Drug-Related Visits to the Emergency Department: How Big Is the Problem?

Payal Patel, Pharm.D., Peter J. Zed, Pharm.D.


Pharmacotherapy. 2002;22(7) 

In This Article


We identified 22 articles that, before review, appeared to meet our inclusion criteria. Four articles were excluded because they described specific patient populations and their risk factors for hospital admission.[8,9,10,11] Another study was excluded because it described drug interactions caused by drugs prescribed in the emergency department.[12] Four articles were excluded because they described the frequency of drug-related hospital admissions rather than emergency department visits.[6,7,13,14] A multidisciplinary conference addressing drug misadventures was excluded because it was not a retrospective or prospective study.[15] Thus, 12 articles met our inclusion criteria: eight retrospective and four prospective studies.[16,17,18,19,20,21,22,23,24,25,26,27]

A retrospective chart review of patients who visited the emergency department of a 517-bed tertiary care facility over a 4-month period evaluated drug-related visits and determined the frequency of these visits, rate of admission for patients with drug-related illness, classes of drugs most often associated with drug-related visits and admissions, demographics of patients whose visits were drug related, causes of drug-related visits and admissions, average length of stay, and cost implications of drug-related admissions.[16] Drug-related illnesses were classified into five categories: adverse drug reaction, intentional overdose or abuse of drugs, toxicity, drug interactions, and noncompliance. To standardize the definition of adverse drug reactions, the authors classified them according to the Food and Drug Administration (FDA) nomogram as possible, probable, and highly probable.[28]

During the 4-month study period, 293 (2.9%) of the 10,184 patients who visited the emergency department were classified as having a drug-related illness. Of these patients, 71 (24.2%) required hospital admission for further evaluation. Average age of those whose visits were drug related was 42.4 years; distribution of men and women was equal. Average length of hospital stay was 5.8 days; average cost was $8888/admission.

Drug classes most often implicated were drugs of abuse (23.2%), anticonvulsants (17.1%), antibiotics (12.6%), respiratory agents (8.9%), and analgesics (8.9%). The most common category of drug-related illness was overdose or abuse (35.5%), followed by noncompliance (28.3%), adverse drug reactions (27.6%), toxicity (7.8%), and drug interactions (0.3%). Of the adverse drug reactions, 40% were classified as probable according to the FDA nomogram; the remaining 60% were deemed possible.

A retrospective chart review of patients admitted to the emergency department of a 238-bed teaching hospital over a 6-month period established the frequency, type, and severity of adverse drug reactions managed in the emergency department.[17] If the recorded diagnosis was an adverse drug reaction, further information was obtained to establish the type and severity of the reaction. Intentional or accidental overdose, poisoning, substance abuse, and insulin reaction were excluded. Reactions were classified as mild (no treatment required), moderate (treatment required for symptom resolution), and severe (hospital admission required). In addition, reactions were classified as type A (dose dependent, predictable, usually an extension of the drug's pharmacology, rarely life threatening) or type B (allergic or idiosyncratic, not an extension of the drug's pharmacology, possibly life threatening).

Review of 13,703 emergency department records revealed that 118 (0.86%) patients had experienced adverse drug reactions of which 58 (49%) were classified as mild, 46 (39%) as moderate, and 14 (12%) as severe. Of the reactions, 68 (58%) were type A, and 50 (42%) were type B. Of the 118 patients, 81 (69%) were women (mean age of women, 51.6 yrs; of men, 46.3 yrs). Drug classes most often associated with adverse drug reactions were antiinfectives (34.7%), analgesics (11.0%), and central nervous system agents (9.3%). Most adverse reactions by body system were dermatologic (29.4%), gastrointestinal (23.5%), neurologic (16.8%), cardiovascular (7.6%), and respiratory (7.5%). The mean cost/emergency department visit was $333.81 ± $241.96 (US $).

A single-investigator retrospective chart review of patients who visited the emergency department of a 560-bed teaching hospital over a 1-month period determined the frequency of drug-related illnesses and the costs associated with the visits.[18] Drug-related illnesses were either documented by the physician at the time of the emergency department visit or suspected by the investigator during chart review if the recorded symptoms or disease had an identifiable or probable relationship to the patient's documented drug therapy. Drug-related illnesses were evaluated only for prescription or nonprescription drugs, thus illnesses due to alcohol or drug abuse were not covered in this study. Causes of drug-related illnesses were classified as inappropriate prescribing (deviation from the recommended dosage, lack of appropriate drug therapy for a medical condition, or incorrect drug selection given a patient's age, organ function, or concurrent diseases or drug therapy), patient noncompliance (based on drug history at the time of the emergency department visit), adverse drug reaction, or drug interaction. Adverse drug reactions were classified as untoward (an extension of the pharmacologic effect or a direct toxic effect on an organ system), idiosyncratic, or hypersensitivity related.

