Mistakes involving medications are among the most common healthcare errors. Medication errors lengthen hospital stays, increase inpatient expenses, and lead to more than 7,000 deaths annually in the United States. Each error costs an estimated $2,000 to $8,750. An error can happen in the home or a healthcare facility; this article focuses on errors in hospitals.
While any medication potentially can cause harm, a select group of drugs—high-alert medications (HAMs)—carries a higher risk of patient injury. According to The Joint Commission (TJC), HAMs frequently are associated with harm, the harm they cause is serious, and when they’re misused, the risk of serious injury or death is high. Even when given correctly, these drugs carry a significant risk of causing harm. The Institute for Safe Medication Practices (ISMP) describes HAMs as drugs “that bear a heightened risk of causing significant patient harm when…used in error.”
HAMs share several characteristics—a narrow therapeutic index and the risk of significant harm if the wrong route is chosen or a system failure occurs. Drugs with a narrow therapeutic index are dangerous because small changes in dosage or blood drug levels can lead to dose- or blood concentration-dependent critical therapeutic failures or adverse drug events. These adverse events are persistent, life-threatening, permanent, or slowly reversible and can lead to disability, the need for hospitalization, or death.
Organizations with guidelines on using HAMs include the ISMP, the Institute for Healthcare Improvement (IHI), and TJC. Some of these organizations also monitor errors involving HAMs. (See Organizations that focus on medication errors.)
The number of drug categories and specific medications identified as high alert varies with the agency or organization. All relevant organizations identify four specific HAM drug classes—anticoagulants, sedatives, insulins, and opioids—because they’re frequently linked to potentially harmful outcomes. (See Major adverse effects of high-risk medications.)
Reports received by ISMP reveal most insulin errors stem from human error (concentration lapses, distractions, and forgetfulness) related to dosage measurement and hyperkalemia treatment. Most of these errors stemmed from knowledge deficits—for instance, related to differences between insulin syringes and other parenteral syringes and the perceived urgency of treating hyperkalemia.
Here’s an example: Treatment for hyperkalemic patients with renal failure may involve dextrose 50% injection (D50) and insulin to help shift potassium from the extracellular space to the intracellular space. This helps stabilize cardiac-cell membranes to allow time for other definitive treatments. In one error reported to ISMP, the physician prescribed 50 mL of D50 along with 4 units of regular insulin I.V. (U-100) for a patient with severe hyperkalemia and renal failure. The nurse mistakenly drew 4 mL (400 units) into a 10-mL syringe and administered it I.V. The patient suffered severe hypoglycemia.
Other reports involving hyperkalemia treatment indicate D50 wasn’t given when the patient received insulin and became profoundly hypoglycemic. Additional confusion can occur with U-500 insulin because it’s given in a tuberculin syringe, not an insulin syringe. A patient who receives 150 units of U-500 twice daily should be taught to withdraw 0.3 mL insulin in a tuberculin syringe. Many patients and nurses mistakenly believe the patient’s receiving 30 units of insulin, not 150 units.
Some insulin errors result from storing multiple concentrations and drug strengths next to one another, similar packaging of some drugs, and design flaws of certain insulin pens. With some pens, the user can easily misread the digital display of the dose when holding the pen upside-down. For example, a dose of 12 units might look like 21 units.
The most common anticoagulant errors are administration mistakes, including incorrect dosage calculation and infusion rates.
- Anticoagulants were linked to 59,316 errors reported to the United States Pharmacopeia MEDMARX registry from 2001 to 2006. Roughly 60% of these errors reached the patient, and about 3% caused death or harm.
- From January 1997 to December 2007, 446 medication-error sentinel events were reported to TJC’s sentinel event database. About 7% were associated with anticoagulants; two-thirds of these involved heparin. Twenty-eight deaths occurred and six patients suffered loss of function.
- In 2005, enoxaparin was associated with four patient deaths and two cases of harm.
Errors involving newer anticoagulants
Since 2010, the Food and Drug Administration (FDA) has approved three target-specific oral anticoagulants. Unlike warfarin, which blocks multiple steps in the coagulation cascade, these newer anticoagulants block just one step. The first of these drugs, dabigatran, was approved in 2010 for stroke prevention in patients with nonvalvular atrial fibrillation and in 2014 for treatment of pulmonary embolism and deep vein thrombosis. Rivaroxaban was approved in 2011; apixaban, in 2012.
