Data Integrity in Medical Writing

By Dr. Zubair Khalid, DVM, MS, PhD · 2026-08-27

Quick Answer

• Verify every numeric value, citation, and image against its primary source before submission, and document the verification trail for each claim.
• Learn to recognize image manipulation red flags such as duplicated background patterns, inconsistent lighting, and irregular borders around gel bands or microscopy fields.
• Escalate suspected data fabrication to the corresponding author or institutional research integrity office immediately, and never alter or destroy original records while an inquiry is pending.

What Data Integrity Means in Medical Writing

Data integrity in medical writing refers to the complete, consistent, and accurate representation of research findings across all written deliverables, including manuscripts, clinical study reports, regulatory submissions, and patient-facing materials. The concept extends beyond avoiding outright falsification. It includes the prevention of inadvertent errors, the transparent reporting of methods and limitations, and the preservation of original data records that support every claim in a document.

Medical writers occupy a unique position in the research enterprise. They are often the last professionals to review a manuscript before submission, yet they rarely have direct access to the raw laboratory data that underpin the results section. This creates a practical tension. Writers must trust the authors who supply the numbers, but they also bear professional responsibility for the accuracy of the text they produce. The resolution to this tension is a systematic verification workflow that treats every data point as unverified until checked against a primary source.

The stakes are substantial. Published literature informs clinical decisions, regulatory approvals, and future research directions. A single incorrect value in a table can mislead clinicians who rely on that figure to assess drug efficacy or safety. An improperly cropped Western blot image can trigger a journal investigation, a retraction, and damage to the careers of everyone listed as an author. Medical writers who understand data integrity principles protect themselves, their employers, and the scientific record.

The Medical Writer Role in the Research Ecosystem

Medical writers translate complex biomedical data into clear, accurate documents for diverse audiences. The U.S. Bureau of Labor Statistics Occupational Outlook Handbook describes technical writers as professionals who communicate complex information clearly and who often work in specialized fields such as medicine and science. Within this occupation family, medical writers typically hold advanced degrees in life sciences, public health, or related disciplines, and many have laboratory or clinical experience before transitioning to writing roles.

The O*NET OnLine database, maintained by the U.S. Department of Labor, lists the core tasks of technical writers to include organizing material, studying drawings and specifications, and selecting photographs or diagrams that support the text. For medical writers, these tasks translate into specific responsibilities. They organize clinical data into logical result narratives, study statistical outputs to ensure the text matches the tables, and select figures that accurately represent the underlying measurements.

Medical writers work in several distinct settings. Pharmaceutical companies employ writers to produce regulatory documents for submission to agencies. Contract research organizations hire writers to support clinical trial reporting. Academic medical centers employ writers to assist investigators with grant applications and journal submissions. Medical communications agencies produce educational materials for healthcare professionals and patients. Each setting carries its own data integrity expectations, but the core principles remain consistent.

The National Institutes of Health Office of Intramural Training and Education emphasizes that research careers require ongoing professional development in responsible conduct of research. Medical writers who pursue certification through professional organizations demonstrate their commitment to these standards. Certification programs typically require coursework in regulatory writing, biostatistics, and ethical publication practices, followed by an examination that tests applied knowledge.

Core Principles of Data Integrity

Accuracy and Verification

Accuracy means that every number, statement, and citation in a medical document matches its source. Verification is the process of confirming that match. A medical writer should never assume that a value provided by an author is correct simply because the author supplied it. Human error is common in data transcription, and authors frequently copy values from statistical outputs into manuscripts without rechecking the original analysis.

The verification process begins with the study protocol or statistical analysis plan. These documents define the endpoints, the analysis populations, and the statistical methods. The writer uses them to understand what the results should represent. Next, the writer compares each value in the results section against the statistical tables or listings provided by the biostatistician. Discrepancies are flagged and resolved before submission.

Transparency in Methods

Transparent reporting means describing what was done in sufficient detail that another researcher could replicate the study. This includes the study design, participant selection criteria, interventions, outcome measures, and statistical methods. The writer must resist the temptation to omit inconvenient details or to present methods in a way that overstates the strength of the evidence.

The National Science Foundation Careers page illustrates how federal research organizations expect scientists to document their work rigorously. The same expectation applies to the written products that emerge from research. A methods section that omits the number of participants excluded from analysis, or that fails to describe how missing data were handled, compromises the integrity of the entire manuscript.

Attribution and Citation

Proper attribution means giving credit to the sources of ideas, methods, and data. Medical writers must ensure that every claim that originates from another publication is cited accurately. This includes verifying that the cited source actually contains the claimed information. A common error is citing a review article for a finding that the review itself cited from a primary study. The writer should trace the claim to its original source whenever possible.

