Health Insurance Companies are Facing a Reckoning on Patient Matching
As technology continues to embed itself into our daily lives, we all create more and more data pertinent to our health – through telemedicine, patient portals, medical devices, behavioral health. Even our socioeconomic information is increasingly – and correctly – being treated as health data. With more information available than ever before, shouldn’t payers and providers have all the data they need to work together and improve the quality of health care delivery?
We’re seeing some gains on integrated delivery systems improving quality and value, but not nearly fast enough. As patient-generated data explodes, payers and providers alike struggle to find the signal from the noise and to share information across their fundamentally siloed technology systems.
We can’t make progress until we are able to answer a core fundamental question: Are we even talking about the same person?
Payers have often been exempt from this patient matching conversation. With the member ID on every member’s insurance card, payers had a pretty good identifier that worked for a while. But now that payers are becoming more embedded in population health and care delivery, they are facing the same issues of patient identification as their provider counterparts.
Data created outside directly billable events becomes difficult to link to the right member record – any maiden name, mis-typed birthdate, or change in address can prevent clinical data from being matched to the member record. But as quality metrics and reporting become increasingly critical to payers’ long-term success, this clinical data becomes more and more valuable, yet frustratingly inaccessible.
Conventional patient matching technologies compare a record from one IT system (say an EHR) with a record in another IT system (say the billing system). But each of these systems will store slightly different demographic data about a person: maybe one has a person’s maiden name, another an old address. These conventional patient matching technologies will therefore struggle to match these records to each other.
Verato approaches this problem in a totally new way. Rather than comparing different records directly to each other, Verato compares them to its large reference database, which contains a vast repository of publicly available demographic data spanning the entire US. This reference database essentially acts as an “answer key” for demographic data. Using this “referential matching” approach, Verato can match records even if they contain old, incorrect, or incomplete demographic data – which traditional approaches can never do.