Cultivating Quality: Reducing the Use of Potentially Inappropriate Medications in Older Adults

Bilyeu, Kathleen M. RN, PCCN; Gumm, Cynthia J. RN, PCCN; Fitzgerald, Jessica M. BSN, RN; Fox, Sherry W. PhD, RN; Selig, Patricia M. PhD, FNP-BC

AJN, American Journal of Nursing:
doi: 10.1097/01.NAJ.0000393060.94063.15
Feature Articles

A nurse-led quality improvement initiative aims to improve patient safety.

Keywords: adverse drug reactions, Beers criteria, elderly, older adults, potentially inappropriate medications, quality improvement

In Brief

A nurse-led quality improvement initiative aims to improve patient safety.

Author Information

Kathleen M. Bilyeu is clinical coordinator at Bon Secours St. Mary's Hospital, Richmond, VA, where Cynthia J. Gumm is a nurse manager, Jessica M. Fitzgerald is a Level 3 RN, and Sherry W. Fox is director of the Cullather Brain Tumor Quality of Life Center. Patricia M. Selig is director of nursing research at Bon Secours Richmond Health System. The authors acknowledge Holly M. Isaacs, RN, and Shaquilla V. Scott, LPN, for their assistance with literature review and data collection and analysis. Contact author, Kathleen M. Bilyeu:

Article Outline

Prescriptions for the elderly reportedly account for 25% to 40% of all prescriptions ordered in the United States1—and that population is steadily growing. The U.S. Census Bureau projects that by the year 2020 there will be nearly 55 million people ages 65 and older in this country, rising to almost 89 million by 2050—more than twice the current number.2 It stands to reason that both the number of hospitalized older adults and the number of prescriptions written for them will also increase proportionally. Improving the safety of medication use is among the Joint Commission's 2010 National Patient Safety Goals for hospitals.3 Indeed, Goal 8 calls for hospitals to "accurately and completely reconcile medications across the continuum of care" from admission and transfer to discharge.

The use of potentially inappropriate medications (PIMs) is a known risk factor for adverse drug reactions in the elderly; other factors include polypharmacy and poor or inconsistent adherence to the drug regimen.1 Older adults are notably vulnerable to adverse drug reactions, particularly during unexpected hospitalizations. Adverse drug reactions and other medication-related problems are associated with significant mortality; in 2000 such problems reportedly caused more than 100,000 deaths at a cost of $85 billion annually.4

Back to Top | Article Outline


Our community hospital has a strong shared governance model; direct care nurses have developed and implemented many quality improvement initiatives. This particular unit-based initiative was developed during a review of internal data that included nurse-sensitive indicators, as defined by the National Database of Nursing Quality Indicators (NDNQI). The nurses were interested in improving medication safety, particularly in older adults on a progressive telemetry unit, because of the large number of elderly patients admitted to this unit, a fall rate slightly above the NDNQI benchmark, and reported adverse drug reactions. Furthermore, in 2006 our hospital had adopted the Nurses Improving Care for Healthsystem Elders (NICHE) program. One of NICHE's medication safety recommendations is to reduce the number of PIMs in older adults at the time of admission; this also prompted our staff to address the issue.

In October 2007, several nursing staff (including three of us: KMB, CJG, and JMF) volunteered to form a medication safety task force. The task force met with an adult NP who had expertise in geriatrics to discuss how to best improve medication administration practices in our hospitalized older adults. Next, the group decided to collectively review the literature.

Literature review. The purpose of the review was to determine the prevalence of adverse drug reactions in the elderly, in particular in those receiving PIMs. We wanted to determine the significance of the problem of prescribing these drugs in this patient population. The task force searched the PubMed database using three keywords, Beers criteria, potentially inappropriate medications, and hospitalized elderly, and limited the search to articles published in English from December 1, 2003 through December 31, 2008. This initial search yielded 589 articles. From that list, we chose 12 studies to use as references for staff to read. Most of the 12 studies had been conducted in patients who were in nursing homes or EDs or were being cared for by primary care physicians. Only four studies addressed inpatient hospital care.

Beers and colleagues published the first set of criteria for identifying PIMs in the elderly in 1991.5 That set has since been revised and updated several times and has been validated as a drug utilization review tool.4,6 The most recent revision, by a panel of 12 experts in geriatric medicine, including Beers, delineates 48 individual medications or classes of medications that should be avoided in older adults.4 Sixty-six of these are categorized as having potential for adverse outcomes of high severity (high-severity rating); the rest are categorized as having potential for less severe outcomes (low-severity rating).

