Brief Report

inappropriate prescribing in an older ED population^B

Fredric M. Hustey MD^a,b,*, Nicole Wallis MD^c, Jonathan Miller MD^c

^aDepartment of Emergency Medicine, Cleveland Clinic, Cleveland, OH 44195, USA

^bCleveland Clinic Lerner College of Medicine, Case Western Reserve University, Cleveland, OH 44195, USA

^cDepartment of Emergency Medicine, MetroHealth Medical Center, Cleveland, OH 44109,USA

Received 4 December 2006; revised 17 January 2007; accepted 18 January 2007

Abstract The objective of this study was to determine the prevalence of potentially inappropriate medication (PIMs) use in older emergency department (ED) patients based on the updated 2002 Beers criteria. This was a retrospective analysis of 352 consecutive ED visits by patients aged 65 years and older. The mean number of medications taken was 8.4 per patient. In the study population, 111 (32%; 95% confidence interval [CI], 27-36) of 352 patients were taking at least 1 PIM at ED presentation. Propoxyphene/acetaminophen (24/352, 7%; 95% CI, 4-10), Muscle relaxants (14/352, 4%; 95% CI, 2-

7), and antihistamines (12/352, 3%; 95% CI, 2-6) were the most common PIMs. Among 101 of 193 patients discharged home from the ED with a new prescription, 13 (13%; 95% CI, 6-19) were also given PIMs. The most common PIMs were propoxyphene/acetaminophen (3/101; 95% CI, 1-8), diazepam (3/ 101; 95% CI, 1-8), cyclobenzaprine (2/101, 2%; 95% CI, 0-7), and diphenhydramine (2/101, 2%; 95% CI, 0-7). Outpatient PIM use in older ED patients is highly prevalent. Further education is needed regarding prescribing practices in this population group.

D 2007

Introduction

The potential for adverse drug events in older emergency department (ED) patients is a growing concern [1]. The physiologic changes associated with aging, increased likeli- hood of comorbidities, and greater number of medications taken all place older patients at higher risk for adverse drug events than their younger counterparts [2]. The Beers criteria

This article was presented in part at the SAEM Annual Meeting in San Francisco, CA, May 2006.

^B Fredric M. Hustey, MD, is supported in part by an AGS/Hartford

Foundation/Atlantic Philanthropies Jahnigen Career Development Scholar Award.

* Corresponding author. Department of Emergency Medicine-E-19 Cleveland Clinic, Cleveland, OH 44195, USA. Tel.: +1 216 445 4546; fax:

+1 216 444 1703.

E-mail address: husteyf@ccf.org (F.M. Hustey).

is one widely accepted list that attempts to define potentially inappropriate medication use in older patients [3]. It was originally developed in 1991 for use in nursing home residents, but has since been applied in a variety of settings. This list, which was last updated in 2002, was developed by a consensus panel of nationally recognized experts using a modified Delphi technique [3]. Concerns have been raised regarding the rigidity of such criteria that cannot take into account all clinical factors affecting high-quality health care (eg, some medications on the list may still be justified under certain circumstances). However, medications on this list have been associated with increased morbidity and mortality in older patients [2], and should generally be avoided with the availability of safer alternatives.

The Beers list has been widely studied in outpatient clinics [4] and extended care facilities [5]. However, there is little available information regarding the Beers criteria

0735-6757/$ – see front matter D 2007 doi:10.1016/j.ajem.2007.01.018

in older ED patients [6,7]. One study focused on medications administered only during the ED visit [6]. This approach may not be as relevant to the principles of the Beers criteria, which have been traditionally applied to potentially inappropriate outpatient prescribing. Single doses of medications and those prescribed from the ED (which are often intended for short-term use) may not carry the same risk as drugs prescribed in the Outpatient setting. The only other publication involving ED patients was conducted before the more recent modifications of the Beers criteria [7]. There is no current evidence that we are aware of regarding the prevalence of inappropriate medication usage based on the Beers criteria in the outpatient Medication lists of older ED patients. Identify- ing a need in this population could lead to targeted interventions to improve care. The primary objective of this study was to determine the prevalence of inappropri- ate medications based on the Beers criteria in outpatient medication lists of older ED patients.

Methods

Study design

This was a retrospective chart review of a consecutive sample of ED visits by older patients. This study was reviewed and approved by the hospital institutional review board.

Study setting and population

Consecutive ED visits were reviewed for a 2-week period in June 2004 at an urban teaching hospital with approxi- mately 55000 ED visits per year and an affiliated ED residency program. All patients Age 65 years or older presenting to the ED during the study period were eligible for enrollment. Patients were excluded if there were incomplete data for analysis.

