01 · The Question
A Participant Has Withdrawn. What Do You Do With Their Data Now?
Withdrawal seems straightforward until the research team opens the dataset. Some information was collected months ago. Other data were obtained yesterday. A laboratory has already analyzed specimens. Some records may have been shared with collaborators. Perhaps the participant now wants everything deleted.
Should researchers remove the participant's row from the dataset? Keep it but stop collecting anything new? Delete identifiers? Exclude the participant from analysis?
There is no universal “delete everything” rule. The correct response depends on what the participant withdrew from, when the data were collected, the regulatory framework, what the consent promised, and how far the information has already moved through the research process.
03 · What You Need to Know
Withdrawal Creates a Data Boundary, Not an Automatic Eraser
Start by Recording Exactly What the Participant Withdrew From
Before changing the dataset, establish the scope of withdrawal. A participant may have stopped only an intervention while agreeing to follow-up, or they may have withdrawn from every component of the study.
OHRP recommends clarifying this distinction because other research activities may continue when the participant withdraws only from the primary interventional component and previously consented to those additional activities.
This is why partial withdrawal from a research study should be resolved before anyone begins deleting, retaining, or continuing to collect data.
If Withdrawal Is Complete, Stop New Identifiable Research Data Collection
For HHS-regulated research, OHRP states that withdrawal from all components requires investigators to stop research interactions and interventions used to obtain data and stop obtaining additional identifiable private information about the participant from other sources. That includes new identifiable information from records and other third parties.
Researchers should therefore distinguish data already in the research record from data that could still be obtained after withdrawal.
Pre-withdrawal data
Information obtained and recorded before the participant withdrew.
Post-withdrawal data
Additional information researchers would obtain after withdrawal, which may no longer be permissible without continued consent or another valid authorization.
OHRP Permits Retention and Analysis of Already Collected Data
For non-exempt HHS-conducted or supported research, OHRP interprets the regulations as allowing investigators to retain and analyze data already collected before withdrawal, including identifiable private information, provided the analysis falls within the IRB-approved protocol.
This does not mean every researcher must retain every withdrawn participant's data. OHRP also states that in research not regulated by FDA, investigators can choose to honor a participant's request to destroy the data or exclude them from analysis.
The research team therefore needs to know whether retention is merely permissible, scientifically necessary, promised in consent, required by another rule, or subject to an approved deletion policy.
FDA-Regulated Trial Data Must Remain
FDA takes a more specific position for FDA-regulated clinical trials. Its longstanding policy is that all data collected up to the point of withdrawal must be maintained in the study database and included in subsequent analyses as appropriate. FDA ties this requirement to maintaining complete clinical study data and an accurate assessment of the investigational product's safety and effectiveness.
For these studies, deleting a withdrawing participant's accrued trial data simply because they request it would conflict with FDA's data-retention policy.
Data status
Question after withdrawal
Typical consideration
Not yet collected
Can researchers obtain it after complete withdrawal?
Generally stop additional identifiable research collection covered by the withdrawal.
Already collected in HHS-regulated research
May it remain in the research record?
OHRP permits retention and analysis within the approved protocol.
Already accrued in an FDA-regulated clinical trial
Can it be removed?
FDA requires the data to remain in the trial database.
Participant requests deletion in non-FDA research
Can the request be honored?
Potentially, depending on protocol, consent, governance, and other obligations.
Already distributed to another approved researcher
Can it be recalled?
Retrieval may not be required or feasible, depending on the applicable framework.
Already incorporated into completed analysis
Must the analysis be undone?
Completed analyses generally are not treated as automatically reversible.
Do Not Quietly Keep Collecting Data From Other Sources
Researchers sometimes interpret withdrawal as meaning only that they should stop contacting the participant. Under OHRP guidance, complete withdrawal goes further: investigators must also stop obtaining additional identifiable private information for the study from sources such as medical records, education records, stored clinical specimens, or other third parties.
If the participant agreed only to stop the intervention while continuing record-based follow-up, that is different. The research team needs to document the actual scope of the participant's decision rather than infer it.
A Deletion Request Should Be Evaluated, Not Automatically Promised or Rejected
When a participant asks for deletion, determine whether the request can be honored under the governing framework. In non-FDA research, OHRP expressly notes that investigators may choose to destroy a participant's data or exclude them from analysis.
But the answer can change if data-retention obligations apply, if information has already been shared, if analyses have been completed, or if the consent explicitly described limits on withdrawal.
