Monday, June 5, 2023
Consider this all too common scenario: Mrs. Smith is a 55-year-old, white woman who has type 2 diabetes, hypertension, hypercholesterolemia, and arthritis. She has been admitted to the hospital to bring her diabetes under control and to be worked up for possible bladder cancer. Yesterday, she was started on a different drug for her diabetes. Today, she is complaining of significantly increased pain in her knee joints, and there is some redness, tenderness, and swelling in her knees. The medical intern thinks that this is a worsening of her arthritis. The medical resident thinks that this is a new medical problem. The pharmacy resident thinks that it is an adverse drug reaction (ADR). Who is right?
Contrary to the public's common belief, all drugs are dangerous. Just because a drug is approved by the US Food and Drug Administration (FDA) does not mean that it won't cause problems; all usually do, from minor side effects to permanent disability, to life threats, and even death.[1-3] The FDA approves drugs on the basis of benefits outweighing the risks, not because there is a complete lack of danger.
The overall incidence of ADRs is unknown. However, studies have found that about 8% of emergency department patients are there because of an ADR, and 3% to 8% of hospitalized patients were admitted because of an ADR.[4] In fact, about 7 of every 100 patients in the hospital will experience a serious ADR during their stay, and about 3 of every 1000 hospitalized patients may die as a result of an ADR.[5]
It is not an easy task to determine whether a patient is experiencing an adverse effect from a drug, let alone determining what type of ADR is occurring; the root cause may lie with the drug's pharmacokinetic profile, a patient allergy, a drug-drug interaction, or even human error. This difficulty may partly explain why many ADRs are never recognized as adverse events. Such oversights can lead to the subsequent use of other drugs to correct a drug-induced condition, which is a contributing factor to polypharmacy.[6] Additionally, unrecognized ADRs may be misdiagnosed as an exacerbation of an existing medical problem or as a new medical problem, with clinicians increasing dosages of current drugs or adding new drugs to treat a medical problem that does not exist. As dosages are increased or more drugs are added to a patient's regimen, the likelihood of drug-drug interactions and other ADRs increases.
Patients who are experiencing an adverse drug effect rarely think the problem was caused by a drug that they are taking. Instead, they often suspect foods they have eaten recently or new products they have used, such as cosmetics or laundry agents.
Many nurses are attuned to looking out for ADRs, especially in older adults.[7] Some pharmacists are more focused on drugs than on patients (although this is improving), so they may actually overestimate the likelihood that a drug is the cause of an adverse effect. How are doctors at recognizing ADRs? Unfortunately, there is scant research there, either.
Unfortunately, the difficulties in assessing drug causality are endless. ADRs are complex clinical problems that involve 3 key variables: the person, the drug, and the event. Each one of these factors has multiple aspects that could be involved in an ADR.
For the person: There are multitudes of variables to consider. For example, risk factors, such as age, sex, illness, severity of illness, comorbid features, poor renal or liver function, etc, may be involved.
For the drug: Dosage, duration of use, known toxicity, timing of administration, and other variables are to be considered. If you are dealing with a new drug, clinical trials may not have been robust enough to detect rare ADRs.[8]
For the event: Previous reports, timing of onset, disease exacerbation, and new medical problems often cloud the picture.
Then there is background noise that may confound the diagnosis. In fact, the clinical picture in the elderly or intensive care patient is often so complex, with the presence of multiple disorders and multiple drugs, that possible drug-related effects are not easily sorted out.[6] Poring over the Physicians' Desk Reference can be bewildering, with long lists of possible but not well-documented ADRs that are associated with a particular drug (often listed without incidence rates). Of course, there is also the possibility that the problem lies not with the drug itself, but rather with something in the drug product, such as a coloring dye or a nonactive ingredient that is an allergen for some people.
Clinicians should always consider the possibility that a drug may be associated with an exacerbation of a patient's condition or with the emergence of a new medical problem. The key is to recognize a change in a person's functioning, as this may be an early sign of an ADR.[7] When a drug reaction is suspected, the clinician should determine whether the drug has been previously associated with that reaction, then rule out alternative explanations, and finally establish a temporal link between the onset of the reaction and drug administration.
It is also wise to review the patient's clinical course, assessing the pertinent characteristics of the patient, the suspected drug, and the adverse event.[9] In particular, the patient's recently administered medications, significant medical problems, and risk factors (such as age, severity of illness, or reduced renal or hepatic function) should be scrutinized for signs of a possible ADR. The patient should be questioned about the use of other agents, such as over-the-counter drugs, herbal supplements, or medications borrowed from others. The ADR also needs to be evaluated with the patient's clinical state in mind: Could the event be explained by known characteristics of the patient's medical illnesses?
