Adverse drug events: Do you know what the warning label really says?

By Katherine A. Hawkins, MD, JD, Associate Professor of Clinical Medicine, Albert Einstein College of Medicine, Bronx, NY.

Editor’s note: Medication errors have been publicized as among the most dangerous risks to emergency department (ED) and hospitalized patients. Newspaper and magazine articles have leaped on the issue. While physicians and hospitals would like to believe this is just news propaganda to increase sales, they are wrong. Studies have shown that patients are dying from preventable adverse drug events (ADEs). Although hospitals, physicians, nurses, risk managers, and pharmacists have made attempts to reduce risks and prevent ADEs, there is much more that must be done. This month’s ED Legal Letter describes different types of ADEs and develops risk management strategies to reduce the chance of medication error. Adopting the guidelines provided in this issue will create a safer environment for our patients.

Introduction

An adverse event is an injury caused by medical management rather than the underlying condition of the patient. An ADE is defined as an injury resulting from medical intervention related to a drug. Not all ADEs are attributable to errors; an adverse event attributable to error is a "preventable adverse event." Medication-related error has been studied extensively for several reasons: It is one of the most common types of error; substantial numbers of individuals are affected; and it accounts for a sizable increase in health care costs.1 More people are affected because of the extensive use of drugs in both out-of-hospital and in-hospital settings.2

The Institute of Medicine’s study, "To Err is Human," estimated that there were at least 44,000 and perhaps as many as 98,000 deaths due to preventable medical errors in the United States each year.3 A review of U.S. death certificates revealed that 7391 people died in 1993 from medication errors.4 High error rates with serious consequences are most likely to occur in intensive care units, operating rooms, and EDs. Errors result in lost lives, physical and psychological discomfort, loss of trust in the health care system, diminished satisfaction by both patients and health professionals, and costs — estimated at between $17 billion and $29 billion per year, including the expense of additional care necessitated by the errors, lost income, household productivity, and disability.5

A total of 2103 medication-related errors thought to have clinical importance were detected at a major medical center during a yearlong study period. The overall rate of errors was 3.99 errors per 1000 medication orders. The 696 errors with potential for adverse patient effects were evaluated for likely related factors. The most common specific factors associated with errors were decline in renal or hepatic function requiring alteration of drug therapy (13.9%); patient history of allergy to the same medication class (12.1%); using the wrong drug name, dosage form, or abbreviation (11.4%); incorrect dosage calculations (11.1%); and atypical or unusual and critical dosage frequency considerations (10.8%).6

The ED is at higher risk for medication errors than other hospital areas because of extreme time pressure, high patient volumes, verbal orders, and multiple staff caring for patients.7 In their study of patients experiencing adverse events admitted to hospitals in Colorado and Utah, Thomas, et al. found that about 30% of events were attributable to negligence and that the hospital location with the highest proportion of negligent adverse events (52.6%) was the ED. The authors note the complexity inherent in emergency medical care and point to the need to improve the teamwork and standardize work procedures.8

ED ADEs may be categorized into those occurring in allergic or hypersensitive patients; those due to sedative/anesthetic/narcotic drugs; those due to drug interactions; and those due to wrong drug/wrong dosage/wrong route/wrong patient errors.

Patients with Allergy/Hypersensitivity

Baylis v. Lourdes Hospital.9 Elizabeth Baylis became ill in 1984 and took the antibiotic Keflex (cephalosporin), which she had left over from an old prescription from her dentist. She suffered a reaction and went to the ED, where it was determined she probably was allergic to cephalosporin. A few weeks later, she called the office of Dr. Wheeler concerning a stomach problem, and was informed that he was in surgery at the hospital but that if she would go to the ED, he might be able to see her.

Upon her arrival at the ED, she was examined and asked if she was allergic to any medication. Ms. Baylis replied that she was allergic to penicillin, which was noted in the ED records. Following further consultation and x-rays, she was given Phenergan (promethazine). Dr. Wheeler also prescribed Mylicon (simethicone) and Keflex (cephalosporin) and left instructions for her to call him. After going home and taking one capsule, the patient collapsed and was returned to the hospital in a state of shock. Upon arrival, she was found to have no blood pressure but was awake. She received epinephrine and was resuscitated, and her vital signs normalized. She was diagnosed with anaphylaxis with cardiovascular collapse. The ED progress note stated, "It was felt that the patient had had an anaphylactic reaction to the Keflex."

The plaintiff’s expert testimony dealt largely with defendants’ negligence in prescribing cephalosporin to a patient with known penicillin allergy and prior reaction to cephalosporin, as well as defendants’ failure to review its own records, which showed that she had been in the same ED before with an allergic reaction to cephalosporin.

The plaintiff’s expert testified, "She was allergic to Keflex, and she never should have got it. She had been in that same ED before with an allergic reaction to Keflex. There is just no excuse for giving a patient in the same ED of the same hospital the same drug that made her allergic before, because there is a danger of almost killing her and it almost did." The Supreme Court of Kentucky held that from the testimony given and the circumstances surrounding the onset of anaphylactic shock in the patient, a jury could reasonably have found that defendants’ negligence was a proximate cause of the condition.

