Chapter 29. Preventing and Managing Medication Errors: The Pharmacist's Role

Matthew Grissinger, R.Ph., FISMP, FASCP is the Director of Error Reporting Programs at the Institute for Safe Medication Practices (ISMP). Prior to joining ISMP, he served as a home care and long-term care pharmacy surveyor for the Joint Commission (TJC). Mr. Grissinger has published numerous articles in the pharmacy literature, including regular columns in P&T and the Patient Safety Advisory (PSA). He serves on the National Quality Form (NQF) Common Formats Expert Panel, the Faculty Advisory Board for the Pharmacy Learning Network (PLN), and the Publications Advisory Board for Davis's Drug Guide for Nurses. He is an adjunct assistant professor for Temple University School of Pharmacy and clinical assistant professor for the University of the Sciences in Philadelphia. Mr. Grissinger received a B.S. in Pharmacy from the Philadelphia College of Pharmacy and Science and is a fellow of the ISMP as well as the American Society of Consultant Pharmacists.

Michael Cohen, R.Ph., M.S., Sc.D., is president of the ISMP. He is editor of the textbook Medication Errors and serves as coeditor of the ISMP Medication Safety Alert!, publications that reach over 2 million health professionals and consumers in the United States and in over 30 countries. Dr. Cohen is a member of the Sentinel Event Advisory Group for TJC and recently served as a member on the Committee on Identifying and Preventing Medication Errors for the Institute of Medicine (IOM). In 2005, he was recognized as a MacArthur Fellow by the John D. and Catherine T. MacArthur Foundation. In 2008, Dr. Cohen was honored by The National Quality Forum (NQF) and The Joint Commission (TJC) with the John M. Eisenberg Patient Safety and Quality Award in recognition of his lifelong professional commitment to promoting safe medication use and a safe medication delivery system.

After completing this chapter, readers should be able to

1. Discuss the role of the pharmacist in preventing medication errors.

2. Define latent and active failures and the role each plays when a medication error occurs.

3. Define the types of medication errors that can occur during the ordering and dispensing process.

4. List some commonly used drugs that can result in medication error-related deaths.

5. Describe what changes are needed at the risk management level to better address medication safety issues, including the use of failure mode and effects analysis to reduce the potential for errors.

6. Identify specific problems in our approach to error prevention and what needs to be changed to ensure patient safety.

7. Describe a variety of methods that can be used to identify risk and provide meaningful data on the relative safety of their facility's medication use process.

8. Select high-leverage error reduction strategies based on sound safety principles.

This scenario is based on a true story that demonstrates the multiple breakdowns that can occur during the medication use process.

An infant was born to a mother with a prior history of syphilis. Despite having incomplete patient information about the mother's past treatment for syphilis and the current health status of both the mother and the child, a decision was made to treat the infant for congenital syphilis. After phone consultation with infectious disease specialists and the health department, an order was written for one dose of “benzathine penicillin G 150,000 units IM.”

The physicians, nurses, and pharmacists, unfamiliar with the treatment of congenital syphilis, also had limited knowledge about this medication. The pharmacist consulted Drug Facts and Comparisons to determine the usual dose of penicillin G benzathine for an infant. However, she misread the dose as 500,000 units/kg, a typical adult dose, instead of 50,000 units/kg. Consequently, the pharmacist also incorrectly read and prepared the order as 1,500,000 units, a 10-fold overdose. Owing to the lack of a consistent pharmacy procedure for independent double checking, the error was not detected. The pharmacy dispensed the 10-fold overdose in a plastic bag containing two full syringes of Permapen® (the brand name of penicillin G benzathine) 1.2 million units/2 mL each, with green stickers on the plungers to “note dosage strength.” A pharmacy label on the bag indicated that 2.5 mL of medication was to be administered intramuscularly to equal a dose of 1,500,000 units. After glancing at the medication sent from the pharmacy, the infant's primary care nurse expressed concern to her colleagues about the number of injections required to give the infant the medication (since there a maximum of 0.5 mL per intramuscular injection allowed in infants, the dose would require five injections).

Anxious to prevent unnecessary pain to the infant, the two nurses decided to investigate the possibility of administering the medication intravenously instead of intramuscularly. The monograph on penicillin G did not specifically mention penicillin G benzathine; instead, it noted the treatment for congenital syphilis with aqueous crystalline penicillin G slow intravenous push or penicillin G procaine intramuscularly. Nowhere in the two-page monograph was penicillin G benzathine mentioned, and no specific warnings regarding “IM use only” for penicillin G procaine and penicillin G benzathine were present. Unfamiliar with the various forms of penicillin G, a nurse practitioner believed that “benzathine” was a brand name for penicillin G and concluded that the drug could be administered safely intravenously. While preparing for drug administration, neither nurse noticed the 10-fold overdose, and neither noticed that the syringe was labeled by the manufacturer, “IM use only.” The nurses began to administer the first syringe of Permapen as a slow intravenous push. After about 1.8 mL was administered, the infant became unresponsive, and resuscitation efforts were unsuccessful (ISMP, 1998).

The three nurses involved in this case were indicted for criminally negligent homicide in the death of the baby. Over 50 different system failures allowed this error to occur, go undetected, and ultimately, reach a healthy newborn child, causing his death. Had even just one of these failures not occurred, either the accident would not have happened, or the error would have been detected and corrected before reaching the infant.

1. At what stages of the medication use process do medication errors occur?

2. What types of contributing factors lead to medication errors during the ordering process?

3. What procedures should be followed when taking a verbal order?

4. What steps should be followed during the prescription-filling process to prevent medication errors?

5. What three important factors play a role in any patient interface?

6. Which types of patients are more at risk for noncompliance?

Patient safety has become a major concern since the November 1999 release of the Institute of Medicine's (IOM) report To Err Is Human. Health care practitioners may have been surprised to learn from this report that errors involving prescription medications kill up to 7,000 Americans a year (Kohn et al., 1999). The IOM released a report in 2006 entitled Preventing Medication Errors and indicated that medication errors are among the most common medical errors, harming at least 1.5 million people every year. The reports concluded that 400,000 preventable drug-related injuries occur each year in hospitals. Another 800,000 occur in long-term care settings, and roughly 530,000 occur just among Medicare recipients in outpatient clinics. Assuming conservatively, an annual incidence of 400,000 in-hospital preventable adverse drug events (ADEs) yields an annual cost of $3.5 billion in 2006. The report noted that these are likely underestimates because the data excluded errors of omission such as the failure to prescribe medications for which there is an evidence base for the ability to reduce morbidity and morbidity (Aspden et al., 2006).

With this increased attention to medication errors, concern has intensified in both the public and health care sectors. Research demonstrates that injuries resulting from medication errors generally are not the fault of individual health care professionals, but rather represent failures in a complex health care system. Medication error prevention starts with recognizing that errors are multifactorial and are faults of the system as a whole rather than results of the acts or omissions of the people in the system. Even when an error can be traced directly to a specific individual (e.g., the pharmacist dispensing or nurse administering the wrong medication), further investigation often determines that a number of factors, such as poor order communication between the physician and pharmacist, dangerous storage practices in pharmacies, and look-alike labeling may have played a role in the error. Protecting patients from inappropriate administration of medications has become an important focus for pharmacists and technicians, including those in ambulatory, acute care, long-term care, home care, and managed-care settings.

Pharmacists and technicians play a pivotal role in assuring medication safety. After summarizing several studies performed in hospitals and long-term care facilities, Allan and Barker (1990) estimated that medication errors occur at a rate of about 1 per patient per day. In a more recent study performed in ambulatory pharmacies, they found an overall dispensing accuracy rate for prescription medications of 98.3 percent (Allan et al., 2003). Although most of these errors probably have minimal clinical relevance and do not affect patients adversely, many experts believe that medication error rates may be higher in the ambulatory care setting because errors may not always be evident to the health professionals who work there. For example, medication errors can occur when a patient purchases nonprescription medications without speaking with the pharmacist about any potential interactions with his or her prescription medications, or if patients fail to verify the appropriate dose of the over-the-counter (OTC) medication.

