TrialPilot has two export paths, matched to two jobs. The Export Study action on the study dashboard produces de-identified datasets for your own analysis — survey responses, assessment results, wearable summaries, the roster, the audit trail. The Data management workspace produces trial-of-record output for studies run to that standard, including CDISC-oriented datasets a statistician or a submission pipeline can consume.
Both paths share one property: everything they produce is de-identified. Participants appear as study-scoped pseudonymous IDs, never as names or contact details — see Privacy and de-identification for how that separation works.
Export study data#
Select Export Study from the study dashboard and choose:
- Data type — which dataset to export (see the table below).
- Format — CSV (loads directly into Excel, R, or Stata), newline-delimited JSON, or a single JSON array.
- Start and end date — an optional time window. Leave the dates empty to export the full study window.
- Wide format — when the data type supports it, one row per participant per day with measures as columns, instead of one row per data point.
The export downloads as a zip containing the data file and a manifest describing what was produced.
| Data type | What it contains |
|---|---|
| Patient-reported outcomes | Daily and periodic survey responses. |
| Objective assessments | Results from guided physical assessments such as the six-minute walk, sit-to-stand, NASA lean, and timed up-and-go tests. |
| Wearable summaries | Daily aggregate wearable metrics per participant. |
| Participant roster | Enrollment status and metadata, keyed by pseudonymous ID. |
| Audit log | The Part 11 style trail of clinical data changes. Available to roles with audit-export permission. |
The export tells you what it left out
Trial-of-record and CDISC output#
For studies using the Data management workspace, the Export tab there produces CDISC-oriented output. CDISC is the standards body whose formats regulators and statisticians expect clinical trial data in; SDTM is its standard layout for collected study data.
A CDISC export is always tied to a specific database lock or snapshot, which makes it reproducible: the same lock always yields the same export. It produces:
- SDTM-aligned domain datasets — demographics, vital signs, adverse events, concomitant medications, and questionnaire domains, with adverse events and medications carrying their MedDRA and WHODrug coding.
- define.xml — the machine-readable data definition file that describes each dataset and its variables, so a receiving system knows how to interpret the export.
- A statistician-loadable CSV bundle — the same domains as plain per-domain CSV files.
Availability of trial-of-record output depends on your study configuration — a study that never uses the data-management workflow exports through the study dashboard path only.
Test data never leaks into a real export#
Exports respect the sandbox boundary described in Test mode:
- A draft study's export is always the watermarked sandbox export. The filename is prefixed
SANDBOX_, the first line of the data file carries a notice, and the manifest is flagged as test data — so a rehearsal dataset can never pass as a real one. - A published study's export excludes test data entirely. The only way to see the sandbox record after publication is to explicitly request it, and then only during its retention window — and what you get is still the watermarked sandbox export.
The retention window is finite
A typical end-of-study export#
- Resolve open queries and finish coding in Data management.
- Take the controlled lock — see Locking and closeout.
- Generate the CDISC export from that lock for the statistician or submission, and any dashboard exports your own analysis needs.
- Archive the downloaded files with your study records.