03 · What You Need to Know
Maintaining Research Contact Across Changing Participant Circumstances
Why Contact Changes Are a Methodological Problem
Longitudinal research depends on obtaining information from participants across defined periods. When contact details become outdated, researchers may lose opportunities to collect scheduled measurements.
These losses can reduce the amount of information available for analysis. More importantly, they may create selection bias if participants who become unreachable differ systematically from those who remain contactable.
For example, students who leave an institution before graduation may lose access to university email accounts. If these students also differ in academic engagement or educational outcomes, relying exclusively on institutional email could produce selective missingness.
Similarly, a study of employment experiences may lose participants who change jobs if follow-up depends on workplace contact information.
These examples illustrate plausible mechanisms, not universal relationships. Changing contact details does not necessarily imply poor outcomes or lower motivation.
The methodological concern is whether the circumstances associated with participant dropout are also related to the outcomes being studied.
Distinguish Unreachable Participants From Those Who Withdraw
A participant who cannot be contacted may still wish to remain in the study.
Researchers should distinguish several situations.
Temporary communication failure
An approved contact method does not work or a message is not delivered, but the participant's willingness to continue is unknown.
Missed follow-up
A scheduled assessment is not completed, whether because of communication difficulties or another reason.
Loss to follow-up
The required follow-up information cannot be obtained under the study's operational definition and permitted procedures.
Participant withdrawal
The participant communicates a decision to discontinue participation or further contact according to the applicable consent arrangements.
These categories should not be used interchangeably.
A bounced email does not establish withdrawal. Likewise, an unanswered telephone call does not prove that the participant has changed numbers.
Researchers should use the protocol's predefined definitions and record what is actually known rather than assigning speculative reasons for missingness.
Collect Contact Information That Can Remain Useful Beyond the Recruitment Setting
Contact information collected at baseline should reflect the expected duration and circumstances of follow-up.
For a short study conducted entirely within one institution, an institutional email address may be sufficient.
For a multi-year study involving students approaching graduation, relying exclusively on that address may be less appropriate.
Researchers might consider a participant-authorized personal email address or another stable communication method, subject to the study's privacy and consent requirements.
However, collecting more contact information is not automatically better. Each additional identifier creates responsibilities concerning access, security, retention, and potential disclosure.
The appropriate objective is to collect sufficient information for legitimate follow-up while respecting data minimization.
The decision to collect alternative contact methods at baseline should therefore be justified by the study's actual follow-up needs.
Establish a Secure Contact-Update Procedure
Participants should know how to update their contact details if they change during the study.
Possible procedures include a secure participant portal, an authorized research email address, a designated study telephone number, or updates collected during scheduled assessments.
The selected method should be accessible to the population and appropriate to the sensitivity of the information.
Researchers should consider how updated details will be verified, who may enter them into the contact database, and how obsolete information will be handled.
Where contact records contain sensitive identifiers, they should be protected through appropriate access controls and organizational security measures.
Study contact information should ordinarily be managed separately from analytical datasets where practical, using coded participant identifiers to connect records when necessary.
Separating contact information from research outcomes can reduce unnecessary exposure of identifiable data, although it does not make the information anonymous.
Confirm Contact Details at Appropriate Follow-Up Points
Contact verification can be incorporated into scheduled research interactions.
For example, participants may be asked whether their preferred email address or telephone number has changed when completing a routine assessment.
This can be more practical than waiting until all available communication methods fail.
However, verification should not become an additional source of excessive contact.
The frequency should reflect the study's duration, expected mobility, and participant circumstances.
A twelve-month study may reasonably use different verification arrangements from a study involving one follow-up after two weeks.
Researchers should integrate contact updates into the approved participant retention plan rather than introducing repeated unscheduled requests.
Use Alternative Contact Channels Only When Authorized
If the primary communication method fails, another approved channel may help maintain follow-up.
For example, a participant who has authorized contact through both email and telephone may still be reachable by telephone after changing email addresses.
However, permission to use one channel does not necessarily authorize every other method.
Researchers should not automatically switch to social media messaging, workplace communication systems, or personal contacts merely because these methods are technically available.
Such contact may disclose research participation to others or conflict with the participant's preferences.
The protocol should specify which alternative methods are permitted, the conditions under which they may be used, and the limits on contact attempts.
What About Contacting Family Members or Other Third Parties?
Some longitudinal studies consider collecting information about a trusted person who may help researchers reconnect with a participant.
This approach can create significant privacy and consent considerations.
