Rural hospital improves meds reconciliation via AI automation into EHR
Like other hospital executives, leaders at King’s Daughters Medical Center, a 99-bed community hospital in Brookhaven, Mississippi, 60 miles south of the state capital in Jackson, were concerned that patient records could potentially contain incomplete or inaccurate medication history data. To mitigate that risk, King’s Daughters focused on improving its medication history reconciliation process.
Discrepancies occur when different EHRs attempt to talk with each other but run into issues for a variety of reasons. One of the most challenging problems complicating the reconciliation process is that each system uses distinct nomenclature that often does not translate accurately, or at all.
“Frequently, this results in missing or indiscernible ‘sigs’ – the critical shorthand prescribing instructions that mean the difference between a patient receiving 1.0 mg and 10 mg of a medication, or ‘qhs’ being interpreted as ‘every hour’ instead of ‘nightly at bedtime,’ for example,” said Joe Farr, RN, clinical applications coordinator at King’s Daughters. “To prevent adverse drug events, staff and clinicians can spend hours on the computer or phone conferring with other providers and pharmacies to gather or fill in missing sig data.”
"The team believes that improved accuracy of medication dosage accounts for at least a significant portion of the decrease in readmissions, possibly due to a decline in post-discharge adverse drug reactions."
Joe Farr, RN, King’s Daughters Medical Center
Many current prescription routing technologies provide free-text sig information for dosing instructions rather than discrete text that is easily translatable to an EHR. Both pharmacists and clinical staff responsible for medication reconciliation during the patient triage/intake process must manually sift through the sig data to create prescriptions, taking precious time away from patient care.
“King’s Daughters was not exempt from this workflow challenge,” Farr said. “Nurses had to manually transcribe sigs from the medication history into the current visit list in the EHR at the point of care – a time-consuming process that is susceptible to human error, with potential safety implications. King’s Daughters sought to improve patient safety and streamline the medication reconciliation process by automating the transcription of sig data into the EHR.”
King’s Daughters’ leadership decided the best way to improve the safety of medication reconciliation was to automate the transcription of sig data into the EHR. Doing that meant finding a way to convert free-text into programmable data yielding discrete sig components within a patient’s medication history. The hospital hoped that doing so would reduce clicks and keystrokes, ensure a more accurate patient medication history, and reduce adverse drug events.
“For help, King’s Daughters turned to health IT vendor DrFirst, its e-prescribing partner, for implementation of an AI-powered solution that uses Natural Language Processing and machine learning to process and validate results, and codifies sigs into each facility’s standard terminology – for example, ‘by mouth’ versus ‘oral’ or ‘PO,’” Farr explained. “The automation operates entirely in the background, without clinician intervention, and uses statistical validation and clinical analysis to translate sigs in real-time.”
The new system also helps staff resolve gaps by supplying alternative drug IDs for best-case drug matching, and details for incomplete or uncommon sigs. Multiple safety checks disqualify transactions that are deemed clinically invalid. The system is designed to prefer no data to wrong data. If it determines that a sig data point poses a safety risk, it errs on the side of not transcribing it.
MEETING THE CHALLENGE
King’s Daughters’ goals were to reduce medication reconciliation clicks and keystrokes, ensure a more accurate patient medication history by converting free-text sigs into hospital-unique nomenclature, and fill gaps by deducing missing sigs.
Implementation of the automated sig-transcription required King’s Daughters’ EHR vendor to update its code to accommodate the new system, which the team accomplished quickly.
The automated sig translation system helped King’s Daughters reduce the number of incomplete or error-filled patient medication records, which in turn minimized pharmacy call-backs, workflow disruptions and patient treatment delays.
It also significantly reduced the average number of keyboard clicks required for medication reconciliation, resulting in additional staff time and cost savings.
“Based on 19,390 annual patient visits and an average of five medications per patient, the resulting time savings of 34 hours per month for clinicians, or 404 hours per year, translates into about $11,000 in recaptured nursing productivity over 12 months,” Farr noted. “This easily justifies the minimal investment of time and resources necessary to deploy the solution.”
More significantly, the new system appears to have contributed to increased patient safety and improved health outcomes, he added.
“In the first seven months following sig translation implementation, King’s Daughters’ overall 30-day readmission rate fell by 11.3%, from 6.2% to 5.5%,” he explained. “The team believes that improved accuracy of medication dosage accounts for at least a significant portion of the decrease in readmissions, possibly due to a decline in post-discharge adverse drug reactions.”
ADVICE FOR OTHERS
As a result of this experience, King’s Daughters advises peer institutions to find ways to automate the transcription of sig data with tools that enhance patient safety and outcomes while creating time savings.
“Another lesson learned was to work closely with nurses to manage the transition to the new process,” Farr advised. “In order to reduce the possibility that nurses responsible for medication reconciliation would revert to the manual method of sig translation because the new process was unfamiliar, a two-minute educational video was prepared to facilitate the transition. This helped drive fast adoption and eliminated typical barriers associated with new technology rollouts.”