Las Vegas—Three very different problems in patient care—the overuse of growth factors in chemotherapy, medication mishaps caused by drug cabinet stocking errors and a lack of quick access to key data during clinical rounds—have a shared solution: the innovative use of pharmacy technology, according to poster presentations at the American Society of Health-System Pharmacists’ (ASHP) Midyear Clinical Meeting.
 
A Better System for Stocking ADCs
Each month, the pharmacy at Baystate Medical Center, a 653-bed tertiary care facility in Springfield, Mass., dispenses about 800,000 IV and oral medication doses. In late 2008, the hospital instituted medication and patient ID bar-code scanning at the bedside, and the number of medication errors that reached patients decreased dramatically, according to Erin Taylor, PharmD, assistant director of inpatient pharmacy services at Baystate. “But we noticed that although actual errors decreased, the number of near misses rose considerably—that is, incidents when the nurse discovered the mistake before administering the drug.” The reason: Pharmacy technicians sometimes stocked the wrong drug or the wrong strength into the automatic dispensing cabinet (ADC) drawers located on the patient care floors. “Nurses were catching the errors, so you could say the system worked,” Dr. Taylor said. “But as a department, we saw an opportunity to add another safety barrier to further decrease the chances of the wrong medicines reaching patients.”
 
In October 2010, the pharmacy instituted a protocol that stipulated scanning of all medications when they were loaded into the ADC. The goal was to attain a 90% scan rate among the technicians as they stocked the cabinets. At first, ADC scans were completed only 30% of the time, but the rate jumped to 90% after only four months, and now averages 93%, according to Dr. Taylor. As a result, near misses caught during bedside bar-code scanning by nurses dropped by 70%. In terms of absolute numbers, she added, dispensing errors went from 20 events per 1 million doses dispensed to six per 1 million.
 
Based on those results, she noted, “we know our patients are safer than they were before we established this procedure.”
 
Mark Heelon, PharmD, Baystate’s medication safety coordinator, emphasized the role pharmacy technicians played in the quality improvement initiative. “What really helped this program to be successful was that the technicians were strongly vested in the outcomes and wanted to make it work,” he said. “They were involved in much of the decision making as we developed the program. We told them, ‘We have a problem, we know we can do better and we need your expertise and devotion to patient safety to devise strategies to reduce these errors,’ and they came through.”
 
G-CSF and Hodgkin’s Lymphoma
Granulocyte-colony stimulating factor (G-CSF) is commonly used to maintain dose intensity and avoid treatment delays in patients who develop neutropenia caused by ABVD (adriamycin, bleomycin, vincristine and doxorubicin) chemotherapy. But a growing body of research has documented that the multidrug regimen can be safely administered without the routine use of G-CSF. Given the growth factor’s high cost and propensity for potentiating bleomycin-associated lung toxicity, pharmacists at Kaiser Permanente Northwest, in Portland, Ore. knew they needed a method for reducing G-CSF use, particularly in patients with Hodgkin’s lymphoma.
 
Their solution: embedding therapeutic messaging about the proper indications for G-CSF into its electronic medical record (EMR).
 
“We noticed that we were using a good deal of G-CSF in our Hodgkin’s disease population that received ABVD, even though [National Comprehensive Cancer Network] guidelines clearly suggest it isn’t recommended,” said Jasen Knudsen, PharmD, BCPS, BCOP, an oncology/hematology pharmacist at Kaiser. “We wanted to align ourselves with the most current treatment guidelines without compromising care.”
 
The embedded messages pop up on the computer screen as prescribers move through the drug-ordering process and are considering G-CSF use. The messages include links to additional, evidenced-based references. In general, they state that patients may be treated at full ABVD dose without G-CSF, on time, regardless of hematologic counts. Dr. Knudsen and his colleagues hypothesized that the prompts in the system (KP HealthConnect, a version of EPIC) would persuade clinicians to be more aggressive in maintaining dose intensity and not delay treatment or reduce chemotherapy dose. Based on their pilot study, they were right.
 
The study was based on a retrospective chart review of adult patients who were treated before and after therapeutic messaging about the updated ABVD protocol appeared in the EMR. Before embedded messaging, 11 of 18 (61%) patients received G-CSF compared with three of 20 (15%) patients after the messaging. Moreover, full-dose intensity was maintained in 12 of 18 patients before the change to all 20 patients after the start of message embedding.  No cases of febrile neutropenia or bleomcyin-induced pulmonary toxicity occurred.
 
In addition to the reduced G-CSF use, “we also found that updating the EMR protocol resulted in a change in clinician workflow, with fewer treatment deferrals and therapy delays, which we didn’t expect,” Dr. Knudsen noted. “But that was certainly a good [outcome].”
 
