As we strike out into 2018, the implementation of All-Payer Claims Databases (APCD) across states remains variable and dynamic. Massachusetts maintains a comprehensive implementation, aggregating data feeds from over 80 public and private payers. Massachusetts has leveraged their APCD to create a state-specific risk adjustment model to meet the ACA provision which balances funds from healthier populations to higher risk pools. Late in 2016, Minnesota concluded a feasibility study which determined their APCD could significantly improve risk adjustment vs. the federal model.
On the other hand, West Virginia and Tennessee have put APCD development on hold. California payers optionally submit claims and encounters to a public benefit corporation. Legal, fiscal and political concerns guarantee a fluid situation for insurers.
This blog post is focused on the technical obstacles that health plans face in states requiring APCD submission. Since these databases have phased in over the last decade through both voluntary and legislated…
The X12 HIPAA transaction set is used across the healthcare industry to transmit claim, enrollment and payment information. Given the importance and ubiquity of these EDI files, you might assume that translating them from ANSI to a relational database format would be well-supported with a range of options.
In practice, a task as common as parsing a claim or encounter and storing it in a database can quickly escalate into a significant problem.
One solution we’ve seen involves archiving a snapshot of the EDI file using filestream storage. This can satisfy some retention requirements, but provides little in terms of fine-grained tracking or analytic capabilities.
A more complete approach is to parse the X12 file into its discrete elements and store them in a relational database. The ideal solution captures the full extent of the EDI transactions while also applying a reasonable leveling of flattening to keep in the number of table joins under control.