Citation accuracy also involves checking that reference details match the published record. Authors, journal names, volume numbers, page ranges, and publication years must all be verified. Errors in references undermine reader confidence and can prevent readers from locating the cited work.

The Verification Workflow

Step 1: Inventory All Data Points

Before writing begins, create a complete inventory of every data point that will appear in the document. This includes numerical values in the text, table entries, figure labels, and statistical notations. The inventory serves as a checklist for verification. Each data point should be traceable to a specific source document, such as a statistical output file, a case report form, or a laboratory report.

Step 2: Compare Against Primary Sources

For each data point in the inventory, locate the corresponding value in the primary source and compare them. This step requires access to the statistical tables, clinical databases, or laboratory records that generated the results. When the writer does not have direct access to these sources, the corresponding author must provide them or confirm the values in writing.

Step 3: Verify All Citations

Check every reference against the published record. Confirm that the citation exists, that the authors and year are correct, and that the cited content actually supports the claim in the text. This step is time-consuming but essential. Many journals now require authors to confirm that they have verified all references before submission.

Step 4: Document the Verification Trail

Maintain a verification log that records the date each data point was checked, the source used for verification, and the outcome of the check. This log serves multiple purposes. It provides evidence of due diligence if questions arise after publication. It also helps the writer track which items remain unverified and require follow-up with the authors.

Step 5: Obtain Written Confirmation for Unverifiable Items

Some data points cannot be independently verified by the writer. This includes patient-level data that the writer cannot access due to privacy restrictions, or proprietary information that the sponsor has not released. For these items, obtain written confirmation from the responsible party that the values are accurate as presented. This confirmation should be retained in the project files.

At a Glance

Image Integrity and Manipulation Red Flags

Why Image Integrity Matters

Scientific images carry evidentiary weight. Western blots demonstrate protein expression, microscopy images show cellular morphology, and clinical photographs document treatment outcomes. Readers trust that these images accurately represent what the researchers observed. Image manipulation that misrepresents the underlying data constitutes research misconduct, regardless of whether the manipulation was intentional.

The American Society for Microbiology Careers page emphasizes that professional integrity is foundational to careers in the microbiological sciences. This principle extends to the images that accompany published research. Journals now routinely screen submitted images for signs of manipulation, and reviewers are trained to examine figures critically.

Common Manipulation Techniques

Image manipulation in scientific publications typically falls into several categories. Cropping removes portions of an image to exclude unwanted bands or cells. Splicing joins images from different experiments to create a composite that appears to come from a single experiment. Duplication copies a region of an image and pastes it elsewhere, often to replicate a band or cell. Brightness and contrast adjustments can obscure background bands or enhance weak signals. Each of these techniques can misrepresent the experimental results.

Red Flags in Western Blots

Western blot images present specific red flags that writers should learn to recognize. Duplicated bands appear as identical patterns in different lanes. Irregular borders around individual lanes suggest that bands were cut from other images and pasted into the figure. Background patterns that do not align across the blot indicate that the image was assembled from multiple sources. Excessive contrast adjustment can create a harsh, unnatural appearance that obscures the original signal.

Red Flags in Microscopy Images

Microscopy images present different challenges. Duplicated cells or organelles within a field suggest that regions were copied and pasted. Inconsistent lighting across the image indicates that the field was assembled from multiple exposures. Irregular borders around the image perimeter suggest cropping that removed original context. The writer should compare the image against the scale bar to ensure that the magnification is consistent with the methods description.

The Writer Response to Suspected Manipulation

When a writer suspects image manipulation, the appropriate response is to request the original uncropped images from the authors. Many journals now require authors to submit original images as supplementary files, and the writer should verify that the submitted figures match these originals. If the original images reveal manipulation, the writer must escalate the concern to the corresponding author and, if necessary, to the institutional research integrity office.

Records and Documentation Standards

The Audit Trail

An audit trail is a chronological record of all changes made to a document or dataset. In regulated environments, such as clinical trials supporting regulatory submissions, audit trails are mandatory. Regulatory agencies require that electronic records include audit trails that capture who made changes, what changes were made, and when they were made. Medical writers working on regulatory documents must ensure that their working files preserve this information.

Version Control

Version control is the practice of tracking document revisions systematically. Each version receives a unique identifier, a date, and a description of the changes made. This practice allows the writer to reconstruct the evolution of a document and to identify when specific data points were introduced or modified. Version control also protects against the accidental overwriting of earlier drafts that may contain important information.

