Chapter 15: Medication Misadventures I: Adverse Drug Reactions

After completing this chapter, the reader will be able to

• Define adverse drug reactions (ADRs).

• Discuss the impact of ADRs on health care systems and patients.

• Explain methods for determining causality and probability of an ADR.

• Identify specialty drug information resources that can be used to locate information related to ADRs.

• Classify ADRs based on type and severity.

• Explain when, where, and how to report an ADR to the United States (U.S.) Food and Drug Administration (FDA).

• Explain how to report adverse reactions related to dietary supplements and vaccines.

• Describe the use of technology in ADR monitoring and reporting.

• Use guidelines from national organizations to implement an ADR reporting program.

1. An adverse drug reaction (ADR) is any unexpected, unintended, undesired, or excessive response to a medicine.

2. One of the first steps in establishing an ADR program is to define what each facility or organization categorizes as ADRs.

3. Several algorithms have been published that try to incorporate information about an ADR into a more objective form.

4. Communication is a critical component throughout the ADR monitoring process.

5. Technology plays an important role in monitoring, identifying, and minimizing ADRs.

6. ADR surveillance serves as a means to primarily provide early signals about possible problems.

The terminology surrounding adverse drug reactions (ADRs) is often confusing. All adverse drug events (ADEs), ADRs, and medication errors fall under the umbrella of medication misadventures. Medication misadventure is a very broad term, referring to any iatrogenic hazard or incident associated with medications. An ADE is the next broadest term, and refers to any injury caused by a medicine. An ADE encompasses all ADRs, including allergic and idiosyncratic reactions, as well as medication errors that result in harm to a patient.^1-5 ADRs and medication errors are the most specific terms. ❶ Adverse drug reaction refers to any unexpected, unintended, undesired, or excessive response to a medicine. A medication error is any preventable event that has the potential to lead to inappropriate medication use or patient harm.¹Figure 15–1 shows one way of graphically classifying these terms. Many of these concepts will be explored in greater depth later in this chapter and in Chapter 16. This chapter focuses on ADRs, highlighting the impact of ADRs, pertinent definitions and classifications, specialty drug information resources, ADR reporting systems, and future approaches to detecting and managing ADRs.

IMPACT OF ADVERSE DRUG REACTIONS IN THE UNITED STATES

All medications, including the inactive ingredients of a product, are capable of producing adverse reactions.⁶ ADRs account for about 5% to 15% of all hospital admissions, cause patients to lose confidence in their health care providers, and lead to a significant increase in morbidity and mortality.^7-13 In 2010, there were over 40,000 deaths due to drug-induced causes, landing ADRs among the 10 leading causes of death in the United States.⁹

ADRs have economic consequences as well. ADRs result in an annual cost of $5 to $7 billion to the U.S. health care system. Each hospital in the United States spends up to $5.6 million annually as a result of ADEs. After experiencing an ADR, patients spend an average of 8 to 12 days longer in the hospital, increasing the cost of their hospitalization by $16,000 to $24,000.¹¹

The incidence of ADRs for hospitalized patients has been reported to be as high as 28%. Of course, ADRs do not affect hospitalized patients alone. Approximately 20% of the ambulatory population receiving medications suffers from ADRs.¹⁰ These outpatient events do not always result in hospitalization, but certainly affect morbidity and patient quality of life. Health care professionals agree that these estimates are somewhat conservative because many ADRs go undetected, unreported, and untreated.¹⁰ One review concluded that only 2% to 4% of all ADRs, and fewer than 10% of serious ADRs, are ever reported.¹¹

Although not all adverse reactions can be prevented, a large proportion of them can; the World Health Organization (WHO) estimates that at least 60% of ADRs are preventable. Preventable ADRs may be caused by such factors as the wrong diagnosis of a patient’s condition, incorrect dosing or selection of a prescription medication, interactions with other drugs including dietary supplements, and patients who do not follow the instructions for taking a medication.¹⁴ This highlights the need for health care practitioners and health care facilities to engage in the process of systematically preventing and responding to ADRs. WHO terms this pharmacovigilance, or the process of preventing and detecting adverse effects from medications.¹²

Several agencies and professional organizations across the country have an active role in pharmacovigilance, and contribute efforts to minimize the occurrence and impact of ADRs. In addition to WHO, some of the key organizations include the FDA, The Joint Commission (TJC), and the American Society of Health-System Pharmacists (ASHP). See Table 15–1. Despite the numerous organizations involved, pharmacovigilance depends heavily on individual health care practitioners, including pharmacists, physicians, and nurses, to take measures to minimize ADRs and report events. Practitioners need to understand the potential for ADRs and be prepared to recognize and prevent such occurrences in order to minimize adverse outcomes.

Pharmacists play a vital role in the medication use process. Multiple studies have highlighted the tremendous impact individual pharmacists can have on minimizing ADRs.^15,16 In one study published in the Journal of the American Medical Association, pharmacists began participating in patient rounds in an intensive care unit. By implementing this measure alone, the hospital significantly reduced the incidence of ADRs and saved an estimated $270,000 per year.¹³

DEFINITIONS

After reading this chapter, the reader should be able to evaluate ADRs and, potentially, develop and implement an ADR surveillance program at their hospital, clinic, or community pharmacy. ❷ One of the first steps in establishing an ADR program is to define what each facility or organization categorizes as ADRs. There are many definitions for ADRs that have been described in the literature, including guidance from national and international organizations, as well as individual facilities and practitioners. Several of these definitions are presented below. Rather than getting bogged down in the differences between these many definitions, use this information for guidance when starting to evaluate ADRs and develop an ADR monitoring and reporting system. Remember that institutions, as well as clinicians, use different definitions depending on their practice needs.

WHO defines an ADR as “a response to a medicine which is noxious and unintended, and which occurs at doses normally used in man.”¹⁷ WHO additionally defines the term side effect, which is “any unintended effect of a pharmaceutical product occurring at doses normally used by a patient which is related to the pharmacological properties of the drug.”¹⁷ Note that, in clinical practice, side effect is a broad term that may be used synonymously with ADR, or may even refer to an ADE.