Frasier v. Department of Health and Human Resources, State of Louisiana.10 In this case, Rosemary Frasier brought an action against the State of Louisiana for alleged negligence associated with her care and treatment at a state hospital. The appellate court found that the defendant’s treatment and administration of medications was "not reasonable, prudent, and skillful . . . (in that) defendant administered Thorazine to plaintiff . . . which resulted in an allergic reaction to the drug. Yet (the following year), plaintiff was again administered Thorazine in complete disregard to her noted allergy to the drug, resulting in another allergic reaction. We find that defendant failed to use reasonable care and diligence in this particular instance and fell below the standard of care required of a physician."11

Discussion. Plaintiffs will have little difficulty proving negligence when health care providers administer drugs to which patients have histories of allergies, especially when the health care institution’s records document such allergies. It will be no defense that the institution’s medical record was incomplete or not available to the prescribing provider at the time the second administration was ordered, nor that the patient failed to give a history of allergy.

Sedatives/Anesthetics/Narcotics

Robinson v. University Medical Center.12 A 41-year-old woman was seen at defendant’s ED suffering from an asthma attack. She was first treated with oxygen and bronchodilators, but when her condition continued to deteriorate, an attending physician attempted to intubate her. When this proved difficult, apparently because she was agitated, one of the physicians administered 5 mg Versed (midazolam). Shortly thereafter, she suffered a fatal cardiac arrest. The patient’s survivors alleged the patient was given twice the amount of midazolam recommended by the manufacturer for conscious sedation. The plaintiffs also accused the staff of failing to monitor the patient’s condition properly. The defense claimed that the patient’s asthma attack and ultimate death were caused by her failure to take her asthma medication on a regular basis. The jury returned a verdict of $2.74 million, with a finding of 25% fault on the part of the patient.

Kilian v. Montgomery Hosp.13 A patient was brought to the ED with injuries suffered in a truck accident. He was intoxicated and became belligerent, and the ED physician ordered an injection of midazolam. The nurse who gave the injection later testified she administered 10 mL of a 1 mg/mL solution; however, the patient’s hospital bill suggested that he might have been given 5 mL of a 5 mg/mL solution. Shortly after receiving the injection, the patient suffered a cardiopulmonary arrest. Attempts to resuscitate the patient were unsuccessful, partly because intubation was delayed due to aspiration of vomitus. As a result of lack of oxygen, the patient suffered brain damage and remained in a persistent vegetative state. Plaintiffs’ main charge was the overdose of midazolam, which they claimed was 10 times the proper amount. The ED physician contended that he had ordered the correct dosage. The jury returned a verdict against the ED physician and the hospital for $2.8 million.

Discussion. Trauma patients such as the patient in this case are at increased risk for ADEs. A study of 4320 trauma patients seen in one hospital during a three-year period revealed that the rate of ADEs was twice that of nontrauma patients. Female trauma patients had ADEs 1.5 times more than males (2.7% vs 1.8%, p = 0.052). Analgesics was the medication class most often associated with ADEs, with 54% of events involving morphine and 20% involving meperidine. The most common ADEs were nausea, vomiting, and itching. The authors concluded that analgesics particularly are associated with ADEs in trauma patients, and their use should be carefully monitored.14

Sedatives, anesthetics, and narcotics frequently are administered in the ED for a number of indications. Many of these drugs have major toxicities, including respiratory depression and hypotension. Care must be taken to correctly order and administer such drugs, as well as to monitor patients for toxicity. Patients at high risk for ADEs from sedatives, anesthetics, and narcotics include patients with chronic obstructive pulmonary disease and asthma; those with hepatic and/or renal insufficiency; children; and the elderly. In addition, sedatives such as midazolam or other benzodiazepines should be administered with caution in patients with alcohol intoxication or other substance abuse. These classes of drugs often display additive effects and have narrow therapeutic windows. The dosage, dosage schedule, and risk of toxicity with repeated dosing should be recognized. Idiosyncratic reactions also occur and should be anticipated. Finally, all patients receiving sedatives, anesthetics, and narcotics should be monitored, ideally with pulse oximetry.

A 2-month-old infant suffered an iatrogenic cardiac arrest when a consulting resident administered too much MPC (meperidine, promethazine, and chlorpromazine) — 10 times the expected dose — by the wrong route (intravenously instead of intramuscularly). Although the child was resuscitated with no apparent neurological deficit, the institution where the incident occurred removed MPC entirely from its ED and instituted a policy restricting ED procedural sedation privileges to emergency physicians.15

Of 1180 pediatric patients who underwent procedural sedation and analgesia in the ED of a large urban pediatric teaching hospital and continually were monitored, 27 (2.3%) experienced adverse events, which included oxygen desaturation less than 90% requiring intervention (10 patients [supplemental oxygen (9), bag-mask ventilation (1)]); paradoxical reactions (7); emesis (3); paradoxical reaction and oxygen desaturation requiring supplemental oxygen (2); apnea requiring bag-mask ventilation (1); laryngospasm requiring bag-mask ventilation (1); bradycardia (1); stridor and emesis (1); and oxygen desaturation requiring bag-mask ventilation with subsequent emesis (1). There was no statistically significant difference found in mean doses for all procedural sedation and analgesia medication regimens between those children who experienced adverse events and those who did not. No single drug or drug regimen was associated with a higher adverse event rate, nor between males and females, among different ages, or among the different indications for procedural sedation and analgesia. Despite the finding that no patient in this series required reversal of sedation with naloxone, flumazenil, endotracheal intubation, or hospital admission because of complications from procedural sedation and analgesia,16 the inability to predict which regimen or age group was most at risk for an adverse drug-related event indicates that all pediatric patients are at risk. Although the risk is small, monitoring is required.

