Using Provider Directory APIs to identify errors in plan directories

One potentially promising method is to use data analytics to compare health plans’ directories, many of which are available in machine-readable formats thanks to federal regulation.

HHS Office of Behavioral Health, Disability, and Aging Policy
State Efforts to Coordinate Provider Directory Accuracy

In our previous post reacting to the HHS Report on State Efforts to Coordinate Provider Directory Accuracy, we observed that the report considered lack of scrutiny as a reason for directory inaccuracy: ‘One reason there is little data on the accuracy of provider directories is that few state regulators conduct regular surveys of all health plan products’ and a need for an ‘alternative to a “secret shopper” method or phone survey method for monitoring may offer efficiency and value in identifying inaccurate listings.’

The authors reference the CMS-required Provider Directory APIs and the HL7 DaVinci Plan-Net standard multiple times within the report. Starting in 2021, CMS required that all Medicare Advantage and Medicaid plans publish provider directory data in a standards-based API. Exchange plan directory data had already been required to be published as part of previous ACA requirements. A small but growing number of states have enacted or are considering legislation to require provider directory transparency for commercial lines of business.

Analytics using Provider Directory API data to identify inaccurate listings

On the APIs, the authors consider performing analytics on the newly available data to serve as an efficient ‘alternative to a “secret shopper” method or phone survey method’ that ‘may offer efficiency and value in identifying inaccurate listings of behavioral health providers in health plans’ provider directories, and seeking enforcement of directory accuracy and health plan provider network adequacy.’

Such an analytical method could:

  • Provide payers with an affordable and scalable feedback loop on directory accuracy.
  • Identify provider organizations contributing the most errors to payer directories, enabling collective intervention and accountability on those provider organizations.
  • Provide regulators with a method to more frequently assess payer directory accuracy.
  • Enable industry to benchmark payers and compare their accuracy relative to each other. This could assist regulators in prioritizing enforcement. This could also help consumers make a decision on both the breadth of a network as well as its accuracy.

It is informative to know how payers, using diverse data sources and cleansing methods, agree and disagree on specific directory data records. When a large proportion of payers who have a particular provider in-network agree on the locations and phone numbers associated with that provider, the confidence around those records increases. When a small proportion of payers agree with a particular directory data record, it follows that the confidence around those data is relatively low. Analytics on the consensus agreement or disagreement of these directories can produce reliable assessments of directory-wide accuracy for each payer.

Defacto Health has integrated with 127+ payers’ Provider Directory APIs and machine-readable files and has built tools to assess the accuracy of directories by comparing payers’ data. We assess the accuracy of a directory, benchmark directory accuracy across payers within a market, and identify problem spots in a payer’s directory. Learn more about Defacto Health’s Directory Risk Report.

The HHS report identifies an opportunity for both regulators and payers to use newly available data from payers’ Directory APIs to identify discrepancies and assess directory accuracy, establishing a feedback loop that can immediately promote greater accuracy across all payer directories.

Importance of emergent Provider Directory data standards

On the FHIR standard itself, the authors state, ‘A truly industry-wide standard would be utilized nationally, obviating the need for each state entity to produce a competing set of standards. FHIR standards seem poised to fill this need.’ Defacto Health strongly supports this assertion. The HL7 DaVinci Plan-Net Implementation Guide is the most widely adopted and used standard for provider directory data, with most Medicare and Medicaid payers using it to publish APIs and a growing number of healthcare data and interoperability vendors also using the standard. This is the closest to a normative, national standard for provider directory data that we have ever seen. Seeing this standard further adopted for data transmission between payers and providers would be a positive development.

In conclusion

As a milestone towards a centralized directory, government and industry should embrace provider directory data standards. This reduces administrative burden in data collection and submission, and future centralized provider directory would need to leverage such endpoints and standards. This would require adoption, over time, among vendors supporting payers and providers to realize the benefit.

More immediately, there is an opportunity for payers to leverage Provider Directory APIs published with those same directory data standards to help identify accurate and inaccurate listings for payers in an efficient, non-abrasive approach. This new, scalable approach can serve as a robust feedback loop to support provider directory data quality improvement initiatives within payer organizations.