This chapter focuses on system enhancements and the checks and balances needed to proactively prevent medication errors as pharmacists and technicians prepare, dispense, and monitor the effects of medications in all practice settings. In addition, focus is placed on the importance of determining latent failures that contribute to medication errors by developing effective reporting programs to discover how latent failures occur and how they can be prevented.

Many organizations take an ineffective approach to preventing medication errors. Investigations tend to focus on the front end or active end of the error (e.g., the front-line practitioner, such as the pharmacist dispensing the medication). When an error occurs, human nature needs to assign blame to someone or something. In addition, health care practitioners work in an environment where they strive for perfection. Individuals involved in the commission of an error may be considered inattentive, incompetent, lazy, and uncaring. They are often subject to punitive action such as disciplinary action, public or private reprimands, remedial education, suspensions, or termination. As a result, the practitioner develops feelings of inadequacy, denial, and embarrassment.

Effective approaches, on the other hand, consider factors that contribute to medication errors that occur at the organizational level, known as the latent end or blunt end of an error. Latent failures are weaknesses in organizational structure, such as faulty information management or ineffective personnel training, that may have resulted from decisions made by upper management (Reason, 1990). Latent failures can also stem from incomplete patient information, such as missing allergy or diagnosis information, unclear communication of a drug order, lack of an independent double check before dispensing, lack of computer warnings, ambiguous drug references, drug storage issues, and look-alike/sound-alike medications. These latent failures are properties of the medication use system. To prevent medications errors, we must change and improve the system and not rely on changing people and their behaviors.

ISMP has identified 10 key elements that have the greatest influence on medication-use system. System-based causes of medication errors can be directly traced to weaknesses or failures in these key elements:

• Patient information
• Drug information
• Communication of drug information
• Drug packaging, labeling, and nomenclature
• Drug device acquisition and use
• Drug storage, stock, and distribution
• Environmental factors
• Staff competency and education
• Patient education
• Quality processes; and risk management (Cohen, 2007).

By themselves, latent failures often are subtle and may not cause problems directly. The potential consequences are often hidden, becoming apparent only when they occur in a certain sequence and combine with the active failures of an individual.

Medication use is a complex process that consists of subprocesses such as the ordering, preparing, dispensing, administration, and the provision of patient education. Because failed communication of medication orders is at the heart of many errors, pharmacists must be aware of the types of breakdowns that can occur during the ordering process that can lead to medication errors.

Physicians or their designees—pharmacists, nurse practitioners, physician assistants, and nurses—initiate the drug dispensing and administration process through a medication order or prescription. While computerized prescriber order entry (CPOE) and electronic prescribing (e-prescribing) systems, each with clinical decision-support tools, are being implemented in more settings, as of 2011 most pharmacists still dispense from handwritten medication orders. Illegible, ambiguous, or incomplete handwritten prescriptions or medication orders contribute to many errors made by nurses, pharmacists, pharmacy technicians, and other health care workers.

Illegible Handwriting

To minimize the chance of misinterpretation due to illegibly handwritten orders, encourage physicians with poor handwriting to print prescriptions and medication orders in block letters. In the institutional setting, have physicians review orders with the nursing staff before leaving the patient care area. More importantly, ask physicians to include the purpose of the medication as part of the prescription or medication order to help readers distinguish drug names when handwriting legibility is less than ideal. Preprinted orders, dictation, and direct order entry into the computer by physicians are other solutions for poor handwriting.

Because even skilled individuals can misread good handwriting, a system of independent double checks for order/prescription transcription should be in place in which several individuals interpret and transcribe an order. In many hospitals, each order is read by a unit secretary and reviewed by a nurse. At the same time, an exact copy of the order is sent to the pharmacy either directly, scanned, or by facsimile. In the pharmacy, pharmacists and technicians have a number of opportunities to check the order, including a double check against labels, printouts, and the drug containers. A technician often screens the order and sometimes enters it into the computer. After data entry, a label is printed, and a pharmacist should interpret the original order/prescription and verifies the technician's computer entry by comparing it with the label. Later, the order and label will be read again by technicians and pharmacists as doses are prepared and dispensed. In the outpatient setting, this system should include a final check when providing counseling to the patient. In no case should pharmacy technicians interpret orders on their own because this process does not offer enough checks to prevent an error. In addition, orders must not be filled only from computer-generated labels. Rather, the original order should accompany the label to serve as another check.

Look-Alike Drug Names

Medications with names that are spelled similarly can be easily misread for one another. Pharmacists and technicians must be alert to this problem and should never guess about a prescriber's intent.

Study the handwriting in Figs. 29-1 and 29-2. These are actual examples of handwritten orders in both the inpatient and outpatient settings, each of which led to medication errors. The problem was not uncertainty. On the contrary, each order was misread from the start. No consideration was ever given to the alternative drug because in each case the pharmacy staff members thought they were reading the order correctly.

When pharmacists and technicians interpret prescriptions and medication orders, newly marketed drugs are a particular problem. Staff members are not as familiar with names of the drugs, and they tend to misinterpret them as older drugs. It is important that up-to-date education on all new medications is provided to the pharmacy staff, including any potential for error that may exist with these new products. In the ambulatory care setting, physicians can write both the generic and trade names legibly on the prescription, and they can add the intended purpose of the medication to further alert pharmacy staff to the correct medication name. Many medications have look-alike names, but very few name pairs that are spelled similarly are used for similar purposes.

Sound-Alike Names

Drug orders communicated orally are often misheard, misunderstood, misinterpreted, or transcribed incorrectly. Celebrex® and Cerebyx® sound alike, as do Celexa® and Zyprexa®, alprazolam and lorazepam, Flomax® and Volmax®, and hundreds of other name pairs. Each of these drugs has been confused for the other, resulting in patients receiving incorrect medications. In many cases, serious injuries have occurred because of misinterpreted verbal orders. This is a good reason for health care facilities such as hospitals and long-term care facilities to establish policies that prohibit verbal requests for medication without pharmacy review of a hard copy of the order. Sound-alike drug names present many of the same problems as look alikes. Obviously, when uncertainties exist, the pharmacist must contact the prescriber for clarification.

To decrease the opportunity for misunderstanding, health care facilities and community pharmacies should discourage verbal orders. TJC, a national accrediting agency for health care organizations, requires in its standards that accredited organizations improve the effectiveness of communication among caregivers by implementing a process for taking verbal or telephone orders that mandates that a verbal order to be transcribed and then “read back” completely by the person receiving the order (TJC, 2011).

Greater use of computerized order entry, scanning technology and facsimile machines among hospital areas, medical offices, pharmacies, and nursing units will help improve communication of medication orders. When verbal communication is unavoidable, strict adherence to these procedures can minimize errors:

• In acute care settings, spoken orders should be limited to true emergencies or circumstances in which the prescriber is physically unable to write or electronically transmit orders (e.g., the prescriber is working in a sterile field).
• Prohibit spoken orders for selected high-alert medications (e.g., chemotherapy, IV insulin for neonates) because of their complexity and potential for serious errors.
• Verbal orders should be taken only by authorized personnel.
• If possible, a second person should listen while the prescription is being given.
• The order should be written down and then read back, repeating exactly what has been ordered, sometimes spelling the drug name for verification and the strength by using a digit-by-digit technique for the dose (1–5, not 15).
• In the acute care setting, record the verbal order directly onto an order sheet in the patient's chart whenever possible. Transcription from a scrap of paper to the chart introduces another opportunity for error.
• Obtain the prescriber's phone number in case it is necessary for follow-up questions.
• The prescribed agent must make sense for the patient's clinical situation. (Cohen, 2007a)

To prevent sound-alike and look-alike errors, physicians must be encouraged to include complete directions, strengths, route of administration, and indication (purpose) for use. With this information, pharmacists and other skilled health care professionals can judge whether the drug ordered makes sense for the patient in the context in which it is written. For example, knowing that the patient has a diagnosis of diabetes would be important in determining that Avandia® is intended by the medication order in Fig. 29-2. Diagnostic procedures along with orders could also provide important information. This is why it is important for pharmacists to verify all orders processed by technicians. When in doubt, technicians should check with the pharmacist, who can contact the prescriber for clarification if the intent is not completely clear.