Drug Interactions

In Eiss v. Lillis,17 a 46-year-old male patient was placed on warfarin after a mild heart attack, and prothrombin time testing was done during his 12-day hospitalization. Upon returning to the hospital’s cardiac treatment center two weeks after discharge, he complained of pain in one of his legs. A physician at the center told him to take aspirin. During the next several days, the patient apparently took at least 12 aspirin. He began feeling "weak" and developed some bleeding from his gums and a "horrible headache." Upon returning to the hospital, he seemed a "little bit disoriented." His prothrombin time was excessively high. He was treated with vitamin K and admitted, but no repeat prothrombin time test was done. On the second hospital day, he went into a coma and died from intracranial bleeding. In a lawsuit against the treating physician, the expert witness testified that the defendant physician was negligent in not having prothrombin time testing performed on the patient following his first hospitalization and in failing to treat the patient aggressively to counteract the bleeding on his readmission, especially in view of his recent aspirin intake. In the expert’s opinion, the patient should have been put into an intensive care unit, given a blood transfusion, and monitored more closely. He further opined that if the patient had been managed appropriately, he would have survived.

Discussion. Drug interactions are a common cause of morbidity and mortality; warfarin, in particular, interacts with multiple classes of drugs, including aspirin, other nonsteroidal anti-inflammatory drugs, and antibiotics. Treatment of patients by multiple providers at multiple sites of care compounds the risk of such interactions. Optimal care requires a thorough drug history, review of medication records, examination of medication lists or pill bottles in possession of the patient, and a check for MedicAlert bracelets. Not only must the ED provider be vigilant for drug interactions in patients seeking care in the ED, the provider must be careful not to cause adverse drug interactions by prescribing a new medication that will interact with one the patient already is taking.

In several studies of visits to hospital EDs, 47-70% of visits led to added medication. In visits in which at least one medication was added, a new medication added a potential adverse interaction in 5-47% of patients. Additionally, drug-disease interactions were found in one study in 21% of patients, with 34% of such interactions being related to ED treatment.18,19

The risk of an adverse drug interaction has been estimated to rise from 13% for patients taking two medications to 82% for patients taking seven or more medications. ED patients taking three or more medications and patients older than 50 years of age taking two or more medications are at substantial risk for adverse drug-drug and drug-disease interactions.20

Patients do not always remember their medications. In one study of patients 65 years old and older, a complete medication history was given by 50.4% of patients, with the number of medications the patient "forgot" ranging from zero to nine. There was an upward trend that correlated with an increased chance for an incomplete medication history as the total number of current medications increased.21

Additional medications prescribed in the ED that accounted for most of the added interactions were theophylline, macrolid antibiotics, digitalis glycosides, NSAIDs, angiotensin converting enzyme inhibitors, and calcium antagonists. Potential adverse drug interactions are more common in elderly patients because of the higher number of concurrent medications rather than age-based factors; safeguards need to be introduced to prevent patients from receiving medications in the ED that have the potential to cause adverse interactions.22

Wrong Drug/Dosage/Route/Patient

Errors involving the wrong drug or patient represented 4% of all medication errors among hospitalized medical patients.23

Wrong drug, wrong dosage — Caron v. United States of America.24 Plaintiff Annette R. Caron took her healthy and normal 4-month-old baby girl, Monique, to the Custer Air Force Base dispensary for routine immunizations.

Without taking any history or making any inquiries, an airman administered to Monique diphtheria-pertussis-tetanus (DPT), oral polio, and typhoid vaccines. Approximately one hour later, Monique’s left arm started jerking, twitching, and she was "burning with fever" the mother immediately returned her to the dispensary.

The dispensary record reads in pertinent part: "Patient convulsed in grand mal fashion for almost 45 minutes; exam essentially negative . . . presumed this is a febrile convulsion secondary to typhoid inoculation."

The child developed permanent mental retardation of marked severity, resulting in the mental age of 4½ years at the age of 12. It was thereafter determined that the infant had been given an adult dose of typhoid vaccine.

The District Court for the District of Rhode Island (410 F.Supp. 378 [1976]) entered judgment in favor of the child and parents, holding that the government was to be held liable for gross negligence in respect to permanent brain damage sustained by the child as a result of a "sublethal" 0.5 mL injection of typhoid vaccine; that a history and complete physical were required in any case; and that an injection, if given, was to be given in a proper amount and then only if the area was endemic to typhoid.

