A CRO change mid-study puts the dataset, the audit trail and the site relationships at risk. Here is a sponsor's plan for what to demand, how to cut over on one date, and how Capture fits for the part of the study that comes next.
Free sandbox · No credit card · 21 CFR Part 11 aligned
Weeks
New data in the new system from this date only
What to know first
The situation
Sponsors change CRO mid-study for ordinary reasons: a missed enrolment target, a key project manager leaving, a cost dispute, an acquisition, or a decision to take data management in-house after the first phase. The handover is difficult because the CRO usually holds the clinical database, the monitoring history and the query log, and its incentive to make the transition smooth is weaker than yours.
What goes wrong is rarely dramatic. A dataset is exported without its audit trail. A query log is missing for the first six months. Site staff keep logging in to the old system for weeks after cutover. Two versions of a form exist and nobody can say which was used for which participant. These are all documentation failures, and each of them shows up later as an inspection finding or an analysis delay.
A transfer of obligations is also a regulatory event. In the US, transferring trial obligations to a CRO has to be in writing (21 CFR 312.52), and ICH E6(R2) expects trial-related duties transferred to a CRO to be specified in writing; take the same care when the duties come back to you. Check the specifics with your regulatory and quality teams. The wider decision is covered in how to choose a CRO, and the EDC side in how to switch EDC vendors mid-study.
Before you give notice
Put these in the transition agreement. The wording is for your legal team; the list is for your project team.
Every form, visit and participant, in a documented structure with a data dictionary, not only a final spreadsheet.
The full audit trail for all records, readable without the old system. An export of values alone does not meet the need.
Open and closed queries, SDV status, monitoring visit reports and deviation logs.
CRF specifications, edit check specifications, randomization specification and user lists with roles.
Handled with care to keep the blind intact; agree who holds the unblinded materials.
Signed consents, regulatory documents and the site contact list.
A person, a date for last data entry in the old system and a date for archive delivery.
The cutover
Data entered before the cutover stays in the legacy archive; data entered after it goes into the new system. The plan below keeps the two cleanly separate and documents the join.
Sites trained
6 of 8
Open legacy queries
14
Options
Choose per study, not per company. Capture's role is on the new-system side.
| Option | What happens to existing data | Risk | Where it fits |
|---|---|---|---|
| Finish the study with the CRO | Stays where it is | Low data risk; keeps the relationship you are trying to leave | Studies near database lock |
| New study or cohort in-house | Legacy archive stays complete and separate | Low; needs a merge plan at analysis | Extensions, next phases, new cohorts |
| Cutover mid-study, archive legacy | Archived with its audit trail; new entries in the new system | Medium; needs mapping and reconciliation | Studies that cannot wait |
| Re-key legacy data into the new system | Copied manually | High; transcription errors and a broken audit history | Rarely worth it; discuss with QA first |
Capture states no import tool on this page. If you need to load historical records, ask us what is supported before you plan around it.
The free sandbox includes every feature. Upload the protocol, let the AI draft visits and forms for your review, and test everything with sample data while the transition agreement is being negotiated.
Going in-house
Taking data management in-house does not require building a department. It requires a system your team can configure and run without a services contract: forms, edit checks, queries, roles, randomization and exports, all from one place. Capture is self-serve, so a small team can build and amend the study itself, with AI drafting visits and forms from the protocol and every output reviewed by a person before it is saved. See clinical trial software without a CRO and how to run a clinical trial without a CRO.
You still need people for the work the CRO did around the software: monitoring, medical review, safety reporting and site management. Decide which of those stay outsourced before you leave. A partial handover to a smaller regional CRO that uses your system is often a sensible middle path; see EDC for CRO-independent trials.
Finally, keep your own record. A dated transition file with the agreement, the export receipt, the mapping and the cutover notice is the document an inspector will ask for first.
This page does not claim an import or migration tool. Capture provides exports. Historical data is normally archived in its original export and merged at analysis. Ask us what is supported before planning around an import.
The complete data export, the audit trail, query and monitoring history, configuration documents, user lists and a named handover contact, all in writing before you give notice.
Often not. If the study is near lock, finishing with the CRO and moving the next study or cohort is lower risk. Switch mid-study only when you have to, with a documented plan.
It depends on the study, the contract and the number of sites. The timeline on this page is illustrative, not a promise; the new study can be built and tested in the sandbox while you negotiate.
Not into the new system. The legacy audit trail should be exported and archived with the legacy data. The new system starts its own trail from the cutover.
Capture is self-serve and the sandbox is free, so a small team can try it. You still need people for monitoring, safety and medical review, either in-house or outsourced.
Yes. Capture is suitable for Phase 1, 2 and 3 studies.
The sandbox is free with every feature and no credit card. You pay only once you go live with real participants. Talk to us for details; this page quotes no prices.
Keep exploring
How to switch EDC vendors mid-study
The system side of a transition.
How to choose a CRO
Selection and oversight.
Clinical trial software without a CRO
Running the data yourself.
EDC for CRO-independent trials
Independent and sponsor-run studies.
Database lock and unlock workflow
Closing the legacy dataset.
Export EDC data to R, SAS and SPSS
Getting data out for analysis.
Free sandbox, every feature, no credit card. Build and test before cutover.