From October 2010 to December 2012, QuarterWatch (published by ISMP) noted an increased incidence of severe and fatal bleeding events in patients with a median age of 80. The FDA has received 7,387 reports of serious events associated with dabigatran, including 1,158 deaths. In 2014, the agency released new information revealing that although dabigatran is less likely than warfarin to cause intracranial hemorrhage, it’s more likely to cause GI bleeding.
In the United States, many patients have switched from warfarin to dabigatran. However, no widely available laboratory test exists in this country to monitor dabigatran blood levels. In Europe and Canada, the hemoclot thrombin inhibitor kit assay determines thrombin clotting time, which correlates to plasma dabigatran levels. Using this test would help clinicians identify patients at higher risk of bleeding.
Another disadvantage of the newer oral anticoagulants is lack of a reversal agent if catastrophic bleeding occurs. To help reduce bleeding risk during anticoagulation therapy, clinicians should use a calculator. (See Online bleeding and stroke risk calculators.) Before prescribing dabigatran, clinicians should weigh the patient’s risks of bleeding and falling against the drug’s potential benefits.
Even when prescribed in appropriate dosages, opioids can cause harm. In 2013, the Society for Critical Care Medicine released new guidelines for the treatment of pain, agitation, and delirium. TJC periodically issues newsletters identifying important sentinel events and steps healthcare organizations should take to mitigate these events. An August 2012 sentinel evert alert reported that 47% of opioid-related adverse events in hospitals from 2004 to 2011 resulted from incorrect dosages, 29% from improper patient monitoring, and 11% from other factors (including drug interactions, excessive dosing, and adverse drug reactions).
Some hospital patients use patient-controlled analgesia (PCA) pumps to control pain. Based on reports submitted to the FDA Manufacturer and User Device Experience (MAUDE) database, PCA pumps carry a threefold higher risk of injury or death than general device infusion pumps. A review of the MAUDE database as of January 31, 2011, revealed 4,230 errors resulting in 826 injuries or deaths associated with PCA use, compared to 48,961 errors and 3,240 injuries or deaths associated with I.V. infusion pumps. Despite the greater number of errors with PCA use, the percentage of injuries associated with errors is three times higher with PCA pumps than with I.V. pumps (19% vs. 6.7%). The majority of these PCA errors stemmed from erroneous pump programming or using the wrong medication in the device. To improve PCA safety, many organizations use standardized pain medication order sets and dilutions for PCA medications. Some also use different order sets for opioid-naïve and opioid-tolerant patients.
Sedatives, such as chloral hydrate and benzodiazepines, commonly are given for procedural sedation and during hospitalization. Inappropriate use can lead to oversedation, lethargy, hypotension, and delirium. Also, sedatives may increase the risk of falling. Sometimes, sedatives and opioids are administered together, with a synergistic effect that leads to central nervous system depression. The IHI publication “How-to Guide: Prevent harm from high-alert medications” describes results of periodic surveys of healthcare professionals conducted by ISMP. When asked about benzodiazepine use in elderly patients, many respondents said they consider this drug class to be high risk with this patient population.
In addition, sedatives may lead to harm if clinicians aren’t familiar with the specific medication. For instance, they may titrate the dosage without knowing the upper dose limits, or they may be unaware of the drug’s onset of action. Harm also may occur if the patient experiences respiratory depression or respiratory arrest in a facility without appropriate safety measures. The Minnesota Department of Health found approximately half of patients who fell and sustained serious injury were prescribed one or more problem medications associated with an increased risk of falling in the 24 hours preceding their fall. The most common problem medications linked to an increased fall risk were sedatives, opioids, antihypertensives, anti-anxiety drugs, and antipsychotics.
Research indicates nurses intercept 50% to 86% of medication errors before these errors reach the patient. In the hospital, medication delivery is a three-tiered process: a practitioner orders the medication, a pharmacist prepares it, and a nurse administers it. If the practitioner makes a mistake when ordering it, the pharmacist and nurse have a chance to intercept it. If the pharmacist makes an error in filling the order, the nurse has an opportunity to intercept it before it reaches the patient. Unfortunately, errors nurses make are likely to reach the patient.
IHI, ISMP, and other organizations recommend various strategies to reduce the risk of errors. One strategy involves multidisciplinary pharmacy and therapeutics committee teams with nurses and pharmacists working together. The Health Research and Educational Trust also has issued specific recommendations. (See Web resources for clinicians.)
Performing independent double-checks
Performing an independent double-check (IDC) helps ensure safe administration of HAMs. According to ISMP, IDCs can prevent up to 95% of errors before they reach the patient. In a properly conducted IDC, the second nurse verifies that the patient, drug, dosage, and route are correct and match the physician’s order. In the scenario below, the second nurse caught her colleague’s mistakes.