One comprehensive literature review found that, despite the use of the Beers criteria and another drug utilization review tool, the prescribing of PIMs to older adults ranges from 12% among community-dwelling elders to up to 40% of nursing home residents.7 More recently, in a large retrospective cohort study using data from 384 hospitals, Rothberg and colleagues investigated the rate of PIM prescribing among older adults admitted with certain common medical diagnoses.8 They found that 49% of the more than 493,000 patients studied had received at least one drug listed as a PIM using the Beers criteria. Laroche and colleagues studied the use of PIMs in patients on admission to an acute care geriatric unit (drugs were prescribed primarily by general practitioners) and again at discharge (drugs were prescribed by geriatricians).9 They found that hospitalization resulted in a reduction in the use of PIMs, and concluded that improved pharmacologic education for general practitioners, in particular on the potential for adverse drug reactions and the "peculiarities" of elderly patients, could further lower the use of PIMs. However, our search found no prospective, randomized, controlled trials to confirm that either the use of drug utilization review tools or pharmacologic education in geriatric medicine results in improved patient outcomes.

The literature didn't identify any specific nursing interventions for reducing adverse drug reactions associated with the use of PIMs or with problems in medication reconciliation. Further research in these areas is needed.

Objectives. Although prescriptive practices have traditionally been in the physicians' domain, we felt that a collaborative, interdisciplinary approach—one that included specific nursing interventions—was needed to ensure safer medication practices for the hospitalized older adults under our care. The task force identified three objectives:

* to determine the prevalence of the use of PIMs in the elderly population on the unit

* to develop a pilot quality improvement project to reduce the prescription and administration of PIMs

* to evaluate the effect of the interventions on prescribing patterns

Back to Top | Article Outline


The task force used the most recent revision of the Beers criteria as a guide in identifying PIMs prescribed for the geriatric patients on the unit.4, 10 In using the criteria, we targeted only those drugs listed with high-severity ratings.

The first step was to determine the prevalence of the use of PIMs in the geriatric population on the unit. First we compiled a list of all patients ages 65 years or older who'd been discharged from the unit between January and June, 2008. Then we conducted a retrospective chart review of 100 patients on that list, starting with the most recent discharge in June and working backward until we reached 100. All medication reconciliation orders were compared with the Beers list of 66 drugs given a high-severity rating. The review showed that of these 66 PIMs, 22 (33%) were routinely prescribed on admission; that is, either the patients were already taking these drugs or the admitting physician ordered them.

After analyzing the data, the task force decided that on admission of elderly patients, nursing staff would identify and seek to replace or discontinue any PIMs with high-severity ratings in an effort to minimize adverse drug reactions. We also agreed that if the prescribing physician felt it best for a particular patient to begin or continue to take a PIM, that patient would be closely monitored for potential adverse drug reactions.

The next step was to develop a strategic plan for implementing the pilot quality improvement project. At this point, we expanded the task force to include pharmacists, physicians, and case managers involved in discharge planning. The primary strategy was to educate the unit staff, including physicians, about the potential adverse events associated with PIMs. The following actions were taken between June and October, 2008:

* An algorithm was developed to guide staff in reviewing medication reconciliation orders for PIMs (see Figure 1).

* The algorithm, along with data from the retrospective chart review, was presented to the hospital's vice president of medical affairs, a physician who's familiar with Beers's work. The pilot project was readily supported.

* The algorithm was prominently placed in every patient's chart.

* Nurses, pharmacists, and physicians were encouraged to use the algorithm and consult the primary care physician on the use of all PIMs with high-severity ratings.

* An executive summary of the pilot project was developed for staff review and as an educational resource. This included the findings of the retrospective chart review, several pertinent research articles, the algorithm, and a list of alternative medications based on our hospital's formulary and preferred personal digital assistant (PDA) software and other sources.4, 11

* All nursing staff attended an educational in-service on the pilot project. This included a review of the Beers criteria and of staff expectations about using the algorithm to foster safer medication practices for our hospitalized elderly.

* Information on the Beers criteria, the algorithm, and the pilot project was included in the unit orientation packets for new nurse hires.

* Pocket cards listing the drug names and classes of the 66 PIMs with high-severity ratings were given to each nurse.

* Information on the NICHE program was featured on the home page of the hospital's Web site.

* A formal presentation on the pilot project was provided to the hospitalists who frequently serviced this population.

* The staff education was completed and the pilot project went live on November 10, 2008.

Once the project went live, data were gathered for the first 100 patients ages 65 or older who were admitted to the unit on or after November 10, 2008. These 100 patients were different from the 100 patients in the retrospective review group. We again compared medication reconciliation orders with the Beers list of 66 drugs given a high-severity rating.

Back to Top | Article Outline


Initial barriers. Achieving staff "buy-in" was a challenge at first, as using the algorithm was seen as another addition to the nurse's workload during the admission process. Once nurses were educated about the algorithm's potential to improve medication safety and therefore patient outcomes, they more readily accepted its use and their added responsibility. Physicians' participation steadily increased as they observed the nursing staff's enthusiasm for the project and recognized an opportunity to improve medication safety of the unit's elderly population.

