The Edge Data Governance Gateway: How Healthcare Data Exchange Shifts Left For Efficiency
bTrade's TDXchange and FlightPath Data team up on security, compliance, correctness, and scale.
Over the years, I have been involved with a few healthcare navigators and analytics firms. From cost transparency to pharma benefits to provider quality. In each case the requirements for partner data exchanges were strict. Strong evidence of controls and correctness was required by law. With PII (personally identifiable information) and the financial details of companies and individuals, the data powering healthcare navigators and healthcare analytics companies is the very definition of sensitive.
Atesta Analytics was formed with the mission to, first and foremost, control data operations costs, end data counterparty risk, and raise data value through quality management. Compliance is a part of that story and deserves highlighting. The cost of compliance with standards like HIPAA and SOC 2 is nothing compared to the impact of adverse events compliance could have forestalled.
And this is the moment to focus on the requirements of healthcare, because Atesta Analytics and its partner bTrade are teaming up to announce an Edge Data Governance Gateway solution to solve these exact challenges. What better way to illustrate the capabilities than with a composite example from the world of healthcare data? We feel that if your solution can cut it in the most sensitive of industries, you pass a test that everyone understands.
What is an Edge Data Governance Gateway? It's a governance-first ingress point that sits at the edge of your data estate, inspecting partner files as they arrive. It combines secure managed file transfer (MFT) with versioned staging, automated data validation, lineage, and metadata tracking, so that only trustworthy data reaches your data lake, applications, analytics, or AI.
Healthcare Navigators Aggregate Data
Most navigators lower costs three ways:
They cut out-of-pocket expenses for individual families
They lower the cost of benefits to employers and captive insurance
They provide tools for managed care organizations and benefits managers to facilitate effective care while protecting margins
The source of a navigator's power is visibility across payers and providers combined with the analytics that enable cost allocation, negotiation, and comparison shopping. Obviously, this requires fresh and trustworthy data, and a lot of it.
Aggregating structured data from multiple partners carries line-of-business risk and unlimited downside. And navigators carry the additional burden of intense industry and regulatory oversight. Fortunately, Atesta delivers the capabilities required in the world's most popular commercial open source data preboarding solution, FlightPath Data and its CsvPath Framework foundation.
I say this a lot because it is key to proof-of-value and adoption: there has never before been a COTS (commercial off-the-shelf) package for the edge data governance of data partnerships built around tabular data exchange. The majority of mid-market and large enterprises, not merely in healthcare, exchange tabular data feeds with data partners. Historically, they rolled their own "good enough" solution and owned the overhead and limitations of systems that quickly become brittle. Commonly they are dissatisfied. Atesta aims to change that.
Introducing an Edge Data Governance Gateway
I'm pleased to announce Atesta Analytics is teaming up with bTrade to offer a new solution combining our expertise: the Edge Data Governance Gateway. The Gateway builds on the long-standing security at scale delivered by bTrade's highly regarded managed file transfer (MFT) platform, TDXchange. Atesta layers on the comprehensive data preboarding infrastructure of its FlightPath Data and FlightPath Server.
What is an Edge Data Governance Gateway, in practice? At a high level, think of it as a customs gate where goods are assessed before they enter the data estate. It is a guard post that collects data with strong post-quantum encryption and defense in-depth, bTrade's specialty. It then inspects the data to assess provenance and assign version identity, does a deep inspection to verify and canonicalize content, tracks lineage and operational metadata, and releases the data to downstream consumers through no-code/low-code integrations and channels.
The Gateway is end-to-end ingestion from partner through internal trusted publishing from an immutable, versioned permanent repository of truth. bTrade's decades of experience handling volume data with unimpeachable security and compliance now has a governance-first layer that extends into the data lake, applications, analytics, and AI.
The core deserves a data partner it can trust. The Gateway is that partner.
The Navigator's Problem In a Nutshell
Back to our composite healthcare navigation company. The company receives sensitive healthcare data from its customers, providers, and external partners through multiple file transfer channels in regular feeds and ad hoc batch processing. As we said, truly sensitive records. They include personally identifiable information (PII). The PII contained confidential details, such as diagnosis and procedures performed by specific providers, costs and payments, and, of course, consumer demographics and locations. Security was paramount. However, securely receiving the files was only the beginning of the challenge.
Healthcare navigators sit at the crossroads of urgently needed care, finance, and long-term outcomes. The company highlights a widely cited industry figure: roughly 80% of medical bills contain at least one error.[^1] Given the complexity of medical treatment and the known difficulty of assigning costs across dozens of components of each visit, I believe it. It tallies with my own impressions from the field. And not only must procedures be correctly coded, costs and coverage must reflect actual patient circumstances.
