03 · What You Need to Know
A missing full paper is a research trail to investigate
First, make sure the full paper really is missing
Failure to find a journal article using the conference abstract's exact title is not enough to conclude that the study was never fully published.
Titles frequently change between conference presentation and publication. Authors may change order. The final article may emphasize a different outcome. A multicenter study may be published under a group name. A preliminary conference sample may later become part of a larger study. The full report may also exist as a dissertation, report, preprint, working paper, or other document rather than as the journal article you expected.
Before classifying the research as abstract-only, search using several identifiers:
- the complete conference abstract title;
- distinctive phrases from the title rather than only the exact title;
- first, corresponding, and senior author names;
- author names combined with the intervention, exposure, population, or main topic;
- institution or study location;
- sample size and distinctive methodological characteristics;
- study acronym or project name;
- trial or study registration number, if reported;
- conference name and year;
- citations to the abstract or study.
The broader techniques used to trace theses, reports, working papers, and other grey literature can be useful here because the next version of the research may not be a journal article.
Do not match publications by title alone
Suppose you find a later paper by the same authors on approximately the same topic. That does not automatically establish that it is the full publication corresponding to the abstract.
Compare characteristics such as:
- participants and eligibility criteria;
- sample size;
- recruitment dates;
- study setting and locations;
- intervention or exposure;
- comparison groups;
- study design;
- outcomes;
- follow-up period;
- registration number;
- funding or project identifiers.
Some differences are expected because the study may have progressed after the conference. A larger sample or longer follow-up does not necessarily mean you have found a different study. You need to judge the pattern of identifying characteristics collectively.
When uncertainty remains and the distinction matters to the review, contact the authors rather than quietly deciding that two reports are either identical or unrelated.
Why do some conference abstracts never become full papers?
Nonpublication does not have one explanation, and it should not automatically be interpreted as evidence that a journal rejected the research.
A systematic review examining investigators' stated reasons for not publishing research presented at biomedical conferences found a mean full-publication rate of 55.9% across 24 reports that provided the relevant data. Among reports examining clinical trials, the corresponding average was 73.0%. Lack of time was the most frequently reported reason for nonpublication and the reason most frequently identified as the most important. Other reported explanations included limited time or resources, publication not being an aim, low priority, an incomplete study, and difficulties involving co-authors.
The investigators concluded that, across the biomedical specialties examined, reasons for nonpublication were more often related to authors than to journal editors.
These figures should not be generalized mechanically to every discipline or conference. Publication rates vary substantially by field, meeting, study design, follow-up period, and publication culture. They do establish an important point: a conference abstract remaining unpublished in full is not an exceptional enough event to justify assuming that something must have been fundamentally wrong with the study.
Nonpublication can nevertheless be related to the findings
Although authors report many practical reasons for failing to produce full papers, another methodological concern remains: research with statistically significant or favorable findings may have a greater probability of full publication or may reach publication sooner.
Methodological research on conference abstracts has found evidence of this selective progression. One Cochrane methodology review reported that statistically significant results were associated with subsequent full publication, as were results favoring the intervention. Statistically significant findings were also reported to reach full publication sooner.
This is why simply deleting abstract-only studies from a systematic review can be consequential. If the probability of full publication is related to the findings, restricting the evidence base to studies that successfully became full papers can contribute to a systematically distorted picture of the available evidence.
Watch Out
Do not infer why one particular study remained unpublished from population-level evidence about publication bias. A null result may remain unpublished, but so may a positive result whose authors lacked time, changed jobs, lost collaborators, or simply never prioritized the manuscript.
Contacting the authors can resolve several different uncertainties
When an abstract appears eligible and no full report can be located, contacting the authors may be worthwhile, particularly in a systematic review where the missing study could influence the synthesis.
A focused request can ask:
- whether the study was subsequently published under another title;
- whether a manuscript, report, thesis, preprint, or other fuller version exists;
- whether the study was completed;
- whether the conference results were preliminary or final;
- for clarification of an eligibility criterion that is not reported in the abstract;
- for numerical outcome data required for synthesis, when appropriate;
- for a study registration number or other identifier.
Keep the request specific. Authors are more likely to understand and respond to a concise question about a particular study than to an open-ended request for "all unpublished data."
Author contact can also reveal that what appeared to be an unpublished study was eventually reported somewhere you had not searched.
No response from the authors is still not evidence about the study's findings
You may receive no reply. Email addresses become obsolete, researchers move institutions, investigators retire, and old studies may no longer be readily accessible even to their authors.