The government appealed. The Court of Appeals held that the action was timely brought within two years after the parents discovered the alleged malpractice; that evidence sustained determination that administration of DPT shot, oral polio, and typhoid shot in an adult dose to a 4-month-old child was negligence, and that the monetary award for pediatric nursing home care for the child, loss of earnings, and pain and suffering were not excessive. Several expert witnesses stated that there was no indication for a typhoid inoculation, that the 0.5 mL dose given was the dosage recommended for an adult and was five times the dosage for a 4-month-old child, and that the combination of DPT and typhoid as administered to Monique was the precipitating cause of her convulsions, with complicating sequelae of brain damage, mental retardation, or hyperactive behavior to follow.

Wrong dosage — Bove v. Ross.25 A 43-year-old man was undergoing a thallium stress test in a cardiologist’s office. During the test, he developed ventricular tachycardia, and the cardiologist requested a technician to administer lidocaine from the office "crash cart." The container of lidocaine was marked "2 grams," but also was marked "for use in dilution only: not for bolus injection." The technician administered the injection, but the patient died before they could get him to the hospital. An autopsy revealed toxic levels of lidocaine in the patient’s brain; it was estimated that he had received 20 times the normal dosage. The patient’s wife and children were awarded $1.03 million.

In Hart v. Landers, 26 a similar ADE occurred in a patient given 1000 mg lidocaine from a prepackaged syringe that was supposed to be diluted in IV fluid for drip administration. The patient suffered a cardiac arrest and died several hours later. Unfortunately, the hospital stored the two forms of dosages next to each other. The parties entered into a structured settlement worth $1.1 million.

Again, in Feldburg v. Sun Towers Hosp.,27 the patient was taken to the Sun Towers Hospital ED suffering from chest pains. An ED physician performed an electrocardiogram and ordered 100 mg lidocaine hydrochloride intravenously. The ED nurse, however, mistakenly administered 20 times the prescribed dosage. The patient immediately went into convulsions and died. The plaintiffs settled with the hospital’s insurance company for $2.02 million.

Discussion. These cases illustrate the risks involved in storage of high-dose and undiluted medications in the ED. Drugs intended for dilution before administration are best prepared by pharmacists. Where nonpharmacist ED personnel are required to prepare and/or administer drugs that are not in single-use containers, labels must be carefully checked and calculations must be checked and rechecked, ideally with a second practitioner. Lookalike and different dosage forms of the same drugs should not be stored in the same area.

Wrong route — Fleming v. Baptist Gen. Convention of Okla.28 The plaintiff had been suffering from chronic back pain and migraine headaches for many years, and for three years had been receiving intramuscular injections of a combined prescription of Atarax (hydroxyzine HCl) and Talwin (pentazocine) at least twice weekly.

On the occasion in question, she received an injection as an ED patient in the defendant’s hospital. The nurse mistakenly gave the injection subcutaneously rather than intramuscularly. The patient experienced immediate pain, and by the time she had returned home, the skin at the injection site in her left thigh had become discolored. The skin and subcutaneous tissues eroded and eventually, gangrene set in. The plaintiff required two skin grafts to close the wound. She brought suit against the hospital under the theory that the injection that precipitated the injury was not given deep into the muscle as it was intended to be but into the subcutaneous tissues.

On the basis of plaintiff’s expert’s opinion that 1) the injection must have been given subcutaneously because there was no evidence at any time later of any disruption or slough of the deeper muscle or the covering of the muscle (the fascia), nor did it slough out, die, or become involved; and that 2) it was not proper, medically accepted practice to inject Atarax subcutaneously, and despite defendant’s defense that patient should have known the injection was being administered improperly yet did nothing about it, the jury returned a verdict for the plaintiff and the appellate court affirmed.

One easily can imagine a verbal order being given in a busy ED for "a shot of" or "injection of" a drug wherein the prescriber does not specify the exact route for administration. These cases stand for the responsibility of nursing staff to possess the requisite knowledge regarding appropriate routes of injection for various drugs.

Wrong patient — Waliczek v. Gutta, et al.29 A 46-year-old man was admitted through the ED at Alexian Brothers Medical Center after having fallen approximately 30 feet while working at a construction site. He was diagnosed in the ED as having suffered multiple fractures, intra-abdominal bleeding, and a small bleed or bruise on the right side of the brain, as determined by computed tomography scan. The patient was placed in a room in the ED where a patient with a cardiac condition had been treated. A bag containing heparin had been brought into the room for the cardiac patient, and hospital personnel had failed to remove the bag containing the heparin from the room. The trauma patient received approximately 350 cc heparin, which extenuated his bleeding. He died two days later at Loyola University Medical Center, where he was transferred after it was determined that he had suffered massive brain hemorrhaging. The defendants agreed to settle the case prior to trial for $6.5 million.

Discussion. When a patient receives a drug intended for another patient, in actuality two errors occur — one, the error of commission (for the patient who receives a drug not ordered/indicated for him or her) and the other, an error of omission (for the patient who fails to receive the ordered medication).

