Study: Assigning Pharmacy Staff to Take Drug Histories in ED Cuts Medication Errors
A persistent problem for hospital emergency departments is medication errors due to incomplete information or miscommunication between providers and patients. A research study led by Cedars-Sinai in Los Angeles found that when pharmacy professionals — rather than doctors or nurses — take medication histories of patients in emergency departments, mistakes in drug orders can be reduced by more than 80 percent. Acting on the findings, Cedars-Sinai now assigns pharmacy staff members to take medication histories for high-risk patients admitted to the hospital through the ED.
In the study, published in the journal BMJ Quality & Safety, the investigators focused on 306 medically complex patients who were taking 10 or more prescription drugs and had a history of heart failure or other serious conditions. The study also involved investigators from Western University of Health Sciences in Pomona, California; University of Toronto and University Health Network in Toronto; the VA Greater Los Angeles Healthcare System; and UCLA.
In their preliminary data working with this population, the Cedars-Sinai researchers found up to seven errors per patient in the medication histories. “Not all of those are clinically significant errors,” said Joshua Pevnick, M.D., associate director of the Division of Informatics at Cedars-Sinai and the study’s first author. “If you forget a multivitamin, it is probably not going to have a huge impact on their health. But there are errors that are clinically important for patients.”
The study found that when pharmacists or pharmacy technicians, instead of medical staff, took these patients’ histories in the Cedars-Sinai Emergency Department, errors in the histories fell drastically, and as a result, significantly fewer drug-order errors were made during hospitalization.
Pevnick, who is also an assistant professor of medicine at Cedars-Sinai, explained the context for the study. These older, sicker patients tend to see multiple physicians, who are often using different kinds of EHR. “Even when people are using the EHR, and even when there is data exchange, the people aren’t recording everything that happens,” he said. “You may get a call at night that goes to a partner who tells the patient to stop taking a medication. That may not get updated in the EHR.”
Time of day of admission is also an issue. Most hospital admissions tend to happen in the late afternoon to evening. The patient’s pharmacy may be closed, if it is a mom-and-pop-type pharmacy, and the physician’s office may be closed. “A lot of times the hospitalist will see the patient after the family has left and sometimes they have taken with them the bag of medications that the person brought in,” he explained. “So placing these pharmacy personnel in the ED allows them to get to the patient frequently when the family is still there and the medications are still there, and the physician’s office and pharmacy are still open.”
Pevnick also said that having someone with a pharmacology background who has the time and the assignment of checking medication history seems to make a big difference. The usual care process is that everyone who touches the electronic medical record should check medications, he noted. “But since it is everyone’s job, you also have the risk of relying on the EHR without checking it that carefully, figuring that people before you have done a good job. So we are trying to assign the task to one person and say to them, ‘I know there is already information in there, but we really want you to carefully confirm everything in there.’”
Cedars-Sinai now assigns pharmacy staff to take medication histories for certain high-risk patients who are admitted to the hospital after first seeking treatment in the ED. It plans to expand the effort by providing pharmacy staff reviews of medications for a wider range of patients in the Emergency Department and inpatient areas.
But freeing up those employees’ time is a challenge, Pevnick admitted. Pharmacists are really valuable in the healthcare system and there are many places where they can contribute. “Part of the idea of this study was to quantify the benefit they are providing in this situation,” he explained, “so that can be weighed against other areas where they can provide value.”
Pevnick also noted that Cedars-Sinai is a large medical center with a lot of volume in the ED, so it can justify having pharmacists or pharmacy technicians stationed in the ED. It may not work as well at hospitals that don’t have as much volume in this type of patient population.
The increasing flow of patient medication data between organizations will make this task easier, he said. For this particular study, they did not access Surescripts data, but in a separate study they did look at the potential benefit of Surescripts data and found there was good error prevention potential for patients who had Surescripts data available. (That study found that Surescripts data probably would have prevented 35 percent of admission medication history errors and 31 percent of resultant inpatient order errors.)
“Since that is nationwide and there is Surescripts data for a lot of patients, that is the first step,” he said. “And for a lot of EHRs it comes right into the workflow.”
Cedars-Sinai also takes advantage of health information exchange between organizations. It tends to require going outside the workflow and is lower yield, but can be incredibly helpful. “I think more and more we will have better data from more places, and it will be coming right into the work flow so it will be easier to incorporate.”