D. Villanyi et al.

The American Journal of Geriatric Pharmacotherapy

Medication Reconciliation: Identifying Medication Discrepancies in Acutely Ill Hospitalized Older Adults
Diane Villanyi, BSc Pharm, MD, FRCPC; Mark Fok, BSc Pharm, MD; and Roger Y. M. Wong, BMSc, MD, FRCPC, FACP
Department of Medicine, Division of Geriatric Medicine, University of British Columbia, Vancouver, British Columbia, Canada

ABSTRACT
Background: Medication discrepancies may occur during transitions from community to acute care hospitals. The elderly are at risk for such discrepancies due to multiple comorbidities and complex medication regimens. Medication reconciliation involves verifying medication use and identifying and rectifying discrepancies. Objective: The aim of this study was to describe the prevalences and types of medication discrepancies in acutely ill older patients. Methods: Patients who were 70 years and were admitted to any of 3 acute care for elders (ACE) units over a period of 2 nonconsecutive months in 2008 were prospectively enrolled. Medication discrepancies were classied as intentional, undocumented intentional, and unintentional. Unintentional medication discrepancies were classied by a blinded rater for potential to harm. This study was primarily qualitative, and descriptive (univariate) statistics are presented. Results: Sixty-seven patients (42 women; mean [SD] age, 84.0 [6.5] years) were enrolled. There were 37 unintentional prescription-medication discrepancies in 27 patients (40.3%) and 43 unintentional over-the-counter (OTC) medication discrepancies in 19 patients (28.4%), which translates to Medication Reconciliation Success Index (MRSI) of 89% for prescription medications and 59% for OTC medications. The overall MRSI was 83%. More than half of the prescription-medication discrepancies (56.8%) were classied as potentially causing moderate/severe discomfort or clinical deterioration. Conclusion: Despite a fairly high overall MRSI in these patients admitted to ACE units, a substantial proportion of the prescription-medication discrepancies were associated with potential harm. (Am J Geriatr Pharmacother. 2011;9: 339 344) 2011 Elsevier HS Journals, Inc. All rights reserved. Key words: acute care, elderly, medication reconciliation.
The data from this article were previously presented in abstract form at the International Association of Gerontology and Geriatrics World Congress; July 59, 2009; Paris, France.

Accepted for publication July 25, 2011. 2011 Elsevier HS Journals, Inc. All rights reserved.

doi:10.1016/j.amjopharm.2011.07.005 1543-5946/$ - see front matter

Volume 9 Number 5

October 2011

339

The American Journal of Geriatric Pharmacotherapy

D. Villanyi et al.

INTRODUCTION
The incidence of adverse events (AEs) in hospitalized patients is reported to be 3.7%, and 27.6% of these are related to medications.1,2 On average, drug-related AEs increase lengths of stay by 4.6 days and increase costs per stay by US $4685.3 Medication discrepancies can occur at any of several points in the medication-administration process during ordering, transcription, dispensing, and administering medications. In 1 study, 39% to 49% of medication discrepancies occurred during the ordering phase.4 A Canadian study by Baker et al5 reported a 7.5% overall incidence rate of AEs. In a subset of patients on a medical service, 42.6% of AEs were considered as secondary to a drug- or uid-related event. Approximately 25% of all medication-related AEs are reported to be due to medication discrepancies and are thus considered preventable.6 Older adults are at particular risk for medication discrepancies for several reasons. They may be on complex medication regimens.7,8 They may have multiple comorbidities, which has been found to be an independent risk factor for drug-related AEs.9 Other factors, such as cognitive impairment, language difculties, age-related changes in pharmacokinetics, and physical disabilities can play signicant roles in predisposing the elderly population to medication discrepancies.10 The Institute for Healthcare Improvement (IHI) (in the United States) and Safer Healthcare Now! (in Canada) have made medication reconciliation a top priority. The Joint Commission on Accreditation of Healthcare Organizations denes medication reconciliation as follows: The process of comparing a patients medication orders to all of the medications that the patient has been taking. This reconciliation is done to avoid medication errors such as omissions, duplications, dosing errors, or drug interactions. It should be done at every transition of care in which new medications are ordered or existing orders are rewritten. Transitions in care include changes in setting, service, practitioner or level of care.11 An accurate and up-to-date medication history is considered an essential component of safe and effective medical practice. Despite this, medication discrepancies on admission, transfer, and discharge are common.12,13 Few published data have specically addressed medication reconciliation in the elderly population in the acute care setting. Thus, the aim of the present study was to determine the rate and nature of medication discrepancies in this high-risk population in the context of spe-

cialized acute care for elders (ACE) units. The primary objective was to describe the prevalence of intentional, undocumented intentional, and unintentional medication discrepancies (for both prescription and over-thecounter [OTC] medications) in older adults admitted to ACE units.