Wrong-patient errors may occur at several stages:

1. Ordering (as when the order is written in the wrong chart or entered into the wrong computerized record).

2. Transcription (as when order is written in the correct chart, but order sheets have the wrong name stamp — or when a new order sheet is used with the patient’s name handwritten; or when the order is correcting written but transcribed to wrong medication administration record by any variety of reasons).

3. Dispensing (as when the pharmacist is distracted during order entry by phone call about another patient; resumes order entry with the wrong computer record open, so that medication becomes part of the pharmacy order record for the wrong patient).

4. Administration (as when a nurse helps out a colleague: "Mr. Smith needs his Haldol; can you give it to him, while I take care of Ms. Jones?").30

The extent of true errors in drug therapy has been a revelation both to health care providers and to the public. Patients rely on providers to correctly prescribe and administer medications. The ordering or administration of the wrong drug, at the wrong dosage, via the wrong route, or to the wrong patient are errors for which there is no justification and no defense.

In a study of 36 institutions, 19% of medication doses (605/3216) were found to be in error. The most frequent errors by category were wrong time (43%), omission (30%), wrong dose (17%), and unauthorized drug (4%). Seven percent of the errors were judged potential ADEs.31

In a study done at a tertiary care hospital, 200 cases of tenfold dosage prescribing errors were detected during an 18-month period. Overdoses were prescribed in 61% of the cases and underdoses in 39% of the cases. Errors were associated with multiple zeros in the dose (45%); use of equations or calculations to determine dose (27% total cases, 92.3% of pediatric cases); and dose amount less than 1 (25%). Tenfold error was produced by a misplaced decimal point in 87 cases (43.5%), adding an extra zero in 63 cases (31.5%), and omitting a zero in 50 cases (25%). Factors identified as enabling a tenfold error to be carried out as (erroneously) ordered were a wide dose range for the drug (76.5%); medication ordered and able to be given by injection (42%); ability to give ordered dose as less than or equal to 5; solid oral dosage forms (36%); and availability of an oral liquid dose form (15%).32

Pediatric patients may be at increased risk for medication errors, often due to incorrect recording of patient weight leading to incorrect medication dose. The most common error was an incorrect dose of medication (35%) or incorrect medication given (30%). In one study, 12% of patients who experienced drug errors required additional treatment, and one patient was admitted to the hospital because of the error. Interestingly, in one-third of the cases, the family was not notified.33 Variables associated with an increased proportion of errors include patients seen between 4 a.m. and 8 a.m.; patients with severe disease; medication ordered by a trainee; and patients seen during weekends.34

One of the major causes of medication errors results from the use of potentially dangerous abbreviations and or symbols. Frequently used to save time when writing prescriptions and documenting in patient records, some abbreviations are especially problematic. Examples include "U" for units and "ug" for micrograms. Writing "U" can give the appearance of a zero. Using the abbreviation of "ug" rather than "mcg" also has been a source of error when the handwritten symbol "u" resembles the letter "m." Another frequent and dangerous occurrence is the inability to clearly distinguish a decimal point. The decimal point sometimes is not visible when orders are handwritten using trailing zeros or no leading zeros. Use of computerized order entry systems can reduce errors associated with illegible handwriting. Realizing that many health care organizations are unable to afford new technologies, other risk reduction strategies are available. Suggestions include: 1) develop a list of unacceptable abbreviations and symbols that is shared with all prescribers; 2) develop a policy to ensure that medical staff refer to the list, and take steps to ensure compliance; and 3) establish a policy that if an unacceptable abbreviation or symbol is used, the prescription order is verified with the prescriber prior to its being filled.35

The emergency physician, generally, is not liable for negligent acts of a nurse or other staff person in the ED performing a task pursuant to the physician’s orders, such as giving an injection, unless the nurse is acting under the physician’s direct supervision. Liability may attach, on the other hand, if the physician is actually directing the conduct of the nurse in a given procedure or, for example, where the emergency services corporation acting as independent contractor for the hospital has the right to hire and fire ED nurses and/or pays their wages.36 However, the hospital and ED physician often are viewed by the courts as jointly liable in cases of negligence involving the physician.37 The ostensible agency doctrine (also known as apparent authority) refers to the accountability of a principal (hospital) for the acts of his agent (physician). Irrespective of the contractual arrangement between the hospital and the physician (either true employees of the hospital or independent contractors), courts have used ostensible agency doctrine to find hospitals vicariously liable for negligent acts of independent ED physicians as well as actual employees.38

Avoiding Liability

Case reports of ADEs published in Clin-Alert during 1976-1995 were reviewed for fatal ADEs. Four hundred forty-seven fatal events were found; 10% were assessed as definite, 46% as probable, and 44% as possible. Forty-five percent of the patients were 40-69 years old, and 40% were healthy. Central nervous system agents, antineoplastics, antimicrobials, and cardiovascular agents accounted for 69% of the deaths. Sixty-eight percent of the deaths were judged to have been preventable. Litigation was reported in 14% of the fatal ADE cases; judgments and settlements averaged $1.1 million.39

Of all significant ADEs reported to Clin-Alert during 1976-1995 (1520 cases), faulty prescribing was the most common reason for medication error, and wrong dosage was the most common type of error. Overall, 52% of the errors were judged to have been prevent-able, and of these, a pharmacist could have prevented half. A lawsuit was reported in 13% of cases, and settlements and judgments averaged $3.1 million.40

A retrospective analysis of a New England malpractice insurance company’s claim records from 1990 to 1999 revealed that ADEs represented 6.3% of claims. ADEs were judged preventable in 73% of the cases and were nearly evenly divided between outpatient and inpatient settings. The most frequently involved classes were antibiotics, antidepressants or antipsychotics, cardiovascular drugs, and anticoagulants. Among these ADEs, 46% were life-threatening or fatal. System deficiencies and performance errors were the most frequent cause of preventable ADEs.41

What steps can be taken by EDs to avoid errors and subsequent liability due to ADEs?