Review of 1260 charts established that 49 (3.9%) patients had a drug-related illness; eight (16.3%) of these patients required hospital admission. A physician documented the drug-related illness at the time of the emergency department visit for 39 patients; the investigator discovered the illness for the remaining 10 while reviewing the charts. Mean age of all patients with a drug-related illness was 41.7 ± 22.5 years; distribution of men and women was equal. Causes of illness were noncompliance (58%), inappropriate prescribing (32%), and adverse drug reaction (10%). Drug-related illness was considered preventable for 33 (67.3%) patients; average cost was $678/visit (US $)

A single-investigator retrospective chart review of all visits to the emergency department of a health maintenance organization medical center over a 12-month period identified the frequency of factors associated with drug misadventures resulting in emergency department visits.[19] Drug misadventure was defined as an unfavorable effect of drug administration, such as poor compliance, inappropriate self-administration, inappropriate prescribing, or drug interaction. Intentional overdose, substance abuse, and myelosuppression from antineoplastic agents were excluded from this definition.

After the charts were reviewed, any event identified as a likely drug misadventure was evaluated further through telephone interview with each patient, if possible. This interview provided further detail regarding the emergency department visit. In addition, the patient's understanding of the drug regimen was assessed and rated as some, good, or poor, depending on degree of awareness of the drug's purpose, proper administration, possible adverse effects, and possible drug interactions.

Of 62,216 emergency department visits that took place during the 12-month study period, 1074 (1.7%) could be attributed to drug misadventure. Of the patients who experienced a misadventure, 62% were women; 38% were aged 15-44 years, and 33% were 65 years or older. The most common reasons for drug misadventure were undesired adverse effect (43%), drug allergy (19%), and underadministration of drug (12.8%). Drug classes most likely associated with misadventure were antiinfectives (23.6%), analgesics (23.3%), and cardiovascular agents (18.5%).

Telephone interviews were conducted with 962 (89.6%) of the patients who had visited the emergency department because of a drug-related problem. Understanding of the drug's purpose and of proper administration was rated as good in 74.5% and 69.4% of patients, respectively. However, only 30.6% patients had good understanding of the potential adverse effects of their regimen, and 29.0% had good understanding of the possible interactions.

Of the 1074 misadventures, 151 (14.1%) resulted in hospital admissions for further care; these accounted for 1% of all hospital admissions. Likelihood of hospital admission increased with age. Admission rates were 2.1% for patients younger than 14 years, 5.2% for patients aged 15-44 years, 16.4% for patients aged 45-64 years, and 24.4% for those 65 years or older.

A single-investigator retrospective study evaluated possible drug-related visits to emergency departments and subsequent hospital admissions in a 528-bed university-affiliated tertiary care institution over a 1-month period.[20] All emergency department treatment records were reviewed to estimate the frequency of any of the eight categories of drug-related problems described above.[1] Drug-related problems were described as mild (requiring no treatment), moderate (requiring drug therapy), or severe (requiring hospital admission).

Results indicated that 244 (4.2%) of 5757 emergency department visits were drug related. A total of 14.8% of the patients had a severe drug-related problem requiring hospital admission, with a mean stay of 7.1 ± 6.99 days. Drug-related problems were moderate in 65.1% of patients and mild in 20.1%. The most common type of drug-related problem was failure to receive a drug (54%), followed by suboptimal therapy (14%), lack of indication (14%), and adverse drug reaction (11%). Of the 26 adverse drug reactions detected, nine were hypersensitivity related and two were idiosyncratic. The most common drug classes leading to drug-related problems were cardiovascular agents (22%), sympathomimetics (20%), hormones or synthetic substitutes (18%), analgesics (16%), and anticonvulsants (11%).

A retrospective review of drug-related visits to hospital emergency departments over a 1-year period evaluated the rate, nature, demographics, and resource use associated with the visits.[21] Data for this study were obtained from the 1992 National Hospital Ambulatory Medical Care Survey, which was designed by the National Health Center for Health Statistics. Data were compiled from 474 hospitals, of which 437 had an emergency department. Patient diagnoses were coded according to the international classification of diseases.[29] Drug-related injuries were defined as injuries with a principal or first-listed diagnosis recorded as adverse events of drugs. Patients with a secondary or tertiary diagnosis of adverse events of drugs were excluded, as were intentional or accidental overdose, underdose, abuse, or poisoning.

Results indicated that an estimated 367,647 drug-related injury visits were made to hospital emergency departments in the United States. Approximately 1.16% of injury-related visits, or 0.41% of all visits to hospital emergency departments, were due to adverse effects of drugs. This represents a rate of 10.64 drug-related visits/10,000 persons in the United States. Women accounted for approximately 64% of drug-related injury visits; their rate of visits was 18.2/10,000 persons. Patients aged 75 years or older accounted for only about 8% of drug-related emergency department visits; however, their rate of visits, 23.8/10,000 persons, was higher than that of any other age-group. The most cited reasons for emergency department visits were skin rash (13.9%), nausea and vomiting (7.5%), stomach pain (6.2%), and dizziness (4.8%). The most common classes of drugs associated with drug-related visits were antibiotics (15.4%), hormones or synthetic substitutes (7.8%), and psychotropic agents (6.9%).