Defined by the FDA, an ADR is any “undesirable experience” associated with the use of a drug or medical product in a patient.¹⁸ This definition is also fairly broad, and includes adverse events occurring from drug overdose as well as situations involving abuse. The FDA goes on to define an unexpected adverse reaction as one that is not listed in the current labeling for a medication as having been reported or associated with the use of the drug. This distinction is important, because unexpected ADRs are those that the FDA would like health care practitioners to report. (Reporting ADRs to the FDA will be discussed later in this chapter.) This definition focuses on reporting unusual, uncommon, or newly identified ADRs. Although common or expected ADRs are relevant and important, they do not provide the FDA with new or additional safety information. Unexpected drug reactions may be symptomatically or pathophysiologically related to an ADR listed in the current labeling, but may differ from the labeled ADR because of greater severity or specificity. For example, a medication may have abnormal liver function listed in its labeling as an adverse reaction. A health care practitioner should still report a case of hepatic necrosis following administration of the drug, because it represents an unexpected ADR of greater severity.¹⁹

Edwards and Aronson propose another definition of an ADR: “An appreciably harmful or unpleasant reaction, resulting from an intervention related to the use of a medicinal product, which predicts hazard from future administration and warrants prevention or specific treatment, or alteration of the dosage regimen, or withdrawal of the product.”²⁰ This is a very practical definition for health care practitioners, because it highlights the need to take action in order to manage the effects of the adverse reaction.

Finally, Karch and Lasagna,²¹ two prominent researchers in the area of ADRs, define a drug, an ADR, and a patient drug exposure as follows:

Many institutions use Karch and Lasagna’s definition because it excludes accidental poisonings as well as problems with drugs of abuse.

Remember, these definitions are intended to provide guidance and food for thought; in daily practice, the differences between these descriptions may not be relevant or clinically meaningful. What is important is that individuals find or develop a definition of ADRs that works for the needs of their pharmacy, clinic, or hospital.

One of the challenges in defining and managing ADRs is determining causality. How can it be determined whether or not a drug caused a patient’s adverse reaction? Cause and effect is difficult to prove in general, and ADRs are no exception. Many investigators have dealt with this problem by developing and publishing definitions, algorithms, and questionnaires that try to determine the probability of a reaction—that is, the likelihood that an ADR was caused by a particular drug or medication. There are over 30 different published methods for assessing adverse drug reaction causality. They fall into three broad categories:²²

1. Expert judgment/Global introspection: This method involves the individual assessment of the event by a health care practitioner, based on his or her clinical knowledge and experience without using any form of standardized tool.

2. Algorithm: This method uses specific questions to assign a weighted score that helps determine the probability of causality in a given reaction.

3. Probabilistic: This method uses Bayesian approaches and epidemiological data to calculate and estimate the probability of causality with advanced statistical methods.

To date, none of these attempts have been able to prove actual causality. These tools, however, are used to determine the probability that a particular drug caused an adverse event and will be described further below.

These algorithms and definitions use several important key concepts.^18,23 Dechallenge and rechallenge are often discussed. Dechallenge occurs when the drug is discontinued and the patient is then monitored to determine whether the ADR abates or decreases in intensity. Rechallenge occurs when the drug is discontinued and, after the ADR abates, the same drug is administered in an attempt to elicit the response again. Dechallenge and rechallenge are effective means for establishing a strong case that the drug was responsible for the ADR. Unfortunately, in clinical practice, a rechallenge may not be practical and may actually cause further harm to the patient. Patients who suffer a serious ADR may not be thrilled about experiencing the reaction again in the name of science. Therefore, a rechallenge may not always be practical, but a dechallenge is often essential.

Another important concept to consider is the temporal relationship between the drug and the event. Does the timeframe for development of the ADR make sense? Did the patient’s exposure to the drug precede the suspected ADR? If there are previous reports of the ADR in the literature, do these reports describe a temporal relationship between the drug and the event? Case reports, specialty drug information resources, and package inserts can be helpful in noting if a drug has been known to cause a certain type of reaction in a certain timeframe in the past. Unfortunately, the literature is not likely to be helpful for rare or new ADRs, but this does not discount the fact that a reaction may have occurred.

Naranjo and colleagues²³ developed the following definitions to assist in determining the probability of a suspected ADR:

USING ALGORITHMS

❸ Several algorithms have been published that try to incorporate information about an ADR into a more objective form. As described above, these algorithms each use a set of specific questions to determine the likelihood that the drug was responsible for the reaction, and establish a rational and scientific approach to what previously required strictly clinical judgment. Although algorithms are helpful in offering a systematic approach to assessing the probability of ADRs, they can be very time consuming and the results vary significantly according to the interpretation of multiple observers.

In 1979, Kramer and coworkers published a questionnaire composed of 56 yes or no questions (Appendix 15–1).²⁴ This questionnaire includes sections about the patient’s previous experience with the drug or related drugs, alternative etiologies, timing of events, drug concentrations, dechallenge, and rechallenge. Responses to each question are given a weighted value and these values are totaled. The total value then correlates to one of four categories: unlikely, possible, probable, or definite. One of the problems with the Kramer method (as well as other algorithms) is that clinicians can disagree on the weighted values because the user must make subjective judgments for some of the questions. Another problem inherent with this questionnaire is that an unexpected ADR may not score well because of lack of literature or previous experience with the ADR. If the reaction is not universally accepted or in the most recent edition of the Physicians’ Desk Reference, the suspected reaction would score a zero in this section. This makes the method devised by Kramer and associates less useful for new medications and for unexpected or emergent ADRs. Overall, however, the questionnaire provides health care professionals with the opportunity to use a standardized tool. Hutchinson and colleagues evaluated the reproducibility and validity of the Kramer questionnaire and concluded that, although the questionnaire was cumbersome to use, the method was superior to clinical judgment alone.²⁵

Naranjo and colleagues developed an alternative algorithm in 1981 (Appendix 15–2).²³ This algorithm asks 10 questions involving the following areas: temporal relationship, pattern of response, dechallenge or administration of an antagonist, rechallenge, alternative causes, placebo response, drug level in the body fluids or tissue, dose-response relationship, previous patient experience with the drug, and confirmation by any other objective evidence. Like the algorithm proposed by Kramer et al, the answer to each question is assigned a score. The score is then totaled and placed into a category from definite to doubtful. In the initial published report of this algorithm, Naranjo and colleagues tested the reproducibility and validity of the algorithm and found that their tool was a valid means of assessing ADRs. Today, the Naranjo algorithm is one of the most commonly used methods to assess adverse drug reaction causality, and has been considered a gold standard of assessing ADR causality. Compared to the method devised by Kramer et al, the Naranjo tool is much quicker to administer. However, it is not without its faults. The tool emphasizes rechallenge and dechallenge, which may pose some problems in evaluating ADRs as described previously. The Naranjo algorithm also asks about a response to placebo, which is rarely (if ever) administered in clinical practice today.