1. Computerized order-entry systems (CPOES) — Medication error, the No. 1 source, is perhaps the area in which technology can have the greatest benefit. CPOES combine the benefits of computerized order entry by physicians with real-time clinical decision support. Computer physician order entry systems have the potential to address problems of ordering errors such as incorrect choice of medication for the patient’s condition, an incorrect dose of the medication given the patient’s physiologic state (for example, renal insufficiency or advanced age), and prescribing a medication to which the patient was known to be allergic. CPOES include patient-specific dosing suggestions, reminders to monitor drug levels, reminders to choose an appropriate drug, checking for drug-allergy and drug-drug interactions, standardized order sets, increased legibility, automated communication to ancillary departments, and easy access to patient data and reference information when ordering.42

• A major reduction in errors was achieved with a computerized physician order entry with decision support system that included drug allergy and drug-drug interaction warnings. Nonmissed-dose medication error rate fell 81%, nonintercepted serious medication errors (those with potential to cause injury) fell 86%. Large differences were seen for all main types of medication errors: dose errors, frequency errors, route errors, substitution errors, and allergies.43

• A pre-/post-retrospective analysis compares errors requiring pharmacist clarification between handwritten and computer-assisted ED prescriptions. During the pre-intervention period, there were 91 clarifications required for 2326 ED prescriptions, with a clarification rate of 3.9%. There were 54 handwritten errors (errors defined as missing information, incorrect information, incorrect dose, nonformulary medication, or illegibility being the reason for clarification), for an error rate of 2.3%. During the post-intervention period, there were 13 clarifications, with a clarification rate of 0.8% and 11 errors, for a computer-assisted error rate of 0.7%. The authors concluded that computer-assisted prescriptions were substantially less likely to contain an error or to require pharmacist clarification than were handwritten prescriptions. Computer-assisted prescriptions were more than three times less likely to contain errors and five times less likely to require pharmacist clarification than handwritten prescriptions.44

2. Electronic medical record— In many cases, important historical information is unavailable to the treating physician in the ED. Patients often forget key details of their medical history or are too ill or injured to provide any information at all. Furthermore, either the patient’s condition is so acute as to require action before data are available or the time lost waiting for the data outweighs the disadvantages of repeating tests or gathering information again. Sharing information between hospitals, private offices, and clinics is often problematic. From a patient safety perspective, every roadblock to critical data access represents an opportunity for error and patient harm.45 A centralized medical record goes a long way toward eliminating these problems.

3. ED pharmacist — Bar coding and scanning/ matching of patient and medication prior to administration work well when drugs are prepared in the pharmacy or under controlled circumstances by a nurse at a nursing station where labels can be printed and applied to the medication. However, this system tends to be especially cumbersome when patients are critically ill (as often is the case in the ED). Alternative approaches during resuscitation include stocking crash carts with pre-mixed, standardized unit doses of resuscitation medication and placement and use of dosage calculators on the carts. Ideally, these units would be interfaced with the master pharmacy system to look for allergies and potential drug interactions.46 There was a 45% decrease in medication errors (1.81-fold decrease) in hospitals that had decentralized pharmacists, compared with hospitals that had centralized pharmacist as well as a 94% decrease in medication errors that adversely affected patient care outcomes (16.88-fold decrease).47

4. Legible orders/verbal order issue — Illegible handwriting is an obvious cause of medication errors. A British study of doctors’ handwriting legibility revealed that while numeric legibility was the same between physicians, nurses, and other professionals and administrators, there was a significant difference between the groups for letters, with doctors making significantly more errors, irrespective of gender.48 While CPOES clearly would eliminate this source of error, many hospitals and EDs still do not have them.

Efforts to encourage prescribing practitioners to print medication orders have not been met with success. Verbal orders, common in an ED, rely on several steps — the prescriber must state the correct drug and dosage and route without the memory assist or seeing it before his or her eyes in writing and the administering nurse must receive the information without benefit of seeing the drug prescription in writing. Distractions, so common in EDs may interrupt this flow of information and lead to potentially lethal errors.

5. Nurses or other practitioners should utilize a double-check system for correct drug/dosage/route/ patient ID. Medication errors often are caused by unclear labeling, lack of a check system for medication dosages, and similar drugs kept in close proximity to each other. Recommended strategies include a check system in which one nurse prepares a dose and another nurse reviews it, and avoiding lookalike and soundalike medications.49

6. Monitoring for patients receiving sedatives — Monitoring with alarms for out-of-range pulse rates or oxygen saturations clearly is helpful and should be the standard of care in EDs.