PATIENTS AND METHODS

Population
This study was conducted at Vancouver General Hospital, Vancouver, British Columbia, Canada. Individuals aged 70 years who were admitted to the emergency department or directly from a long-term care facility to any of 3 ACE units and who were taking at least 1 regular prescription or OTC medication within 2 weeks before admission were eligible for enrollment in the study. Patients were excluded if they had a Mini-Mental State Examination (MMSE) score 20, which may have affected the ability to provide informed consent, or if they were unable to provide consent for other reasons. No formal medication-reconciliation process was in place on these units when the study was conducted.

Protocol
All patients admitted in the months of January 2008 and October 2008 were identied by means of daily census reports and were screened for eligibility. Data collection was performed in 2 nonconsecutive months to obtain a representative sample of patients and to capture physicians practices as accurately as possible. A letter of information was provided to eligible patients within 24 hours of admission. In all patients who consented to participate, scores on the Best Possible Medication History (BPMH) and on the MMSE were obtained through an interview with the patient and/or caregiver. The BPMH was obtained by a health care provider and included a thorough history of all regular medication use (prescribed and OTC) using a number of different sources of information. The interviews were conducted by 2 internal medicine residents (D.V. and M.F.). Other sources for the BPMH were relatives, prescription vials, PharmaNet (a provincial medication database that maintains individual patient prescription records), and the medication-administration record from the long-term care facility. Data for the following additional scales were collected from each patients electronic or nonelectronic medical record: Timed Up and Go,14 Clinical Frailty Score,15 Cumulative Illness Rating Scale,16 and Geriatric Prognostic Index.17 On admission, patients medication orders were compared with the preadmission BPMH. Medication dis-

340

D. Villanyi et al.

The American Journal of Geriatric Pharmacotherapy

crepancies were classied as intentional, undocumented intentional, and unintentional, based on the denitions established by the Institute for Healthcare Improvement and Safer Healthcare Now! campaign.18 An intentional medication discrepancy was one in which the physician intentionally changed, added, or omitted a medication that the patient was taking before admission, transfer, or discharge. An undocumented intentional discrepancy was one in which the physician added, changed, or discontinued a medication that the patient was taking before admission, transfer, or discharge, but he or she did not clearly document the change. An unintentional discrepancy was one in which the physician unintentionally changed, added, or omitted a medication that the patient was taking before admission, transfer, or discharge. Unintentional medication discrepancies were classied by a pharmacist (M.F.) for the potential to cause harm using a scale identied in the literature.19 The grading system identied class 1 as medication discrepancies unlikely to cause patient discomfort or clinical deterioration, class 2 as those with the potential to cause moderate discomfort or clinical deterioration, and class 3 as those with the potential to cause severe discomfort or clinical deterioration. The Medication Reconciliation Success Index (MRSI) is a measure of the percentage of good or acceptable orders.18 For example, a success index of 90% means that 90 in 100 orders had no discrepancy between the preadmission medications and the physician orders. This index was used in the present study and was calculated as follows: (Number of No discrepancies Number of Intentional documented discrepancies) (Number of No discrepancies total number of discrepancies) The institutional research ethics boards approved the study protocol before data were collected.

Table I. Baseline characteristics of the patients included in this study of medication discrepancies in older adults (N 67).
Characteristic Age, y Mean (SD) Range Sex, no. (%) Female Male Scale scores, mean (SD) MMSE Clinical Frailty Scale Cumulative Illness Rating Scale TUG (n 32) Admitted from, no. (%) Home Long-term care facility Hospital Medications per patient, mean (SD) Prescription Over the counter
MMSE Mini-Mental State Examination; TUG

Value

84.0 (6.5) 7096 42 (63.7) 25 (37.3) 25.6 (2.3) 4.6 (1.2) 17.7 (3.6) 13.8 (5.4) 54 (80.6) 11 (16.4) 2 (3.0) 5.9 (2.7) 1.8 (1.6)
Timed Up and Go.

