How to Handle Suspected Data Fabrication in Your Lab

By Dr. Zubair Khalid, DVM, MS, PhD · 2026-08-28

Quick Answer

• Suspected data fabrication requires immediate documentation of observations and secure preservation of original records before any conversation with the accused individual.
• Report concerns through your institution's designated research integrity officer or office, following the formal process outlined in your institutional policy.
• The process is confidential and fact-finding, not accusatory, and you should not confront the individual or share suspicions with colleagues.

Understanding Data Fabrication in Research

Data fabrication is the making up of data or results and recording or reporting them as if they were real. This is distinct from data falsification, which involves manipulating or changing existing data to obtain a desired result, and from plagiarism, which is the appropriation of another person's ideas, processes, results, or words without giving appropriate credit. The Committee on Publication Ethics Core Practices identifies data fabrication and falsification as forms of research misconduct that journals, institutions, and funding agencies are expected to address through clear policies and procedures.

Research institutions and funding agencies treat fabrication as a serious breach of professional conduct because it undermines the reliability of the scientific record and erodes public trust in research. The National Institutes of Health Grants and Funding pages describe the expectations for responsible conduct of research that apply to all NIH-funded investigators, including the accurate recording and reporting of results. When you suspect fabrication, you are not being asked to prove it yourself. Your responsibility is to report a concern in good faith so that the institution can conduct a fair and thorough inquiry.

The distinction between a genuine error and intentional fabrication matters for how you respond. A lab notebook entry with a transcription mistake, a calculation error, or a mislabeled sample is a quality problem that can be corrected through normal lab management. Fabrication is different because it involves intent to deceive. You cannot know intent from a single observation, which is why the reporting process is designed to gather information before any conclusion is reached.

At a Glance

Core Principles of Research Integrity

Research integrity rests on a set of shared expectations that apply to every person involved in the research process. The Committee on Publication Ethics Core Practices list the responsibilities of authors, reviewers, editors, and institutions in maintaining the integrity of the published record. For a lab setting, these principles translate into concrete practices that make fabrication harder to hide and easier to detect.

Authorship and Accountability

Authorship carries responsibility for the accuracy and integrity of the work. Every person listed as an author on a paper must have made a significant contribution to the work and must be prepared to take public responsibility for the content. The Committee on Publication Ethics Core Practices state that all authors should agree to be listed and should approve the final version of the manuscript. When you suspect fabrication, the authorship question becomes relevant because the person who fabricated data may be an author, and the other authors may not have seen the raw data.

Lab managers should make authorship expectations clear at the start of a project. A written authorship agreement that states who will be an author, what contribution each person will make, and who will hold the raw data can prevent disputes later. This agreement does not prevent fabrication, but it does establish a baseline of accountability that makes it easier to identify who is responsible for which parts of the work.

Data Management and Sharing

The NIH Data Management and Sharing Policy requires researchers to plan for how data will be managed and shared. The policy applies to NIH-funded research and expects researchers to describe their data management practices in a plan submitted with their application. For a lab, this means having a written data management plan that covers where raw data are stored, who has access, how files are named and versioned, and how long records are kept.

A well-managed lab has a clear chain of custody for data. Raw instrument files should be stored in a read-only location, analysis scripts should be versioned, and the final dataset used for a paper should be archived. When a suspicion of fabrication arises, the existence of these records makes it possible for the institution to conduct a meaningful review. Without them, the inquiry has nothing to examine.

Researcher Identity and Records

The ORCID for Researchers pages describe how researchers can maintain a persistent digital identifier that links their work across publications and grants. While ORCID is primarily a tool for identity and record keeping, it also supports research integrity by making it easier to verify that a person's publication record is accurate. A researcher who has a complete ORCID record is less able to claim work that is not theirs, and the record provides a transparent history of their contributions.

The Role of Reporting Guidelines

Reporting guidelines are checklists that tell authors what information should be included in a research paper so that the methods and results are transparent and reproducible. The EQUATOR Network maintains a comprehensive collection of these guidelines for different study types, including randomized trials, observational studies, and systematic reviews. When a lab uses the appropriate reporting guideline, the paper must describe the methods in enough detail that a reader can understand how the data were collected and analyzed.

Reporting guidelines do not prevent fabrication, but they make it more difficult to hide. A paper that follows the CONSORT guideline for a randomized trial must report the number of participants screened, the number randomized, and the number analyzed at each stage. If the reported numbers do not match the raw data, the discrepancy is visible to a careful reader. The EQUATOR Network provides a searchable database of guidelines, and lab managers should identify the relevant guideline for each project and require authors to follow it.

