Among the many benefits of a robust medication reconciliation program—including increased standardization and consistency in treatment, improved documentation, and reduced health care costs—the capacity to reduce medication errors during transitions of care is clearly the most significant. Because medication reconciliation has a clear impact on increasing patient safety, it  has been a Joint Commission Patient Safety Goal since 2006.¹ However, developing and implementing effective medication reconciliation programs remains challenging for many hospitals. One effective option for maximizing medication reconciliation is to train pharmacy technicians to perform this function during the patient admission process. 

Research has shown that 54% of patients have at least one unintended discrepancy on their admission medication orders.² Although most of these discrepancies are minor, 39% have the potential to cause moderate to severe discomfort or clinical deterioration.² These ADEs may contribute to increased morbidity and mortality, prolonged hospital stays, and increased readmissions after discharge. Far from being a perfunctory administrative task, medication reconciliation should be a patient-centered process that supports optimal medication management in the context of a complex health care system with numerous health care providers. 

Lehigh Valley Health Network (LVHN), located in Allentown and Bethlehem, Pennsylvania, includes two full-service hospitals and a children’s hospital. To strengthen our medication reconciliation process upon admission, in 2012 LVHN created the new position of medication reconciliation technician (MRT)—pharmacy technicians who are specially trained to perform only one function—collecting an accurate medication list upon patient admission. As patients are admitted to the hospital, the MRT conducts a patient interview to gather medication information and develops a list of pre-admission medications, which the provider then reviews prior to the initial consultation. The provider reconciles this list and the MRT closes the loop by comparing the reconciled list with the provider’s admission medication orders. 

The Challenge of Medication Reconciliation
One reason hospitals struggle with medication reconciliation is that the information required to compile an accurate medication list is not available in a centralized database; rather, it must be gleaned from consultation with numerous sources of varying quality. In addition to the patient’s medication list, information may be provided by family members, nursing homes, outside providers, and pharmacies. Compounding the problem, patients may be taking a dozen or more medications prescribed by multiple providers. For these reasons, outdated information often remains on patient lists. In a study of one unit at LVHN, 98.3% of patients had incorrect data on their medication list upon admission. 

Developing a New Approach
Prior to the development of the MRT program, the process for medication reconciliation at LVHN included the duplication of efforts and had a high potential for miscommunication. The nurse and provider collected separate—and sometimes conflicting—medication histories from the patient. The providers relied on the history they collected to write the orders, while nursing checked against their own list. Using this process, it was not possible to determine if discrepancies were deliberate changes or simple mistakes. The resulting error rate was 45.7% and the majority of providers and nurses were dissatisfied with the process (see FIGURE 1).

Click here to view a larger version of this Figure

Instead of reevaluating individual steps in our fragmented process, we chose to define a new procedure and implement it all at once. Physician support for overhauling our approach was critical to our success. Progress at LVHN began in earnest when the vice-chair of medicine brought physicians to the table with the nurses and pharmacists, so that each health care provider would fully understand their role in the new medication reconciliation process. 

We decided to focus our efforts on hospital admission and re-defined medication reconciliation as a three-step process: 

• After a patient is admitted, one person obtains a medication list with the goal of obtaining the best possible medication history 
• A provider (physician or physician assistant) enters the orders
• At the time of order entry, the provider reconciles the medications with the original medication list 

This streamlined approach saves time, minimizes miscommunication, expedites documentation in the electronic medical record (EMR), and generates higher quality data. 

Critical to the success of this approach is compiling the most accurate admission medication list possible. To this end, we defined the actions required to obtain a thorough medication history. Specifically, we stated that appropriate personnel would review pre-hospital medications and confirm the medication use and dose with at least one validated information source, such as the patient, a family member, pharmacy, or the patient’s primary care physician. 

Staffing Considerations
To obtain the most accurate medication lists, it was necessary to dedicate specific staff to procuring this information. Therefore, we sought and obtained approval for nine new pharmacy technician FTEs dedicated solely to medication reconciliation. While trials at other hospitals have used nurses, pharmacists, and pharmacy technicians for medication reconciliation, we decided it would be most cost-effective to hire pharmacy technicians as MRTs and train them specifically for this function. 

Training MRTs
Among the most important considerations for hiring MRTs was to identify candidates who had direct experience working with people and who would be comfortable working with patients face to face. This quality is important to identify in potential MRT employees, as it is not unusual for many hospital-based pharmacy technicians to have no experience interacting with patients. Some patients may be intimidated by the health care system and may not be truthful about their medications. During the month-long training period for the first group of MRTs, we taught strategies for asking open-ended questions, eliciting information, and investigating inconsistencies in patients’ reports to get to the facts. The MRTs also received training on the top 200 drugs used in our hospital. 

An extensive training checklist was developed to establish MRT competency in using electronic systems to gather medication information available from physician offices in our network, as well as prior hospitalizations (see SIDEBAR. To view the checklist in its entirety, see http://pppmag.com/MRTchecklist). MRTs collect the name of a patient’s primary care office and the pharmacies they use. For patients on warfarin, the MRTs specifically ask which doctor controls their warfarin. Because warfarin dosing can change quickly for hospitalized patients due to sickness, diet modifications, or the introduction of new medications, we have found this part of the continuum of care to be particularly critical. 

