SuperTruth and imaware: What 105,000 Diagnostic Records Taught Us About Cancer Data Trust

imaware builds at-home diagnostic tests. Their products measure biomarkers that matter for cancer detection, cardiovascular risk, autoimmune conditions, and metabolic health. By the time SuperTruth began working with them, they had collected over 105,000 diagnostic records.

They could not trust a single one of them.

Not because the records were fraudulent. Not because the lab work was poor. But because the data existed across inconsistent systems with no unified provenance, no standardized consent framework, and no way to query across sources without a team of data engineers spending weeks on each analysis.

That is the data trust problem in oncology diagnostics. The data exists. It is even good data. But without a trust layer, it cannot be used.

What SuperTruth Found

When SuperTruth standardized the imaware data and applied the Data Trust Index, the first thing that happened was speed.

Analysis that had taken three weeks took two hours. Not because the analysis got simpler, but because the data was now queryable, scored, and trustworthy enough to ask questions of directly. The 95 percent time reduction was not a technology trick. It was what happens when you can trust your data enough to move fast.

imaware: data processing time before and after SuperTruth DTI scoring

The second thing that happened was efficiency. 200 hours of manual work per month were eliminated. The work of tracing where records came from, validating that consent covered the intended use, checking whether findings corroborated across sources: all of it automated by the DTI scoring infrastructure.

The third thing that happened was discovery.

In the standardized, scored data, a customer segment emerged that had been invisible in the fragmented data: males aged 30 to 35, contributing 20 percent of imaware total revenue. This was not a demographic the company had explicitly targeted. It was a pattern that existed in the data, obscured by the inconsistency.

imaware CEO Brodie Flanders said it directly: the lab industry has never had a trust standard. DTI created one.

What the DTI Scored

For imaware diagnostic data, the eight DTI dimensions produced different insights at different weights.

Provenance (25 percent of the score) addressed the lab chain of custody. imaware tests are conducted through CLIA-certified labs. That certification established a provenance baseline. But the records from different tests on the same patient, run through different processing pipelines, had different provenance scores until they were standardized.

Consent (20 percent) addressed whether the explicit use permissions attached to each record covered the intended analysis. Consumer health data, once in a warehouse, tends to drift from its original consent scope. The DTI scoring process identified records whose consent language did not cover certain analytical uses.

Recency (15 percent) addressed how the data aged. A CBC drawn 18 months ago is not the same as one drawn last week, even if the numbers look similar. The time-decay applied differently to different biomarker types.

Concordance (10 percent) was where the most interesting patterns emerged. Records that corroborated across multiple imaware test types, or between test results and customer-reported symptoms, scored significantly higher than those that existed in isolation. High-concordance clusters were the most reliable inputs for population-level analysis.

The Oncology Diagnostic Landscape

imaware operates at the intersection of consumer health and clinical diagnostics. Many of the biomarkers they test have direct relevance to cancer screening and monitoring: PSA for prostate cancer risk, thyroid markers for thyroid cancer risk, inflammatory markers that appear in multiple cancer contexts.

The data trust problem in this context is not abstract. An elevated PSA from an untrusted source could trigger unnecessary intervention. A suppressed inflammatory marker from a record with poor provenance could create false confidence. The trust score is not a quality-of-life improvement for data teams. It is a patient safety question.

SuperTruth's work with imaware established that the same infrastructure that makes data trustworthy for business intelligence also makes it trustworthy for clinical intelligence. The DTI dimensions do not change based on the intended use. Provenance, consent, recency, quality, concordance, validation, breadth, and stability are the right questions whether you are running a revenue analysis or building a cancer screening model.

What This Means for Health and Wellness Organizations

The imaware case is not unique. Across direct-to-consumer health, digital health platforms, and at-home diagnostics, the same pattern repeats: rich data, poor infrastructure, no trust layer.

Health and wellness organizations sit on behavioral data, biomarker data, wearable data, and self-reported symptom data that, with the right trust infrastructure, becomes population-level intelligence. Without it, the data stays locked in the warehouse, useful only to the data engineering team that built it.

SuperTruth works with health and wellness organizations to apply the same infrastructure imaware used. The Data Trust Index is not a healthcare-only tool. It scores health data from any source: consumer apps, wearables, at-home labs, coaching platforms, employer health programs.

Further reading: Read the full imaware case study

Further reading: See our federated research solution If your organization collects health or wellness data and wants to understand what that data is actually worth with a trust layer underneath it, that conversation starts at louis@supertruth.ai or (215) 918-4140.