Practical Workflow for Handling Suspected Fabrication

The process for handling a suspicion of fabrication follows a structured path that protects the reporter, the accused, and the integrity of the research. The steps below are based on the expectations described in the Committee on Publication Ethics Core Practices and the institutional responsibilities described in the NIH Grants and Funding pages.

Step 1: Document What You Observed

The first action is to write down what you observed, when you observed it, and why it raised a concern. This documentation should be factual and specific. For example, you might write that a data file contains identical values for what should be independent measurements, or that a figure in a draft paper does not match the raw data file. Include the date, the time, the file names, and the location of the original data.

Do not alter or delete any files. Preserve the original data in its current state, and make a copy of the relevant files for your own records. If the data is on a shared drive, do not move or rename the files. The institution will need to examine the original records, and any change you make could compromise the review.

Step 2: Review the Concern with a Trusted Colleague

Before filing a formal report, discuss the concern with a trusted senior researcher who is not involved in the project. This person can help you determine whether the observation is a genuine concern or a misunderstanding. For example, a data pattern that looks suspicious to you might have a legitimate explanation that you have not considered.

This discussion should be confidential. Do not discuss the concern with other lab members, the accused person, or anyone outside the small circle of people who need to know. The purpose of this step is to test your own judgment, not to gather support for an accusation.

Step 3: File a Formal Report

If the concern survives the review, file a formal report with the institutional research integrity office. The Committee on Publication Ethics Core Practices state that institutions should have clear procedures for handling allegations of misconduct, and the report should be made through that formal channel.

Your report should include your documentation, the names of the people involved, and the specific data or records that raise the concern. You do not need to prove that fabrication occurred. You need to describe what you observed and why it raises a concern.

Step 4: Cooperate with the Inquiry

After you file a report, the institution will conduct an inquiry to determine whether the concern warrants a full investigation. The inquiry is a fact-finding process, not a judgment. You may be asked to provide additional information, to identify the original data, or to explain your observations.

Cooperate fully with the inquiry. Provide the records you have preserved, answer questions honestly, and do not discuss the case with anyone outside the process. The institution will protect your identity as a reporter to the extent permitted by law and policy.

Records and Measurements

The quality of an investigation depends on the quality of the records available. A lab that maintains complete and accurate records makes it possible to determine whether a concern is valid. A lab with poor records leaves the institution with no way to verify the data.

The NIH Data Management and Sharing Policy describes the expectations for data management and sharing that apply to NIH-funded research. The policy expects researchers to describe how data will be preserved and shared, and it expects the data to be available for verification. A lab that follows this policy has a clear record of what data exists and where it is stored.

Common Failure Patterns in Handling Fabrication

The most common failure in handling suspected fabrication is not the fabrication itself but the response to it. Lab managers and researchers often make mistakes that turn a manageable concern into a serious problem.

Confronting the Accused Person Directly

A lab manager who confronts the accused person directly may be acting on good intentions, but the approach is risky. The accused person may destroy evidence, may pressure the reporter to withdraw the concern, or may create a hostile environment in the lab. The Committee on Publication Ethics Core Practices describe the institutional process for handling allegations, and that process is designed to protect all parties. Confrontation outside that process is not a substitute for it.

Promising Confidentiality to the Reporter

A lab manager who tells a reporter that the report will be kept confidential may be making a promise that the institution cannot keep. The institution may need to share the report with the accused person, with the research integrity office, or with a funding agency. The NIH Grants and Funding pages describe the responsibilities of institutions to handle allegations of misconduct, and those responsibilities may require disclosure of the report.

Destroying or Altering Evidence

A lab manager who deletes a file, edits a notebook, or changes a data record to "clean up" the situation is committing a separate act of misconduct. The evidence must be preserved in its original state. The NIH Data Management and Sharing Policy expects that data will be preserved and made available, and the same expectation applies to an investigation.

Delaying the Report

A lab manager who waits to see whether the concern resolves itself is taking a risk. The delay gives the accused person time to destroy evidence, and it makes the investigation more difficult. The Committee on Publication Ethics Core Practices expect that allegations will be handled promptly, and a delay can be seen as a failure to act.

Welfare and Safety Context

The handling of suspected fabrication has a welfare dimension that is often overlooked. The reporter, the accused, and the lab team all experience stress during an investigation, and the lab manager has a responsibility to support the people involved.

The reporter may fear retaliation, loss of position, or damage to their career. The institution should have policies that protect reporters from retaliation, and the lab manager should be aware of those policies. The NIH Grants and Funding pages describe the responsibilities of institutions to protect whistleblowers, and the lab manager should ensure that the reporter knows about these protections.

