Skip to Content

Adverse Drug Reactions: A New National Perspective on Risk

November 2006

According to a landmark new study, more than 700,000 Americans annually are diagnosed in hospital emergency departments (EDs) with adverse reactions to drugs – and, according to lead author Dr Daniel S. Budnitz*, these numbers are likely “on the low side”.

“[T]he problem is large and can be expected to increase,” write the authors, who cite an aging population and increases in outpatient service-delivery, a move to over-the-counter availability of prescription drugs, and use of chemoprevention drugs as main contributors to the expected rise.

Allergic reactions to drugs and accidental overdosing were the most common causes of adverse reactions to drugs and a significant cause of serious illness, according to the study. Established drugs such as insulin and warfarin were among the 18 drugs most frequently implicated in adverse drug events (ADEs), and people aged 65 years and over were in the highest risk-category.

While adverse drug events in hospitalized patients have been well studied, ADEs occurring outside the hospital setting have not received as much attention, in part because of the challenges in obtaining timely, nationally representative data.

“Outpatient use of drug therapies in the United States is common and may confer serious risks along with substantial therapeutic benefits,” they wrote. “Historically, the health burden of adverse events from therapeutic drug use in community-dwelling, nonhospitalized patients has been difficult to estimate, but the problem is large and can be expected to increase.”

Dr Budnitz and colleagues’ ongoing study†,‡ – the National Electronic Injury Surveillance System–Cooperative Adverse Drug Event Surveillance project (NEISS-CADES) – is being conducted to estimate and describe the national burden of ADEs that lead to emergency department visits. It is the first study to gather such a large pool of national data on outpatient ADEs.

Frequency of ADEs in the US Population

Over a two-year period, researchers studied 21,298 cases of ADEs in adults and children that occurred in a representative sample of 63 hospital emergency departments nationwide. Adverse events were defined as allergic reactions, undesirable effects at recommended doses, unintentional overdoses, or secondary effects, such as falls or choking.

According to the researchers, these data translate to an annual estimate of 701,547 people – or 2.4 people per 1,000. Dr Budnitz noted that the estimate is “on the low side” because it could not capture data on ADEs for which people sought treatment outside the ED, or for people who died from ADE-related events either before arriving at the hospital or in the ED. Moreover, some ADEs leading to ED visits (e.g., uncommon ADEs and ADEs with gradual symptom-onset) may go unrecognized in the ED by ED physicians.

Elderly people (aged 65 years and over) were found to have the highest risk for ADEs, with twice as many ED admissions and nearly seven times as many hospitalizations related to ADEs, compared with people under age 65 years. This age-group accounted 37.0% of estimated unintentional-injury visits requiring hospitalization, but accounted for 48.9% of estimated ADE visits requiring hospitalization.

Drugs Commonly Implicated in ADEs

A total of 18 drugs were implicated in 1% or more of ADEs treated in emergency departments (see table). Overall, five drug classes were most commonly implicated in ADEs: insulins, analgesics (containing opioids), anticoagulants, amoxicillin-based drugs and cold remedies (containing antihistamine).

The top two drugs implicated in ADEs – insulins (for diabetes) and warfarin (an anticoagulant used to treat cardiac arrhythmia and other conditions) – were implicated in 1 in 7 ADEs. The list also included seven antibiotics which together accounted for 1 in 8 ADEs.

Both insulin and warfarin have been used for many years and are considered very safe when taken at the correct doses. However, both typically require regular monitoring to prevent overdose or toxicity because of their narrow therapeutic dosing-range.

Number of Cases and Annual Estimate of Drugs Most Commonly Implicated in Adverse Events Treated in Emergency Departments United States, 2004-2005, NEISS-CADES, CDCa,b


Adverse Drug Events

  Annual Estimate
No.  (%)     
Insulin 1,577 55,819 (8.0)
Warfarin 1,234 43,401c (6.2)
Amoxicillin 1,022 30,135 (4.3)
Aspirin 473 17,734 (2.5)
Trimethoprim-sulfamethoxazole 447 15,291 (2.2)
Hydrocodone-acetaminophen 420 15,512 (2.2)
Ibuprofen 526 14,852 (2.1)
Acetaminophen 497 12,832 (1.8)
Clopidogrel 241 10,931c (1.6)
Cephalexin 293 10,628 (1.5)
Penicillin 270 9,275 (1.3)
Amoxicillin-clavulanate 274 8,959 (1.3)
Azithromycin 255 8,794 (1.3)
Levofloxacin 230 8,682 (1.2)
Naproxen 245 8,634 (1.2)
Phenytoin 238 7,937 (1.1)
Oxycodone-acetaminophen 227 7,328 (1.0)
Metformin  179 6,678 (1.0)

Table reproduced from original article, which is a US Government work and is in the public domain in the USA.
b Drugs implicated in >=1% of adverse events. For 434 cases (annual estimate 15,784 persons, 2.2%) two of these 18 drugs were implicated in the adverse event. Therefore, these 18 drugs accounted for ADEs in 8,214 cases (annual estimate 277,636 persons, 39.6%).
c Estimates with coefficient of variation >30%: warfarin 32.5%, clopidogrel 36.6%.