To ensure this fidelity, every file was checked for canonical values and completeness, reformatted according to schemas, deduplicated, verified for correlated fields and commercial ties, and marked ready for downstream processing. Data that did not pass standards was returned to upstream sources or channeled to BizOps for triage. The validation required was extensive. Each tabular data file required hours to days to fully vet.
Unfortunately, due to the flexibility of CSV formats that both enabled volume exchanges yet adhered only loosely to partner-by-partner agreed formats, collecting healthcare data in traditional ways is burdensome. Even the switch to NDJSON doesn't fix the problems of errors, drift, restatement, identity, etc. It forces team effort to scale linearly with the business. That problem alone is unsustainable. But it gets worse.
Given the complexities of medical access tracking, provider rosters and directories, and coding and billing, the skill level required to manually check data was significantly higher than in less intensive domains. I have personally yet to work with a navigator that doesn't start out using MDs, PharmDs, or MPHs to do manual data checking in Excel. That's a gross misuse of skills and training, but it is the only responsible resort where rigorous, scalable, and maintainable data preboarding is not available.
Moreover, many navigators are paid for performance. If their numbers don't add up or fail to have impact, the company doesn't get paid. Similarly, healthcare analytics providers, whether in billing, procurement, utilization, or outcomes, are paid to bend the trendline. Trusting that line is nonnegotiable, so the tabular data collected must be verified correct immediately. Faulty data at the source is never an excuse.
The Solution
FlightPath Data stands above its peers on two stilts: the preboarding process of data handling and tabular data validation. No other tool offers a prescriptive identification, immutable staging, validation, and metadata publishing linear pipeline that adapts seamlessly, even transparently, into existing file feed processes. And there is no tabular data validation tool that can compare to FlightPath's ability to lower costs by stopping manual data checking.
FlightPath's validation works by combining SQL-database like schema validation with business-rules-based validation that encodes your business requirements. Does a patient need a follow-up visit within 45 days? A data structure schema can't check that. Likewise, a schema cannot warn you that Albuterol and Propranolol are not customary bedfellows. But if a rule can be written down in English, 90% of the time it can be encoded as a FlightPath validation rule. And FlightPath's AI can even read the English and write the rules for you. Not only do you lessen the SME workload, you do it without stressing the technology team. Savings from reducing manual data checking, and the skill level and human judgement it requires, can add up quickly.
Nevertheless, no system is completely foolproof. Some errors will always get through. Two statistics are commonly called out to highlight how damaging even a greatly reduced flow of errors can be. First, errors compound in diagnosis and remediation cost with every hop downstream, echoing the well-known "1-10-100 rule" of data quality.[^2] An error caught at the point of entry might cost $1 to fix, the same error caught downstream can cost 10x to 100x more. Second, that 50% or more of the total work of data engineering and BizOps teams is consumed by firefighting errors.[^3] Obviously, these are two sides of the same coin. How does FlightPath address this?
FlightPath has an intentionally simple, linear workflow. It relies on immutable, versioned data and durable byte-level data identification. At each point of data capture or change, a complete metadata snapshot is created and distributed. When a rule or processing flow changes, a new version is created so that every ingestion run can be 100% replicated at any time from any point in its workflow. At all key lifecycle points, from setup to running to errors, more metadata is captured. These metadata events are broadcast using open standards to any of over 80 observability systems, both open source and commercial, that provide tracing and alerting. And all this precise configuration, data handling, and metadata capturing happens exactly the same way for every data partnership. There are no one-off scripts, no lost knowledge, no buried logs with inscrutable indications, and in fact no magic to break. This is how tabular data operations become efficient.
By integrating TDXchange with FlightPath Data, we deliver on a promise: faster customer onboarding, faster time to resolution, and agile change as the business evolves. Security and compliance evidence becomes rote. And the level of skills required drops as quality management shifts left.
The Edge Data Governance Gateway comes from two companies that know how to deliver, rapidly install, and speed time to value. In fact, I believe our products enable among the fastest time to value in their respective classes. They tie together with no-code interfaces and downstream connections that mirror how most companies already work. Our partners at bTrade are used to re-platforming their new customers onto TDXchange with minimal disruption. Likewise, Atesta's distribution strategy prioritizes a fast start.
