How Do You Know Whether Two Papers Report the Same Study?

Are These Really Two Studies, or Two Papers From One Study?

You find two papers that look sufficiently different to survive ordinary reference deduplication. They have different titles, perhaps different author lists, and they may even emphasize different outcomes. Should they enter your review as two studies?

Not necessarily. A single study can generate a protocol, conference abstract, primary results paper, secondary analysis, follow-up paper, harms report, or other publication. PRISMA 2020 therefore distinguishes a report, which is a document supplying information about a study, from the underlying study itself. One study may have multiple reports.

This distinction matters because treating several reports of the same underlying investigation as independent studies can duplicate evidence and, in quantitative synthesis, may result in the same participants contributing more than once. The difficult part is that related papers do not always announce their relationship clearly.

Look for a Pattern of Matching Study Characteristics

How to Work Out Whether Publications Belong to the Same Underlying Study

Start by Separating the Paper From the Study

The first conceptual step is simple but consequential: a paper is not necessarily a study. A paper is one report of research, while the study is the underlying investigation that generated the data. Conversely, a single paper can sometimes report more than one study.

This is why bibliographic deduplication is not enough. Reference-management software can usually detect two database records describing the same article. It cannot reliably determine that two genuinely different articles came from the same research project.

Look First for Unique Study Identifiers

The strongest clue is often an identifier that explicitly connects publications to the same research project. Depending on the field, this might include a trial registration number, protocol registration, study acronym, grant number, ethics approval number, cohort name, or another persistent study identifier.

If two papers report the same trial registration number, for example, that is strong evidence that they are reports associated with the same registered trial. You should still inspect what each paper actually analyzes because one publication might use a subset, secondary analysis, or later follow-up of the original participants.

Identifiers are particularly useful because titles and author lists can change substantially between publications.

Compare Where and When Participants Were Recruited

Recruitment information is one of the most useful fingerprints of a study. Compare the institutions, hospitals, schools, communities, countries, laboratories, registries, or other settings from which participants were obtained. Then compare recruitment dates and study duration.

Suppose Paper A reports recruiting participants at three hospitals between January 2021 and June 2022. Paper B reports the same hospitals and recruitment period. That combination is considerably more informative than discovering that the papers merely share an author.

Dates also help explain apparent discrepancies. One publication might describe an interim analysis after 300 participants, while a later paper reports the completed sample of 500. Different sample sizes therefore do not automatically establish that the studies are different.

Compare Participant Numbers and Baseline Characteristics

Examine the number recruited, allocated, analyzed, or followed up. Then look beyond the total sample size to the characteristics of the participants.

Useful details may include age distributions, sex or gender composition, diagnostic criteria, eligibility requirements, baseline measurements, treatment-group sizes, school grade, occupational category, or other population-specific characteristics.

Near-identical baseline profiles can be revealing. If two randomized trials conducted at the same sites during the same period both contain 214 participants allocated 107 versus 107, with matching demographic characteristics, the probability that this happened independently deserves scrutiny.

Still, participant counts should not be treated as a universal fingerprint. Attrition, missing data, subgroup analyses, different analytic populations, and extended recruitment can produce different numbers across reports. Determining whether papers contain overlapping study populations may therefore require more detailed comparison.

Compare the Design and Study Procedures

Look at characteristics that would have been determined when the underlying research was conducted rather than when the paper was written. These may include the study design, intervention or exposure, comparator, allocation ratio, intervention dose or duration, measurement schedule, inclusion criteria, and follow-up period.

A distinctive combination is particularly informative. Two papers describing the same unusual intervention, delivered according to the same schedule at the same institutions during the same recruitment period, are much more likely to be related than two papers that merely use the same broad research design.

Compare Authors, but Do Not Treat Authorship as Proof

Shared authorship is a useful clue, particularly when the same principal investigator or research group appears across publications. It is not a requirement.

Cochrane specifically cautions that reports from the same study may not share all authors and may not explicitly reference one another. Author order can also change when different papers emphasize different outcomes or disciplinary contributions.

The reverse problem occurs as well. A prolific research team may conduct several similar studies in the same population. Shared authorship therefore supports an identification only when considered alongside other evidence.

Look for Explicit Cross-References

Read the methods section carefully. Authors may state that the analysis is based on a previously described trial, an existing cohort, a larger research project, or data reported elsewhere. Citations to a protocol or earlier publication can reveal connections that are invisible from the title and abstract.

Phrases such as “secondary analysis,” “follow-up of participants,” “data were drawn from,” or “the study protocol has been published previously” deserve attention. Follow the cited references rather than assuming the relationship from the wording alone.

Build a Study Fingerprint Instead of Looking for One Perfect Match

In ambiguous cases, it helps to compare papers systematically rather than relying on an impression that they “look similar.”

The goal is not to accumulate an arbitrary number of matches. Consider how specific each clue is. The same country and research design provide relatively weak evidence. The same registration number, recruitment sites, dates, allocation pattern, and baseline characteristics provide a much stronger case.

Think in Terms of Same Study, Different Study, or Uncertain

Not every pair of publications can be classified confidently from the available reports. Forcing a binary decision too early can create errors in either direction.

A practical approach is to classify candidate reports provisionally as belonging to the same study, different studies, or having an uncertain relationship. Uncertainty can then trigger further checking of protocols, registries, supplementary files, cited papers, institutional repositories, or other available sources.

If uncertainty remains consequential to the review, contacting the study investigators may be appropriate. Cochrane specifically recommends correspondence with authors or principal investigators when the relationship between reports cannot otherwise be resolved.

Tracing Two Apparently Different Papers Back to One Study

Common Mistakes When Linking Study Reports

Treat Study Identification as a Documented Review Decision

When you suspect that papers are related, compare them at the study level before extracting or synthesizing their results as independent evidence. Record why you linked them. This creates an audit trail if another reviewer later questions the decision.

Once reports have been linked, organize your extraction around the underlying study. Cochrane recommends collating multiple reports so that the study rather than the report remains the unit of interest. This does not mean that every piece of information must come from one publication. Different reports may contribute different methodological details or outcomes.

This becomes particularly important when deciding whether multiple papers represent one study or several pieces of evidence and when using companion papers without double-counting participants.

What to Compare Before Treating Papers as Independent Studies

Questions About Identifying Multiple Reports of the Same Study

Identify the Underlying Investigation, Not Just the Publication

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.