Study protocol

A consecutive sample of ED patient visits was obtained via chart review. This ED uses a Tracking system (EMSTAT, A4 Health Systems, Austin, TX) that maintains a record of all ED patient visits, including dates and times of ED presentation. All patient details are entered into this system during the registration and triage process. Information maintained in this tracking database includes medical record number, patient name, demographic data, date and time of ED visit, and discharge diagnosis. This system is also used to generate prescriptions at patient discharge, and maintains a record of each prescription given by the emergency physician. For those medications not included in the EMSTAT list, the system prints a blank prescription to be completed by the emergency physician and records them as such (blank). This tracking system was reviewed for all eligible ED visits during the study period. Patient demo-

graphics, discharge diagnoses, and medications prescribed at ED discharge were abstracted from this database. Potentially inappropriate medications prescribed from the ED were identified by using the 2002 explicit Beers criteria. As Beers recommendations were developed for outpatient prescribing, only ED discharge medications were included in this portion of the analysis.

The study hospital also uses an electronic medical record system (EPICARE, Madison, WI) for maintaining information on all patient encounters, including ED visits. This information includes a record of current medica- tions, which is documented by the ED physician at the time of the ED visit. Current medication lists were obtained via review of the electronic medical record system for the date of the ED visit. If there was no record of current medications in the ED encounter, the most recent hospital visit within the past 30 days was used to obtain a current medication list. If there was no visit within the past 30 days, then this was considered to be incomplete data and these patients were excluded from the study. Potentially inappropriate medications in current medication lists of older patients were identified with the 2002 explicit Beers criteria.

Chart abstractions were performed by trained reviewers using standardized chart abstraction forms. Fifteen percent of all chart abstractions were reviewed by the primary investigator for PIMs in usual medication lists to assess interrater reliability.

Outcome measures and data analysis

Primary outcome measures include the prevalence of PIMs in current medication lists at the time of the ED visit as well as the prevalence of potentially inappropriate prescribing by ED physicians. Secondary outcome measures include a description of common PIMs in each group. The j values are reported as a measure of Interrater agreement. Prevalence data are reported as proportions with 95% confidence intervals (CI).

Results

A total of 382 eligible patient visits were reviewed, of which 30 were subsequently excluded because of incom- plete data. Of the remaining 352 patients, 192 (55%) were women, 180 (51%) white, and 153 (44%) black. Mean age was 75.13 years (F7.05). The j value for interrater agreement regarding PIM abstractions was 1.

The mean number of medications in current medication lists at the time of the ED presentation was 8.4. Of 352 patients, 111 (32%; 95% CI, 27-36) had at least 1 PIM in the routine medication list at the time of ED presentation. Of the

111 patients, 106 had at least 1 PIM independent of comorbidities and 11 had PIMs considering comorbidities (6/11 overlap both categories). The most common PIMs are listed in Table 1.

Table 1 Most common PIMs in routine medication lists of

older ED patients

Medication

Propoxyphene/acetaminophen (Darvocet)

Muscle relaxants Antihistamines

(diphenhydramine and hydroxyzine)

Amiodarone Estrogens only Amitriptylene

Long-acting benzodiazepines (eg, diazepam, chlorazepate, chlordiazepoxide)

No. of patients

24/352 (7%; 95% CI, 4-10)

14/352 (4%; 95% CI, 2-7)

12/352 (3%; 95% CI, 2-6)

10/352 (3%; 95% CI, 1-5)

9/352 (3%; 95% CI, 1-5)

9/352 (3%; 95% CI, 1-5)

8/352 (2%; 95% CI, 1-4)

One hundred ninety-three of 352 patients were dis- charged home from the ED. Of these patients, 101 (52%; 95% CI, 45-59) were given a new prescription at discharge by the emergency physician. Thirteen (13%; 95% CI, 6-19) of these 101 patients were prescribed a PIM. The most common PIMs prescribed from the ED were propoxyphene/ acetaminophen (Darvocet) (3/101; 95% CI, 1-8), diazepam (Valium) (3/101; 95% CI, 1-8), cyclobenzaprine (Flexeril) (2/101; 2%, 95% CI, 0-7), and diphenhydramine (Benadryl)

(2/101; 2%, 95% CI, 0-7). The most common conditions among patients prescribed PIMs in the ED were musculo- skeletal pain or injury-related (5/13, 39%; 95% CI, 14-68).

