Introduction
Medication therapy is becoming increasingly more complex as new drugs are developed and more therapeutic targets are elucidated. In addition, polypharmacy (≥5 scheduled medications) has become exceedingly common in geriatric patients and in patients with chronic disease states. As the complexity of drug therapy and the number of medications increase, patients are at a high risk for medication errors and adverse drug events (ADEs), or injuries resulting from medication. The National Coordinating Council for Medication Error Reporting and Prevention (NCC MERP) defines a medication error as “any preventable event that may cause or lead to inappropriate medication use or patient harm while the medication is in the control of the health care professional, patient, or consumer.”1 The Institute of Medicine (IOM) simplifies the definition by stating that a medication error is “any error occurring in the medication-use process.”2 Medication errors are particularly problematic when they lead to preventable ADEs, or injuries resulting from medication. The IOM estimated in 2006 that 25% of medication errors led to preventable ADEs. While the majority of ADEs are nonpreventable, it is still imperative that hospitalists recognize the seriousness of preventable ADEs and how these ADEs can be prevented.
Incidence
Accounting for the true number of medication errors and ADEs is difficult since many go unrecognized or are not reported. Even though many databases are anonymous and nonpunitive, many professionals do not report medication errors and ADEs out of fear; institutions then cannot identify system failures and address the problems. In 2006, the IOM released Preventing Medication Errors: Quality Chasm Series and published staggering statistics about the number and cost of medication errors. The IOM estimated that each hospitalized patient experiences an average of one medication error daily. Though not all of these medication errors result in preventable ADEs, the IOM estimated that 1.5 million ADEs occur annually with 390,000–450,000 occurring within the hospital setting. Another study at a large academic medical center found that 2 of every 100 admitted patients experienced a preventable ADE resulting in $2.8 million in additional costs to the institution. Due to underreporting and underrecognition, these estimates likely under-estimate the true number of medication errors and ADEs. Billions of additional dollars are spent each year as a result of medication errors and preventable ADEs. Identifying common error types and implementing strategies that reduce preventable errors and ADEs are the first steps to reducing the number and cost of errors.
Common Error Types
There are two ways to classify medication errors: commission errors that result from an action (ie, wrong dose or wrong route) and omission errors that result from no action (ie, missed doses or untreated indications). Both of these error types may occur during any step in the medication use process; however, most errors occur during prescribing and medication administration.
Hospital prescribing errors occurred at a rate of 0.3 per patient per day as reported by the IOM in 2006. Another study reported that approximately 1% of prescribing errors resulted in harm to hospitalized patients.3 Commission-related prescribing errors include selecting the wrong drug, wrong dose, wrong route of administration, wrong frequency, wrong dosage form, or entering an order on the wrong patient. Prescribing errors of omission include missing pertinent patient information (allergies, height, weight, previous ADEs, diagnoses, and home medications). Writing a safe medication order involves several steps, and physicians can reduce prescribing errors by clearly printing handwritten orders and including all necessary components: medication name, dose, route, frequency, and any special instructions, while avoiding ambiguous abbreviations and phrases. The physician can improve accuracy and completeness by taking additional time to review the order prior to signing it and sending it to the pharmacy.
Prescribing errors and preventable ADEs may also arise from deficits in convenient drug information. The Internet program Epocrates provides a free drug information database that is downloadable to handheld devices, allowing convenient and efficient access to dosing guidelines and basic drug information. More comprehensive drug information databases, such as Lexi-Drugs by Lexi-Comp and Micro-medex by Thomson Reuters, are available for purchase and download to handheld devices. Some health care institutions have subscriptions to these databases that allow providers to download handheld versions free of charge. In addition, physicians may consult with pharmacists for help with complex medication reconciliation, polypharmacy minimization, complicated medication selection and dosing, and drug information requests. The pharmacist can provide valuable information regarding medication therapy management, therapeutic drug monitoring, and drug interactions. Team-based or unit-based clinical pharmacists can provide point-of-care drug information, are experts on safe medication use, and can aid physicians in appropriate drug selection and dosing. The Institute for Safe Medication Practices (ISMP) publishes a list of high-alert medications that generate significant patient harm when used incorrectly (Table 10-1). Prescribers should exercise caution when prescribing these agents and ensure that the patient orders are clearly, accurately, and appropriately written. Prescribers should also comply with institution-specific cautions and guidelines for prescribing high alert medications.
