KEY POINTS
* Implementation of a clinical reminder created in collaboration with a hospital's information technology department, within the electronic health record, improved medication reconciliations from 68% to 92% within a small rural hospital.
* Nurses agreed that accuracy of medication reconciliations would increase and improve patient safety, but also cited barriers to the use of clinical reminders.
* Clinical reminders are a cost-effective way to prompt healthcare providers to perform important tasks.
By 2050, current estimates indicate that there will be approximately 2 billion people older than 65 years worldwide.1 Concurrent with this increase, studies have emerged on medication reconciliations (MRs) and their impact on patient outcomes. Medication reconciliation is defined as the process of obtaining an accurate list of a patient's current medications and comparing that list with physician orders upon admission and discharge.2 Globally, patients admitted to hospitals encounter greater than a 50% chance of at least one prescribing error during the process of MR.3
Longevity and comorbidities play an important role in the growing issue of polypharmacy, defined as taking five or more medications, among those older than 65 years.4,5 Nobili et al5 found that more than 50% of patients 65 years and older are admitted to hospitals with an average of five diagnoses and take five or more medications. While polypharmacy is becoming more commonplace with the aging population, patient safety is of utmost concern to nursing and other healthcare professionals. Nurses are viewed as playing a major role in the prevention of drug-related issues. Studies indicate that nurses can identify potential patient safety concerns during the MR process.6
Chan and colleagues2 conducted a study on 470 patients 75 years of age and older taking five or more medications. Through a comparison of medication histories taken on admission by the physician and a history taken 24 hours after admission by a member of the MR team, 71.9% of participants experienced at least one medication discrepancy. While 66.8% of discrepancies were defined as unlikely to harm, 31.7% had potential to cause mild to moderate harm, and 1.5% were determined to have the potential to cause severe harm, deterioration, and prolonged hospital stay.2
A National Patient Safety Goal identified by the Joint Commission is the creation of policies to decrease medication errors, including the use of an accurate MR, in accredited facilities.7 In addition, the Health Information Technology for Economic and Clinical Health Act was developed to encourage the adoption of information technology (IT) systems by healthcare facilities to improve quality and safety of patient care.8 In addition, hospitals and physician offices must also meet "meaningful use" criteria as established by the Centers for Medicare and Medicaid Services to increase the adoption process of the integration of IT in patient care.9
Gimenez Manzorro and colleagues10 compared MR discrepancies before and after the implementation of an electronic tool. Before tool implementation, 1823 reconciled drugs were listed with a total of 132 discrepancies. After the electronic tool was incorporated into the computerized provider order entry (CPOE) system, 1958 medications were reconciled for a total of only 82 discrepancies, a decrease from 7.24% to 4.18%. Electronic tools for MRs have decreased medication errors and are useful reminders for nurses during reconciliation.10
Medication reconciliation is an issue that will continue to be addressed by healthcare organizations. Evidence suggests that discrepancies in MRs cause harm to patients. Adequate MR is a standard for patient care, and processes must be implemented within organizations to ensure compliance with accrediting bodies and ensure patient safety.
Clinical Reminders
The use of clinical reminders (CRs) in practice serves as a cue that an event that should have occurred has not. Clinical reminders can be simple in form such as an email or paper note placed on a chart or may be a system itself. The anesthesia information management system is a system that has been used in operating rooms to track patient data and alert anesthesia when abnormal data are received.11
Systems, such as computerized clinical decision support systems (CDSSs), have been instrumental in providing computer-based CRs with the overall goal of improvement of patient outcomes through the integration of technology. In a review examining CDSSs, 21 reminder-based studies revealed a 76% improvement in patient outcomes and a 55% decrease in medication errors in inpatient and outpatient settings when compared with pre-CDSS data.10 Wanderer et al11 found that, after a CDSS was incorporated into a task list for anesthesia, timely administration of antibiotics increased from 69% to 92%. Shelley et al12 implemented alerts, forms, order sets, and CRs in a CDSS to improve outcomes for patients with hypertension. The average systolic blood pressure decreased from 135.79 to 132.83 postintervention, and diastolic blood pressure decreased from 82.79 to 81.94.12
A concern for end users of CDSS is the integration of an information system. The success of CDSS will depend on how well it is implemented and accepted by those involved. Implementation of CDSS requires strategic planning to ensure ease of use and minimal to no interruption in clinician workflow.11
The adoption of IT systems in smaller, rural hospitals has been a slow, problematic process. Many rural hospitals have adopted basic systems that may be lacking in functions that are present in more comprehensive IT systems.9 The aim of this quality improvement study was to assess the improvement of MRs for patients 65 years and older in a rural hospital after a CR was incorporated in the electronic health record (EHR).
METHODS
Study Sample
This pilot study was approved by a university institutional review board and the chief nursing officer at the rural hospital. Specific aims for the study included (1) survey of nurses working on the medical-surgical unit before the CR implementation and (2) retrospective chart reviews preimplementation and postimplementation of the CR. The sample for the second goal included 100 charts of patients who had been discharged from the medical-surgical unit at a rural hospital during a 5-month period from June to November. Inclusion criteria for chart selections were adults 65 years or older, discharged from a medical-surgical unit in June and July (preimplementation), and discharged from a medical-surgical unit in October and November (postimplementation).
Study Design
This study used a descriptive pre/post design to evaluate the improvement of MRs after implementation of a CR.
