ParagonRx® | Minimize Risk and Optimize Use

By: Ted Agres

A demonstration project at nine hospitals in southern Pennsylvania, aimed at improving the safe use of hydromorphone through a combination of educational programs, computerized order entry improvements and enhanced prescribing oversight, has yielded mixed results, according to its developers.

The nine-month project, developed in part by the Institute for Safe Medication Practices (ISMP), found that practitioners’ knowledge about prescribing differences between hydromorphone and morphine improved, and the number of patients who required a rescue agent or rapid response call decreased. But the percentage of cases in which concomitant medications contributed to adverse drug reactions (ADRs) increased from 20% before the initiative to 41% afterward.

“Education is not a long-term strategy to effect change,” said Matthew Fricker, RPh, ISMP’s program director, “but it’s a starting point.”

Mr. Fricker described results of the regional initiative during a Dec. 14, 2011, webinar hosted by ISMP. “We wanted to determine, if hospitals developed specific risk-reduction strategies, would we see a positive impact,” he said. Overall, staff physicians, nurses and pharmacists demonstrated a 15% to 19% improvement in their knowledge of hydromorphone equianalgesic potency, but their understanding of comorbidities did not meaningfully improve, and misinformation persisted about the effects of obesity, respiratory disease, opioid tolerance and sleep apnea. “More work must be done in these areas,” Mr. Fricker noted.

Still, there was a 51% reduction in the number of patients who required a rescue agent or rapid response call, indicating that improvements can in fact be made as a result of these types of safe medication use initiatives, Mr. Fricker stressed.

A Continuing Drug Safety Issue

Hydromorphone (Dilaudid), the morphine derivative commonly used to manage moderate to severe pain, continues to be implicated in medication errors and patient harm despite efforts to educate pharmacists and clinicians about its safe use. According to the Pennsylvania Patient Safety Authority, nearly 1,700 hydromorphone errors were reported between January 2008 and October 2009, with almost 70% of them reaching the patient. A 2006 Canadian study reported 251 deaths in the previous decade due to oral and IV overdoses with hydromorphone (J Anal Toxicol 2006;30:202-209).

“Hydromorphone continues to be associated with patient harm in many hospitals and health systems,” said Matthew Grissinger, RPh, director of ISMP’s Error Reporting Programs. “Some organizations have begun to recognize this risk but may not have developed effective strategies for reducing the potential for patient harm. Others may not be aware of the number of hydromorphone-related events occurring in their institutions because they are not visible through the usual medication event reporting system.”

Why the gap in awareness? “For the most part, bad events that happen with hydromorphone are not being reported as medication errors through the ‘traditional’ error reporting process,” Mr. Grissinger said. “For example, I’ve seen far too many events that were reported as ‘unpreventable’ adverse drug reactions when, in fact, they were more than likely preventable. We’ve seen patients get hydromorphone 2 mg and higher doses (which is more than likely inappropriately too much), the patient has a bad response to the drug (e.g., their respirations decrease dramatically) and the doctor/nurse gives the patient a reversal agent/antidote like naloxone. From my vantage point, it almost comes across like ‘this type of stuff happens all the time, so we just give naloxone and everything is better.’ But hospitals don’t realize that this is NOT okay practice and those types of events are not thought of or reported as medication errors.”

Another problem is that hydromorphone is approximately seven times more potent than morphine IV, but both drugs come in the same dosage forms and in similar packaging. “There are still a number of practitioners who believe [hydromorphone and morphine] are one and the same drugs, or that hydromorphone is a generic name for morphine,” Mr. Fricker said. “It’s complicated by the fact that both of these drugs have been available in exactly the same dosage forms. It’s easy to mix them up,” he said, adding that the use of “tall man” lettering for the painkiller (i.e., HYDROmorphone) can help reduce errors due to name confusion.

Establishing the Baseline

Participating hospitals in the demonstration project—done in conjunction with the Health Care Improvement Foundation—included Abington Memorial Hospital, Doylestown Hospital, Hahnemann University Hospital, Holy Redeemer Hospital, LifeCare Hospitals of Chester County, Nazareth Hospital, Riddle Hospital, the Chester County Hospital and Thomas Jefferson University Hospital. The process began with hospitals surveying their physicians, nurses and pharmacists to determine their baseline knowledge about proper hydromorphone dosing and administration. The hospitals also analyzed their internal databases to determine the numbers of pharmacist interventions, medication errors, ADRs and patients requiring naloxone rescue.

They found that the vast majority of ADRs occurred among acutely ill patients in medical-surgical units (76%), followed by the emergency department (11%). The lowest incidence rate (1%) was found in critical care units, but Mr. Fricker cautioned about under-reporting, especially among patients who are not well or who are not expected to improve. “That’s a common error reporting problem in critical care units,” he said.

