Health1st AI Logo
Biostatistics & Reporting

SDTM/ADaM Dataset Agent

H1ST-AI-STAT-004 · Task 3.1

Every data cut shouldn't restart your mapping

When the datasets are late, everything downstream waits with them, your TLFs, your define, your submission, and every mid-study transfer means re-running work you thought was done. This agent maps collected CRF data to SDTM domains and derives the ADaM analysis datasets that feed your TLFs, applying CDISC Controlled Terminology and traceability metadata as it goes. Weeks of manual SAS mapping and derivation become a governed, repeatable pipeline that regenerates cleanly on every data cut.

The grind you know

Every data cut shouldn't restart your mapping

Hand-building SDTM domains and ADaM datasets is the slowest, most error-prone step you own, and every mid-study transfer drags you back to re-run and re-validate work you thought was finished. You know the findings that follow, inconsistent controlled terminology, an undocumented derivation, a broken trace, are exactly what Pinnacle 21 and FDA data-quality queries catch, so the pressure to get it right by hand never lets up.

Part of the Statistical Programming family

SDTM/ADaM, Define.xml, TLFs & analytics.

When the datasets are late, everything downstream waits with them, your TLFs, your define, your submission, and every mid-study transfer means re-running work you thought was done. This agent maps collected CRF data to SDTM domains and derives the ADaM analysis datasets that feed your TLFs, applying CDISC Controlled Terminology and traceability metadata as it goes. Weeks of manual SAS mapping and derivation become a governed, repeatable pipeline that regenerates cleanly on every data cut.

  • SDTM
  • ADaM
  • Define-XML 2.1
  • Pinnacle 21
Explore the Statistical Programming family

How it works

  1. 1

    Connect the source

    Point the agent at your EDC export or SDTM source and provide the study CRF annotations and mapping specification.

  2. 2

    Map and derive

    The agent generates SDTM domains, then derives ADaM datasets with controlled terminology and traceability metadata applied throughout.

  3. 3

    Package and validate

    Datasets are written to XPT, run through Pinnacle 21 conformance checks, and released with a findings report for programmer sign-off.

Capabilities

What it takes off your plate

35+ SDTM domains

Maps collected data to interventions, events, findings, and special-purpose domains including DM, AE, LB, VS, EX, CM, and trial-design datasets.

ADaM derivation engine

Produces ADSL, ADAE, ADLB, ADEFF, and ADTTE with documented derivation logic and full one-to-one traceability back to SDTM.

Controlled terminology

Applies the correct CDISC CT version to every coded variable and flags non-conformant values before they reach validation.

Incremental refresh

Re-runs transformation rules against each new data transfer so datasets stay current without a full manual rebuild.

What you get

  • SDTM domain datasets in SAS Transport (XPT) format
  • ADaM analysis datasets (ADSL, ADAE, ADLB, ADEFF, ADTTE) in XPT
  • Documented data-transformation and derivation specifications
  • Value-level metadata and traceability mapping to source

In practice

What this looks like on a real study

Where the agent shows up in the day-to-day of a live trial — the moments the grind usually lives in.

A mid-study data transfer lands

Your third interim transfer arrives with new sites and updated labs. Instead of re-running SAS mapping by hand, the stored transformation and derivation rules refresh SDTM and ADaM incrementally, so ADSL and ADAE stay current for the next TLF run.

Controlled terminology mismatch

A coordinator entered a lab test not on the study's CDISC CT version. The agent flags the non-conformant value before it reaches Pinnacle 21, so you correct it at the data-management level instead of chasing a conformance finding at submission.

Traceability question from a reviewer

An FDA reviewer asks how an ADEFF endpoint was derived. Because every ADaM variable carries documented derivation logic and one-to-one traceability to its SDTM source, you point straight to the value-level metadata rather than reconstructing the logic weeks later.

Proof

The impact on your study

0 errors
Pinnacle 21 error findings on generated datasets
ADaM IG v1.3
Analysis dataset model conformance
< 2 hrs
SDTM + ADaM generation for a Phase 3 study

Figures shown are pre-launch targets based on internal benchmarks, not guaranteed outcomes.

Works with your stack

  • Pinnacle 21

    Validation

Who it's for

Built for the teams who run trials

More from Biostatistics & Reporting

Peace of mind

Built to the standards inspectors expect

Every output is generated inside a validated, audit-ready platform, kept under human-in-the-loop control, and mapped to the regulatory and CDISC standards this agent supports.

  • SDTM
  • ADaM
  • Define-XML 2.1
  • Pinnacle 21

Frequently asked questions

See the SDTM/ADaM Dataset Agent on your study

Walk through it on your own workflow with a clinical-trials expert — no pressure, no obligation, and honest answers, including on the limits.