Statistical Analysis
This study was primarily qualitative, and descriptive (univariate) statistics are presented. Means (SD) were calculated for continuous variables, and frequencies and percentages were calculated for categorical values.

tients; MMSE 20, 54; unable to communicate due to a language barrier or a medical condition (eg, delirium, sepsis), 30; and age 70, 4. The mean (SD) age of study participants was 84.0 (6.5) years. The majority of patients (80.6%) were admitted to hospital from home; 16.4% came from long-term care. On admission, patients were taking a mean (SD) of 7.7 (3.5) medications. Of these, the number of prescription medications per patient was 5.9 (2.7), and patients were taking 1.8 (1.6) OTC medications.

Medication Discrepancies
Of the 396 prescription medications that were taken by the study patients, 273 (68.9%) were ordered accurately by the admitting physician. There were 79 intentional discrepancies (19.9%), 7 intentional undocumented discrepancies (1.8%), and 37 unintentional undocumented discrepancies (9.3%). The unintentional undocumented medication discrepancies included 32 drug omissions (86.5%) and 5 medications being ordered with a different dose/frequency than the preadmission regimen (13.5%) (Table II).

RESULTS
Demographic and Medication Prole
A total of 168 patients were admitted to the ACE units in the 2 months of the study. Of these, 67 patients (42 women; 25 men) were interested in study enrollment (Table I). A total of 101 patients were excluded from analysis for the following reasons: declined, 13 pa-

341

The American Journal of Geriatric Pharmacotherapy

D. Villanyi et al.

Table II. Undocumented unintentional medication discrepancies in this study of medication discrepancies in older adults. Data are number (%) of discrepancies.
Prescription Discrepancies (n 37) 32 (86.5) 5 (13.5) 0 Over-the-Counter Discrepancies (n 43) 43 (100) 0 0

Discrepancy Drug omission Incorrect dose/frequency Incorrect drug

These discrepancies occurred in 27 patients (40.3%). On classication of the unintentional undocumented medication discrepancies, 21 (56.8%) were class 3 (potential to cause moderate to severe potential discomfort or clinical deterioration) (Table III). There were 123 OTC medications being taken by patients on admission. Seventy-three of these (59.3%) were accurately ordered by the admitting physician. Seven OTC medications (5.7%) were intentionally changed. The orders for 43 OTC medications (35.0%) were classied as unintentional undocumented discrepancies. Unintentional undocumented medication discrepancies occurred in the following drug classes: vitamins/ minerals (39), cardiovascular agents (18), bisphosphonates (5), asthma bronchodilators/corticosteroids (4), analgesia (3), and other (11).

Medication Reconciliation Success Index

The MRSI was 89% for prescription medications and 59% for OTC medications. Overall, the MRSI was 83%.

term discontinuation of these medications may result in a severe adverse outcome; however, the duration of admission in older adults may be difcult to predict, resulting in a longer interruption of therapy and a potentially greater risk for adverse outcomes. Previously published studies have reported that the most common type of medication discrepancy is drug omission. Orrico et al20 conducted a nurse-led study that compared medication lists from electronic medical records with those obtained by patient interview. Half of the discrepancies comprised antibiotics, anti-inammatory agents, analgesics, and vitamins. The most common types of discrepancies were medications recorded in the electronic medical record that the patient was no longer using (70.4%), followed by omission from the electronic medical record of a medication being taken by the patient (15.5%). Up to 61% of hospitalized patients have at least 1 drug omitted from their regimens,13,19,21 which underscores the importance of an effective medication-reconciliation process. Successful medication reconciliation depends on accurate medication history taking on admission, which serves as guidance for admitting physicians when writing their orders. In the older adult population, obtaining this medication history can be challenging. Dunham and Makoul22 reported that, in the outpatient setting, complete agreement between the medication list and what the patient is actually taking occurred in 5% of patients. A study by Pippins et al23 reported that unintentional medication discrepancies were more often due to errors in recording medication history than errors reconciling the history with medication orders. Based on these ndings, accurate medication history taking is crucial for improving medication safety. In the present study,

DISCUSSION
Despite the high MRSI in this patient population, more than half of the unintentional undocumented medication discrepancies had the potential to cause moderate/ severe discomfort or clinical deterioration. A baseline audit in Winnipeg, Manitoba, Canada, reported MRSIs in the range of 57% to 74%.18 The MRSI may also vary depending on the service to which a patient is admitted. On the surgical service at Vancouver General Hospital, a baseline MRSI was 66% (personal communication, J. De Lemos, 2008). Antihypertensives, corticosteroid inhalers, antiplatelet agents, and bisphosphonates were the major pharmacologic classes in which unintentional undocumented medication discrepancies were found. Arguably, short-

Table III. Grading of undocumented unintentional medication discrepancies in this study of medication discrepancies in older adults. Data are number (%) of discrepancies.
Prescription Discrepancies (n 37)* 16 (43.2) 12 (32.4) 9 (24.3) Over-the-Counter Discrepancies (n 43) 41 (95.3) 2 (4.7) 0

Class 1 2 3

*Percentages do not total 100% due to rounding. Unlikely to cause patient discomfort/clinical deterioration. May cause moderate discomfort/clinical deterioration. May cause severe discomfort/clinical deterioration.