This is the practical counterpart to the question of whether participants can withdraw permission for continued use of their data . Researchers should answer the request based on the actual data lifecycle rather than a generic withdrawal policy.
Previously Distributed Data May Not Come Back
SACHRP states that withdrawal of stored specimens from future research does not generally require retrieval of specimens already distributed to secondary users, and completed analyses generally will not be destroyed or removed from datasets. It recommends explaining these practical limitations during consent.
Once data have legitimately moved to another approved research environment, a later withdrawal request may therefore stop some future uses without literally reversing every prior disclosure.
Data That Have Already Been Analyzed Create Another Boundary
A participant may withdraw after their observations have already contributed to a statistical model, coded qualitative themes, derived variables, pooled estimates, or other analytic products. At that point, removing one person's underlying record may not be equivalent to simply deleting a row.
The next question becomes whether data can be withdrawn after analysis has begun . The answer depends on the regulatory framework, study commitments, feasibility, and stage of analysis.
Anonymization Can Make Later Individual Removal Impossible
When data have been genuinely anonymized so that researchers can no longer connect a record to a particular participant, the team may no longer be able to identify which observations belong to the withdrawing individual.
This creates an important practical limit that should be disclosed when relevant. Whether participants can withdraw data after anonymization therefore depends heavily on whether the data can still be linked back to them.
The Consent Process Should Explain the Data Consequences Before Withdrawal Happens
OHRP recommends that consent materials explain whether previously collected data will be retained and analyzed after withdrawal. For FDA-regulated research, participants should be told that data collected up to withdrawal will remain in the trial database and be included in analysis.
This is much better than introducing the limitation only after someone asks to leave. A participant's right to discontinue participation is easier to understand when the consequences for already collected information are stated from the beginning.
Watch Out
Never delete a withdrawing participant's data reflexively. Depending on the study, doing so may conflict with regulatory retention requirements, the approved protocol, scientific-integrity obligations, or the study's data-management plan. Withdrawal should trigger a defined data-handling process, not an improvised deletion.
06 · What This Means for You
Treat Withdrawal as a Data-Management Event as Well as a Consent Event
A robust withdrawal procedure should tell the research team what to stop, what to retain, what to flag, what may be deleted, and who makes those determinations. This should not depend on whoever happens to receive the participant's email that day.
A simple data-handling framework
If the participant withdrew from all components
Stop future interactions, interventions, and additional identifiable research data collection as required.
If data were collected before withdrawal
Determine whether they may, must, or should remain under the applicable protocol, consent, regulatory, and privacy framework.
If the study is FDA-regulated
Retain already-accrued trial data in accordance with FDA requirements.
If the participant requests deletion in research where deletion may be possible
Evaluate the request against the consent promises, data status, IRB requirements, data-management plan, and other obligations.
If data have already been anonymized, distributed, analyzed, or otherwise transformed
Determine what remains technically and legally possible and communicate those limits accurately.
07 · A Quick Checklist
After a Participant Withdraws
Before changing the participant's data, check:
Document the effective withdrawal point and whether withdrawal is complete or partial when documentation is appropriate.
Stop future research activities and additional identifiable data collection covered by the withdrawal.
Inventory what data and specimens were obtained before withdrawal.
Determine whether the research is subject to FDA or another framework requiring retention of accrued data.
Review what the consent form promised about retention, deletion, future use, anonymization, and sharing after withdrawal.
Identify whether any data have already been analyzed, anonymized, distributed, published, or incorporated into derived results.
Evaluate any deletion or exclusion request under the applicable ethics, regulatory, privacy, and data-governance requirements.
Record the resulting data-handling decision so the entire research team applies the participant's withdrawal consistently.
09 · The Bottom Line
Withdrawal Stops Future Participation, but Existing Data Need Their Own Decision
The Bottom Line
After complete withdrawal, researchers should stop the prospective research activities and additional identifiable data collection covered by that decision, but they should not automatically delete information already collected.
Determine what happens to existing data from the applicable protocol, consent promises, regulatory and privacy requirements, and the data's current status. OHRP generally permits retention and analysis of pre-withdrawal data, FDA requires accrued clinical-trial data to remain, and some non-FDA studies may be able to honor participant requests for deletion or exclusion.
11 · Cite this Guide
How to Cite This Guide
This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.
Recommended (Field Guide)
APA
MLA
Chicago
Copy Citation
