has been mistaken as a zero or a four. An insulin order for 3u can result in the administration of 30 or 34 units, possibly causing hypoglycemia.
  • It's easy to understand that QOD, QID, and QD can be mistaken for each other when written. An order for amlodipine 10 mg po QD can be misinterpreted as QID, especially if written as Q.D. The period between the q and d can be misinterpreted as the i, and the medication mistakenly administered four times daily instead of once daily.
  • Using the g abbreviation can be misinterpreted as mg.
  • D/C can be interpreted as "discharge" or "discontinue." Improperly interpreting this abbreviation can lead to prematurely discontinuing critical medications.

Use of dangerous abbreviations is a widespread and difficult problem to solve. Heightening clinician's awareness of the problem by publicizing a list of dangerous abbreviations is the first step in addressing this problem. Auditing the use of abbreviations and giving feedback to prescribers is another option in eliminating the use of dangerous abbreviations. As mentioned before, preprinted orders and CPOE are both effective strategies to minimize or eliminate the use of abbreviations.

Errors Involving Zeroes
Two other error-prone practices associated with prescription writing are the use of trailing zeros and the absence of a leading zero. Including a trailing zero is particularly dangerous in situations where multiple strengths of the medication are available, such as warfarin. If a physician writes a prescription as "1.0mg", there is a risk that the pharmacist will misinterpret this order as "10mg" resulting in an overdose and possible bleeding complications. This error is particularly problematic for outpatient prescriptions, as the error could go undetected for several days or weeks increasing the likelihood of adverse outcomes. In the hospitalized patient, intensive monitoring of the patient's anticoagulant response is possible increasing the likelihood that the error will be discovered and corrected.

The absence of a leading zero results in similar 10-fold dosing errors. In a study by Lesar9 on tenfold dosing errors related to prescribing, numerous 10-fold errors could be tracked to the use of trailing zeros or absence of leading zeros. In one case, a patient misinterpreted the prescription label that included a trailing zero thus providing an incorrect dosing history to the physician. This illustrates how violating safety rules can lead to errors even when there is not a legibility issue. Pharmacy systems should default to using a leading zero and delete a trailing zero on labels thereby preventing patient confusion as to the dose of medication they are taking.

Look-alike Drug Names
Look-alike names (brand-brand, brand-generic, and generic-generic) are another source of medication errors (Table 1). Recently, JCAHO added a national patient safety goal (NPSG) addressing the look-alike name issue (NPSG 3C). This new goal will require institutions to identify certain look-alike drug names and implement strategies to prevent errors associated with these medication name mix-ups. The JCAHO website is a valuable resource for look-alike and sound-alike drug names and includes information on preventing mix-ups for many problematic medications.10 These look-alike names cause errors in all aspects of the medication-use process: prescribing, transcribing, and dispensing of medications. There are many issues that contribute to an error associated with a look-alike name.5,10-12

  • Poor handwriting-leads to misinterpretation of a written order.
  • Nonformulary medications-pharmacy and/or nursing staff may be unfamiliar with nonformulary medications and may mistake the order for a look-alike formulary medication.
  • Newly marketed medication-clinicians may not be familiar with a new medication and can mistake it for a look-alike medication that has been in use for a longer period of time.

