compose a paper

FRONTLINE PHARMACIST

270 AM J HEALTH-SYST PHARM | VOLUME 73 | NUMBER 5 | MARCH 1, 2016

The Frontline Pharmacist column gives staff pharmacists an oppor- tunity to share their experiences and pertinent lessons related to day-to-day practice. Topics include workplace innovations, coop- erating with peers, communicating with other professionals, dealing with management, handling technical issues related to pharmacy practice, and supervising technicians. Readers are invited to submit manuscripts, ideas, and comments to AJHP, 7272 Wisconsin Ave- nue, Bethesda, MD 20814 (301-664-8601 or [email protected]).

Implementation of active surveillance in electronic health records at pediatric institutions

Pharmacotherapy services provided by clinical pharmacists within the hospital setting have demonstrated optimal patient care outcomes, and the use of active surveillance programs has helped to improve the care provided to patients.1 Active surveillance consists of a set of process- es for the continued systematic compilation, analysis, and interpretation of data on benefits and harms. These active surveillance processes help to identify, evaluate, and communicate pre- viously unknown effects of healthcare products, or new aspects of known effects. The aim of the process is to harness any beneficial effects and prevent or mitigate effects that may cause harm.2 Some current forms of active surveillance that are widely accepted and used by hospital pharmacists include the reporting of adverse drug reactions and the optimization of antimicrobial therapy through an antimicrobial stewardship program. The benefits of active surveillance have been shown through the use of such programs and others, such as the U.S. Vaccine Adverse Event Reporting System and institution-specific programs targeting pharmacokinetic drug monitoring and drug dos- ing in organ dysfunction.3-7 However, many of these pro- grams are not integrated into a single application, making a comprehensive evaluation cumbersome. Fortunately, with newer, widely utilized electronic health record (EHR) systems, there is the potential for improvement in active surveillance programs and their applicability in pharma- ceutical care.

The implementation of an active surveillance pro- gram in a pediatric population is particularly difficult, because all necessary components of a variable popula- tion must be generalized into a single program or process. One of the biggest obstacles in developing a pediatric- focused active surveillance process is the need to ac- count for the changes in pharmacokinetic and pharma- codynamic parameters associated with the growth and

development of this population.8 The clinical pharmacy group at the Children’s Hospital of Philadelphia custom- ized an electronic program based on its pediatric patient population to provide an improved and more compre- hensive approach to the prioritization and monitoring of a patient’s drug therapy.

Background. The Children’s Hospital of Philadelphia is an urban, tertiary care hospital with 535 beds (203 intensive care and 46 surgical), a level 1 trauma center, and a level 3 neonatal intensive care unit. At the time of writing, 8 postgraduate year 2 specialty trained pediatric clinical pharmacists provided care for this medically complex population. The clinical pharmacy team previously used a paper-based documentation system to provide ongoing patient monitoring of therapeutic drug levels and organ dysfunction and to notate other pertinent findings in the ongoing care of patients. This paper documentation was then passed among the clinical pharmacist staff to facil- itate cross-coverage in times of colleague absence. This tedious, time-consuming, and potentially error-prone approach relied on the clinical pharmacist to ensure that all laboratory values were transcribed properly, with no pertinent values omitted. A user-friendly, interactive, and customizable tool was necessary to help prioritize patient care, provide the data necessary for continual pharmacokinetic drug monitoring, identify and assess organ dysfunction, and provide alerts for patient- specific factors that require assessment by a clinical pharmacist. With an average patient: clinical pharmacist ratio of 60:1, a decision was made to take advantage of a new process that would be integrated into the EHR and

FRONTLINE PHARMACIST

AM J HEALTH-SYST PHARM | VOLUME 73 | NUMBER 5 | MARCH 1, 2016 271

allow for the efficient and effective provision of quality pharmaceutical care.

Methods. An existing tool within the integrated EHR—the Electronic Pharmacy Acuity System (ERxAS, Epic Systems)—was customized to take the place of the previously descr ibed paper-based system of active surveillance. The ERxAS allows for real- time, active collection and analysis of data entered into the EHR. To customize this system, the clinical pharma- cy group, comprising pediatric specialists from general pediatrics, intensive care (cardiac, neonatal, and pedi- atric), oncology, solid organ transplantation, and drug information, came to a consensus regarding the clinical criteria that were most relevant to the provision of efficient pharmaceutical care. An analysis of the selected criteria within a patient’s medical record are presented in a single screen for the clinical pharmacist to review (Figure). The criteria included considerations suggestive of organ impairment (acute rises in or elevated serum creatinine [SCr] level, abnormal creatinine clearance, elevated Child-Pugh score), drug level results requiring assessment by a clinical pharmacist (antimicrobials, anticoagulants, anticonvulsants, immunosuppressants, antineoplastics, and cardiac medications), active anticoagulant orders, pertinent patient-specific problems (positive serum or urine human chorionic gonadotropin, ketogenic diet, Q-T interval prolongation, renal replacement therapy,

extracorporeal membrane oxygenation), and active pharmacy intervention notes (ongoing pharmacokinetic monitoring assessments, active organ impairment or patient-specific alert assessments, or nonspecific phar- macist intervention assessments).