7. Prescribers should recheck dosage calculations. Electronic prescribing systems provide easy-to-use calculation tables. Prescribers also should try to calculate and write prescriptions in a setting where interruptions are less likely, hard as this might be in a busy ED.

8. EDs should institute procedures to ensure identity verification of patients. Patients with similar-sounding names present challenges to the best-functioning health care systems. Some automated verification systems (e.g., bar-coding technology) may help to reduce the likelihood of misidentification. But technology still requires a protocol to be effective; a particular team member or members must be charged with matching the bar code on the patient’s identity bracelet to the bar code on the medication.50

9. On-line resources should be available in all EDs and offer a wealth of information useful to practitioners. In addition to medical and pharmacy databases, there are a number of systems useful for analyzing and tracking medication errors:

www.hospitalconnect.com — A set of web-based educational tools designed to reduce medication errors. Developed by the American Hospital Association, the Health Research and Educational Trust, and the Institute for Safe Medication Practices (ISMP), with support from the Commonwealth Fund.

Medication Errors Reporting (MER) Program — A nationwide program that makes it possible for health care professionals who encounter actual or potential medication errors to report confidentially to the United States Pharmacoepia (USP). USP reviews each report for health hazards and forwards all information to the FDA and the product manufacturer. The program is presented in cooperation with ISMP. Information is available at www.usp.org/patientSafety/reporting/mer.html.

Medmarx is an Internet-accessible, standardized program used by hospitals nationwide to report and track medication errors. Data from participating hospitals are stored in a central database at USP for educational and research purposes. Each participating facility has the ability to access its own Medmarx data. Information is available at www.usp.org.

ISMP, www.ismp.org.

CDERCenter for Drug Evaluation and Research of the U.S. Food and Drug Administration (FDA). FDA monitors medication error reports that are forwarded to it from USP and ISMP. The agency also reviews MedWatch reports for possible medication errors. www.fda.gov/cder/drug/MedErrors/default.htm.

Conclusion

New patient safety standards that went into effect July 1, 2001, require hospitals to initiate specific efforts to prevent medical errors and to tell patients when they have been harmed during their treatment from a medical error. A major focus of the new standards is on the prevention of medical errors through the prospective analysis and redesign of vulnerable patient care systems (e.g., the ordering, preparation, and dispensing of medications).

As a result of the uncontrollable patient volume, high acuity and complexity of illness and injury, and frequent lack of longitudinal patient data, emergency medical care presents unique patient safety challenges. Recognition and prevention of ADEs is one of those challenges. To avoid liability and provide the best of patient care, the staff must adhere to good professional practices in the ED’s sometimes hectic environment. Innovations such as CPOES and electronic medical record systems will be of assistance in avoidance of ADEs.

Endnotes

1. Kohn, LT, Corrigan JM, Donaldson MS, eds. To Err is Human: Building a Safer Health System. Washington, DC: Institute of Medicine, National Academy Press; 2000, p. 28.

2. Id., at 32.

3. Id.

4. Id.

5. See Institute of Medicine web site, www.iom.edu.

6. Lesar TS, Briceland L, Stein DS. Factors related to errors in medication prescribing. JAMA 1997;277:312-317.

7. Reports spotlight medication errors: Make changes before tragedy strikes. ED Manag 2000;12:61-66, suppl 1-2.

8. Thomas EJ, Studdeert DM, Newhouse JP, et al., Costs of Medical Injuries in Utah and Colorado. Inquiry 1999;36: 255-264.

9. Elizabeth Baylis, Appellant v. Lourdes Hospital Inc., William Wheeler, MD, and Credit Bureau of Paducah Inc. Appellees. Ky 805 S.W.2d 122-126 (Supreme Court of Kentucky, 1991).

10. Rosemary Frasier v. Department of Health and Human Resources, State of Louisiana, et al., 500 So. 2d 858 (Court of Appeal of Louisiana, First Circuit, 1986).

11. Id. at 863.

12. Robinson v. University Med. Center, No. 89-05290-CA (Circuit Court, Duval Cty, Fla., Feb 28, 1991 (reported by ATLA professional negligence law reporter 6:129, 1991).

13. Kilian v. Montgomery Hosp., No. 91-CV-5890 (U.S. Dist. Ct., E.D. Pa., Oct 1, 1992) (reported by Shepherd’s/McGraw-Hill, Verdicts, Settlements & Tactics; LEXIS, Verdicts library, Shepjv file).

14. Lazarus HM, Fox J, Evans RS, et al. Adverse drug events in trauma patients. J Trauma 2003;54:337-343.

15. Brown ED, Corbett SW, Green SM. Iatrogenic cardiopulmonary arrest during pediatric sedation with meperidine, promethazine, and chlorpromazine. Pediatr Emerg Care 2001;17:351-353.

16. Pena BM, Krauss B. Adverse events of procedural sedation and analgesia in a pediatric emergency department. Ann Emerg Med 1999;34:483-491.