File Naming Conventions

Consistent file naming conventions support data integrity by making files easy to identify and locate. A recommended convention includes the project identifier, the document type, the version number, and the date. For example, a clinical study report for protocol ABC-123 might be named ABC-123_CSR_v03_2025-01-15.docx. This convention prevents confusion between versions and ensures that the most recent version is readily identifiable.

Retention of Source Documents

Source documents include the statistical outputs, laboratory reports, and clinical databases that support the data in a medical document. These documents must be retained for the duration required by the applicable regulations or institutional policies. The writer should maintain a file that maps each data point in the document to its source document, so that any question about the data can be resolved quickly.

Common Failure Patterns in Data Integrity

Transcription Errors

Transcription errors occur when a value is copied incorrectly from one document to another. A common pattern involves the misreading of decimal points, the transposition of digits, or the confusion of similar-looking numbers. These errors are often detected only when a reviewer compares the manuscript against the statistical output. The writer can reduce transcription errors by using a two-person verification process for critical values.

Inconsistent Numbers Across Sections

A manuscript may report the same value in the abstract, the results section, a table, and a figure. Inconsistencies arise when one of these sections is updated without updating the others. For example, an author may revise the primary outcome analysis and update the results section but forget to update the abstract. The writer should check all sections for consistency before submission.

Citation Errors

Citation errors include incorrect author names, wrong publication years, and references that do not support the claims they are cited for. These errors are common because authors often copy references from other papers without verifying them. The writer should verify every reference against the original publication or a reliable database such as PubMed.

Unsupported Conclusions

An unsupported conclusion occurs when the discussion section makes claims that the results do not actually support. This pattern often involves overgeneralization from a limited sample, or the attribution of causal relationships to correlational findings. The writer should compare each conclusion against the results and flag any claims that exceed the evidence.

Selective Reporting

Selective reporting occurs when authors present only the analyses that support their hypothesis and omit those that do not. This pattern is difficult for a writer to detect without access to the full analysis plan. The writer should request the statistical analysis plan and compare the reported analyses against the planned analyses to identify any omissions.

Practical Implementation Steps

Establish a Verification Protocol

Create a written verification protocol that describes the steps the writer will follow to verify data accuracy. The protocol should specify who is responsible for each verification task, what sources will be used, and how discrepancies will be resolved. This protocol should be shared with the authors and the project team before writing begins.

Use a Data Verification Checklist

Develop a checklist that covers all the verification tasks required for a typical medical document. The checklist should include items such as confirming that all values in the text match the tables, verifying that all references are accurate, checking that all figures have original images on file, and confirming that the methods section matches the study protocol.

Conduct a Pre-Submission Review

Before the document is submitted, conduct a final review that focuses specifically on data integrity. This review should be separate from the content review and should involve a fresh look at the document with attention to numbers, citations, and images. If possible, have a second person conduct this review to provide an independent check.

Document All Discrepancies

When a discrepancy is identified, document it in the verification log. Record the nature of the discrepancy, the source that revealed it, and the resolution. This documentation provides evidence of the verification process and helps identify patterns of error that may indicate larger problems with the underlying data.

Records and Measurements for Quality Control

Verification Log Metrics

Track the number of data points verified, the number of discrepancies found, and the time required for verification. These metrics help the writer estimate the resources needed for future projects and identify areas where the verification process can be improved.

Discrepancy Resolution Rate

Measure the percentage of discrepancies that are resolved before submission. A high resolution rate indicates that the verification process is effective. A low rate may indicate that the authors are not providing the necessary source documents or that the verification process is not being followed consistently.

Citation Verification Rate

Track the percentage of references that are verified against the original source. This metric is particularly important for regulatory documents, where citation accuracy is critical. A citation verification rate below 100 percent indicates that some references remain unverified and require follow-up.

Image Verification Rate

Track the percentage of figures for which original uncropped images have been obtained and compared against the submitted version. This metric is especially important for journals that require original image submission. A low image verification rate increases the risk of undetected manipulation.

Welfare and Safety Context

Patient Safety Implications

Data integrity in medical writing has direct implications for patient safety. Clinical trial results that are reported inaccurately can lead to incorrect dosing recommendations, the approval of ineffective treatments, or the rejection of safe and effective therapies. The medical writer's verification work is a safeguard for the patients who will ultimately receive the treatments described in the documents.

Research Participant Protection

Research participants consent to participate in studies based on the expectation that their data will be handled responsibly. Inaccurate reporting of study results violates this expectation and can undermine trust in the research enterprise. The writer's commitment to data integrity honors the contribution of research participants.