Once the navigator installs the Edge Data Governance Gateway, the connection and validation rules that mirror and automate then-current flows are added. Both of the Gateway's product components are AI-accelerated. Tabular data exchanges are often highly specified. The access, rules, and layouts can be complex. Having an AI capability that is able to take plain-English instructions and generate the matching configuration is huge. And FlightPath's ability to set up and easily sync local development with server-based production operations is key.
Together, TDXchange and FlightPath Data enabled the company to:
Secure sensitive healthcare records exchanges with advanced, post-quantum-ready encryption and controls
Automate partner onboarding and file-processing workflows over MFT
Validate incoming records automatically before downstream processing
Quarantine, reject, or return files that failed defined requirements
Maintain lineage, provenance, and processing metadata, and
Improve operational visibility and audit readiness
Moreover, and somewhat obviously given the industry we're talking about, the Gateway increased the robustness of the navigator's compliance stance. No individual technology platform makes an organization compliant in and of itself. Compliance also depends on policies, people, risk management, and documented processes. TDXchange and FlightPath simply provided the firm technical foundations and operational evidence needed to support those programs. And as governance-first products, the package speeds compliance rollout and evidence collection, giving partners confidence from the start.
It's All About Outcomes
Like healthcare navigators, the Edge Data Governance Gateway is all about practical outcomes. IT workload dropped. Quality shifted left even as the skills required for quality management shifted down. And teams faced dramatically less firefighting and significantly faster forensics. With those improvements, scalability was unlocked. The company was able to take a more ambitious approach to a fast-moving market.
These are not small wins. When your mission is critical to people's well-being and you're winning the game, you are making a difference every day. As they say, winning is fun, particularly when you know you are saving lives and protecting families' financial future.
If your company is scale-challenged due to difficult data exchanges and the lack of a COTS preboarding architecture, you probably need an Edge Data Governance Gateway solution too. It's a common problem. And data partnership doesn't have to be so hard, or even sometimes feel so existential.
When you have the right purpose-built infrastructure in place, aggregating data becomes more manageable quickly. If you want to see how, reach out to bTrade or Atesta Analytics. We'll be glad to give you more information and help you get started.
Frequently Asked Questions
What is an Edge Data Governance Gateway? It's a combined security-and-governance layer for partner data exchange: secure managed file transfer (MFT) for encrypted, controlled ingestion, paired with automated data validation, lineage tracking, and metadata publishing before data ever reaches downstream systems.
How is this different from just using an MFT server? An MFT server like TDXchange secures the transport of files — who can send what, over which protocol, with what encryption. It doesn't tell you whether the contents of the file are correct, complete, or safe to use. The Gateway adds that layer: schema and business-rule validation, canonicalization, deduplication, and quarantine of bad records, on top of the secure transport bTrade already provides.
How does FlightPath Data's validation differ from schema validation alone? A schema can confirm a field is the right type or format. It can't tell you a patient needs a follow-up visit within 45 days, or that two prescribed medications shouldn't be combined. FlightPath layers business-rules-based validation on top of schema checks, and can generate many of those rules directly from plain-English requirements.
Does adopting the Gateway replace our existing MFT server? Not necessarily. The Gateway is built on bTrade's TDXchange for the secure transport layer, with FlightPath Data adding preboarding and validation on top. Organizations already on TDXchange add governance capability; organizations on other MFT platforms can discuss integration options with bTrade and Atesta directly.
Is this specific to healthcare? No. The composite example here is healthcare because it's one of the most regulated, highest-stakes environments for tabular partner data exchange — and because FlightPath Data with its NDJSON support maps well to current FHIR bulk data exchange mandates. But the same problem (and the same Gateway) applies to any mid-market or enterprise organization exchanging structured data feeds with outside partners.
By David Kershaw, Head of Product, Atesta Analytics
[^1]: The 80% figure is widely cited across healthcare billing and revenue-cycle industry analyses. For example see: Beckers Hospital Review. Similarly CMS found 79% of incorrect billings were due to incomplete or inaccurate information. See: [Centers for Medicare & Medicaid Services](https://www.cms.gov/newsroom/fact-sheets/fiscal-year-2024-improper-payments-fact-sheet?utm\_source=chatgpt.com
[^2]: The 1-10-100 rule of data quality was introduced by George Labovitz and Yu Sang Chang in Making Quality Work (1992) and remains a standard reference point in data quality management.
[^3]: Wakefield Research and Monte Carlo Data did a 2022 survey that found 40% of data engineering time spent was on data quality checking. See Monte Carlo Data. Other claims, such as 53% of data engineering capacity gets consumed by maintenance and troubleshooting and the DataKitchen/Data.world data engineering burnout survey, support the same conclusion.