In 1982, Jones and colleagues published an algorithm that allows health care practitioners to answer a series of yes or no questions to determine the probability that an ADR occurred (Appendix 15–3).²⁶ The Jones algorithm asks similar questions as those used in the methods described above, but uses a dichotomous key design. Like the Naranjo algorithm, the tool developed by Jones et al is shorter and quicker to complete than Kramer’s questionnaire. Unlike either Kramer or Naranjo algorithms, the Jones method does not require summation of a score. Of note, the Jones algorithm was originally designed for use in a community health setting, illustrating that the assessment of ADRs is not limited to inpatient or institutional settings only.

All of the algorithms described above possess a certain degree of observer variability. However, each can be used to help determine whether an adverse event was precipitated by a certain drug or drug–drug combination. Michel and Knodel compared the three algorithms by Kramer, Jones, and Naranjo.²⁷ Their study found that the Naranjo algorithm was simpler and less time consuming, and compared favorably to the 56 questions asked by Kramer. Although there was agreement between the Naranjo and Jones algorithms, the study found a higher correlation between the Naranjo algorithm and the Kramer questionnaire. The authors concluded that more data were needed to support the use of the algorithm developed by Jones and colleagues.

A modern approach to assessing the causality of ADRs is being developed by researchers at Butler University College of Pharmacy and Health Sciences.²⁸ Their algorithm, the Butler University Adverse Drug Reaction Causality Assessment Tool (BADCAT), has been proposed as an easy-to-use alternative to traditional methods, and aims to improve the interrater reliability of ADR assessment techniques. BADCAT comprises 11 questions that ask the health care provider to assess the temporal relationship, previous reports of the potential ADR, patient-specific factors such as co-morbidities, dechallenge, rechallenge, and so on. BADCAT is unique in that it also includes an item related to clinical judgment—that is, the clinician’s professional opinion of whether or not the suspected drug caused or contributed to the patient’s adverse reaction. This item is factored into the overall score, which is interpreted as either a highly likely, likely, or probable ADR. BADCAT has not yet been published in its final form, but preliminary results indicate that the tool may demonstrate higher interrater agreement than the Naranjo algorithm.

At this time, no algorithm for assessing the causality or probability of an ADR has been proven superior.²⁹ Therefore, it is reasonable to select a method based on factors such as availability, speed and ease of completing the tool, and clinician preference.

PROBABILISTIC METHODS

A Bayesian approach to assessing adverse reactions has been developed by Lane.³⁰ Using the Bayesian approach, relevant information is collected and a quantitative measure of the odds that a particular drug caused a particular event is calculated. The Bayesian approach has the potential to be an outstanding tool for predicting populations that may be at higher risk for ADRs. However, this method requires complex calculations, significant time investments to develop the model, and the involvement of a statistician.³¹ Therefore, the Bayesian approach is likely not practical for implementation in individual clinics or hospitals. But WHO is using these methods to analyze data in their international drug monitoring network.²⁰

In addition to the methods for assessing adverse drug reaction causality and probability discussed above, there are also tools for assessing specific types of ADRs. For example, different methods exist for assessing drug-related liver toxicity.³²

SPECIALTY RESOURCES FOR ADVERSE DRUG REACTIONS

Methods used to assess the causality of ADRs, including the Kramer and Naranjo algorithms as well as the BADCAT tool, rely in part on previous documentation of the reaction occurring in response to administration of the suspected drug. For common or well-known ADRs, the prescribing information (also called the package insert) and major drug information compendia can be useful. However, for other suspected ADRs, specialty resources may be necessary.

Chapter 3 introduces ADR specialty resources, including gold standards references Meyler’s Side Effects of Drugs and Side Effects of Drugs Annual. This section briefly describes a few additional specialty resources that may be useful.

FDA’s MedWatch program for voluntary reporting of ADRs is discussed later in this chapter and Chapter 16. The results of this program are disseminated through a special section of the FDA Web site dedicated to ADRs and other medication safety topics (http://www.fda.gov/Safety/Medwatch). This online reference contains the FDA’s latest safety alerts and recalls as well as resources for health care professionals. The site also provides monthly summaries of changes to drug labeling that the FDA has made in response to reports from health care providers and others involved in pharmacovigilance.³³ Users can sign up to receive safety alerts by e-mail, or follow FDA MedWatch on Twitter using Internet-enabled mobile devices such as smartphones or tablet computers (http://www.twitter.com/FDAMedWatch).

Because the FDA’s Web site can be difficult to navigate, it is perhaps best used for current hot topics or news items related to ADRs. The search function is rudimentary, and searching MedWatch for previous reports of a suspected ADR can be challenging. An alternative solution is DrugCite (http://www.drugcite.com), an independent Web site that uses a series of algorithms to search and report on the FDA’s adverse events database. Although the depth of information provided by DrugCite is minimal, the site can be used to determine whether an adverse reaction has been previously reported to MedWatch in response to a particular drug and, if so, how often.^34,35 A mobile DrugCite application is available for Android-based devices.³⁶

Published case reports can also be useful in assessing and evaluating ADRs (see Chapter 5 for more information on evaluating case reports). Reactions Weekly is a weekly publication that indexes and abstracts ADR case reports and ADR-related news from around the world (see also Chapter 3). Because it pulls information from biomedical journals, scientific meetings, and the WHO International Drug Monitoring Programme, Reactions Weekly can be useful for efficiently searching the available information related to a suspected ADR.³⁷Clin-Alert is a similar publication that summarizes reports of adverse clinical events from more than 100 key research journals, although its focus is primarily on reports of drug–drug interactions (DDIs).³⁸

Various definitions and terminology have been used to classify ADRs. As discussed above, algorithms such as those developed by Naranjo, Kramer, and Jones have used the definite, probable, possible, and unlikely categories to classify ADRs by their probability. Other classification systems rank ADRs by severity or by their mechanism. When developing an ADR monitoring program, these various systems can be used to determine probability (cause and effect) and severity of ADRs and help describe and quantify data. The data may help identify severity of reactions that are occurring and which medications cause the most severe reactions. Classification systems can also help health care practitioners organize and present data, and facilitate monitoring of ADR trends and potential causative agents. These trends can be used to change prescribing habits or to alert institutions and organizations to potential problems with medications.

CLASSIFYING BY SEVERITY

Karch and Lasagna²¹ developed a method of classifying ADRs by severity, from minor to severe, as defined below:

• Minor: No antidote, therapy or prolongation of hospitalization is required in response to the ADR.

• Moderate: The management of the ADR requires a change in drug therapy, specific treatment, or an increase in hospitalization by at least 1 day.

• Severe: The ADR is potentially life threatening, causing permanent damage, or requiring intensive medical care.

• Lethal: The ADR directly or indirectly contributes to the death of the patient.