Abbreviations to Avoid

Certain abbreviations are easily misinterpreted. Controlling dangerous abbreviations can reduce communication errors. Although many health care facilities have lists of abbreviations that are approved for use by professional staff, it would be far safer if each hospital also developed a list of abbreviations that should never be used. TJC has recommended that accredited organizations standardize abbreviations, acronyms, and symbols used throughout their facilities, including a list of abbreviations, acronyms, and symbols not to use.

The ISMP has developed a list that contains several easily misinterpreted abbreviations from actual medication errors, some of which resulted in patient harm (ISMP, 2010). These abbreviations should never be used in medication orders, on pharmacy labels, in newsletters, or in other communications that originate in a pharmacy or pharmacy computer systems. Examples of some of the most problematic abbreviations used to communicate orders include:

• The abbreviation U for units. Errors have occurred when the letter U was mistaken for the numerals 0, 4, 6, and 7, and even cc, resulting in disastrous drug overdoses of insulin, heparin, penicillin, and other medications whose doses sometimes are expressed in units. For example, an order written as “Humalog 6U” has been misinterpreted as “60 units.” (see Fig. 29-3).
• Q for “every,” as well as other abbreviations with this letter (QD, QID, and QOD; or qd, qid, and qod). QD for “daily” can result in fourfold overdosages if seen as QID (q.i.d.) for “four times daily,” or subtherapeutic doses if seen as QOD for “every other day.” Errors associated with QD and QOD have been reported to PA Patient Safety Authority (PSA) [Pennsylvania Patient Safety Authority (PSA, 2005)]. In one case, an order for Zithromax® (azithromycin) 500 mg written as QD was misinterpreted as QID. In another report, an order was written for digoxin 0.125 mg po QOD (every other day), but the medication was given QD (every day).
• D/C. It has been written to mean either “discontinue” or “discharge,” sometimes resulting in premature stoppage of a patient's medications. In Fig. 29-4, the “d/c” order was incorrectly interpreted as “discontinuation” of an antibiotic that the patient had never even received. In reality, the “d/c” is really “OK,” meaning that the drug was approved for use by the infectious disease physician. Drug names also should not be abbreviated.

Abbreviating drug names have led to frequent medication errors, including:

• Magnesium sulfate (MgSO4 or simply Mg) and morphine sulfate (MSO4 or simply MS). In one example, a prescriber used an abbreviation for magnesium sulfate and wrote “MgSo4 2g IV × 1 dose” for a 45-year-old female patient. However, the unit clerk and nurse misinterpreted the order as morphine sulfate (MSO4) “2 mg IV × 1 dose.” The patient received a 2 mg dose of morphine sulfate. Contributing to this error was the fact that the patient was having pain, so morphine seemed reasonable [Pennsylvania Patient Safety Authority (PSA, 2005)].
• MTX. This means “methotrexate” to some health professionals, but others understand it as “mitoxantrone.”
• AZT. This has been misunderstood as “azathioprine” (Imuran®) when “zidovudine” (Retrovir®) was intended. In one case, this misinterpretation led to a patient with AIDS receiving azathioprine, an immunosuppressant, instead of the intended antiretroviral agent. The patient's immune system worsened, and he developed an overwhelming infection.
• HCT and HCTZ. Orders for “HCTZ50 mg” (hydrochlorothiazide) can be mistaken for “HCT 250 mg” (hydrocortisone).

Ambiguous Orders

Errors can result when ambiguous orders are interpreted in a manner other than what the prescriber intended. Proper expression of doses is vital in a drug order. Pharmacists should be able to recognize improper expressions of doses and their potential for error. When the order is not clear, the pharmacist must contact the prescriber for clarification. Pharmacists and technicians should avoid using dangerous expressions of doses as they process orders, type labels, and communicate with others. The following examples include several improperly expressed orders that were reported to the ISMP:

• Zeros and decimal points. When listing drug doses on labels or in other communications, never follow a whole number with a decimal and a zero. For example, “Coumadin 1.0 mg” is a very dangerous way to express this dose. If the decimal point were not seen, the dose would be misinterpreted as “10 mg,” and a 10-fold overdose would result. The same could happen when “Dilaudid 1.0 mg” is written. The proper way to express these orders would be “Coumadin 1 mg” and “Dilaudid 1 mg.”
• Leading zeros. Always place a leading zero before a decimal point when the dose is smaller than 1. For example, “Vincristine .4 mg” was seen as “Vincristine 4 mg” due to a poor impression of the decimal, as is common on faxes or copies of orders. Avoid using decimal expressions when recognizable alternatives exist because whole numbers are easier to work with. For example, “Digoxin 0.125 mg” is acceptable, but “Digoxin 125 mcg” is a better alternative. So too would be to indicate a dose as “500 mg” instead of “0.5 g.”
• Tablet strengths. Orders specifying both strength and number of tablets are confusing when more than one tablet strength exists. For example, “Metoprolol 1/2 (one-half) tablet 25 mg once daily” appears clear enough. However, when you realize that this product is available in both 25- and 50-mg tablets, the ambiguity of this order becomes apparent. What is the intended dose, 25 or 12.5 mg? Orders are clearer if the dose is specified regardless of the strengths available (e.g., “Metoprolol 12.5 mg once daily”). For doses that require several tablets or capsules, the pharmacy label should note the exact number of dosage units needed. For example, the label on an 800-mg dose of Asacol® (mesalamine), which is available only in 400-mg tablets, should read “2 × 400-mg tablets = 800 mg.” For a 6.25-mg dose of captopril, which is available in 12.5-mg tablets, the label should read one-half (1/2) × 12.5-mg tablet = 6.25 mg.” Fractions should be displayed in fraction format and not as 1 slash 2 (1/2) (ISMP, 2011). The same type of notation should be used on computer-generated medication administration records (MARs) for nurses.
• Liquid dosage forms. Expressing the dose for liquid dosage forms in only milliliters or teaspoonfuls is dangerous. For example, acetaminophen elixir is available in many strengths, including 80, 120, and 160 mg per 5 mL. If the prescriber wrote “5 mL,” the intended number of milligrams would be unclear. However, an order of “80 mg” states the appropriate dose. The amount of drug by metric weight, as well as the volume, should always be included on the pharmacy label (e.g., “Acetaminophen elixir 80 mg/5 mL”). Further, the patient dose should also be included. For a 320-mg dose, the label should read, “320 mg = 20 mL.” The same holds true for unit dose and bulk labels.
• Injectable medications. The same rules apply for injectable drugs. Because solution concentrations can vary, list the metric weight of the dose, or the metric weight and volume of the dose, but never the volume alone. An example of this error occurred at a hospital where hepatitis B vaccines were being administered. A preprinted physician's order form was used to prescribe the vaccine, listing only the volume to be given. When the clinic switched to another brand of vaccine, containing a different concentration of vaccine, the same preprinted forms continued to be used. This resulted in the underdosing of hundreds of children until the error was discovered. This could have been avoided had the amount of vaccine been prescribed in micrograms rather than just the volume in milliliters.
• Variable amounts. A drug dose should never be ordered solely by number of tablets, capsules, ampules, or vials because the amounts contained in these dosage forms vary. Drug doses should be ordered with proper unit expression (e.g., “20 mEq potassium chloride”). A patient whose doctor orders “an amp” of potassium chloride might get 10, 20, 30, 40, 60, or 90 mEq. Under certain circumstances, the higher doses of this drug could be lethal.
• Spacing. Overdoses have been reported because a lowercase l (ell) was the final letter in a drug name and was misread as the number 1. In one case, an order for 300 mg Tegretol® (carbamazepine) twice daily appeared as “Tegretol 300 mg bid” and was misinterpreted as “1300 mg bid” (see Fig. 29-5). In another case, a nurse misread an order for 2 mg Amaryl® (glimepiride) as 12 mg because there was insufficient space between the last letter in the drug name and the numerical dose. In addition, when labels are printed, make sure that there is a space after the drug name, the dose, and the unit of measurement.
• Apothecary system. Use the metric system exclusively. The apothecary system (grains, drams, minims, and ounces) has been found to be very inaccurate. For example, symbols for dram have been misread as “3,” and minim has been misread as “55.” Orders for phenobarbital 0.5 gr (30 mg) have been mistaken as “0.5 gram (500 mg).” Use of the apothecary system is no longer officially recognized by the United States Pharmacopeia (USP).

