The volume of data in our daily lives continues to surge. By 2025, the compound annual growth rate of data in healthcare is projected to reach 36%, outpacing other industries like manufacturing, financial services, and media and entertainment. The digital revolution has spurred advancement in areas like diagnostic testing and wearable devices, generating even more data.
What is identity data and why does it matter?
At the heart of all this data is one constant – the person. You. Who you are and how you’re represented across medical systems is important. Identity data is often thought of as protected health information (PHI), which includes your full name, phone number, birthdate, and more. In the US this type of information is protected under HIPAA. Across the world there are other regulations like GDPR in Europe that also ensure data privacy and protection.
Identity data also encompasses more than just your health-related or medical encounters. Sometimes this is called personally identifiable information (PII), and it can be details about your lifestyle, behaviors, or credit, if available. Really, identity data is any information that helps determine you are you across disparate systems and records. The more identifying information that is available – in a secure, compliant format – the better the chances of accurately recognizing the same person across different systems and entities.
By comparing and matching pieces of identifying information, a health system can create one single record that users trust is accurate and complete. Additionally, knowing you are the patient in the exam room avoids errors and delays in treatment. It matters that your identity data is accurately linked, and this is challenging work.
Let’s say when you were younger you went by a nickname, and now that you’re a professional you’ve started using your full given name. And, on top of that, you recently changed your last name and moved. Your name is also really difficult to spell. What this means is you are the same person with different names and different addresses across various systems. It is likely that the next time you go in for your annual checkup, your doctor does not have a complete picture of you and your health. This might mean they call you by the wrong name, and it might mean your potentially life-threatening allergy isn’t accounted for.
Why should I care?
Health tech innovators like you are building digital innovations to accelerate administrative processes, deliver analytical insights, and improve efficiency and quality of care. Accurate, dependable data is the backbone of any successful application, AI algorithm, or machine learning model. We all know the quality of the input determines the quality of the output. The same is true when it comes to identity data, especially for patient- and person-centered innovations.
Yet, healthcare data is dirty, fraught with duplications, misspellings, transpositions, and discrepancies. Think about the earlier examples of nicknames and maiden or married names. Sometimes data is also entered incorrectly because someone didn’t understand what the patient said, or hit the wrong number, or was so tired they couldn’t concentrate. These data quality nuances can create havoc for individuals managing data and for the users downstream.
As an industry, it’s crucial to integrate multiple data sources. Rarely do individuals only visit one hospital or even one health system, and we all have different wearables now. If you see different specialists or have to go to urgent care on vacation, you want your record to follow you – completely, accurately, and securely. This requires making clinical and non-clinical systems interoperable. It also means conducting person matching and record linking, comparing identity data across systems.
What can I do?
You can prioritize data integrity by ensuring data quality and data auditability within your own product.
Digital health innovations, especially those using AI and machine learning (ML), are only as solid as the data they’re built upon. Therefore, data integrity is essential. That means developers must, themselves, be strong data stewards, prioritizing data lineage and auditability – having a detailed record of how data moves, where it originates, where it goes, and how it transforms along the way. Data lineage gives visibility into changes that occur to the data after migrations or system updates, for example.
In a perfect world, you have all the resources and time you need to do it all – innovate, sell, develop your roadmap, ensure data quality, and enjoy outings with friends and family. We know that is not the case, however, and often developing enhancements is prioritized over improving data quality (of course, both after family and friends!). But, doing so and putting data quality off until you have more time and resources may lead to lower quality and, in turn, subpar customer service and delivery.
Is there an app for that?
An alternative to manual data quality management is letting technology do the heavy lifting. Instead of linking person records and matching identities manually, which could take up to 10 minutes to compare just two records, you can employ an Enterprise Master Person Index (EMPI). An EMPI is designed to facilitate real-time data exchange and accurate record-linking across an ecosystem.
An EMPI becomes a critical technology foundation for organizations needing to make aggregated data usable across multiple products. Perhaps your company has grown through M&A activity or organically. Either way, you’re trying to build a comprehensive data set for analytics, AI, or reporting, and the person is the center of that data.
Or, maybe you’re looking to connect data across a workflow that interacts with multiple data sources. You need to make sure a person’s data makes it from the medical device to an internal system or a customer’s EHR. You could also be supporting clinical trial patient identification and tracking or facilitating claims and prior authorization workflows.
In all these instances, it is crucial that the person’s data – whether they used a nickname, full name, maiden, or dead name when filling out their information – is accurately matched and aggregated. Instead of spending your time, energy, and resources on this work, there’s a tool for that. At its core, an EMPI needs integration. Integration data sources effectively and securely is foundational to using an EMPI solution.
Now is the time to prioritize your team’s nights and weekends. Go ahead and put roadmap development ahead of integration and data quality, knowing you have an EMPI solution to ease the person data management burden. Take a look at how EMPI helps solve data quality, generates economic value, and enables you to focus on your core competency and differentiators.