An oral chemotherapy prescription-writing module grafted to a shared electronic medical record is part of a series of quality improvement efforts undertaken at Dana-Farber Cancer Institute in Boston to prevent errors in prescribing oral chemotherapy agents. While oncologists have readily accepted the module, modifications are planned to make it easier to prescribe drugs with complex dosing regimens and ensure that dose-limit warnings reflect current clinical practice, according to an article published in the Journal of Oncology Practice (JOP).¹

The prescription-writing module builds on findings from previous efforts to identify and address key vulnerabilities associated with the use of oral chemotherapy agents² and uses some of those findings to enhance the longitudinal medical record. The longitudinal medical record is described in the JOParticle as “a multi-feature electronic medical record shared with clinicians across the affiliated Partners Healthcare system, to document encounters, check laboratory and diagnostic test results, and generate prescriptions that can be printed, faxed, or transmitted electronically to pharmacies.” The Partners Healthcare system includes Dana-Farber, as well as Brigham and Women’s Hospital and Massachusetts General Hospital in Boston and affiliated community hospitals and practices in eastern Massachusetts and New Hampshire.

New features added to the existing system made it more specific for oral chemotherapy drugs. “With guidance from physician, nurse, pharmacy, and information technology colleagues, we designed several safety-oriented enhancements to the existing module to support safe prescribing of oral chemotherapy. The enhancements included the ability to order oral chemotherapy based on a fixed dose, a weight-based dose, or a dose based on body surface area,” the authors explained. The enhanced system provided a calculator to support toxicity-related dose adjustments.

A Welcomed Enhancement, Though Work in Progress

An analysis of all prescriptions for chemotherapy agents during the first 17 months after the module was introduced showed that “clinicians used the enhanced oral chemotherapy prescription module extensively. Oncologists accepted the changes without resistance or complaint,” said the authors.

“Many saw [the new module] as an enhancement,” Saul N. Weingart, MD, PhD, Vice President for Quality Improvement and Patient Safety at Dana-Farber Cancer Institute, said in an interview with The ASCO Post. Dr. Weingart is also lead author of the JOParticle. “We sent out some notification in advance that it was coming and people just used it. Clinicians are used to electronic prescribing, and we just added some bells and whistles for them to take advantage of. I think some of the changes were welcome,” Dr. Weingart said.

During the analysis period, “600 oncology prescribers entered 6,673 prescriptions into the electronic medical record for 18 different oral chemotherapies,” according to the analysis. Six drugs accounted for 83% of the total prescriptions—temozolomide, capecitabine (Xeloda), lenalidomide (Revlimid), hydroxyurea, imatinib (Gleevec), and erlotinib (Tarceva).

“The enhanced oral chemotherapy prescribing module offered clinicians the opportunity to specify the intent of therapy (curative or palliative) and display the diagnosis, replicating safety features required for infusion therapy,” the article’s authors reported. These optional fields on the prescribing module were used inconsistently. Intent of therapy was specified 13% of the time, and cancer diagnosis, 46% of the time.

A free-text field allowed prescribers to customize their prescriptions, adding information about treatment dates, dose calculations, whether patients should take medications with food, and the number of tablets of different strength that should be taken together. This field was used in 4,242 (63.6%) of 6,673 prescriptions.

The investigators concluded that modifications are needed to make it easy for prescribers to incorporate information currently provided in free-text fields. “Offering a streamlined way to include this information in the prescription may permit quicker prescription writing, facilitate complete prescriptions, and offer more comprehensive data capture,” they stated.

‘Inappropriate Alerts’

As with other prescriptions in the system, “oral chemotherapies could generate drug allergy and interaction alerts and trigger geriatric and renal dose warnings,” Dr. Weingart and colleagues explained. “We enhanced these decision support tools by creating, with the assistance of oncologists and oncology pharmacists, daily and weekly dose-limit warnings that would alert the prescriber to potential overdoses.” During the first 17 months only, dose-limit warnings were generated.