Nurse Laura is caring for a postoperative diabetic patient in the intensive care unit. When his blood glucose value soars to 472 mg/dL, she notifies the intensivist, who orders an insulin infusion and 10 units of regular insulin by I.V. push. Laura knows the insulin syringe has an affixed needle and the I.V. tubing has needleless ports. She draws up the insulin in a tuberculin syringe and asks another registered nurse to verify the dose, per hospital policy for high-risk medications. The second nurse notes that Laura has drawn up the wrong amount of insulin and is using a tuberculin syringe instead of an insulin syringe. She refuses to electronically sign the verification. If she hadn’t caught Laura’s mistakes, the patient would have received 1 mL of U-100 insulin, or 100 units I.V. push. As a result, the patient’s blood glucose value would have dropped significantly, necessitating hypoglycemia treatment.
When nurses explain the IDC process to patients, the level of trust increases as patients witness the nursing staff taking measures to ensure their safety. So why don’t all nurses perform IDCs if it’s a proven way to reduce drug errors? ISMP research shows IDCs are time consuming. Medication administration already takes up about 25% of a nurse’s typical shift. Also, with staffing shortages and increasing workloads, a second nurse may not always be available to perform an IDC.
What’s more, not all electronic medication administration record (MAR) systems allow IDC documentation. Some organizations are using creative ways to rectify this situation. When staff at one Veterans Health Administration facility noted its electronic MAR didn’t permit IDC documentation, they worked with the clinical documentation staff to make IDC documentation a “hard stop” with certain HAMs. Also, to speed the IDC process, they made personalized barcode identification cards for each nurse that can be scanned quickly instead of requiring manual entry for the second nurse’s information.
Limiting interruptions during medication administration
Interruptions contribute to medication errors by disrupting the clinician’s concentration and focus. In one study, the chance of making a medication error increased approximately 12% with each interruption during a single administration episode. The error rate doubled when four or more interruptions occurred. High workloads, high patient acuity, poor staffing ratios, and a chaotic environment contribute to interruptions and distractions.
A review of 54 studies on hospital medication administration errors found that in 16 studies, interruptions and distractions contributed significantly to errors. Wrong drug, wrong dosage calculation, and wrong administration time were common errors caused by interruptions. Lapses leading to omitting a drug dose may range from a minor delay (forgetting what the nurse intended to get from the medication cabinet) to a nurse thinking she gave a drug even though she didn’t. Dangerous lapses include forgetting to discontinue an insulin infusion after an enteral feeding is halted and giving an I.V. push medication too rapidly, thinking it’s the saline flush instead.
Several methods have been studied to decrease interruptions. In 1981, the Federal Aviation Administration instituted the “sterile cockpit” rule after it found nonessential conversations and interruptions contributed to some airplane crashes. Several hospitals have applied portions of the sterile cockpit rule to nursing units and medication administration procedures. One hospital created a no-interruption zone by encircling the area around the medication cart with red tape to indicate nurses inside the red zone weren’t to be interrupted. Over a 3-week period, interruptions decreased by 40.9%. In some hospitals, nurses administering medications wear yellow or red vests to serve as a visual reminder to others not to interrupt them. Also, hospitals can place warning signs on medication-dispensing machines and medication carts to limit distractions.
One nursing unit determined interruptions during medication administration came primarily from patients’ families, transporters, physicians, and telephone calls. They limited communications during medication administration by screening telephone calls and placing removable warning signs on medication carts during medication administration.
Reducing confusion around drug names
Many drug names look or sound like those of other drugs. Confusing drug names are a leading cause of medication errors. Several HAMs are on ISMP’s confused-drug name list, including these drug pairs:
- Pavulon/Peptavlon. (For the complete list, visit www.ismp.org/tools/confuseddrugnames.pdf.)
Strategies for reducing errors related to look-alike, sound-alike, or confusing medications include using tall-man lettering (such as DOPamine and DOBUTamine) and separating the pairs in dispensing machines and on storage shelves.
Similar packaging from one drug to another, and even from one dosage strength to another,
also causes confusion that can lead to medication errors. Here’s an example:
While caring for Morris Wilson, age 72, Nurse Jessica notices his heart rhythm has suddenly changed to ventricular tachycardia. She reaches into the medication cabinet and grabs a prefilled lidocaine syringe in a red and white box. In her haste, she has mistakenly picked up the 1-g box instead of the 100-mg box. As she pushes the last of the lidocaine into Mr. Wilson’s I.V. line, she realizes her mistake. Unfortunately, he suffers cardiopulmonary arrest and requires resuscitation.
