When health records don’t capture patient deaths
Electronic health records (EHRs) are meant to record patients’ accurate and complete medical history, yet it is no secret that these systems have their share of problems. Common issues include duplicated patient information, prescription authorization errors, billing mistakes, and patient misidentification. While simply copying and pasting patient information from previous visits to update patient charts may initially help in saving staff time, what happens when a shortcut like this fails to capture that a patient has died?
It turns out that inconsistencies around patient death status in the EHR is a pretty common occurrence. A recent report in JAMA Internal Medicine investigated the prevalence of outdated EHR information using records from seriously ill patients at UCLA Health. From the 12,000 patients identified in this category, 676 were marked alive in their health records but were actually dead in state public records.
It gets worse.
As these patients were still assumed to be alive, 541 of them still had appointments scheduled after their death. These patients continued to be contacted by health care staff and received approximately 221 calls and 338 portal messages unrelated to their deceased status. Additional contact attempts urged these patients to get preventative care like flu shots and cancer screenings, and medications were still being authorized for at least 88 dead patients.
This mistaken follow-up is not only wasteful for the health system, it could be upsetting for family members to receive these messages. Inconsistencies in EHR data could also be perpetuated in AI algorithms trained on this data, which would make these tools less accurate.
Drivers of EHR inconsistencies
A part of the problem lies in the fragmented nature of EHRs. The average health system uses 18 different EHR vendors across affiliated providers. But these EHR systems aren’t always able to talk to one another, meaning patients may still struggle to access their data, doctors may order duplicate testing, or need to transfer health data with fax or a CD. With so many data sources trying to provide a total view of the patient, it is easy to see why a patient may be marked as deceased in an EHR used for inpatient services but their specialist or primary care doctor may have no idea. And the fact that many EHRs cannot link up easily with federal and state records makes recording patient deaths a much harder task.
Even without interoperability issues, verifying patient death is not easy. Health systems have to access state records to confirm patient death, and ease in state accessibility varies. Even if the patient is found in state records, if they have a common name, there may not be enough information for health systems to verify that it’s the right person. Research services have to be contracted out to do a deeper search, but if the patient can not be linked to a social security number, or the information in their EHR is insufficient, a deceased status can not be officially confirmed. Internal departments often do not have an organized system or incentive to verify a flag in a patient’s chart indicating their death.
Dr. Eric Cheng, chief medical informatics officer at UCLA, explained challenges with coordinating this information, in an interview with StatNews:
“If a patient were to call the clinic or a doctor and say a family member died, we don’t necessarily do the best job in documenting that the same way. Physicians don’t know whether that’s stored, the front desk clinic may not be comfortable if they’ve never heard of the patient — they would all document in the note, but not in the official place where it should be.”
Dr. Eric Cheng, chief medical informatics officer at UCLA, in StatNews
How can EHRs be improved?
Creating requirements and standards for interoperability has been a goal for CMS for many years—and now it appears real change is finally on its way. CMS launched the Trusted Exchange Framework and Common Agreement (TEFCA) in December 2023, providing much-needed standards and allowing public-private collaboration to address gaps in health information exchange across EHR systems. Five organizations known as Qualified Health Information Networks (QHINs) are officially signed on to use this framework and can start exchanging data immediately. Hopefully, many others will follow.
Patient misidentification not only contributes to poor tracking of death records but can also cause trauma and delayed care. To prevent this, another potential solution is creating unique patient identification numbers, a policy supported by some specialty groups, health systems, and industry organizations.
A large part of making hospitals more accountable starts with having an efficient EHR system that keeps patient values as the focus. As solutions are being pushed, we hope to see advantages also impact entire health systems by preventing physician burnout, reducing administration costs, and providing high quality care.