In addition to general confusion about differences between the two drugs, the practitioner surveys and databases revealed other factors that contributed to ADRs. These involved believing that a patient was opioid-naive when he or she was not (46%), inadequate patient monitoring (34%), believing that obese patients require higher doses (30%) and failure to recognize concomitant use of potentiating drugs (19%). Still other contributing factors involved the patient’s age, presence of respiratory disease, orders being placed for excessive doses and too-short intervals between doses.

The hospitals then developed and conducted a range of educational efforts for their staff and implemented a number of other changes. These included removing higher doses of hydromorphone (2 and 4 mg forms) from patient care units, revising paper and computerized physician order entry (CPOE) order sets to list lower doses and warnings about comparative doses, removing the ability to override hydromorphone doses, reducing initial doses and eliminating the administration of frequent doses (two to three times per hour). Other interventions included requiring an independent double-check of the dose and dosage form if an override of the automatic dispensing cabinet is entered and requiring a consultation with pain management or pharmacy intervention for higher doses.

“Confusing hydromorphone with morphine, and gaps in knowledge about the potency and efficacy of hydromorphone, have frequently contributed to preventable adverse drug events described in the patient safety literature,” said Dave Ehlert, PharmD, vice president, Pharmacy Practice Resources, McKesson Pharmacy Optimization. “Therefore, it is important for hospitals and other health care settings to develop and implement effective strategies that address these issues and minimize the incidence and severity of hydromorphone-related adverse drug events,” noted Dr. Ehlert, who was not a webinar participant but who was asked to comment on the ISMP pilot project.

Christiana Care Update

“Overwhelmingly, it’s the pharmacists who catch ADRs at our institutions, and that’s pretty much the case elsewhere,” said Dean Bennett, RPh, medication safety specialist at Christiana Care Health System in Wilmington, Del. During the webinar, Mr. Bennett presented updated data from an ongoing project to improve injectable hydromorphone safety at 913-bed Christiana Hospital in Newark, and at 241-bed Wilmington Hospital, both in Delaware. The project received a Cheers Award from the ISMP in 2010 (see Pharmacy Practice News, January 2011).

When the project began in 2004, Christiana Care did not have a CPOE system and ADE data showed that hydromorphone was a contributing or causative factor in a number of moderately severe and severe events, and use of the drug was increasing. “This was very troubling for us because we knew a preference was developing for hydromorphone versus morphine,” Mr. Bennett said. “Our prescribers told us, ‘It’s just a much better drug for treating pain.’”

Christiana assembled a multidisciplinary task force to focus on the safe use of injectable hydromorphone. As at other facilities, the task force found the most concerning and risky behavior revolved around an under-appreciation of the drug’s potency. This was exacerbated by the fact that the 1 mg dose was the smallest available, equivalent in potency to about 7 mg of morphine. Despite a variety of educational efforts from 2006 to 2009, including newsletter articles, distribution of pocket cards and use of preprinted order forms to which prescribing statements had been added, the number of ADEs related to IV hydromorphone continued to fluctuate. “There was no sustained improvement,” Mr. Bennett said.

When Christiana installed a CPOE system in 2009, one of the pharmacy’s first priorities involved hydromorphone. The system was populated only with doses that fell within the hospital’s recommended ranges for opioid-naive patients (<65 years, 0.5-1.0 mg IV every four to six hours; ≥65 years, 0.2-0.3 mg IV every four to six hours). Any other dose required an active alteration. Additionally, an “acute pain module” was designed and placed within all electronic order sets needing pain management orders. The pain module included baselines and acceptable dosages for all agents. Finally, a custom alert system was embedded in the clinical decision support system to notify prescribers when orders fell outside the recommended range.

The effort bore results: The percentage of 1 mg initial doses fell from a baseline of 46% in 2005 to 34% post-education to 18% post-CPOE. During the 21-month period after CPOE was installed, the hospitals experienced only one hydromorphone-related ADR, and that was in an area that wasn’t connected to the CPOE. “CPOE allows us to hammer home recommended prescriber doses. It’s right there in their faces,” Mr. Bennett said.

In the future, Mr. Bennett said that Christiana plans to modify the order entry format to require a data entry about whether a patient is opioid-naive or tolerant. Also in the works are clinical documentation modifications to include such risk factors as sleep apnea and high body mass index. Ultimately, the system will be programmed to issue alerts based on specific risk factors and recommendations for each patient, he said.

In June 2011, in response to ongoing medication errors, the FDA approved some revisions to the prescribing information, container labels and carton labels for 1, 2, 4 and 10 mg hydromorphone products. The dosing information and warnings now appear at the front of the prescribing information. Additionally, the IV starting dose has been reduced to 0.2 to 1 mg from a previous 1 to 2 mg. Finally, container and carton information now display the generic name using tall man lettering to better distinguish it from morphine.

Concerning all of these efforts, “no one of these strategies by [itself] is going to make a huge difference,” Mr. Fricker said. “But as with any medication error reduction effort, when you start stacking them and getting additive effects, the result is greater than the sum of the parts.”

[Link to Article]