Real provider data inaccuracies at two health plans - and how to solve them

In the world of healthcare and provider data, the term “accuracy” is thrown around often. Organizations want the most up-to-date and accurate provider data they can get their hands on, and many solutions claim to offer the most accurate data on the market. That begs the question: how do we measure accuracy? 

We know how deeply this matters - 80% of entries in 5 top health plan directories are inaccurate, a staggering percentage. Inaccurate provider data causes massive operational headaches, with tech and admin teams spending significant time and resources trying to manually validate and resolve these data issues. At Ribbon, our customers tell us that they previously spent over 120+ hours a week on this type of work. And it’s not just a directory issue - if the data on a provider is incorrect, it can cause downstream issues in authorizations, claims processing, contracting, credentialing, network adequacy, and more. 

Ribbon Provider Data Platform measures accuracy by assigning an Accuracy Score to every provider at a location with at least one phone number. This score is used within our model on the backend and is a percentage (from 0-100) that represents the likelihood that a given provider can be found at a location and is contactable using the phone number provided.

Let’s take a look at how that could come to life for your organization. 

Real provider data inaccuracies at two health plans

Below are two examples of provider records that we’ve pulled straight from major health plan directories. You can see exactly where directory accuracy issues occur. 

In our first example, Dr. Alex Hernandez has multiple addresses listed. With a combination of Ribbon machine learning methodology and manual validation, we can quickly determine the accuracy of each address. Additionally, we can enhance the data by adding a verified Board Certification to the provider’s record. 

Let’s explore a listing for Dr. Amanda Kit-Ling Tse in another national health plan’s directory. Although we are not seeing data that’s outright inaccurate, there is a significant amount of information missing. This health plan’s directory keeps only one address per provider, leaving substantial opportunity to enrich the record with other addresses where Dr. Kit-Ling Tse practices. The directory does not share important information like languages spoken by the provider or sub-specialty, for example. 

Is Accuracy Scoring the answer? 

Inaccurate data and missing information like the above can have a ripple effect across an organization. These examples show one provider record, at one organization. Many enterprise healthcare companies contract with over one million providers - imagine the scale of solving these inaccuracies at that level. Often these providers don’t just have two or three discrepancies  - they can have hundreds, with data coming from a multitude of sources. 

If it sounds nearly impossible to scale this data, it is. Ribbon exists to solve this problem. 

With the Accuracy Scoring feature of Ribbon Provider Data Platform, organizations can differentiate between good and bad data in minutes, whether scoring their own data sources or using Ribbon’s established provider data asset. Teams can track the confidence of specific sources and overall accuracy over time, and can quickly triage data that needs to be further investigated. Ribbon also offers Manual Validation services that can be used to strategically lift the data accuracy of target markets or fields that are hard to verify through technology alone. 

If you’re curious about your organization’s provider data quality, how your accuracy compares to similar organizations, and actionable steps to improve, connect with our team below. 

‍