Strategies for Managing Patient Identification and Remote COVID-19 Testing

The demand for COVID-19 testing has increased the volume of test results sitting in organizations’ master patient index (MPI) or enterprise master patient index (EMPI) work queue.

Because factors such as a person’s employer, insurance company, or address determines where they go to get tested for COVID-19, there’s great variability when it comes to which data elements get collected at registration, and how data are entered into the ordering system. There is currently no national strategy requiring collection of certain patient demographic data elements, which can lead to inconsistent, dirty data at the beginning of the patient registration process.

Health information professionals who manage the MPI/EMPI confront the following issues daily—all of which are something that the presence of strong, cross-functional teams could help address in these challenging times:

Data discrepancies

• Human mistakes happen, especially when data are manually entered. Examples of keystroke mistakes include incorrectly entering or transposing the patient’s date of birth (e.g., 12/01/80 instead of 01/12/80), entering data in the wrong field (such as an insurance company name in the patient’s last name field), or entry of the incorrect gender.
• Educating staff and business associates about the frequency of human error in patient identification—and how to guard against it—is foundational, especially today with the increased volume of COVID-19 lab results.

Rejection queue

• The dirty data described above will create mismatches and prohibit the results from being uploaded to the electronic health record source system.
• As a result of the dirty data, COVID-19 lab results are placed in a “rejected” result queue for the data integrity team to remediate. This action is necessary several times per day not only to stay current with the increasing demand, but to post results in a timely manner to the patient’s unique health record.
• Health information professionals are aware of the possibility that rejections may be due to an existing duplicate record in the system. When this occurs, results are matched to both records, creating a rejection. Two charts must be merged together to take the entry off the rejection queue and release the results to the correct health record.
• To avoid further risk, medical record numbers (MRNs) should not be added to the HL7 message to push the results to the provider or health information exchange. If the MRN is entered incorrectly, the entire medical record is compromised. This action can result in separating the accurate data from the correct chart and require the creation of a help desk ticket to fix the feed, including other labs and reference ranges in the same HL7 message.

Individuals from information technology, registration, and senior leadership need to collaborate with health information professionals to explore matching workflows and processes for handling remote COVID-19 testing results, which is crucial to taking care of people who contract the infection.

Additionally, because patient identification and matching is vital to the delivery of safe and efficient care, and misidentification errors have been a recurring challenge resulting in administrative inefficiencies, serious injuries, and even death, AHIMA is recommending an bold goal of achieving and maintaining a 1 percent duplicate error rate. Our 2020 Patient Identification Survey found 22 percent of the health systems that responded are already achieving that goal. In anew white paper, AHIMA recommends a cycle approach to achieve a 1 percent duplicate record error rate. You can download the free whitepaper here.

Julie A. Pursley (Julie.Pursley@ahima.org) is director, health information thought leadership at AHIMA.