Do not interpret silence as evidence that the study was poor, abandoned because of null findings, or deliberately suppressed. It means only that your attempt to obtain additional information was unsuccessful.
For a systematic review, document the contact attempt and proceed according to the review methods using the information that is actually available.
Can you include the abstract if no full report exists?
Possibly. The answer depends on your eligibility criteria and what "include" means in your review.
An abstract may provide enough information to determine that the study meets your inclusion criteria. It may even contain sufficient numerical data for a particular analysis. In other cases, it may establish only that a potentially eligible study exists.
Eligible study with usable abstract data
The available report establishes eligibility and contains enough information for the specific descriptive or quantitative use planned in the review.
Potentially eligible study with insufficient information
The abstract indicates relevant research exists, but essential information needed to establish eligibility or use the results remains unavailable.
This distinction matters because including conference abstracts in a literature review is not necessarily an all-or-nothing decision. A study can be recognized and documented without forcing incomplete numerical information into a meta-analysis.
Do not invent what the abstract does not report
Conference abstracts often omit methodological details because they are short. If randomization is mentioned but allocation concealment is not, you cannot assume either that allocation concealment occurred or that it did not. If attrition is unreported, you cannot treat it as zero. If only a p-value is provided, you should not manufacture the underlying group statistics.
Similarly, if the abstract does not report whether the study had finished recruitment, do not assume the conference sample was final.
Missing information should remain missing unless another source or the investigators supply it.
Be particularly cautious with preliminary results
Conference presentations often occur before the research record is final. An abstract may describe interim recruitment, early follow-up, preliminary analyses, or a subset of outcomes.
Empirical comparisons between conference abstracts and later full publications have documented changes in participant numbers and analyses. One study of randomized trials found that only 46% of abstract-publication pairs reported the same number of participants randomized, while only 22% reported the same number analyzed. The authors concluded that information in conference proceedings can be unstable, particularly when abstracts describe early or preliminary results.
If the abstract explicitly labels the analysis preliminary, interim, or ongoing, treat those words as methodologically meaningful. A preliminary estimate is not automatically wrong, but it is not equivalent to a final analysis.
Abstract-only evidence may limit critical appraisal
Even when eligibility is clear, a conference abstract may provide too little information for a confident assessment of methodological quality or risk of bias.
This creates an important distinction between a flaw in the study and a flaw in the available reporting. If the abstract does not explain how participants were allocated, for example, the problem for the reviewer is lack of information. It is not proof that allocation was inappropriate.
Your appraisal should preserve that uncertainty rather than converting unreported information into assumptions.
Do not count the abstract again if a later publication appears
If you eventually identify a full paper, the abstract and article usually represent multiple reports of the same study rather than two independent studies.
Link them. Do not count participants twice. Compare the versions for changes in sample size, outcomes, analyses, follow-up, and conclusions. Use the fuller or more appropriate source for each piece of information while preserving any consequential discrepancies.
This becomes especially important when a later publication emerges after a review has already been completed. An abstract-only study may need to be updated rather than added as a new independent study.
A study can matter even when its results cannot be synthesized
Suppose the abstract clearly describes an eligible randomized trial but does not provide enough numerical data for meta-analysis. The study's existence may still matter when interpreting the evidence base.
If several eligible studies are known to exist but their results are unavailable, a synthesis based only on the fully reported studies deserves more caution than one in which essentially all known eligible research is represented.
This is one reason the broader task of finding unpublished and ongoing studies matters even when every identified study cannot ultimately contribute data.
Do not quietly erase the study from the review record
For transparent evidence synthesis, an abstract-only study should not vanish simply because the reviewers could not obtain sufficient information.
Its final disposition depends on the review methodology. It might be included, excluded for a documented reason, retained as awaiting classification, described as an ongoing or incomplete study, or acknowledged as an eligible study lacking usable outcome data.
The exact categories vary among review frameworks. What matters is that the reader can understand that the study was identified, what information was available, what attempts were made to clarify it, and why it did or did not contribute to the synthesis.
Update the search before concluding that no full publication exists
Publication takes time. A conference abstract identified early in a review may become a full article before the manuscript is submitted or while the review itself is undergoing peer review.
For reviews with an updated search near completion, recheck unresolved conference abstracts. Search the authors and study identifiers again rather than relying indefinitely on the result of the first search.
A study that was genuinely abstract-only twelve months ago may no longer be abstract-only today. Academic publishing does occasionally move while the reviewer is looking away.