Table 1.  Examples of Look-Alike and Sound-Alike Medication Names*

Adriamycin – Idamycin

Celebrex – Celexa – Cerebra

Hydralazine – Hydroxyzine

Phenobarbital – Pentobarbital

Aggrenox – Aggrastat

Dopamine – Dobutamine

Imdur – Imuran

Quinidine – Quinine

Akarpine – Atropine

DiaBeta – Zebeta

Inderal – Isordil

Relafen – Rezulin

Alkeran – Leukeran

Diprivan – Diflucan

Klonopin – Clonidine

Retrovir – Ritonavir

Brevibloc – Brevital

Doxorubicin - Daunorubicin

Lamisil – Lamictal

Rifabutin – Rifampin

Bupivacaine – Ropivacaine

Efudex – Eurax

Lodine – Codeine

Seroquel – Serentil – Serzone

Bupropion – Buspirone

Eldepryl – Enalapril

Lopid – Slo-bid

Sulfasalazine – Sulfadiazine

Cafergot – Carafate

Equagesic – EquiGesic

Lorazepam – Clonazepam

Tobrex – TobraDex

Calcitriol – Calciferol

Etomidate – Etidronate

Mitomycin – Mitoxantrone

Vanceril – Vancenase

Carboplatin – Cisplatin

Folic acid – Folinic acid

Nelfinavir – Nevirapine

Vancomycin – Vecuronium  

Cardene – Cardizem

Glipizide – Glyburide

Nicardipine – Nifedipine – Nimodipine

 Vesicare – Visicol

Cardura – Coumadin

Guaifenesin – Guanfacine

Oxycodone – OxyContin

 Viramune – Viracept

Ceftazidime – Ceftizoxime

Humalog – Humulin

Paxil – Plavix

 Zosyn – Zofran

*For a more complete list of look-alike/sound-alike drug names visit: www.usp.org

Prior to the increased attention to medication safety, little consideration was given to patient safety when developing brand and generic names for new drug products. Marketing was the driving force when developing drug names. The IOM Report, as well as newer FDA standards requiring safety testing of trademarks, has become a motivating factor for many pharmaceutical companies to improve the selection of new drug names with an emphasis on safety. The Med-ERRS, Inc, a subsidiary of the ISMP established in 1997, will evaluate proposed medication names in order to identify and prevent potential sound-alike and look-alike confusion with existing products. When pharmaceutical companies contract with Med-ERRS for a fee, feedback is obtained from physicians, pharmacists, and nurses about possible similarity between the proposed name and existing medication names.13 This group also evaluates the design of drug packaging and labeling to maximize safety.
In addition to the Med-ERRS group, the Office of Postmarketing Drug Risk Assessment (OPDRA) was formed by the FDA in October 1999. In 2002, the OPDRA was renamed the Office of Drug Safety (ODS). The ODS consists of three divisions: the division of Drug Risk Evaluation (DDRE); the division of Medication Errors and Technical Support (DMETS); and the division of Surveillance, Research and Communication Support (SRCS). DMETS provides pre-marketing review of all proprietary names, labels, and labeling. Part of this process includes an expert panel review of the proposed name. Safety evaluators conduct literature and computer database searches and send approximately 100 health professionals simulated written prescriptions to assess the proposed name for look-alike problems. Names are compared to currently marketed drug names, other medicinal products, and to commonly used medical abbreviations, procedures, and/or lab tests. The drugs are evaluated for overlapping strengths and similar regimens that could increase the likelihood of an error with a look-alike medication. The drugs are also evaluated for similar indications, use in the same patient population, and/or storage in the same area as a sound-alike or look-alike medication.

While there is a mechanism now in place to address the process of assigning new drug names, the risk of medication errors from look-alike, sound-alike drugs still exists. One strategy aimed at minimizing prescribing errors associated with look-alike drug names is for the physician to include the medication's indication on the prescription. This gives the pharmacist additional information to validate his/her interpretation of the prescription. For example, an order for hydroxyzine 50 mg PO q6h for itching is not likely to be misinterpreted as hydralazine. As discussed earlier, preprinted orders or physician order entry systems will prevent those look-alike medication errors where handwriting plays a role.

Sound-alike Drug Names
Verbal orders are a source of errors with sound-alike drug names. Oftentimes the actual transmittal of a verbal order is problem prone.14 Speech patterns (speaking too fast, too softly, with an accent, etc.) affect the ability of a pharmacist to accurately receive and transcribe the order. Further, the environment of the healthcare provider receiving the order can be filled with distracting noises (faxes, equipment alarms, co-worker conversations, etc.). Strengths and directions for use can also be misinterpreted when taking verbal orders (eg, 15 confused for 50). The entire process is fraught with the potential for miscommunication and errors.

Ambulatory clinics and physicians' offices will oftentimes "call in" prescriptions. This process usually starts with a written order (which is subject to all of the problems associated with written prescriptions) that is then phoned to a pharmacy. The nurse calling in the prescription may be unfamiliar with the drug name, doses, and indications. This can lead to mispronunciation of the drug name or misreading of dosing instructions to the pharmacy. The more steps and intermediaries (eg, nurse or clerk) introduced into the process, the more likely an error will occur. This is also true when a patient requests a transfer of a prescription from one pharmacy to another. The procedure of transmitting the prescription from pharmacy 1 to pharmacy 2 is yet another opportunity for introducing error into the process.