Click here to view a larger version of this Sidebar

We use a hybrid system for documenting medication reconciliation activities. The MRTs document the medication history electronically, and also deliver a paper copy of the list to the providers for the actual reconciliation. The providers mark each medication on the list as continue, discontinue, or change. The MRTs then input the decision in the EMR and compare it with the provider’s orders. If there are any discrepancies, the MRT contacts the provider for clarification. Because provider involvement in the process is minimal, very little training for the providers or nurses has been necessary.

Emergency Department Pilot Program
Our MRT pilot program began in September 2012 (see FIGURE 2). MRTs were stationed in the emergency department (ED) from 9AM to 9PM. It quickly became apparent that one of the most significant challenges would be identifying which patients ultimately would be admitted. For example, beginning the process once a bed was assigned was often too late as the physician may already have entered an order. Therefore, MRTs work closely with the triage nurses and collect medication lists for patients who meet criteria that indicate likely admission, including age and chief complaint. They also scan the electronic board for patients marked as intend to admit who do not meet our criteria. 

Click here to view a larger version of this Figure

At the end of the six-week pilot period, the medication discrepancy rate dropped from 45.7% to 2.5%. This reduction was due to multiple factors, including having one person take ownership of the process, ensuring that person understands the medication use process, and has a firm understanding of the medications given to patients.

Results
Currently, the rate of medication discrepancies is 4.3%, and is monitored monthly. Based on the estimated cost of ADEs (calculated by comparing the variable costs of patients who had a documented medication error compared with the costs of patients who did not—at LVHN this is approximately $1100), LVHN is saving about $1,736,738 per year due to the more robust medication reconciliation program. Data indicate that readmissions have been reduced by 3–7%, depending on the unit. Interestingly, in cases where the MRT obtains a medication history but the provider does not complete the reconciliation, readmission rates still are lower. These results suggest that the thorough medication history performed by the MRTs is key to the program’s success. Moreover, by reducing readmissions, the MRTs have saved $150,000 in excess of their salaries. We have seen a half-day reduction in length of stay, but this result is not statistically significant, likely due to the sample size, and hospital admission costs are unchanged. Each medication reconciliation takes an MRT about 18 minutes. Thus, LVHN would requires 10.9 FTEs to obtain medication histories on every patient. However, because our program has not been fully implemented, we are not yet capturing medication lists for every patient. We chose to focus initially on the ED; as such, we are not currently catching admissions that come directly from a physician’s office or from pre-admission testing. 

Surveys of the providers and nurses show that they are highly satisfied with the MRT program. Providers have more time to spend interacting with patients now and need not be concerned with obtaining the medication list. While the nurses must still review the medication list, they are no longer responsible for generating one and attempting to resolve any discrepancies. By freeing the nurses to perform other tasks, LVHN realizes an estimated $780,000 in soft savings.

In 18 months, LVHN will transition to an EMR that can pull information from an electronic network containing patient data, such as prescription history and insurance information. We anticipate that this change will significantly increase the speed and accuracy of the medication reconciliation process. The MRT will begin the process by generating a medication list using the network, then add information from other sources, and create the admission medication list in the EMR. Providers will rely on this list to create the admission medication order, completing the reconciliation during this step. Thus, providers will no longer need to reconcile the list on paper or re-transcribe the list to enter their orders. Our new EMR also will alert a user who has entered the wrong unit of measurement, which is not possible using our current system.

Future Plans
Our future goals include moving resources to the patients at highest risk of harm due to the potential for medication errors; we are incorporating the psychiatric population into the system, and then we plan to increase our involvement with heart failure, anticoagulation, and diabetes mellitus patients. MRTs also will perform follow-up telephone calls to anticoagulation patients to ensure the patient is compliant with their medication, lab testing, and follow-up physician visits.

Conclusion
Use of pharmacy technicians in a streamlined medication reconciliation process has dramatically reduced discrepancies between the patients’ pre-admission medications and their medication orders at LVHN. The result is a cost savings of more than $1.7 million in averted ADEs, a reduction in readmission rates, and increased staff satisfaction. The support of hospital administrators and physicians was critical to the implementation of this program. We look forward to the future, as our goal is to roll out the program to capture admissions in all areas of the hospital.

References

1. Boockvar KS, Santos SL, Kushniruk A, Johnson C, Nebeker JR. Medication reconciliation: barriers and facilitators from the perspectives of resident physicians and pharmacists. J Hosp Med. 2011;6:329-337. 
2. Cornish PL, Knowles SR, Marchesano R, Tam V, Shadowitz S, Juurlink DN, Etchells EE. Unintended medication discrepancies at the time of hospital admission. Arch Intern Med. 2005;165:424-429.

Leroy Kromis, PharmD, BCPS, is the medication safety officer at Lehigh Valley Hospital. Prior to joining Lehigh Valley, Leroy worked as a clinical pharmacist at Duke University Hospital. He obtained his BS in pharmacy and PharmD from the University of North Carolina at Chapel Hill. Leroy has a strong interest in critical care and health care quality, and is a lecturer at DeSales University and serves on the adjunct faculty for Wilkes University and the University of the Sciences in Philadelphia.