342

D. Villanyi et al.

The American Journal of Geriatric Pharmacotherapy

49.2% of patients were able to provide their own medication histories. In 89.5% of this group, the MMSE score was 24. To obtain an accurate medication history in the remaining patients, alternate sources of information were used, such as a family member or PharmaNet, the province-wide electronic prescription database. Pippins et al23 reported that relying on family members or caregivers as sources of medication information was a risk factor for drug-related AEs. A metaanalysis estimated that 27% to 54% of patients have at least 1 unintentional medication discrepancy due to errors in medication history.13 The accuracy of medication history taking cannot be ascertained. It is paramount to assess how to obtain an accurate medication history and how to achieve medication reconciliation. Pharmacists have been identied as the ideal team member to assume these roles.24 Unfortunately, resource limitations may make exclusive pharmacist involvement virtually impossible. Given this constraint, a multidisciplinary medication-reconciliation process appears most feasible. One study, by Varkey et al,25 had nurses, pharmacists, and physicians complete a medication-reconciliation form on admission. The forms were reviewed by a pharmacist, and the medications were reconciled with the attending physician. The results from this study showed the mean number of admission medication discrepancies decreased from 0.5 per patient to 0 per patient and the mean number of discharge medication discrepancies decreased from 3.3 per patient to 1.8 per patient. Rates of compliance with medication reconciliation are not as robust as they could be.26 In a study by Coffey et al,26 62% to 77% of patients were screened by a pharmacist, and in 65% to 75% of high-risk patients, medications were reconciled by a pharmacist. An exclusively pharmacist-driven admission and discharge medication-reconciliation process at an academic center involved having a pharmacist obtain a complete preadmission medication history within 24 hours of admission.27 On the medicine unit, medication discrepancies were reduced from 57% to 33% (P 0.001). At Vancouver General Hospital, a nurse or physician obtains the medication history, and there is no routine involvement of the pharmacist in the medication-reconciliation process. There were several limitations to the present study. Patients were included from 1 hospital site. Due to the large number of patients excluded, only a relatively small sample of patients was analyzed. The patients who were excluded were arguably the more vulnerable patients, and capturing their information is paramount to minimizing medication discrepancies. The study was primar-

ily qualitative and may not be representative of hospitalized elderly patients admitted to other services. Recall bias may have been introduced into some patients BPMHs. The acute stress of being in a compromised state in a hospital may interfere with a patients or caregivers ability to perform objectively during the interview. Vancouver General Hospital is developing a medication-reconciliation pathway that will incorporate some strategies based on the ndings from the present study and from those from previously published studies. The next phase of this study will involve an assessment of medication discrepancies once the new medication-reconciliation protocol is in place.

CONCLUSIONS
Despite a fairly high overall mean MRSI in these patients admitted to ACE units, a substantial proportion of the prescription-medication discrepancies were associated with potential harm. Future studies are needed to investigate the effectiveness of a robust medication-reconciliation program in acutely ill older adults.

ACKNOWLEDGMENTS
Dr. Wong has been involved as a CME speaker, consultant, or member of advisory board for Janssen-Ortho, Lundbeck Canada, Merck Frosst, and Pzer. The authors have indicated that they have no other conicts of interest regarding the content of this article. All authors contributed equally to the conduct of the study and creation of the manuscript.

REFERENCES
1. Leape LL, Bates DW, Cullen DJ, et al. Systems analysis of adverse drug events. JAMA. 1995;274:35 43. 2. Brennan TA, Leape LL, Laird NM, et al. Incidence of adverse events and negligence in hospitalized patients. Results of the Harvard Medical Practice Study. N Engl J Med. 1991;324:370 376. 3. Bates DW, Spell N, Cullen DJ, et al. The costs of adverse drug events in hospitalized patients. JAMA. 1997;277: 307311. 4. Bates DW, Cullen DJ, Laird N, et al. Incidence of adverse drug events and potential adverse drug events. JAMA. 1995;274:29 34. 5. Baker GR, Norton PG, Flintoft V, et al. The Canadian Adverse Events Study: the incidence of adverse events among hospital patients in Canada. CMAJ. 2004;170; 1678 1686. 6. Aspden P, Wolcott J. Preventing Medication Errors. Washington, DC: National Academic Press; 2007.

