Medication reconciliation improvement in a private psychiatric inpatient hospital

Introduction

Medication reconciliation is an important part of hospital care. In 2005, medication reconciliation was added as a National Patient Safety Goal by the Joint Commission, and facilities were to conduct this process to be eligible for accreditation. By 2010, many hospitals were unable to meet this standard, and it was temporarily suspended.¹ This standard has since been moved to National Patient Safety Goal 3, and facilities must make a good faith effort to obtain the most up-to-date medication list possible.²

Medication reconciliation is defined in the United States by the Institute for Healthcare Improvement as “the process of identifying the most accurate list of a patient's current medicines—including the name, dosage, frequency, and route—and comparing them to the current list in use, recognizing any discrepancies, and documenting any changes, thus resulting in a complete list of medications, accurately communicated.”³ This process compares the patient's home medications with their medications on admission, transfer, or discharge to and from an inpatient facility.⁴ It is estimated that the medication reconciliations completed on admission in a large health care system were accurate approximately 40% of the time.⁵ The most common type of reconciliation discrepancy is omission of medications.^4,6,7 Other errors contributing to discrepancies include commission (medications added during intake), frequency, and dosing errors.⁸ Lizer and Brackbill⁹ completed an evaluation of medication reconciliation by a pharmacist within a behavioral health unit and found that discrepancy rates for omitted or incorrect medication, omitted or incorrect dose, and omitted or incorrect frequency were 48%, 31%, and 13%, respectively.

Performing high-quality, effective medication reconciliation is pivotal to identifying discrepancies, decreasing the number of adverse drug events, and improving patient safety.⁶ A study conducted by Cornish et al⁷ discovered that 54% of general internal medicine patients had at least 1 discrepancy, with 39% of those discrepancies having the potential to cause adverse problems with care. Examples of potentially harmful adverse events included continuation of diclofenac while being admitted for an upper gastrointestinal bleed and omission of a home supply of prednisone. Events such as these have been associated with increased financial burden within the United States, costing the country approximately $5.6 million per hospital year.¹⁰

Despite knowing that medication reconciliation on admission is vital to patient care, many barriers still exist to obtaining accurate medication lists.⁸ These pitfalls include time availability to conduct the interview, language barriers, severity of illness, cognitive status, and the patient's health literacy.⁷ These limitations during admission medication reconciliation may lead to potentially serious adverse events. Discrepancies on admission have occurred in up to 70% of patients, with almost one-third having the potential to cause harm to patients.^11,12

Methods

The purpose of this study was to evaluate improvements made to the medication-reconciliation process at an 80-bed, private, inpatient psychiatric hospital. Previously, this process required that the intake nurse obtain the medication, dosage, route, frequency, and indication for the medication for newly admitted patients. The medication-reconciliation form was then placed in the chart. Accuracy of the previous admission medication-reconciliation process was 45%. This led to the development and implementation of a new policy. Changes included revisions to the Admission Medication Reconciliation Form to include a “Verified With Outpatient Pharmacy” column as well as recording the reconciliation type (accurate, discrepancy, or no record). An in-service training was also provided to all nursing staff, highlighting proper documentation of the medication-reconciliation process and reviewing the previous study's discrepancies. In addition, a period of 24 hours after admission was specified to reconcile and verify home medications with the patient's pharmacy. The nursing supervisor ensured compliance with this time frame and addressed any inaccurate medication reconciliations with the appropriate nursing staff. This study evaluated the effectiveness of the updated policy.

The investigator used the same methods in this study as those used in the previous study. Data were collected from 44 randomly selected patients (or patient guardians), 13 years and older, admitted to the hospital during a 2-month period. The investigator collected data at least 24 hours after admission. This allowed nursing staff adequate time for clarification and verification of admission medication orders. This period also permitted the normal processes of care (ie, admitting the patient to the unit, collecting collateral information, and contacting the patient's pharmacy) to correct any additional problems that might arise during the admissions process, specifically in medication reconciliation. Data collected included the number of medications, patient demographics, and comorbidities (Table 1). The number and type of medication discrepancies were also recorded (Table 2).