Professional Reputation

A medical writer whose name appears on a manuscript that is later retracted for data integrity issues may face professional consequences. These can include damage to reputation, loss of clients, and difficulty obtaining future employment. The Occupational Outlook Handbook notes that technical writers must be detail-oriented and able to work independently. These qualities are essential for maintaining data integrity.

Institutional Accountability

Research institutions have policies governing research misconduct, and they are responsible for investigating allegations of fabrication, falsification, and plagiarism. The National Institutes of Health Office of Intramural Training and Education provides training on the responsible conduct of research, which includes instruction on data management and reporting standards. Medical writers should be familiar with their institution's policies and with the procedures for reporting suspected misconduct.

Professional Escalation Criteria

When to Escalate a Concern

A medical writer should escalate a data integrity concern when the discrepancy cannot be resolved through normal verification processes. This includes situations where the author refuses to provide source documents, where the discrepancy suggests intentional manipulation, or where the discrepancy affects the interpretation of the study results.

How to Escalate

The first step in escalation is to raise the concern with the corresponding author in writing. The writer should describe the specific discrepancy, the evidence that supports the concern, and the proposed resolution. If the author does not respond satisfactorily, the writer should escalate to the institutional research integrity officer or the journal editor.

Documentation of Escalation

All escalation communications should be documented in writing. This includes the initial concern, the response received, and the final resolution. This documentation protects the writer if questions arise later about the handling of the concern.

Protection Against Retaliation

Medical writers who report suspected misconduct may fear retaliation from authors or employers. Many institutions have policies that protect individuals who report research misconduct in good faith. The writer should be aware of these policies and should report concerns through the appropriate channels.

Frequently Asked Questions

What qualifications do medical writers need?

Medical writers typically hold a bachelor's or master's degree in a life science field, and many have advanced degrees such as a PhD or PharmD. The Occupational Outlook Handbook notes that technical writers often need a bachelor's degree and that those in specialized fields may need additional education or experience. Relevant laboratory or clinical experience is highly valued by employers.

How do I start a career in medical writing?

Start by gaining experience in scientific communication through internships, academic writing, or volunteer work. The American Society for Microbiology Careers page offers career development resources that can help you identify opportunities and build professional connections. Consider pursuing certification through a professional organization to demonstrate your qualifications.

Is medical writer certification required?

Certification is not legally required, but it can enhance your credentials and demonstrate your commitment to the field. Certification programs typically require coursework and an examination. The National Institutes of Health Office of Intramural Training and Education provides professional development resources that can help you prepare for certification.

What is the difference between medical writing and regulatory writing?

Medical writing encompasses a broad range of documents, including manuscripts, educational materials, and conference presentations. Regulatory writing focuses specifically on documents submitted to regulatory agencies, such as clinical study reports and investigator brochures. Regulatory writing requires knowledge of specific regulations and submission formats.

How do I verify data when I do not have access to the raw data?

When you do not have access to raw data, request written confirmation from the responsible party that the values are accurate. Document this confirmation in your verification log. If the responsible party cannot provide confirmation, escalate the concern to the project lead or the corresponding author.

What should I do if I suspect image manipulation?

Request the original uncropped images from the authors and compare them against the submitted figures. If the comparison reveals manipulation, escalate the concern to the corresponding author and, if necessary, to the institutional research integrity office. Do not alter or destroy any files related to the concern.

How long should I retain source documents?

Retention requirements vary by institution and by the type of document. Regulatory documents may have specific retention requirements that are defined by the applicable regulations. Check your institution's policies and the requirements of the relevant regulatory agency to determine the appropriate retention period.

What are the career prospects for medical writers?

The Occupational Outlook Handbook projects employment for technical writers to grow as organizations continue to produce complex documents that require specialized communication skills. Medical writing is a specialized niche within this field, and demand is driven by the ongoing need for clinical trial reporting, regulatory submissions, and scientific publications.

Using the Evidence

Related Articles

• Research Data Management Careers: Librarians and Specialists
• Clinical Research Data Management: Core Skills and Career Pathways
• Research Data Management Jobs: Roles, Skills, and Career Opportunities
• Research Data Management Plan: From File Naming to Long-Term Sharing
• california medical

References and Further Reading

• Occupational Outlook Handbook. U.S. Bureau of Labor Statistics.
• O*NET OnLine. U.S. Department of Labor.
• Office of Intramural Training and Education. National Institutes of Health.
• Careers at NSF. U.S. National Science Foundation.
• Career Services. American Society for Microbiology.
• AAVMC Career Resources. Association of American Veterinary Medical Colleges.
• 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.