Another study by the same investigator retrospectively reviewed drug-related visits to hospital outpatient departments over a 2-year period using the 1995-1996 National Hospital Ambulatory Medical Care Surveys database.[22] During that time an estimated 251,017 visits due to drug-related injury were made to hospital outpatient departments in the United States, or 125,508 drug-related visits/year. This corresponds to an overall rate of 4.77 drug-related visits/10,000 persons in the United States. Approximately 1.79% of injury-related visits, or 0.19% of all visits to hospital outpatient departments, were due to adverse effects of drugs. Women accounted for approximately 65% of drug-related injury visits, with a visit rate of 6.09/10,000 women versus 3.38/10,000 men. Patients aged 75 years or older accounted for approximately 9% of drug-related emergency department visits, but their rate of visits, 8.24/10,000 persons, was higher than that of any other age-group. The most frequent reasons for outpatient department visits were adverse effects of drugs (22.9%) and skin rash (18.8%). The most common classes of drugs associated with drug-related visits were antibiotics and antiinfectives (25.5%), hormones and synthetic substitutes (18.8%), and analgesics (9.1%).

The most recent retrospective study evaluated 300 randomly selected patients older than 65 years to determine degree of polypharmacy and frequency of adverse drug-related events resulting in an emergency department visit.[23] An adverse drug-related event was defined as any unfavorable medical event related to drug administration or misadministration, such as adverse drug interaction, drug withdrawal reaction, or an adverse event occurring after prescription errors or noncompliance. Charts were reviewed by a single investigator. All charts indicating an adverse drug reaction were reviewed a second time by a different investigator to ensure accuracy of data collection. A standard algorithm was used to determine cause and effect.[30]

A total of 283 patients were included in the final analysis; 17 were excluded. Of these, 143 (50.7%) were women; average age of all patients was 78.6 ± 8.4 years. Of the 283 visits, 30 (10.6%) were associated with adverse drug reactions; four were definite, 16 probable, and 10 possible. Drug classes most frequently implicated were nonsteroidal antiinflammatory drugs, anticoagulants, diuretics, hypoglycemic agents, -blockers, calcium channel blockers, and chemotherapeutic agents.

One prospective study evaluated prescription noncompliance as a reason for emergency department visits in a random survey of 100 patients from a large urban university hospital.[24] Over a 6-day period, all patients who came to the emergency department were surveyed about the existence of, and reasons for, prescription noncompliance. Noncompliance was considered if the patient had taken no drugs for at least 48 hours before the visit; the drug, when previously taken, had controlled the condition for which the patient visited the emergency department; and no other significant cause or illness was believed to have precipitated the visit.

Results indicated that noncompliance accounted for 22% of emergency department visits; 18% of patients required hospital admission; mean age of patients was 42 years (range 19-68 yrs); and 59% were women. The most common medical conditions were asthma and seizures, and the most common reason for prescription noncompliance was cost; other reasons were inappropriate self-advice and prescription supply depleted. Mean cost/emergency department visit was $576; average cost/hospital admission was $4834 (US $).

A prospective observational study described the contribution of adverse drug events to the overall number of referrals or visits to an emergency department, the proportion of severe events requiring hospital admission, and the causes of the emergency department visits or hospital admissions.[25] Data were collected for 1 week/month over a 1-year period in the emergency department of a 700-bed public hospital in Milan, Italy. Emergency department physicians obtained standard patient information regarding demographics, diagnosis coded according to the International Classification of Diseases, 9th revision,[29] drug history, type of adverse drug event, and clinical condition. A trained nurse reviewed the medical records of all patients the day after each emergency department visit, and events that were possibly related to drug administration were identified.

Hospital admission was considered drug related if some aspect of drug therapy was suspected to be responsible, and if the admission was not likely a result of disease progression. Details of all identified cases were reviewed by at least two investigators in an attempt to classify adverse drug events into any of the following categories: adverse drug reaction -- any noxious, unintended drug reaction that occurs with the dosage normally administered for prophylaxis, diagnosis, or therapy; dose-related therapeutic failures or clinical events that could be related to a prescribed dosage that is subtherapeutic secondary to a recent dosage reduction or discontinuation, total noncompliance (taking no drug), or partial noncompliance (taking less than or inadvertently taking more than prescribed); drug interactions; and interactions between drugs and alcohol.

