Geriatricians and geriatric pharmacists are knowledgeable about potential medication problems among elders, but other practitioners often need a little help. Elders' risk for adverse drug events (ADEs) is especially high because of increased prescription use and age-related metabolism and excretion changes. Patients =65 years consume approximately one third of all prescription drugs.1 Almost half (46%) take 5 or more agents, and 12% use 10 or more,2 frequently prescribed by different physicians.3 It is estimated that 35% of elders experience ADEs and that almost half of these events are preventable.3 Patients and clinicians may summarily dismiss ADEs?such as falls, appetite loss, fatigue, and mild confusion?as age-related changes. Unfortunately, these effects may be ADEs.
The Beers Criteria
The Beers Criteria, named after the project's lead researcher, Mark H. Beers, MD, represent a new type of consensus-based guideline?a guideline that looks specifically for potentially inappropriate medications in a specific population.4 The term potentially inappropriate medications might apply to other populations, but to date it has been used almost exclusively with older populations. In this case, the researchers defined elderly as 65 years or older.
In 1991, Dr. Beers and a panel of 12 geographically diverse experts?including geriatricians, pharmacists, and geriatric psychiatrists?began providing criteria for identifying potentially inappropriate drugs for elders.5 They updated their work in 1997 and in 2003.4,6 The authors stress the word potentially, because many factors can affect a drug's appropriateness.
In 1999, the Centers for Medicare & Medicaid Services adapted the Beers Criteria, incorporating them into nursing home regulatory guidelines, although this use was never the panel's intent. This development has generated considerable controversy. Additionally, many patient advocacy groups have publicized the Beers Criteria, and patients may be the first to educate clinicians about the Beers Criteria when they appear with questions prompted by this publicity. Patients may want to discontinue a medication because it is on this list, but they should be advised not to do so without first talking with their physician.
Guidelines were developed using the Delphi method, an analytical tool for formulating group consensus. Using a panel of subject matter experts, group consensus was established for 2 key issues: (1) medications or medication classes that should generally be avoided in elders 65 and older and (2) medications that should be avoided in elders with specific medical conditions. Each identified agent was assigned a severity rating of high or low based on its potential negative impact. The most recent Beers Criteria identify 48 agents or classes that should be avoided, as well as 20 disease states in which designated agents should be avoided. Pharmacists should note that the Beers Criteria are not evidence-based, but they represent thoughtful consensus.
Truncating the Beers list does the document an injustice, because it is well-written and concise. It has few absolute contraindications, instead tempering recommendations by describing detrimental dose ranges or comorbidities. An online version of the 2003 update, published in the Archives of Internal Medicine, is available at http://archinte.amaassn.org/cgi/content/full/163/22/2716. The Table lists drugs identified on the list as potentially inappropriate. Note that several of them are available over the counter, alone or in combination with other drugs.
Included drugs are accompanied by certain caveats, and with each revision of the list drugs are added, modified, or removed. For example, in 2005 the panel modified oxybutynin by retaining the immediate-release formulation but dropping the extended-release formula from the list because it tends to have fewer adverse effects. A significant change recently has been the removal of betablockers from the list in light of their utility in heart failure.7 Evidence has emerged that has proven that patients previously excluded from beta-blocker therapy (people with diabetes, chronic obstructive pulmonary disease, bronchospastic disease, and lipid abnormalities)8-11 can benefit and generally have few adverse events.12
The list also addresses medications that should not be used in older patients known to have specific medical conditions, but it makes no exceptions for palliative care or severe chronic disease. This is a limitation. Most of the disease-drug contraindications are well-known, such as the use of bupropion in patients with seizure disorders, olanzapine in those who are obese, or alphablockers in elders who have stress incontinence. Clinicians who treat elders must prescribe based on the elder's clinical presentation and must consider the entire medication regimen, history of medication use, comorbidities, and prognosis. Because functional status varies widely among elders, the choice of medications also will be predicated on each patient's robustness or frailty.
The Beers Criteria are not standards of care, and in many instances the benefits of a particular agent exceed the risk. Some drugs on the list can be used appropriately in specific elders. When listed drugs are used inappropriately in elderly patients and must be discontinued, finding alternatives can be difficult. Many elders have multiple conditions and take many medications. With that fact in mind, members of the American Society of Consultant Pharmacists' Fleetwood Project developed treatment algorithms that can guide decision making.13 The algorithms describe appropriate uses for drugs listed in the Beers Criteria. When the use is judged as inappropriate, they offer alternatives and the conditions under which the alternative is acceptable.
Despite the criteria's acceptance by the medical community, many physicians and pharmacists remain unaware of the criteria or the updates. Up to 23% of elders take at least one agent on the list.4 One recent Canadian study found that, among 3300 deaths among elders, one third were taking at least one agent on the Beers list.14 Studies directly linking negative outcomes and the Beers list have had mixed results, but most have demonstrated that avoidance of these drugs is associated with decreased iatrogenic complications among elders.4
Along with the Beers Criteria, pharmacists should be sensitive to other factors that may contribute to elders' negative outcomes, especially nonadherence. A 2003 survey of 17,685 seniors found that 52% do not take agents as prescribed. Two main reasons are (1) a belief that the drug made them feel worse or was not helping (25%) and (2) skipping or taking a smaller dose because of cost (26%).3 Researchers found that prescription drug coverage significantly impacted adherence?37% nonadherence among those without coverage, compared with 22% nonadherence for those with coverage.3
Few agents are tested specifically in the elderly, and, although elders are now actively recruited for clinical trials, representation remains inadequate. Whereas the Beers Criteria attempt to fill this knowledge gap, they are not derived from evidence-based studies?which is a frequent criticism. Their utility rests in drawing attention to those agents that require thoughtful clinical justification and vigilant monitoring.
Ms. Wick is a senior clinical research pharmacist at the National Cancer Institute, National Institutes of Health. The views expressed are those of the author and not those of any government agency.
For a list of references, send a stamped, self-addressed envelope to: References Department, Attn. A. Stahl, Pharmacy Times, 241 Forsgate Drive, Jamesburg, NJ 08831; or send an e-mail request to: firstname.lastname@example.org.