The FDA and WHO also use severity to classify ADRs. According to the these organizations, an ADR is serious when it results in death, is life threatening, causes or prolongs hospitalization, causes a significant persistent disability, results in a congenital anomaly, or requires intervention to prevent permanent damage.^14,39

CLASSIFYING BY MECHANISM

Karch and Lasagna also described various mechanisms by which ADRs occur.²¹ These mechanisms are related to the pharmacologic or pharmacodynamic properties of drugs, and can be used to classify the type of reaction that occurs:

• Idiosyncrasy: an uncharacteristic response of a patient to a drug, usually not occurring on administration.

• Hypersensitivity: a reaction, not explained by the pharmacologic effects of the drug, caused by altered reactivity of the patient and generally considered to be an allergic manifestation.

• Intolerance: a characteristic pharmacologic effect of a drug produced by an unusually small dose, so that the usual dose tends to induce a massive overaction.

• Drug interaction: an unusual pharmacologic response that could not be explained by the action of a single drug, but was caused by two or more drugs.

• Pharmacologic: a known, inherent pharmacologic effect of a drug, directly related to dose.

Classifying the ADR by its mechanism may aid in identifying similar drugs that can be expected to cause a reaction, or may help explain patient-specific reactions to medications.

Well-designed programs that monitor and identify ADRs, as well as broadcast information to the medical community, are essential. A later section of this chapter will discuss the importance of reporting suspected ADRs to the FDA and other stakeholders. But before ADRs can be reported, a program for detecting and capturing these reactions must be implemented at a facility or institution.

Prior to implementing an ADR program, the health care facility must educate its staff on the importance and significance of the program. The pharmacy department is in an excellent position to provide this education because of its involvement in the pharmacy and therapeutics (P&T) committee, pharmacokinetic dosing, drug utilization evaluation (DUE), and drug distribution. The pharmacy department can be an excellent resource for developing an ADR program, as well as providing data about ADRs to the P&T committee.

STEPS FOR IMPLEMENTING A PROGRAM

Accrediting bodies like TJC⁴⁰ require that hospitals and health care organizations have an ADR reporting program. These programs are generally a function of the P&T committee (or other medical staff committee) and the department of pharmacy. ASHP also encourages pharmacists and health care practitioners to take an active role in monitoring adverse events. ASHP has published very specific guidelines on ADR monitoring and reporting as part of its practice standards.⁴¹ TJC and ASHP standards can be used as a basis for starting an ADR monitoring program. In addition to the standards, the pharmacy and medical literature are rich with examples of successful programs, some of which will be reviewed in this chapter.

Steps for implementing an ADR monitoring program include the following:

1. Develop definitions and classifications of ADRs that work for the institution. The definitions and classifications in this chapter provide a good starting point for discussion.

2. Assign responsibility for the ADR program within the pharmacy and throughout other key departments. A multidisciplinary approach is an essential factor. This will improve awareness of the monitoring program and increase ADR reporting at all levels of patient care.

3. Develop a program with approval from the pharmacy department, medical staff and nursing department, as well as other appropriate areas within the facility. Cooperation is essential in initiation of a successful program.

4. Promote awareness of the program. Newsletters, e-mails, in-services, grand rounds presentations, and other educational programs are opportunities to increase awareness and garner support for the program.

5. Promote awareness of ADRs and the importance of reporting such events in order to increase patient safety. Again, newsletters, e-mails, in-services, grand rounds presentations, and other educations programs can be utilized to increase awareness of specific ADRs.

6. Establish mechanisms for screening ADRs continuously. These mechanisms should include retrospective reviews and concurrent monitoring, as well as prospective planning for high-risk groups. It is worthwhile to educate pharmacists to check for ADRs when they see orders for certain indicator drugs that are often used in treating an ADR (see Table 15–2), orders to discontinue or hold drugs, and orders to decrease the dose or frequency of a drug.⁴² Also, electronic screening methods to check for laboratory tests that are indicative of ADRs (e.g., drug levels, Clostridium difficile toxin assays, elevated serum potassium, low white blood cell counts) can be helpful.⁴³ Emergency box usage is another event that may trigger investigation by the pharmacist to determine if an ADR has occurred (e.g., use of epinephrine, diphenhydramine).⁴⁴ The previous examples highlight specific approaches to monitor for ADRs continuously. Additionally, retrospective review of indicator drug orders, laboratory tests, and/or emergency box usage may also bring to light previously unreported ADRs.

7. Develop internal forms or other mechanisms for data collection and reporting of ADRs. A quick Internet search will identify a number of ADR report forms used at specific institutions across the country. Some institutions use computer reporting, as well as hotline phone numbers.

8. Develop policies and procedures for handling ADR reports. Indicate who is responsible for sending them to the FDA.

9. Establish procedures for evaluating the causality and probability of ADRs, usually based on one of the classification systems previously discussed.

10. Routinely review ADRs for trends.

11. Develop preventive interventions. Examples of these include flagging patients who are at high risk of developing ADRs and identifying specific drugs which are likely to cause ADRs. Monitoring these patients and drugs more closely will help prevent or reduce the severity of possible ADRs.

12. Report all findings to P&T and/or other appropriate committee(s).

13. Develop strategies for decreasing the incidence of ADRs, depending on the opportunities presented by the ADRs reported. This vital step has often been ignored in the literature; however, for an ADR program to be part of the quality assurance process, it must be included wherever possible.

❹ Communication is a critical component throughout the ADR monitoring process. To ensure that the suspected drug is not administered again, and that patients receive necessary treatment and monitoring, each ADR should be documented thoroughly in the patient’s medical record. However, this action alone is not sufficient, as the information is often poorly visible and may be difficult to access. The Institute for Safe Medication Practices (ISMP) recommends that ADRs be communicated to the ADR monitoring team by documenting the reaction on a standardized order form, either in paper or electronic form, just as if prescribing drug therapy or ordering lab draws.⁴⁵ This increases visibility of the information and facilitates a timely response by all individuals involved in the patient’s care. In addition, ASHP recommends that patients and their caregivers be notified when a suspected ADR has occurred.⁴¹ Well-informed patients can help prevent ADRs from recurring in the future.⁴⁵ Finally, previously unreported or clinically important ADRs should be disseminated to the medical community by publication and/or presentation in an appropriate forum.⁴⁴

