The Death of the Flat File: Why APIs are the Future of PDM
Moving from monthly batch reports to a continuous “Data Stream“
Mark Fabiano
CEO of Cúratus
Why health plans are replacing monthly data reports with real-time provider feeds
A member calls to find an in-network OB-GYN. Your directory says Dr. Smith is accepting patients at her current address. But Dr. Smith left that practice six weeks ago. The member drives 40 minutes, gets turned away, and files a complaint. Your plan pays a corrective claim. And somewhere, a spreadsheet is to blame.
This happens every day because 5% of provider data changes every month because of addresses move, licenses expire, panels close, or group affiliations shift. For a network of 10,000 providers, that means roughly 500 records are wrong at any given moment.
Why is our current vendor or batch process still failing us?
Because it was never built for this pace of change. Whether you’re working with a quarterly data feed, an automated file drop, or a vendor extract, the core problem is the same: these are snapshots. A snapshot taken Monday is wrong by Friday.
The workflow requires someone to notice a change, capture it, send it, and wait for it to be processed. This is a chain with multiple places to break. Even when every step works, you’re still operating on data that could be 30 days stale.
Why can't we just use spreadsheets for provider data?
You can, but the math works against you. If 5% of your provider records change every month and you’re updating them monthly, you’re perpetually one step behind. For a 10,000-provider network, that’s 500 wrong records at any given time flowing into claims systems, member directories, and call center scripts.
The spreadsheet isn’t just slow. It’s a single point of failure that quietly touches every department.
What does bad provider data actually cost us?
More than most plans realize. CMS estimates that provider directory errors contribute to millions of dollars in improper claims payments annually. Beyond the financial hit:
- Compliance exposure — The No Surprises Act and state-level directory accuracy mandates have made “we updated it last month” an increasingly risky answer to a regulator.
- Member churn — Directory inaccuracies consistently rank among the top drivers of member complaints and plan-switching.
- Operational drag — Every team downstream from the data, claims, member services, network contracting, absorbs the cost of corrections.
How does real-time provider data work?
Instead of waiting for a monthly file, your systems connect directly to live sources: state licensing boards, DEA registries, claims systems, and credentialing platforms. When a provider’s license renews or an address changes, that update flows straight into your directory — in seconds, not weeks.
This is possible today because federal standards now require health systems and providers to share data in a common, machine-readable format. Plans don’t need to build this infrastructure from scratch; they need a curation layer that connects those sources and keeps internal systems current, automatically.
How do we make the internal business case?
Start with what you’re already spending. Tally the staff hours spent chasing data corrections, the claims that had to be rerouted or reprocessed, and the complaints that required follow-up. For most plans, the cost of bad data is already embedded in the budget, it’s just scattered across departments and hard to see as a single line item.
The shift to real-time data isn’t a new expense. It’s a reallocation away from reactive cleanup toward proactive accuracy.
How hard is it to implement?
Less disruptive than most plans expect. The integration connects to sources you already rely on — your credentialing system, your claims platform, state licensing feeds — and layers on top of your existing workflows rather than replacing them. Most plans see their first accuracy improvements within the first few weeks of going live.
Ready to see what always-current provider data looks like for your plan? Explore Curatus.