This post was written by Rachel Witalec, Redox’s Chief Product Officer.
In our last “Connections” post, we discussed how interoperability isn’t a one-time destination—it’s an ongoing journey. In today’s world of fragmented systems and shifting requirements, the only way forward is a flexible foundation: an interoperability platform that can bend without breaking as you connect more tools and data sources.
Even with a flexible foundation in place, the data journey is far from smooth. You expect data to arrive ready to go. Instead, it often hits turbulence along the way—different formats, mismatched code sets, and missing context. Your data arrives like you do after a red-eye flight—disheveled and exhausted. And just like you need a day to recover, your data needs hours of manual cleanup, mapping, and reconciliation before it’s remotely useful.
What if your data could arrive in the right condition from the start? This could include normalization of clinical codes, adding missing context from third party sources, or validating quality/accuracy. In this post, we’ll explore how solving data quality problems “in-flight” (or upstream) doesn’t just save time—it creates a lasting strategic advantage.
To make this vision a reality, we’ve introduced new capabilities that let your data fly first class. It’s part of our evolution from simply powering connections to powering intelligence—where every transaction becomes an opportunity to improve data, not just move it. These enhancements make it possible to 1) enrich data in-flight to resolve quality issues and 2) orchestrate complex workflows across systems. The result? Your data arrives at its destination clean and ready to get to work.
Today, we’ll focus on in-flight enrichment. Stay tuned for more on orchestration (and follow us on LinkedIn to be the first to know).
Data enrichment is essential…and exhausting
Clean, normalized data is essential, and the success of downstream processes depend on it. But healthcare data comes from everywhere—EHRs, labs, payers, digital health apps, devices—and each source brings its own formats, structures, and terminology.
To make matters worse, much of the most valuable clinical information doesn’t live in tidy rows and columns. Physician notes, discharge summaries, and care plans are filled with critical insights that are often stored as unstructured or semi-structured data. Unlocking that information requires natural language processing (NLP) and clinical concept extraction to identify medications, diagnoses, and more buried in unstructured formats.
Even when data is properly coded, consistency remains elusive. There are more than a dozen major clinical codesets in play, each designed for a specific purpose. Consider some of the most widely adopted codesets:
Source: IMO Health, October 2025. https://www.imohealth.com/health-it-101/
It is exhausting (and expensive!) to go through the effort of developing data integrations to access the right data, just to realize the data you receive requires further enriching in order to be useful. It begs the question of when the data cleanup should happen, before or after you receive it?
Post-processing of data causes delays
Once the data has arrived and is sitting in your database or application, the mess is baked in. Just like the ground crew needs time to clean a plane between flights, now your team is faced with an endless backlog of “data janitorial work.”
Data engineering teams often report spending upwards of 60-80% of their time just cleaning and organizing data, before useful analysis or ML even begins. This may include writing transformation logic, maintaining brittle mapping tables, and chasing down mismatched codes across systems that were never designed to work together.
The costs don’t stop at wasted hours. Every new vendor or app introduces another one-off workflow to maintain, creating further delays as teams work to normalize and stitch together records. The risks go beyond inefficiency: inconsistent or incomplete data slows interventions, undermines decision-making, and can even compromise patient safety.
Take a proactive approach to data quality
Imagine if the plane were cleaned while still in the air, so by the time it landed passengers could board right away without waiting for the ground crew. In the same way, you can address data quality issues “in-flight”: cleaning, standardizing, and enriching information while it’s still moving, before it lands in your systems.
That’s exactly what Redox is designed to do. By partnering with industry leaders like IMO Health, a clinical data intelligence business, and Verato, identity intelligence experts, we ensure the heavy lifting happens upstream.
In-flight enrichment acts as a data quality layer, improving information in real-time as it moves across systems and proactively addressing some of the toughest data quality challenges:
- Codeset normalization: Through IMO Health, data is normalized across standards such as ICD-10, SNOMED CT, LOINC, CPT, and RxNorm, ensuring that every record arrives mapped to standardized clinical vocabularies.
- Free-text extraction: Through IMO Health, natural language processing (NLP) transforms unstructured information like physician notes or discharge summaries into standardized, structured, queryable data.
- Missing context: When critical details are missing (like lab results, allergy histories, or immunization records), Redox can reach out to Health Information Exchanges (HIEs) in real time to fill those data gaps, bundling enriched data back into the same format (for example, FHIR).
- Patient identity management: Partnering with Verato, Redox ensures every record is accurately linked to the right individual, resolving duplicates and identity mismatches as data flows across systems.
The result is a complete, enriched patient record delivered directly into your system. Your teams gain immediate access to high-quality data that drives better care, sharper analytics, and faster decisions. This is the first-class upgrade your data deserves.
High quality data is now a competitive edge in the age of AI
Clean, enriched data isn’t just about efficiency—it’s the foundation for the next generation of AI-powered healthcare. The organizations that get this right today will define what’s possible tomorrow. Algorithms are only as strong as the information they’re trained on and the signals they process. If data is messy, incomplete, or inconsistent, AI outputs suffer—creating risk instead of value.
Margins are thin, regulations are tightening, and patient expectations are rising. Healthcare organizations that can act on data in real time will outpace those still mired in downstream clean-up.
The benefits of in-flight enrichment are clear:
- Speed: Less data janitorial time accelerates interventions, reporting, and AI-driven insights. This means faster reimbursements, earlier clinical decision support, and more agile operations.
- Accuracy: Data arrives trustworthy and standardized, reducing errors, avoiding rework, and giving confidence to clinicians, operational leaders, and algorithms alike.
- Completeness: By tapping HIEs and other upstream sources, data arrives with the context needed to make better decisions, without chasing down gaps later.
- Efficiency: IT staff aren’t stuck in endless reconciliation, and clinical/analytics teams can focus on innovation, growth, and patient outcomes powered by AI.
In-flight enrichment isn’t just a technical upgrade; it’s about reclaiming your time and unlocking your potential. Your teams can spend less time cleaning up messy data and more time driving innovation—using clean, complete data to make real-time decisions that can save lives. That’s the payoff of getting it right.
Today’s first-class upgrade focuses on data quality. Next, we’ll explore how orchestration gives you the power to route and act on that clean data—automatically, intelligently, and at scale.
If your data still lands disheveled, it’s time to rethink your flight plan. Reach out to Redox to see how in-flight enrichment can help your data—and your business—arrive ready for action.