THE ROLE OF TECHNOLOGY AND EXEMPLAR PROGRAMS

❺ Technology plays an important role in monitoring, identifying, and minimizing ADRs. Information systems are available that can identify and alert practitioners to potential ADRs and detect potential DDIs that may contribute to ADRs. These systems search patient records and medical data to find drug names, drug levels, or drug–lab interactions (sometimes called triggers) that frequently indicate an ADR has occurred.⁴⁶ See Chapter 24 for more information on this topic. Sophisticated systems may also search through International Classification of Diseases (ICD-9) codes, nursing notes, or outpatient medical records. For example, a medication profile that includes an angiotensin-converting enzyme inhibitor (ACE-I) and a note in the medical chart that mentions cough might trigger the ADR surveillance system. This information is then used to alert a pharmacist or prescriber who can investigate the suspected ADR and manage any adverse effects.³⁹

An example of one such system was developed at Brigham and Women’s Hospital. This system was programmed to detect a combination of patient-specific factors and medications that may indicate a patient who has the potential to experience an ADR. For example, patients taking medications that required renal function-based dosing and who had elevated serum creatinine were flagged as patients at risk for development of an ADR. Various other screening rules were also programmed. When compared to traditional methods of identifying ADRs, this system detected more ADRs than spontaneous reports, but fewer than retrospective chart review. Interestingly, the errors detected by the computer system were different than those detected by chart review, indicating that a combination of multiple ADR detection systems may provide the best results. As expected, using the computer system saved work time, requiring five times fewer person-hours than the chart review method.⁴⁷ In another study, a similar ADE detection system identified potential ADEs in 64 of every 1000 admissions. The prescribing physician did not recognize 44% of the ADEs detected by the system.⁴⁸

At LDS Hospital in Salt Lake City, Utah, the Health Evaluation through Logical Processing (HELP) system monitors medical records around the clock and automatically identifies patients who may have experienced an ADR. The system does this by identifying certain flags such as stop orders, orders for antidotes, and abnormal laboratory values. The flags or signals are then reported to the pharmacy so that pharmacists can follow-up to determine if an actual adverse event has truly occurred.⁴³

Because hospitals are required by accrediting bodies to have an ADR reporting system, the use of technology for identifying, reporting, and minimizing ADRs is very common in the inpatient setting. Unfortunately, this is quite different from the ambulatory care setting. In general, the reporting of ADRs and other drug-related events has experienced minimal uptake in the primary care setting. Reasons for this include busy physician schedules, high cost of implementation, and complexity of outpatient prescribing practices. Because of this, the Agency for Healthcare Research and Quality (AHRQ) recently initiated a project to develop and test a reporting system for use in ambulatory clinics nationwide. Called the Medication Error and Adverse Drug Event Reporting System (MEADERS), it was designed with input from a group of experts in patient and medication safety in the primary care setting. An intuitive electronic system was developed that allowed health care professionals to report events using both simple data entry and open text boxes in order to capture the complete story regarding each medication error or ADE. MEADERS also includes the option of sending each report to the FDA’s MedWatch program.⁴⁹ ADR reports are documented via entry into an Internet-based system or a stand-alone system using Microsoft^® Access^™. MEADERS also includes an Internet-based report viewing system that allows users to browse reports from ambulatory setting across the country using the MedMarx^SM database. This provides capabilities for practitioners to compare their local or organizational reports to national data.⁵⁰ Initial field testing of MEADERS involved 24 practice sites who submitted a total of 507 reports during the pilot period. Of these reports, 27% involved ADEs. While physicians maintained that time was the primary barrier to reporting medication errors and ADEs, the average time spent per report was just 4.3 minutes. In addition, only eight reports were electronically forwarded to the FDA’s MedWatch program. Thus, the development team is currently adapting the nuances of the system to make reporting a seamless component of the normal work flow.⁴²

AHRQ had previously concluded that the data supporting ADR monitoring and reporting programs in the outpatient setting was limited, but initiatives like MEADERS demonstrate that evidence is still emerging in this practice setting.⁴⁶ Traditionally, hospital-based pharmacists have been the health care professional to provide the greatest number of spontaneous ADR reports to the FDA.⁵¹ However, community pharmacists are in a unique position to report on ADRs relating to nonprescription medications and dietary supplements. Community pharmacists can also report on reactions that patients are likely to detect themselves, such as ophthalmic and dermatological disorders. In fact, in countries such as the Netherlands, Japan, and Spain, the majority of ADR reports originate from community, rather than hospital, pharmacists.⁵¹

One of the finest examples concerning the use of information technology for ADR monitoring and reporting comes from the Department of Veterans Affairs (VA). In 2007, the VA refined its methods for collecting and reporting patient-specific ADR information. This was accomplished by developing two distinct, yet complementary, databases. The first involves patient-specific information related to ADRs. Health care providers within the VA system are responsible for entering ADR information into a specific part of the patients’ electronic medical record, called the Adverse Reaction Tracking (ART) package. This locally entered ADR information is then automatically extracted to construct a national ART database. Since its inception, the ART database has received over 50,000 entries per month. The second database, the VA Adverse Drug Event Reporting System (VA ADERS), is external to the electronic medical record and requires providers to report detailed information related to specific, preselected ADRs. VA ADERS also allows for direct submission to the FDA’s reporting programs MedWatch and the Vaccine Adverse Event Reporting System (VAERS), discussed later in this chapter. The impact of VA ADERS on ADR reporting is demonstrated by the number of reports generated compared to the VA’s previous system (the legacy Adverse Drug Event System). In the 5 years prior to the implementation of the VA ADERS, only 21,000 ADR reports were generated. In the 3 years after the implementation of the VA ADERS, nearly 150,000 ADR reports were generated.⁵²

Because DDIs may comprise as many as 59% of ADRs, computer systems that detect clinically significant interactions are also important for reducing ADRs.⁵³ Over the past 20 years, the integration of drug information databases with clinical information has allowed for increased screening of DDIs upon medication dispensing. As pharmacists are entering orders into patient profiles, the integrated drug information database is checking for DDIs with all concurrent medications. While these clinical decision support systems are a good first-line of defense in monitoring for DDIs, studies have suggested the interaction tools often perform at a suboptimal level, either by misclassifying the severity of specific DDIs or failing to provide sufficient evidence to support the supposed interaction. Additionally, studies have shown that clinicians override the majority of alerts, ignoring potentially important safety warnings, and may develop alert fatigue if they receive drug interaction alerts too frequently.⁵⁴ In response, Warholak and colleagues have developed a tool which allows pharmacies to evaluate the performance of their specific DDI detection system.⁵⁵

Waller proposed a potentially significant advancement in the use of technology for reporting ADRs. With this system, information regarding ADRs would be automatically captured from the computers of health care professionals.⁵⁶ When a trigger event occurred, such as the administration of a drug antidote, reports could be generated and sent to the FDA or other regulatory bodies without any action at all from the practitioner. This proposal, while not without its challenges for implementation, has the capability to greatly reduce the underreporting of ADRs.