343

The American Journal of Geriatric Pharmacotherapy

D. Villanyi et al.

7. Shelton P, Fritsch M, Scott M. Assessing medication appropriateness in the elderly: a review of available measures. Drugs. 2000;6:437 450. 8. Osterberg L, Blaschke T. Adherence to medication. N Engl J Med. 2005;353:487 497. 9. Field T, Gurwitz J, Harrold L, et al. Risk factors for adverse drug events among older adults in the ambulatory setting. J Am Geriatr Soc. 2004;51:1349 1354. 10. Walker J, Wynne H. Review: the frequency and severity of adverse drug reactions in elderly people. Age Ageing. 1994;23:255259. 11. The Join Commission. Issue 35: Using medication reconciliation to prevent errors. http://www.jointcommission. org/assets/1/18/SEA_35.PDF. Accessed August 25, 2011. 12. Vira T, Colquhoun M, Etchells E, et al. Reconcilable differences: correcting medication errors at hospital admission and discharge. Qual Saf Health Care. 2006;15:122126. 13. Tam VC, Knowles SR, Cornish PL, et al. Frequency, type and clinical importance of medication history errors at admission to hospital: a systematic review. CMAJ. 2005; 173:510 515. 14. Bohannon R. Reference values for the timed up and go test: a descriptive meta-analysis. J Geriatr Phys Ther. 2006; 29:64 68. 15. Rockwood K, Song X, MacKnight C, et al. A global clinical measure of tness and frailty in elderly people. CMAJ. 2005;173:489 495. 16. Parmalee P, Thuras P, Katz I, et al. Validation of the cumulative illness rating scale in a geriatric residential population. J Am Geriatr Soc. 1995;4:130 137. 17. Walter L, Brand R, Counsell S, et al. Development and validation of a prognostic index for 1-year mortality in older adults after hospitalization. JAMA. 2001;285: 29872994. 18. Quebec campaign. Getting Started Kit: Medication Reconciliation Prevention of Adverse Drug Events. How-to

19.

20.

21.

22. 23.

24.

25.

26.

27.

Guide. http://ismp-canada.org/download/Medication ReconciliationGettingStartedKit-Version2.pdf. Accessed August 25, 2011. Cornish PL, Knowles SR, Marchasano R, et al. Unintended medication discrepancies at the time of hospital admission. Arch Intern Med. 2005;165:424 429. Orrico KB. Sources and types of discrepancies between electronic medical records and actual outpatient medication use. J Manag Care Pharm. 2008;14:626 631. Lau HS, Florax C, Porsius AJ, et al. The completeness of medication histories in hospital medical records of patients admitted to general internal medicine wards. Br J Clin Pharmacol. 2000;49:597 603. Dunham D, Makoul G. Improving medication reconciliation in the 21st century. Curr Drug Saf. 2008;3:227229. Pippins JR, Gandhi TK, Hamann C, et al. Classifying and predicting errors of inpatient medication reconciliation. J Gen Intern Med. 2008;23:1414 1422. Gleason KM, Groszek JM, Sullivan C, et al. Reconciliation of Medication Discrepancies In Medication Histories and Admission Orders of Newly Hospitalized Patients. Am J Health Syst Pharm. 2004;61:1689 1695. Varkey P, Cunnningham J, OMeara J, et al. Multidisciplinary approach to inpatient medication reconciliation in an academic setting. Am J Health Syst Pharm. 2007;64: 850 854. Coffey M, Cornish P, Koonthanam T, et al. Implementation of admission medication reconciliation at two academic health sciences centres: challenges and success factors. Healthc Q. 2009;12 Spec No Patient: 102109. Murphy EM, Oxencis CJ, Klauck JA, et al. Medication reconciliation at an academic medical centre: implementation of a comprehensive program from admission to discharge. Am J Health Syst Pharm. 2009;66:2126 2131.

Address correspondence to: Roger Y. M. Wong, MD, Room 7153, 2775 Laurel Street, Vancouver, BC V5Z 1M9, Canada. E-mail: rymwong@interchange.ubc.ca

344