A total of 5497 patients were involved in the study, with a mean of 458 patients/week of observation. The male:female patient ratio was 0.95; mean age was 54.8 years. The frequency of emergency department visits, total hospital admissions, and hospital admissions due to adverse drug events was constant over time, except for July, when no patients were admitted for adverse drug events. A total of 235 (4.3%) emergency department visits were drug related; 45 (19.1%) patients required hospital admission. A total of 115 drugs were implicated in the adverse drug events that led to the 235 visits. Aspirin accounted for 5.1% of the visits; amoxicillin and phenobarbital each accounted for 3.8%. The most frequent type of adverse drug event related to emergency department visits was adverse drug reaction (64%), followed by drug-related therapeutic failure (31%), drug interaction (4%), and interaction between drugs and alcohol (2%). The most common drug-related reason for hospital admission was therapeutic failure, followed by adverse drug reaction and drug interaction. Five hospitalized patients died as a result of an adverse drug event during the study, a mortality of 2.7/1000 admissions. Fifteen cases were deemed life threatening. Among the 45 patients admitted because of adverse drug events, 1.4% of these events were considered preventable.

Another prospective observational study evaluated the impact of drug-related visits to the emergency department at a 665-bed tertiary care center.[26] The objectives were to determine the prevalence of drug-related visits to the emergency department, determine the frequency of preventable versus nonpreventable drug-related visits, classify drug-related visits by severity, identify the most common drugs associated with drug-related visits, identify categories of drug-related problems associated with individual drug-related visits, and calculate the institution's cost associated with drug-related visits and subsequent hospital admissions.

Patients were randomly selected by a single investigator from 8:00 A.M.-11:30 P.M. for 35 nonconsecutive days over a 2-month period. Patients were excluded if they bypassed the triage nurse (e.g., ambulance), took illicit drugs, drank alcoholic beverages, or were admitted because of minor injuries. Patients also were excluded if the initial investigator was unable to obtain a drug history. A second investigator and the emergency department treating physician together reinterviewed patients to assess the reason for the emergency department visit. In addition, the medical director of the emergency department and two pharmacists not involved in the data collection phase reviewed all completed patient interview data collection forms. Consensus was required to establish a causal relationship between the drug and the reason for the emergency department visit. The cause of the adverse drug reaction was graded according to the Naranjo Adverse Drug Reaction Probability Scale.[31] Each drug-related visit subsequently was subcategorized as preventable or nonpreventable, graded according to severity, and classified as one of the eight drug-related problem categories listed earlier.[1]

Of 253 patients, 71 (28.1%) patients were identified as having a drug-related reason for their visit. Among these 71 patients, 50 (70.4%) had problems that were deemed preventable; the remaining 21 (29.6%) had unavoidable problems. Some drug-related visits were placed in more than one category. The most common drug-related problems were adverse drug reactions (29.8%), followed by overprescribing of the correct drug (16.0%), subtherapeutic dosage (13.8%), and either untreated indication or drug administration without an indication (10.6%). Severity of drug-related visits most often was classified as moderate. Drug classes most frequently associated with emergency department visits were cardiovascular agents and analgesics (18.3% each), antibiotics (11.2%), and asthma drugs (9.9%). Preventable drug-related visits primarily were due to noncompliance (46%), inappropriate prescribing and monitoring (44%), lack of patient education (8%), and dispensing errors (2%). The average cost/drug-related visit was approximately $1444 (range $253-$17,488)/preventable event versus approximately $1847/nonpreventable event (US $).

A third prospective observational study evaluated the contribution of adverse drug events to the overall number of emergency department visits and determined the rate of hospital admissions.[27] The study was conducted over an 8-month period in a 1500-bed tertiary care hospital in Chandigarh, India. Adverse drug events were classified into five categories: adverse drug reactions, drug interactions, patient noncompliance, physician noncompliance (e.g., inaccurate prescription, dose, or duration; patient not given sufficient information regarding prescribed drugs; or treatment not modified according to change in patient's condition), and overdose. To minimize interpersonal variability and bias in reporting, the Naranjo Adverse Drug Reaction Probability Scale[31] was used to correlate the emergency department visit with the adverse drug event.

A total of 4764 patients were included in the study. Of these, 280 (5.9%) were considered to have a drug-related problem; 24 (8.6%) required hospital admission. Mean patient age was 46.7 ± 19.4 years; most were younger than 20 and older than 80 years. Most adverse drug events were adverse drug reactions (45%), followed by patient noncompliance (28%), physician noncompliance (13%), accidental or intentional overdose (11%), and drug interactions (3%). All adverse drug events were classified as definite or probable (24%), possible (46%), or a contributing factor (30%). Of all emergency department visits and hospital admissions due to an adverse drug event, 52% and 55%, respectively, were considered preventable. Drug classes most often implicated were nonsteroidal antiinflammatory drugs, antidiabetic agents, and antibiotics.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: