03 · What You Need to Know
Establishing Approval Authority and Change Control for Research SOPs
Why Do Research SOP Changes Require Approval?
An SOP may govern procedures that affect participant eligibility, consent, intervention delivery, assessment administration, equipment operation, data recording, or correction of research records.
Changing these instructions can alter how the study is implemented.
For example, revising an assessment SOP to permit additional working time may affect score comparability. Changing a data entry procedure may alter how missing values are represented or how records are linked.
Approval provides a mechanism for determining whether a proposed revision is scientifically appropriate, operationally feasible, and consistent with existing authorizations.
However, not every editorial correction carries the same risk. Approval processes should distinguish minor document maintenance from substantive procedural changes.
Who Can Approve SOP Changes?
Approval authority depends on the institution, research design, applicable quality system, and nature of the proposed change.
| Authority |
Possible responsibility |
Important limitation |
| SOP owner |
Maintain the procedure and coordinate proposed revisions. |
Ownership does not necessarily confer authority to approve every substantive change. |
| Principal investigator |
Evaluate scientific implications and authorize changes within delegated authority. |
Cannot replace required independent ethics or regulatory approval. |
| Quality manager or designated reviewer |
Review document control, quality requirements, and implementation safeguards. |
May not have authority to change scientific commitments independently. |
| Technical specialist |
Evaluate equipment, software, measurement, or other specialist implications. |
Technical endorsement is not necessarily final governance approval. |
| Research ethics committee |
Review changes requiring ethics approval under the applicable framework. |
Not every minor SOP edit requires ethics committee review. |
| Sponsor or regulatory authority |
Provide required review or authorization for changes within its remit. |
Requirements vary by study type and jurisdiction. |
| Coordinating center |
Manage common procedures and cross-site implementation. |
Local institutional requirements may still apply. |
A clear approval matrix can establish which authority is responsible for each category of change.
What Is the Difference Between Drafting, Reviewing, and Approving an SOP?
These are distinct responsibilities, although smaller research teams may assign several roles to the same person.
The author prepares the proposed revision. A reviewer evaluates its technical, methodological, or operational suitability. The approver authorizes the document for implementation.
For example, a research assistant may identify a recurring problem and draft a clearer data entry instruction. A data manager may review the revised procedure, while the designated principal investigator or quality authority approves it.
For high-consequence changes, independent technical or scientific review may strengthen the process.
However, separation of duties should remain proportionate to the study's risks and available governance structure.
Which Changes Can Usually Be Approved Internally?
Internal approval may be sufficient when a revision remains within the approved research design and applicable institutional requirements.
Examples may include correcting typographical errors, clarifying responsibilities, improving the layout of instructions, or replacing an obsolete software screenshot without changing the procedure.
Some operational changes may also be authorized internally when the protocol already permits the relevant alternatives.
However, the research team should verify that a seemingly minor change does not alter a critical requirement.
For example, changing the wording of an assessment instruction may appear editorial but could affect how participants understand the task.
Classification should depend on the change's actual implications, not merely its apparent size.
When Is Principal Investigator Approval Necessary?
The principal investigator commonly has responsibility for scientific oversight and ensuring that the study is conducted according to its approved arrangements.
PI review is particularly important when a proposed SOP revision may affect research objectives, measurements, sampling, intervention implementation, or analysis.
For example, changing the procedure for resolving discrepant primary outcome scores may influence the dataset used for the principal analysis.
The PI should assess whether the change is consistent with the approved protocol and whether additional review is necessary.
However, PI approval alone does not necessarily authorize changes requiring ethics, institutional, sponsor, or regulatory approval.
When Does an SOP Change Require Ethics Review?
Ethics review may be necessary when a proposed change affects approved participant-facing procedures, risks, consent arrangements, confidentiality safeguards, or other matters within the ethics committee's jurisdiction.
For example, replacing an approved consent procedure with a materially different process may require prior review.
Changing participant eligibility or introducing additional sensitive questions may also have ethical implications.
Requirements differ among institutions and jurisdictions. Researchers should consult the conditions of their ethics approval and applicable policies.
Not every formatting correction or internal administrative revision requires a formal ethics amendment.
Where uncertainty exists, the appropriate ethics office or committee should clarify the required process before implementation.
Can an SOP Change Require a Research Protocol Amendment?
Yes. If the proposed SOP revision changes a substantive commitment in the research protocol, a protocol amendment may be necessary.
For example, an SOP cannot independently authorize a different primary outcome instrument when the approved protocol specifies the original measure.
Similarly, revising the intervention delivery procedure to remove an essential component may alter the study itself.
The research team should assess whether the change affects the research design, population, outcomes, measurement procedures, or analysis commitments.
If a protocol amendment is necessary, the SOP should remain consistent with the authorized revised protocol.
Researchers should not use operational documentation to bypass methodological or governance review.
How Should Proposed Changes Be Classified?
A proportionate change control system can distinguish revisions according to their consequences.
| Illustrative change category |
Example |
Possible approval route |
| Editorial |
Correcting a typographical error without changing meaning. |
Designated document control authority. |
| Routine operational |
Clarifying an existing data transfer instruction. |
SOP owner or authorized operational reviewer. |
| Technical |
Changing equipment settings or software configuration. |
Technical review and designated approval. |
| Methodologically consequential |
Changing assessment administration or scoring rules. |
Scientific review and any required protocol amendment. |
| Ethically consequential |
Changing consent or participant protection procedures. |
Relevant ethics and institutional review where required. |
| Regulated or contractually controlled |
Revising a procedure subject to external requirements. |
Applicable sponsor, regulatory, or other required authorization. |
These categories are illustrative rather than universal. Institutions may use different terminology or approval thresholds.
What Should Be Included in an SOP Change Request?
A change request should provide enough information for reviewers to understand the proposal and its implications.
Useful elements include the current SOP identifier and version, proposed revision, reason for change, affected procedures, scientific and ethical impact assessment, required approvals, and intended implementation date.
Where the change affects software or equipment, the request may include testing or validation evidence.
Where multiple sites are involved, the request should identify whether the revision applies centrally or only to particular locations.
Supporting documentation should be proportionate to the consequences of the change.
Should the Scientific Impact Be Assessed Before Approval?
Yes, particularly when the change could influence how research evidence is generated or interpreted.
For example, modifying the timing of a follow-up assessment may affect the estimated intervention effect or the interpretation of longitudinal change.
Reviewers should examine whether the revision alters measurement comparability, intervention exposure, participant selection, or other important design features.
They should also consider whether data collected before and after implementation remain comparable.
Some changes may require sensitivity analysis or additional reporting even when they are approved prospectively.
Approval establishes authorization, not proof that the change has no scientific consequences.
Can Research Assistants Propose SOP Changes?
Yes. Personnel implementing procedures may be particularly well positioned to identify ambiguous instructions, recurring errors, or impractical requirements.
Research teams should provide a clear process for proposing improvements.
However, the authority to recommend a change is distinct from the authority to approve and implement it.
For example, a research assistant may recommend revising a data entry procedure after observing repeated identifier mismatches.
The proposal should be evaluated by the appropriate responsible personnel before the revised instructions become effective.
Encouraging staff to report procedural weaknesses can improve quality without permitting uncontrolled changes.
What Happens When an Urgent Change Is Necessary?
Some circumstances require immediate action, particularly when continuing the existing procedure could endanger participants or compromise safety.
Applicable governance frameworks may permit urgent protective measures before the ordinary amendment process is completed.
However, researchers should not assume that every operational inconvenience qualifies as an emergency.
Urgent action should be limited to what is necessary, documented promptly, and reported or reviewed according to applicable requirements.
For example, an equipment safety failure may justify suspending a measurement procedure immediately rather than continuing to follow an unsafe SOP.
Subsequent authorization may be necessary before a revised procedure becomes routine.
Should SOP Changes Be Approved Before Implementation?