Each identified clinical criterion was assigned a point value based on its potential impact on a patient’s pharmacotherapy. The goal of assigning points was to identify those patients with a higher acuity, as they likely require a greater level of attention and clini- cal pharmacist review and intervention. The clinical pharmacy group chose the department-specific val- ues and weighting based on existing values prepop- ulated in the system as purchased from the vendor and then altered the point value based on the opin- ion of the clinical pharmacy staff. For example, one clinical criterion is the presence of a drug level for a predetermined group of medications. When a drug level is reported in the patient’s chart, it triggers the ERxAS to assign a point value determined by the re- sult (i.e., more points are assigned for levels outside of normal limits and fewer for those within normal limits).

ERxAS features. The ERxAS list is displayed in a table for- mat with the patient’s total score appearing first (Figure). The system highlights a patient’s ERxAS score with one of three colors to designate the patient’s level of acuity—high

Figure. A screenshot in the Electronic Pharmacy Acuity System, or ERxAS. The clinical criteria are displayed across the top of the screen, with individual patient total scores shown on the left. The change in total score since the previous review along with the time since that review is displayed side by side. The components of a patient’s total score are shown within the individual contributing criterion. Below the patient scores, the patient-specific comment box is shown with examples of information that can be shared among clinical pharmacists.

FRONTLINE PHARMACIST

272 AM J HEALTH-SYST PHARM | VOLUME 73 | NUMBER 5 | MARCH 1, 2016

(red), moderate (orange), or low (yellow)—to direct the attention of the clinical pharmacist to patients with the highest acuity first. Another column indicates the change in a patient’s ERxAS score since a clinical pharmacist last reviewed the profile; the time that has elapsed since that review occurred is presented in another column. These data notify the clinical pharmacist of acute changes in a patient’s profile and allow for the rapid identification and assessment of changes. This information also assists the clinical pharmacist in maintaining a consistent workflow by allowing for easier identification of patients who still require an assessment. Furthermore, these data serve as a notification to other clinical pharmacists that the profile has been reviewed and does not require further attention, which prevents duplication of work. Other data specific to the ERxAS include patient identifiers and the scoring criteria comprising the patient’s total ERxAS score (Figure). The scoring criteria are separat- ed into individual columns to easily identify the rea- sons for an elevated total score. Any column, including the total score, can be sorted from highest to lowest to help prioritize patient review.

A detailed view of the patient’s overall score (the ERxAS profile) is displayed when a specific patient is se- lected, which shows the breakdown of the patient’s indi- vidual scoring components. For example, if a patient has a serum vancomycin level within normal range, a score would be triggered in the ERxAS and a point value would populate in the “Drug Level” column on the ERxAS list, causing the patient’s total ERxAS score to rise. While the ERxAS list only notates a drug level, the detailed ERxAS profile will provide the reason for the patient’s score such as a serum vancomycin level. By having all of this information readily available in a single window di- rectly linked to the patient’s EHR, the ERxAS helps aug- ment the efficiency of the clinical pharmacy staff.

The ERxAS profile also was constructed to allow the clinical pharmacist to input information in a patient- specific comment box to facilitate internal communi- cation and transition of care between clinical pharma- cists. This communication method allows one clinical pharmacist to cross cover for another with a basic un- derstanding of the pertinent patient-specific pharmaco- therapy information included in the comment field. For example, a patient who undergoes cardiac bypass and sustains a notable elevation in SCr will have an ERxAS score assigned based on this change. The clinical phar- macist who identifies cardiac bypass as the reason for the elevated SCr level can note this in the comment box as the causative factor for the increased ERxAS score and the clinical significance of the elevation.

Numerous features of the ERxAS streamline patient assessment by the clinical pharmacists. The ERxAS is accessed within the hospital’s EHR system, enabling the use of a window-in-window feature for faster iden-

tification of pertinent data without leaving the EHR. There is no need to access multiple programs in multi- ple windows, as all necessary data can be found in the EHR. Real-time analysis of each patient is made possi- ble through the continual updating of patient data, de- creasing the chance of an error due to delayed or out- dated data. Further, the ERxAS allows users to group patients, such as those on a specific unit, into separate lists in order to reduce search time.

