Page 21 - FLIPMAG.NET

Email addresses aren’t typically used for matching, al-
tients, according to Pew’s Moscovitch. That’s despite the
fact email addresses are one of a few “important data ele-
ments” recent research has suggested could help to “facili-
n The ONC stopped short of requiring
organizations to use a specific format
tate greater match rates,” Moscovitch said.
And, according to a study published in the Journal of the
when collecting patient addresses,
American Medical Informatics Association last year, stan-
although many healthcare stakeholders had
dardizing last names and addresses—specifically to the
submitted public comments requesting the
format used by the U.S. Postal Service—has proved helpful
agency require organizations to use the U.S.
for improving matching sensitivity.
Postal Service standard.
After a new medical record is created at Children’s
Health, data integrity specialists manually review each
one to determine if it’s a potential duplicate that needs to
To match patients across facilities, UHIN uses probabi- though most hospitals do collect the information from pa-
be merged with another file, or if it really is a new patient.
listic matching algorithms that compare multiple demo- If a data integrity specialist unearths a problem—a re-
graphics, including name, date of birth, gender, address, cord that’s been erroneously created or merged—“imme-
phone number and Social Security number, to determine diately, the registrar knows that they’ve made a mistake,”
the likelihood of two records being from the same patient. Lusk said, as the data integrity specialist sends the registrar
If enough of those items match closely enough, the soft- feedback on what policies they might have missed.
ware will link the two records. Children’s Health also shares quarterly reports with a
If some of the demographics match, but the system is un- health information management committee, showing du-
sure, it’ll flag the records for additional review. plicate record rates by department.
UHIN runs those possible matches through software
that uses external, third-party data—such as information Quickly alerting registrars
from public records—to determine whether to merge two Hospitals need to set up these feedback loops, so regis-
records. This method, called referential matching, works trars are alerted when they’ve pulled the wrong patient
by creating a more comprehensive view of a patient with record or input a patient’s demographic information incor-
information like previous names and addresses. rectly. That should include training registrars not only on
Regularly reviewing those possible matches is important the technical aspects of their responsibilities, but also on
to avoid a long queue of records needing confirmation. “the importance of what they’re doing,” such as the possi-
Still, Johansen stressed that while matching algorithms ble clinical ramifications if something goes awry, AHIMA’s
are important, health systems and HIEs need to ensure Pursley said.
they’re working with high quality and consistent data. “Show them how to correct it in the future,” Pursley said
“You can do so much with an algorithm, but if the source of acknowledging mistakes. But beyond that, explain how
data isn’t clean to start out with then there are limitations “identifying the patient upfront is a core function to every-
on what you can do,” Johansen said. thing that we do.”
Another component of Children’s Health’s patient-
Back to the basics matching strategy has involved figuring out when standard
To get clean source data, it’s not just about the IT un- demographic data—even if input perfectly—can still trip
derpinnings—there’s also a human component that up some algorithms.
demands special training. That’s been key to helping In the case of a newborn, for example, a hospital will need
Children’s Health in Dallas maintain a low duplicate re- to develop protocols for how to create a medical record for
cord rate. The system boasts a duplicate record rate of just an infant who hasn’t been named yet. There’s also chal-
0.1% or 0.2%, according to Katherine Lusk, chief health lenges that arise with young twins or triplets, since many
information management and exchange officer at Chil- will share the same date of birth, address and phone num-
dren’s Health. ber. In those cases, a data integrity specialist might need to
“We really focus on data integrity,” Lusk said, which delve in to the clinical data for assurance that they’re look-
means the system gives immediate feedback to registrars ing at the right patient.
when a mistake they’ve made results in a duplicate or erro- Twins and triplets proved challenging when working
neously merged patient record. with regional HIEs, Lusk said. HIEs would often try to er-
To create a new medical record, registrars at Children’s roneously merge young siblings with one another, since
Health are prompted to input complete legal name and their patient-matching algorithms and practices were de-
address in the format used by the U.S. Postal Service for veloped for adults—so Children’s Health learned it needed
patients, as well as date of birth, gender, race, phone num- to keep a close eye on those cases and flag patients who are
ber and even email address. That information—along with part of multiple births.
previous names and addresses—is the same data used to “In our very fragile population, oftentimes we have to
match patients that visit the system in the future. look just a little bit deeper,” Lusk said. l
June 8, 2020 | Modern Healthcare 19