It isn’t surprising that many health plans are quickly updating their models to improve provider data inaccuracies, enable members to find the right care, and avoid costly fines for data that isn’t compliant. With ghost networks recently in news headlines and incorrect data impacting patients’ ability to find care, health plans across the country need to act quickly to rethink their current data solutions to improve member experience and save costs via automation.

In 2017, The Centers for Medicare & Medicaid Services (CMS) conducted a study of online provider directories and found that 46% of examined locations had inaccurate information resulting in a lack of access to care. Despite inaccurate data being a big culprit in patients’ ability to find care, patients still trust health plans to provide accurate information to help them identify the right provider, according to the J.D. Power 2021 U.S. Commercial Member Health Plan Study.

So what does this mean for health plans? Health plans have an opportunity to innovate their current data infrastructure to not only update provider data in real-time but improve the provider search experience to increase member retention and satisfaction, and ultimately help patients find better care.

The way to reach this end-goal is by consumerizing data. Consumerized data, in the case of health plans, would involve reorienting current products and services to model similar consumer experiences, think online shopping, for members. In short, health plans’ provider networks should show insightful information about providers that allows each user to personalize their care decisions. In this article, I explore a few ways health plans can back into this solution with their existing data. This approach, if done correctly, can help health plans deliver more customer-centric services that help patients find differentiated, high-quality care that meets their specific needs.

Single source of truth for data from multiple sources

More data does not necessarily mean accurate data or better patient experiences. It’s what health plans do with that data that drives member retention and acquisition. Health plans should see gaps in their current data as a driving point to refine their current infrastructure and draw in diversified data outside of traditional provider sources. This can help health plans position their provider search platforms as the single source of truth for provider information across a variety of data points. For this to happen, health plans must leverage rich, external data networks to pull in different provider data outside of phone numbers, email, and addresses.

Because consumers know that health plan provider directories are not always accurate, they seek information from many other sources like search engines, doctor rating sites, and social media. The rise of ghost networks and outdated provider data has created a frustrating experience for members and can result in additional costs to health plans if a member goes to a provider that’s actually out of network, and undergoes unnecessary and costly procedures and services. Instead, data should be validated, synthesized, and distributed from payer to member in a clear format to help members make confident decisions regarding their care provider.

The rise in telehealth is a relatively new but also critical factor to consider. Virtual care visits are cost-effective for members, providers, and health plans alike. An analysis conducted by McKinsey & Company in May 2020 suggested that offering virtual urgent care assistance could cut emergency room visits by 20%. Including information like virtual care availability and improving the overall comprehensiveness of a health plan’s provider directory can position that directory as the single source of truth. This reduces the friction and frustration members experience when they’re forced to search through multiple sources to determine if a doctor really takes their insurance or actually treats knee joint pain, rather than back pain.

Leveraging machine learning to drive actionable data

Machine learning can be a health plan’s secret weapon when it comes to consumerizing existing data. Health plans can improve their information on providers by pulling in data from external sources — there is also a clear opportunity for plans to use this data to arm consumers with more personalized provider options.

In addition to phone numbers and addresses, health plans should expand their provider data to include considerations like cultural background, languages spoken, specialty and focus areas of expertise, and quality indicators. Health plans should also look to optimize their current provider search experience to allow patients to drive their provider search with personalized filters that include ratings and social/cultural parameters.

For example, if I am looking for a provider who specifically treats joint pain, is located in Idaho, has 4.5-star reviews from other patients, and can speak Spanish, I may have a challenge in finding this exclusively through my health plan’s current provider data. However, if this health plan were leveraging data informed by machine learning, I may experience a much more seamless and accurate search experience.

Machine learning models would sift through the thousands of search results and predict that Dr. Camila Velasquez who practices at 41 Broadway is the same provider as Dr. Camila Velasquez-Smith practicing at 41 W Broadway and also no longer practicing at 66 Main Street. Rather than seeing 20 similar results for the same person, I’m served one accurate and consolidated listing that has the information I’m looking for. Machine learning would also recognize the numerous terms and specialty focus areas that “joint pain” could be categorized as, and pull these results in to ensure I’m seeing all options available.

The results I get from a provider search powered by machine learning are personalized, centralized, and filtered to match my expectation of quality. As a patient, this saves me time, frustration, and ensures I’m directed to the right provider for my specific needs.

Enriching existing provider data to enable members to make personalized care decisions