17. Eiss v. Lillis, 357 S.E.2d 539 (Supreme Court of Virginia, 1987).

18. Beers MH, Storrie M, Lee G. Potential adverse drug interactions in the emergency room. An issue in the quality of care. Ann Intern Med 1990;112:61-64.

19. Goldberg, RM, Mabee J, Chan L, et al. Drug-drug and drug-disease interactions in the ED: Analysis of high-risk population. Am J Emerg Med 1996;14:447-450.

20. Gaddis GM, Holt TR, Woods M. Drug interactions in at-risk emergency department patients. Acad Emerg Med 2002;9: 1162-1167.

21. Hancock D, Kennington JM, Beckner RR, Quick G. Emergency department medication and drug interaction evaluation. Hosp Pharm 1992;27:129-132.

22. Heininger-Rothbucher D, Bischinger S, Ulmer H, et al. Incidence and risk of potential adverse drug interactions in the emergency room. Resuscitation 2001;49:282-288.

23. Bates DW, Boyle DL, Vander Vliet MB, et al. Relationship between medication errors and adverse drug events. J Gen Intern Med 1995;10:1999-2005.

24. Monique G. Caron, et al., plaintiffs-appellees v. United States of America, defendant-appellant, 548 F. 2d 366 (U.S. Court of Appeals, First Circuit, 1976).

25. Bove v. Ross, No. 91-429, Circuit Ct., Arundel Cty., Md., July 17, 1993 (reported by Shepard’s/McGraw-Hill, Verdicts, Settlements & Tactics; LEXIS verdicts library, Shepjv file; Metro Verdicts Monthly; LEXIS verdicts library, Dcjury file).

26. Hart v. Landers. No. 87-15921-H, 160th Judicial Dist.Court, Dallas Cty., Tex., May 25, 1988 (reported by Professional Negligence Law Reporter 4:10, 1989).

27. Feldburg v. Sun Towers Hosp., No. 87-8421, 65th Judicial District Court, El Paso Cty., Tex., Mar. 1, 1989.

28. Fleming v. Baptist Gen. Convention of Okla., 742 P.2d 1087 (Okla., 1987).

29. Zofia Waliczek, Special Administrator of the Estate of Stanislaw Waliczek v. Gandhi Gutta, MD, et al. Case no. 97L 8110; Cook County, Ill. Judge Richard Elrod, 6-28-01 (reported by Zarin’s Medical Liability Alert 2001;10:1. Springfield, NJ: Zarin’s Professional Liability Publications, Jury Verdict Review Publications Inc.).

30. Archived Cases & Commentaries: Medicine. Agency for Healthcare Research & Quality, www.ahcpr.gov/, accessed 7/1/2003.

31. Barker KN, Flynn EA, Pepper GA, et al. Medication errors observed in 36 health care facilities. Arch Intern Med 2002; 162:1897-1903.

32. Lesar TS, Tenfold medication dose prescribing errors. Ann Pharmacother 2002;36:1833-1839.

33. Slebst SM, Fein, JA, Osterhoudt K, Ho W. Medication errors in a pediatric emergency department. Pediatr Emerg Care 1999;15:1-4.

34. Kozer E, Scolnik D, Macpherson A, et al. Variables associated with medication errors in pediatric emergency medicine. Pediatrics 2002;110:737-742.

35. See JCAHO Sentinel Event Alert, issue 23, September 2001.

36. Rothenberg MA. Emergency Department Standards of Care: A Lawyer’s Guide. Professional Education Systems Inc.; 1989, p. 3).

37. Id.

38. Id.

39. Kelly WN. Potential risks and prevention, Part 1: Fatal adverse drug events. Am J Health Syst Pharm 2001;58:1317-1324.

40. Kelly WN. Potential risks and prevention, Part 4: Reports of significant adverse drug events. Am J Health Syst Pharm 2001;58:1406-1412.

41. Rothschild JM, Federico FA, Gandhi TK, et al. Analysis of medication-related malpractice claims: Causes, preventability, and costs. Arch Intern Med 2002;162:2414-2420.

42. Kuperman GJ, Gibson RF. Computer physician order entry: Benefits, costs, and issues. Ann Int Med 2003;139:31-39, at 34.

43. Bates DW, Teich JM, Lee J, et al. the impact of computerized physician order entry on medication error prevention. J Am Med Inform Assoc 1999;6:313-321.

44. Bizovi KE, Beckley BE, McDade MC, et al. The effect of computer-assisted prescription writing on emergency department prescription errors. Acad Emerg Med 2002;9:1168-1175.

45. Patient Safety in the Emergency Department Environment; Patient Safety Task Force — 2001, accessed on ACEP web site, p. 30, at www.acep.org.

46. Id. at 29.

47. Bond CA, Raehl CL, Franke T. Medication errors in United States hospitals. Pharmacotherapy 2001;21:1023-1036.

48. Lyons R, Payne C, McCabe M, et al. Legibility of doctors’ handwriting: Quantitative comparative study. Br Med J 1998;317:863-864.

49. ED Manag 2000;12:61-66, suppl 1-2.

50. Chassin, MR, Becher EC. The wrong patient. Ann Intern Med 2002;136:826-833 at 830.