Due to high costs and the time required to educate health care professionals on the changes, the health care industry has traditionally lagged behind other industries in the implementation of high-end technology information systems.⁵⁷ The 2009 passage of the American Recovery and Reinvestment Act (ARRA), also known as the stimulus package, called for the development of a nationwide health record system in the United States. The ARRA both places pressure on the health care community to adopt health information technologies and provides funding to make this transition a reality.⁵⁸ If and when this national health record system is implemented, it has the potential to significantly change the way ADRs are identified and minimized.

❻ ADR surveillance serves as a means to primarily provide early signals about possible problems, according to Gerald A. Faich, MD, MPH, Former Director, Office of Epidemiology and Biostatistics, Center for Drugs and Biologics, of the FDA. He goes on to explain that “neither industry nor the FDA should consider its scientific job complete when a new drug is approved.”⁵⁹ Faich is referring to the importance of postmarketing surveillance, or the process of continually monitoring and reviewing suspected adverse reactions associated with medications once they reach the market and are available to the public (see Chapter 22 for more about postmarketing surveillance). In order to detect an ADR that occurs once in every 10,000 patients exposed to the drug, at least 30,000 people would need to be treated with the medication.¹² Clinical trials that are required to approve a new drug may not enroll a sufficient number of patients, and often exclude special groups such as children, pregnant women, and the elderly. Due to the limited sample size and populations included in these trials, the FDA relies on postmarketing information to establish a better understanding of adverse events. In addition, the short duration of most clinical trials means that they will not detect ADRs with a delayed onset.⁶⁰ Pharmaceutical companies are required by the FDA to submit quarterly reports of all ADRs for the first 3 years that a drug is on the market, followed by annual reports thereafter, as part of the postmarketing surveillance system. However, the success of postmarketing surveillance is directly dependent on the active involvement of health care practitioners.¹⁴ Postmarketing ADR reports from health care providers have led to changes in prescribing habits, changes to drug labeling, and the withdrawal of various drugs from the market.

REPORTING ADRS TO THE FDA

With the passage of the Kefauver–Harris Amendment of 1962, the FDA was required to maintain a Spontaneous Reporting System (SRS). The SRS allows for a relatively inexpensive monitoring system of ADRs for all drugs marketed in the United States, throughout their entire life cycle.^53,61 All health care professionals and consumers can use this program to report ADRs and participate in the postmarketing surveillance process.

In the past, however, the problem with the SRS has been the lack of voluntary reporting by the medical community. In 1988, for example, a survey demonstrated that only 57% of community-based physicians were aware of the voluntary system of reporting, which used FDA Form 1639 to allow anyone to report an adverse event through the SRS.⁶² In June 1993, the FDA switched to a new program called MedWatch: The FDA Medical Products Reporting Program. With MedWatch, the FDA receives reports via mailings, phone calls, faxes, and the Internet. From 2000 to 2010, the number of reports submitted annually through MedWatch to the FDA’s Adverse Event Reporting System (AERS) database grew from 266,866 to 758,890.⁶³

The current MedWatch system allows health care providers to report suspected ADRs using FDA Form 3500 (Appendix 15–4), which can be submitted by mail, fax, phone, or online.⁶⁴ The MedWatch form asks for information related to the suspected ADR, the patient’s relevant medical history and medication regimen, and pertinent laboratory tests, although this information does not need to be complete in order for an ADR report to be submitted. From start to finish, the FDA estimates that it takes 36 minutes to complete a MedWatch form, including gathering the patient’s data, reviewing the instructions, and submitting the report.

Once submitted through the MedWatch system, ADR reports are received by a unit of the FDA called the Central Triage Unit. The Central Triage Unit screens reports and forwards them to the appropriate FDA program within 24 hours of receipt. The report becomes part of a database used by the FDA to identify signals or warnings related to drug safety that require further study or regulatory action (see the section Future Approaches to Pharmacovigilance). As a reminder, MedWatch is interested in capturing reports of serious ADRs, which the FDA defines as death, life-threatening events, hospitalization, disability, congenital anomaly, or requiring intervention to prevent permanent impairment or damage, as well as unexpected ADRs. The MedWatch program asks individuals to report an event even if they are not certain that the drug product was the cause.⁶⁵

The MedWatch program does not overcome the traditional lack of reporting seen with voluntary systems. It is important to recall that pharmaceutical manufacturers are required to report all adverse events to the FDA, whereas individual health care practitioners do so only voluntarily. Various explanations can account for the failure of practitioners to participate in the FDA program. Hoffman elucidates several reasons that physicians do not voluntarily report ADRs, which are as follows:

1. Failure to detect the reaction due to a low level of suspicion.

2. Fear of potential legal implications.

3. Lack of training about drug therapy.

4. Uncertainty about whether the drug causes the reaction.

5. Lack of clear responsibility for reporting.

6. Paperwork and time involved.

7. No financial incentive to report.

8. Unaware of reporting procedure or little understanding of it.

9. Lack of readily available reporting forms.

10. Desire to publish the report.

11. Fear that a useful drug will be removed from the market or given a bad name.

12. Complacency and lethargy.

13. Guilty feelings because of patient harm.

14. Reaction not worth reporting.⁶⁵

In a systematic review of available evidence, Lopez-Gonzalez and colleagues identified additional reasons why ADRs are underreported, including the beliefs that a single ADR report cannot meaningfully contribute to medical knowledge and only safe drugs are allowed on the market.⁶⁶ Another potential explanation for the lack of reporting is that medical record personnel, whose job might be to categorize and report data, are not familiar with ADRs and/or their method of documentation.⁶⁵

Many of the reasons noted above also explain pharmacists’ barriers to ADR reporting. A recent survey of Texas pharmacists investigated attitudes about voluntary reporting of serious ADEs in both community and hospital settings. While nearly 90% of respondents felt reporting could improve patient safety, 72.6% noted that the reporting process was time consuming. Furthermore, 55.5% of respondents felt the reporting process disrupted their normal workflow. Some of the pharmacists in this survey also mentioned the reporting would increase the risk of malpractice, compromise relationships with physicians, and potentially break trust with their patients.⁶⁷ Ultimately, the benefit of increased patient safety should be enough to overcome the barriers to reporting faced by health care practitioners.