The accused person is also in a difficult position. The person may be innocent, and the accusation may be a misunderstanding. The institution's process is designed to determine the facts, and the lab manager should treat the accused person with respect until the process is complete. The lab manager should not isolate the accused person, should not discuss the case with other lab members, and should not make any judgment before the institution has reached a conclusion.

The lab itself is affected by the investigation. The work may be paused, the team may be distracted, and the trust within the team may be damaged. The lab manager should maintain a professional environment, should keep the team focused on the work that can continue, and should communicate only what the institution permits.

Limitations and Professional Escalation

The process described in this article is a general framework, and the specific requirements vary by institution and by jurisdiction. The Committee on Publication Ethics Core Practices describe the expectations for institutions, but they do not prescribe a single procedure. Your institution may have a specific office, a specific form, and a specific timeline for handling allegations.

The NIH Grants and Funding pages describe the requirements for NIH-funded research, but they do not apply to all research. If your research is not funded by NIH, the requirements may be different. The NIH Data Management and Sharing Policy applies to NIH-funded research, and the expectations for data management may be different for other funders.

The EQUATOR Network provides reporting guidelines, but the guidelines are not regulations. A journal may require a specific guideline, but the requirement is a journal policy, not a legal requirement.

When the concern involves a publication, the journal may have its own process for handling allegations. The Committee on Publication Ethics Core Practices describe the responsibilities of journals to handle allegations of misconduct, and the journal may contact the institution to request an investigation. The lab manager should be aware that the journal may be involved, and the institution will coordinate with the journal as needed.

Decision Framework for Triage and Escalation

A structured triage framework helps lab managers and PIs sort observations into three response categories before any formal report is filed. This framework prevents both overreaction to benign discrepancies and underreaction to genuine fabrication concerns. The Committee on Publication Ethics Core Practices describe the institutional responsibilities for handling allegations, and a triage step ensures that only appropriate concerns reach the formal process.

Category One: Quality Control Issue

The first category covers discrepancies that have a plausible non-intentional explanation. Examples include a mislabeled sample, a transcription error in a notebook, a calculation mistake in a spreadsheet, or an instrument calibration problem. These issues are handled through normal lab management. You document the discrepancy, review the protocol, and correct the process. No institutional report is needed.

The decision rule for this category is simple. If you can identify a mechanism by which the discrepancy occurred without intent, and the discrepancy is isolated instead of repeated across independent experiments, treat it as a quality issue. Correct the record, note the correction in the lab notebook, and move on.

Triage Category: Possible Fabrication

This category applies when the discrepancy has no plausible innocent explanation or when the pattern repeats across multiple experiments. Examples include identical values across independent replicates, results that are too consistent to be biologically plausible, or a figure that does not match the raw data file. The NIH Data Management and Sharing Policy expects that data will be preserved and available for verification, and this category is where that expectation becomes operational.

The decision rule for this category is to document the observation, preserve the original records, and discuss the concern with a trusted senior researcher who is not involved in the project. This discussion is confidential and is designed to test your own judgment. If the concern survives this review, you move to the formal reporting step.

Category: Confirmed or Admitted Fabrication

This category applies when a person admits to making up data or when you have direct evidence that data were fabricated. The admission may come during a private conversation, or you may have access to files that show the fabrication directly. This category requires immediate escalation to the institutional research integrity office. The Committee on Publication Ethics Core Practices state that institutions should have clear procedures for handling allegations, and this category triggers those procedures without delay.

The decision rule for this category is that you do not need to investigate further. You do not confront the person, you do not gather more evidence, and you do not discuss the matter with colleagues. You document what you observed or heard, and you file the formal report.

Applying the Framework in Practice

The triage framework works best when it is written down and shared with the lab team before a concern arises. A one-page document that describes the three categories and the decision rules for each can be distributed during lab orientation and reviewed at annual lab meetings. This document is not a substitute for the institutional policy, but it gives the team a common language for discussing concerns.

The framework also helps the lab manager respond consistently. When a team member raises a concern, the manager can walk through the categories with the person and identify which category applies. This process is transparent and reduces the risk that the manager will dismiss a legitimate concern or escalate a minor issue.

Escalation Criteria

The escalation criteria are the specific conditions that move a concern from one category to the next. These criteria should be written down and shared with the lab team.

The Committee on Publication Ethics Core Practices describe the institutional process for handling allegations, and the escalation criteria in the framework align with that process. The framework does not replace the institutional policy, but it gives the lab manager a clear path from observation to formal report.