Generally, yes. Nonurgent changes should complete the required review and authorization process before personnel begin using the revised instructions.
Implementing an unapproved procedure can create uncertainty about which instructions governed particular observations.
For example, if one site begins using revised scoring rules before central approval, its results may become difficult to compare with those from other sites.
Where urgent protective action is necessary, the applicable exception and reporting process should be followed.
Approval should therefore precede ordinary implementation, with narrowly justified exceptions for circumstances requiring immediate action.
How Should SOP Versions Be Controlled?
Each controlled SOP should identify its version, approval status, effective date, and responsible owner.
When a revised version becomes effective, personnel should be able to distinguish it from superseded instructions.
Earlier versions may need to be retained to establish which procedure governed previous research activities.
For example, if an assessment SOP changes halfway through a longitudinal study, the research record should identify which participants were assessed under each version.
Version control supports traceability and allows researchers to evaluate whether procedural changes affect interpretation.
Should Personnel Be Retrained After Every SOP Change?
Training requirements should reflect the significance of the revision.
A minor spelling correction may not require formal retraining. A change to assessment administration, equipment operation, or data correction procedures may require practical instruction and competency reassessment.
Researchers should identify affected roles and ensure that personnel understand the revised procedure before independently implementing it.
Training records should identify the relevant SOP version when necessary.
The objective is demonstrated readiness, not automatic attendance at another training session for every editorial update.
Who Approves SOP Changes in Multisite Research?
Multisite studies may require coordination between the central research team and local institutions.
A coordinating center may approve changes to master SOPs, while local authorities manage permitted site-specific supplements.
However, changes affecting local ethics approvals, institutional systems, or regulated activities may require additional site-level authorization.
The approval process should establish which changes can be made locally and which require central review.
Researchers should avoid allowing sites to independently alter critical procedures without evaluating cross-site comparability.
The principles for standardizing SOPs across research sites
How Do SOP Changes Relate to Registered Reports?
Registered Reports involve prospective peer review of consequential research design and analysis decisions.
After in-principle acceptance, changes to accepted methodological commitments may require editorial consultation or approval according to the journal's policies.
However, routine SOP revisions that do not alter the accepted design may not require journal review.
The research team should distinguish administrative document maintenance from changes affecting the registered protocol.
Where a consequential change is necessary, the procedures for design changes after Stage 1 acceptance may apply.
What If an SOP Change Is Rejected?
A rejected proposal should not be implemented merely because the personnel involved believe it would be more convenient.
The decision should identify the reason for rejection and whether additional evidence or a different proposal could address the concern.
For example, a proposed assessment platform may be rejected because its scoring equivalence has not been established.
The team may retain the existing procedure, seek additional validation evidence, or propose an alternative solution.
If the current SOP is unsafe or impossible to implement, the issue should be escalated rather than ignored.
How Can Change Control Avoid Becoming Excessively Bureaucratic?
Approval processes should be proportionate to the consequences of the proposed revision.
Requiring full committee review for every formatting correction may consume resources without improving research quality.
Conversely, allowing consequential methodological changes through informal messages may undermine accountability.
A tiered approval matrix can establish streamlined review for minor changes and stronger oversight for substantive revisions.
Researchers should periodically assess whether the process remains effective, timely, and consistent with applicable requirements.
How Should SOP Changes Be Reported in the Final Study?
Changes that materially affect measurement, intervention delivery, sampling, or analysis should be described sufficiently for readers to understand their implications.
Minor editorial or administrative revisions may not need to appear in the manuscript.
Researchers should distinguish approved prospective amendments from earlier unplanned deviations.
A change approved after an earlier departure does not automatically make that departure compliant with the instructions previously in effect.
For further consideration, see documenting research SOP deviations.
Watch Out
Principal investigator approval does not automatically replace ethics, institutional, sponsor, editorial, or regulatory authorization. Before implementing a consequential SOP revision, verify which authorities have jurisdiction over the change and whether the research protocol must also be amended.