Historically, consumer reporting of adverse events has provided little data to MedWatch in comparison to the number of reports submitted by health care professionals. With the passage of the Food and Drug Administration Amendments Act (FDAAA) in 2007, however, consumer reports have been increasing. The FDAAA called for increased measures to ensure the safety of marketed drugs. In an effort to increase consumer reporting of adverse events to the FDA, the act requires direct-to-consumer advertisements of prescription drugs to include the following statement: “You are encouraged to report negative side effects of prescription drugs to the FDA. Visit http://www.fda.gov/medwatch, or call 1-800-FDA-1088.”⁶⁸

In fact, in 2006, before the FDAAA was passed, consumers reported only 127,475 adverse event reports to MedWatch. In 2010, consumers were responsible for reporting 403,746 adverse event reports, nearing the total provided by all health care providers that year.⁶⁹ While this trend is encouraging, there is still room for an increased rate of reporting by consumers. Efforts to educate the public on its role in ADR reporting could significantly impact public health.⁷⁶

Although consumers have the option of submitting ADR reports directly to the MedWatch program, FDA Form 3500 may be difficult for patients to navigate and complete. For example, they may not have ready access to their medical records, laboratory test results, and so on. Fortunately, a simplified, consumer-focused Form 3500B is available for patients to report ADRs to the FDA. A reporting mechanism is available for these individuals through the FDA’s Consumer Complaint Coordinators. Available by telephone, these coordinators listen to consumer complaints, document the suspected problem, and follow up when needed. In the past, the FDA has recalled nonprescription products after just two or three reports of serious problems were received through the Consumer Complaint Coordinators.⁷⁰

DIETARY SUPPLEMENT ADR REPORTING

Dietary supplements, including herbs, vitamins, minerals, and other so-called “nutraceuticals,” can and do cause ADRs. Many of these products have powerful pharmacological effects and, therefore, can cause ADRs. The medical literature contains many case reports that describe ADRs related to dietary supplements. However, the exact incidence of ADRs with dietary supplements is not known. In one survey of 2743 adults, 4% who took a dietary supplement reported an adverse event in the previous 12 months.⁷¹

Dietary supplements are regulated much differently than pharmaceuticals. The most striking difference is that these supplements can reach pharmacies and grocery store shelves without FDA approval and without any proof of safety or effectiveness. Until recently, manufacturers of these products were not required to monitor safety of their products through postmarketing surveillance, nor were they required to share information about safety with the FDA. However, the Dietary Supplement and Nonprescription Drug Consumer Protection Act of 2006 changed this. This legislation now requires that any manufacturer, packager, or distributor of a dietary supplement whose name appears on a product label must collect and report ADR information related to their product(s). Manufacturers or marketers of dietary supplements who receive reports or information about severe adverse reactions related to their products are now required to share these reports with the FDA.⁷² Retailers who sell these products may report events upstream to the manufacturer, packer, or distributor.

An interesting case related to dietary supplements and ADRs involves the herb ephedra, also known as ma huang. This herb was marketed as a dietary supplement and promoted primarily for weight loss and enhancing athletic performance. It received lots of negative media attention when a Minnesota football player died of heat exhaustion during a training session. It turned out that the football player was using ephedra for weight loss. Over a period of several years, there were well over 100 reports to the FDA of life-threatening ADRs linked to ephedra, including heart attacks, strokes, seizures, and death. In March 2004, the FDA banned the sale of dietary supplements containing ephedra,⁷³ although this ban was lifted by a judge in 2010.⁷⁴ Still, the FDA could not prove that ephedra was the cause of these numerous ADRs. Because there were no reporting standards or requirements for manufacturers to collect data on their products’ safety at this time, it would be unlikely that the FDA would ever have enough data to scientifically prove causality; they had to act based on the best available evidence. The FDA’s actions on ephedra will have long-lasting effects, and has ultimately set the precedent by which other cases against dietary supplements will be decided.

Currently, ADRs related to dietary supplements can be submitted through the FDA MedWatch program using the same approach that is used for prescription drugs. There are also third-party systems for collecting adverse event data such as Natural Medicines Watch. Natural Medicines Watch allows consumers or health care professionals to complete an electronic form reporting an adverse event related to any dietary supplement. This system uses a database of around 90,000 commercially available dietary supplements that allows the user to select the specific commercial product used by the patient. Since many commercially available dietary supplements contain multiple ingredients, this system may improve the reliability of adverse reaction reporting for supplements. Natural Medicines Watch is integrated with the drug information resource Natural Medicines Comprehensive Database (http://www.naturaldatabase.com), which increases its accessibility for health care practitioners. Natural Medicines Watch asks similar questions to those on the FDA MedWatch Form 3500, and all reports submitted through this system are also simultaneously shared with the FDA’s MedWatch program.⁷⁵

Just as with conventional drugs, reporting ADRs related to dietary supplements is voluntary for health care professionals. Practitioners and consumers have many options for reporting dietary supplement ADRs. As described above, they may submit them directly to the FDA, through a third-party system such as Natural Medicines Watch, or to the manufacturer or product distributor. Each of these avenues ultimately results in the information being provided to the FDA for evaluation and analysis. Within the FDA, these reports are sent to the Center for Food Safety and Applied Nutrition (CFSAN), which maintains a special group charged with evaluating these adverse events. The CFSAN Adverse Event Reporting System (CAERS) allows for a more efficient evaluation of reports specifically related to dietary supplements.⁷⁶

Despite the availability of this reporting infrastructure, the number of adverse events submitted by health professionals is minimal. From 2007 to 2010, an average of only 440 voluntary reports related to dietary supplement ADRs were submitted to the FDA each year.⁷⁷ During the same time frame, about 790 per year were reported through mandatory mechanisms, such as manufacturer or distributor reporting. Most of these reports originate from consumers rather than health professionals.⁷⁶

There may be a variety of reasons why few health professionals submit reports on dietary supplements to the FDA. Some of these may include:

• Patients often do not tell their providers about supplements use, or about the adverse effects they experience.

• Health professionals often do not ask about the use of dietary supplements.

• Consumers do not believe health professionals are able to answer questions about dietary supplements.

• Health professionals do not know where or how to submit reports of adverse reactions.

• There is a lack of information available to assess dietary supplement adverse reactions.

• Health care professionals may face a lack of time or support from management.

• Practitioners may have a fear of litigation.

ADRs related to dietary supplements can be assessed in much the same way as conventional drugs. However, there are some unique considerations. For example, when it comes to supplements, there is no such thing as generic equivalency. In other words, two supplement products made by different manufacturers cannot be considered equivalent. Different manufacturing practices and different extraction methods can dramatically change how a substance acts in the body, and this is especially true when it comes to herbal extracts. Because herbal extracts are made from plant material, variables such as extraction method, time of harvest, growing conditions, and so on can affect the chemical makeup of the extract. Therefore, two supplements containing Echinacea should not necessarily be considered equivalent. This may also mean that an ADR caused by one version of a supplement may not occur with another version.