An important safety enhancement for preventing dispensing errors is the development of a system of redundant checks from the time a prescription order first is written in the physician's office or on the nursing unit to receipt in the pharmacy and through dispensing and administration. The more “looks” an order receives (while efficient workflow is maintained), the better. Health professionals can review orders at several checkpoints and thereby maximize the chances of errors being discovered. Pharmacies with computerized drug distribution systems have an advantage because labels and reports can be printed so that order interpretation and order entry can be verified at various steps.

Steps in Prescription Filling

It is important that all health care professionals, including prescribers, nurses, and pharmacy personnel, work together when filling prescriptions so that the lines of communication are kept open if questions should arise. It is also important that pharmacy personnel understand the processes involved in the communication of orders in their particular practice setting so that breakdowns can be addressed and improvements can be made to improve both efficiency and safety.

Community Pharmacy

1. The prescriber sees the patient; performs an assessment; determines the appropriate medication, dose, route, and frequency; and writes the order or communicates the order verbally to nursing personnel or the pharmacy.

2. A direct copy of the order is transported by the patient, faxed/scanned to the pharmacy, or the prescriber's computer entry system (ePrescribing) directly reaches the pharmacy order entry system.

3. The pharmacist or pharmacy technician reads the order and enters it in the pharmacy computer system.

   1. If the technician finds a duplicate order, incorrect dose, drug allergy or other problem, it is documented and called to the attention of the pharmacist during the clinical screening.

4. A pharmacist reviews the prescriber's or pharmacy technician's computer entry, compares it with the original prescription (handwritten or electronic), and performs a clinical screening of the prescription with respect to the need for the drug, allergies or other contraindications, proper dose, and proper route of administration.

5. A label and/or medication profile is printed. It is important that a copy of the original prescription or medication order continues to accompany the label or medication profile while the order is filled. Orders should never be filled solely on the basis of what appears on the label or medication profile because the computer entry may have been in error.

6. To choose an item for dispensing, a technician reviews both the label and the medication order for possible discrepancies.

7. A pharmacist checks the technician's work, reviewing the label against the medication order and the dose that has been prepared. The drug is labeled and dispensed.

8. The pharmacist uses the patient counseling session to further assess that the correct medication is being dispensed and that the patient has a condition treatable with the product being provided.

   1. Pharmacists should not simply ask patients if they have any questions about their medication, but confirm their understanding of the medication and its proper use.

   2. For refills and medications patients have received in the past, patients should be encouraged to ask the pharmacist about any changes in the appearance of the product.

   3. If patients or caregivers are provided with administration device, such as an oral syringe or inhaler, pharmacists must ensure that the patient or caregiver understands how to use the device properly. Demonstrate for patients how to use the device, and follow with a return demonstration by the user.

9. Patients in the community setting should be educated about the common adverse effects of medications they are taking and understand what clinical signs to watch for and report to health professionals.

10. The final step in the process is the assurance of adherence to medication therapy. If the patient is taking too much medication or is not taking the drug as frequently as prescribed, the pharmacist should speak with the patient to determine the reasons for and address the variation. In addition, patients should be asked about common adverse effects and about signs of serious drug toxicities.

Hospital Pharmacy

1. The prescriber sees the patient; performs an assessment; determines the appropriate medication(s), dose(s), route(s), and frequency(ies); and writes orders or communicates the orders verbally to nursing personnel or the pharmacy.

2. A unit secretary reads and transcribes the order onto the MAR. A nurse checks the unit secretary's transcription for accuracy. This step is unnecessary in hospitals where computers generate an electronic medical record (eMAR), where prescribers can enter orders directly into a computerize physician-order entry system (CPOE) or pharmacy prints MARs for the nursing staff.

3. A direct copy of the order is transported, faxed/scanned to the pharmacy, or the prescriber's computer entry system (CPOE) reaches the pharmacy order entry system. The pharmacist or pharmacy technician reads the order and enters it in the pharmacy computer system. If the technician finds a duplicate order, incorrect dose, drug allergy or other problem, it is documented and called to the attention of the pharmacist during the clinical screening.

4. A pharmacist reviews the prescriber's or pharmacy technician's computer entry, compares it with the original prescription (handwritten or electronic), and performs a clinical screening of the prescription with respect to the need for the drug, allergies or other contraindications, proper dose, and proper route of administration.

5. The pharmacy label is printed. A copy of the medication order(s) continues to accompany the label while the order is filled. Orders should never be filled solely on the basis of what appears on the label because the computer entry may have been in error.

6. To choose an item for dispensing, a technician reviews both the label and the medication order for possible discrepancies and obtains the product from its storage area.

7. A pharmacist checks the technician's work, reviewing the label against the medication order and the dose that has been prepared. The drug is labeled and dispensed.

8. The medications are delivered to the patient's care area. Medications are stored in a variety of areas that include individual patient cassettes/bins or automated dispensing cabinets (ADCs).

9. The nurse obtains the drug and compares the medication and pharmacy label against the copy of the physician's order, as well as the MAR/eMAR.

10. The nurse administers the dose, giving the drug's name, explaining the drug's purpose and potential adverse effects, and answering questions and concerns raised by the patient.

Selecting Medications

The importance of reading the product label while selecting medications and filling prescriptions cannot be overemphasized. Too often the wrong drug, strength or concentration is dispensed. Such errors often stem from failure to read the label thoroughly. During drug preparation and dispensing, the label should be read three times:

1. When the product is selected

2. When the medication is prepared

3. When either the partially used medication is disposed of (or restored to stock) or product preparation is complete

Selecting the correct item from the shelf, drawer, or bin can be complicated by many factors. Similar labeling and packaging, as well as look-alike names, are common traps that lead to medication errors (see Figs. 29-6 and 29-7). Restocking errors are common and can lead to repeated medication errors before being detected.

ADCs (e.g., Pyxis, Omnicell) have become common on nursing units in many hospitals. The nurse must enter a security code and a password into the dispensing device, along with the name of the patient and the name of the medication, before the machine will allow access to remove the medication. This system allows better control of items kept on the nursing unit and serves as a check for the nurse who retrieves the medication. One important feature that should be included in these dispensing units is online communication with the pharmacy system that allows a pharmacist to review medication orders before nursing accesses the medication.

ADCs create several situations that can result in errors. While these machines are routinely restocked, the wrong drug still can be placed into the wrong bin during this process. Devices that have multiple medications in each drawer and/or that do not require pharmacist review of orders before access have drawbacks that are identical to the flaws in traditional floor-stock systems:

• The pharmacy can replenish the system with an incorrect medication.
• The nurse can retrieve either the wrong item or additional items to use for other patients.
• Lack of pharmacist double checking and screening orders before a nurse withdraws a medication from the cabinet allows prescribing errors, wrong dosages, incorrect routes of administration, and other clinical errors to occur.

The term confirmation bias is used to describe the phenomenon that when choosing an item, people see what they are looking for, and once they think they have found it, they stop looking any further. Often health professionals choose a medication container based on a mental picture of the item. Staff members may be looking for some characteristic of the drug label, the shape and size or color of the container, or the location of the item on a shelf, in a drawer, or in a storage bin instead of reading the name of the drug itself. Consequently, they may fail to realize that they have the wrong item in hand.

A number of approaches can be used to minimize the possibility of such errors in the pharmacy, care areas, and in automated dispensing machines. Physically separating drugs with look-alike labels and packaging reduces the potential for error. Some pharmacies also separate drugs with similar names and overlapping strengths, especially those labeled and packaged by the same manufacturer. For example, chlorpromazine 100-mg tablets and chlorpropamide 100-mg tablets, both from the same unit-dose packager (see Fig. 29-8), might pose a problem. So might tramadol 50 mg and trazodone 50 mg or metformin 500 mg and metronidazole 500 mg. Another strategy would be to change the appearance of look-alike product names on computer screens, pharmacy shelf labels and bins, and pharmacy product labels by highlighting, through boldface, color, or the use of “tall man” letters, the parts of the names that are different (e.g., hydrOXYzine and hydrALAzine). In fact, the Food and Drug Administration (FDA) Office of Generic Drugs requested that the manufacturers of 16 look-alike name-pair drug products voluntarily revise the appearance of their established names to minimize medication errors resulting from look-alike confusion. Manufacturers were encouraged to differentiate their established names visually with the use of “tall man” letters. Examples of established names involved include chlorproMAZINE and chlorproPAMIDE, vinBLASTINE and vinCRISTINE, and NICARdipine and NIFEdipine.

In institutional and community pharmacies with several staff members, everyone should have input in deciding how and where drugs are available, how doses are prepared, who is responsible for preparing them, the appearance of the storage containers, and how they are labeled. In addition, staff members should be encouraged to use a technique known as failure mode and effects analysis (FMEA) to examine the use of new products to determine points of potential failures and their effects before any error actually happens (see Chapter 8). In this regard, FMEA differs from root-cause analysis (RCA). RCA is a reactive process, employed after an error occurs, to identify its underlying causes. In contrast, FMEA is a proactive process used to carefully and systematically evaluate vulnerable areas or processes. FMEA can be employed before the purchase and implementation of new products to identify potential failure modes so that steps can be taken to avoid errors before they occur [Institute for Safe Medication Practices (ISMP), 2011a, 2011b]. Procedures to ensure safe medication use must be written, and the importance of adhering to the guidelines must be shared by all involved pharmacy, medical, and nursing personnel.

Selecting Auxiliary Labels

To help prevent errors and improve patient outcomes, pharmacists and technicians should apply auxiliary labels in certain circumstances, especially in the community setting. For example, amoxicillin oral suspension is available in dropper bottles for pediatric use. When the suspension is used for an ear infection, some parents have been known to place the suspension in the child's ear rather than to give it properly (orally). An auxiliary label, “For oral use only,” would help to prevent this administration error. However, this practice can be unsafe if the patient is unable to understand the warning. The application of an auxiliary label should not be done in lieu of speaking with the patient. A study that appeared in the American Journal of Health-System Pharmacy showed that there is a high level of misunderstanding of auxiliary labels among adults with low literacy, defined as a reading level at or below the sixth-grade level (Wolf et al., 2006). The rate of correct interpretation of these labels ranged from 0 to 78.7 percent. With the exception of the label “Take with food,” less than half of all patients were able to provide adequate interpretations of the warning labels' messages. In fact, no one was able to correctly interpret the label, “Do not take dairy products, antacids, or iron preparations within 1 hour of this medication.” Studies also have shown that a combination of a verbal description of a warning along with visual symbols improves the overall comprehension of the warning (Lesch, 2003).

In preparing fluids for injectable administration, the potential for grave errors is increased for several reasons. First, patients often are sicker when they need intravenous drugs, so the medications used have more dramatic effects on the body's function and physiology. Further, most injectable solutions are clear, colorless, water-based fluids that look alike, regardless of what drug and how much of it are actually in the fluid.

A five-hospital observational study on the accuracy of preparing small- and large-volume injectables, chemotherapy solutions, and parenteral nutrition showed a mean error rate of 9%, meaning almost 1 in 10 products was prepared incorrectly and then dispensed (Flynn et al., 1997). Error rates for complex solutions such as parenteral nutrition were especially high—37% for manual preparation and 22% for preparations that were partly automated. More recently, a 2009 State of Pharmacy Compounding Survey showed that 30% of hospitals have experienced a patient event involving a compounding error in the past 5 years (Pharmacy Purchasing & Products, 2009). In 2006, an infant received a lethal dose of zinc stemming from an error that occurred during the order entry and compounding of a total parenteral nutrition (TPN) solution. TPN was prescribed for a preterm infant born at 26 weeks gestation. On the day of the event, the physician's TPN order included directions to add zinc in a concentration of 330 mcg/100 mL. Because the automated compounder used for TPN required entry of zinc in a mcg/kg dose, the pharmacist converted the mcg/mL dose to a mcg/kg dose. She performed this calculation correctly, but accidentally entered the zinc dose in the pharmacy computer in mg, not mcg. This resulted in a final concentration of 330 mg/100 mL—a 1,000-fold overdose (ISMP, 2007).

In the sterile admixture preparation setting, dosage miscalculations and measurement errors must be minimized by systems designed with procedures that require independent double checks by two staff members. In many pharmacies, independent double checks are required for all calculations or measurements, whereas others may require it only for calculations falling in special categories, such as dosage calculations for admixture compounding for any child under age 12, critical care drug infusions requiring a dose in micrograms per kilogram per minute, insulin infusions, chemotherapy, and patient-controlled analgesia. Calculators and computer programs may improve accuracy, but they do not eliminate the need for a second person to review the calculations and solution concentrations used.

Another important way to minimize calculation errors is to avoid the need for calculations in the first place. This can be accomplished by using the unit-dose system exclusively as follows:

• Using commercially available unit-dose systems such as premixed critical care parenteral products.
• Standardizing doses and concentrations, especially of critical care drugs such as heparin, insulin, dobutamine, dopamine, and morphine.

Similar steps can be taken in community pharmacies that provide sterile admixtures to physician offices, home care programs and patients, long-term care facilities, and other clients.

The use of standard dosage charts on nursing units in institutions and standard formulations in the pharmacy minimizes the possibility of error and makes calculations much easier. For example, in critical care units, physicians need to order only the amount of drug they want infused and list any titration parameters. No one has to perform any calculations because dosage charts can be readily available for choosing appropriate flow rates by patient weight (in kilograms) and dose ordered.

Standard concentrations for frequently prepared formulations should be recorded and be readily accessible for reference in the admixture preparation area of the pharmacy. Of course, all calculations must be double-checked and documented by the pharmacist. Diluents as well as active drugs must be checked before they are added to the base solution. The stock container of each additive with its accompanying syringe should be lined up in the order it appears on the container label to facilitate the checking procedure. When compounding TPN solutions, at least three verification processes should occur in the pharmacy: after initial order entry of TPN; before manually injecting additives into the TPN; and once the TPN has been compounded. Each verification should require a pharmacist to compare the actual prescriber's order to the printed labels, and the printed labels to the additives and final product, as appropriate. Verification of manual additives should include inspection of the actual vials and syringes that contain the additives. The final verification of the compounded TPN should include a comprehensive review of the TPN order, the label on the product, and the work label.

In many hospitals and home infusion pharmacies, automated compounding equipment is being used for admixing both large- and small-volume parenterals. While automated equipment improves the accuracy of compounding parenterals, it is not perfect and has been known to occasionally fail. Also, accidents can occur when solutions are placed in the wrong additive channel. These can result in serious medication errors. Therefore, it is important that the pharmacy have an ongoing quality assurance program for the use of automated compounding equipment. This program should include double checks and documentation of solution placement within the compounder, final weighing or refractometer testing of the solution to ensure that proper concentrations have been compounded, and ongoing sampling of electrolyte concentrations. Pharmacists who prepare special parenteral solutions in batches (e.g., TPN base solutions or cardioplegic solutions) should have additional quality assurance procedures in place, including sterility testing and quarantine until confirmation.

The patient is the last individual involved in the medication use process. The pharmacist–patient interface can play a significant role in catching medication errors before they occur. Unfortunately, many health care organizations do not take advantage of this key interaction. Three important factors play a role in any patient interface and often determine the outcome of error-prevention efforts. These include direct patient education, health care literacy, and patient compliance.

Community pharmacies in the United States filled 3.6 billion prescriptions in 2009, and prescription volumes continue to increase at a rate of approximately 3% per year (StateHealthFacts, 2011). This increase in prescription volume often results in a decrease in the amount of time available for direct pharmacist involvement in patient education. A 1999 study involving community pharmacies in eight states revealed that 87 percent of all patients received written information with their prescriptions. However, only 35 percent of pharmacists made any reference to the written leaflet, and only 8 percent actually reviewed it with the patient (Svarstad, 2000). Contributing to this gap in patient education is the failure to provide patients with understandable written instructions.

The second factor is patient literacy, which includes general literacy levels and health care literacy. Many people have difficulty understanding their illness or disease, proper management of disease, and their role in maintaining their health. Whether limited by knowledge, socioeconomic factors, emotional or clinical state, or cultural background, patients' level of health literacy (i.e., the ability to read, understand, and act on health care information) often is much lower than many health care providers appreciate. Examples of patients who have had difficulty reading and understanding medication directions are plentiful. For example, an elderly patient could not tell the difference between his bottle of Coumadin® (warfarin) and Celebrex® (celecoxib). A mother, who after reading the label on a bottle of acetaminophen, could not accurately state her child's dose. And a teenager who misunderstood directions for contraceptive jelly and ate it on toast every morning to prevent pregnancy.

According a report published by the American Medical Association Ad Hoc Committee on Health Literacy, more than 40 percent of patients with chronic illnesses are functionally illiterate, and almost a quarter of all adult Americans read at or below a fifth grade level. Unfortunately, medical information leaflets typically are written at or above a tenth grade reading level. Further contributing to the dilemma is the fact that an estimated three-quarters of patients throw out the medication leaflet stapled to the prescription bag without reading it, and only one-half of all patients take their medications as directed (AMA, 1999).

One reason for this lack of understanding may be that people who have difficulty reading or understanding health information are too embarrassed or ashamed to acknowledge their deficits. Instead, they are hesistant to ask questions of their health professionals and often pretend to understand instructions. In addition, low literacy is not obvious. Researchers have reported poor reading skills in some of the most poised and articulate patients (ISMP, 2011a).

Compliance is the third patient-related factor contributing to medication errors. One study found a 76 percent difference between medications patients actually are taking when compared with those recorded in their charts as prescribed. Two factors that contribute to this high rate of discrepancy include confusion that may accompany advancing age and the increase in the number of prescribed medications (Bedell et al., 2000). Another study demonstrated that patient noncompliance played a role in 33 percent of hospital admissions (McDonnell et al., 2002).

Noncompliance may be exhibited by patients in many ways, such as not having a prescription filled initially or refilled, dose omissions, taking the wrong dose, stopping a medication without the physician's advice, taking a medication incorrectly or at the wrong time, taking someone else's medication, and financial inability to purchase their medications. Patients at risk for being noncompliant include those taking more than one drug, those with a chronic condition who are on complex drug regimens that may result in bothersome side effects, those who take a drug more than once daily, and those who have a condition that produces no overt symptoms or physical impairment such as hypertension or diabetes (National Council on Patient Information and Education (NCPIE), 2002). In addition, elderly patients are more at risk owing to factors such as decrease in mental acuity and increased confusion, lack of family or caregiver support, decreased coordination and dexterity, and impaired vision (Lombardi and Kennicutt, 2001). Pharmacy managers must consider these factors in developing and providing patient education tools or methodologies.

A New Model of Accountability

Despite a growing awareness of the system-based causes of errors, many in health care are still struggling to come to terms with the role of individual accountability. Even when we seem to understand the system-based causes of errors, it's still hard to let individuals “off the hook.” We ask, “How can we hold individuals accountable for their actions without punishment?” Some have suggested that a nonpunitive approach to error reduction leads to increased carelessness as people learn that they will not be punished for their mistakes. However, another perspective is that staff awareness of safety issues, as well as their enthusiasm for changing systems and practices associated with errors, actually grows in a nonpunitive system. A nonpunitive, system-based approach to error reduction does not diminish accountability; it redefines it and directs it in a more productive manner by focusing on the most manageable component of the error: the medication-use system itself.

Typically when an error happens, only those individuals at the “sharp end” of an error (for example, pharmacists and pharmacy technicians) are held accountable. But, we must shift from this thinking and realize that accountability must be shared among all health care stakeholders. In the new patient safety model, each individual becomes accountable, not for zero errors (which is an unrealistic and unattainable expectation for any human), but for making patient safety a part of every aspect of their job. In addition, all become accountable for identifying safety problems, implementing system-based solutions, and inspiring and embracing a culture of safety.

Because we are not capable of practicing without making errors, health care practitioners should be held accountable for speaking out about patient safety issues, voluntarily reporting potential and actual errors as well as hazardous situations, and sharing personal knowledge of what went wrong when an error occurs. Also, practitioners must be empowered to ask for help when needed, consistently provide patient education, and be willing to change their practices to enhance safety.

Management should be held equally accountable for making it safe and rewarding for practitioners to openly discuss errors and patient safety issues. They must hold regular safety briefings with staff to learn about improvement needs, discuss strategic plans, and identify new potential sources of error. When practitioners recommend error prevention strategies, leaders must support them and provide the means necessary within a reasonable timeframe to implement system enhancements to improve efficiency and safety. Leaders should be held accountable for understanding and addressing barriers to safe practice such as distractions and unsafe workloads. Leaders should incorporate patient safety as a value in the organization's mission and engage the community and staff in proactive continuous quality improvement (CQI) efforts, including an annual self-assessment of patient safety. All health care personnel should be held accountable for working together as a team, not as autonomous individuals. Finally, leaders and staff alike need to review and share safety literature frequently and offer visible support to their colleagues who have been involved in errors.

This model of shared accountability spreads far beyond the walls of individual health care settings to encompass licensing, regulatory, and accrediting bodies; institutions that educate and train students and health care professionals; professional associations; governments and public policy makers; the pharmaceutical industry; medical device and technology vendors; and even the public at large. These often-overlooked participants share equal accountability for doing their part to make health care safer. For example, licensing, regulatory, and accrediting bodies should be held accountable for adopting standards related to error reduction recommendations that arise from expert analysis of adverse events and scientific research. Educators should seek out patient safety information and use it in curriculum design. Professional organizations should support local and national reporting systems and disseminate important patient safety information to their members. Companies that produce medical devices, pharmaceutical products, health care computers and software, and other health-related products should be held accountable for pre- and postmarket evaluation, continuous improvement in the design of devices and products as well as labels and packages. Purchasers of health care should provide incentives and rewards for patient safety initiatives. And all health care professionals, including pharmacists, should encourage the public to ask questions and stay informed about their care and ways to avoid errors.

The system-based model of accountability requires that all who interact with the health care system help to define its weaknesses and find ways to make it stronger. Organizational leaders and other stakeholders who simply hold the workforce accountable when an error happens are inappropriately delegating their own responsibility for patient safety. Managers must stop blaming and punishing those closest to an error, and instead accept a model of shared accountability to collectively translate our sincere concern for patient safety into effective system-based error solutions. Implementing such solutions and inspiring and embracing a culture of change to reach the goal of safety may not be easy, but is certainly necessary.

Effective Risk-Reduction Strategies

Selecting the best error-prevention strategies is not an easy task. Often, the most effective action is not obvious, and the best error prevention tools to use in each situation are not clear, even when system-based causes have been identified. In addition, pharmacists should focus their efforts on medications that may lead to harm if involved in an error. Although most medications have a wide margin of safety, a few drugs have a high risk of causing patient injury or death if they are misused. Special precautions are needed to reduce the risk of error with these “high-alert” medications. Errors with these drugs may not be more common than with other medications, but their consequences can be more devastating (ISMP, 2008).

Listed below are examples of error-prevention strategies for creating lasting system changes for safe medication use (See Table 29-1). Those listed first are considered to be “high leverage” strategies, such as constraints and forcing functions, are more powerful because they focus on changes to the system in which individuals operate. As the list descends, strategies that target system changes but rely in some part on human vigilance and memory, considered to be “low leverage” strategies, such as education and information, are presented. Strategies toward the end are familiar and often easy to implement, but rely entirely on human vigilance (ISMP, 1999; ISMP, 2006).

• Fail-safes and constraints are among the most powerful and effective error-prevention strategies. They involve true system changes in the design of products or how individuals interact within the system. In the hospital setting, one example includes removing concentrated potassium chloride for injection from all patient care areas. In this situation, a nurse could not accidentally administer this medication because it would not be available to select from stock. Another acute care example includes eliminating nursing access to the pharmacy when it is closed by establishing a carefully selected nighttime formulary and dispensing cabinet. At a community pharmacy where the pharmacy computer system is integrated with the cash register, a fail-safe would prevent the clerk from “ringing up” the prescription unless final verification by a pharmacist was noted in the system.
• Forcing functions are procedures that create a “hard stop” during a process to help ensure that important information is provided before proceeding; often referred to as a “lock and key” design. Examples outside of health care would include the inability to drive your car unless you first place your foot on the brake before shifting the car into “drive.” For example, a pharmacy order entry system that requires a weight to be entered for each patient receiving an order for a weight-based medication (e.g., heparin, enoxaparin) before it is processed, or a bar-code scanning system that does not allow final verification of a product without a positive match between the selected product and the profiled medication.
• Automation and computerization of medication-use processes and tasks can lessen human fallibility by limiting reliance solely on vigilance or memory. Examples include use of electronic prescribing software that includes clinical decision support; pharmacy computer systems that can receive prescriptions sent electronically from a prescriber's hand-held device or computer and thus eliminate transcriptions and misinterpretations; robotic prescription preparation and dispensing technology; and computer systems that provide accurate warnings related to allergies, significant drug interactions, and excessive doses.
• Standardization creates a uniform model to adhere to when performing various functions and it tends to reduce the complexity and variation of a specific process. For example, standardized processes could be created to guide the pharmacist's final verification of a medication or to enhance the safety of giving or receiving a telephoned medication order. Prescriber use of carefully designed preprinted prescription blanks or standardized order forms that contain commonly used protocols (e.g., steroid tapers, insulin regimens, weight-based heparin) or frequently prescribed medications can reduce problems with confusing or missing instructions and illegible handwriting. On its own, standardization relies on human vigilance to ensure that a process is followed; therefore, it is less effective than the strategies mentioned previously.
• Redundancies incorporate duplicate steps or add another individual to a process to force additional checks in the system. Involving two individuals in a process reduces the likelihood that both will make the same error with the same medication for the same patient. However, the potential for error still exists since the redundant step may be omitted or ignored. Examples of redundancies include use of both brand and generic names when communicating medication information or requiring independent double checks of high-alert medications before dispensing. Patient counseling is often an underutilized redundancy that can detect many errors.
• Reminders and checklists help make important information readily available. For example, using auxiliary labels to distinguish products; building look- and sound-alike alerts into order entry systems; and using preprinted prescription blanks that include prompts for important information (e.g., medication indication, allergies, patient birth date).
• Rules and policies are useful and necessary in organizations. Effective rules and policies should guide staff toward an intended positive outcome. However, some may add unnecessary complexity and may be met with resistance, even rightfully so, especially when implemented in response to an error. Because their use relies on memory, they should be used as a foundation to support more effective strategies that target system issues.
• Education and information are important tactics when combined with other strategies that strengthen the medication-use system. But, the effectiveness of these tactics relies totally on an individual's ability to remember what has been presented. Thus, on their own, they offer little leverage to prevent errors.

While each tool mentioned above can play an important role in error prevention, beware of those that, on the surface, seem to provide the easiest and fastest solution. While strategies at the bottom of the list may be used initially, we must realize that they will not be effective for long-lasting error prevention when used alone. In order to do a better job at preventing errors, we need to employ a variety of strategies that focus on the medication-use system issues and address human factors issues for those who work within that system. Since people cannot be expected to compensate for weak systems, managers should routinely evaluate the error-prevention strategies being used in their organizations. Consider if there are more powerful strategies that could be implemented to enhance medication safety.

Risk Identification Methods

All drug-dispensing procedures should be examined regularly, both proactively and retrospectively, and the potential problems as well as causes of system breakdowns must be discovered so that prevention measures can be designed. Pharmacy staff needs to communicate clearly to managers what it takes to do the job correctly in terms of personnel, training programs, facilities design, equipment, drug procedures and supplies, computer systems, and quality assurance programs. In addition, reducing medication errors requires using a number of effective risk identification methods including a nonpunitive environment and a voluntary medication error reporting system.

Currently, many pharmacies have an ineffective approach to error reduction. Often their primary means of identifying risk involves investigations that occur during the error reporting process. These investigations tend to focus their attention on the front end or active end of the error, such as the front-line practitioner. Human nature tends to assign blame to these front-line practitioners involved in medication errors. It is easier and in our nature to blame individuals and resort to familiar solutions: disciplinary action, individual remedial education, placing error information into personnel files, or developing new rules. While these actions may not seem outwardly punitive, these forms of reprimand lead to underreporting of errors. In fact, punishing individuals for errors actually can be dangerous to an organization. It inhibits open discussion about errors, creates a defensive and reactive environment, and hinders careful and unbiased consideration of the system-based root causes of errors. Pharmacies are weakened further by punitive actions, especially if the sole responsibility for safe medication practices rests on individuals rather than on strong systems that make it difficult for practitioners to make errors.

The goal of patient safety is best served with a nonpunitive environment that places more value on a variety of risk identification methods, in addition to reporting problems so that they can be remedied, rather than on pursuing the unprofitable path of disciplining employees for errors. Methods to proactively identify the potential for problems in an organization include performing FMEAs to see where breakdowns may occur, using self-assesssment tools, and using external sources of information such as reviewing the literature for errors that have taken place in other organiztions and applying the “lessons learned” and determine if your organization could experience the same problems.

Self-assessments can help organizations assess the safety of medication practices throughout the facility, identify opportunities for improvement, and possibly compare their experiences with the aggregate experiences of demographically similar hospitals (ISMP, 2011b). One key consideration when undergoing this type of process is the establishment of an interdisciplinary team that should consist of representatives from all aspects of an organization, ranging from leadership (CEO, CMO, and nurse executives), directors of pharmacy as well as front line staff including nursing, pharmacists and physicians.

It is also important for pharmacy staff not just to focus on their own internal errors but also to look at other pharmacies' errors and methods of prevention and to learn from them. Organizations such as the ISMP, TJC, and many others provide ongoing features to facilitate these reviews in publications such as Hospital Pharmacy, Pharmacy Today, and Pharmacy and Therapeutics or newsletters that report on current medication safety issues and offer recommendations for changes.

The direct observation technique, in which a trained observer accompanies the person giving medications and witnesses the preparation and administration of each dose, was developed by Barker and McConnell (1962). The observer writes down in detail what the subject does when preparing and administering the drug (Barker et al., 2002). The notes are then compared with the prescriber's orders. An error is counted if the subject did not carry out the order accurately. The error rate is the percentage of doses administered in error, and the accuracy rate is the percentage of doses accurately administered. This method has many advantages. For example, it is not affected by lack of awareness of errors, lack of willingness to report errors, faulty memory, poor communication skills, or selective perception of the subject observed. Its reliability has been established through replication over many years. A large quantity of data can be accumulated at a rate 24 times that of chart review and 456 times that of incident reports. It can detect more errors and more types of error per unit of time than other methods, and it can identify clues about the causes so that future errors can be prevented (Cohen, 2007b).

A retrospective, voluntary, confidential reporting program provides pharmacists and pharmacy technicians with the opportunity to tell the complete story without fear of retribution. The depth of information contained in these stories is critical to understanding the error. This information is critical to identification of system deficiencies that can be corrected to prevent future errors. However, successful and sustained improvement of error-prone processes cannot occur if there is little information available about factors that contribute to an error.

Many organizational factors inhibit the reporting of medication errors. Examples include inconsistent definitions of a medication error within an organization, a punitive approach to medication errors, failure to improve the medication system or address reported problems reported by staff, lack of feedback to staff, over concern with medication error rates when reporting of an error may increase a facilities “error rate”, complex reporting processes, a perception that reporting is a low priority andconcern for personal liability. A voluntary program encourages practitioners to report hazardous situations and errors that have the potential to cause serious patient harm. A confidential reporting system where everyone understands that errors will not be linked to performance appraisals is critical. Many pharmacy organizations have regular meetings where medication errors are addressed. The results of these meetings often are not shared with the front-line staff, therefore giving the impression that “nothing is being done” when errors are reported. In addition, busy practitioners tend to avoid reporting errors owing to the cumbersome nature of their organizations' reporting forms and processes. It is important to make error reporting easy, reward error reporting, and provide timely feedback to show what is being done to address problems. Consistently applying a nonpunitive approach to errors is important. If even one person is disciplined for an error, mistakes will be hidden. Employees should not be evaluated based on errors or lack of making mistakes but on positive measures that evaluate an employee's overall contribution to the organization. Armed with these tools, pharmacists can become aware of the deficiencies in their organizations and make performance-improvement changes. Without them, pharmacies can only address errors when they surface rather than at the root cause.

To be successful, medication error-reduction efforts must result in system improvements that are identified through a four-pronged analysis of errors. The first two prongs, both reactive in nature, include analysis of organization-specific errors that have caused some degree of patient harm and analysis of aggregate medication error data (e.g., trends by drugs or location of drugs involved in errors). Equally important, the other two prongs, both proactive in nature, include analysis of “close calls” (errors that have the potential to cause patient harm) and analysis of errors that have occurred in other organizations. Each prong contains valuable information about weaknesses in the system that, collectively, can lead to effective error-reduction strategies. Yet many organizations focus primarily on the first two prongs of error analysis and action. Most often proactive efforts are not given high priority. As a result, organizations may be busy “fighting fires” rather than preventing them. A “close call” should be clear evidence that a tragic event could occur. Unfortunately, too often this wakeup call is not heard. Little attention is focused on thorough analysis of errors that, fortunately, do not cause actual patient harm, especially if organizations identify errors that require analysis by a severity rating that is based on actual patient outcome. For example, a serious overdose detected before administration may not be given the same priority and analysis as a similar error that actually reached and possibly harmed the patient. Worse, some organizations fail to use errors that have occurred elsewhere as a road map for improvement in their own organization. Staff will be more comfortable discussing a serious external error than one that has occurred within its own organization. Because blame is not an issue, defensive posturing and other obstacles to effective discussion will not be present. Staff can identify possible system-based causes of the error more easily and the likelihood of a similar error in their facility and make suggestions for improvement. As improvements are made, enthusiasm builds for identifying, reporting, and analyzing errors that are actually occurring within the organization. In the end, discussion about external errors leads to more effective analysis of internal errors.

Multidisciplinary educational programs should be developed for health care personnel about medication error prevention. Because many errors happen when procedures are not followed, this is one area on which to focus through newsletters, discussions during staff meetings, and in-service training.

Lastly, organizations should report medication errors externally so that the lessons learned from these events can be shared with others to prevent future occurances in their facilities. In 1999, the IOM issued a landmark report entitled, “To Err is Human: Building a Safer Health System” that highlighted critical areas of research and activities needed to improve the safety and quality of health care delivery. One critical area of the IOM report addressed the reporting and analysis of data on adverse events.

The IOM report and its findings spotlighted a serious need to capture information that would help to improve quality and reduce harm to patients. Addressing this need, Congress passed The Patient Safety and Quality Improvement Act of 2005 (Patient Safety Act). To implement the Patient Safety Act, the Department of Health and Human Services issued the Patient Safety and Quality Improvement final rule. The Patient Safety Act and the Patient Safety Rule authorize the creation of Patient Safety Organizatoins (PSOs) to improve quality and safety through the collection and analysis of data on patient events [Agency for Healthcare Research and Quality (AHRQ), 2011].

PSOs are organizations that share the goal of improving the quality and safety of health care delivery. Organizations that are eligible to become PSOs include: public or private entities, profit or not-for-profit entities, provider entities such as hospital chains, and other entities that establish special components to serve as PSOs. By providing both privilege and confidentiality, PSOs create a secure environment where clinicians and health care organizations can collect, aggregate, and analyze data, thereby improving quality by identifying and reducing the risks and hazards associated with patient care. ISMP is certified as a Patient Safety Organization (PSO) by the Agency for Health care Research and Quality, which offers the highest level of legal protection and confidentiality for medication error reporting and analysis

Reporting errors to organizations like ISMP can also inform pharmaceutical companies of labeling and packaging problems. Many pharmaceutical companies have responded to suggestions made by technicians and pharmacists. Health professionals can report all of the aforementioned close calls and actual medication errors by using the ISMP Medication Error Reporting Program (MERP). Reports are forwarded to the individual pharmaceutical company and the FDA, and ISMP provides follow-up when appropriate. Call 1-800-FAILSAFE, visit ISMP's Web site at www.ismp.org, or complete an ISMP MERP report (see Fig. 29-9).

Pharmacists play a key role in the drug use process throughout any health care organization. Pharmacists and other members of the pharmacy department should lead a multidisciplinary effort to examine where errors arise in this process. Pharmacy department administrators should encourage their staff to collaborate in designing quality assurance programs to obtain information that helps to establish priorities and make changes. For example, joint reviews of the accuracy of unit-dose cart fills are of great help in detecting reasons for missing or inaccurate doses and changing the drug-dispensing system accordingly. Programs can be established to monitor the accuracy of order entry into computers in the pharmacy. Quality assurance efforts that include a review of medication error reports help to develop a better understanding of the kinds of systemic or behavioral defects being experienced so that necessary corrections can be identified. The medication error problem will never be eliminated completely, but pharmacy managers, working together with their staffs and other health care providers, can use their expertise to address issues of safety and thus ensure best outcomes in the safest environment possible.

1. Have you witnessed or been involved in a medication error, and what actions did you take to prevent any future errors?

2. Have you counseled a patient whom you know did not understand his or her doctor's instructions? What steps did you take to ensure that the patient thoroughly knew his or her medication, directions, and indication for use?

3. Will you include medication error prevention strategies in the orientation process for new employees and ongoing education for pharmacy staff?

4. Do you think that you will be involved in reporting medication errors both internally within your organization and externally to reporting agencies such as the ISMP or the FDA? Do you think this will make a difference in your pharmacy practice setting?