Safe practices would dictate:

  • Use verbal orders only for emergency situations when a written order is not possible.
  • Always read back a verbal order so that the prescriber can validate the order.
  • Obtain the indication for the prescribed drug; as with written prescriptions, this piece of information can serve to validate the drug choice.
  • Prohibit the use of verbal orders for certain high-risk medications, such as chemotherapy.
  • Establish policies on who can receive verbal orders.
  • Whenever possible, the prescriber should call in prescriptions to the pharmacy.
  • The pharmacist should ask to speak to the prescriber directly if there are any questions regarding a prescription that is being called in by office personnel.

Incomplete Information
The final problems mentioned that might be encountered during the prescribing process, incomplete prescriptions and unavailable patient information, are obvious potential sources of error if the dispenser merely guesses at what the prescriber intended without verifying the actual intent and obtaining complete information before processing.


DISPENSING ISSUES

Look-alike names also plague the dispensing aspect of the medication-use system. Pharmacy generated labels can use small fonts that may be difficult to read. This can increase the likelihood of an error due to look-alike names. In March 2001, the FDA Office of Generic Drugs asked manufacturers of 16 pairs of drugs with look-alike names to voluntarily change the appearance of the name on the label using "tall man" letters to visually distinguish the names (Table 2).15


Table 2.  OGDs “Tall Man” Name Differentiation Project

Established Appearance

Proposed Appearance

Acetohexamide
Acetazolamide

AcetaHEXAMIDE
AcetaZOLAMIDE

Bupropion
Buspirone

BuPROPion
BusPIRone

Chlorpromazine
Chlorpropamide

ChlorproMAZINE
ChlorproPAMIDE

Clomiphene
Clomipramine

ClomiPHENE
ClomiPRAMINE

Cyclosporine
Cycloserine

CycloSPORINE
CycloSERINE

Daunorubicin
Doxorubicin

DAUNOrubicin
DOXOrubicin

Dimenhydrinate
Diphendydramine

DimenhyDRINATE
DiphenhydrAMINE

Dobutamine
Dopamine

DOBUTamine
DOPamine

Glipizide
Glyburide

GlipiZIDE
GlyBURIDE

Hydralazine
Hydroxyzine

HydrALAZINE
HydrOXYzine

Medroxyprogesterone

MedroxyPROGESTERone

Methylprednisolone
Methyltestosterone

MethylPREDNISolone
MethylTESTOSTERone

Nicardipine
Nifedipine

NiCARdipine
NIFEdipine

Prednisone
Prednisolone

PredniSONE
PrednisoLONE

Sulfadiazine
Sulfisoxazole

SulfADIAZINE
SulfiSOXAZOLE

Tolazamide
Tolbutamide

TOLAZamide
TOLBUTamide

Vinblastine
Vincristine

VinBLAStine
VinCRIStine  


The changes proposed by the FDA could be duplicated in pharmacy medication dispensing processes to minimize errors associated with look-alike medication names. Some of strategies to prevent dispensing errors due to look-alike names include:

  • Incorporate alerts in the pharmacy computer system;
  • Incorporate the FDA proposed changes (using tall man letters) onto pharmacy-generated labels or into pharmacy computer systems;
  • Flag shelves where the look-alike products are stored;
  • Do not store problem medications alphabetically;
  • Use both brand and generic names on medication labels and provide both names to the patient;
  • Use a formulary system to limit the availability of problem-prone, look-alike/sound-alike medications. If sound-alike/look-alike medications must be added, risk assessment and error prevention strategies should be developed as part of the formulary addition process.

It is important to note that look-alike errors affect all aspects of the medication-use system: prescribing, transcribing, dispensing, and administering. One goal in the prevention of errors due to look-alike medication names is to increase the "visibility" of the error, thereby increasing the likelihood that the error will be discovered. It is essential to ensure that at least a double check of the order occurs. The nurse and pharmacist need to independently review the order, which increases the likelihood that a written order subject to misinterpretation will be discovered. In the hospital, requiring a pharmacist's review of orders prior to nurse administration (ie, removing the medication from floor stock, prohibiting the nurse from obtaining the medication from an automatic dispensing machine prior to pharmacist review) is an essential part of making sure the double check takes place. Alerts that flag look-alike names in automated dispensing machines can serve to heighten the nurse's awareness to the potential for misinterpretation.