Dr. Knudsen acknowledged that the study’s small sample size limits definitive conclusions about the efficacy of message embedding. His study team plans to examine a larger population drawn from pooled data of several Kaiser Permanente regions.
 
As for why so many prescribers at Kaiser in this initial study seemed to favor G-CSF therapy before message embedding, “significant practice changes take some time for clinicians to fully adopt,” Dr. Knudsen said—a message that itself should be heeded by any hospital undertaking similar quality improvement efforts.
 
iPad Can Help With Data Access
Clinical pharmacists at Boston Children’s Hospital routinely participate in daily rounds with multidisciplinary care teams, and until recently, all had been equipped with laptops, which allow immediate access to pharmacy order systems, the EMR, and other key information. But the laptops, weighing in at a hefty five pounds, proved to be somewhat unwieldy for pharmacists on the move, according to Tsingyi Koh-Pham, PharmD, BCPS, a clinical pharmacist. To lighten their load and speed access to crucial patient care data, rounding pharmacists were given iPads and surveyed on their experiences with the devices.
 
In a pilot trial, 10 of the devices were given to rounding pharmacists, who were then asked to gauge the tablet’s relative utility, how well it could be incorporated into daily work patterns and whether using the iPads enabled them to overcome some of the functional limitations associated with laptops. The response was overwhelmingly favorable and led the department to issue 31 additional iPads to all decentralized pharmacists—as well as to some centralized pharmacists—who wished to use them. Surveys of 41 iPad users were conducted in September 2011 (n=23) and again in June 2012 (n=25). In the first survey, 70% of respondents said they were extremely or moderately satisfied with their iPad experience. Ninety percent responded similarly in the second survey. Dr. Koh-Pham attributes the increase in satisfaction to growing levels of comfort and proficiency with the device over time. The device’s connectivity was another key benefit that contributed to its high marks—specifically, its ability to link seamlessly to the hospital’s Cerner EMR, Dr. Koh-Pham said.
 
Additional benefits of the device include much longer battery life than most laptops, Dr. Koh-Pham said. The devices also cost about one-third as much as a laptop: $562 versus about $1,560.
 
But although iPad users consistently preferred their new devices over laptops for mobility and accessibility, they found that the iPads did not necessarily increase work efficiency; an average of the two surveys showed that overall scores for ease of use favored the laptops in all categories measured, including multiple-order verification, document viewing, and responding or composing emails.
 
One comment from an iPad user represented the broader sentiment: “Verifying orders and looking up lab data and references are still definitely easier on the laptop if you have access to one. However it’s much easier to be mobile and verify orders on the iPad.”
 
Dr. Koh-Pham acknowledged that “some of the tasks that are easily completed on a laptop are more cumbersome on an iPad, particularly anything that requires complex finger movements.” Typing and multitasking on iPads, he added, “are often more difficult.”
 
Despite the disadvantages, pharmacists said that the iPads allow them to more consistently attend patient care rounds and conduct timely interventions, Dr. Koh-Pham noted. Only one pharmacist chose to forgo use of the iPad. The Pharmacy Department plans to continue supporting the device and will purchase iPad-specific medication management software.
—Steve Frandzel
 
Commentary
Karl Gumpper, RPh, BCPS, FASHP, director of ASHP’s Section of Pharmacy Informatics & Technology, offered his views on the three quality improvement initiatives detailed in the poster presentations.
 
Scanning medications prior to ADC stocking. “This research shows the next steps in the progression of bar-code technology within the pharmacy,” Mr. Gumpper said. “Many hospitals have focused on the use of bar-code technology at the point of care. The … technology can and should be integrated into other parts of the medication-use process. This work shows that there is a gain in patient safety when combined with the use of automated dispensing cabinets.” The ASHP’s Statement on Bar-Code Verification During Inventory, Preparation, and Dispensing of Medications “encourages hospital and health-system pharmacies to incorporate bar-code scanning into inventory management, dose preparation and packaging, and dispensing of medications” (http://www.ashp.org/DocLibrary/BestPractices/AutoITStBCVerif.aspx).
 
Therapeutic messaging on G-CSF. “This a good example of how clinical decision support can work to change physician practice and potentially improve patient care and outcomes,” Mr. Gumpper said. “The presentation of laboratory values in the decision-making process is much more effective than alerts alone. There may also be less ‘alert’ fatigue with therapeutic messaging.”
 
iPads and clinical rounding. “There are emerging studies using tablet devices by physicians for patient care, so it is encouraging to see the same type of uptake by pharmacy,” he said. “The need to have ready access to patient health information and drug information resources will allow us to meet the goals of the Pharmacy Practice Model Initiative” (http://www.ashp.org/PPMI).
—S.F.