Permission from the participant to name a contact person does not necessarily establish that the third party has agreed to receive communications or that researchers may disclose the participant's involvement in a sensitive study.
Any third-party contact procedure should have a clear purpose, appropriate authorization, and an approved communication script that minimizes disclosure.
Researchers should also consider whether contacting another person could expose the participant to interpersonal, employment, or safety risks.
In some studies, third-party tracing may be inappropriate regardless of its potential usefulness for retention.
Researchers should follow the applicable ethics and privacy requirements rather than treating third-party contact as a routine administrative convenience.
Can Researchers Search Public Records or Social Media for Missing Participants?
Public availability does not automatically make personal information ethically appropriate for research tracing.
Searching social media profiles, public directories, institutional records, or other databases to locate participants may create privacy concerns and may exceed the scope of the original consent.
The appropriateness of such procedures depends on the research context, applicable law, institutional authorization, and ethics approval.
Researchers should not independently expand tracing methods after a participant becomes unreachable.
If external record linkage or tracing is necessary, the protocol should define its purpose, permitted sources, safeguards, and limits before implementation.
The existence of a technically feasible search method is not sufficient justification for using it.
Relocation Does Not Necessarily Prevent Continued Participation
A participant may move away from the original recruitment location while remaining able and willing to provide research information.
Depending on the study, follow-up may be possible through secure online questionnaires, telephone interviews, approved remote assessments, or collaboration with authorized study sites.
However, remote procedures are not always equivalent to in-person measurement.
Some outcomes require physical examination, standardized equipment, or assessment conditions that cannot be reproduced remotely.
Researchers should evaluate whether alternative measurement arrangements preserve the intended construct, timing, and data quality.
Changes to assessment methods may require protocol amendments and should not be introduced solely to improve retention without considering measurement validity.
Changing Institutions Can Create Additional Follow-Up Barriers
Participants may leave the organization through which they were originally recruited.
Students graduate or transfer. Employees change workplaces. Patients receive care from different providers.
These transitions may affect access to institutional email accounts, research facilities, or authorized communication systems.
Researchers should consider whether follow-up can continue independently of the original institution and whether the relevant permissions allow that continuation.
For example, a university study might obtain an approved personal contact method before students graduate, rather than attempting to retrieve their new information from institutional records afterward.
Institutional transitions should be anticipated where reasonably foreseeable, particularly when they are related to the study population or outcomes.
Document Contact Attempts Without Overcollecting Information
Contact logs can help research staff avoid duplicate messages and determine whether approved follow-up procedures have been completed.
Useful information may include the date, authorized communication channel, delivery status where known, assessment involved, and whether the participant requested a change in contact preferences.
Researchers should avoid recording unnecessary personal details or speculative explanations for nonresponse.
For example, a log may accurately state that an email was returned as undeliverable. It should not infer that the participant deliberately avoided the research team.
Access to contact logs should be limited to authorized personnel, and retention periods should follow applicable requirements.
How Contact Changes Affect Missing-Data Analysis
When participants become unreachable, the resulting missingness may or may not be related to the study outcomes.
For example, residential relocation might be unrelated to a particular laboratory measurement but closely connected to employment or educational outcomes.
The implications depend on the research question and the circumstances producing missingness.
Researchers should examine whether contact failures differ across relevant baseline characteristics, study groups, or follow-up periods.
Such comparisons may reveal patterns but cannot establish the missing-data mechanism with certainty.
Statistical approaches such as multiple imputation or inverse probability weighting may be appropriate under specified assumptions, but they cannot guarantee correction for unobserved differences.
Preventing avoidable contact failures remains preferable to assuming that missing outcomes can be reconstructed later.
When Should Researchers Stop Trying to Reestablish Contact?
The protocol should establish reasonable limits on contact attempts and tracing procedures.
Attempts should stop when participants explicitly withdraw from further contact, when approved procedures have been exhausted, or when continued contact is otherwise inappropriate under the applicable arrangements.
Researchers should also consider whether the relevant assessment window has closed.
Continuing to pursue a measurement that can no longer provide scientifically valid information may impose burden without meaningful benefit.
Participants who miss one assessment may still be eligible for later follow-up if they have not withdrawn and continued contact remains authorized.
The appropriate frequency and limits of follow-up contact should be established before these situations arise.
Watch Out
Do not treat a participant's disappearance from an approved contact channel as permission to search for them through unrelated personal accounts, employers, relatives, or external databases. Any tracing procedure must be justified, authorized, proportionate, and consistent with the applicable consent and privacy requirements.