Intervention
Before the CR implementation, medical-surgical nurses participated in an annual skills fair including a presentation on the importance of accurate MRs, as well as the inclusion of a CR in the EHR. The nurses were surveyed immediately after the presentation using a seven-item, 5-point Likert questionnaire based on the elements of relative advantage, compatibility, and complexity within Rogers' Diffusion of Innovation Theory.13 Rahimi et al14 incorporated these elements of Rogers' theory when evaluating physician and nurse attitudes in the implementation of a CPOE system. Various concepts from Rogers' Diffusion of Innovation Theory have been used when examining technology acceptance in healthcare.15-17
Through collaboration with the hospital's IT department, the CR was created within the EHR. The words "medication reconciliation" were placed on the top line of the EHR flow sheet where nurses' notes began. This was strategically placed in hopes that every nurse who took care of the patient would see and use the CR. Once a nurse clicked on the MR tab, a drop-down box appeared that required further information. The nurse could check "yes" if the MR was completed or "no" if it was not completed and then had the opportunity under "no" to explain why it had not been completed.
Data Collection
Twenty-six nurses completed the survey on attitudes about use of CRs to improve MR. Assistance from the Director of Quality and Education was needed to retrieve the charts from the electronic database at the facility. A data collection sheet was used to collect data from 50 charts preimplementation and 50 charts postimplementation during a retrospective chart review. The preimplementation chart reviews occurred in July, which was approximately 3 months before implementation. The postimplementation chart reviews occurred in November, 6 weeks after the CR was implemented.
Descriptive statistical analyses were performed using IBM SPSS Statistics version 20.0 (IBM, Armonk, NY).
Results
Nurse attitudes toward using a CR to improve MR were generally positive. One hundred percent of nurses agreed that "accuracy of MRs would increase with the use of CR and a CR will improve patient safety," and 91.3% agreed that "a CR would provide an opportunity to recognize discrepancies in the prescribing of medications." These results are similar to other studies as regards the importance of MR in increasing patient safety. Medication reconciliation is being viewed as an important component of healthcare practice to reduce the incorrect use of medication, as well as decrease errors and help alleviate adverse effects.6 However, 38.4% of nurses also agreed that "a CR will lead to computer problems that will impact care time," and 53.8% agreed that "technology interventions such as the clinical reminder increase workload," indicating that barriers to the use of CR persist.
Frequencies of pre- and post-chart reviews for the number of prescribers and home medications addressed at discharge are found in Table 1. There was little change in the number of prescribers preimplementation and postimplementation. Most charts reviewed had one prescriber for medications, 42% preimplementation (n = 21) and 50% postimplementation (n = 25) of CR.
Frequencies on completion of MRs pre- and post-chart review are found in Table 2. After the implementation of the CR, completion of MRs improved. Sixty-eight percent of MRs (n = 34) were completed before the CR compared with 92% (n = 46) after implementation. Further analysis revealed that the mean (SD) age of patients pre-CR was 81.09 (9.18) years for those with MR completed and 80.00 (8.19) years for those who did not have a completed MR. However, post-CR, the mean (SD) age was 78.65 (9.00) years with MR completed and 67.50 (2.12) years for those without a completed MR. In other words, after implementation of the CR, patients at a younger age had fewer completed MRs.
Discussion
Using the EHR for implementation of CRs is supported by other studies that have implemented technology to improve patient outcomes. Avery and colleagues18 conducted a double-blind randomized controlled trial using a pharmacist-led IT intervention for medication errors (PINCER) to determine its effect on medication errors. If a prescribed medication conflicted with a diagnosis or other medications the patient was taking, PINCER was used to alert providers. Participants assigned to the PINCER group experienced a 50% decrease in adverse effects.18
Clinical reminders are requested by healthcare professionals to improve continuity of care and patient outcomes. In a study of 247 general practitioners (GPs), a concern that emerged among the GPs was the follow-up care of chronic diseases and multiple disease conditions. The GPs even requested CRs to be added to the software program to guide them when providing care to the patients.11 While CRs are recognized as valuable, barriers to their use may persist, as evidenced by 53.8% of nurses agreeing that technology interventions increase workload and 38.4% agreeing that care time may be affected by the occurrence of a CR. These findings are similar to those of other studies, which found that nurses' attitudes toward technology innovations and the impact the innovation has on workload are a major factor in successful implementation.19,20
After implementation of the CR in this study, a decrease was noted in the mean age of patients who did not receive a completed MR (mean [SD], 67.50 [9.0]) compared with patients who received a completed MR (mean [SD], 78.65 [2.12]). This finding could be a result of fewer prescribed medications or a shorter length of stay with less time to reconcile medications among patients at a younger age.
Limitations
Increasing the number of charts reviewed pre- and post-CR implementation may have yielded data of further statistical significance. Data collection may have been affected by seasonal variations in illness as post-CR data were collected during autumn when influenza was becoming prevalent. It was also noted on several charts that the first medication entered had a prescriber listed, while the remainder of medications did not, making it difficult to determine whether the prescriber was the sole prescriber for all the medications or only the first medication listed. Since there was no penalty for failure to complete a reconciliation on each medication, nurses may have reconciled only the first medication and assumed all medications were ordered by this prescriber. With 38.4% of participants agreeing that computer problems associated with CR would affect care time, there was no place for them to identify specific computer concerns. Therefore, these problems could not be addressed.
Conclusions
This study represents an effort to improve MRs among those 65 years and older and thus improve patient safety among patients admitted to a healthcare facility. The implementation of a CR into the existing software at the rural hospital occurred at no cost to the facility, which is important for healthcare facilities that are struggling to meet meaningful use requirements. The continued sustainability of the CR practice change has been ensured by the collaborative efforts of nursing and quality and assurance staff at the hospital. This quality improvement study provides evidence that the use of CRs implemented in EHRs can improve MRs and ultimately lead to improved patient outcomes. While the importance of MRs is well documented, barriers to the use of MRs should be addressed as technology continues to evolve. Clinical reminders are an easy, cost-effective way to prompt healthcare providers to perform important tasks such as MR.
References