Synthetic Data for
Workflow Utility
in Clinical Trials
Literature Gap Analysis · May 22, 2026
synthetic-dataclinical-trialsworkflowPHI
Provenance Labs · Generated by Gambit & Rook Medium-High Confidence
Executive Summary

No paper directly addresses synthetic data as a PHI-avoiding substrate for pre-data-lock analysis code development. The literature recognizes simulation for trial prep, synthetic data validity, and pipeline validation — but no one has connected these into a "workflow utility" paradigm distinct from clinical mimicry.

JCTS Conversation

Taekman + Dadiz/Guillet (2024–2026) on simulation for trial preparation — operational focus, not data-centric.

Synth Data Methods

TrialSynth, GAN tutorials, validation studies — mature but optimized for clinical realism, not workflow utility.

Pipeline Validation

DosiTest, Petalcorin — validate workflows with simulated data, but target analytic outputs not code development.

The Gap

Synthetic data that exercises analysis code paths without PHI exposure: a novel concept with structural fidelity as its core insight.

The JCTS Conversation
Simulation for Trial Preparation

Key insight: JCTS editors are actively receptive to pre-trial methodology papers. But all three focus on operational simulation — not data-centric pipeline preparation.

Sources 1–3: Taekman 2026, Guillet/Dadiz 2026, Dadiz et al. 2024 — all in JCTS
Synthetic Data Methods
Mature Generation, Wrong Objective
TrialSynth (2024)

VAE + Hawkes Processes. State-of-the-art fidelity for time-sequence clinical trial data. arXiv.

Azizi et al. (2021)

Validates synthetic data as statistical proxy for real trial data. BMJ Open. 94 citations.

Mangino et al. (2026)

GAN tutorial for synthetic clinical prediction data. JCTS.

Zazzetti et al. (2025)

Longitudinal synthetic data for breast cancer translational research. JCO CCI.

The misalignment: Every paper evaluates synthetic data by clinical realism. None asks: what if fidelity to real patients isn't the goal?

Sources 4, 5, 7, 8
Pipeline & Workflow Validation
The Closest Conceptual Neighbors

But: DosiTest evaluates existing workflows for uncertainty, and Petalcorin targets exploratory analytics. Neither addresses pre-data-lock code development.

Sources 6, 9: Kayal et al. 2022 (Physica Medica), Petalcorin 2026 (medRxiv)
The Literature Gap
Workflow Utility ≠ Clinical Mimicry
Clinical MimicryWorkflow Utility
Goal: synthetic ≈ real patientsGoal: synthetic exercises analysis code
Requires distributional fidelityRequires structural fidelity only
High generation complexityLower bar — simpler methods viable
Evaluated by statistical similarityEvaluated by pipeline readiness

Structural fidelity: matching variable names, data types, visit structures, missingness patterns, and analysis metadata. Enough to write and validate all analysis code before data lock — without touching PHI.

Synthesis of Sources 1–9
JCTS Is the Natural Venue
Why This Journal?

Strategic fit: This paper would extend the existing JCTS conversation from operational simulation into the synthetic data domain — a natural and novel bridge.

Sources 1–4: all JCTS papers
Risks, Gaps & Uncertainty

Recommended Next Actions

Sources & Process Provenance
9 sources · 5 Tier 1 · 4 Tier 2

Tier guide:  Tier 1 = Primary institutional source (peer-reviewed journal or official publication)  ·  Tier 2 = Preprint or non-peer-reviewed source — verify independently

Generated May 22, 2026 · DeepSeek V4 Pro · research-rapid v1.0 · Gambit / brief-to-slides v2.0