Remember, if the morbidity was caused by an agent, the offending substance was either swallowed, inhaled, or touched. Timing is a key factor because most adverse drug events occur within 12 hours of exposure, and often within an even shorter period. Therefore, we must ask: What has the person swallowed, inhaled, or touched during that time frame?
The most common ADR targets are the skin and the gastrointestinal tract. The most common responsible agents are antibiotics and allergens. With a possible allergen, checking for eosinophilia is helpful. A number of studies and texts have identified ways to identify particular cutaneous or gastrointestinal reactions.[10-15]
An empirical approach to identifying ADRs is sadly lacking because of the complex set of variables discussed earlier. In such complex situations, computer decision programs can be helpful. Electronic medical record systems can be programmed to fire alerts when a potential adverse drug event is about to occur or has already occurred.[16,17] Automated adverse drug event monitors can search for keywords or phrases throughout the patient's medical record to identify drug therapies, laboratory results, or problem lists that may indicate that a patient has already been treated for an ADR. This detection method uncovers significantly more adverse events, including medication errors, than relying on empirical methods or incident reports.[18,19]
Empirical methods to assess the likelihood that an ADR has taken place are also lacking. A more formal, logical analysis can help distinguish events that are attributable to a drug from those associated with underlying diseases or other factors.[6]
Several investigators, including researchers at the FDA, have developed such logical evaluation procedures, or algorithms, for evaluating the probability of an ADR.[6, 20-24] Almost all of these methods employ critical causation variables identified by Sir Austin Bradford Hill in 1965.[25] The most widely accepted of these instruments is the Naranjo algorithm[22] (Table). This method has been tested for internal validity with between-rater reliability testing, and its probability scale has consensual, content, and concurrent validity as well as ease of use.
Table. The Naranjo Algorithm
| To assess the adverse drug reaction, please answer the following and give pertinent score. | ||||
|---|---|---|---|---|
| Question | Yes | No | Do Not Know | Score |
| 1. Are there previous conclusive reports on this reaction? | +1 | 0 | 0 | |
| 2. Did the adverse event appear after the suspected drug was administered? | +2 | -1 | 0 | |
| 3. Did the adverse reaction improve when the drug was discontinued or a specific antagonist was administered? | +1 | 0 | 0 | |
| 4. Did the adverse reaction reappear when the drug was readministered? | +2 | -1 | 0 | |
| 5. Are there alternative causes (other than the drug) that could on their own have caused the reaction? | -1 | +2 | 0 | |
| 6. Did the reaction reappear when a placebo was given? | -1 | +1 | 0 | |
| 7. Was the drug detected in the blood (or other fluids) in concentrations known to be toxic? | +1 | 0 | 0 | |
| 8. Was the reaction more severe when the dose was increased, or less severe when the dose was decreased? | +1 | 0 | 0 | |
| 9. Did the patient have a similar reaction to the same or similar drugs in any previous exposure? | +1 | 0 | 0 | |
| 10. Was the adverse event confirmed by any objective evidence? | +1 | 0 | 0 | |
Naranjo CA, Busto U, Sellers EM, et al. A method for estimating the probability of adverse drug reactions. Clin Pharmacol Ther. 1981;30:239-245. Reprinted with permission.
Naranjo scores of 9 or 10 indicate that an event was "definitely" an ADR; scores of 5-8 rate the likelihood as "probable"; scores of 1-4 are "possible"; and scores of less than 1 are "doubtful."
It is important to recognize when a patient is experiencing an ADR, or the oversight may lead to subsequent prescribing of more drugs to correct the drug-induced disease. However, it may be difficult to distinguish an ADR from an exacerbation of an existing disease or a new medical problem because the clinical picture is so complex. Such challenges test the diagnostic expertise of the best clinicians.
However, to recognize an ADR, you must first entertain the possibility of an adverse event. If you don't even look, you won't find them.
When a drug reaction is suspected, the clinician should investigate whether that particular drug is known to cause such a reaction, rule out alternative explanations, and establish a temporal link between the onset of the reaction and drug administration. When empirical methods fail or produce cloudy causation results, the more formal process of using a probability assessment tool, such as the Naranjo probability scale, will often produce clearer results.
The bottom line of recognizing ADRs is this: Whenever a patient experiences what looks like an exacerbation of an existing condition, or when a patient develops what seems like a new medical problem while being treated for something else, the possibility of an ADR must be added to the differential diagnosis. It just may be the drug!
This activity is supported by an independent educational grant from PhRMA.
