In last month’s blog, I drew a comparison between health information exchange (HIE) organizations, like Hixny, and Ray Kinsella in the movie Field of Dreams. I described how SMART on FHIR technology has the potential to do for an HIE what a baseball field in the middle of a cornfield did for Ray Kinsella: bring people to the front door.
I like analogies, so I continued to think about this comparison and wondered if we, as an HIE, would continue to have the same success Ray did. Then I remembered that there was no sequel to the movie, so we don’t know if Ray saved his farm. I’m an optimist, so I like to believe the lines of cars we saw continued—but for all we know, people left and never came back after their first visit because, well, it wasn’t that great after all. HIEs want—and need—to avoid that uncertainty.
The current environment increasingly offers providers options to access remote data within their workflow—enabled by SMART on FHIR technology, for example. To keep people coming back, though, we need to do more than just make our applications available. We need to ensure the data our users find is worth a return visit.
No matter what data an HIE chooses to make available through its SMART on FHIR application, that data must be of the highest quality. This includes data that comes through the HIE from external sources, as well as data created by the HIE.
In fact, in my opinion, the single biggest indicator of data quality is the master patient index (MPI) created by an HIE, which demonstrates the HIE’s ability to properly match patients with their healthcare records. Unmatched records split a single patient’s data into multiple patient records, making it impossible for a provider to view a patient’s complete record in one place.
I think it’s also important to mention one major mistake an HIE can make when it comes to data quality: mismatching records. Beyond inconvenience, mismatched records present a real risk to patients, because they may lead providers to make care decisions based on the wrong patient’s health history. HIEs are most likely to encounter a mismatch in the records of same-sex twins and patients with the same name and Jr./Sr. suffixes—and with the same address. A mismatch can also occur when data sources—like electronic health records systems (EHRs)—reuse medical record numbers (MRNs).
Beyond the MPI
In addition to ensuring the information in the MPI they’re reviewing is properly matched, providers must also pay attention to the quality of the data they’re sharing from their own EHRs through the HIE. There are six categories in which to measure quality:
- How well does the data reflect what’s actually going on with the patient?
- Is the data up-to-date and does it reflect the patient’s current state?
- Are data values consistent across sources? HIEs commonly refer to this as semantic interoperability or standard coding of data content (e.g., LOINC for labs, SNOMED, and RXNorm).
- Is the data the same across systems (e.g., has the matching allowed the HIE to merge records or is data split into multiple patient records)?
- Did the EHR send a complete patient record through the HIE or is the HIE missing data that is relevant to the patient’s other providers?
- Does the data make sense for that patient (e.g., does a child’s health data describe a condition only present in geriatric patients)?
While we have no concrete evidence that Ray Kinsella successfully saved his farm, HIEs aren’t left in the dark when it comes to their SMART on FHIR applications. We can look at use trends and analyze the information to understand whether we’re giving providers the data they want and need. If we’re not, we can adjust and improve.
While quality is at the top of the list for review, there is also unique data HIEs can make available to confirm the value of their application over the numerous others available. Unlike Field of Dreams, I’ll write a sequel to this blog next month to discuss those differentiators.