Classes/categories of medications
|
Specific medications
|
Background
|
Benefits: Computerized physician order entry (CPOE) has been shown to reduce inpatient prescribing er-rors; however, the evidence is limited. One study reported that CPOE could decrease serious medication prescribing errors by up to 55%.4 A systematic review of multiple studies evaluating CPOE showed that 23 of 25 studies reported a significant relative risk reduction in medication errors, 6 of 9 studies reported a significant relative risk reduction in potential ADEs, and 4 of 6 studies reported a significant relative risk reduction in ADEs. Though the authors reported several study limitations and included studies with weak designs, the overall conclusion was that CPOE appears effective in reducing medication errors and ADEs.5 Data regarding error severity are scant; however, some literature reports CPOE decreases both major and minor prescribing errors. Computerized physician order entry provides the prescriber with preset choices for commonly used medications and doses, which may aid in proper dose selection and reduce errors in order writing. Prescribers may also benefit from a CPOE-based clinical decision support system, such as allergy and drug interaction screening, dosing calculators, and drug information resources, which can result in safer medication and dosing selection. A subgroup analysis from a systematic review of CPOE studies showed that 14 studies with advanced clinical decision support systems reported greater relative risk reductions in medication errors compared to studies with limited to no decision support.
Risks: Though CPOE systems may reduce many types of medication errors, they may also be responsible for generating new error types. Overuse of low-importance and irrelevant clinical decision support alerts may lead to prescriber alert fatigue, which results in prescribers ignoring potentially serious alerts. High-sensitivity alerts should be implemented for critical warnings to minimize alert fatigue. Prescribers may accidentally select the wrong medication from the drop-down menu, especially when look-alike medications are listed in alphabetical order. Many institutions have included “Tall Man” letters into their CPOE systems to reduce the likelihood of a prescriber selecting the wrong drug (eg, levoFLOXACIN and leveTIRACETAM or hydrALAzine and hydrOXYzine). Duplicate orders have been a reported problem with CPOE systems since order entry screens may make it difficult to see what has already been entered for a patient. Computerized physician order entry alone will not alleviate all types of errors. A study in France showed that even after implementation of CPOE with a clinical decision support system, team-based pharmacists were still able to identify 33 drug-related problems per 100 admissions. Pharmacists sent identified drug-related problems through CPOE to a patient’s medical record for physician review. Therefore, the authors of the study recommended the inclusion of pharmacists on medical rounds to identify drug-related problems.6
Transcribing errors occur when a written or verbal order is inappropriately translated and dispensed. Transcribing errors often result from illegible handwriting, look-alike/sound-alike medications, and inappropriate abbreviations. Computerized physician order entry can significantly reduce transcription errors by eliminating handwritten orders and prompting prescribers to enter all pertinent information to avoid ambiguous or incomplete orders.
The Joint Commission compiles lists for problematic look-alike/sound-alike medications based on specific practice settings. Table 10-2 lists problem medications for the hospitalist. The look-alike/sound-alike medications pose significant risk for generating medication errors, especially when combined with illegible handwriting and unclear verbal communication. In addition, the U.S. Food and Drug Administration (FDA) utilizes people and computer software to review medication names submitted for new drugs prior to approval to reduce the number of look-alike/sound-alike medications. The FDA also monitors medication error reports resulting from drug name confusion on approved drugs. The FDA may require pharmaceutical companies to change a product’s brand name after initial marketing when errors based upon the product name being too similar to another product already on the market are reported. For example, Alto-cor (lovastatin) was being confused with Advicor (lovastatin/niacin) and, subsequently, Altocor’s name was changed to Altoprev. Galan-tamine, an acetylcholinesterase inhibitor for Alzheimer disease, was initially marketed as Reminyl but was subsequently changed to Razadyne when errors resulted from confusion with Amaryl (glimepiride), a sulfonylurea for diabetes.
Generic Name |
---|
Concentrated morphine liquid and conventional morphine liquid |
Ephedrine and epinephrine |
Hydromorphone and morphine injection |
Hydroxyzine and hydralazine |
Insulin preparations |
Liposomal daunorubicin and doxorubicin and conventional daunorubicin and doxorubicin |
Liposomal amphotericin B and conventional amphotericin B |
Meformin and metronidazole |
Oxycodone CR (OxyContin) and Oxycodone IR |
Vinblastine and vincristine |
These safeguards help to reduce the risk of medication errors; however, clear written and verbal communication is also imperative. The Joint Commission publishes a “Do Not Use” list to reduce use of error-prone abbreviations and symbols in medical practice, as listed in Table 10-3. For example, the popular abbreviation QD, has been mistaken for QOD or QID