Common Triage Mistakes

The most common mistake in triage is treating a quality issue as a fabrication concern. A lab manager who escalates a transcription error to the research integrity office creates unnecessary stress for the team and may damage trust. The framework prevents this by requiring a plausible mechanism for the quality category.

The second most common mistake is treating a fabrication concern as a quality issue. A lab manager who explains away a repeated pattern of identical values as a coincidence is failing to act. The framework prevents this by requiring escalation when the pattern repeats.

The third mistake is delaying the decision. A lab manager who is unsure which category applies may wait to see whether the concern resolves itself. The framework requires a decision within a defined period, typically one to two weeks, and the decision is documented in the lab record.

Documentation of the Triage Decision

The triage decision should be documented in a confidential file that is separate from the lab notebook. The file should contain the date of the observation, the category assigned, the reason for the category, and the action taken. This documentation is not shared with the team, but it is available if the institution later asks about the handling of the concern.

The documentation also protects the lab manager. If the concern is later found to be valid, the documentation shows that the lab manager acted promptly and followed a structured process. If the concern is found to be invalid, the documentation shows that the lab manager did not act on a rumor or a misunderstanding.

When to Escalate Beyond the Institution

The triage framework covers the internal handling of a concern. There are situations where the concern must also be reported to an external body. The NIH Grants and Funding pages describe the responsibilities of institutions to report misconduct to the funding agency, and the institution will handle this reporting. The lab manager does not report directly to the funding agency.

If the concern involves a publication, the journal may have its own process for handling allegations. The Committee on Publication Ethics Core Practices describe the responsibilities of journals to handle allegations, and the journal may contact the institution to request an investigation. The lab manager should be aware that the journal may be involved, and the institution will coordinate with the journal as needed.

The framework is a practical tool for the lab manager, but it does not replace the institutional policy. The lab manager should always follow the institutional policy when a concern reaches the formal category, and the framework is a supplement to that policy, not a substitute for it.

Frequently Asked Questions

What is the difference between data fabrication and data falsification?

Data fabrication is making up data or results and recording or reporting them as if they were real. Data falsification is manipulating or changing existing data to obtain a false result. Both are forms of research misconduct, and the Committee on Publication Ethics Core Practices describe both as behaviors that institutions should address.

Should I confront the person I suspect of fabrication?

No. Confronting the accused person directly is not the recommended approach. The institution has a formal process for handling allegations, and that process is designed to protect all parties. The Committee on Publication Ethics Core Practices describe the institutional process, and the lab manager should follow that process instead of confronting the person.

What should I do if I am not sure whether a discrepancy is fabrication?

Document the observation and discuss it with a trusted senior researcher who is not involved in the work. If the concern survives that review, file a formal report with the research integrity office. The institution will determine whether the concern warrants an inquiry.

Will my report be kept confidential?

The institution will protect the identity of the reporter to the extent of the law and policy, but the report may need to be shared with the accused person, the investigators, or the funding agency. The Committee on Publication Ethics Core Practices describe the expectations for confidentiality, and the institution's policy will define the specific protections.

What happens after I file a report?

The institution will conduct an inquiry to determine whether the concern warrants a full investigation. The inquiry is a preliminary finding of fact, and the institution will notify you of the outcome. The Committee on Publication Ethics Core Practices describe the process that institutions should follow.

Can I be punished for reporting a concern in good faith?

Institutions have policies that protect reporters from retaliation, and the NIH Grants and Funding pages describe the requirements for protecting whistleblowers. If you report a concern in good faith, you should be protected from retaliation. If you report a concern that you know is false, the protection may not apply.

What should I do with the original data files?

Preserve the original data files in their current state. Do not alter, delete, or move them. The institution will need to examine the original data as part of the inquiry, and any change you make could affect the review.

How long does an investigation take?

The timeline for an investigation varies by institution and by the complexity of the case. The Committee on Publication Ethics Core Practices expect that allegations will be handled promptly, but the specific timeline is set by the institution's policy.

Related Articles

• Protein Synthesis Step By Step
• Protein Synthesis: A Step-by-Step Guide to Transcription and Translation
• Lentivirus Protocol
• Protein Data Bank
• Statistical Data Analysis

References and Further Reading

• Research Methods Resources. National Library of Medicine.
• EQUATOR Network. EQUATOR Network.
• Core Practices. Committee on Publication Ethics.
• NIH Grants and Funding. National Institutes of Health.
• ORCID for Researchers. ORCID.
• Data Management and Sharing Policy. National Institutes of Health.
• NCBI Literature Resources. National Center for Biotechnology Information.
• PubMed. National Library of Medicine.

This article is educational and does not replace institutional policy, professional advice, or applicable safety and regulatory requirements.