Four electronic reports are generated by the EHR throughout the day (one to identify patients taking an- ticoagulant medications and three to report all drug levels of medications requiring pharmacokinetic moni- toring). These reports are used by the clinical pharmacy staff in combination with the ERxAS to reduce the risk of omission of patients requiring a clinical pharmacist’s as- sessment. The reports also serve as a redundancy in the case of an ERxAS reset or downtime, helping provide the greatest level of surveillance for patients.

Results. The ERxAS system was implemented in Janu- ary 2011 and fully replaced the paper documentation system. From January to June 2014, the clinical phar- macy staff used the ERxAS to identify and assess 6741 medication levels. In addition, 61 courses of heparin therapy and 51 courses of warfarin therapy were iden- tified, monitored, and adjusted. A total of 732 organ impairment cases were identified and assessed in 514 patients, with further ongoing assessments provided until the patients were discharged. Altogether, this equates to a mean of 41 patient interventions daily. It is important to note that these numbers represent a conservative estimate, as many organ impairment cases and anticoagulant treatment courses necessitate additional ongoing clinical pharmacist interventions, but only the initial identification and assessment were captured in our reporting.

Potential barriers to implementation. There are some potential barriers preventing other institutions from utilizing an electronic active surveillance system. Most importantly is an EHR system that is able to support a similar scoring system. If an institution’s EHR system can support an electronic active surveillance system, the time necessary to develop criteria, customize the program, and educate the staff on its use could be an- other potentially major barrier. This barrier is likely due to a lack of familiarity with the program, the ability to gain consensus on appropriate criteria, and the ability to build a template.

Conclusion. The integration of the active surveillance program into the clinical pharmacy workflow has ben- efited the care of patients at the Children’s Hospital of Philadelphia. The clinical pharmacy staff reports in- creased workflow efficiency and a decrease in wasteful redundancies, and the time saved has been allotted for

FRONTLINE PHARMACIST

AM J HEALTH-SYST PHARM | VOLUME 73 | NUMBER 5 | MARCH 1, 2016 273

other clinical activities. The system allows for hospi- talwide coverage on the weekends by a single clinical pharmacist. Finally, the customization of the ERxAS allows the program to stay current as clinical pharmacy practice advances in providing optimal pharmaceutical care.

1. Anderson SV, Schumock GT. Evaluation and justification of clinical pharmacy services. Expert Rev Pharmacoecon Outcomes Res. 2009; 9:539-45.

2. Aronson J, Hauben M, Bate A. Defining “surveillance” in drug safety. Drug Saf. 2012; 35:347-57.

3. Jha A, Laguette J, Seger A, Bates D. Can surveillance systems identify and avert adverse drug events? A prospective eval- uation of a commercial application. J Am Med Inform Assoc. 2008; 15:647-53.

4. Di Pentima M, Chan S, Eppes S, Klein J. Antimicrobial prescription errors in hospitalized children: role of antimi- crobial stewardship program in detection and intervention. Clin Pediatr. 2009; 48:505-12.

5. Haber P, Patel M, Pan Y et al. Intussusception after rotavirus vaccines reported to US VAERS, 2006-2012. Pediatrics. 2013; 131:1042-9.

6. Ratanajamit C, Kaewpibal P, Setthawacharavancih S, Faroonsarng D. Effect of pharmacist participation in the health care team on therapeutic drug monitoring uti- lization for antiepileptic drugs. J Med Assoc Thai. 2009; 92:1500-7.

7. Hassan Y, Al-Ramahi R, Aziz N, Ghazali R. Impact of a renal drug dosing service on dose adjustment in hospitalized pa- tients with chronic kidney disease. Ann Pharmacother. 2009; 43:1598-605.

8. Bhatt-Mehta V, Buck M, Chung A et al. Recommenda- tions for meeting the pediatric patient’s need for a clinical pharmacist: a joint opinion of the Pediatrics Practice and Research Network of the American College of Clinical Phar- macy and the Pediatric Pharmacy Advocacy Group. Phar- macotherapy. 2013; 33:243-5.

Neil Patel, Pharm.D. Pediatric Oncology

Michael Reedy, Pharm.D. [email protected]

E. Zachary Ramsey, Pharm.D. Pediatric Cardiology/Cardiac Intensive Care Unit

Department of Pharmacy Services The Children’s Hospital of Philadelphia Philadelphia, PA

The authors have declared no potential conflicts of interest.

DOI 10.2146/ajhp140887

Copyright of American Journal of Health-System Pharmacy is the property of American Society of Health System Pharmacists and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.