Prescriptions for seven drugs generated all 395 alerts (Table 1, page 86). Temozolomide alone accounted for 374 alerts (95%). “A clinician whose prescription generated an alert had the option of aborting the prescription attempt or completing the prescription as planned,” the JOParticle noted. Clinicians ignored most (96%) of these alerts. Only 17 (4%) of the 395 prescriptions that generated an alert were aborted (Table 2, page 86). “In retrospect, it is surprising that clinicians aborted any prescriptions. Current dosing recommendations exceeded the dose-limit warnings for the alerted medications in every case, thus generating a significant number of inappropriate alerts,” the authors noted.

“When we created the alerts, they were based on current published guidelines, but by the time the system was implemented, practice had evolved so quickly that the dose limits were out of date,” Dr. Weingart explained. “If we are providing alerts to providers, we should make sure that they are helpful, rather than being an interruption or a distraction. Other studies have shown that “in general ambulatory care, the override rate of alerts is about 90% in most settings,” Dr. Weingart added.

“Overridden alerts don’t necessarily mean the alerts were not useful,” he continued. “Sometimes the physician will decide to proceed with the prescription despite an alert because it has reminded him or her to monitor a lab parameter or to bring the patient in a little more quickly. But some of the alerts are really not appropriate, and we need to be careful about providing decision support that really doesn’t support decisions, that actually is outdated, inaccurate, or inappropriate.”

Dr. Weingart said that he was a bit dismayed that no one notified him or others in his department about the outdated dose limit alerts. “We heard nothing about it,” he said, discovering the problem only during the analysis. He speculated that maybe clinicians “are just used to getting unhelpful alerts in a variety of systems.” The analysis showed that “we needed to circle back and make sure” that the dose-limit alerts were based on “real-time clinical input,” Dr. Weingart said.

No Drug Interaction Alerts          

Although the enhanced system “could flag interactions between the oral chemotherapy medicines and any of the other medicines that are on the patient’s medication list,” Dr. Weingart said, “there wasn’t a single alert that was generated about an interaction between an oral chemotherapy agent and one of the patient’s other drugs. The alert rate in ambulatory care is around 2% to 3%, so with 6,000 prescriptions, you would expect there would be a bunch of drug interaction or drug allergy alerts that would pop up, but they didn’t. It is really quite odd. I think the most likely reason is that we haven’t discovered a lot of interactions between oral chemotherapies and other drugs.”

The lack of interactions “raises this question about whether our electronic decision support is built in a way that picks up those interactions. It is possible that there really aren’t many, or it is possible that the vendors creating these alerts just haven’t built that section out yet,” he continued.

“We have started to look at alerts generated in the chemotherapy infusion order-entry system for interactions with drugs in the patient’s ambulatory medical record, including oral chemotherapies. Those alerts often do change physician behavior,” Dr. Weingart said. Drugs that can have interactions with infusional chemotherapy agents include blood pressure medicines, cholesterol medicines, seizure medicines, and antidepressants, Dr. Weingart noted.

Other Medication Safety Updates

Hormonal chemotherapies and investigational agents are not included in the enhanced oral chemotherapy prescription module, but Dr. Weingart said he expects that investigational agents will be added in the near future. “It is complicated because we have over 600 trials in place, and each drug needs to be coded separately.” Dr. Weingart noted. “There still is no enthusiasm for incorporating hormonal therapy into this system. The thought is that there is less danger in their use, and safety features aren’t as important for this relatively benign intervention,” he added.

“The other thing that we are going to be rolling out shortly is an alert that reminds the provider to complete or update an informed consent form for oral chemotherapy. One of the best practice recommendations is that patients complete a written informed consent for oral agents in just the way they do for infusion therapy. There is some inconsistency in practice across the country, so we are trying to build in a feature that ensures clinicians get written informed consent consistently.” ■

Disclosure: Dr. Weingart and coauthors reported no potential conflicts of interest.

References

1. Weingart SN, Mattsson T, Zhu J, et al: Improving electronic oral chemotherapy prescription: Can we build a safer system? J Oncol Pract 8(6):e168-e173, 2012.

2. Weingart SN, Spencer J, Buia S, et al: Medication safety of five oral chemotherapies: A proactive risk assessment. J Oncol Pract 7(1):2-6, 2011.