ADEs with insulin include people with diabetes passing out because of low blood-sugar; people on warfarin experiencing excessive bleeding, and people taking amoxicillin breaking out in severe skin rashes (due to an allergic reaction).

“These are old drugs which are known to be extremely effective. We could not and would not want to live without them. But you’ve got to get the dose exactly right. Variations, especially on the high side, are really dangerous,” patient safety specialist Bruce Lambert, a professor at the University of Illinois at Chicago’s college of pharmacy, commenting on the study’s findings in an interview with the Associated Press.

Dr Budnitz said in an interview that another study concluded monitoring “was not occurring as often as might be appropriate, particularly for elderly people living in the community and taking medications on their own.” Drugs that require regular monitoring were implicated in 85.4% of ADEs due to overdoses in people over 65 years of age, the authors wrote.

The drug-monitoring issue is complicated by the fact that national monitoring standards for drugs such as warfarin do not exist: “Clinical practice for monitoring [warfarin] varies,” said Dr Budnitz in an interview with “In some cases it’s conducted by a doctor in an office or a lab, and sometimes in specialty anti-coagulation clinics whose only purpose is to monitor and adjust anticoagulants. The bottom line is that variation exists throughout the nation on how monitoring is done.”

An Ongoing Study

Initial results from the NEISS-CADES study provide a snapshot of the national incidence of ADEs and their burden on the US healthcare system. While remarkable for its scope, the study raises as many questions as it answers. For example, in addition to capturing data only on ADEs recognized within the ED, the study thus far has not parsed out ADEs associated with prescription versus over-the-counter drugs.

“We did not look at prescription status for this study, but we will look at this in the future,” said Dr Budnitz. “It’s complicated, because some drugs are available over-the-counter and by prescription. From our data, it’s difficult to determine if the active drug is part of a prescription or bought over the counter.”

As NEISS-CADES continues, some of these questions will likely be answered. Dr Budnitz and colleagues write, “Ongoing data collection…will enable more detailed examination of the epidemiology of ED-treated outpatient ADEs, focusing on specific patient populations, drug classes, conditions, and circumstances.”

The authors add that, until electronic health records make national, real-time data on outpatient drugs-safety available, theirs is one efficient, cost-effective way to monitor outpatient drug-safety and target appropriate prevention strategies.

Don’t Stop Taking Your Medications!

If the possibility of an ADE causes you concern, don’t stop taking your pills or injections – instead, contact your doctor.

“The most important point for your readers is, they should not stop taking the medications prescribed by their doctors,” Dr Budnitz cautioned. “What they should do is learn about their medications – learn about proper use and dosing, proper monitoring – to try to become knowledgeable patients, because these are often life-saving medications and should be used in the safest manner possible.”

It’s Not Just Pharmacy Drugs!

Many people think that alternative therapies and health supplements – such as vitamins and herbs – can do no harm. In fact, many natural products affect how prescription and non-prescription drugs are absorbed and used by your body, altering their intended effects on your health.

For people who take multiple medications, drug interactions may pose a problem – particularly as many people taking prescription drugs don’t bother to tell their doctors about their alternative health products.

Some drug combinations can lead to adverse events. Before taking any medication or combination of medications – prescription or non-prescription – talk to your doctor about correct dosing and the possibility of adverse effects, including allergic reactions.

To check the safety of your combination of medications and associated possible adverse events, visit the Drug Interactions Checker.

Make sure to include the full list of all your medications – including:

  • prescription drugs
  • non-prescription drugs
  • supplements (e.g., vitamins)
  • herbs.

*Dr Daniel S. Budnitz, MD, MPH, Medical Epidemiologist, Division of Healthcare Quality Promotion, Centers for Disease Control and Prevention.
†The NEISS-CADES program was developed by the Centers for Disease Control and Prevention (CDC), the US Consumer Product Safety Commission (CPSC), and the US Food and Drug Administration (FDA).
‡Results of the first two years of NEISS-CADES were published in the October 18 issue of the Journal of the American Medical Association.

National Surveillance of Emergency Department Visits for Outpatient Adverse Drug Events. Daniel S Budnitz et al, Journal of the American Medical Association, volume 296(15), pages 1858-1866, October 18, 2006.
Drug Reactions Send 700,000 Yearly to ER, The Associated Press, October 17, 2006.

Internet Resources:

Posted: November 2006