In this example, the 1-g and 100-mg dosage forms came in look-alike packages and were stored side by side. After the incident, staff requested the hospital obtain premixed infusions and remove the higher-dose lidocaine from the unit stock to help prevent similar errors.
A goal within reach
In the 15 years since the Institute of Medicine (IOM) published To Err Is Human: Building a Safer Health System, medication errors remain near the top of the list of harmful events for hospital patients. Such errors persist even after the IOM’s 2006 report Preventing Medication Errors found medication errors harm 1.5 million patients each year. Interventions designed to reduce or eliminate interruptions during medication administration have been shown to reduce errors—but only when staff use them properly.
Standardization of drug labeling, storage, concentrations, and dosages has significantly decreased errors involving HAMs. IDCs and barcode scanning have helped, too. Combining multiple strategies can reduce errors even further. Staff education and buy-in are important to ensure compliance with error-reduction strategies and overall success of interventions.
Understanding the medications we administer and the correct way to administer them are vital to preventing errors. While great strides have been made in error prevention, we still haven’t reached our goal—to make HAM errors “never” events. With vigilance, knowledge, standardization, and automatic safeguards, we can achieve our goal and ensure safety for the hospital patients in our care.
Baldwin K, Walsh V. Independent double-checks for high-alert medications: essential practice. Nursing. 2014;44(4):65-7.
Committee on Identifying and Preventing Medication Errors, Aspden, P, Wolcott J, et al. Preventing Medication Errors: Quality Chasm Series. Washington, DC: The National Academies Press; 2006.
Health Research & Educational Trust. Implementation guide to reducing harm from high-alert medications. 2012. www.ihconline.org/UserDocs/Pages/HRET_HEN_Change_Packages_AllMay2012.pdf
Institute for Healthcare Improvement. How-to Guide: Prevent Harm from High-Alert Medications. Cambridge, MA: Institute for Healthcare Improvement; 2012.
ISMP Medication Safety Alert. How FDA can help reduce dabigatran (PRADAXA) bleeding risks. Institute for Safe Medication Practices. February 13, 2014. www.ismp.org/newsletters/acutecare/showarticle.aspx?id=71
ISMP Medication Safety Alert. Misadministration of I.V. insulin associated with dose measurement and hyperkalemia treatment. Institute for Safe Medication Practices. August 11, 2011. www.ismp.org/newsletters/acutecare/articles/20110811.asp
ISMP Medication Safety Alert. Survey suggests possible downward trend in identifying key drugs/drug classes as high-alert medications. Institute for Safe Medication Practices. July 3, 2014. www.ismp.org/newsletters/acutecare/showarticle.aspx?id=83
ISMP Medication Safety Alert. Your high-alert medication list—relatively useless without associated risk-reduction strategies. Institute for Safe Medication Practices. April 4, 2013. www.ismp.org/Newsletters/acutecare/showarticle.aspx?id=45
Joint Commission, The. Sentinel Event Alert 49. Safe use of opioids in hospitals. August 8, 2012. www.jointcommission.org/assets/1/18/SEA_49_opioids_8_2_12_final.pdf
Keers RN, Williams SD, Cooke J, Ashcroft DM. Causes of medication administration errors in hospitals: a systematic review of quantitative and qualitative evidence. Drug Saf. 2013;36(11):1045-67.
Mattox EA. Strategies for improving patient safety: linking task type to error type. Crit Care Nurse. 2012;32(1):52-78.
Ritter III HTM. Making Patient-Controlled Analgesia Safer for Patients. Pa Patient Saf Advis. 2011;8(3):94-9.
Tham E, Calmes HM, Poppy A, et al. Sustaining and spreading the reduction of adverse drug events in a multicenter collaborative. Pediatrics. 2011;128(2):e438-45.
Thomas M. Evaluation of the personalized bar-code identification card to verify high-risk, high-alert medications. Comput Inform Nurs. 2013;31(9):412-21.
Trbovich P, Prakash V, Stewart J, Trip K, Savage P. Interruptions during the delivery of high-risk Medications. J Nurs Adm. 2010;40(5):211-8.
Westbrook JI, Woods A, Rob MI, Dunsmuir WR, Day RO. Association of interruptions with an increased risk and severity of medication administration errors. Arch Intern Med. 2010;170(8):683-90.
Pamela Anderson is a nurse practitioner for the peripheral vascular surgeons of St. Vincent Physician Network in Indianapolis, Indiana. Terri Townsend is a medical-surgical staff educator for Community Hospital Anderson in Anderson, Indiana.