VACCINE ADR REPORTING

While MedWatch accepts adverse event reports for all FDA-regulated drugs, biologics, and medical devices, as well as dietary supplements, a separate process has been established for the reporting of adverse reactions related to vaccinations. The VAERS is a postmarketing surveillance program supported by the FDA and the Centers for Disease Control and Prevention (CDC). It was established after the passage of the National Childhood Vaccine Injury Act in 1986. This act requires health care providers to report specific adverse events that follow vaccination. These events include those listed by the vaccine manufacturer as a contraindication to subsequent doses of the vaccine, as well as certain other reactions defined by the FDA and CDC. A list of these reportable reactions is available on the VAERS Web site (http://vaers.hhs.gov).⁷⁸ Around 30,000 reports are submitted to VAERS annually. Of these, approximately 13% are considered serious.⁷⁹ Similar to the MedWatch program, patients and practitioners can submit reports of adverse reactions to VAERS online, by fax, or by mail. Detailed instructions regarding this process are also available on the VAERS Web site.

The FDA has recognized that voluntary postmarketing evaluations of drug safety data alone are not sufficient to identify serious ADRs. High-profile adverse events related to drugs, such as rofecoxib (Vioxx^®) and rosiglitazone (Avandia^®), are prime examples of the insufficiency of the voluntary reporting system. The FDA, working with private companies and academic researchers, has developed new methodologies to get early detection signals about safety issues with medications. These latest techniques include the development of additional pharmacovigilance methods as well as the improvement of the existing SRS. For example, the FDA is using data mining approaches for a more rapid identification of potential problems. Data mining is a statistical process that attempts to find a drug-associated event that appears more often in the ADR databases than would normally be expected to occur in the general population. When a higher-than-expected event occurs, this is referred to as signal detection. Statistics are used to identify these signals using both Bayesian and non-Bayesian methods. When an association is flagged in the system, the FDA then requires a clinical review involving experts to decide if any further action is needed for these events.⁸⁰ Although data mining and signal detection cannot determine a causal relationship between a drug and an adverse reaction, it is an efficient means of identifying ADRs that may warrant further investigation by the FDA.

With these new techniques, some hospitals have investigated pharmacovigilance methods for evaluating the safety profile of drugs at their institutions. The New York Presbyterian Hospital implemented a pharmacovigilance system that involved the electronic health record and used natural language computer processing. Various statistical techniques were used to establish associations between drugs and ADRs, as well as cutoff thresholds. In the hospital’s feasibility study, seven drug classes were evaluated for novel adverse events. The researchers believed that the use of a comprehensive, unstructured data-monitoring program of electronic health records was a feasible option for computerized pharmacovigilance at the local health-system level. This study demonstrates that data mining and signal detection, even at the local level, provide an effective mechanism for ADR reporting.⁸¹

The traditional randomized controlled clinical trial (RCT) is primarily designed to test the effectiveness of a drug or therapy, and is not usually powered to detect relevant safety information. With this in mind, it is easy to see why adverse reactions may go undetected in these trials. As discussed previously, it has been estimated that any ADR that occurs in fewer than 1 in 10,000 people will not be detected in most RCTs. Therefore, some experts have recommended the use of what is known as a large, simple trial (LST) to increase sample sizes and to monitor larger numbers of patients during clinical testing. LSTs have less stringent eligibility criteria than traditional RCTs, and subjects are typically studied for shorter periods of time. LST designs still use randomization techniques and continue to have important safety monitoring thresholds; they are not powered to determine efficacy. The larger sample sizes seen with the LST design provide more data for ADR monitoring methods such as signal detection. In addition, the larger number of subjects increases the probability of picking up adverse events that are less common.⁸² However, because LSTs are of relatively short duration, they do not overcome the challenge of detecting ADRs with a long latency.

Pharmacovigilance has traditionally relied on retrospective reports of ADRs, and relatively little information is available that helps health care professionals prospectively decide whether or not to use a drug in a particular patient. Pharmacogenetics may offer a solution for identifying characteristics and risk factors that predispose an individual to an ADR.⁵³ Genetic variations in enzymes that metabolize and eliminate drugs can lead to a patient’s increased exposure to a medication, potentially increasing their risk for adverse effects.⁸³ Genome-wide association studies scan for genetic markers across the DNA of many individuals to find variations that may be associated with serious ADRs.^84,85 As of 2012, 10 such studies had been published, providing evidence for specific genes associated with drug-induced hepatotoxicity, carbamazepine-related skin rash, and muscle toxicity following simvastatin administration.⁸⁵

Adverse drug reactions are a serious problem in the U.S. health care system, causing significant morbidity and mortality as well as costing billions of dollars annually. The problem is undeniably widespread, with the FDA receiving over 830,000 spontaneous reports of ADRs in 2010 alone.⁶² From 1998 to 2005, the number of reports documenting serious ADRs and fatal ADRs increased almost threefold. Astonishingly, during that time reports related to biological drugs grew almost 16-fold.⁸⁶ Because of documented underreporting, these numbers represent only a small fraction of the ADRs occurring annually. Recognition of the problem is an important step in developing strategies to minimize the occurrence of ADRs. Reporting of these reactions by health care professionals is vital to gauging progress and directing our efforts in medication safety and patient care.

Pharmacists and other health care practitioners play a vital role in developing, maintaining, and promoting ADR monitoring programs. Key steps in the development of these programs include identifying an institution-wide definition of ADRs, obtaining buy-in throughout the facility, establishing a mechanism to continuously screen for ADRs, and assigning responsibility for transmitting ADR reports. These programs provide valuable information about ADRs within the institution, as well as spontaneous reports of ADRs that can be forwarded to the FDA to aid in signal detection. ADR monitoring programs improve communication channels, generate additional education on adverse events, and positively impact patient care.

Technology is essential for monitoring, identifying, and minimizing ADRs, with most institutions and facilities utilizing computerized systems to screen for and alert practitioners to potential ADRs. The role of technology continues to expand as the FDA develops new approaches for signal detection and data mining. In addition, advances in pharmacogenetics may soon help practitioners proactively identify risk factors that predispose patients to an ADR, optimizing prescribing and medication management practices.

Despite the impact of ADRs and the tools available to assist in their management, ADRs remain underreported. Physicians and pharmacists report several barriers to reporting ADRs, including the time and hassle involved in reporting, as well as the beliefs that a single ADR report cannot meaningfully contribute to medical knowledge and that only safe drugs are allowed on the market. ADR monitoring programs can only be successful if health care providers believe in their potential to improve patient safety and are willing to be active participants. Therefore, future efforts should focus on strategies for overcoming barriers to ADRs reporting in order to improve medication safety–related outcomes and patient care.