TABLE 1: Demographic comparison

View Fullscreen

TABLE 2: Comparison of medication-reconciliation discrepancies

View Fullscreen

The investigator verified medications by contacting the patient's community pharmacy, obtaining medication administration records from other institutions and/or discharge summaries from other locations. Patients were then interviewed by the pharmacist to determine whether any discrepancies existed from the medication-reconciliation form collected by the nursing staff. Identified discrepancies were documented on the medication-reconciliation form, and the physician was notified to make necessary adjustments to the inpatient medication regimen.

Study data were recorded in a standardized data-collection form and were entered into Microsoft Excel® spreadsheet (Richmond, WA) for data-handling purposes. Statistical comparisons of proportions were assessed using χ² analysis, whereas comparisons of mean values between groups were assessed using Student t test. All statistical analyses were performed using SAS® 9.3 software (SAS Institute, Cary, NC).

Results

Baseline demographic information was similar between study 1 and study 2 (Table 1). Regarding gender (P = 0.073), race (P = 0.407), and average number of medications (P = 0.156), no statistical differences were found. The study before the implementation of the new medication-reconciliation process showed that the percentage of accurate medication reconciliations, defined as no discrepancies with medication name, dosage, route of administration, and frequency, was 45% (17 of 38 patients). After implementation of the updated processes, the investigators found that the percentage of accurate medication reconciliations increased to 80% (35 of 44 patients) (P = 0.0011).

Additionally, there was a reduction in the total number of discrepancies (Table 2). The most discrepancies detected by the investigators for study 1 and study 2 were omitted medications (12 and 9, respectively). Other discrepancies detected were incorrect dose (6 and 5), incorrect frequency (2 and 2), excess medication (1 and 1), incomplete orders (1 and 0), and no record at the pharmacy (1 and 0). The distributions of discrepancy types were similar between studies.

We also assessed whether discrepancies were more commonly observed during specific times of the day (Table 3). Study 1 showed more errors specifically around peak outpatient pharmacy hours. The total number of errors was 8 (39%) between the hours of 7:00 am and 5:00 pm. We had postulated that outpatient pharmacies would be easier to reach and verify medications during these hours. Despite that, other barriers may have existed during that period to prevent proper medication reconciliation. After implementation of the new medication-reconciliation protocol (study 2), the number of errors decreased to 3 (33%) during the same period.

TABLE 3: Total patients with discrepancies captured during specified times

View Fullscreen

Discussion

Improvements to admission medication-reconciliation procedures have increased the overall effectiveness of the process at the study site. The accuracy of medication-reconciliation verification increased from 45% to 80% (P = 0.0011). Despite the overall decrease in errors, there were still areas for improvement identified. Similar to the previous study, the medication-reconciliation process still had omitted medications and incorrect doses. Moving forward, additional training for all staff involved with medication reconciliation will aid in improving the overall effectiveness of the process. Incorporating pharmacists and student pharmacists into the process would also provide a valuable service, helping to identify over-the-counter and herbal medications, not currently included in the medication-reconciliation process, and resolving home and admission medication discrepancies.

In addition to the demonstrated clinical benefits of medication reconciliation by a pharmacist at admission, there are possible economic benefits to consider as well. Timely medication reconciliation strategies have the potential to prevent costly adverse drug events, which can lead to readmission, need for additional therapies, disability, and death.¹⁰ Researchers have previously studied interventions aimed at reducing adverse drug events in hospitals, and estimated that pharmacist-led reconciliation interventions had the highest expected net benefits. This analysis further stated that pharmacist-led reconciliation interventions had a significant probability of being cost-effective when compared with no medication reconciliation.¹³

Limitations exist in this study. First, this study was conducted at an inpatient psychiatric facility. This makes it difficult to generalize findings across other areas of health care. Most of this patient population was in need of rapid stabilization, making it difficult to perform prompt medication reconciliation. Additionally, patients may be reluctant to speak with nursing or pharmacy staff regarding their medications because of disease-specific communication barriers.⁹ Second, this was a small study with only 44 patients. Thus, a larger study will need to be conducted to validate these findings. Third, recall bias existed. Patients in an acute exacerbation of their diseases have a difficult time communicating specific information. However, verifying medications with the outpatient pharmacies helped to eliminate recall bias.

Despite the Joint Commission recognizing and addressing the importance of reliable medication reconciliation, facilities are only required to make a good faith effort to complete a medication record. In turn, medication errors still have a potential to occur at a high rate. Although the implementation of a new medication-reconciliation protocol proved effective for this facility, further improvements in the process are still needed.