Packaging can also contribute to dispensing errors. Vials of similar size and shape, or products with similar appearing labels, can be confused resulting in medication errors and adverse drug reactions. The ODS also has responsibility for reviewing packaging issues as well as drug names. The packaging review attempts to determine if multiple strengths are distinguishable, if there is a possibility of confusion with another product, and if an error with a product could have a serious outcome. While some error-reduction strategies such as labeling shelves or posting warnings about look-alike packaging can have some impact, the technology that will significantly reduce packaging related errors is bar coding (which will be covered later in this article).

Other factors have been found to affect the accuracy of dispensing. Process errors have been studied by Grasha and psychological patterns identified that could be applied in developing error reduction strategies.16 Here are some of his findings and strategies.

  • More process errors occurred when going from high to low workload, or while working under low workload conditions. Findings suggest that there can be a lack of focus, boredom, or interruption of work rhythm. Regulating workload to prevent dramatic differences in workload is one possible strategy for error reduction.
  • Adequate lighting is important in reducing errors. Findings indicated that fewer process errors were identified if lighting was perceived as adequate.
  • Eye-level script holders resulted in a 30% reduction in reported process errors.

HIGH-ALERT MEDICATIONS

Medications are deemed high alert not because they are more prone to errors than other medications but because of the serious harm that can result from administering the drug in error. As early as 1993, certain medications were tagged as "today's poisons." Some of those same medications still remain as high-alert medications (Table 3).6 JCAHO has made one high-alert medication a priority by including it in one of its patient safety goals.6 In order to improve the safety of high-alert medications, the goal states that concentrated electrolytes should be removed from patient care units and that institutions should standardize and limit the number of drug concentrations available.


Table 3.  High-Alert Medications

Adrenergic Agonists

Digoxin

Lidocaine

Theophylline

Heparin, thrombolytics and warfarin

*Intravenous magnesium

Benzodiazepines

*Hypertonic and hypotonic saline

Narcotic and opiates

*Intravenous calcium

Insulin

Neuromuscular blocking agents

Chemotherapeutic agents

Oral hypoglycemic agents

Cardioplegia

Chloral hydrate

*Potassium (Chloride or phosphate

Colchicine

*Included in JCAHO patient safety goal

Each high-alert medication has its own set of breakdown points and possible solutions. Some of the measures implemented to prevent problems with high-alert medications can include:

  • Limit availability of medication (remove from floor stock);
  • Utilize double checks; can occur at the dispensing or administering point of the medication-use process;
  • Utilize caution labels;
  • Review storage practices for those items that must remain available for use;
  • Standardize ordering procedures; use pre-printed orders when possible;
  • Use premixed solutions when possible;
  • Require double checks on calculations;
  • Use only pumps that are protected from free flow;
  • Implement maximum dosing alerts in pharmacy computer systems;
  • Develop standards for monitoring of some high-alert medications;
  • Prohibit bolus doses from infusion bags.

One of the most effective methods of preventing errors from the high-alert medications is to perform a Failure Mode and Effects Analysis (FMEA) for ordering, storing, dispensing, or administering these drugs. The FMEA process is a systematic approach to prevent process problems before they occur. An institution can look at a new drug, high-alert drug, and dispensing process and proactively identify system flaws before an error or bad outcome occurs. The VA National Center for Patient Safety has developed a system of utilizing the FMEA process that can be applied to health care.17 Steps in the process include:

  1. Define the topic;
  2. Form the team;
  3. Describe the process (flow diagram);
  4. Identify where and how the process can fail;
  5. Conduct a hazard analysis that includes developing a severity-rating tool (Table 4) and a probability-rating tool (Table 5) to identify the high-risk aspects of a process. Table 6 identifies how the severity and probability rating can then be used to determine an overall hazard score for a process.
  6. Develop an action plan for processes identified by the team as posing a significant hazard;
  7. Determine outcome measures.

Table 4. VA NCPS’ Severity Rating Tool

Severity Rating

Rank

Catastrophic

4

Major

3

Moderate

2

Minor

1


Table 5.  VA NCPS’ Probability Rating Tool

Probability Rating

Rank

Frequent

4

Occasional

3

Uncommon

2

Remote

1


Table 6. Overall Hazard Score Based on Probability and Severity

 

Severity of Effect

Probability

Catastrophic

Major

Moderate

Minor

Frequent

16

12

8

4

Occasional

12

9

6

3

Uncommon

8

6

4

2