Discussion

These results suggest a high prevalence of inappropriate prescription medication use in older ED patients. In contrast to recent previous studies that addressed only medications administered in the ED [6], our study focused on medi- cations reported in the usual medication lists of older ED patients. These patients may be at increased risk for future adverse drug events [2]. Unlike medications prescribed from the ED, which are often intended for short-term use, many of the drugs in our study are likely to be taken on a longer- term basis. Patients with painful conditions including Musculoskeletal problems may be at even higher risk. Propoxyphene/acetaminophen and muscle relaxants accounted for 11% of all inappropriate prescriptions in outpatient medication lists. Although the sample of patients receiving inappropriate prescriptions from the ED at discharge was small, the findings were similar. In addition, the most common conditions among patients prescribed inappropriate medications at ED discharge were also Musculoskeletal pain or injury-related.

The ED has been identified as an important site for case finding of at risk older adults [8,9]. However, reducing the prevalence of inappropriate prescribing in older patients may be challenging. These medications are likely to come from several sources. Patients who were recently hospital-

ized may have received them as part of their discharge instructions. Others may have received them during outpatient clinic visits. Some may have been prescribed them during prior ED visits or visits to urgent care type settings for acute injury. Education to enhance awareness of medication risks in older patients is likely to be needed across health care groups. In spite of such obstacles, the emergency physician-patient interaction in the ED still provides an opportunity for improving care. Even if emergency physicians in our study had focused on the 3 most common inappropriate medications in our population (propoxyphene/acetaminophen, muscle relaxants, and rou- tine outpatient antihistamine use), they could have poten- tially reduced inappropriate prescription medication usage by nearly 15%.

There are several limitations to this study. Medication lists were obtained via chart review instead of direct patient interviews. Chart reviewers were not blinded to the purposes of the study, possibly introducing reviewer bias. Small sample size limits statistical significance in certain sub- groups. This was a single site study at an urban tertiary care center and therefore the results may not be generalizable to other settings. Finally, Beers criteria do not encompass all potentially dangerous medications, specifically adverse drug-Drug interactions. This may have resulted in an underestimation of the prevalence of inappropriate prescrip- tion medication use in our population. In addition, some medications classified as inappropriate by Beers criteria may have been justified based on the clinical circumstances.

These results suggest a high prevalence of inappropriate prescription medication use in older ED patients. Further education may be needed regarding prescribing practices in this population. Future studies should address whether the presence of inappropriate prescription medication use in older ED patients is associated with an increased risk of adverse drug events, and whether this population may benefit from targeted interventions aimed at reducing inappropriate outpatient medication use.

References

1. Hohl CM, Dankoff J, Colacone A, et al. Polypharmacy, adverse drug- related events, and potential adverse drug interactions in elderly patients presenting to an emergency department [see comment]. Ann Emerg Med 2001;38:666 – 71.
2. Terrell KM, Heard K, Miller DK. Prescribing to older ED patients. Am J Emerg Med 2006;24:468 – 78.
3. Fick DM, Cooper JW, Wade WE, et al. Updating the Beers criteria for potentially Inappropriate medication use in older adults: results of a US consensus panel of experts [see comment] [erratum appears in Arch Intern Med. 2004 Feb 9;164(3):298]. Arch Intern Med 2003;163:2716 – 24.
4. Chang CM, Liu PY, Yang YH, et al. Use of the Beers criteria to predict Adverse drug reactions among first-visit elderly outpatients. Pharma- cotherapy 2005;25:831 – 8.
5. Beers MH, Ouslander JG, Fingold SF, et al. inappropriate medication prescribing in skilled-nursing facilities [see comment]. Ann Intern Med 1992;117:684 – 9.
6. Caterino JM, Emond JA, Camargo Jr CA. Inappropriate medication administration to the acutely ill elderly: a nationwide emergency department study, 1992-2000. J Am Geriatr Soc 2004;52:1847 – 55.
7. Chin MH, Wang LC, Jin L, et al. Appropriateness of medication selection for older persons in an urban academic emergency department [see comment]. Acad Emerg Med 1999;6:1232 – 42.
8. Rubenstein LZ. The emergency department: a useful site for CGA? [comment]. J Am Geriatr Soc 1996;44:601 – 2.
9. McCusker J, Bellavance F, Cardin S, et al. Screening for geriatric problems in the emergency department: reliability and validity. Identification of Seniors at Risk (ISAR) Steering Committee. Acad Emerg Med 1998;5:883 – 93.