Project champions. A supportive nursing administration and an engaged group of professional nurses encouraged staff interest and participation in the project. The assistance of an advanced practice nurse in the project design and implementation was invaluable because of her expertise in this patient population and her ability to empower the staff nurses to use and conduct research. The nursing staff was proud that the project was initiated by direct care nurses.

Back to Top | Article Outline


In the initial retrospective chart review, there were 81 instances in which one of 22 PIMs with high-severity ratings had been routinely prescribed to patients ages 65 years or older on admission. Of these 22 drugs, alprazolam (Xanax) and digoxin (Lanoxin) were the most commonly prescribed, followed by amiodarone (Cordarone), ferrous sulfate (iron), and naproxen (Naprosyn and others). (See Table 1.)

For the patients admitted in November and December, 2008, there were 93 instances in which one of 26 PIMs with high-severity ratings was prescribed to elderly patients on admission; the most commonly prescribed were propoxyphene (Darvon), digoxin, and clonidine (Catapres), followed by ferrous sulfate (iron) and lorazepam (Ativan). (See Table 2.) In these 93 instances, 50 medications were continued because they were deemed medically necessary and had proven safe for those particular patients; seven were continued but at a lower dosage; two were replaced with a safer, alternative medication; and 34 were immediately discontinued (see Table 3). Thus, as a result of this pilot project, nearly half (46%) of these PIMs identified on admission were either discontinued or changed to a safer dosage or safer medication. Although for some PIMs (such as digoxin), the number of instances the drug was prescribed was higher in this second group of patients, in each case the prescriber had determined that the drug was safe and medically necessary for that patient. The strategies initiated for this pilot project modified or changed the prescribing practices of our prescribers with regard to PIMs. Additional data analysis is necessary to understand any correlations between changes in prescribing practice and the number of adverse drug reactions.

There were several unexpected outcomes of the pilot project. In addition to fostering safer medication practices, it increased compliance with the medication reconciliation process, according to anecdotal staff reports. As a result of the education component, nurses report more confidence in their ability to care for hospitalized older adults. They also report heightened awareness of the risks associated with the use of PIMs, and they are intervening more aggressively to prevent adverse drug reactions. For example, patients taking PIMs might be placed nearer to the nurses' station and monitored more closely.

Back to Top | Article Outline


The practice implications of this pilot project go beyond improved medication practices. As a result of the project's success, the staff has expressed feeling a stronger commitment to addressing adverse patient outcomes. Surveillance for PIMs and for adverse drug reactions is continuing, so that we can better understand how the new practice changes are affecting patient outcomes. The project leaders and staff nurses continue to disseminate the findings of the pilot project to other units, assisting them in using the Beers criteria in the medication reconciliation process with regard to hospitalized older adults.

This initiative occurred because direct care nurses took steps to evaluate their practice, consult the literature and seek out expertise, develop and execute a pilot project, gather data, and evaluate the outcomes. Its undertaking has helped to further empower nurses, building knowledge and confidence, and its success has stimulated discussion about other potential projects at the unit level. As one of the nurses on the task force stated during the final evaluation of the project, "It wasn't just what best practice did for our patients; it was also the experience of being involved in a project that changed practice."

Back to Top | Article Outline


1. Ferrario CG. Geropharmacology: a primer for advanced practice acute care and critical care nurses, part I. AACN Adv Crit Care 2008;19(1): 23-35.
2. Vincent GK, Velkoff VA. The next four decades: the older population in the United States: 2010 to 2050. 2010.
3. Joint Commission. National Patient Safety Goals. Accreditation program: hospitals. 2010.
4. Fick DM, et al. Updating the Beers criteria for potentially inappropriate medication use in older adults: results of a US consensus panel of experts. Arch Intern Med 2003;163(22): 2716-24.
5. Beers MH, et al. Explicit criteria for determining inappropriate medication use in nursing home residents. Arch Intern Med 1991;151(9): 1825-32.
6. Beers MH. Explicit criteria for determining potentially inappropriate medication use by the elderly: an update. Arch Intern Med 1997; 157(14):1531-6.
7. Gallagher P, et al. Inappropriate prescribing in the elderly. J Clin Pharm Ther 2007;32(2):113-21.
8. Rothberg MB, et al. Potentially inappropriate medication use in hospitalized elders. J Hosp Med 2008;3(2): 91-102.
9. Laroche ML, et al. Impact of hospitalisation in an acute medical geriatric unit on potentially inappropriate medication use. Drugs Aging 2006; 23(1):49-59.
10. Duke Clinical Research Institute. Beers Criteria (medication list): potentially inappropriate medications for the elderly according to the revised Beers criteria. n.d.
11. Pharmacist's Letter/Prescriber's Letter. Potentially harmful drugs in the elderly: Beers list and more. Therapeutic Research Center. 2007.
© 2011 Lippincott Williams & Wilkins, Inc.